Docstoc

Diabetes in pregnancy

Document Sample
Diabetes in pregnancy Powered By Docstoc
					                                                                                              DRAFT FOR CONSULTATION




Appendix D             Evidence tables

Q1 - What information should be offered in relation to outcomes and risks for the mother and the baby?

No specific searches were conducted for this clinical question and so there are no evidence tables




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                             Page 1 of 441
                                                                                              DRAFT FOR CONSULTATION




Q2 - What information should be offered in relation to the importance of planning a pregnancy and the role of
contraception?

No specific searches were conducted for this clinical question and so there are no evidence tables




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                             Page 2 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Q 3 - What information should be offered in relation to diet, dietary supplements, body weight and exercise?

Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &            Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                   Comments
                 Level                 cs                                                Measures
Kaplan JS;Iqbal   Study Type:       31 pregnant                 Intervention: Folate   Follow-up period: Results for                  There were no significant
S;England         Case-control      women with                  metabolism in                            unsupplemented               differences between the
BG;Zawacki                          diabetes         Country:   pregnant women         Outcome           participants (15 diabetic,   pregnant diabetic and
CM;Herman WH;     Evidence level:   54 nondiabetic              with diabetes          Measures:         34 control)                  nondiabetic pregnant
                  2+                pregnant                                           Measures of                                    women for any measures
1999 Jul                            women                       Comparison:            folate            Dietary folate               of folate metabolism after
                                                                Nondiabetic            metabolism:       Diabetic: 294±154            accounting for folate
39
                                                                pregnant women         Dietary folic     Control:455±334,             supplementation. In
                                                                                       intake            p=0.03                       addition, among diabetic
                                                                                       serum folate                                   women, there was no
                                                                                       red blod cell     Serum folate (ng/ml)         associations among
                                                                                       folate            Diabetic: 17.6±10.3          parameters of folate
                                                                                       urinary folate    Control:14.1±5.7, NS         metabolism and
                                                                                       homocysteine                                   glycaemic control.
                                                                                                         Red blood cell folate
                                                                                                         (ng/ml)
                                                                                                         Diabetic: 385±73
                                                                                                         Control:353±184, NS

                                                                                                          Urinary folate (ng/ml)
                                                                                                          Diabetic: 8.8±5.2
                                                                                                          Control:8.8±4.3

                                                                                                          Urinary folate
                                                                                                          (ng/ml)/urinary creatine
                                                                                                          (mg/dl)
                                                                                                          Diabetic: 0.08±0.05
                                                                                                          Control:0.07±0.03, NS

                                                                                                          Homocysteine (umol/l)
                                                                                                          Diabetic: 8.0±8.1
                                                                                                          Control:6.5±1.9, NS

                                                                                                          Vitamin B12 (pg/ml)
                                                                                                          Diabetic: 493±275
                                                                                                          Control:383±152, NS




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 3 of 441
                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                       Follow-up &           Effect Size            Study Summary            Reviewer
 Information & Evidence   Patients Characteristi Comparison                            Outcome                                                             Comments
                 Level                 cs                                              Measures
Ray               Study Type:       428 women                   Intervention:        Follow-up period: Results are rate and      Maternal obesity and
JG;Vermeulen      Cohort            with gestational            Prepregnancy                            adjusted OR (adjusted    excessive weight gain
MJ;Shapiro                          diabetes         Country:   weight               Outcome            for type of diabetes,    were independent risk
JL;Kenshole AB;   Evidence level:   196 women                   Weight gain during   Measures:          maternal age, parity,    factors for Caesarean
                  2++               with                                             Outcome            prepregnancy BMI, net    delivery, hypertensive
2001                                pregestational              Comparison:          Measures:          weight gain during       disorders of pregnancy,
                                    diabetes                                         Caesarean          pregnancy and obsteric   admission to neonatal
41
                                                                                     delivery           history)                 intensive care, LGA and
                                                                                     Shoulder dystocia                           preterm birth.
                                                                                     or cephalopelvic Caesarean delivery
                                                                                     disproportion      Prepregnancy BMI <20:
                                                                                     Gestational        30.2% (OR 1.0)
                                                                                     hypertension or    20.0-24.9: 39.0% (OR
                                                                                     toxaemia           1.2, 95%CI 0.5-2.8)
                                                                                     Admission to       25.0-29.9: 43.4% (OR
                                                                                     neonatal intensive 1.3, 95% CI 0.6-3.3)
                                                                                     care unit (NICU) ≥30: 55.3% (OR 3.5,
                                                                                     LGA                95% CI 1.4-8.6)
                                                                                     Preterm birth      Weight gain (5kg
                                                                                     (before 37 weeks) increments): OR 1.2,
                                                                                                        95%CI 1.0 -1.4.

                                                                                                       Shoulder dystocia or
                                                                                                       cephalopelvic
                                                                                                       disproportion
                                                                                                       Prepregnancy BMI <20:
                                                                                                       14% (OR 1.0)
                                                                                                       20.0-24.9: 21.1% (OR
                                                                                                       1.5, 95%CI 0.6-3.8)
                                                                                                       25.0-29.9: 16.4% (OR
                                                                                                       1.3, 95%CI 0.5-3.5)
                                                                                                       ≥30: 15.4% (OR 1.2,
                                                                                                       95%CI 0.4-3.4)
                                                                                                       Weight gain (5kg
                                                                                                       increments): OR 1.1
                                                                                                       (0.9-1.3)

                                                                                                       Hypertensive disorders
                                                                                                       of pregnancy
                                                                                                       Prepregnancy BMI <20:
                                                                                                       4.6% (OR 1.0)
                                                                                                       20.0-24.9: 10.6% (OR




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 4 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary         Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         1.8, 95%CI 0.4-8.1)
                                                                                         25.0-29.9: 13.2% (OR
                                                                                         2.6, 95% CI 0.6-11.8)
                                                                                         ≥30: 17.1% (OR 4.1,
                                                                                         95% CI 0.9-18.9)
                                                                                         Weight gain (5kg
                                                                                         increments): OR 1.4
                                                                                         (95% CI 1.2-1.7)

                                                                                         NICU admission:
                                                                                         Prepregnancy BMI <20:
                                                                                         39.5% (OR 1.0)
                                                                                         20.0-24.9: 54.1% (OR
                                                                                         1.2, 95% CI 0.5-2.7)
                                                                                         25.0-29.9: 61.8% (OR
                                                                                         1.9, 95% CI 0.8-4.4)
                                                                                         ≥30: 65.0% (OR 2.4,
                                                                                         95% CI 1.0-5.9)
                                                                                         Weight gain (5kg
                                                                                         increments):OR 1.2 (1.0-
                                                                                         1.4)

                                                                                         LGA
                                                                                         Prepregnancy BMI <20:
                                                                                         9.3% (OR 1.0)
                                                                                         20.0-24.9: 24.8% (OR
                                                                                         2.5, 95% CI 0.8-7.7)
                                                                                         25.0-29.9: 24.3% (OR
                                                                                         2.6, 95% CI 0.8-8.2)
                                                                                         ≥30: 26% (OR 3.3, 95%
                                                                                         CI 1.0-10.6)
                                                                                         Weight gain (5kg
                                                                                         increments):1.3 (1.1-1.6)

                                                                                         Preterm birth <37 weeks
                                                                                         Prepregnancy BMI <20:
                                                                                         7% (OR 1.0)
                                                                                         20.0-24.9: 28.0% (OR
                                                                                         5.0, 95% 1.4-17.6)
                                                                                         25.0-29.9: 28.3% (OR
                                                                                         5.5 (1.5-19.4)
                                                                                         ≥30: 27.6% (OR 5.1,




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 5 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary         Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                      Comments
                 Level                 cs                                   Measures
                                                                                         95% CI 1.4-18.6)
                                                                                         Weight gain (5kg
                                                                                         increments):OR 1.0 (0.8-
                                                                                         1.2).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 6 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                Follow-up &              Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                     Outcome                                                                     Comments
                 Level                 cs                                                       Measures
Moore, H; et al.   Study Type:       6 studies        Type 2 diabetes.   Intervention: Dietary Follow-up period:   The quality of the trials   The evidence suggests
                   Systematic        compared         Adults. Not        advice plus exercise                      was assessed to be at       that dietary advice plus
2005               review - meta-    dietary advice   pregnant.                                Outcome             high risk of bias           exercise has the potential
                   analysis          with dietary                        Comparison: Dietary Measures:                                         to reduce weight and
46
                                     advice plus      Country:           advice                Weight loss         Weight change was           improve glycaemic
                   Evidence level:   exercise                                                  HbA1c               reported in four trials.    control.
                   1++               (n=322)                                                                       More weight was lost on
                                                                                                                   average in the diet and
                                                                                                                   exercise groups. HbA1c
                                                                                                                   decreased more in the
                                                                                                                   participants in the
                                                                                                                   dietary advice and
                                                                                                                   exercise groups than in
                                                                                                                   those in the dietary
                                                                                                                   advice group alone.

                                                                                                                   At six months, dietary
                                                                                                                   advice plus exercise
                                                                                                                   was associated with a
                                                                                                                   statistically significant
                                                                                                                   mean (pooled weighted
                                                                                                                   mean difference)
                                                                                                                   decrease in glycated
                                                                                                                   haemoglobin of 0.9%
                                                                                                                   (95% confidence interval
                                                                                                                   0.4 to 1.3). At 12 months
                                                                                                                   dietary advice plus
                                                                                                                   exercise was associated
                                                                                                                   with a statistically
                                                                                                                   significant mean
                                                                                                                   decrease of 1% (95% CI
                                                                                                                   0.4 to 1.5).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 7 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size             Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                   Comments
                 Level                 cs                                                      Measures
Brand-Miller, J.   Study Type:       14 studies     203 participants  Intervention: Low GI   Follow-up period: Low GI diets reduced        Choosing low GI foods in
                   Systematic        comprising 356 had type 1        diets                                    HbA1c by 0.43% points       place of conventional or
2003               review - meta-    subjects       diabetes, 153 had                        Outcome           (CI 0.72-0.13) over and     high GI foods has a
                   analysis                         type 2 diabetes.  Comparison: High       Measures:         above that produced by      clinically useful effect on
38
                                                                      GI diets               HbA1c             high GI diets. Taking       medium-term glycaemic
                   Evidence level:                  Country:                                 Fructosamine      both HbA1c and              control in patients with
                   1++                                                                                         fructosamine data           diabetes
                                                                                                               together and adjusting
                                                                                                               for baseline differences,
                                                                                                               glycated proteins were
                                                                                                               reduced 7.4% (8.8-6.0)
                                                                                                               more on the low GI diet
                                                                                                               than on the high GI diet.
                                                                                                               Systematically taking out
                                                                                                               each study from the
                                                                                                               meta analysis did not
                                                                                                               change CIs.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 8 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &            Effect Size            Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                            Comments
                 Level                 cs                                                Measures
Gillmer        Study Type:       15      Women routinely       Intervention: A 24 h    Follow-up period: Mean plasma glucose                          Insulin therapy more likely
MD;Maresh      RCT                       screened at 28-30     profile (hourly blood   Treatment         concentration of the                         achieved normoglycaemia
M;Beard                                  weeks’ gestation      sampling during         continued until   diet-treated women                           in these women. (very
RW;Elkeles     Evidence level:           and revealed          daytime and 2           delivery          significantly greater than                   small sample size)
RS;Alderson    1-                        abnormal glucose      hourly during night)                      those of the controls
C;Bloxham B;                             tolerance test with   performed during        Outcome           (pregnant women with a
                                         a total area under    which a diet of 2800    Measures: Mean normal glucose
1986                                     the GTT curve         kcal and 260 g          plasma glucose of tolerance who ate
                                         between 42 and        carbohydrate            women,            according to appetite) at
40
                                         65 mmol/l             provided.               Mean neonatal     10am, 2pm and 8pm as
                                                               Patients then           plasma glucose, compared to significantly
                                         Country: UK           randomly allocated      Mean plasma 3- lower in the insulin-
                                                               to treatment with       hydroxybutyrate treated group at 6pm,
                                                               diet alone or insulin   concentrations of 2am, 4am and 6am than
                                                               and dietary advice.     women.            the controls. Mean 2 h
                                                                                       Gestational age   neonatal plasma
                                                               Comparison:             at delivery, Mean glucose concentration of
                                                               Controls were           percentile infant the diet-treated group
                                                               normal women and        birthweight.      was significantly higher
                                                               no intervention                           than that of other
                                                               given.                                    groups. The neonatal
                                                                                                         skinfold thickness was
                                                                                                         similar in the groups.
                                                                                                         Diet therapy alone and
                                                                                                         insulin treatment
                                                                                                         lowered the plasma
                                                                                                         concentrations of 3-
                                                                                                         hydroxybutyrate.
                                                                                                         No significant difference
                                                                                                         between the groups in
                                                                                                         terms of gestational age
                                                                                                         at delivery or mean
                                                                                                         percentile infant
                                                                                                         birthweight.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 9 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &             Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                     Comments
                 Level                 cs                                                Measures
Ceysens G,     Study Type:       114     Pregnant women        Intervention: Studies   Follow-up period: No significant difference     There is insufficient
Rouiller D,    Systematic                with gestational      (RCTs) comparing                           between exercise and         evidence to recommend,
Boulvian M     review - meta-            diabetes              any type of exercise    Outcome            the other regimen in any     or advise against,
               analysis                  None included         programme with no       Measures:          of the outcomes              diabetic pregnant women
2006                                     pregnant women        exercise programme      Caesarean          evaluated.                   enrolling in exercise
               Evidence level:           with type 1 or        or other therapy.       section, perinatal The outcome 'use of          programmes. Trials, with
47
               1+                        type 2 diabetes.      Programnes              death (death       insulin therapy' was only    larger sample size,
                                         Women were            generally consisted     occurring during reported by Avery 1997         involving women with
                                         recruited during      of exercising three     pregnancy or the and Brankston 2004,            gestational diabetes, and
                                         the third trimester   times a week for 20     first six days of  where the difference         possibly type 1 and 2
                                         and the               to 45 minutes.          life), admission   between the two groups       diabetes, are needed to
                                         intervention was                              and length of stay was not significant RR       evaluate this intervention.
                                         performed for         Comparison:             in neonatal        0.98, 95% CI 0.51to
                                         about six weeks.                              intensive care     1.87). No woman
                                                                                       unit.              required insulin therapy
                                         Country:                                                         in the study by
                                                                                                          Jovanovic 1989. In Bung
                                                                                                          1991, exercise was
                                                                                                          compared to insulin
                                                                                                          treatment; therefore, this
                                                                                                          study was included in a
                                                                                                          separate analysis (diet +
                                                                                                          exercise versus diet +
                                                                                                          insulin).

                                                                                                          The occurrence of
                                                                                                          macrosomia was
                                                                                                          defined in two of the
                                                                                                          trials as birthweight >
                                                                                                          4000 g (Avery 1997;
                                                                                                          Bung 1991). No
                                                                                                          information was given in
                                                                                                          the third trial (Jovanovic
                                                                                                          1989).

                                                                                                          No significant difference
                                                                                                          between exercise and
                                                                                                          no exercise and
                                                                                                          between exercise and
                                                                                                          insulin in all the
                                                                                                          outcomes evaluated.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 10 of 441
                                                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                    Follow-up &             Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                         Outcome                                                                    Comments
                 Level                 cs                                                           Measures
Kieffer EC;Tabaei   Study Type:       1041 mother-   Exclusion criteria:    Intervention:         Follow-up period: 42% of women in this          There is an increased
BP;Carman           Cohort            infant pairs   Twin                   Maternal                                study entered                 risk of adverse maternal
WJ;Nolan                                             pregnancies,           sociodemographic,     Outcome           pregnancy overweight or       and infant outcomes
GH;Guzman           Evidence level:                  miscarriages,          prenatal care,        Measures: Infant obese; at least 36%            associated with
JR;Herman WH;       2+                               stillbirths, missing   anthropometric, and   birthweight       exceeded weight-gain          excessive maternal
                                                     records, women         metabolic                               recommendations.              weight, weight gain, and
2006 Dec                                             who entered care       characteristics                         Twenty-seven percent of       glucose intolerance
                                                     in the final weeks                                             the women had at least        among Latinas. This
43
                                                     of pregnancy and       Comparison:                             some degree of glucose        gives public health
                                                     women who had                                                  abnormality, including        professionals a unique
                                                     participated                                                   6.8% who had                  opportunity for prevention
                                                     during their                                                   gestational diabetes.         through prenatal and
                                                     previous                                                       Maternal multiparity,         postpartum interventions
                                                     pregnancy                                                      height, weight, weight
                                                                                                                    gain, and 1-hour
                                                     Country: USA                                                   screening glucose levels
                                                                                                                    were significant
                                                                                                                    independent predictors
                                                                                                                    of infant birthweight after
                                                                                                                    adjustment for
                                                                                                                    gestational age.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                    Page 11 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &             Effect Size             Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                  Comments
                 Level                 cs                                                Measures
Stotland         Study Type:       20,465   Women without      Intervention:           Follow-up period: Gestational weight gain       Gestational weight gain
NE;Cheng         Cohort                     diabetes, term,    Associations                                above the Institute of      above guidelines was
YW;Hopkins                                  singleton births   between gestational     Outcome             Medicine guidelines was     common and associated
LM;Caughey AB;   Evidence level:                               weight gain and         Measures:           43.3% vs 20.1% in           with multiple adverse
                 2-                         Country: USA       neonatal outcomes.      Neonatal            gestational weight gain     neonatal outcomes,
2006 Sep                                                       Gestational weight      outcomes: birth     below. Gestational          whereas gestational
                                                               gain categorised by     trauma, 5-min       weight gain above           weight gain below
44
                                                               the Institute of        Apgar score < 7, guidelines was                 guidelines was only
                                                               Medicine guidelines     assisted            associated with a low 5-    associated with SGA
                                                               as well as extremes     ventilation, SGA, minute Apgar score            status.
                                                               of gestational weight   LGA, cord arterial (AOR 1.33, 95% CI
                                                               gain (less than 7 kg    pH < 7.1, NICU      1.01-1.76), seizure
                                                               and more than 18        admission, SCN (AOR 6.50, 95% CI
                                                               kg)                     admission,          1.43-29.65),
                                                                                       neonatal            hypoglycemia (AOR
                                                               Comparison:             infection, seizure, 1.52, 95% CI 1.06-2.16),
                                                                                       hypoglycaemia,      polycythemia (AOR
                                                                                       polycythemia,       1.44, 95% CI 1.06-1.94),
                                                                                       Meconium            meconium aspiration
                                                                                       aspiration          syndrome (AOR 1.79,
                                                                                       syndrome,           95% CI 1.12-2.86), and
                                                                                       Respiratory         large for gestational age
                                                                                       distress            (AOR 1.98, 95% CI
                                                                                       syndrome,           1.74-2.25) compared
                                                                                       hospital stay > 5 with women within
                                                                                       days and hospital weight gain guidelines.
                                                                                       stay > 10 days.     Gestational weight gain
                                                                                                           below guidelines was
                                                                                                           associated with
                                                                                                           decreased odds of
                                                                                                           neonatal intensive care
                                                                                                           unit admission (AOR
                                                                                                           0.66, 95% CI 0.46-0.96)
                                                                                                           and increased odds of
                                                                                                           small for gestational age
                                                                                                           (SGA; AOR 1.66, 95%
                                                                                                           CI 1.44-1.92).
                                                                                                           Gestational weight gain
                                                                                                           less than 7 kg was
                                                                                                           associated with
                                                                                                           increased risk of




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 12 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                        Comments
                 Level                 cs                                   Measures
                                                                                         seizure, hospital stay
                                                                                         more than 5 days, and
                                                                                         SGA. Gestational weight
                                                                                         gain more than 18 kg
                                                                                         was associated with
                                                                                         assisted ventilation,
                                                                                         seizure, hypoglycemia,
                                                                                         polycythemia, meconium
                                                                                         aspiration syndrome,
                                                                                         and large for gestational
                                                                                         age.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 13 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                  Comments
                 Level                 cs                                                Measures
Ricart W;Lopez      Study Type:       9,270   Inclusion criteria:   Intervention:      Follow-up period: Both prepregnancy BMI       In terms of population
J;Mozas J;Pericot   Cohort                    Spanish pregnant      Prepregnancy BMI                     and abnormal glucose        impact, prepregnancy
A;Sancho                                      women with            and glucose        Outcome           tolerance categories        maternal BMI exhibits a
MA;Gonzalez         Evidence level:           singleton             tolerance status   Measures:         were independent            much stronger influence
N;Balsells M;Luna   2++                       pregnancies and                          Primary outcome: predictors of pregnancy      than abnormal blood
R;Cortazar                                    without a former      Comparison:        fetal macrosomia, outcomes.                   glucose tolerance on
A;Navarro                                     diagnosis of                             Caesarean         The upper quartile of       macrosomia, Caesarean
P;Ramirez                                     diabetes                                 section           BMI accounted for 23%       section, pregnancy-
O;Flandez                                                                              Secondary         of macrosomia, 9.4% of      induced hypertension
B;Pallardo                                    Exclusion criteria:                      outcome:          Caesarean section, 50%      and large-for-gestational-
LF;Hernandez-                                 Pregnancies                              diabetes-related of pregnancy-induced         age newborns.
Mijas A;Ampudia                               with preterm                             pregnancy         hypertension and 17.6%
J;Fernandez-Real                              delivery (at less                        outcomes.         of large-for-gestational-
JM;Corcoy                                     than 28 weeks)                                             age newborns. In
R;Spanish Group                               and the                                                    contrast, NDDG GDM
for the Study of                              second                                                     accounted
the Impact of                                 pregnancy of                                               for 3.8% of macrosomia,
Carpenter and                                 women with two                                             9.1% of pregnancy-
Coustan GDM                                   pregnancies in                                             induced
Thresholds.;                                  the                                                        hypertension and 3.4%
                                              same year.                                                 of preterm births.
2005 Sep
                                              Country: Spain
45




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 14 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Q4 - What are the target ranges for blood glucose in the preconception period?

Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &             Effect Size            Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                             Comments
                 Level                 cs                                                      Measures
Diabetes and    Study Type:       435                Preconception       Intervention:       Follow-up period: 166 infants were                              Participating centres are
Pregnancy Group Cohort            pregnancies,       care was            Prepregnancy                             delivered before 37                        tertiary units recruiting
F;                                289 in women       provided in 48.5%   glycaemic control   Outcome              weeks. Multivariate                        high risk pregnancies.
                Evidence level:   with type 1        women with type                         Measures:            analysis found first
2003            2++               diabetes and       1 diabetes and      Comparison:         Perinatal mortality trimester HbA1c >8% to
                                  146 from           24% of women                            (Fetal death +       be associated with
54
                                  women with         with type 2                             Neonatal death) preterm delivery (OR
                                  type 2 diabetes.   diabetes.                               Major congenital 2.2; 95%CI 1.4-
                                                     Preconception                           malformations        3.7;p=0.002).
                                                     capillary blood                         Preterm delivery
                                                     glucose targets                         Macrosomia           First trimester
                                                     were <5.3mmol/l                         Mode of delivery HbA1cwas >8% in 120
                                                     before meals and                        Neonatal             women (27.6%). Women
                                                     <6.7 mmol/l 2 h                         complications        whose HbA1c was >8%
                                                     postprandial                                                 had higher rates of
                                                                                             HbA1c (normal        perinatal mortality (9.2
                                                                                             4.9±0.6%)            vs 2.5%; OR 3.9; 95%
                                                                                             obtained during      CI 1.5-9.7; p<0.005)
                                                                                             the first trimester. major congenital
                                                                                             Actual values        malformations (8.3 vs.
                                                                                             were not available 2.5%; OR 3.5; 95%CI
                                                                                             as only              1.3-8.9;p<0.01) and
                                                                                             values>8% were preterm delivery (57.6%
                                                                                             recorded.            vs 24.8%; OR 1.4; 1.1-
                                                                                             Therefore values< 1.7; P<0.005).
                                                                                             8% was
                                                                                             considered as a      First trimester HbA1c
                                                                                             surrogate marker was more frequently
                                                                                             for effective        >8% in women who did
                                                                                             prepregnancy         not receive
                                                                                             control.             preconception care than
                                                                                                                  in those who did (43.5
                                                                                                                  vs 4.0%; OR 18.5; 8.3-
                                                                                                                  40.9; p<0.001).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 15 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                  Comments
                 Level                 cs                                                      Measures
Dicker D;Feldberg   Study Type:       94 women. 59    Type 1 diabetes   Intervention:         Follow-up period: Preconception group:       Miscarriage rates among
D;Samuel            Cohort            attending                         Preconception care.                     Normoglycaemia and         patients with wit diabetes
N;Yeshaya                             preconception   Pregnancies       Glyceamic control     Outcome           normal HbA1c levels        not seen before
A;Karp              Evidence level:   clinic, 35      resulting in      was obtained by       Measures:         were achieved in this      pregnancy were
M;Goldman JA;       2++               nonattenders.   congenital        intensified insulin   Miscarriages      group before conception    significantly higher when
                                                      malformation      therapy and           (losses that      and were maintained        compared with women
1988 May                                              were excluded     monitored by blood    occurred before   throughout pregnancy.      who attended the clinic ,
                                                      from the study    glucose self          22 weeks)                                    whose rates represented
63
                                                      population.       monitoring.                             Miscarriage occurred in    the normal frequencies in
                                                                                              HbA1(normal       5/59 women attending       the general population.
                                                                        Comparison: 35        range 6.0% to     pre-conception care and
                                                                        pregnant women        7.5%)             in 10/35 not attending.    Poor metabolic control
                                                                        who did not receive                     (p<0.001)                  around conception and in
                                                                        preconception                                                      the early weeks of
                                                                        glycaemic control                       HbA1c Initial visit:       pregnancy may be the
                                                                                                                Preconception group:       determining factor
                                                                                                                Pregnancy>22 weeks:        favouring abortion.
                                                                                                                10.70 ± 0.49
                                                                                                                Miscarriage: 10.68 ±
                                                                                                                0.49

                                                                                                                HbA1c at conception
                                                                                                                Preconception group:
                                                                                                                Pregnancy>22 wk: 7.54
                                                                                                                ± 0.34
                                                                                                                Miscarriage: 7.5 ± 0.34

                                                                                                                First trimester HbA1c:
                                                                                                                Preconception group:
                                                                                                                Pregnancy>22wks: 7.4 ±
                                                                                                                0.34
                                                                                                                Miscarriage: 7.39 ± 0.34
                                                                                                                Later group:
                                                                                                                Pregnancy >22 wks
                                                                                                                10.42 ± 0.47
                                                                                                                Miscarriage: 10.68 ±
                                                                                                                0.49, p<0.001.

                                                                                                                Second trimester HbA1c
                                                                                                                Preconception group:
                                                                                                                Pregnancy>22 wks:
                                                                                                                7.20 ± 0.33




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 16 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                        Comments
                 Level                 cs                                   Measures
                                                                                         Miscarriage:
                                                                                         7.21 ± 0.33
                                                                                         Later group:
                                                                                         Pregnancy>22 wks:
                                                                                         8.14 ± 0.37
                                                                                         Miscarriage: 8.20 ± 0.38.
                                                                                         P<0.50
                                                                                         Blood glucose level
                                                                                         (mmol/l) initial visit:
                                                                                         Preconception group:
                                                                                         Pregnancy>22 wks: 9.7
                                                                                         ± 0.6
                                                                                         Miscarriage: 9.6 ± 0.4

                                                                                         Blood glucose level at
                                                                                         conception:
                                                                                         Preconception group:
                                                                                         Pregnancy>22 wks: 6.4
                                                                                         ± 0.4
                                                                                          Miscarriage: 6.5 ± 0.4

                                                                                         First trimester blood
                                                                                         glucose control:
                                                                                         Preconception group:
                                                                                         Pregnancy>22 wks: 6.05
                                                                                         ± 0.4
                                                                                         Miscarriage: 6.05 ± 0.4
                                                                                         Later group:
                                                                                         Pregnancy>22 wks:
                                                                                         9.02 ± 0.4
                                                                                         Miscarriage: 9.57 ± 0.6.
                                                                                         P<0.001

                                                                                         Second trimester blood
                                                                                         glucose control
                                                                                         Preconception group:
                                                                                         Pregnancy>22 wks: 6.05
                                                                                         ± 0.4
                                                                                         Miscarriage: 6.05 ± 0.4
                                                                                         Later group:
                                                                                         Pregnancy>22 wks: 7.7
                                                                                         ± 0.5




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 17 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         Miscarriage: 8.14 ± 0.5.
                                                                                         P<0.50




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 18 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                  Follow-up &           Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                       Outcome                                                                 Comments
                 Level                 cs                                                         Measures
Fuhrmann        Study Type:       200 pregnant       Pregnant women        Intervention: Pre-   Follow-up period: 39/56 women in             The findings stress the
K;Reiher        Cohort            women, 56          with type 1           conception care                        preconception group        importance of a
H;Semmler                         received           diabetes                                   Outcome           had 87% of readings        reasonably strict
K;Glockner E;   Evidence level:   preconception      In all patients the   Comparison:          Measures:         between 2.3 and 7.7        metabolic control, started
                2++               care,144           aim was to keep                            Glycaemic         mmol/l, 9 patients had     well before conception, to
1984 Apr                          referred after 8   blood glucose                              control: based on 77% of readings in this    prevent excess rates of
                                  weeks              levels between                             24 hr capillary   range, 8 patients failed   congenital malformation.
52
                                  gestation          3.3 and 6.6                                blood glucose     to achieve sufficient
                                                     mmol/l                                     profiles derived  metabolic control.
                                                                                                from 9 samples
                                                                                                which were taken 1/56 (1.8%) of infants of
                                                                                                1 to 3 times a    women in preconception
                                                                                                week. The         care group had a
                                                                                                therapeutic goal congenital malformation.
                                                                                                was to keep       This occurred in one of
                                                                                                values between    the eight mothers who
                                                                                                2.3 and 7.7       failed to achieve
                                                                                                mmol/l. Values in sufficient metabolic
                                                                                                this range, but   control. 9/136 (6.6%)
                                                                                                below or above    infants in the late
                                                                                                normal, were only treatment group had
                                                                                                considered for    congenital
                                                                                                correction if     malformations.
                                                                                                repeatedly
                                                                                                determined at the When data pooled with
                                                                                                same time of day. earlier findings (from
                                                                                                Deviations        1977-1981) 2/185
                                                                                                outside the range (1.1%) infants had
                                                                                                were corrected    malformations in
                                                                                                immediately by    preconception care
                                                                                                changing insulin group compared to
                                                                                                dose and/or diet 31/473 (6.6%) in late
                                                                                                                  treatment group p<0.01.
                                                                                                Congenital
                                                                                                malformation




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 19 of 441
                                                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &            Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                                  Comments
                 Level                 cs                                                        Measures
Fuhrmann K;    Study Type:       620 women         In all patients the   Intervention:         Follow-up period: In the preconception         Strict metabolic control
               Cohort            with type         aim was to            Preconception                           group mean daily blood       established before
1986                             1diabetes         achieve a blood       metabolic control     Outcome           glucose was <6.1 mmol/l      conception can reduce
               Evidence level:   receiving         glucose                                     Measures:         in 88.3% of subjects         the rate of malformations
53
               2++               education and     concentration         Comparison:           Congenital        compared with only           in infants of diabetic
                                 intensive         comparable to         Metabolic control     malformations     20.7% of the later           mothers.
                                 treatment. 184    that found in non     optimised after 8th                     registrants (χ2 = 110.6;
                                 recruited prior   diabetic pregnant     week gestation.       Normalboold       p<0.001).                    Normal blood glucose
                                 to conception     women                                       glucose readings                               levels can be achieved in
                                 436 recruited 8                                               (2.5-7.8 mmol/l)  Among the 622 infants        about 85% of all diabetic
                                 weeks after                                                                     born to 620 diabetic         women if they attend a
                                 conception.                                                                     patients 33 had              preconception
                                                                                                                 malformations (5.3%).        programme.
                                                                                                                 31/437 women who
                                                                                                                 began intensive
                                                                                                                 metabolic control after
                                                                                                                 the 8th week of
                                                                                                                 gestation gave birth to
                                                                                                                 an infant with a
                                                                                                                 malformation (7.1%)
                                                                                                                 compared to 2/185
                                                                                                                 (1.1%) infants born to
                                                                                                                 mothers who optimised
                                                                                                                 metabolic control prior to
                                                                                                                 conception, p<0.01.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                Page 20 of 441
                                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &            Effect Size             Study Summary                   Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                                      Comments
                 Level                 cs                                                     Measures
Gold AE;Reilly      Study Type:       57 deliveries   Type 1 diabetes   Intervention:       Follow-up period: Two groups of               Glycaemic control           A failure to find an
R;Little J;Walker   Cohort                                              Glyceamic control                     approximately equal         assessed by glycosylated    association between third
JD;                                                   Only viable       preconception and   Outcome           size: Group 1: Birth        haemoglobin at the time     trimester glycaemic
                    Evidence level:                   pregnancies were in early pregnancy   Measures: Birth weight z score < 1 SD         of conception and in the    control and birthweight
1998                2+                                included as major                     weight            above nondiabetic mean      early weeks of pregnancy    may be due to the fact
                                                      outcome variable Comparison:                            Group 2: Birth weight z     is a more powerful          that all of the women
67
                                                      of interest was                       HbA1              score >1 SD above           predictor of birth weight   achieved glycaemic
                                                      birth weight                                            nondiabetic mean            than is glycaemic control   control between a narrow
                                                                                            (In Edinburgh the The study had a >80%        measured during the         range
                                                                                            nondiabetic       chance of detecting a       later weeks of
                                                                                            nonpregnant       large difference (0.8 SD)   pregnancy.                 Glycaemic control during
                                                                                            normal range is   between the groups                                     different periods are
                                                                                            5.2-6.8% and the (p=0.05)                     A multiple r value of 0.48 interrelated
                                                                                            normal range for                              suggests that glycaemic
                                                                                            pregnancy is 4.8 HbA1 6-12 mths               control as measured by
                                                                                            to 6.4%)          prepregnancy:               HbA1 only contributes
                                                                                                              Group 1(n=21): 8.6 ± 1.4    23% of the variance in
                                                                                                              Group 2 (n=25):             birth weight.
                                                                                                              10.0 ± 2.3, p=0.02

                                                                                                              HbA1 0-6 mths
                                                                                                              prepregnancy:
                                                                                                              Group 1(n=23):8.7 ± 2.0
                                                                                                              Group 2(n=26): 10.2 ±
                                                                                                              2.4, p=0.03

                                                                                                              HbA1 at booking:
                                                                                                              Group 1(n=25): 8.4 ± 1.6
                                                                                                              Group 2(n=26): 9.5 ±
                                                                                                              2.2, p=0.04

                                                                                                              HbA1 0-12 weeks:
                                                                                                              Group 1 (n=26): 8.0 ±
                                                                                                              1.3
                                                                                                              Group 2 (n=31): 9.5 ±
                                                                                                              2.2, p = 0.04

                                                                                                              Hb1 12-24 weeks:
                                                                                                              Group 1 (n=26) : 7.1±
                                                                                                              1.0,
                                                                                                              Group 2 (n=31): 7.4 ±
                                                                                                              0.9 (NS)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 21 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         HbA1 24 weeks-term
                                                                                         Group 1 (n=26): 7.4 ±
                                                                                         0.8
                                                                                         Group 2 (n=31): 7.2 ±
                                                                                         1.1 (NS)

                                                                                         Significant correlation
                                                                                         between HbA1c during 6
                                                                                         mths before pregnancy
                                                                                         (r=0.44, p=0.002) at
                                                                                         booking (r=0.43,
                                                                                         p=0.001), the first 12
                                                                                         weeks of pregnancy
                                                                                         (r=0.48, p=0.001) weeks
                                                                                         12-4 of pregnancy
                                                                                         (r=045, p=0.001) and
                                                                                         weeks 24 to term
                                                                                         (r=0.34, p=0.009) and z
                                                                                         score of the birth weight.

                                                                                         Regression analysis
                                                                                         only HbA1 at 0-12
                                                                                         weeks entered in the
                                                                                         equation (multiple r-
                                                                                         0.48, sig F = 00005).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 22 of 441
                                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                            Follow-up &           Effect Size                Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                 Outcome                                                                     Comments
                 Level                 cs                                                   Measures
Goldman           Study Type:       75 pregnant      Type1 diabetes   Intervention:        Follow-up period: The mean HbA1c was           We confirm the evidence
JA;Dicker         Cohort            women with                        Preconception care                     10.68% ± 0.49% and           accumulated in the
D;Feldberg                          type 1 diabetes.                                       Outcome           the mean blood glucose       recent literature that
D;Yeshaya         Evidence level:   44 women                          Comparison: 31       Measures:         level was 174±10.8           congenital malformations
A;Samuel N;Karp   2++               attending                         Nonattenders of the HbA1c (normal      mg/100ml in patients         in pregnancy complicated
M;                                  preconception                     preconception clinic mean (SD) 6.6     attending the                by diabetes may be
                                    clinic, 31                                             (0.3))            preconception clinic.        linked to disturbances in
1986 Aug                            nonattenders of                                                                                       maternal metabolism
                                    the                                                    Malformation     Women attending               during the period of
50
                                    preconception                                                           preconception clinic:         embryogenesis.
                                    clinic                                                                  HbA1c:                        Consequently we concur
                                                                                                            Initial visit: 10.68 ± 0.49   with the recommendation
                                                                                                            At conception: 7.56 ±         that tight diabetic control
                                                                                                            0.34                          is required before the
                                                                                                            First trimester: 7.39 ±       patient attempts to
                                                                                                            0.34                          conceive.
                                                                                                            Second trimester 7.21±
                                                                                                            0.33                          Our study reveals a
                                                                                                            Third trimester 7.14 ±        relation between
                                                                                                            0.32                          preconception blood
                                                                                                            Blood glucose level           glucose regulation and
                                                                                                            (mmol/l):                     the degree of improved
                                                                                                            Initial visit: 9.6 ± 0.6      diabetic control.
                                                                                                            At conception: 6.4 ± 0.4
                                                                                                            First trimester: 6 ± 0.4
                                                                                                            Second trimester: 6± 0.4
                                                                                                            Third trimester: 5.3 ± 0.3
                                                                                                            Insulin dose:
                                                                                                            Initial visit: 0.60 ± 0.03
                                                                                                            At conception: 0.62 ±
                                                                                                            0.03
                                                                                                            First trimester: 0.71 ±
                                                                                                            0.04
                                                                                                            Second trimester: 0.80 ±
                                                                                                            0.04
                                                                                                            Third trimester: 0.95 ±
                                                                                                            0.05

                                                                                                            Nonattenders:
                                                                                                            HbA1c:
                                                                                                            First trimester: 10.42 ±
                                                                                                            0.47,p<0.001




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 23 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         Second trimester:8.13 ±
                                                                                         0.37, p<0.05
                                                                                         Third trimester: 7.44 ±
                                                                                         0.3, NS
                                                                                         Blood glucose level
                                                                                         (mmol/l):
                                                                                         First trimester: 163 ±
                                                                                         10.2,p<0.001
                                                                                         Second trimester: 148 ±
                                                                                         9.2,p<0.05
                                                                                         Third trimester: 124 ±
                                                                                         7.8, NS
                                                                                         Insulin dose:
                                                                                         First trimester: 0.83 ±
                                                                                         0.04, p<0.001
                                                                                         Second trimester: 0.92 ±
                                                                                         0.05, p<0.001
                                                                                         Third trimester: 0.96 ±
                                                                                         0.05, NS

                                                                                         3 congenital
                                                                                         malformations occurred,
                                                                                         all in the nonattending
                                                                                         group (incidence in this
                                                                                         group 9.6%, p<0.07)).
                                                                                         First trimester HbA1c
                                                                                         values of mothers were
                                                                                         8.8%, 9.2% and 10.2%.

                                                                                         Other outcomes
                                                                                         Preconception group:
                                                                                         CS: 22.7%
                                                                                         Preeclampsia: 9.1%
                                                                                         Gestational age
                                                                                         (mean±SD) (wk):
                                                                                         39.3±0.5
                                                                                         Birthweight (mean ±SD)
                                                                                         (gm): 3296.2 ± 234.1
                                                                                         1 minute Apgar score
                                                                                         (mean ± SD): 8.1±0.7
                                                                                         Hypoglycaemia: (<1.65
                                                                                         mmol/l) (%) 11.4




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 24 of 441
                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &       Effect Size          Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                     Comments
                 Level                 cs                                   Measures
                                                                                         Hypocalcemia (<0.44
                                                                                         mmol/l) (%) 9.6
                                                                                         Hyperbilirubinemia
                                                                                         (>0.825 mmol/l) (%)
                                                                                         22.7
                                                                                         Respiratory distress
                                                                                         syndrome (%): 2.3

                                                                                         Nonattending group
                                                                                         CS (%): 41.9 (SS not
                                                                                         reported)
                                                                                         Preeclampsia (%): 22.5
                                                                                         (SS not reported)
                                                                                         Gestational age
                                                                                         (mean±SD) (wk): 36.9 ±
                                                                                         0.5, p<0.05
                                                                                         Birthweight (mean ±SD)
                                                                                         (gm): 3681.4 ± 261.5,
                                                                                         p<0.05
                                                                                         1 minute Apgar score
                                                                                         (mean ± SD): p<0.001
                                                                                         Hypoglycaemia: (<1.65
                                                                                         mmol/l) (%)25.8, NS
                                                                                         Hypocalcemia (<0.44
                                                                                         mmol/l) (%): 9.6, NS
                                                                                         Hyperbilirubinemia
                                                                                         (>0.825 mmol/l) (%):
                                                                                         32.2, NS
                                                                                         Respiratory distress
                                                                                         syndrome (%): 12.9, NS




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 25 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &            Effect Size              Study Summary                      Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                                          Comments
                 Level                 cs                                                     Measures
Greene MF;Hare    Study Type:       303 pregnant   Type 1 diabetes.   Intervention: First   Follow-up period: 35% of women had first       Although the risks for        Ten patients lost to follow
JW;Cloherty       Cohort            women          All patients       trimester gylcaemic                     trimester HbA1 >9SD          both adverse outcomes         up (mean HbA1 12.0%).
JP;Benacerraf                                      presenting at 12   control               Outcome           above the mean               were markedly elevated
BR;Soeldner JS;   Evidence level:                  weeks gestation                          Measures: First (>11.0%)                       following a first trimester   Rate of Miscarriages may
                  2++                              or less            Comparison:           trimester HbA1                                 in very poor metabolic        be higher
1989 Mar                                                                                    (In nondiabetic   There were 20 major          control, there was a
                                                                                            population mean malformations.                 broad range of control
55
                                                                                            (SD) is 5.9%                                   over which the risks were
                                                                                            (0.57)).          Miscarriages are             not substantially
                                                                                                              significantly increased at   elevated. To keep
                                                                                            Miscarriage       >9SD above the mean          malformations and
                                                                                                              (HbA1 >11.0%) the risk       Miscarriages to a
                                                                                            Congenital        of major malformations       minimum among diabetic
                                                                                            malformations     is significantly elevated    women does not require
                                                                                                              at >12 SD above the          ‘excellent’ control; there
                                                                                                              mean (HbA1 >12.7%)           seems to be a fairly
                                                                                                                                           broad range of
                                                                                                               Miscarriage by first        ‘acceptable’ control.
                                                                                                               trimester HbA1:
                                                                                                               ≤9.3%: 14/113 (12.4%)
                                                                                                               RR=1

                                                                                                               9.4-11.0%: 7/85 (8.3%)
                                                                                                               RR=0.7 (95% CI =0.3-
                                                                                                               1.6)

                                                                                                               11.1-12.7%: 15/61
                                                                                                               (24.6%) RR=1.98 (95%
                                                                                                               CI = 1.03-3.38)

                                                                                                               12.8-14.4%: 10/28
                                                                                                               (35.7%) RR=2.9 (1.4-
                                                                                                               5.8)

                                                                                                               >14.4: 6/16 (37.5%)
                                                                                                               RR=3 (95% CI=1.3-7.0)

                                                                                                               Major malformations by
                                                                                                               first trimester HbA1:
                                                                                                               ≤9.3% 3/99 (3%) RR=1

                                                                                                               9.4-11.0%: 4/77 (5.2%)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 26 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         RR=1.7 (95% CI=0.4-
                                                                                         1.7)

                                                                                         11.1-12.7%: 2/46 (4.3%)
                                                                                         RR=1.4 (95% CI 0.3-
                                                                                         8.3)

                                                                                         12.8-14.4%: 7/18
                                                                                         (38.9%) RR 12.8 (95%
                                                                                         CI 4.7-35.0)

                                                                                         >14.4 %: 4/10 (40%) RR
                                                                                         13.2 (95% CI 4.3-40.4)

                                                                                         First trimester HbA1:
                                                                                         Miscarriage (n=52): 11.6
                                                                                         ± 2.5
                                                                                         Major malformation
                                                                                         (n=20):12.4±2.9
                                                                                         No malformation
                                                                                         (n=230): 9.9±1.9,
                                                                                         p<0.005.

                                                                                         Gestational age (wk) at
                                                                                         first visit
                                                                                         Miscarriage: 8.2±2.0
                                                                                         Major malformation:
                                                                                         8.6±2.1
                                                                                         No malformation:
                                                                                         7.5±1.9, p=0.01

                                                                                         Years of diabetes prior
                                                                                         to pregnancy:
                                                                                         Miscarriage: 12.0±7.7
                                                                                         Major malformation: 8.9
                                                                                         ± 5.1
                                                                                         No malformation: 13.4 ±
                                                                                         7.0, p=0.025




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 27 of 441
                                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                       Follow-up &           Effect Size             Study Summary             Reviewer
 Information & Evidence   Patients Characteristi Comparison                            Outcome                                                               Comments
                 Level                 cs                                              Measures
Jensen           Study Type:       1218             Type 1diabetes   Intervention:   Follow-up period: HbA1c 0-3 mths prior to    Poor glucose control
DM;Damm          Cohort            pregnancies in                    Preconception                     conception (median and     before and during
P;Moelsted-                        990 women                         HbA1c           Outcome           interquartile range):      pregnancy is associated
Pedersen         Evidence level:                                                     Measures:         Adverse outcome: 8.0       with perinatal morality
L;Ovesen         2++                                                 Comparison:     Serious adverse (7.3-9.1)                    and congenital
P;Westergaard                                                                        outcome           Others: 7.6 (6.8-8.5),     malformations
JG;Moeller                                                                           (perinatal death  p=0.005.
M;Beck-Nielsen                                                                       and/or congenital
H;                                                                                   malformation)     First trimester HbA1c:
                                                                                                       Adverse outcome:7.6
2004                                                                                                   (6.6-8.6)
                                                                                                       Others 7.3 (6.6-8.1),
66
                                                                                                       p=0.037

                                                                                                       Second trimester
                                                                                                       HbA1c:
                                                                                                       Adverse outcome: 6.9
                                                                                                       (6.2-8.0)
                                                                                                       Others: 6.6 (6.0-7.3),
                                                                                                       p=0.012

                                                                                                       Third trimester HbA1c:
                                                                                                       Adverse outcome: 7.1
                                                                                                       (6.5-7.9)
                                                                                                       Others: 6.7 (6.2-7.4)
                                                                                                       p<0.001

                                                                                                       Preconception
                                                                                                       guidance:
                                                                                                       Adverse outcome: 38
                                                                                                       (42%)
                                                                                                       Others: 631 (59.2%)
                                                                                                       p=0.002

                                                                                                       Daily glucose monitoring
                                                                                                       at conception
                                                                                                       Adverse outcomes: 18
                                                                                                       (22.5%)
                                                                                                       Others: 363 (34.6%)
                                                                                                       p=0.019




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 28 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &             Effect Size            Study Summary            Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                               Comments
                 Level                 cs                                                    Measures
Key TC;Giuffrida   Study Type:       83 pregnant   Women with          Intervention: HbA1c Follow-up period:   Pregnancy outcomes in   Abnormal glucose control
R;Moore TR;        Cohort            women         diabetes (63        (normal nondiabetic                     type I and type II      during early pregnancy,
                                                   types I and 20      mean 5.1 (1.1)      Outcome             diabetes were similar   as reflected by
1987 May           Evidence level:                 type II) who came                       Measures:           (χ2=2.8, p>0.05). 25/63 abnormalHbA1c, is
                   2++                             for care before     Comparison:         Congenital          (39%) of women with     associted with an
59
                                                   the 15th week of                        malformations       type 1 diabetes had an  increased incidence of
                                                   gestation                               Miscarriages        abnormal outcome and    congenital malformations
                                                                                                               6/20 (30%) of women     and Miscarriage. The
                                                                                                                                       type of diabetes does not
                                                                                                               with type II diabetes had
                                                                                                               an abnormal outcome.    appear to be important in
                                                                                                                                       that type 1and type 2 had
                                                                                                               There were 31/83 (37%) similar pregnancy
                                                                                                               abnormal outcomes. 22 outcomes.
                                                                                                               pregnancies terminated
                                                                                                               as Miscarriages and     The present study
                                                                                                               there were 9 congenital establishes that fair to
                                                                                                               malformations. The      good diabetic control
                                                                                                               congenital malformation (HbA1c<9.5%) during
                                                                                                               rate in continuing      early gestation is highly
                                                                                                               pregnancies was 15%; predictive of normal
                                                                                                               the overall Miscarriage pregnancy outcome.
                                                                                                               rate was 26.5%.

                                                                                                               Malformations:
                                                                                                               82% of patients in the
                                                                                                               study has HbA1c levels
                                                                                                               ≥9.5%, 42% had values
                                                                                                               ≥11.5%. All anomalies
                                                                                                               were seen in patients
                                                                                                               whose initial HbA1c was
                                                                                                               >9.5%. 7/9 were seen in
                                                                                                               mothers whose HbA1c
                                                                                                               was>11.5% and 6/9 in
                                                                                                               the 20% of the study
                                                                                                               population whose
                                                                                                               HbA1c was >13.5%.

                                                                                                               Miscarriages:
                                                                                                               No Miscarriages
                                                                                                               occurred in mothers with
                                                                                                               HbA1c <7.5%, 1/22
                                                                                                               occurred in mothers




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 29 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         whose HbA1c was
                                                                                         7.5%-9.4%. 21/22 (95%)
                                                                                         occurred in mothers
                                                                                         whose HbA1c was
                                                                                         >11.5%, 19/22 (86%)
                                                                                         occurred in mothers
                                                                                         whose HbA1c >13.5%.

                                                                                         Higher HbA1clevels
                                                                                         predict abnormal
                                                                                         outcomes and low
                                                                                         HbA1c values predict
                                                                                         normal outcomes
                                                                                         (p<0.005).
                                                                                         Predictive values:
                                                                                         >7.5%: Sensitivity
                                                                                         100%,Specificity 14%,
                                                                                         PPV 0.41, NPV 1

                                                                                         >9.5%: Sensitivity
                                                                                         97%,Specificity 42%,
                                                                                         PPV 0.50, NPV 0.96

                                                                                         >11.5%: Sensitivity
                                                                                         84%, Specificity
                                                                                         82%,PPV0.74, NPV
                                                                                         0.89

                                                                                         >13.5%: Sensitivity
                                                                                         55%, Specificity
                                                                                         94%,PPV 0.85, NPV
                                                                                         0.77.

                                                                                         When HbA1c exceeded
                                                                                         15.5% (n=6) all
                                                                                         pregnancies resulted in
                                                                                         an abnormal outcome, a
                                                                                         predictive value of 1.00




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 30 of 441
                                                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &              Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                                    Comments
                 Level                 cs                                                        Measures
Kitzmiller        Study Type:       84 women           Women with          Intervention:        Follow-up period:   One major congenital        Education and intensive
JL;Gavin LA;Gin   Cohort            recruited prior    diabetes            Recruitment prior to                     malformation occurred in    management for
GD;Jovanovic-                       to conception,     undergoing          conception           Outcome             84 infants (1.2%) of        glycaemic control of
Peterson L;Main   Evidence level:   110 women          education and                            Measures:           women treated before        diabetic women before
EK;Zigrang WD;    2++               referred at 6 to   intensive           Comparison:          Congenital          conception compared         and during early
                                    8 weeks            management.         Recruitment at 6 to malformations        with 12 anomalies in        pregnancy will prvent
1991 Feb 13                         gestation          Type 1 diabetes     8 weeks conception                       110 infants (10.9%) of      excess ratesof congenital
                                                       present in 68% of                        HbA1c               mothers in the post         malformations.
36
                                                       women in                                 (normalnondiabeti   conception group.
                                                       preconception                            c mean (SD) 6.3     (P=0.01).
                                                       group and 60% in                         (0.7)               Χ2=40.9, p<0.0001 for
                                                       postconception                                               elevated initial HbA1c
                                                       group.                                                       levels in the
                                                                                                                    postconception
                                                                                                                    compared with the
                                                                                                                    preconception group
                                                                                                                    χ2=10.1, p=0.04 for
                                                                                                                    association of congenital
                                                                                                                    anomalies with elevated
                                                                                                                    HbA1c level in the
                                                                                                                    postconception group
                                                                                                                    only
                                                                                                                    χ2=19.5, p=0.001 for
                                                                                                                    association of congenital
                                                                                                                    anomalies with elevated
                                                                                                                    HbA1c level in both
                                                                                                                    groups

                                                                                                                    Of the 12 mothers who
                                                                                                                    had infants with
                                                                                                                    congenital
                                                                                                                    malformations 7 had
                                                                                                                    HbA1c >10.5% at first
                                                                                                                    prenatal visit and 4 had
                                                                                                                    levels >9.2%.

                                                                                                                    Frequency of
                                                                                                                    hypoglycemia in
                                                                                                                    preconception group
                                                                                                                    (No of episodes per
                                                                                                                    week):
                                                                                                                    0, 42%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 31 of 441
                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size        Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                    Comments
                 Level                 cs                                   Measures
                                                                                         1-2, 35%
                                                                                         3-4, 14%
                                                                                         5-7, 6%
                                                                                         ≥8, 3%
                                                                                         Requiring third party
                                                                                         intervention n=2




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                              Page 32 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &           Effect Size          Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                             Comments
                 Level                 cs                                                     Measures
Miodovnik        Study Type:       116              Type 1 diabetes.   Intervention: First   Follow-up period: Two groups based on  Women with type 1
M;Skillman       Cohort            pregnancies in                      trimester HbA1                          method of HbA1       diabetes who
C;Holroyde                         75 women         Goal of glycaemic                        Outcome           assessment.          experienced
JC;Butler        Evidence level:                    control were a    Comparison:            Measures:         Group 1 (1981-1984)  Miscarriages had
JB;Wendel        2++                                fasting blood                            Miscarriage       n=79:                significantly higher levels
JS;Siddiqi TA;                                      glucose <5.5                             (between 5 and    Miscarriages:        of HbA1 in the first
                                                    mmol/l and a 90                          15 weeks)         HbA1 <12%: 11/68     trimester than did women
1985 Oct 15                                         minute                                                     (16%)                whose pregnancies
                                                    postprandial                                               HbA1 ≥ 12%: 5/11     progressed beyond 20
65
                                                    glucose of less                                            (45%) p<0.05         weeks. This indicates an
                                                    than 7.7 mmol/l                                                                 increased risk of fetal
                                                                                                              Group 2 (1978-1980)   loss in insulin diabetic
                                                    Pregnancies that                                          n=37:                 mothers with poor control
                                                    resulted in                                               Miscarriages:         of glycaemia in the first
                                                    congenital                                                HbA1 <12%: 3/21 (14%) trimester. To reduce the
                                                    malformations                                             HbA1 ≥12%: 7/9 (44%) frequency of
                                                    were excluded                                                                   Miscarriages in women
                                                    from the study                                                                  with type 1 diabetes good
                                                                                                                                    metabolic control is
                                                                                                                                    essential prior to
                                                                                                                                    conception and in the
                                                                                                                                    early weeks of
                                                                                                                                    pregnancy.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 33 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                            Follow-up &             Effect Size           Study Summary                   Reviewer
 Information & Evidence   Patients Characteristi Comparison                                 Outcome                                                                     Comments
                 Level                 cs                                                   Measures
Miodovnik       Study Type:       84 pregnancies Type 1 diabetes     Intervention:        Follow-up period:   Initial HbA1:           HbA1 is higher in diabetic   Only pregnant women
M;Mimouni       Cohort            in 68 women                        Glycaemic control in                     Miscarriage (n=18):     pregnancies leading to a     were enrolled which may
F;Tsang                                          Goal of glycaemic   early pregnancy      Outcome             12.0± 0.6               spontaneous aborton          underestimate abortion
RC;Ammar        Evidence level:                  control were a                           Measures: HbA1      Pregnanciy >20 weeks    compared with those          rate
E;Kaplan        2++                              fasting blood       Comparison:          at 8-9 wks          (n=66): 10.7± 0.3       progressing to viability.
L;Siddiqi TA;                                    glucose <5.5                             gestation (normal   (p<0.05)                However glycosated
                                                 mmol/l and a 90                          range 5.5-8.5%)                             proteins and glycosated
1986 Sep                                         minute                                                       Mean initial glycosated albumin were not
                                                 postprandial                             Glycosated          proteins(%):            significantly different in
64
                                                 glucose of less                          proteins (at 9      Miscarriage: 22.7 ± 2.0 abortion and control
                                                 than 7.7 mmol/l                          wks).               Pregnancy>20 wks: 22.2 groups. Because the half
                                                                                                              ± 0.7 (NS)              life of albumin
                                                 Pregnancies that                         Glycosated                                  approximates 14 days,
                                                 resulted in                              albumin (9          Mean initial glycosated the percentage of
                                                 congenital                               weeks)              albumin(%):             glycosylated albumin
                                                 malformations                                                Miscarriage: 14.2 ± 1.1 reflects the metabolic
                                                 were excluded                            Miscarriage         Pregnancy >20wks 15.1 control during the 2 to 4
                                                 from the study                                               ± 0.6 (NS)              weeks before
                                                                                                                                      measurement. Similarly,
                                                                                                                                      glycosylated total
                                                                                                                                      proteins reflect control
                                                                                                                                      during the previous one
                                                                                                                                      to two weeks before
                                                                                                                                      measurement. It appears
                                                                                                                                      therefore that poor
                                                                                                                                      metabolic control around
                                                                                                                                      conception and/or in the
                                                                                                                                      early weeks of
                                                                                                                                      pregnancy, but not in the
                                                                                                                                      few weeks preceeding
                                                                                                                                      the abortive event, is the
                                                                                                                                      determining factor
                                                                                                                                      favouring abortion.

                                                                                                                                       The authors speculated
                                                                                                                                       that an improved
                                                                                                                                       metabolic control at the
                                                                                                                                       time of conception and in
                                                                                                                                       the early pregnancy will
                                                                                                                                       will minimise the risk of
                                                                                                                                       both major malformations
                                                                                                                                       and Miscarriages.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 34 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                           Follow-up &             Effect Size            Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                Outcome                                                                   Comments
                 Level                 cs                                                  Measures
Suhonen        Study Type:       691             Type 1 diabetes.   Intervention:        Follow-up period:   There were 30 (4.2%)      Even a slightly raised
L;Hiilesmaa    Cohort            pregnancies     98% Finnish        Glycaemic control in                     malformations in          HbA1c during early
V;Teramo K;                      and 709 infants caucasians.        early pregnancy      Outcome             patients with type I      pregnancy in women with
               Evidence level:   of 488 women                                            Measures:           diabetes and 10 (1.2%)    Type 1 diabetes carries
2000 Jan       2++               with type 1                        Comparison:          Congenital          in controls (RR 3.1;95%   an increased risk for fetal
                                 diabetes. 729                                           malformations       CI 1.6-6.2).              malformations.Therefore
35
                                 non-selected                                                                                          normoglycaemia should
                                 control                                                 HbA1c (mean         HbA1c (%):                be strived for during early
                                 pregnancies in                                          normal              <5.6, RR=1.6 (0.3-9.5)    pregnancy
                                 women without                                           nondiabetic 4.93    5.6-6.8, RR=3.0 (1.2-
                                 diabetes.                                               (0.32)              7.5)
                                                                                                             6.9-8.0, RR=2.3 (1.0-
                                                                                                             5.7)
                                                                                                             8.1-9.3, RR=3.3 (1.3-
                                                                                                             8.6)
                                                                                                             ≥9.4, RR=4.8 (1.6-13.9)

                                                                                                             Regression:
                                                                                                             HbA1c associted with
                                                                                                             occurance of
                                                                                                             malformations after
                                                                                                             adjusting for White's
                                                                                                             class, age at onset of
                                                                                                             diabetes, duration of
                                                                                                             diabetes,
                                                                                                             parity,smoking, and
                                                                                                             participation in
                                                                                                             prepregnancy
                                                                                                             counselling. (p=0.02).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 35 of 441
                                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                    Follow-up &           Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                         Outcome                                                               Comments
                 Level                 cs                                           Measures
Miller E;Hare       Study Type:       116   women with type   Intervention: HbA1c Follow-up period: 15/116 (13%) had infant   Our results revealed a
JW;Cloherty         Cohort                  1 diabetes                                              with major congenital     significantly higher
JP;Dunn                                                       Comparison:         Outcome           malformation. The mean    incidence of major
PJ;Gleason          Evidence level:                                               Measures: A       initial HbA1c was         congenital anomalies in
RE;Soeldner         2++                                                           major congenital significantly higher in    the offspring of women
JS;Kitzmiller JL;                                                                 malformation was women with infants with    with elevated HbA1c
                                                                                  defined as one    major congenital          values, and we conclude
1981 May 28                                                                       causing death or malformations than         that poorly controlled
                                                                                  serious handicap those without (8.4 ± 1.6   diabetes is associated
57
                                                                                  or one requiring  vs 9.5± 1.0, p<0.01).     with an increased risk of
                                                                                  surgery.                                    such anomalies.
                                                                                                    Congenital
                                                                                                    malformations by first
                                                                                                    trimester HbA1c
                                                                                                    ≤6.9%: 0/19
                                                                                                    7.0-8.5%: 2/37(5.1%)
                                                                                                    8.6-9.9%: 8/35 (22.9%)
                                                                                                    ≥ 10.0%: 5/23 (21.7%)

                                                                                                    Group 1 (≤8.5%) 2/57
                                                                                                    (3.4%)
                                                                                                    Group 2 (>8.5%) 13/58
                                                                                                    (22.4%) p<0.01.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                Page 36 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &            Effect Size              Study Summary                  Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                                      Comments
                 Level                 cs                                                     Measures
Mills JL;Simpson   Study Type:       386 type 1      75.9% enrolled      Intervention:       Follow-up period: There were                  Diabetic women with        Diabetes in early
JL;Driscoll        Cohort            diabetes, 432   before              Glycaemic control                     62/386(16.1%)               good metabolic control     pregnancy (DIEP) study
SG;Jovanovic-                        control women   conception,                             Outcome           Miscarriages among          are no more likely than
Peterson L;Van     Evidence level:                   24.1% enrolled      Comparison:         Measures:         women with diabetes         nondiabetic women to       Because in the majority of
AM;Aarons          2++                               before 21 days                          Glycosylated      and 70/432 (16.2%) in       lose a pregnancy but       women with diabetes
JH;Metzger                                           after conception.                       hemoglobin        the control group.          diabetic women with        blood glucose levels were
B;Bieber                                                                                     (Nanomoles of                                 elevated blood glucose     well controlled the effect
FR;Knopp                                                                                     fructose/10g of   Within the normal range     and glycosylated           of poor diabetic control
RH;Holmes LB;                                                                                hemoglobin)       there was no relation       hemoglobin levels in the   was not apparent in the
                                                                                                               between the level of        first trimester have a     overall loss rates.
1988 Dec 22                                                                                                    glycosylated hemoglobin     significantly increased
                                                                                                               and the rate of             risk of having a
61
                                                                                                               pregnancy loss. In          Miscarriage.
                                                                                                               diabetic or control
                                                                                                               subjects. Above the
                                                                                                               normal range loss rates
                                                                                                               in the diabetic group
                                                                                                               increased in an
                                                                                                               approximately linear
                                                                                                               fashion with increasing
                                                                                                               levels of glycosylated
                                                                                                               hemoglobin (y=0.7x –
                                                                                                               19.1; p=0.015).

                                                                                                               Altough the numbers are
                                                                                                               small they suggest an
                                                                                                               increase in the risk of
                                                                                                               abortion of 3.1% (95%
                                                                                                               CI 0.6-5.6) for each
                                                                                                               increase of 1 SD above
                                                                                                               the mean.

                                                                                                               Fasting (p=0.01) and
                                                                                                               postprandial (p=0.004)
                                                                                                               blood glucose levels
                                                                                                               were significantly higher
                                                                                                               in women who aborted
                                                                                                               and were significant
                                                                                                               predictors of loss when
                                                                                                               analysed by logistic
                                                                                                               regression.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 37 of 441
                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size       Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                   Comments
                 Level                 cs                                   Measures
                                                                                         No relationship with
                                                                                         hypoglycaemia




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                             Page 38 of 441
                                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                    Follow-up &             Effect Size              Study Summary                    Reviewer
 Information & Evidence   Patients Characteristi Comparison                         Outcome                                                                         Comments
                 Level                 cs                                           Measures
Rosenn          Study Type:       215    Type 1 diabetes.    Intervention: HbA1   Follow-up period: 52/215 women who              Women with type 1            The study uses HbA1 with
B;Miodovnik     Cohort                   Glycaemic control   (normal range 5.5-                       enrolled before 9 weeks     diabetes with initial HbA1   a normal range 5.5-8.5%
M;Combs                                  achieved with       8.5)                 Outcome             gestation (24%) had         >12% or median first         which is higher than the
CA;Khoury       Evidence level:          dietary regimen                          Measures:           Miscarriages. The           trimester preprandial        normal range for HbA1c
J;Siddiqi TA;   2++                      and split, mixed    Comparison:          Miscarriage         threshold for increase      glucose concentrations
                                         dosage regimens                          defined at          risk of abortion was a      >6.6 mmol/l have an
1994 Oct                                 of short and                             miscarriages        preprandial glucose         increased risk of abortion
                                         intermediate-                            before 20 weeks concentration of 7.15           and malformations.
60
                                         acting insulin.                          gestation.          mmol/l or an initial HbA1   Below these glycaemic
                                         Goals of                                                     of 12% (6.2 SD above        thresholds the risks are
                                         glycaemic control                        Major congenital the normal mean)               comparable to those in
                                         were a fasting                           malformations       p<0.05.                     nondiabetic women.
                                         blood glucose <                          were defined as
                                         5.6 mmol/l, and a                        gross physical or 55/284 women who
                                         90 min                                   anatomical          enrolled before 14
                                         postprandial                             developmental       weeks gestation had
                                         glucose                                  anomalies           Miscarriages and were
                                         <7.8mmol/l.                              resulting in death, excluded from analysis.
                                         When necessary                           requiring major     14 of the remaining 229
                                         women were                               surgery or          (6.1%) infants were born
                                         hospitalised to                          substantial         with congenital
                                         maintain these                           medical             malformations (including
                                         goals.                                   treatment, or       3 stillborn and one
                                                                                  causing a           neonatal death). The
                                                                                  substantial         threshold for increased
                                                                                  physical or         risk of malformations
                                                                                  psychological       was a preprandial
                                                                                  handicap.           glucose concentration of
                                                                                                      6.6 mmol/l or an initial
                                                                                  HbA1 normal         HbA1 of 13% (6.2 SD
                                                                                  range 5.5-8.5%      above the normal mean)
                                                                                                      p<0.05.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                     Page 39 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &             Effect Size                Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                                      Comments
                 Level                 cs                                                    Measures
Ylinen K;Aula   Study Type:       146 fetuses in   Women with type Intervention: HbA1c Follow-up period:       17/142 pregnancies            Increased incidence of
P;Stenman       Cohort            142 women        1 diabetes      (mean nondiabetic                           complicated by                fetal malformations are
UH;Kesaniemi-                                                      HbA1c 5% (SD 0.5)) Outcome                  malformation - 6 minor        associated with maternal
Kuokkanen       Evidence level:                                                        Measures:               and 11 major. The mean        hyperglycaemia during
T;Teramo K;     2++                                                Comparison:         Malformation            initial HbA1c was             organogenesis. Hence
                                                                                       defined as major        significantly higher in the   diabetic women who are
1984 Aug 11                                                                            if fatal or likely to   group with minor              planning to have a child -
                                                                                       cause serious           malformations (9.3, SD        especially those with a
56
                                                                                       handicap to the         1.9, p<0.05) and the          high HbA1c value -
                                                                                       child. Other            group with major              should receive intensified
                                                                                       malformations           malformations (9.6, SD        metabolic control.
                                                                                       classified as           1.8,p<0.001)) than in the
                                                                                       minor.                  group without
                                                                                                               malformations (8.0, SD
                                                                                                               1.4%)

                                                                                                               Congenital
                                                                                                               malformations by initial
                                                                                                               HbA1c:
                                                                                                               <8.0%: 3/63 (4.8%)
                                                                                                               8.0-9.9%: 8/62 (12.9%)
                                                                                                               ≥ 10.0%: 6/17(35.3%)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                Page 40 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &            Effect Size            Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                            Comments
                 Level                 cs                                                 Measures
Lucas MJ;Leveno Study Type:       105 women       GDM=23, White Intervention:           Follow-up period: The mean HbA1a+b+c
KJ;Williams     Cohort            with diabetes   Class A= 4, B=27, Glycaemic control                      of the 105 study patients
ML;Raskin                         14 women        C=25, D=18, F,H                       Outcome            was 9.2% (range 5.2%
P;Whalley PJ;   Evidence level:   without         and R =8.         Comparison:         Measures:          to 14.2%).
                2++               diabetes                                              congenital
1989 Aug                                                                                malformations      The mean ± SD
                                                                                        Miscarriages       HbA1a+b+c levels of the
58
                                                                                                           73 mothers of infants
                                                                                        Total glycosylated with no malformations
                                                                                        haemoglobin        was 8.9%±2.3%
                                                                                        (HbA1a+b+c).       compared with
                                                                                        Mean in            10.3%±1.9% for mothers
                                                                                        nondiabetic        of infants with major or
                                                                                        women was 6.0% minor malformations
                                                                                        (range 5.0% -      (p=0.05).
                                                                                        6.9%).
                                                                                                           In pregnancies that
                                                                                                           progressed beyond
                                                                                                           abortion there was a
                                                                                                           statistically significant
                                                                                                           (p=0.05) distribution of
                                                                                                           malformation above the
                                                                                                           HbA1a+b+c of 9.2%;
                                                                                                           10/14 malformations
                                                                                                           occurred in 42 women
                                                                                                           whose levels exceeded
                                                                                                           this value, compared
                                                                                                           with four malformations
                                                                                                           in the 45 women with
                                                                                                           levels below this value.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 41 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size              Study Summary                   Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                       Comments
                 Level                 cs                                                      Measures
Rosenn          Study Type:       99 pregnant      Significantly more   Intervention:        Follow-up period: HbA1c were significantly     Women in the pre-           High drop-out rate in
B;Miodovnik     Cohort            women with       patients in the      Preconception care                     lower in the                 conception program          preconception clinic
M;Combs                           Type 1           study group had                           Outcome           preconception group at       achived good glycaemic
CA;Khoury       Evidence level:   diabetes. 28     advanced disease     Comparison:          Measures:         the first prenatal visit     control before
J;Siddiqi TA;   2++               had attended a   compared with        Women enrolled       Pregnancy loss    and at 9 and 14 weeks        conceiving; this was
                                  preconception    the control group    after conception                       gestation p=0.0008,          associted with a
1991 Jun                          clinic and 71    (p=0.04)                                  Glycohemoglobin p=0.002,p=0.02). There         significantly lower rate of
                                  had enrolled                                               (normal range     was no significant           Miscarriages.
62
                                  after                                                      5.5-5.8)          difference at 26 weeks
                                  conception.                                                                  or delivery. The rate of
                                                                                                               Miscarriage was
                                                                                                               significantly lower in the
                                                                                                               preconception group
                                                                                                               (7% compared to 24%,
                                                                                                               p=0.04). One major
                                                                                                               malformation
                                                                                                               (microcephaly) was
                                                                                                               diagnosed in the
                                                                                                               preconception group
                                                                                                               and none in the
                                                                                                               postconception group.

                                                                                                                HbA1c at the first
                                                                                                                prenatal visit and at 9
                                                                                                                and 14 weeks gestation
                                                                                                                were significantly higher
                                                                                                                among those who had
                                                                                                                Miscarriages (p=0.005,
                                                                                                                p=0.02, p=0.02).

                                                                                                                Rate of Miscarriages
                                                                                                                according to HbA1c at 9
                                                                                                                weeks gestation:
                                                                                                                HbA1c (SD above
                                                                                                                nondiabetic mean):
                                                                                                                <1, 0.08
                                                                                                                1-3, 0.05
                                                                                                                3-5, 0.11
                                                                                                                5-9, 0.57
                                                                                                                >9, 0.5




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 42 of 441
                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &              Follow-up &            Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                   Outcome                                                                    Comments
                 Level                 cs                                     Measures
Schaefer-Graf      Study Type:       4180               Intervention:       Follow-up period: The initial fasting serum     Congenital malformations
UM;Buchanan        Cohort            pregnancies        glycaemic control                     glucose, highest fasting      in offspring of women
TA;Xiang                             (3764                                  Outcome           serum glucose and and         with gestational
A;Songster         Evidence level:   complicated by     Comparison:         Measures:         HbA1c levels were             andtype2diabetes affect
G;Montoro M;Kjos   2++               gestational                            congenital        significantly higher in       the same organ systems
SL;                                  diabetes, 416                          malformations     pregnancies with major        that have previously
                                     complicated by                                           (143) and minor (n=112)       been described in
2000 Feb                             type 2 diabetes)                                         anomalies (p<0.0001 for       pregnancies complicated
                                                                                              all glycemic                  by type 1 diabetes
402
                                                                                              parameters).                  .Increasing
                                                                                                                            hyperglycaemia at
                                                                                               Malformation rate by         diagnosis or presentation
                                                                                               initial fasting serum        for care was associated
                                                                                               glucose level                with an increasing risk of
                                                                                               ≤6.6mmol/l, 64/3060          malformations.
                                                                                               (2.1%)
                                                                                               6.65-11 mmol/l, 53/905
                                                                                               (5.9%)*
                                                                                               >11 mmol/l 26/185
                                                                                               (12.9%)*†
                                                                                               * p<0.0001, compared
                                                                                               with initial fasting serum
                                                                                               glucose level of ≤6.6
                                                                                               mmol/l
                                                                                               † p<0.0001 when
                                                                                               compared with initial
                                                                                               fasting serum glucose
                                                                                               level of 6.65-11mmol/l




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                              Page 43 of 441
                                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                       Follow-up &           Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                            Outcome                                                                 Comments
                 Level                 cs                                              Measures
Steel          Study Type:       143 women      type1 diabetes   Intervention:       Follow-up period: Mean HbA1 during the       Tight control of the
JM;Johnstone   Cohort            attending a                     Prepregnancy care                     first trimester was        maternal blood glucose
FD;Hepburn                       prepregnancy                                        Outcome           significantly lower        concentration in the early
DA;Smith AF;   Evidence level:   clinic and 96                   Comparison:         Measures:         among attenders at the     weeks of pregnancy can
               2++               women                                               Congenital        prepregnancy clinic        be achieved by the
1990 Nov 10                      managed over                                        malformations     (8.4%) than among the      prepregnancy clinic
                                 the same                                                              non attenders (10.5%,      approach and is
51
                                 period who had                                      Maternal          p<0.0001)                  associated with a highly
                                 not received                                        hypoglycaemia                                significant reduction in
                                 specific                                                              Hypoglycaemia was          the risk of serious
                                 prepregnancy                                        HbA1(normal       significantly more         congenital abnormalities
                                 care                                                range 5.9-8.0%) common in all patients       in the offspring.
                                                                                                       who had attended the       Hypoglycaemic episodes
                                                                                                       prepregnancy clinic than   do not seem to lead to
                                                                                                       non attenders (38/143      fetal malformation even
                                                                                                       women v 8/96; χ2=11.1,     when they occur during
                                                                                                       p<0.001). One out of 12    the period of
                                                                                                       mothers who had an         organogenesis.
                                                                                                       infant with malformation
                                                                                                       had experienced
                                                                                                       hypoglycaemia in the
                                                                                                       first 9 weeks. The other
                                                                                                       45 patients who had
                                                                                                       severe hypoglycaemia
                                                                                                       had no abnormal
                                                                                                       neonatal outcomes.

                                                                                                       There were 12 infants
                                                                                                       with major congenital
                                                                                                       malformations. The
                                                                                                       malformations were
                                                                                                       significantly more
                                                                                                       common in babies born
                                                                                                       to patients who had not
                                                                                                       attended the
                                                                                                       prepregnancy clinic
                                                                                                       (10/96; 10.4%) than in
                                                                                                       babies of mothers who
                                                                                                       had attended (2/143;
                                                                                                       1.4%, χ2=8.0, p<0.005;
                                                                                                       RR 7.4 95% CI 1.7-
                                                                                                       33.2). Median HbA1 in




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                     Page 44 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         first 9 weeks of mothers
                                                                                         of infants with
                                                                                         malformations 12.5%
                                                                                         (range 8.8-14.9) .




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 45 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &             Effect Size              Study Summary                 Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                      Comments
                 Level                 cs                                                Measures
Rosenn           Study Type:       84 women   Type 1 diabetes.    Intervention:        Follow-up period: 71% of patients had at        Hypoglycaemia is a
BM;Miodovnik     Cohort                       Recruited before    Intensive insulin                       least one moderate           frequent complication
M;Holcberg                                    9 weeks gestation   therapy throughout   Outcome            symptomatic episode          that significantly affects
G;Khoury         Evidence level:              and received        pregnancy            Measures:          and 34% had at least         the majority of women
JC;Siddiqi TA;   2++                          intensive insulin                        Hypoglycaemia      one severe symptomatic       who receive intensive
                                              therapy             Comparison:          Biochemical        episode.                     insulin therapy during
1995 Mar                                      throughout                               hypoglycaemia                                   pregnancy.
                                              pregnancy.                               determined from 19 patients had a total of
68
                                              Patients                                 glucometer data 54 episodes of loss of          This may be related to
                                              monitored                                                   consciousness and 12         the institution of intensive
                                              glucose                                  Symptomatic        patients had a total of 15   therapy which diminishes
                                              concentrations                           hypoglycaemic      seizures.                    counteregulaltory
                                              with memory                              episodes                                        responses to
                                              glucometers and                          recorded weekly Most symptomatic                hypoglycaemia and
                                              insulin doses                            on standardised episodes occurred               increases hypoglycaemic
                                              were adjusted                            forms              during the first half of     unawareness.
                                              weekly                                                      pregnancy. The peak          Alternatively it may be
                                              accordingly.                             Moderate           incidence of severe or       that pregnancy
                                                                                       symptomatic        moderate symptomatic         independently increases
                                                                                       hypoglycaemia:     hypoglycaemia was            the risk of
                                                                                       Any episode that between 10 and 15              hypoglycaemia, possibly
                                                                                       could not be       weeks.                       due to hormonal
                                                                                       resolved by the                                 changes, diminished
                                                                                       patient herself    There was no                 counterregulatory
                                                                                       and required       association between          responses to
                                                                                       active assistance hypoglycaemia during          hypoglycaemia or the
                                                                                       from another       early pregnancy and          nausea and vomiting of
                                                                                       person             adverse pregnancy            pregnancy.
                                                                                                          outcome.
                                                                                       Severe                                          At present, a measure of
                                                                                       symptomatic        When hypoglycaemic           caution should be
                                                                                       hypoglycaemia:     and nonhypoglycaemic         exercised for the
                                                                                       an episode         groups were compared         occasional patient who
                                                                                       resulting in       there was no differences     demonatrates a tendency
                                                                                       seizures, loss of in any of the commonly        to have recurrent
                                                                                       consciousness,     used indices of              episodes of severe
                                                                                       injury, IV glucose glycaemic control,           hypoglycaemia that
                                                                                       treatment or       including preprandial        cannot be resolved with
                                                                                       emergency          glucose, postprandial        the usual tactics of
                                                                                       glucagon           glucose and                  modifying insulin
                                                                                       administration.    glycohaemoglobin             regimens, caloric intake,
                                                                                                          concentrations.              and physical activity.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 46 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type             Aim of Study          Number of      Population                     Outcome      Results & Comments               Study         Reviewer
 Information & Evidence                                    Patients &    Characteristic                  measures                                     Summary        Comment
                 Level                                       Patient          s
                                                          Characteristic
                                                                s
The Diabetes     Study Type:        Intervention: Women   180 women             13-39 years,          Miscarriage    Intensive treatment group     Timely
Control and      OtherNon           originally assigned   completed 270         type1diabetes, in     congenital     mean HbA1c at conception      institution of
Complications    randomised         to intensive          pregnancies. 94       good health.          malformation   7.4%±1.3%. Conventional       intensive
Trial Research   controlled trial   treatment             women originally                                           treatment group mean HbA1c    therapy is
Group;                                                    assigned to           Goals for all         HbA1c          at conception 8.1%±1.7%,      associated with
                 Evidence Level: Comparison:              intensive             pregnant women                       p=0.0001.                     ratesofMiscarri
1996 Apr         2++             Women originally         treatment. 86         fasting level 3.85-                                                age and
                                 assigned to              women originally      5.5, 1 hour                          Nine congenital malformations gongenital
49
                                 conventional             assigned to           postprandial level                   were identified, 8 in the     malformations
                                 treatment                conventional          7.7 mmol/l                           conventional treatment group similar to those
                                                          treatment (26                                              (p=0.06).                     in the
                                                          changed to                                                                               nondiabetic
                                                          intensive treatment                                        There were 18 (13.3%)         population.
                                                          before conception,                                         Miscarriages in the intensive
                                                          60 after                                                   group and 14 (10.4%) in the
                                                          conceptiont).                                              conventional group (NS).

                                                                                                                     Overallthe median HbA1c level
                                                                                                                     was higher in those women
                                                                                                                     who had abnormal outcomes
                                                                                                                     (ectopic pregnancy,
                                                                                                                     Miscarriage,induced abortion
                                                                                                                     formedicalreasons,congenital
                                                                                                                     malformations, intrauterine
                                                                                                                     death, neonatal death) than in
                                                                                                                     thosewomen with normal
                                                                                                                     outcomes (p=0.05). At the 3
                                                                                                                     months prior to conception
                                                                                                                     p=0.018 and at 5moths
                                                                                                                     p=0.019.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 47 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Q5 - How should blood glucose and ketones be monitored in the preconception period?

     Bibliographic    Study Type &       Number of        Patient            Intervention &       Follow-up &            Effect Size               Study Summary            Reviewer Comments
      Information    Evidence Level       Patients     Characteristics        Comparison            Outcome
                                                                                                    Measures
The Diabetes         Study Type:       711 intensive   Type 1 diabetes Intervention:           Follow-up period: A statistically significant   Intensive therapy with the
Control and          RCT               therapy         aged 13-39 years. Intensive therapy     mean 6.5 years    difference in the             goal of achieving blood
Complications                          730             Absence of        (IT): Aimed at                          average HbA1c was             glucose concentrations
Research Group       Evidence level:   conventional    hypertension,     achieving             Outcome           maintained after              as close as possible to
                     1++               therapy         hypercholesterole preprandial blood     Measures:         baseline between the IT       the nondiabetic range
1993                                                   mia and severe    glucose               HbA1c             and CT groups                 delays onset and slows
                                                       diabetic          concentrations        Progression of    (p<0.001).                    the progression of long-
69
                                                       complications or between 3.9 and 6.7 retinopathy:         The mean (SD) value for       term diabetic
                                                       medical           mmol/l, postprandial change of at least all glucose profiles in the   complications.
                                                       conditions.       concentrations of     three steps from IT group was 8.6
                                                                         <10 mmol/l, a         baseline that was ±1.7mmol/l vs. 12.8±3.1
                                                       Country: US and weekly 3am              sustained for at  mmol/l in the CT group
                                                       Canada            measurement >3.6 least six months (p<0.0001).
                                                                         mmol/l and HbA1c      (early treatment
                                                                         measured monthly      diabetic          Primary prevention
                                                                         within the normal     retinopathy Study cohort (i.e. no
                                                                         range (<6.05%).       scale).           retinopathy at baseline,
                                                                         Intensive therapy     Proliferative     n=348 IT, 378 CT):
                                                                         included the          retinopathy       Intensive therapy
                                                                         administration of     Severe            reduced the adjusted
                                                                         insulin 3 or more     nonproliferative  mean risk for developing
                                                                         times a day by        retinopathy       retinopathy by 76%
                                                                         injection or external                   (95% CI 62-85%) as
                                                                         pump. The dosage Secondary              compared with
                                                                         was adjusted          outcomes          conventional therapy.
                                                                         according to the      Nephropathic      Secondary intervention
                                                                         results of self       Neuropathic       cohort:
                                                                         monitoring of blood Neuropsychologic Intensive therapy
                                                                         glucose performed al                    slowed the progression
                                                                         at least four times   Macrovascular     of retinopathy by 54%
                                                                         per day, dietary      Quality of life   (95% CI 14-67%)
                                                                         intake and            outcomes
                                                                         anticipated exercise.                   Intensive therapy
                                                                                                                 reduced the occurrence
                                                                         Comparison:                             of microalbuminuria by
                                                                         Conventional                            39%, albuminuria by
                                                                         therapy (CT). CT                        54% and clinical
                                                                         consisted of 1 or 2                     neuropathy by 60%.
                                                                         daily injections of
                                                                         insulin, daily self-                    In the IT group there




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 48 of 441
                                                                                                                               DRAFT FOR CONSULTATION




 Bibliographic    Study Type &    Number of      Patient          Intervention &      Follow-up &         Effect Size            Study Summary    Reviewer Comments
  Information    Evidence Level    Patients   Characteristics      Comparison          Outcome
                                                                                       Measures
                                                                monitoring of urine                 were 62 hypoglycaemic
                                                                or blood glucose,                   episodes requiring
                                                                and education about                 assistance per 100
                                                                diet or exercise.                   patient years vs. 19
                                                                                                    episodes per 100 patient
                                                                                                    years in the CT group
                                                                                                    (p<0.001).There were
                                                                                                    two fatal motor vehicle
                                                                                                    accidents (1 from each
                                                                                                    group).

                                                                                                    There was no difference
                                                                                                    in QOL between the two
                                                                                                    groups.

                                                                                                    Early worsening:
                                                                                                    There was a transient
                                                                                                    worsening of retinopathy
                                                                                                    with intensive therapy
                                                                                                    (22% in IT vs. 13% in
                                                                                                    CT). The abnormalities
                                                                                                    often disappeared by 18
                                                                                                    months. Patients with
                                                                                                    early worsening who
                                                                                                    underwent intensive
                                                                                                    therapy had a 74%
                                                                                                    (95% CI 46-88%)
                                                                                                    reduction in the risk of
                                                                                                    subsequent progression
                                                                                                    as compared with
                                                                                                    patients with early
                                                                                                    worsening who received
                                                                                                    conventional therapy
                                                                                                    (p<0.001).

                                                                                                    Secondary analysis:
                                                                                                    There was a
                                                                                                    continuously increasing
                                                                                                    risk of sustained
                                                                                                    progression by three
                                                                                                    steps with increasing
                                                                                                    mean HbA1c.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                             Page 49 of 441
                                                                                                                                              DRAFT FOR CONSULTATION




     Bibliographic    Study Type &       Number of          Patient           Intervention &        Follow-up &             Effect Size           Study Summary          Reviewer Comments
      Information    Evidence Level       Patients       Characteristics       Comparison             Outcome
                                                                                                      Measures
The Diabetes         Study Type:       180 women         Type 1 diabetes,   Intervention:         Follow-up period: Intensive treatment       Timely institution of
Control and          Cohort            completed 270     in good health.    Women originally      9 months          group mean HbA1c at       intensive therapy is
Complications                          pregnancies. 94                      assigned to                             conception 7.4%±1.3%.     associated with rates of
Trial Research       Evidence level:   women             Country: US and    intensive treatment   Outcome           Conventional treatment    spontaneous abortion
Group;               2++               originally        Canada                                   Measures:         group mean HbA1c at       and congenital
                                       assigned to                          Goals for all         Spontaneous       conception 8.1%±1.7%,     malformations similar to
1996 Apr                               intensive                            pregnant women        abortion          p=0.0001.                 those in the nondiabetic
                                       treatment. 86                        fasting level 3.85-   congenital                                  population.
49
                                       women                                5.5 mmol/l, 1 hour    malformation      Nine congenital
                                       originally                           postprandial level                      malformations were
                                       assigned to                          7.7 mmol/l            HbA1c             identified, 8 in the
                                       conventional                                                                 conventional treatment
                                       treatment (26                        Comparison:                             group (p=0.06).
                                       changed to                           Women originally
                                       intensive                            assigned to                             There were 18 (13.3%)
                                       treatment                            conventional                            spontaneous abortions
                                       before                               treatment                               in the intensive group
                                       conception, 60                                                               and 14 (10.4%) in the
                                       after                                                                        conventional group
                                       conception).                                                                 (NS).

                                                                                                                    Overall the median
                                                                                                                    HbA1c level was higher
                                                                                                                    in those women who
                                                                                                                    had abnormal outcomes
                                                                                                                    (ectopic pregnancy,
                                                                                                                    spontaneous abortion,
                                                                                                                    induced abortion for
                                                                                                                    medical reasons,
                                                                                                                    congenital
                                                                                                                    malformations,
                                                                                                                    intrauterine death,
                                                                                                                    neonatal death) than in
                                                                                                                    those women with
                                                                                                                    normal outcomes
                                                                                                                    (p=0.05). At the 3
                                                                                                                    months prior to
                                                                                                                    conception p=0.018 and
                                                                                                                    at 5 months p=0.019.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                Page 50 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Q6 – How cost effective are self management programmes for women with diabetes who are planning a
pregnancy?

No specific searches were conducted for this clinical question (this is a health economics question), however, the
following study was identified in searches for other questions.

Bibliographic Study Type          Aim of Study       Number of      Population                 Outcome               Results & Comments                Study              Reviewer
 Information & Evidence                              Patients &    Characteristic              measures                                               Summary             Comment
                 Level                                 Patient          s
                                                    Characteristic
                                                          s
Feig DS;Cleave   Study Type:     Intervention:       64           Inclusion criteria:      Diabetes self-           Significant improvement in all   Women              Decrease in
B;Tomlinson G;   Other           Retrospective chart              Women with type 1        management               diabetes self-management         participating in   HbA1c may be
                 Retrospective   review done to                   or type 2 diabetes,      behaviours:              behaviours including frequency   an intensive       explained by the
2006                             obtain information               who were at least 1      frequency of self-       of self-monitoring of blood      diabetes           loss of contact
                 Evidence Level: on self-management               year postpartum          monitoring of blood      glucose, frequency of insulin    management         with the diabetes
119
                 2-              behaviours at entry              Exclusion criteria: if   glucose, frequency of    injections, and frequency and    programme          care team and/or
                                 to the programme                 women gave birth         insulin injections and   complexity of insulin dose       during             the
                                 and at delivery and              before 20 weeks’         frequency and            adjustment from entry to the     pregnancy          discontinuation of
                                 compared with the                gestation                complexity of insulin    programme to delivery.           improve            frequent self-
                                 present.                                                  dose adjustments)        Significant improvement in the   significantly      monitoring of
                                                                  Country: Canada          HbA1C                    HbA1c from entry to delivery.    from entry to      blood glucose
                                 Comparison:                                                                        While comparing entry to the     delivery in        that necessary
                                                                                                                    present, a significant           diabetes self-     for achieving
                                                                                                                    improvement in frequency of      management         optimal
                                                                                                                    insulin injections, frequency    behaviours and     glycaemic
                                                                                                                    and complexity of insulin dose   glycaemic          control.
                                                                                                                    adjustment No significant        control and in
                                                                                                                    change in frequency of self-     the long term
                                                                                                                    monitoring of blood glucose      retention of
                                                                                                                    from before pregnancy to the     some of the
                                                                                                                    present, and a significant       behaviours and
                                                                                                                    decrease in HbA1c by 0.015 (p    knowledge.
                                                                                                                    < 0.0001, 95% CI 0.009-0.021)    No
                                                                                                                    from entry to the programme to   improvement in
                                                                                                                    the present.                     HbA1c level




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                       Page 51 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Q7 - Which medications for diabetes are suitable for use during pregnancy and which should be discontinued?

Bibliographic Study Type Number of   Patient     Intervention &                            Follow-up &              Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                 Outcome                                                                    Comments
                 Level                 cs                                                   Measures
Langer O;Conway Study Type:      404 women          Women were        Intervention:       Follow-up period:    Blood glucose (mmol/l):     We found that among
DL;Berkus       RCT              with gestational   screened for      Glibenclamide                            Fasting                     women with gestational
MD;Xenakis                       diabetes           diabetes with a                       Outcome              Glibenclamide: 5.7±1.4      diabetes, the degree of
EM;Gonzales O; Evidence level:   201 assigned to    one-hour, 50g     Comparison: Insulin Measures: Blood      Insulin:5.9±1.4 (p=0.12)    glycaemic control and
                1++              receive            oral glucose                          glucose              Preprandial:                perinatal outcomes were
2000 Oct 19                      glibenclamide      challenge. The                        Fasting (mmol/l)     Glibenclamide: 5.7± 1.1     essentially the same for
                                 and 203 to         women ranged in                       Preprandial          Insulin:5.9±1.3 (p=0.16)    those treated with
73
                                 receive insulin.   age from 18 to 40                     (mmol/l)             Postprandial:               glibenclamide and those
                                                    and were all                          Postprandial         Glibenclamide: 7.2±1.4      treated with insulin.
                                                    medicaid                              (mmol/l)             Insulin:7.1±1.5 (p=0.69)
                                                    recipients. 83%                       Mean (mmol/l)        Mean:                       Glibenclamide is an
                                                    Hispanic, 12%                         Glycosated           Glibenclamide: 6.3±1.05     effective alternative to
                                                    non Hispanic                          haemoglobin (%)      Insulin:6.4±0.1.2           insulin in women with
                                                    white, 5 % black.                                          (p=0.33)                    gestational diabetes.
                                                                                          Neonatal             Glycosated
                                                    Country: USA                          outcomes:            haemoglobin (%):
                                                                                          Birth weight (g)     Glibenclamide: 5.7±1.3
                                                                                          LGA (no,%)           Insulin: 5.6±1.2 (p=0.42)
                                                                                          Macrosomia
                                                                                          (no,%)               Birth weight (g)
                                                                                          Cord-serum           Glibenclamide:3256±54
                                                                                          insulin              3
                                                                                          Intravenous          Insulin:3194±598
                                                                                          glucose therapy      (p=0.28)
                                                                                          (no,%)               LGA (no,%)
                                                                                          Hypoglycaemia        Glibenclamide:25 (12)
                                                                                          Hypocalcaemia        Insulin:26 (13) p=0.76
                                                                                          Hyperbilirubinemi    Macrosomia (no,%)
                                                                                          a                    Glibenclamide:14 (7)
                                                                                          Polycythemia         Insulin:9 (4) p=0.26
                                                                                          Lung                 Cord-serum insulin
                                                                                          complications (no,   Glibenclamide:15±13
                                                                                          %)                   Insulin:15±21 (p=0.84)
                                                                                          Respiratory          Intravenous glucose
                                                                                          support (no,%)       therapy (no,%)
                                                                                          Admission to         Glibenclamide:28 (14)
                                                                                          neonatal intensive   Insulin:22 (11)p=0.36
                                                                                          care                 Hypoglycaemia
                                                                                          Congenital           Glibenclamide:18 (9)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 52 of 441
                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &           Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                                           anomaly          Insulin:12 (6) p=0.25
                                                                           Stillbirth       Hypocalcaemia
                                                                           Neonatal death   Glibenclamide:2 (1)
                                                                                            Insulin:2 (1) p=0.99
                                                                                            Hyperbilirubinemia
                                                                                            Glibenclamide:12 (6)
                                                                                            Insulin:8 (4) p=0.36
                                                                                            Polycythemia
                                                                                            Glibenclamide:4 (2)
                                                                                            Insulin:6 (3) p=0.52
                                                                                            Lung complications (no,
                                                                                            %)
                                                                                            Glibenclamide:16 (8)
                                                                                            Insulin:12 (6) p=0.43
                                                                                            Respiratory support
                                                                                            (no,%)
                                                                                            Glibenclamide:4 (2)
                                                                                            Insulin:6 (3)p=0.52
                                                                                            Admission to neonatal
                                                                                            intensive care
                                                                                            Glibenclamide:12 (6)
                                                                                            Insulin:14 (7) p=0.68
                                                                                            Congenital anomaly:
                                                                                            Glibenclamide: 5 (2)
                                                                                            Insulin:4 (2) p=0.74
                                                                                            Stillbirth
                                                                                            Glibenclamide:1 (0.5)
                                                                                            Insulin:1 (0.5) p=0.99
                                                                                            Neonatal death
                                                                                            Glibenclamide:1 (0.05)
                                                                                            Insulin: 1 (0.05) p=0.99



                                                                                            Glibenclamide was not
                                                                                            detected in the cord
                                                                                            serum of any infant.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                    Page 53 of 441
                                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &           Effect Size              Study Summary                  Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                                     Comments
                 Level                 cs                                                     Measures
Hague, W.,          Study Type:       14 Insulin     Women with        Intervention:       Follow-up period: Mode of birth No (%)         The outcome measure of Pilot study. A larger trial is
Davoren, P.,        RCT               16 Metformin   gestational       Metformin                             Insulin:                     fetal β cell activity,    underway (MiG study).
Oliver,J., Rowan,                                    diabetes                              Outcome           Vaginal delivery 11 (79)     assessed by cord C-
J.                  Evidence level:                  diagnosed         Comparison: Insulin Measures: Mode Induction of labour 9           peptide concentration
                    1-                               according to                          of birth          (64)                         was not different in the
2003                                                 criteria of the                       Birth weight (g)  Elective caesarean 2         two groups (p=0.31).
                                                     Australasian                          No birth weight   (14)                         Perinatal outcomes were
74
                                                     Diabetes in                           >4000g            Emergency caesarean 1        not different but numbers
                                                     Pregnancy                             Median cord C-    (7)                          are too small to comment
                                                     Society (ADIPS)                       peptide                                        further.
                                                                                           Cord glucose      Metformin
                                                     Country:                              Cord glucose/C- Vaginal delivery 5 (31)
                                                                                           peptide           Induction of labour 5(31)
                                                                                           No of neonates    Elective caesarean 8
                                                                                           with IV dextrose (50)
                                                                                           Median time in    Emergency caesarean 2
                                                                                           special care      (13)
                                                                                           nursery
                                                                                           No of neonates    Birth weight (g) Mean
                                                                                           with jaundice     (SD)
                                                                                                             Insulin:3450 (510)
                                                                                                             Metformin: 3560 (50)

                                                                                                              No birth weight >4000g
                                                                                                              Insulin: 2
                                                                                                              Metformin:2

                                                                                                              Median cord C-peptide
                                                                                                              (pmol/ml) (range) (No)
                                                                                                              Insulin: 0.66 (0.45-1.71)
                                                                                                              (7)
                                                                                                              Metformin:0.53 (0.35-
                                                                                                              2.86) (10)

                                                                                                              Cord glucose (mmol/l)
                                                                                                              Mean (SD) (No)
                                                                                                              Insulin:4.2 (1.1) (11)
                                                                                                              Metformin:4.2 (1.9) (14)

                                                                                                              Cord glucose/C-peptide
                                                                                                              Insulin:5.7 (2.67)
                                                                                                              Metformin:7.4 (1.69)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 54 of 441
                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size         Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                     Comments
                 Level                 cs                                   Measures
                                                                                         No of neonates with IV
                                                                                         dextrose
                                                                                         Insulin: 1
                                                                                         Metformin:4

                                                                                         Median time in special
                                                                                         care nursery
                                                                                         Insulin:24 (0-102)
                                                                                         Metformin:48 (0-360)

                                                                                         No of neonates with
                                                                                         jaundice
                                                                                         Insulin: 6
                                                                                         Metformin: 3




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 55 of 441
                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &             Follow-up &            Effect Size           Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                  Outcome                                                                Comments
                 Level                 cs                                    Measures
Hawthorne, G.   Study Type:       11 small, non-            Intervention:   Follow-up period: No congenital            metformin appears to be No evidence tables
                Systematic        randomised,               Metformin                         malformations were       safe and non-teratogenic
2006            review - meta-    non-controlled Country:                   Outcome           reported. In one study   during pregnancy
                analysis          studies, 5 in             Comparison:     Measures:         metformin was
71
                                  women with                                Congenital        associated with an
                Evidence level:   type 2 diabetes                           malformations     increased incidence of
                2+                and 6 in women                                              perinatal mortality and
                                  with polycystic                                             preeclampsia. However
                                  ovary syndrome                                              this study was poorly
                                  (PCOS).                                                     controlled and outcomes
                                                                                              could not be attributed
                                                                                              to the use of metformin.

                                                                                              The use of metformin
                                                                                              during pregnancy in
                                                                                              women with PCOS is
                                                                                              associated with a
                                                                                              reduction in miscarriage
                                                                                              in early pregnancy,
                                                                                              weight loss, a reduction
                                                                                              in fasting serum insulin
                                                                                              levels and in the
                                                                                              incidence of gestational
                                                                                              diabetes.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                        Page 56 of 441
                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &             Follow-up &           Effect Size            Study Summary                   Reviewer
 Information & Evidence   Patients Characteristi Comparison                  Outcome                                                                    Comments
                 Level                 cs                                    Measures
Checa              Study Type:           Type 2 diabetes, Intervention:    Follow-up period: 2 small RCTs of          The safety profile of       Systematic review does
MA;Requena         Systematic            gestational                                         metformin in gestational metformin has been          not provide evidence
A;Salvador C;Tur   review - meta-        diabetes or PCOS Comparison:      Outcome           diabetes.                sufficiently established    tables or give details for
R;Callejo          analysis                                                Measures:                                  for the use of this drug    the majority of studies
J;Espinos                                Country:                                            12 cohort studies        during gestation. In        included in the review.
JJ;Fabregues       Evidence level:                                                           undertaken between       women currently treated
F;Herrero J;       2+                                                                        1979 and 2004 in         with metformin who
                                                                                             women with type 2        become pregnant, the
2005                                                                                         diabetes, gestational    administration of
                                                                                             diabetes or PCOS.        metformin should be
403
                                                                                                                      maintained during the
                                                                                             No association with      whole pregnancy to
                                                                                             adverse maternal or      prevent the risk of
                                                                                             neonatal outcomes.       abortion during the first
                                                                                                                      trimester and the
                                                                                             One prospective study in development of
                                                                                             72 women with PCOS       gestational diabetes.
                                                                                             who conceived on
                                                                                             metformin found
                                                                                             treatment with metformin
                                                                                             was safely associated
                                                                                             with reduction in
                                                                                             spontaneous abortion
                                                                                             and in gestational
                                                                                             diabetes, was not
                                                                                             teratogenic and did not
                                                                                             adversely affect birth
                                                                                             length and weight,
                                                                                             growth or motor-social
                                                                                             development in the first
                                                                                             18mths of life.

                                                                                             Metformin treatment of
                                                                                             pregnant women with
                                                                                             PCOS may reduce
                                                                                             pregnancy
                                                                                             complications.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                         Page 57 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                    Comments
                 Level                 cs                                                      Measures
Gutzin SJ;Kozer Study Type:       471 women         Most women in       Intervention: oral   Follow-up period: There were 3                 There was no significant
E;Magee LA;Feig Systematic        exposed to oral   the studies had     hypoglycaemic                          prospective cohort           difference in the rate of
DS;Koren G;     review - meta-    hypoglycaemic     type II diabetes.   agents               Outcome           studies, 3 retrospective     major malformations or
                analysis          agents (OHAs)     Women with type                          Measures: Major cohort studies, 3 case         neonatal death among
2003                              1344 women        I diabetes and      Comparison: No       and minor         series and 1 case-           women with first-
                Evidence level:   not exposed.      gestational         exposure             malformations     control study. The oral      trimester exposure to
70
                2++                                 diabetes were                            neonatal death    hypoglycaemics used in       OHAs when compared to
                                                    also present in                          Some studies      the studies included         nonexposed women.
                                                    some studies.                            also reported     chlorpropamide (8)
                                                                                             Birthweight       tolbutamide (6)              Because of the
                                                    Country:                                 Gestational age   glibenclamide (4)            weakness of the
                                                                                             at delivery       metformin (5) and            available data this
                                                                                             Minor neonatal    phenformin (3). 6            systematic review cannot
                                                                                             malformations     studies were rated           be relied upon for clinical
                                                                                             Neonatal          ‘poor’ quality, 2 were       decision making.
                                                                                             hypoglycaemia,    ‘fair’ and 2 were ‘good’.
                                                                                             hypocalcemia and
                                                                                             hyperglycaemia    There was no significant
                                                                                             Preeclampsia      difference in the rate of
                                                                                             labour and        major malformations
                                                                                             delivery          between those exposed
                                                                                             complications     to OHA and those not
                                                                                                               exposed (n=10 studies,
                                                                                                               OR 1.05, 95% CI 0.65-
                                                                                                               1.70).

                                                                                                                The test for
                                                                                                                heterogeneity was
                                                                                                                significant (p=0.05)

                                                                                                                The rates of neonatal
                                                                                                                death did not differ
                                                                                                                significantly between the
                                                                                                                two groups (n=6 studies,
                                                                                                                OR 1.16, 95% CI 0.67-
                                                                                                                2.00).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 58 of 441
                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                      Follow-up &         Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                           Outcome                                                                 Comments
                 Level                 cs                                             Measures
Simmons, D.    Study Type:       12                                              Follow-up period: Congenital
                                                             Intervention: Insulin                                               There is no suggestion of
               Systematic        observational               lispro                                malformations:                any risk to the fetus from
2002           review - meta-    studies (294     Country:                       Outcome           12 observational studies      antenatal use of insulin
               analysis          women with T1               Comparison: Insulin Measures:         involved 294 women            lispro.
86
                                 diabetes and 9                                  Congenital        with T1 diabetes, 9
               Evidence level:   women with T2                                   malformations     women with T2
               1+                diabetes                                        HbA1c             diabetes. Among 170
                                 received insulin                                Maternal          women treated with
                                 lispro during                                   hypoglycaemia     insulin lispro there were
                                 embryogenesis)                                  Patient           7 (4.1%) congenital
                                 1 RCT of 42                                     satisfaction      malformations. Among
                                 women with                                                        133 women treated with
                                 GDM (19                                                           regular soluble insulin
                                 insulin lispro,                                                   there were 14 (10.5%)
                                 23 regular                                                        congenital
                                 insulin)                                                          malformations.

                                                                                                    Other outcomes:
                                                                                                    An RCT compared 19
                                                                                                    women with GDM
                                                                                                    receiving insulin lispro
                                                                                                    with 23 women with
                                                                                                    GDM receiving regular
                                                                                                    insulin. No insulin lispro
                                                                                                    was detected in the
                                                                                                    umbilical cord blood .
                                                                                                    Antibodies to regular
                                                                                                    and insulin lispro were
                                                                                                    within the reference
                                                                                                    range (with one
                                                                                                    exception). The women
                                                                                                    receiving insulin lispro
                                                                                                    had significantly lower
                                                                                                    glucose excursions after
                                                                                                    a test meal and
                                                                                                    experienced fewer
                                                                                                    episodes of
                                                                                                    hypoglycaemia before
                                                                                                    breakfast. They also
                                                                                                    experienced fewer
                                                                                                    hyperglycaemic
                                                                                                    episodes. Obstetric and




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 59 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         fetal outcomes were
                                                                                         similar.
                                                                                         Data from observational
                                                                                         studies also suggests
                                                                                         improved glycaemic
                                                                                         control (3 studies), fewer
                                                                                         maternal hypoglycaemic
                                                                                         episodes (2 studies) and
                                                                                         greater maternal
                                                                                         satisfaction (1 study)
                                                                                         with insulin lispro
                                                                                         compared with regular
                                                                                         insulin.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 60 of 441
                                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &        Effect Size            Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                              Comments
                 Level                 cs                                                        Measures
Cypryk          Study Type:       25 pregnant       All patients were                     Follow-up period: There was no significant
                                                                        Intervention: Insulin                                           The course of pregnancy
K;Sobczak       Cohort            women with        treated with the    lispro                              differences in outcomes     and pregnancy outcome
M;Pertynska-                      type 1 diabetes   same type of                          Outcome           between the two groups.     is comparable in
Marczewska      Evidence level:   taking insulin    insulin as before   Comparison: Human Measures:                                     intensively treated
M;Zawodniak-    2++               lispro            conception. NPH     insulin           HbA1c             HbA1c                       diabetic women
Szalapska                         (Humalog)         or ultralente was                     Pregnancy         First trimester:            regardless of the short-
M;Szymczak                        46 pregnant       used in both                          duration          Lispro: 7.8±1.4%            acting insulin used.
W;Wilczynski                      women with        groups as long                        Caesarean         Regular                     Humalog appears to be a
J;Lewinski A;                     type 1 diabetes   acting insulin.                       section           human:7.5±1.5%              safe alternative to human
                                  taking regular                                          Birthweight       Second trimester:           insulin in the treatment of
2004 Feb                          insulin (n=46)    Country: Poland                       Neonatal          lispro: 6.4±0.8%            diabetes during
                                                                                          hypoglycaemia     Regular human:              pregnancy.
91
                                                                                          Hypertension      6.5±1.6%
                                                                                          Urinary tract     Third trimester
                                                                                          infections        Lispro: 6.7±0.7%
                                                                                                            Regular human:
                                                                                                            6.3±1.2%

                                                                                                              One malformation in the
                                                                                                              human insulin group




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 61 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &           Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                               Comments
                 Level                 cs                                                    Measures
Jovanovic, L.,   Study Type:       27 women with Age 29.7±6.9      Intervention: Insulin   Follow-up period:Mean reductions in         The safety and
Howard, C.,      RCT               GDM           years, HbA1c      Aspart                  From the         HbA1c:                     effectiveness of Insulin
Pettitt, D.                                      <7% at diagnosis.                         diagnosis of GDM Insulin aspart; 0.3±0.5%   Aspart was comparable
                 Evidence level:                                   Comparison: Human       (18-28 weeks) to Human                      to Human Insulin in
2005             1+                              Country: USA      Insulin                 6 weeks          insulin:0.1±0.4%, NS       pregnant women. The
                                                                                           postpartum       Upper respiratoy tract     ease of use of Insulin
79
                                                                                                            infection:                 Aspart injected just
                                                                                           Outcome          Insulin aspart: 14%        before meals rather than
                                                                                           Measures: Mean Human insulin:23%            30 minutes prior to meals
                                                                                           reductions in    Hypoglycaemic events:      may offer a more
                                                                                           HbA1c            Insulin aspart: 0          convenient therapy for
                                                                                           Hypoglyaemic     Human insulin: 0           management of diabetes
                                                                                           events           Average insulin-specific   for patients with GDM
                                                                                           Insulin specific antibodies
                                                                                           antibodies       Insulin aspart:0.97%
                                                                                           Birthweight      Human insulin: 0.07%
                                                                                                            Average percentile
                                                                                                            birthweights:
                                                                                                            Insulin aspart:40%
                                                                                                            Human insulin:44%
                                                                                                            Overall safety profiles
                                                                                                            were similar for Insulin
                                                                                                            aspart and Human
                                                                                                            insulin treatment groups




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 62 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                      Follow-up &        Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                           Outcome                                                                Comments
                 Level                 cs                                                             Measures
Kinsley, B., et al   Study Type:       20 women with    Insulin regimens                       Follow-up period: HbA1c at booking
                                                                             Intervention: Insulin                                             Glycaemic control was
                     Cohort            t1 diabetes      were 4 times daily   aspart                              Insulin aspart:7.0±1.0%       better in the insulin
2005                                   (10 insulin      basal NPH and 3                        Outcome           Human                         aspart group at booking
                     Evidence level:   aspart, 10       times daily short    Comparison: Human Measures:         insulin:8.6±1.0%, p<0.05      and through pregnancy.
80
                     2+                human insulin)   acting bolus         insulin           Gylcaemic control                               FPG and pre-lunch
                                                        insulins                                                 HbA1c at delivery             capillary glucose reading
                                                                                                                 Insulin Aspart: 5.8±0.8%      were lower in the ASP
                                                        Country: Ireland                                         Human Insulin                 group. These data
                                                                                                                 6.7±0.7%, p<0.05              suggest that the use of
                                                                                                                                               insulin aspart as part of a
                                                                                                                   Postprandial glucose        basal bolus regimen is as
                                                                                                                   levels                      effective as short acting
                                                                                                                   No difference               human insulin in
                                                                                                                                               pregnancy in T1
                                                                                                                   Fasting blood glucose       diabetes.
                                                                                                                   Insulin Aspart:4.3±1.4
                                                                                                                   Human Insulin:5.4±2.0,
                                                                                                                   p<0.01

                                                                                                                   Pre-lunch blood glucose
                                                                                                                   Insulin Aspart: 4.5±1.7
                                                                                                                   Human Insulin:5.1±2.3
                                                                                                                   mmol/l, p<0.01

                                                                                                                   Birthweight
                                                                                                                   Insulin aspart: 3.4±0.5kg
                                                                                                                   Human
                                                                                                                   insulin:3.89±0.7kg,
                                                                                                                   p=0.13




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 63 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                     Follow-up &        Effect Size            Study Summary             Reviewer
 Information & Evidence   Patients Characteristi Comparison                                          Outcome                                                           Comments
                 Level                 cs                                                            Measures
Mathiesen et al    Study Type:       157 women          Type 1 diabetes,                      Follow-up period: Severe night-time
                                                                            Intervention: Insulin                                            The risk of
                   RCT               treated with       age 19-34 years,    aspart                              hypoglycaemia RR 0.48,       hypoglycaemia tended to
2006                                 Insulin aspart     HbA1c ≤ 8%at                          Outcome           95% CI 0.2-1.14),            be lower in the insulin
                   Evidence level:   165 treated with   confirmation of     Comparison: Human Measures:         p=0.10.                      aspart group with a
78
                   1++               human insulin      pregnancy. Meal     insulin           Severe                                         strong trend towards
                                                        time insulin in a                     hypoglycaemia     Observed rates of            fewer nocturnal events
See also Hodd et                                        basal bolus                           (requiring        severe hypoglycaemia
al 2006 (33656)                                         regimen with                          assistance)       (per yr)
                                                        NPH.                                                    Insulin aspart: 1.4
                                                                                                                Human insulin: 2.1
                                                        Country: Ireland,                                       RR 0.72, 95% CI 0.36-
                                                        Spain, United                                           1.46) p=0.36
                                                        Kingdom,
                                                        Denmark                                                   Prandial increments
                                                                                                                  (mean: breakfast, lunch
                                                                                                                  and dinner)
                                                                                                                  1st trimester:
                                                                                                                  Insulin aspart:0.73
                                                                                                                  Human insulin:1.49 mM,
                                                                                                                  p<0.05
                                                                                                                  3rd trimester:
                                                                                                                  Insulin aspart:1.10
                                                                                                                  Human insulin:1.50 mM,
                                                                                                                  p<0.05

                                                                                                                  24hr mean PG
                                                                                                                  No difference

                                                                                                                  HbA1c
                                                                                                                  2nd trimester:
                                                                                                                  Insulin aspart:6.0%
                                                                                                                  Human insulin:6.04%
                                                                                                                  3rd trimester:
                                                                                                                  Insulin aspart:6.18%
                                                                                                                  Human insulin:6.26%

                                                                                                                  No significant progress
                                                                                                                  of retinopathy

                                                                                                                  Perinatal complications
                                                                                                                  were similar with either
                                                                                                                  insulin aspart or human




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 64 of 441
                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size   Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                               Comments
                 Level                 cs                                   Measures
                                                                                         insulin




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                         Page 65 of 441
                                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &        Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                               Comments
                 Level                 cs                                                        Measures
BALAJI, V. et al.   Study Type:       5 women with                                        Follow-up period: There was no
                                                                        Intervention: Insulin                                             Both insulin aspart and
                    RCT               GDM given                         aspart                              statistically significant     human insulin achieved
2005                                  insulin aspart   Country: India                     Outcome           difference between the        satisfactory glycaemic
                    Evidence level:   5 women with                      Comparison: Human Measures: FPG two groups in the mean            control and good fetal
81
                    1+                GDM given                         insulin           2 hr PG           plasma glucose level or       outcome. Insulin aspart is
                                      human insulin.                                      Insulin dose      insulin dose at entry or      safe during pregnancy.
                                                                                          Fetal and         before confinement.           Pregnant women were
                                                                                          maternal adverse                                comfortable with insulin
                                                                                          outcomes          No adverse event in the       aspart as it is being
                                                                                          Birthweight       maternal and fetal            administered just before
                                                                                                            outcome of these two          a meal. Insulin aspart
                                                                                                            groups.                       can be safely tried out in
                                                                                                                                          a larger trial.
                                                                                                              No statistical difference
                                                                                                              in mean birthweight
                                                                                                              between the two groups.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 66 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &        Effect Size               Study Summary             Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                              Comments
                 Level                 cs                                                        Measures
Hod, M. et al   Study Type:       322 women         Age 19-43 years.                      Follow-up period: The risk of major
                                                                        Intervention: Insulin                                              Treatment with insulin
                RCT               with type 1       HbA1c <8% at        aspart                              hypoglycaemia was 28%          aspart was at least as
2006                              diabetes          confirmation of                       Outcome           lower for insulin aspart       safe as Human Insulin
                Evidence level:   Insulin aspart:   pregnancy. Meal     Comparison: Human Measures:         than human insulin and         regarding perinatal
404
                1++               157               rime basal-bolus    insulin                             even lower at night, with      outcomes
                                  Human insulin:    regimen with                                            similar HbA1c values.
                                  165               NPH.                                                    Overall 268 live births
                                                                                                            and 35 fetal losses:
                                                    Country: Israel,                                        Insulin aspart: 14
                                                    Netherlands,                                            Human insulin: 21
                                                    Finland, Ireland,
                                                    Denmark                                                   No difference in fetal
                                                                                                              mortality or major
                                                                                                              congenital
                                                                                                              malformations was
                                                                                                              reported between the
                                                                                                              two groups. Neonatal
                                                                                                              hypoglycaemia leading
                                                                                                              to treatment was
                                                                                                              reported in 98 of the live
                                                                                                              births:
                                                                                                              Insulin aspart: 46
                                                                                                              Human insulin: 52

                                                                                                              Mean birth weight (g)
                                                                                                              corrected for gestational
                                                                                                              age:
                                                                                                              Insulin aspart: 3438±72
                                                                                                              Human insulin: 3555±73

                                                                                                              Median gestational age:
                                                                                                              Insulin aspart: 37.9
                                                                                                              Human insulin: 37.7

                                                                                                              Emergency caesarean
                                                                                                              Insulin aspart: 15.9%
                                                                                                              Human insulin: 12.7%

                                                                                                              Preeclampsia
                                                                                                              Insulin aspart: 8.3%
                                                                                                              Human insulin: 6.7%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 67 of 441
                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size      Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                  Comments
                 Level                 cs                                   Measures
                                                                                         Imminent premature
                                                                                         labour:
                                                                                         Insulin aspart:3.8%
                                                                                         Human insulin: 4.2%

                                                                                         Premature delivery:
                                                                                         Insulin aspart: 20%
                                                                                         Human insulin: 30%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                            Page 68 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                  Follow-up &         Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                       Outcome                                                                 Comments
                 Level                 cs                                                         Measures
Poyhonen-alho,   Study Type:       47 pregnant      Women in the                             Follow-up period: There was no difference
                                                                         Intervention: Insulin                                               Treatment with glargine
M. et al         Cohort            women with t1    intervention and     glargine                                for gestational age at      insulin during pregnancy
                                   diabetes using   control group                            Outcome             delivery, pregnancy         seems to be as safe and
2006             Evidence level:   glargine         were matched for     Comparison: Human Measures:             complications and           efficient as with
                 2++               50 pregnant      age, parity,         long acting insulin Gestational age     perinatal outcomes          conventional protophan
92
                                   women with t1    duration of                              at delivery                                     insulin.
                                   diabetes using   diabetes and                             pregnancy           One newborn (Glargine
                                   Human long       diabetes                                 complications,      group) had congenital       There is a tendency
                                   acting insulin   complications. All                       birthweight         malformations               towards better diabetic
                                   (Protophan)      women injected                           shoulder dystocia (anencephaly)                 control during pregnancy
                                                    short-acting                             respiratory                                     with glargine insulin
                                                    insulin analogues                        distress            None of the women           treatment without risk for
                                                    before meals.                            infections          experienced significant     hypoglycaemic events.
                                                                                             first plasma        progression of              Larger RCTs are needed
                                                    Country: Finland                         glucose after birth retinopathy or              to confirm these results.
                                                                                                                 nephropathy

                                                                                                                Although HbA1c was
                                                                                                                similar at first trimester
                                                                                                                of pregnancy (6.9% in
                                                                                                                protaphan group, 7.4%
                                                                                                                in Glargine group) the
                                                                                                                decrease from first
                                                                                                                trimester to third
                                                                                                                trimester was greater in
                                                                                                                the glargine group
                                                                                                                (p=0.04).

                                                                                                                There was a tendency
                                                                                                                towards lower number of
                                                                                                                hypoglycaemic episodes
                                                                                                                in the glargine group
                                                                                                                (11/47) than in the
                                                                                                                protophan group (21/50,
                                                                                                                p=0.07).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                Page 69 of 441
                                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                      Follow-up &             Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                           Outcome                                                                     Comments
                 Level                 cs                                                             Measures
Mecacci, F., et al   Study Type:       25 women with    Diagnosed           Intervention: Insulin   Follow-up period: 1 hr postprandial blood       In women with GDM the
                     RCT               GDM received     according to C&C    lispro                                     glucose values (Total)       use of insulin lispro
2003                                   insulin lispro   criteria. Three                             Outcome            Insulin lispro:108.4±10.7    enabled the attainment of
                     Evidence level:   24 women with    main meals at       Comparison:             Measures: Blood Regular                         near-normal glucose
84
                     1++               GDM received     8.00am, 12.00,      Regular insulin         glucose profiles   insulin:121.0±13.2           levels at the 1-hr
                                       normal insulin   8.00pm.Insulin                              performed using a Controls:105.6±4.7,           postprandial time point
                                       50 non GDM       given before each                           reflectance meter p<0.01                        and was associated with
                                       controls         meal. No NPH                                and consisted of                                normal anthropometric
                                                        insulin.                                    9 determinations: Cranial-thoracic              characteristics; the use of
                                                                                                    fasting/preprandia circumference (CC/CT)        regular insulin was not
                                                        Country: Italy                              l, 1 and 2 hour    ratio (percentile rank, n,   able to blunt the 1 hr
                                                                                                    post-prandial.     %)                           postprandial response to
                                                                                                    HbA1c              <10                          a near-normal extent and
                                                                                                    Birthweight        Lispro:1(4)                  resulted in infants with a
                                                                                                    Ponderal index     Regular:2(8.3)               tendency toward
                                                                                                    Cranio-thoracic    Control:1(2)                 disproportionate growth.
                                                                                                    circumference      10-25                        Insulin lispro can be
                                                                                                                       Lispro:3 (12)                considered as a valuable
                                                                                                                       Regular:9 (37.5)             option for the treatment
                                                                                                                       Control:7 (14), L,C vs R     of gestational diabetes.
                                                                                                                       <0.05
                                                                                                                       26-50
                                                                                                                       Lispro:9 (36)
                                                                                                                       Regular:6 (25)
                                                                                                                       Control:19 (38)
                                                                                                                       51-75
                                                                                                                       Lispro:8 (32)
                                                                                                                       Regular:4 (16.7)
                                                                                                                       Control:17 (34)
                                                                                                                       76-90
                                                                                                                       Lispro:4 (16)
                                                                                                                       Regular:3 (12.5)
                                                                                                                       Control:5 (10)
                                                                                                                       >90
                                                                                                                       Lispro:-
                                                                                                                       Regular:-
                                                                                                                       Control:1 (2)



                                                                                                                       No other significant
                                                                                                                       differences between




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                       Page 70 of 441
                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size     Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                 Comments
                 Level                 cs                                   Measures
                                                                                         groups in neonatal
                                                                                         outcome.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                           Page 71 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                   Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                        Outcome                                                                  Comments
                 Level                 cs                                                          Measures
Gilbert C;Valois   Study Type:       Eight studies   Pregnant women     Intervention:            Follow-up period: Eight studies were          There is no evidence of
M;Koren G;         Systematic        were included   with diabetes or   Exposure to                                included in the meta-       an increased risk for
                   review - meta-    in the meta-    polycystic ovary   metformin in the first   Outcome           analysis.                   major malformations
2006               analysis          analysis        syndrome treated   trimester of             Measures:                                     when metformin is taken
                                                     with metformin.    pregnancy.               Protective effect Metformin treatment in      during the first trimester
72
                   Evidence level:                                                               against           the first trimester was     of pregnancy.
                   2+                                Country:           Comparison:              malformation.     associated with a
                                                                        Pregnant women                             protective effect of 57%.
                                                                        not receiving
                                                                        metformin.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 72 of 441
                                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &             Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                                    Comments
                 Level                 cs                                                 Measures
Plank              Study Type:       7933   People with          Intervention: Short-   Follow-up period: The weighted mean             The systematic review
J;Siebenhofer      Systematic               diabetes who         acting insulin                            difference between           suggests that only a
A;Berghold         review - meta-           were treated with    analogues (lispro      Outcome            HbA1c values obtained        minor benefit to HbA1c
A;Jeitler          analysis                 the currently        and aspart).           Measures:          using short-acting insulin   values in adults with type
K;Horvath K;Mrak                            available short                             1. Glycaemic       analogues and regular        1 diabetes but no benefit
P;Pieber TR;       Evidence level:          acting insulin       Comparison:            control            insulin was –0.12%           in the remaining
                   1++                      analogues lispro     Regular insulin.       2. Hypoglycaemic (95% CI –0.17% to –            population with type 2 or
2005                                        or aspart vs                                episodes           0.07%) for adults with       gestational diabetes from
                                            human regular                               3. Quality of life type 1 diabetes and –        short-acting insulin
83
                                            insulin regardless                          4. Diabetes-       0.02% (95% CI –0.10%         analogue treatment.
                                            of dose or                                  specific           to 0.07%) for patients
                                            schedule and                                complications      with type 2 diabetes
                                            whether insulin                                                mellitus.
                                            was injected
                                            subcutaneously                                                 The standardised mean
                                            via syringe, pen,                                              difference for overall
                                            or pump.                                                       hypoglycaemia
                                                                                                           (episodes per person
                                            Country:                                                       per month) was –0.05
                                                                                                           (95% CI –0.22 to 0.11)
                                                                                                           and –0.04 (95% CI –
                                                                                                           0.12 to 0.04) comparing
                                                                                                           short-acting insulin
                                                                                                           analogues to regular
                                                                                                           insulin in adults with
                                                                                                           type 1 and type 2
                                                                                                           diabetes, respectively.

                                                                                                           No differences between
                                                                                                           treatments were
                                                                                                           observed in children with
                                                                                                           type 1 diabetes,
                                                                                                           pregnant women with
                                                                                                           type 1 diabetes, and
                                                                                                           women with gestational
                                                                                                           diabetes.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 73 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &             Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                                   Comments
                 Level                 cs                                                        Measures
Persson B;Swahn   Study Type:       33 pregnant      33 pregnant        Intervention: Short- Follow-up period:     Blood glucose was           The results suggest that
ML;Hjertberg      RCT               women with       women with type    acting insulin (lispro)                    significantly lower         it is possible to achieve
R;Hanson U;Nord                     type 1 diabetes. 1 diabetes.                                Outcome            (p<0.01) after breakfast    at least as adequate
E;Nordlander      Evidence level:                                       Comparison:             Measures:          in the lispro group, but    glycaemic control with
E;Hansson LO;     1+                1. lispro insulin   Country: Sweden Regular insulin         1. Glycaemic       there were no significant   lispro as with regular
                                    (n=16)                                                      control            group differences in        insulin therapy in type 1
2002 Nov                            2. regular                                                  2. Hypoglycaemia   glycaemic control during    diabetic pregnancies.
                                    insulin (n=17)                                              3. Perinatal       the rest of the day.
85
                                                                                                mortality
                                                                                                                   HbA1c values at
                                                                                                                   inclusion were 6.5% and
                                                                                                                   6.6% in the lispro and
                                                                                                                   regular insulin groups,
                                                                                                                   respectively. HbA1c
                                                                                                                   values declined during
                                                                                                                   the study period and
                                                                                                                   were similar in both
                                                                                                                   groups.

                                                                                                                   There was no perinatal
                                                                                                                   mortality or
                                                                                                                   complications during
                                                                                                                   pregnancy, route of
                                                                                                                   delivery and fetal
                                                                                                                   outcome did not differ
                                                                                                                   between the groups.

                                                                                                                   Retinopathy progression
                                                                                                                   was similar in both
                                                                                                                   groups.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 74 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type              Aim of Study            Number of      Population          Outcome            Results & Comments                 Study           Reviewer
 Information & Evidence                                       Patients &    Characteristic       measures                                             Summary          Comment
                 Level                                          Patient          s
                                                             Characteristic
                                                                   s
Masson, J.,        Study Type:       Intervention: Insulin   76 women with                   Miscarriage           There were 6/76 first trimester   This is the
Patmore, P.,       Other case        lispro                  type 1 diabetes                 Congenital            miscarriages. All 71 babies       largest case
Brash, P. et al.   series                                                      Country: UK   malformation          were born live. There were 4      series of
                                     Comparison:                                             Perinatal mortality   (5.6%) congenital                 pregnancies
2003               Evidence Level:                                                                                 malformations (2 in mothers       treated with
                   3                                                                                               exposed to insulin lispro after   insulin lispro.
89
                                                                                                                   the period of embryogenesis).     An excellent
                                                                                                                   No women developed                perinatal
                                                                                                                   retinopathy de novo during        outcome has
                                                                                                                   pregnancy.                        been
                                                                                                                                                     documented.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 75 of 441
                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type              Aim of Study            Number of      Population          Outcome           Results & Comments               Study            Reviewer
 Information & Evidence                                       Patients &    Characteristic       measures                                          Summary           Comment
                 Level                                          Patient          s
                                                             Characteristic
                                                                   s
Garg, S., Frias, J., Study Type:     Intervention: Insulin   62 women with                    HbA1c               Mean HbA1c was reduced          Insulin lispro
Anil, S., et al      Other case      lispro                  type 1 diabetes                  Progression of      from 7.2±0.2% at the time of    therapy during
                     series                                                    Country: USA   retinopathy         conception to 5.8±0.1% at the   pregnancy
                                     Comparison:                                              Albumin excretion   time of delivery.               resulted in
                   Evidence Level:                                                            rate                                                normalisation of
88
                   3                                                                          Congenital          Congenital malformations (cleft glycaemic
                                                                                              malformations       lip and palate, sensorineural   control and had
                                                                                                                  hearing loss) occurred in 2/62 no adverse
                                                                                                                  infants (3.2%).                 effects on
                                                                                                                                                  maternal or
                                                                                                                  No significant change was       fetal outcomes
                                                                                                                  found in mean eye grade score
                                                                                                                  for retinopathy or albumin
                                                                                                                  excretion rate.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                      Page 76 of 441
                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type         Aim of Study           Number of      Population             Outcome     Results & Comments                 Study            Reviewer
 Information & Evidence                                 Patients &    Characteristic          measures                                      Summary           Comment
                 Level                                    Patient          s
                                                       Characteristic
                                                             s
Al-Shaikh AA;   Study Type:    Intervention: Insulin   11 women with     Aim was to lower                There were no fetal anomalies.    We conclude
                Other          glargine                type 1 diabetes   fasting BG                      Mean birthweight 3.26kg           from these
2006 Apr                                                                 <5.5mmol and                    (range 2.8-4.3kg). All except     cases that
                Evidence Level: Comparison:                              postprandial <7.15              one patient were discharged       glargine insulin
98
                3                                                        mmol/l.                         within 24 hours.                  is safe and
                                                                                                                                           effective during
                                                                         Country:                        The dose of glargine insulin      pregnancy but
                                                                                                         ranged between 30-80 units        more studies
                                                                                                         per day. Mean first trimester     are needed.
                                                                                                         HbA1c was 9.93% (range 7.8-
                                                                                                         12.4%). Mean third trimester
                                                                                                         HbA1c was 6.54% (range 5.9-
                                                                                                         7.4%).

                                                                                                         Normal spontaneous delivery
                                                                                                         in 7 patients, 3 caesarean
                                                                                                         section due to poor
                                                                                                         progression and fetal distress.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                              Page 77 of 441
                                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type           Aim of Study           Number of      Population               Outcome     Results & Comments              Study            Reviewer
 Information & Evidence                                   Patients &    Characteristic            measures                                   Summary           Comment
                 Level                                      Patient          s
                                                         Characteristic
                                                               s
Caronna S;Cioni   Study Type:    Intervention: Insulin   A pregnant woman Obesity,                           A healthy baby was born by     Both the
F;Dall'Aglio      Other          glargine                with type 1      hypertension and                   caesarean weighing 3.54kg      mother and
E;Arsenio L;                                             diabetes         16 years duration of               with an APGAR score of 9.      newborn
                  Evidence Level: Comparison:                             diabetes. Regimen                                                 resulted in
2006              3                                                       of 3 meal time                                                    perfect health
                                                                          injections and 1                                                  conditions
100
                                                                          injection of glargine                                             confirming the
                                                                          at bedtime                                                        possibility of
                                                                          continued                                                         using glargine
                                                                          throughout                                                        insulin profiles
                                                                          pregnancy and after                                               during
                                                                          childbirth.                                                       pregnancy in
                                                                                                                                            selected cases
                                                                            Country: Italy                                                  with close
                                                                                                                                            monitoring may
                                                                                                                                            exist.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                  Page 78 of 441
                                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type              Aim of Study          Number of      Population              Outcome         Results & Comments               Study            Reviewer
 Information & Evidence                                     Patients &    Characteristic           measures                                        Summary           Comment
                 Level                                        Patient          s
                                                           Characteristic
                                                                 s
Di CG;Volpe        Study Type:     Intervention: Insulin   5 women with type                    Major and minor   Babies were delivered at mean   Insulin glargine
L;Lencioni         Other           glargine                1 diabetes                           malformations     gestational age of 36.6±1.1     does not seem
C;Chatzianagnost                   (Used from                                  Country: Italy   Gestational age   weeks by caesarean section in   to affect
ou K;Cuccuru       Evidence Level: preconception to 6-                                          Mode of birth     4/5 women. Mean weight was      embryo-fetal
I;Ghio A;Benzi     3               12 weeks                                                     Birthweight       3066±898g with one baby         development.
L;Del PS;                          postconception).                                             Neonatal          >4kg. There were no             However the
                                                                                                complications     malformations and no neonatal   small number
2005 Apr                            Comparison:                                                                   complications.                  of women and
                                                                                                                                                  the
95
                                                                                                                                                  discontinuation
                                                                                                                                                  of therapy does
                                                                                                                                                  not allow firm
                                                                                                                                                  conclusions to
                                                                                                                                                  be drawn. This
                                                                                                                                                  observation as
                                                                                                                                                  well as other
                                                                                                                                                  anecdotal
                                                                                                                                                  observations
                                                                                                                                                  emphasizes the
                                                                                                                                                  need for
                                                                                                                                                  properly
                                                                                                                                                  planned
                                                                                                                                                  investigations.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 79 of 441
                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type           Aim of Study    Number of      Population      Outcome     Results & Comments                Study            Reviewer
 Information & Evidence                            Patients &    Characteristic   measures                                     Summary           Comment
                 Level                               Patient          s
                                                  Characteristic
                                                        s
Graves DE;White Study Type:       Intervention:   4 women with                               All four women reported          Insulin glargine
JC;Kirk JK;     Other                             gestational                                successful pregnancy             offers an
                                  Comparison:     diabetes       Country:                    outcomes.                        alterative in
2006            Evidence Level:                                                                                               GDM
                3                                                                            Morning dosing of insulin
97
                                                                                             glargine contributed to target
                                                                                             postprandial readings
                                                                                             throughout the day with no
                                                                                             nocturnal hypoglycaemia
                                                                                             reported. Higher than
                                                                                             recommended starting doses
                                                                                             were required to achieve blood
                                                                                             glucose goals during
                                                                                             pregnancy, especially in obese
                                                                                             women.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                Page 80 of 441
                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type            Aim of Study    Number of      Population               Outcome     Results & Comments                Study            Reviewer
 Information & Evidence                             Patients &    Characteristic            measures                                     Summary           Comment
                 Level                                Patient          s
                                                   Characteristic
                                                         s
Woolderink       Study Type:       Intervention:   7 women with type Mean age 34 (29-                  Glycaemic control was            The present
JM;van Loon      Other                             1 diabetes        39) and a diabetes                excellent in 6 patients (HbA1c   observational
AJ;Storms F;de                     Comparison:                       duration of 12 years              5.2-6.9%) and suboptimal in      study puts
HL;Hoogenberg    Evidence Level:                                     (5-8). Five patients              one (6.4-8.1%). Mean HbA1c       forward the
K;               3                                                   continued their                   was 6.4%. The occurrence of      possibility that
                                                                     preconception use                 hypoglycaemia reduced in the     insulin glargine
2005 Oct                                                             of glargine for the               two patients who converted to    is safe in
                                                                     entire pregnancy.                 glargine.                        pregnancy.
96
                                                                     Two converted to                  All women delivered at term,
                                                                     glargine during                   three vaginally and four by
                                                                     pregnancy because                 caesarean section. Mean
                                                                     of recurrent                      birthweight 4180g.
                                                                     nocturnal                         There were no malformations
                                                                     hypoglycaemia.                    and no neonatal complications.

                                                                      Country:
                                                                      Netherlands




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                          Page 81 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type             Aim of Study           Number of      Population                Outcome              Results & Comments               Study            Reviewer
 Information & Evidence                                     Patients &    Characteristic             measures                                             Summary           Comment
                 Level                                        Patient          s
                                                           Characteristic
                                                                 s
Wyatt JW;Frias     Study Type:     Intervention: Insulin   533 pregnancies in Age 29.9±5.2 years Major congenital        HbA1c at first prenatal visit   The rate of
JL;Hoyme           Other           lispro                  496 women (481                        malformations           was 8.9%±4.2% and               major
HE;Jovanovic                                               T1, 15 T2)         Country:           (according to           decreased to 6.2±2.4% in the    congenital
L;Kaaja R;Brown    Evidence Level: Comparison:                                                   definitions of          third trimester (p<0.001).      malformations
F;Garg S;Lee-      3               Published rates of                                            Spranger et al 1982).                                   was 5.4% (95%
Parritz A;Seely                    major anomalies in                                                                    There were 27 major and 2       CI 3.45%-
EW;Kerr L;Mattoo                   infants born to                                                                       minor malformations             7.44%). The
V;Tan M;IONS                       women with                                                                                                            current
study group.;                      diabetes                                                                                                              published rates
                                                                                                                                                         of major
2005 Jun                                                                                                                                                 anomalies in
                                                                                                                                                         infants born to
90
                                                                                                                                                         mothers with
                                                                                                                                                         diabetes
                                                                                                                                                         treated with
                                                                                                                                                         insulin are
                                                                                                                                                         between 2.1
                                                                                                                                                         and 10.9%.
                                                                                                                                                         This suggests
                                                                                                                                                         that the
                                                                                                                                                         anomalies rate
                                                                                                                                                         with insulin
                                                                                                                                                         lispro treatment
                                                                                                                                                         does not differ
                                                                                                                                                         from the
                                                                                                                                                         published major
                                                                                                                                                         congenital
                                                                                                                                                         malformation
                                                                                                                                                         rates for other
                                                                                                                                                         insulin
                                                                                                                                                         treatments.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 82 of 441
                                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type             Aim of Study           Number of      Population              Outcome     Results & Comments                Study        Reviewer
 Information & Evidence                                     Patients &    Characteristic           measures                                     Summary       Comment
                 Level                                        Patient          s
                                                           Characteristic
                                                                 s
Gallen, I., JAAP,   Study Type:    Intervention: Insulin   115 women T1    Mean maternal age                  HbA1c fell from 8.1%±0.2% at The use of
I.                  Other          Glargine                diabetes        30.5±0.5. Insulin                  booking to 6.7%±0.1% during glargine in
                                                           5 T2 diabetes   glargine was used                  the third trimester.          Type 1
2006                Evidence Level: Comparison:            7 gestational   prior to pregnancy                                               diabetes during
                    3                                      diabetes        in 82 (65%) women,                 Background retinopathy        pregnancy
93
                                                                           started in 44 (35%)                developed in 1, progressed in results in
                                                                           and stopped at                     3 and laser photocoagulation outcomes
                                                                           booking in 1. Insulin              was required in 7 women.      comparable to
                                                                           aspart was the                                                   recently
                                                                           bolus insulin in 58                Hypoglycaemia requiring       published
                                                                           (46%) lispro in 48                 assistance occurred in 9 (7%) studies with no
                                                                           (38%) and human                    with 16 (12%) having ≥2       evidence of
                                                                           soluble in 9 (7%) of               episodes.                     increased
                                                                           women.                                                           congenital
                                                                                                              Pre-eclampsia occurred in 16 malformations
                                                                           Country:                           (12.5%) cases.                or adverse
                                                                                                                                            maternal or
                                                                                                              There were 7 (6%) early       fetal outcomes.
                                                                                                              miscarriages.

                                                                                                              All 122 babies were live born,
                                                                                                              with a mean gestational age of
                                                                                                              37.5 weeks, and a median
                                                                                                              birthweight of 3500g.

                                                                                                              24 (20%) babies weighed ≥4kg

                                                                                                              There were 3 congenital
                                                                                                              malformations (positional
                                                                                                              talipes, ventricular septal
                                                                                                              defect and transposition of the
                                                                                                              great arteries) 2 occurring in
                                                                                                              women taking glargine before
                                                                                                              pregnancy, giving a
                                                                                                              malformation rate of 2.5%.

                                                                                                              There were no neonatal
                                                                                                              deaths.

                                                                                                              5 babies had Apgar scores (1




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 83 of 441
                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type     Aim of Study     Number of      Population      Outcome     Results & Comments              Study     Reviewer
 Information & Evidence                       Patients &    Characteristic   measures                                   Summary    Comment
                 Level                          Patient          s
                                             Characteristic
                                                   s
                                                                                        minute) of <5. Neonatal
                                                                                        hypoglycaemia was seen in 51
                                                                                        (42%) and hyperbilirubinaemia
                                                                                        in 28 (23%) babies. Other
                                                                                        neonatal complications were
                                                                                        seen in <5% of babies.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                          Page 84 of 441
                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                    Follow-up &              Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                         Outcome                                                                     Comments
                 Level                 cs                                           Measures
Price,N.;      Study Type:       64 pregnant   64 pregnant   Intervention: Insulin Follow-up period:   There was no significant    Insulin glargine treatment
Bartlett,C.;   Case-control      women.        women with    glargine during their Not stated          difference between the      during pregnancy does
Gillmer,M.D.                                   diabetes      pregnancy (glargine                       birthweight or centile      not appear to be
               Evidence level:                               group)                Outcome             birthweight of babies       associated with
               2+                              Country: UK                         Measures:           born to the women           increased fetal
                                                             Comparison:           1. Birthweight      treated with insulin        macrosomia or neonatal
94
                                                             Intermediate-acting                       glargine during             morbidity.
                                                             human insulin         2. Centile          pregnancy and those
                                                             (isophane or insulin birthweight          born to the control group
                                                             zinc suspension)                          treated with
                                                                                   3. Gestation at     intermediate-acting
                                                                                   delivery            human insulin.

                                                                                  4. Mode of        The overall incidence of
                                                                                  delivery.         fetal macrosomia was
                                                                                                    12/32 (37.5%) in the
                                                                                  5. Apgar score at insulin glargine group
                                                                                  least 1 and 5     and 13/32 (40.6%) in the
                                                                                  minutes           control group.

                                                                                  6. Congenital        There was no significant
                                                                                  anomalies            difference in neonatal
                                                                                                       morbidity between the
                                                                                  7. Neonatal          groups.
                                                                                  hypoglycaemia
                                                                                                    For type 1 diabetes,
                                                                                  8. Admission to   there was no significant
                                                                                  special care baby difference in the mean
                                                                                  unit              maternal weight gain
                                                                                                    during pregancy
                                                                                                    between the insulin
                                                                                                    glargine and control
                                                                                                    groups.

                                                                                                       For women with
                                                                                                       gestational diabetes the
                                                                                                       mean maternal weight
                                                                                                       gain during pregnancy
                                                                                                       was higher for those on
                                                                                                       insulin glargine than
                                                                                                       control.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                      Page 85 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         There was no significant
                                                                                         difference in the
                                                                                         incidence of daytime or
                                                                                         nocturnal
                                                                                         hypoglycaemia in cases
                                                                                         and controls.

                                                                                         No woman in the insulin
                                                                                         glargine or control
                                                                                         groups experienced
                                                                                         hypoglycaemia following
                                                                                         birth.

                                                                                         Gestational age at
                                                                                         delivery, birthweight,
                                                                                         centile birthweight or
                                                                                         Apgar score in neonates
                                                                                         of women on insulin
                                                                                         glargine or intermediate
                                                                                         insulin group did not
                                                                                         differ significantly.

                                                                                         There was no difference
                                                                                         in the mode of birth
                                                                                         between intervention
                                                                                         and control groups.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 86 of 441
                                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                     Follow-up &              Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                          Outcome                                                                     Comments
                 Level                 cs                                                            Measures
Ekpebegh,C.O.;    Study Type:       379 singleton      Type 2 diabetic      Intervention:          Follow-up period: HbA1c was similar at            The two types of oral
Coetzee,E.J.; van Cohort            pregnancies        singleton            1. Oral                Approximately 16 booking and throughout           hypoglycaemic agents
der,Merwe L.;                       (between 1991-     pregnant women       hypoglycaemic          weeks               pregnancy in all the          used (metformin and
Levitt,N.S.       Evidence level:   2000)              of gestation ≥ 24    agents (OHAs)                              study groups.                 glibenclamide) are not
                  2+                                   weeks. The           2. Converted from      Maternal                                          teratogenic.
2007                                1. Oral            women were           OHAs to insulin        outcomes:           In oral hypoglycaemic
                                    hypoglycaemic      mainly of mixed      3. Insulin alone or    1. Glycaemic        agents alone, converted       However, it is advisable
75
                                    agents (OHAs)      ancestry or black    converted from diet    control measured from oral hypoglycaemic          to replace oral
                                    = 93               Africans aged        alone to insulin       by mean fasting     agents to insulin and         hypoglycaemic agents, in
                                                       32.5-35 years and                           plasma glucose      insulin alone/converted       particular glibenclamide,
                                    2. Converted       delivered at         The oral               (FPG),              from diet alone to insulin    with insulin when women
                                    from oral          Groote Shuur         hypoglycaemic          postprandrial       groups, fetal anomaly         book for pregnancy care
                                    hypoglycaemic      Tertiary Hospital.   agents used were       plasma glucose      rates were similar: 5.7%,     to reduce perinatal
                                    agents to                               metformin and          (PPG), and          2.0% and 0.0%,                mortality.
                                    insulin = 249      Country:             glybenclamide          glycosylated        respectively, p = 0.2,;
                                                       South Africa                                haemoglobin         whereas perinatal
                                    3. Insulin alone                        Comparison:            (HbA1c). These      mortality rates (per 1000
                                    or converted                            1. Oral                were calculated     births) were higher for
                                    from diet alone                         hypoglycaemic          for each trimester. oral hypoglycaemic
                                    to insulin= 37                          agents versus Oral                         agents alone: 125, 28,
                                                                            hypoglycaemic          Fetal outcomes      33, respectively, p =
                                                                            agents converted to measured:              0.003,.
                                                                            insulin                1. Fetal growth
                                                                            2. Oral                2. Fetal anomaly Booking HbA1c was
                                                                            hypoglycaemic          rates               independently
                                                                            agents versus          3. Perinatal        associated with fetal
                                                                            insulin alone or diet- mortality rates     anomaly (OR 1.48, 95%
                                                                            insulin                4. Hypoglycaemia CI 1.11 to 1.97, p =
                                                                            3. Oral                5. Polycythaemia 0.006).
                                                                            hypoglycaemic          6.
                                                                            agents then            Hyperbilirubinaem The specific oral
                                                                            converted to insulin ia                    hypoglycaemic agent
                                                                            versus insulin alone 7. Respiratory        used in the first trimester
                                                                            or diet-insulin        distress            (metformin or
                                                                                                                       glybenclamide) was not
                                                                                                                       associated with the
                                                                                                                       occurrence of fetal
                                                                                                                       anomaly.

                                                                                                                       Last HbA1c was
                                                                                                                       independently




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                        Page 87 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         associated with perinatal
                                                                                         mortality and fetal
                                                                                         anomaly (OR 1.65, 95%
                                                                                         CI 1.16 to 2.42, p =
                                                                                         0.005) and (OR 15.18,
                                                                                         95% CI 2.43 to 93.37, p
                                                                                         = 0.005), respectively.

                                                                                         Conversion from OHA to
                                                                                         insulin was protective for
                                                                                         perinatal mortality
                                                                                         compared to oral
                                                                                         hypoglycaemic agents
                                                                                         alone (OR 0.220, 95%
                                                                                         CI 0.061 to 0.756, p =
                                                                                         0.024).

                                                                                         No perinatal mortality
                                                                                         was observed in women
                                                                                         on metformin alone.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 88 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                   Follow-up &            Effect Size             Study Summary             Reviewer
 Information & Evidence   Patients Characteristi Comparison                                        Outcome                                                                Comments
                 Level                 cs                                                          Measures
Lapolla A;Dalfra    Study Type:       93 women           Women with type     Intervention:       Follow-up period: Women treated with          Insulin pump therapy is
MG;Masin            Cohort                               1 diabetes who      Continuous          Not reported       insulin pump more          useful in problematic,
M;Bruttomesso                         1. Treated with    underwent insulin   subcutaneous                           frequently had             complicated cases of
D;Piva I;Crepaldi   Evidence level:   continuous         treatment during    insulin infusion.   Outcomes:          background retinopathy     diabetes during
C;Tortul C;Dalla    2++               subcutaneous       pregnancy in                                               and clinical neuropathy.   pregnancy.
BB;Fedele D;                          insulin infusion   northeast Italy.   Comparison:          Metabolic:
                                      (n = 25).                             Conventional         1. Fasting and 1- No significant
2003                                                     68 were treated intensive insulin       hour post-prandial differences were
                                      2. Treated with    with conventional therapy.              plasma glucose     observed between the
228
                                      conventional       insulin therapy                         2. HbA1c levels    two groups in metabolic
                                      insulin therapy    (multiple daily                         3. Spontaneous     control and maternal
                                      (n = 68).          injections), and                        or induced         outcome.
                                                         25 with insulin                         abortions
                                                         pump infusion (i.e                      4. Time and mode Progressive reduction in
                                                         continuous                              of delivery        HbA1c levels: indicates
                                                         subcutaneous                                               glycaemic control was
                                                         insulin infusion).                      Maternal:          not optimal in the pre-
                                                                                                 1. Pregnancy-      pregnancy state, but
                                                         Country: Italy                          induced            improved significantly
                                                                                                 hypertension       during pregnancy.
                                                                                                 2. Pre-eclampsia
                                                                                                 3. Placental       In terms of fetal
                                                                                                 insufficiency      outcome, no differences
                                                                                                 4. Hydramnios      were observed between
                                                                                                 5. Hypoglycaemic the two groups in
                                                                                                 coma               morbidity and especially
                                                                                                 6. Ketoacidosis    in malformation rate.

                                                                                                 Fetal:              Women with malformed
                                                                                                 1. Fetal weight     babies did not have
                                                                                                 2. Hypoglycaemia    optimal metabolic
                                                                                                 3. Hypocalcaemia    control at conception.
                                                                                                 4.                  Thus, maternal and
                                                                                                 Hyperbilirubinemi   perinatal outcomes were
                                                                                                 a                   similar in patients
                                                                                                 5. Fetal distress   treated with insulin
                                                                                                 6. Asphyxia         pump and continuous
                                                                                                 7. Hyaline          subcutaneous insulin
                                                                                                 membrane            therapy, and depended
                                                                                                 disease             on metabolic control.
                                                                                                 8. Polycythaemia




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 89 of 441
                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &             Follow-up &        Effect Size   Study Summary          Reviewer
 Information & Evidence   Patients Characteristi Comparison                  Outcome                                               Comments
                 Level                 cs                                    Measures
                                                                           9. Shoulder
                                                                           dystocia
                                                                           10. Malformations




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                          Page 90 of 441
                                                                                                                                                      DRAFT FOR CONSULTATION




Q8 - Which medications for diabetic complications are suitable for use during pregnancy and which should be
discontinued?

Bibliographic Study Type Number of   Patient     Intervention &                                        Follow-up &             Effect Size              Study Summary                      Reviewer
 Information & Evidence   Patients Characteristi Comparison                                             Outcome                                                                           Comments
                 Level                 cs                                                               Measures
Cooper, W.,        Study Type:       209 infants with     Infants enrolled in   Intervention:         Follow-up period: Infants with only first       Exposure to ACE               High quality study
Hernandez-Diaz, cohort               exposure to          Tennessee             Exposure to ACE       1 year old           trimester exposure to      inhibitors during the first   RR were adjusted for all
M., Arbogast,P. et                   ACE inhibitors       Medicaid and          inhibitor in first                         ACE inhibitors had an      trimester cannot be           known confounders. The
al.                Evidence level:   in first trimester   born between          trimester alone       Outcome              increased risk of major    considered safe and           robustness of study
                   2+                alone,               1985 and 2000.                              Measures:            congenital                 should be avoided             definitions was tested by
2006                                 202 infants with     Exclusions:           (Determined by        Presence of a        malformations (RR 2.71;                                  secondary analysis:
                                     exposure to          Maternal              Medicaid pharmacy major congenital 95% CI 1.72-4.27)                                                1) restricting the group
104
                                     other                diabetes,             files, which included malformation not                                                              exposed to ACE inhibitors
                                     antihypertensiv      exposure to           the date the          related to a         Fetal exposure to any                                    to infants whose mothers
                                     e medications        angiotensin -         prescription was      chromosomal          hypertensive                                             filled a prescription for
                                     in the first         reception             filled and the        defect or a clinical medications during the                                   ACE inhibitors14 or more
                                     trimester alone,     antagonists,          number of days for genetic                 first trimester did not                                  days after the last
                                     29096 infants        exposure to           which the medicine syndrome.               confer an increased risk                                 menstrual period (thus
                                     with no              hypertensive          was supplied)                              (RR 0.66, 95% CI 0.25-                                   excluding women who
                                     exposure to          beyond first                                                     1.75)                                                    were likely to have
                                     antihypertensiv      trimester,            Comparison:                                                                                         stopped use of ACE
                                     e drugs at any       exposure to other     Exposure to other                          Infants exposed to ACE                                   inhibitors before
                                     time during          potential             hypertensive in first                      inhibitors were at an                                    conception)
                                     gestation.           teratogens.           trimester alone                            increased risk for                                       2) Using a broader
                                                                                No exposure to                             malformations of the                                     definition of diabetes
                                                                                hypertensive at any                        cardiovascular system                                    (further excluding women
                                                                                time during                                (RR 3.72; 95% CI 1.89-                                   with a single prescription
                                                                                gestation                                  7.30) and the central                                    for a hypoglycaemic
                                                                                                                           nervous system (RR                                       agent or one outpatient
                                                                                                                           4.39; 95% CI 1.37-                                       visit with a diabetes
                                                                                                                           14.02).                                                  diagnosis through the first
                                                                                                                                                                                    trimester)
                                                                                                                                                                                    3) Excluding patent
                                                                                                                                                                                    ductus arterious as a
                                                                                                                                                                                    major congenital
                                                                                                                                                                                    malformation (since this
                                                                                                                                                                                    condition infrequently
                                                                                                                                                                                    persists)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                          Page 91 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type          Aim of Study        Number of      Population                       Outcome      Results & Comments                   Study                Reviewer
 Information & Evidence                               Patients &    Characteristic                    measures                                         Summary               Comment
                 Level                                  Patient          s
                                                     Characteristic
                                                           s
Feldkamp, M.   Study Type:     Intervention: First   66 women with          Women who self        Live birth      Among the 48 live births, there     The number of        It is unknown
               Other           trimester only        ACE inhibitors         reported to ACE                       were 3 cases of intrauterine        exposures            whether first
1997                           exposure to ACE       limited to the first   inhibitor registry.   Miscarriage     growth restriction (IUGR). One      reported so far      trimester
               Evidence Level: inhibitors            trimester of                                                 infant with IUGR was from           to the registry is   exposure to ACE
107
               3                                     pregnancy                                    Congenital      twins delivered at 36 weeks         too small to         inhibitors was
                               Comparison:                                                        malformations   gestation; the other two were       determine            associated with
                                                                                                                  full-term. Another child had        conclusively         the development
                                                                                                                  patent ductus arteriosus (PDA)      that exposure        of IUGR in the
                                                                                                                  that required surgical litigation   to an ACE            three infants in
                                                                                                                  at 18 months. That infant was       inhibitor            this study
                                                                                                                  born at 40 weeks gestation to       exclusively          because other
                                                                                                                  a woman who discontinued            during the first     known risk
                                                                                                                  therapy with an ACE inhibitor       trimester is not     factors were
                                                                                                                  at 71/2weeks gestation. The         associated with      present. Because
                                                                                                                  mother was also treated with        the features of      there were no
                                                                                                                  digoxin throughout her              fetopathy.           controls the rate
                                                                                                                  pregnancy and with warfarin                              of IUGR could
                                                                                                                  sodium for the first 5 weeks        Whenever             not be compared
                                                                                                                  followed by heparin throughout      possible             with that in the
                                                                                                                  the remainder of her                pregnant             unexposed
                                                                                                                  pregnancy.                          women should         group. Register is
                                                                                                                                                      be changed to        voluntary which
                                                                                                                                                      another              may cause
                                                                                                                                                      hypertensive         selection bias.
                                                                                                                                                      medication to
                                                                                                                                                      maintain
                                                                                                                                                      normal blood
                                                                                                                                                      pressure




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 92 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type        Aim of Study         Number of      Population                    Outcome             Results & Comments                Study             Reviewer
 Information & Evidence                              Patients &    Characteristic                 measures                                             Summary            Comment
                 Level                                 Patient          s
                                                    Characteristic
                                                          s
Bar J;Hod      Study Type:    Intervention: First   8 women treated     3 women treated for   Preterm delivery       There were 8 live births (one    No major          Prospective case
M;Merlob P;    Other          trimester exposure    with ACE inhibitors chronic                                      set of twins). Mean length of    malformations     series with
                              to ACE inhibitors                         hypertension, 2       Low birth weight       gestation was 34.5 (±1.6)        were detected.    definitive timing
1997           Evidence Level:                                          women treated by                             weeks and average weight         Two newborns      of exposure.
               3               Comparison:                              diabetic              IUGR                   was 2288 (± 991)g, which may     were growth
108
                                                                        nephropathy, 2 for                           be explained by the twin         retarded and
                                                                        cardiomyopathy        fetal/neonatal renal   pregnancy and the severe         four were born
                                                                        and one treated for   failure                disease in two patients (one     prematurely,
                                                                        systemic lupus                               with diabetic nephropathy and    mainly because
                                                                        erythematosus. All    skull ossification     renal deterioration, one with    of maternal
                                                                        women were taking     abnormalities          severe preeclampsia and          disease.
                                                                        one, two or three                            systemic lupus erythematosus)
                                                                        additional            Major congenital       which resulted in preterm        The results are
                                                                        antihypertensive      malformations          deliveries, IUGR and low birth   reassuring for
                                                                        agents.                                      weight.                          women who
                                                                                                                                                      inadvertently
                                                                                                                     Exposure to ACE inhibitors       become
                                                                                                                     ranged from 2 and 12             pregnant whilst
                                                                                                                     embryonic weeks (mean 7.2        on this drug.
                                                                                                                     weeks).

                                                                                                                     No major malformations were
                                                                                                                     detected in the 9 newborns.
                                                                                                                     Two cases of intra-uterine
                                                                                                                     growth restriction (IUGR) were
                                                                                                                     observed, one of them ended
                                                                                                                     in intra-uterine death, which
                                                                                                                     may be attributed to maternal
                                                                                                                     illness and not to drug effect
                                                                                                                     (renal function deterioration
                                                                                                                     with massive proteinuria in a
                                                                                                                     diabetic nephropathy patient).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 93 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type            Aim of Study           Number of      Population                  Outcome             Results & Comments                    Study              Reviewer
 Information & Evidence                                    Patients &    Characteristic               measures                                                 Summary             Comment
                 Level                                       Patient          s
                                                          Characteristic
                                                                s
Bar J;Chen        Study Type:     Intervention: ACE       24 women with     Maternal age at        Maternal renal        The six months of captopril          The                The stillbirth was
R;Schoenfeld      Other           inhibitor (captopril)   type 1 diabetes   conception ranged      function              treatment combined with an           administration     an intrauterine
A;Orvieto                         for a minimum of 6      and diabetic      from21to 30 years.     Short and long-term   intensive insulin regimen led to     of captopril for   death at 24
R;Yahav J;Ben-    Evidence Level: months prior to         nephropathy       The nullparity rate    (2 year) pregnancy    significantly decreased mean         at least 6         weeks secondary
Rafael Z;Hod M;   3               conception.                               was 71% and the        outcome               proteinuria at conception            months prior to    to early IUGR
                                  Treatment was                             duration of diabetes                         (202±141 mg/day vs.                  pregnancy led
1999 Sep                          discontinued                              was 10 to 23 years.                          1292±656 mg/day, p=0.001).           to a significant   Two infants had
                                  immediately after a                       9 patients were                                                                   decrease in        cerebral palsy
103
                                  missed period and a                       categorised as RF                            HbA1c decreased from 9.8%            proteinuria at     owing to birth
                                  positive pregnancy                        (retinopathy and                             before treatment to 7.9% at          conception.        trauma
                                  test.                                     nephropathy) and                             conception (p=0.01).                 There was a        secondary to
                                                                            15 as class F                                                                     concomitant        LGA
                                   Comparison:                              (nephropathy only).                          Renal function during                improvement in
                                                                            Renal function was                           pregnancy:                           blood glucose
                                                                            normal to mildly                             mean serum creatine and uric         control.
                                                                            impaired in all the                          acid levels increased                Moderately
                                                                            patients. 11                                 significantly in the 3rd trimester   good
                                                                            patients (46%) had                           from the 1st and 2nd                 outcomes. Poor
                                                                            hypertension.                                trimesters (0.82±0.12 to             pregnancy
                                                                                                                         0.94±0.13 mg/dl, p=0.02 and          outcomes
                                                                                                                         3.8±0.7 to 5.4±0.7 mg/dl,            attributed to
                                                                                                                         p=0.0001). No significant            pre-existing
                                                                                                                         changes were noted in the            hypertension.
                                                                                                                         grade of proteinuria from            No apparent
                                                                                                                         conception. Creatine clearance       decline in renal
                                                                                                                         and potassium levels remained        function at 2
                                                                                                                         stable throughout pregnancy.         year follow up
                                                                                                                                                              however all
                                                                                                                         All patients maintained close to     patients had
                                                                                                                         normal glycaemic control             preserved or
                                                                                                                         throughout pregnancy.                only mildly
                                                                                                                                                              decreased
                                                                                                                         Pregnancy outcome:                   renal function
                                                                                                                         Superimposed preeclampsia            at conception. .
                                                                                                                         11(46%)
                                                                                                                         Preterm delivery 4(17%)
                                                                                                                         IUGR 5(21%)
                                                                                                                         Caesarean delivery 15 (62.5%)
                                                                                                                         Hospitalisation in NICU 1
                                                                                                                         (4.2%)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                Page 94 of 441
                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type     Aim of Study     Number of      Population      Outcome     Results & Comments                 Study       Reviewer
 Information & Evidence                       Patients &    Characteristic   measures                                      Summary      Comment
                 Level                          Patient          s
                                             Characteristic
                                                   s
                                                                                        Stillbirth 1 (4.2%)

                                                                                        2 year outcomes:
                                                                                        Significant increase in serum
                                                                                        creatine (>1mg/dl) 0
                                                                                        End stage renal disease 0
                                                                                        Severe disability or late infant
                                                                                        death 2(8.4%)

                                                                                        Prediction of outcome:
                                                                                        Only preexisting hypertension
                                                                                        was a significant factor in
                                                                                        worse pregnancy outcome
                                                                                        (superimposed preeclampsia,
                                                                                        preterm delivery and IUGR,
                                                                                        p=0.0004). No other factors
                                                                                        were significant (serum
                                                                                        creatine, uric acid, urinary
                                                                                        creatine clearance, grade of
                                                                                        proteinuria).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 95 of 441
                                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type             Aim of Study           Number of      Population               Outcome            Results & Comments                Study           Reviewer
 Information & Evidence                                     Patients &    Characteristic            measures                                            Summary          Comment
                 Level                                        Patient          s
                                                           Characteristic
                                                                 s
Hod M;van Dijk    Study Type:     Intervention: ACE        Eight women with   Maternal age at     Renal function      Duration of treatment was >12   Prepregnancy
DJ;Karp           Other           inhibitor (captopril).   diabetic           conception ranged                       mths in 5 patients and 8, 9.and captopril
M;Weintraub                       Treatment was            nephropathy        from 23 to 29 years Glycaemic control   11 mths in the remaining three  treatment that
N;Rabinerson      Evidence Level: discontinued             planning           and the duration of                     patients. At conception the     is discontinued
D;Bar J;Peled     3               immediately              pregnancy          diabetes between    Pregnancy outcome   level of proteinuria was        at conception
Y;Erman A;Boner                   following a missed                          11 and 21 years.                        <500mg/day which was            and is
G;Ovadia J;                       period and a positive                       Three patients were                     significantly lower than at the combined with
                                  pregnancy test.                             diagnosed as class                      beginning of captopril therapy  strict glycaemic
1995 Dec                                                                      RF (retinopathy and                     (range 760-3000, mean 1633±     control in
                                   Comparison:                                nephropathy) and 5                      666mg/day, p<0.001).            preparation for
102
                                                                              patients as class F                                                     a planned
                                                                              (nephropathy only).                     Blood sugar levels were         pregnancy
                                                                                                                      significantly lower at          protects against
                                                                                                                      conception than at entry to the acceleration of
                                                                                                                      intensified programme (mean diabetic
                                                                                                                      HbA1c 9.3% vs. 7.9%)            nephropathy
                                                                                                                                                      during
                                                                                                                      At conception all patients were pregnancy and
                                                                                                                      normotensive.                   results in a
                                                                                                                                                      favourable
                                                                                                                      All patients experienced a      maternal and
                                                                                                                      gradual increase in the amount fetal outcome.
                                                                                                                      of protein excreted during
                                                                                                                      pregnancy (first trimester      ACE inhibitor
                                                                                                                      mean: 593±515mg/day,            treatment
                                                                                                                      second trimester mean:          should be used
                                                                                                                      783±813mg/day, third            in patients with
                                                                                                                      trimester mean: 1000±1185       diabetic
                                                                                                                      mg/day). In only two patients   nephropathy
                                                                                                                      did the amount of excreted      attempting to
                                                                                                                      protein exceed 1000mg/day.      become
                                                                                                                                                      pregnant and
                                                                                                                      Serum creatine, creatine        discontinued at
                                                                                                                      clearance serum, uric acid and conception,
                                                                                                                      K+ remained stable throughout concomitantly
                                                                                                                      the study.                      with intensive
                                                                                                                                                      insulin
                                                                                                                      BP remained normal              management.
                                                                                                                      throughout the study except for
                                                                                                                      an abrupt onset of




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 96 of 441
                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type     Aim of Study     Number of      Population      Outcome     Results & Comments                Study      Reviewer
 Information & Evidence                       Patients &    Characteristic   measures                                     Summary     Comment
                 Level                          Patient          s
                                             Characteristic
                                                   s
                                                                                        preeclampsia toxaemia (PET)
                                                                                        just prior to delivery in three
                                                                                        patients.

                                                                                        All patients had good
                                                                                        glycaemic control.

                                                                                        Outcomes of pregnancy:
                                                                                        1 patient delivered preterm at
                                                                                        35 weeks due to PET. There
                                                                                        were three cases of PET just
                                                                                        prior to delivery that resolved
                                                                                        immediately postpartum. 6
                                                                                        patients were delivered by
                                                                                        caesarean. There was one
                                                                                        SGA infant and one LGA infant
                                                                                        delivered without
                                                                                        complications.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                             Page 97 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &           Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                                  Comments
                 Level                 cs                                                     Measures
Lip,G.Y.;       Study Type:       18 women (19   Pregnant women       Intervention:         Follow-up period: Two women, one with          ACE inhibitors taken at
Churchill,D.;   Cohort            pregnancies)   who conceived        Angiotensin-                            type 1 diabetes and the      the time of conception or
Beevers,M.;                                      while taking         converting enzyme     Outcome           other mitral valve           in early pregnancy do not
Auckett,A.;     Evidence level:                  angiotensin-         (ACE) inhibitor.      Measures:         replacement, had a           lead to adverse
Beevers,D.G.    2++                              converting           Treatment with ACE    1. Gestational    miscarriage (at 7 and 8      outcomes as long as
                                                 enzyme (ACE)         inhibitor was         age at delivery   weeks, respectively).        these drugs are
1999                                             inhibitors and who   stopped at either     2. Birthweight                                 discontinued as soon as
                                                 were seen at the     before or at a mean   3. Apgar score    17 pregnancies               pregnancy is confirmed.
105
                                                 antenatal            gestation of 10.3                       proceeded to live birth
                                                 hypertension         weeks (range 6-25                       with a mean gestational
                                                 clinic between       weeks).                                 age at birth of 34.1
                                                 1980 and1997.                                                weeks (range 28-41
                                                                      Comparison: No                          weeks).
                                                 Country: UK          angiotensin-
                                                                      converting enzyme                       No congenital
                                                                      (ACE) inhibitor.                        abnormalities were
                                                                                                              reported with no evident
                                                                                                              neonatal renal
                                                                                                              dysfunction.

                                                                                                              The mean birthweight
                                                                                                              was 2.58 kg (range
                                                                                                              1.26- 3.82kg); mean
                                                                                                              Apgar scores were
                                                                                                              initially 5.9 (range 2-10)
                                                                                                              and subsequently 8.6.

                                                                                                              Even in the six
                                                                                                              pregnancies in which
                                                                                                              ACE inhibitors were
                                                                                                              continued to more than
                                                                                                              12 weeks (including one
                                                                                                              who continued therapy
                                                                                                              until 25 weeks) there
                                                                                                              were no congenital
                                                                                                              abnormalities or
                                                                                                              neonatal problems.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 98 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &           Effect Size            Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                 Comments
                 Level                 cs                                                      Measures
Steffensen,F.H.;   Study Type:       21 pregnant   Pregnant women       Intervention:        Follow-up period: None of the 21 infants    Given the results, there is
Nielsen,G.L.;      Cohort            women         who had a            angiotensin-                           were stillborn.           no reason to deny
Sorensen,H.T.;                                     prescription for     converting enzyme    Outcome                                     women of childbearing
Olesen,C.;         Evidence level:                 ACE inhibitors       (ACE) inhibitors.    Measures:         One preterm infant born   age these effective
Olsen,J.           2++                             during their first   The drug was         1. Preterm        at 27 weeks’ gestation    drugs, especially in
                                                   trimester between    prescribed at 5-15   delivery          to a mother with          countries where most
1998                                               1991 and 1996.       gestational weeks    2. Still birth    diabetes died later.      pregnancies are planned
                                                                        (median 8 weeks).    3. Gestational                              and contraception
106
                                                   Country:                                  age at delivery   There were no             methods make it possible
                                                   Denmark              Comparison:          4. Birthweight    congenital                to stop treatment when
                                                                                             5. Congenital     malformations.            pregnancy is confirmed.
                                                                                             malformation
                                                                                                               Mean gestational age
                                                                                                               was 38.6 weeks (range
                                                                                                               36-41)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 99 of 441
                                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                      Follow-up &            Effect Size             Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                           Outcome                                                                 Comments
                 Level                 cs                                             Measures
Belfort,M.A.;    Study Type:       9 women   Women with      Intervention:          Follow-up period: Verapamil produced a        Verapamil is efficacious
Anthony,J.;      Cohort                      severe          Calcium antagonist                       statistically significant   and justifies further
Buccimazza,A.;                               gestational     verapamil              Outcome           reduction in mean           investigation.
Davey,D.A.       Evidence level:             proteinuric     intravenously          Measures:         arterial pressure and
                 2+                          hypertension.   infused after plasma    1.               systemic vascular
1990                                                         volume expansion       Haemodynamic      resistance without
                                             Country:        with dextran-70.       response          adversely affecting the
110
                                             South Africa                           2. Adverse fetal  cardiac output.
                                                             Comparison:            effects
                                                                                                      The decrease in blood
                                                                                                      pressure was smooth
                                                                                                      and controlled and was
                                                                                                      associated with an
                                                                                                      insignificant increase in
                                                                                                      heart rate.

                                                                                                       There were no adverse
                                                                                                       fetal effects as
                                                                                                       evidenced by
                                                                                                       cardiotocographic
                                                                                                       monitoring.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 100 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &             Effect Size          Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                                Comments
                 Level                 cs                                                        Measures
Magee,L.A.;        Study Type:       78 women with     Women with first-   Intervention:       Follow-up period:   There was no increase Calcium channel
Schick,B.;         Cohort            first-trimester   trimester           First-trimester                         in major malformation   blockers do not represent
Donnenfeld,A.E.;                     exposure to       exposure to         exposure to calcium Outcome             (2/66=3.0% [calcium     a major teratogenic risk.
Sage,S.R.;         Evidence level:   calcium           calcium channel     channel blockers.   Measures:           channel blockers] vs 0%
Conover,B.;        2++               channel           blockers.                                1. Major           [nonteratogenic
Cook,L.;                             blockers.                             Comparison:         malformation        controls], p=0.27); a
McElhatton,P.R.;                                       Country:            No first-trimester  2. Preterm          fivefold increase was
Schmidt,M.A.;                                                              exposure to calcium delivery            ruled out (baseline 2%,
Koren,G.                                                                   channel blockers.                       alpha = 0.05, beta =
                                                                                                                   0.20).
1996
                                                                                                                   The defects reported
109
                                                                                                                   were attributable to
                                                                                                                   maternal diabetes or
                                                                                                                   coingestion of
                                                                                                                   teratogens.

                                                                                                                   The increase in preterm
                                                                                                                   delivery 28% [calcium
                                                                                                                   channel blockers] vs 9%
                                                                                                                   [nonteratogenic
                                                                                                                   controls], p=0.003),
                                                                                                                   attributed to maternal
                                                                                                                   disease by stepwise
                                                                                                                   regression, was the
                                                                                                                   most important factor
                                                                                                                   responsible for the
                                                                                                                   observed decrease in
                                                                                                                   birth weight (mean -334
                                                                                                                   gm vs nonteratogenic
                                                                                                                   controls, p=0.08).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 101 of 441
                                                                                               DRAFT FOR CONSULTATION




Q 9 - What are the barriers to uptake of preconception care?

Q 10 - When should information be offered to i) women of reproductive age with diabetes? ii) women with
diabetes who are planning a pregnancy?

These two clinical questions were addressed together.

Bibliographic    Study       Aim of Study     Number of      Population      Outcome    Results &    Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures   Comments                   Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                            Page 102 of 441
                                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic        Study         Aim of Study         Number of      Population          Outcome                     Results &                Study Summary Reviewer
 Information         Type &                             Patients &    Characteristic       measures                    Comments                               Comment
                    Evidence                              Patient          s
                      Level                            Characteristic
                                                             s
Casele HL;Laifer   Study Type:   What factors influence 55 women with                  HbA1c - determined        33/55 (60%) of women had       Patients may not be     Limitations:
SA;                Other         preconception control pre-existing                    before conception or      suboptimal control at          sufficiently educated   Small sample
                                 of glycaemia in        diabetes                       in the first trimester.   conception. 6/55(11%) had      about the target        size
1998 Jun 22        Evidence      women with diabetes?                                  Optimal glycaemic         HbA1c within the normal        glucose levels          underpowered
                   Level: 2+                                                           control defined as        range. Women with prior        recommended for         to detect
114
                                                                                       within the normal         poor outcome of pregnancy      preconception           difference.
                                                                                       range at the              were significantly more        control.                Unvalidated
                                                                                       institution (4% to 6%)    likely to enter pregnancy                              questionnaire.
                                                                                       Adequate control          with poor glycaemic control   Recommendations:
                                                                                       defined as between        (p=0.02). There was no        Propose that the
                                                                                       6% and 8%                 difference with regard to     implementation of an
                                                                                       Suboptimal control        other variables (payer        intensive effort to
                                                                                       >8%                       status,                       control blood glucose
                                                                                                                 age,race,weight,parity,vasc   levels in all women
                                                                                                                 ular complications, duration  with diabetes of
                                                                                                                 of diabetes).                 childbearing age,
                                                                                                                                               based on the findings
                                                                                                                 Logistic regression found     of the DCCT, would
                                                                                                                 not being advised to          obviate the need for
                                                                                                                 achieve target glucose        preconception
                                                                                                                 values (from questionnaire regimen changes and
                                                                                                                 responses) was                would result in
                                                                                                                 significantly associated with patients entering
                                                                                                                 entering pregnancy with       pregnancy with better
                                                                                                                 poor glycaemic control        glycaemic control.
                                                                                                                 (P=0.02).There was no
                                                                                                                 difference with regard to
                                                                                                                 other questionnaire
                                                                                                                 responses( Did not see
                                                                                                                 diabetes physician before
                                                                                                                 conception, did not plan
                                                                                                                 pregnancy, did not make
                                                                                                                 changes in glycaemic
                                                                                                                 control to prepare for
                                                                                                                 pregnancy, did not monitor
                                                                                                                 blood glucose, did not
                                                                                                                 follow diet, was not aware
                                                                                                                 of complications of diabetes
                                                                                                                 in pregnancy, was not
                                                                                                                 advised to plan pregnancy,




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 103 of 441
                                                                                                          DRAFT FOR CONSULTATION




Bibliographic    Study       Aim of Study     Number of      Population      Outcome          Results &                Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures         Comments                               Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s
                                                                                        was not advised to see a
                                                                                        physician prior to
                                                                                        pregnancy, was not
                                                                                        advised to change
                                                                                        glycaemic control to
                                                                                        prepare for pregnancy).

                                                                                        Overall 29 women (53%)
                                                                                        stated that they had
                                                                                        planned their pregnancy
                                                                                        but only 12 (22%) saw a
                                                                                        physician before
                                                                                        conception to modify insulin
                                                                                        intake or glycaemic control.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 104 of 441
                                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic       Study         Aim of Study        Number of      Population                Outcome                 Results &               Study Summary Reviewer
 Information        Type &                            Patients &    Characteristic             measures                Comments                              Comment
                   Evidence                             Patient          s
                     Level                           Characteristic
                                                           s
Holing EV;Beyer   Study Type:   Why don’t women      85 women with   85/122 women with     Planned pregnancy      35/85 (41%) of pregnancies   Recommendations:
CS;Brown          Other         with diabetes plan   pre-existing    pre-existing          (defined as a          were planned                 Pregnancy
ZA;Connell FA;                  their pregnancies?   diabetes        diabetes who gave     pregnancy that was     The average SD above the     counselling should
                  Evidence                                           birth at 15 centres   desired before         laboratory mean for          involve the women's
1998 Jun          Level: 2+                                          in Washington         conception and in      glycohaemoglobin at the      partner.
                                                                     State.                which conception       first prenatal visit was
111
                                                                     Nonparticipants       was stopped or         significantly lower in       Data suggests
                                                                     included 13 women     avoided for the        planned than unplanned       degrees of
                                                                     who were excluded     purpose of becoming    pregnancies (3.1 vs. 5.8,    pregnancy planning
                                                                     because of an         pregnant and in        p=0.004).                    rather than a
                                                                     adverse pregnancy     which the woman        Age:                         dichotomous
                                                                     outcome.              stated that she        Planned:31.5                 variable. Some
                                                                                           attempted to achieve   Unplanned:28.3, p=0.003      unplanned
                                                                                           optimal blood          Married:                     pregnancies may not
                                                                                           glucose control        Planned:35(100%)             be unexpected.
                                                                                           before becoming        Unplanned:24(48%),           Therefore pregnancy
                                                                                           pregnant.)             p=0.001                      issues should be
                                                                                                                  Annual income: >$20,000      discussed with all
                                                                                           Medical record         Planned:32(94%)              women of
                                                                                           review - HbA1c         Unplanned:17(40%),           childbearing age.
                                                                                                                  p<0.0001
                                                                                           Self administered      > 12 years education         Couples should be
                                                                                           questionnaire -        Planned:29(83%)              reassured that with
                                                                                           information on         Unplanned:31(63%),           preconception
                                                                                           demographics,          p=0.05                       glucose control
                                                                                           health insurance and Race: White                    women with diabetes
                                                                                           access to health       Planned:32(91%)              can have healthy
                                                                                           care. Includes marital Unplanned:33(66%),           babies.
                                                                                           satisfaction scale.    p=0.007
                                                                                                                  Medical coverage: None
                                                                                           Interview - open       Planned: 0(0)
                                                                                           ended questions to     Unplanned: 14(28%),
                                                                                           elicit responses on a p<0.001
                                                                                           range of topics        Provider seen within 6
                                                                                           potentially related to months before
                                                                                           pregnancy planning contraception for diabetes
                                                                                           behaviour.             care:
                                                                                                                  Planned:33(94%)
                                                                                                                  Unplanned:31(67%),
                                                                                                                  p=0.003




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                       Page 105 of 441
                                                                                                          DRAFT FOR CONSULTATION




Bibliographic    Study       Aim of Study     Number of      Population      Outcome          Results &                Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures         Comments                               Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s
                                                                                        Prior pregnancy with
                                                                                        diabetes:
                                                                                        Unplanned:17(49%)
                                                                                        Planned: 28(56%)

                                                                                        Unplanned pregnancy:
                                                                                        35/50(70%) used
                                                                                        contraception less than half
                                                                                        the time (included 5 women
                                                                                        who though diabetes made
                                                                                        it more difficult to get
                                                                                        pregnant). 35/50 (70%)
                                                                                        said they were very happy
                                                                                        to be pregnant.

                                                                                        Partner relationship:
                                                                                        Women with planned
                                                                                        pregnancies were more
                                                                                        satisfied with their partner
                                                                                        relationship (OR 3.86,
                                                                                        p=0.0002).
                                                                                        Planned pregnancy: 28
                                                                                        (80%) believed their partner
                                                                                        was well informed about
                                                                                        diabetes and pregnancy
                                                                                        issues before the
                                                                                        pregnancy. Many couples
                                                                                        had attended counselling
                                                                                        together. In almost all
                                                                                        cases women expressed a
                                                                                        feeling of being supported.
                                                                                        Unplanned pregnancies:
                                                                                        8(16%) felt their partners
                                                                                        were well informed about
                                                                                        diabetes and pregnancy
                                                                                        before the pregnancy. Most
                                                                                        felt that their partners did
                                                                                        not understand the risks or
                                                                                        the enormity of effort
                                                                                        required to achieve good




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 106 of 441
                                                                                                            DRAFT FOR CONSULTATION




Bibliographic    Study       Aim of Study     Number of      Population      Outcome           Results &               Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures          Comments                              Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s
                                                                                        diabetes control.

                                                                                        Locus of control:
                                                                                        Women unplanned
                                                                                        pregnancies more likely to
                                                                                        have external locus of
                                                                                        control (OR 2.28, p<0.004).
                                                                                        No association between
                                                                                        pregnancy planning and
                                                                                        'internal' or 'chance' locus
                                                                                        of control.

                                                                                        Knowledge:
                                                                                        Aware that should be in
                                                                                        good diabetes control
                                                                                        before pregnancy:
                                                                                        Planned:33(94%)
                                                                                        Unplanned:34(68%)
                                                                                        No knowledge about
                                                                                        diabetes and pregnancy:
                                                                                        Planned:0(0)
                                                                                        Unplanned:8(16)

                                                                                        Advice from health care
                                                                                        provider
                                                                                        Positive, encouraging:
                                                                                        planned:26(75%)
                                                                                        unplanned:7(14%)
                                                                                        Negative/advised not to get
                                                                                        pregnant:
                                                                                        Planned:4(11)
                                                                                        Unplanned:19(38)

                                                                                        Relationship with hc
                                                                                        provider
                                                                                        Positive:
                                                                                        Planned:25(71%)
                                                                                        Unplanned:14(29%)
                                                                                        negative. felt judged or
                                                                                        disliked:




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 107 of 441
                                                                                                       DRAFT FOR CONSULTATION




Bibliographic    Study       Aim of Study     Number of      Population      Outcome         Results &       Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures        Comments                      Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s
                                                                                        Planned:2(6%)
                                                                                        Unplanned:17(35%)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                    Page 108 of 441
                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic       Study         Aim of Study      Number of      Population                   Outcome                  Results &              Study Summary Reviewer
 Information        Type &                          Patients &    Characteristic                measures                 Comments                             Comment
                   Evidence                           Patient          s
                     Level                         Characteristic
                                                         s
Janz NK;Herman    Study Type:   What factors are   57 women who       Preconception care    sociodemographic       Race:                        Providers must treat
WH;Becker         Other         associated with    sought             (PC) group: 53        characterisitcs,       PC: White 100%               every visit with a
MP;Charron-                     seeking pre-       preconception      Type 1 diabetes, 4    medical factors,       AC: White 63.9%, African-    diabetic woman as a
Prochownik        Evidence      conception care?   care, 97 women     type 2 diabetes       knowledge, attitudes   American 35.1%, Other        preconception visit.
D;Shayna          Level: 2+                        who did not seek   Antenatal care (AC)   and health-related     1.0%, p<0.001.               In counselling the
VL;Lesnick                                         pre-conception     group: 79 type 1      behaviours             Living with partner:         benefits of
TG;Jacober                                         care.              diabetes, 18 type 2                          PC: 96.2%                    preconception care
SJ;Fachnie                                                            diabetes                                     AC: 59.8%, p<0.001           should be stressed
JD;Kruger                                                                                                          Not high school graduate     and the support of
DF;Sanfield JA;                                                                                                    PC: 3.6%                     families and friends
                                                                                                                   AC: 22.7% p<0.001            should be elicited.
1995 Feb                                                                                                           Employed:
                                                                                                                   PC: 78.2%
112
                                                                                                                   AC: 41.2%, p<0.001
                                                                                                                   Income <$6000:
                                                                                                                   PC: 2%
                                                                                                                   AC: 12.5%, p<0.01

                                                                                                                   Note: As there were only 4
                                                                                                                   women in the
                                                                                                                   preconception care group
                                                                                                                   with type 2 diabetes these
                                                                                                                   were excluded from the
                                                                                                                   following analysis:
                                                                                                                   Discussed PC with health
                                                                                                                   care provider:
                                                                                                                   PC (Type 1, n=53): 97.6%
                                                                                                                   AC (type 1, n=79): 51.4%
                                                                                                                   AC (Type 2, n=18): 35.7%,
                                                                                                                   p<0.05
                                                                                                                   Health provider encouraged
                                                                                                                   preconception care:
                                                                                                                   PC: 77.4%
                                                                                                                   AC (Type 1): 43.0%
                                                                                                                   AC (type 2): 5.9%, p<0.05
                                                                                                                   Patient perceived very
                                                                                                                   good diabetes control in
                                                                                                                   past 6 months:
                                                                                                                   PC: 26.9%
                                                                                                                   AC (Type 1): 8.9%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 109 of 441
                                                                                                           DRAFT FOR CONSULTATION




Bibliographic    Study       Aim of Study     Number of      Population      Outcome          Results &                Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures         Comments                               Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s
                                                                                        AC (Type 2): 11.8%,
                                                                                        p<0.05

                                                                                        Prior pregnancy with
                                                                                        diabetes:
                                                                                        PC: 52.8%
                                                                                        AC (Type 1): 57.7%
                                                                                        AC (Type 2): 70.6%
                                                                                        Prior neonatal death:
                                                                                        PC: 0.0
                                                                                        AC (Type 1): 4.6%
                                                                                        AC (Type 2): 7.7%

                                                                                        Knowledge, attitudes,
                                                                                        beliefs and behaviours
                                                                                        questionnaire:
                                                                                        Awareness of
                                                                                        preconception care:
                                                                                        PC: n/a
                                                                                        AC (Type 1): 72.7%
                                                                                        AC (Type 2): 58.8%
                                                                                        Knowledge of benefits to
                                                                                        mother (% high):
                                                                                        PC: 94.3%, p<0.05
                                                                                        AC (Type 1): 76.6%
                                                                                        AC (Type 2): 70.6%
                                                                                        Knowledge of benefits to
                                                                                        baby:
                                                                                        PC: 94.3%, p<0.05
                                                                                        AC (Type 1): 73.1%
                                                                                        AC (Type 2): 82.4%
                                                                                        Instrumental social support
                                                                                        score (1-5):
                                                                                        PC:1.3 (p<0.05)
                                                                                        AC (type 1): 1.6
                                                                                        AC (type 2): 1.7
                                                                                        Adherence with clinic visits
                                                                                        (% all/most)
                                                                                        PC:93.9 (p<0.05)
                                                                                        AC (type 1): 80.5%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 110 of 441
                                                                                                            DRAFT FOR CONSULTATION




Bibliographic    Study       Aim of Study     Number of      Population      Outcome           Results &                Study Summary Reviewer
 Information     Type &                       Patients &    Characteristic   measures          Comments                               Comment
                Evidence                        Patient          s
                  Level                      Characteristic
                                                   s
                                                                                        AC (type 2): 92.3%

                                                                                        Significant variables after
                                                                                        logistic regression
                                                                                        (adjusted odds ratios):
                                                                                        Education: 4.81, p=0.01
                                                                                        Living with partner: 11.25,
                                                                                        p=0.01
                                                                                        Diabetes clinic visit in last
                                                                                        year: 8.25, p=0.01
                                                                                        Provider encouraged
                                                                                        preconception care: 3.39,
                                                                                        p=0.02
                                                                                        Adherence with diabetes
                                                                                        regimen: 3.03, p=0.01




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                Page 111 of 441
                                                                                                                              DRAFT FOR CONSULTATION




Bibliographic     Study         Aim of Study         Number of      Population          Outcome                   Results &                Study Summary Reviewer
 Information      Type &                             Patients &    Characteristic       measures                  Comments                               Comment
                 Evidence                              Patient          s
                   Level                            Characteristic
                                                          s
St James        Study Type:   What factors relate to 66 women with                  Use of birth control    Use of birth control           Some women may
PJ;Younger      Other         the family planning    type 1 diabetes                                        Diabetes group: 82%            have misconceptions
MD;Hamilton                   behaviour of diabetic                                 Knowledge about         Nondiabetic group: 88%         about fertility and
BD;Waisbren SE; Evidence      women?                 Comparison: 207                fertility and           Diabetic women were more       contraception.
                Level: 2+                            non diabetic                   contraception tested    likely to use condoms
1993 Dec                                             women                          by a questionnaire      (P=0.05).                      Recommendations:
                                                                                    adapted from            Interview themes:              Because social
113
                                                                                    Kempton and             Belief that oral               support for birth
                                                                                    Foreman.                contraceptives were unsafe     control is important
                                                                                                            for diabetic women             clinicians should
                                                                                    Knowledge about         Belief that diabetes made it   emphasize to parents
                                                                                    diabetes tested by a    difficult to conceive          of young diabetic
                                                                                    questionnaire                                          women the
                                                                                    developed for the       Pregnancies:                   importance of birth
                                                                                    study.                  Women with diabetes:           control. Educational
                                                                                                            17/23 (78%) were               programs for diabetic
                                                                                    Locus of control        unplanned                      women should
                                                                                    measured by the         Nondiabetic: 16/33 (48%)       engender positive
                                                                                    Adult Nowicki-          were unplanned.                attitudes towards
                                                                                    Strickland Internal-                                   birth control use.
                                                                                    External Locus of       Factors associated with        Family and friends
                                                                                    Control Scale.          consistent use of birth        should be involved in
                                                                                                            control                        this process as much
                                                                                    Attitudes and beliefs   Social support for birth       as possible because
                                                                                    regarding               control (mean score/SD)        their attitudes are
                                                                                    contraception, sex,     Always used birth              important to the
                                                                                    and childbearing        control:6.51±0.95              young woman's
                                                                                    were measured by        Others:5.39±1.77, p<0.05       behaviour.
                                                                                    subject ratings on a
                                                                                    7-point scale of        Attitudes about birth
                                                                                    statements adapted      control:
                                                                                    from Geis and           Always use: 19.65±2.45
                                                                                    Gerrard and             Others: 17.03±3.26, p<0.05
                                                                                    Nathanson and
                                                                                    Becker.               (no significant difference
                                                                                                          between groups in scores
                                                                                    Scores for social     for knowledge of family
                                                                                    support and attitudes planning, knowledge of
                                                                                    toward birth control  diabetes, locus of control,
                                                                                    were also calculated. self esteem).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 112 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic     Study         Aim of Study            Number of      Population                    Outcome                    Results &                Study Summary Reviewer
 Information      Type &                                Patients &    Characteristic                 measures                   Comments                               Comment
                 Evidence                                 Patient          s
                   Level                               Characteristic
                                                             s
Harris, K.,     Study Type:   Qualitative study        45 women with     Non diabetic. No        Secondary gain          Women with unplanned            Some women with
Campbell, E.    Other         exploring hypothesis     planned           differences in social   (benefit or advantage   pregnancies were                unplanned pregnancy
                              that women with          pregnancies       class between           from becoming           significantly more likely to    may in fact fall into a
1999            Evidence      unplanned                43 women with     groups.                 pregnant) rated on a    be rated as having a ‘high’     group of ‘semi
                Level: 2+     pregnancies are more     unplanned                                 scale                   chance of secondary gain        planned’ pregnancy
115
                              likely to have           pregnancies                               (high/some/little or    than women with planned         who are indifferent or
                              secondary gain           40 non pregnant                           none).                  pregnancies (x2 =29.41,         at worst ambivalent
                              particularly in the      women                                                             p<0.0001). Examples of          about the idea of
                              domain of partnership.                                             Quality of sexual       secondary gain included         pregnancy.
                                                                                                 partnership             autonomy from parents,
                                                                                                 (measured by the        solidification of a shaky
                                                                                                 self esteem and         relationship, an excuse to
                                                                                                 social support          leave a boring job and
                                                                                                 schedule).              begin a new and important
                                                                                                                         phase in their lives.
                                                                                                 Interview data
                                                                                                                         Women with planned
                                                                                                                         pregnancies were
                                                                                                                         significantly more likely to
                                                                                                                         have a partnership rated
                                                                                                                         high in overall quality (x2 =
                                                                                                                         7.10, p<0.05).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 113 of 441
                                                                                                                                DRAFT FOR CONSULTATION




Bibliographic     Study         Aim of Study           Number of      Population          Outcome                   Results &               Study Summary Reviewer
 Information      Type &                               Patients &    Characteristic       measures                  Comments                              Comment
                 Evidence                                Patient          s
                   Level                              Characteristic
                                                            s
Confidential    Study Type:   To assess the quality   213/222 maternity               Standards set out in    NSF criteria: A
Enquiry into    Other         of maternity services   units in England,               the National Services   preconception clinic should
Maternal and                  for women with          Wales and                       Framework for           be run jointly by the adult
Child Health    Evidence      diabetes in 2002        Northern Ireland                Diabetes                diabetes service and the
                Level: 2+                                                                                     maternity service for
2004                          Comparison: Data                                                                women wishing to become
                              from 1994                                                                       pregnant
32
                                                                                                              Units meeting criteria:
                                                                                                              2002: 17%
                                                                                                              1994: 16%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 114 of 441
                                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic     Study         Aim of Study           Number of      Population                 Outcome                     Results &                 Study Summary Reviewer
 Information      Type &                               Patients &    Characteristic              measures                    Comments                                Comment
                 Evidence                                Patient          s
                   Level                              Characteristic
                                                            s
Barrett, G.,    Study Type:   Do women use           47 nondiabetic   6 in the first         Interview topic guide:   When discussing the              Survey questions
Wellings, K.,   Other         particular concepts or women            trimester, 13 in the   1) Background            circumstances of their           eliciting information
                              terms when                              2nd trimester, 10 in   information              pregnancies, women               on women's
2002            Evidence      discussing                              the 3rd trimester.     2) current pregnancy     tended not to use the terms      circumstances of
                Level: 2+     pregnancy?                              Ages 15-43. 15         situation                planned/unplanned/intende        pregnancy do not rely
116
                                                                      married, 1             3) earliest awareness    d/unintended/wanted/unwa         on the above terms in
                              How do women                            separated, 1           of pregnancy             nted spontaneously. When         isolation. A more
                              understand the terms                    divorced, 9 co-        4) confirming            asked to explain the terms       circumspect use of
                              planned/unplanned/int                   habiting, 21 single.   pregnancy                there was considerable           the term in policy and
                              ended/unintended/wa                     13 already had         5) contraception         variation in understanding.      clinical settings is
                              nted/unwanted?                          children. 21 had       around the time of       Women applied the term           required.
                                                                      been or were about     pregnancy                'planned' only if they had
                                                                      to be in higher        6) feelings about        met four key criteria.
                                                                      education, 14 had      being pregnant           Intending to become
                                                                      been in full time      7) decision about        pregnant and stopping
                                                                      education to the       pregnancy                contraception were not
                                                                      age of 18, 4 were      8) orientation to        sufficient criteria, in
                                                                      studying for           motherhood               themselves, to apply the
                                                                      GNVQs, 8 had left      9) timing of             term; Agreement from their
                                                                      school at 16 or        childbearing             partner and reaching the
                                                                      under and 1 was        10) nature of            right time in terms of
                                                                      still at school.       partnership              lifestyle/life stage were also
                                                                                             11) understanding of     necessary. In contrast
                                                                                             terms                    'unplanned' was a widely
                                                                                             planned/unplanned/i      applied term and covered a
                                                                                             ntended/unintended/      variety of circumstances of
                                                                                             wanted/unwanted.         pregnancy.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 115 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &             Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                                     Comments
                 Level                 cs                                                    Measures
Charron-       Study Type:       80 cases      Young women          Intervention: Theory   Follow-up period: Cases appeared to lack         Having diabetes did not
Prochownik     Case-control      37 controls   (aged 16-23          (Expanded Health                          an understanding of           appear to significantly
D;Sereika                                      years) with type 1   belief model) based    Outcome            critical information that     decrease the risk-taking
SM;Falsetti    Evidence level:                 diabetes of ≥ 1 yr   structured telephone   Measures:          could prevent unplanned       behaviour of young
D;Wang         2-                              duration, no other   interview conducted    Knowledge,         pregnancies and               women. Early sexual
SL;Becker                                      chronic illness,     for 80 cases with      attitudes,         pregnancy-related             activity and some unsafe
D;Jacober                                      not pregnant,        type 1 diabetes        intentions, and    complications. Although       sexual practices may
S;Mansfield                                    English speaking                            behaviours         they scored significantly     increase their risk for an
J;White NH;                                                         Comparison: 37         regarding          higher than the controls      unplanned pregnancy
                                               Country: USA         matched controls       diabetes and       on diabetes-related           that could result in
2006 Oct                                                            without diabetes       reproductive       information, the cases        pregnancy-related
                                                                                           issues, sexuality, had their lowest mean         complications.
117
                                                                                           and contraception average of 59% for the
                                                                                                              diabetes and pregnancy
                                                                                                              score. They did not
                                                                                                              appear to have greater
                                                                                                              protective attitudes
                                                                                                              regarding reproductive
                                                                                                              health issues than the
                                                                                                              control group. The cases
                                                                                                              group felt that they were
                                                                                                              only moderately
                                                                                                              susceptible to becoming
                                                                                                              pregnant and that
                                                                                                              severe complications
                                                                                                              would not happen to
                                                                                                              them. The cases group
                                                                                                              perceived greater
                                                                                                              severity to sex-related
                                                                                                              outcomes (p = 0.001).
                                                                                                              The cases group did not
                                                                                                              report safer and more
                                                                                                              effective family-planning
                                                                                                              behaviours (mean age
                                                                                                              at first coitus = 15.7 yr),
                                                                                                              which for them could be
                                                                                                              more detrimental.
                                                                                                              Similar trends were
                                                                                                              noted between groups
                                                                                                              regarding contraceptive
                                                                                                              methods; only a single




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 116 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                        Comments
                 Level                 cs                                   Measures
                                                                                         method (e.g., pill only)
                                                                                         rather than a dual
                                                                                         method (e.g., pill and
                                                                                         condom) was most
                                                                                         frequently used.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 117 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type            Aim of Study        Number of      Population                Outcome               Results & Comments              Study           Reviewer
 Information & Evidence                                 Patients &    Characteristic             measures                                             Summary          Comment
                 Level                                    Patient          s
                                                       Characteristic
                                                             s
Charron-          Study Type:     Intervention: A       80           16-23 yr old women      Knowledge,               The response "don't know" or   The study
Prochownik        QualitativeDesc multisite, structured              with type 1 diabetes    attitudes, intentions,   "never heard about it" was     sample lacked
D;Sereika         riptive         1 hr telephone                     for at least 1 yr, no   and behaviours.          most frequently given. Most    awareness of
SM;Wang                           interview with an                  other chronic           Awareness of issues      young women in this sample     pregnancy-
SL;Hannan         Evidence Level: open-ended section.                illnesses or mental     related to diabetes      were unaware of the term       related
MF;Fischl         3                                                  retardation, not        and pregnancy,           preconception counselling      complications
AR;Stewart                        Comparison:                        pregnant, English       preconception                                           with diabetes,
SH;an-McElhinny                                                      speaking                counselling (PC),        65% (n = 52) indicated they    knowledge of
T;                                                                                           and contraception.       knew nothing about PC.         the term and
                                                                     Country: USA                                                                    role of PC in
2006 Mar                                                                                                              Many were not aware of the     preventing
                                                                                                                      risks of pregnancy-related     these
118
                                                                                                                      complications in women with    complications,
                                                                                                                      diabetes.                      and the
                                                                                                                                                     importance for
                                                                                                                      25% were aware of              women with
                                                                                                                      preplanning a pregnancy and diabetes to use
                                                                                                                      the importance of good         a highly
                                                                                                                      metabolic control.             effective birth
                                                                                                                                                     control method
                                                                                                                      Many knew where to seek        for preventing
                                                                                                                      information about diabetes and unplanned
                                                                                                                      pregnancy, and birth control   pregnancy.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 118 of 441
                                                                                                                                DRAFT FOR CONSULTATION




Q11 - Which women are at high risk of gestational diabetes?

Bibliographic Study Type Number of   Patient     Intervention &                 Follow-up &              Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                      Outcome                                                                     Comments
                 Level                 cs                                        Measures
Scott, D.,      Study Type:              Pregnant women   Intervention:        Follow-up period:    135 studies included in     The principal risk factors
Loveman,E.,     Systematic                                Screening for                             the review. In 16 studies   for gestational diabetes
McIntyre, L.,   review - meta-                            gestational diabetes Outcome              all women were given a      are overweight, age and
Waugh, N.       analysis                                                       Measures: GTT        diagnostic test             ethnicity.
                                                          Comparison:          (gold standard for   regardless of screening
2002            Evidence level:                                                diagnosis of         result.                   The author concludes
                2++                                                            gestational                                    that the use of risk
123
                                                                               diabetes)            Risk factors for          factors as a screening
                                                                                                    gestational diabetes are test for gestational
                                                                                                    obesity, advanced         diabetes has led to high
                                                                                                    maternal age, family      proportions (up to 50%)
                                                                                                    history of diabetes, non- of women with
                                                                                                    white ethnic origin,      gestational diabetes
                                                                                                    obesity, increased        being missed.
                                                                                                    weight gain in early
                                                                                                    adulthood and current
                                                                                                    smoker.

                                                                                                    Using risk factors alone
                                                                                                    as a screening test
                                                                                                    produces low
                                                                                                    sensitivities (50-69%) (8
                                                                                                    studies).

                                                                                                    One study found that
                                                                                                    four risk factors (age,
                                                                                                    BMI, ethnic group and
                                                                                                    family history) gave
                                                                                                    most of the information
                                                                                                    and that adding other
                                                                                                    items added little.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 119 of 441
                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type        Aim of Study       Number of      Population              Outcome              Results & Comments                Study            Reviewer
 Information & Evidence                            Patients &    Characteristic           measures                                              Summary           Comment
                 Level                               Patient          s
                                                  Characteristic
                                                        s
Ostlund, I.,   Study Type:    Intervention: Age   4918 pregnant   All pregnant         Gestational diabetes   3616/4918 (73.5%) had an         Prior GDM and
Hanson, U      Other                              nondiabetic     nondiabetic women    Diagnosis made         OGTT performed. Women who        macrosomic
                               Weight             women           in a defined         using 75g OGTT         did not have an OGGT were        infant had the
2003           Evidence Level:                                    geographic area in   according to WHO.      more likely to have had          highest
               2+              BMI                                Sweden               Test offered between   previous births, be of           association with
125
                                                                                       weeks 28-32.           nonNordic origin and have a      GDM. Women
                              NonNordic origin                                         Gestational diabetes   lower rate of traditional risk   below age 25
                                                                                       diagnosed as fasting   factors.                         have a low risk
                              Heredity                                                 blood glucose ≥6.7                                      of GDM.
                                                                                       mmol/l or 2hr blood    Of the 3616 women who had
                              Prior infant                                             glucose ≥9.0 mmol/l.   an OGTT, 61 (1.7%) were      Using
                                                                                                              diagnosed with GDM.(47 IGT,  traditional risk
                              Prior GDM                                                                       14 DM)                       factors such as
                                                                                                                                           heredity
                              Prior macrosomic                                                                Risk factors:                diabetes,
                              infant                                                                          i. Weight                    obesity, prior
                                                                                                              ≥80kg:11.1%, OR 1.98 (1.05- macrosomic
                              Comparison:                                                                     3.77)                        infant or GDM
                                                                                                              ≥90kg: 4.5%, OR 3.33 (1.56- to perform a
                                                                                                              7.13)                        OGTT will
                                                                                                              ≥100kg: 1.8%, OR 4.09 (1.44- identify less
                                                                                                              11.6)                        than half of
                                                                                                              ii. BMI                      women with
                                                                                                              ≥25: 28.3%, OR 2.05 (1.23-   GDM.
                                                                                                              3.41)                        Introducing
                                                                                                              ≥28: 12.3%, OR 1.66 (0.88-   ethnicity
                                                                                                              3.15)                        improves
                                                                                                              ≥30: 7.9%, OR 2.65 (1.36-    sensitivity but
                                                                                                              5.14)                        will markedly
                                                                                                              iii. Age (years)             increase the
                                                                                                              <25: 26.4%, OR 0.30 (0.13-   number of
                                                                                                              0.70)                        women who
                                                                                                              ≥25: 73.6%, OR 3.37 (1.45-   have to perform
                                                                                                              7.85)                        an OGTT. A
                                                                                                              ≥30:33.3%, OR 2.40 (1.45-    prior
                                                                                                              4.00)                        macrosomic
                                                                                                              ≥35:10%, OR 3.03 (1.68-5.49) infant or GDM
                                                                                                              iv. Heredity: 9.4%, OR 2.74  are of




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                 Page 120 of 441
                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type     Aim of Study     Number of      Population      Outcome     Results & Comments               Study            Reviewer
 Information & Evidence                       Patients &    Characteristic   measures                                    Summary           Comment
                 Level                          Patient          s
                                             Characteristic
                                                   s
                                                                                        (1.47-5.11)                      significant
                                                                                        v. Nonnordic origin: 11.2%, OR   predictors of
                                                                                        2.19 (1.18-4.08)                 GDM but are
                                                                                        vi. Prior macrosomia: 3.2%,      restricted to
                                                                                        OR 5.59 (2.68-11.7)              multiparas
                                                                                        vii. Prior GDM: 1.3%, OR 23.6    women.
                                                                                        (11.6-48.0)                      Especially for
                                                                                                                         primiparas
                                                                                        Screening model using            there is a lack
                                                                                        traditional risk factors (family of good risk
                                                                                        history of diabetes, obesity     indicators for
                                                                                        ≥90kg, prior macrosomic infant detecting GDM.
                                                                                        or GDM):
                                                                                        Occurrence: 15.8%
                                                                                        Sensitivity: 47.5% (29/61)
                                                                                        Specificity: 84.7%
                                                                                        PPV: 5.1%
                                                                                        OR:5.02 (3.01-8.36)
                                                                                        Traditional risk
                                                                                        factors+nonNordic origin:
                                                                                        Occurrence: 25.2%
                                                                                        Sensitivity: 60.7% (37/61)
                                                                                        Specificity: 75.4%
                                                                                        PPV: 4.1%
                                                                                        OR 4.74 (2.82-7.96)
                                                                                        Traditional risk
                                                                                        factors+nonNordic
                                                                                        origin+age≥25years
                                                                                        Occurrence: 79.7%
                                                                                        Sensitivity: 57/61 (93.4%)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                            Page 121 of 441
                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                       Follow-up &            Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                            Outcome                                                                    Comments
                 Level                 cs                                              Measures
Doherty          Study Type:       2827   Recruited             Intervention: BMI    Follow-up period: Pre-pregnancy BMI            Pre-pregnancy obesity is
DA;Magann        Cohort                   pregnancies           evaluated                                classified 331 women as    a risk factor for
EF;Francis                                between 16 and        underweight,         Outcome             underweight (11.7%),       gestational diabetes,
J;Morrison       Evidence level:          18 weeks’             BMI<18.5, normal,    Measures:           1982 normal (69.9%),       preeclampsia, labour
JC;Newnham JP;   2+                       gestation.            BMI 18.5-25,         Antenatal and       326 overweight (11.5%),    induction, caesarean
                                          Inclusion criteria:   overweight BMI 25-   intrapartum         and 188 as obese           section for fetal distress,
2006                                      singleton             30, and obese        outcomes:           (6.6%). Obese women        postpartum hemorrhage
                                          pregnancies, pre      BMI>30 women         Gestational         were more likely to        and neonatal
129
                                          pregnancy BMI                              diabetes,           develop gestational        hypoglycaemic and need
                                          available for         Comparison:          hypertension,       diabetes (p<0.001),        for resuscitation.
                                          analysis.                                  preeclampsia,       hypertension (p<0.001),
                                                                                     antenatal           preeclampsia (p<0.001),
                                          Country: USA                               admissions,         need labour induction
                                                                                     induction of        (p<0.001), caesarean
                                                                                     labour, caesarean section for fetal distress
                                                                                     section,            (p<0.001), postpartum
                                                                                     caesarean           hemorrhage (p=0.003),
                                                                                     section for fetal   need neonatal
                                                                                     distress.           resuscitation (p=0.001)
                                                                                     Postpartum and      and deliver
                                                                                     neonatal            hypoglycemic infants
                                                                                     outcomes:           (p=0.007). Being
                                                                                     postpartum          underweight is
                                                                                     hemorrhage, any correlated with fetal
                                                                                     perineal trauma, growth restriction
                                                                                     maternal            (p=0.001).
                                                                                     infection, retained
                                                                                     placenta,
                                                                                     intrauterine
                                                                                     growth
                                                                                     retardation, time
                                                                                     to spontaneous
                                                                                     respiration > 1
                                                                                     min, any
                                                                                     resuscitation and
                                                                                     hypoglycaemia.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                       Page 122 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                           Follow-up &           Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                Outcome                                                               Comments
                 Level                 cs                                                  Measures
Keshavarz        Study Type:       1310   Exclusion criteria:   Intervention: At first   Follow-up period: Incidence of GDM was Screening for GDM is
M;Cheung         Cohort                   twin pregnancies,     antenatal visit urine                        4.8%. There were          recommended in Iran
NW;Babaee                                 miscarriages,         tested for               Outcome             statistically significant
GR;Moghadam      Evidence level:          terminations and      glycosuria. If this      Measures: Fetal, (p<0.001) differences in
HK;Ajami         2+                       women with pre-       was positive or          maternal and        risk factors: age >30
ME;Shariati M;                            existing diabetes     women had risk           neonatal            years, family history of
                                                                factors (age >30 yrs,    outcomes            diabetes, obesity,
2005                                      Country: Iran         obesity, BMI ≥ 30        Macrosomia,         previous macrosomia,
                                                                kg/m2, family history    gestational         glycosuria between the
130
                                                                of diabetes,             hypertension,       two groups. Women with
                                                                previous                 hydramnios,         GDM had a significantly
                                                                macrosomia,              stillbirth,         higher rate of stillbirth
                                                                unexplained              congenital          (p<0.001; OR 17.1, 95%
                                                                recurrent                malformation,       CI 4.5-65.5), hydramnios
                                                                miscarrriages,           pre-eclampsia,      (p<0.001; OR 15.5, 95%
                                                                previous congenital      pylonephritis, pre- CI 4.8-50.5), gestational
                                                                malformations,           term delivery, low hypertension (p<0.001;
                                                                previous stillbirth,     birth weight,       OR 6, 95% CI 2.3-15.3),
                                                                unexplained              intrauterine        macrosomia (p<0.05;
                                                                neonatal death,          growth retardation OR 3.2, 95% CI 1.2-8.6)
                                                                previous gestational                         and caesarean section
                                                                hypertension and                             (p<0.001).
                                                                preeclampsia,
                                                                previous
                                                                hydramnios and
                                                                glycosuria on 2
                                                                successive
                                                                occasions in the
                                                                current pregnancy)
                                                                early screening with
                                                                50 g oral glucose
                                                                challenge test was
                                                                done. If initial
                                                                screening test was
                                                                negative, and for all
                                                                women without risk
                                                                factors, another
                                                                screening test was
                                                                done at 24-28
                                                                weeks’ gestation. A
                                                                positive screen if 1h




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                       Page 123 of 441
                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                Follow-up &   Effect Size   Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                     Outcome                                           Comments
                 Level                 cs                                       Measures
                                                        serum glucose value
                                                        of ≥ 130 mg/dl on
                                                        glucose challenge
                                                        test. Women with
                                                        positive screen
                                                        underwent a 100g 3
                                                        h OGTT and the
                                                        diagnosis of GDM
                                                        was made with the
                                                        Carpenter and
                                                        Coustan criteria.
                                                        Group 1 women with
                                                        normal glucose
                                                        tolerance

                                                        Comparison: Group
                                                        2 women with
                                                        gestational diabetes




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                        Page 124 of 441
                                                                                                                                                     DRAFT FOR CONSULTATION




Q12 - How should gestational diabetes be diagnosed?

Q13 - Does diagnosis, monitoring and intervention for gestational diabetes improve outcomes in mothers and
babies?

The evidence tables for these clinical questions have been combined.

Bibliographic Study Type Number of   Patient     Intervention &                                      Follow-up &             Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                           Outcome                                                                    Comments
                 Level                 cs                                                             Measures
Crowther            Study Type:       490 women      Women between        Intervention:             Outcome             The rate of serious          Treatment of gestational
CA;Hiller JE;Moss   RCT               with IGT in    16-30 weeks          Treatment                 Measures:           perinatal outcomes           diabetes in the form of
JR;McPhee                             intervention   gestation. One or    (institution's clinical   A composite         among infants was            dietary advice, blood
AJ;Jeffries         Evidence level:   group          more risk factors    practice for              measure of          significantly lower in the   glucose monitoring and
WS;Robinson         1++               510 women in   for gestational      gestational               serious perinatal   intervention group than      insulin therapy as
JS;Australian                         routine care   diabetes on          diabetes) including       complications       the routine care group       required for glycaemic
Carbohydrate                          group          selective            individualised            (death, shoulder    (1% vs. 4%; p=0.01,          control reduces the rate
Intolerance Study                                    screening or a       dietary advice,           dystocia, bone      adjusted for maternal        of serious perinatal
in Pregnant                                          +ve 50g GCT and      instructions on self-     fracture, nerve     age, ethnic group and        complications , without
Women                                                had a 75g OGTT       monitoring (4xday),       palsy)              parity).                     increasing the rate of
(ACHOIS) Trial                                       at 24-34 weeks       target levels for                                                          caesarean delivery.
Group.;                                              gestation in which   blood glucose of          Admission to        The number needed to
                                                     venous plasma        between 3.5 and 5.5       neonatal nursery    treat to prevent a
2005 Jun 16                                          glucose              (fasting) <5.5                                serious outcome in an
                                                     <7.8mmol/l after     (preprandial) and         Jaundice            infant was 34 (95% CI
149
                                                     an overnight fast    <7.0 (postprandial)       requiring           20-103).
                                                     and 7.8-11.0         and insulin therapy.      phototherapy
                                                     mmol/l at 2 hours.                                               Admission to neonatal
                                                     Exclusions:          Comparison:               Induction of      nursery
                                                     Women                Routine care              labour            Intervention: 71%
                                                     previously treated                                               Routine care: 61%,
                                                     for gestational                                Caesarian section adjusted p=0.01
                                                     diabetes or active
                                                     chronic systemic                               Maternal health     Jaundice requiring
                                                     disease (except                                (SF-36)             phototherapy:
                                                     essential                                                          No significant difference
                                                     hypertension).                                 Anxiety
                                                                                                    (Spielberger       Induction of labour:
                                                                                                    State-Trait        Intervention: 39%
                                                                                                    Anxiety Inventory) Routine care: 29%,
                                                                                                                       adjusted p<0.001
                                                                                                    Depression




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                       Page 125 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                           (Edinburgh    Caesarean:
                                                                           Postnatal     No significant difference
                                                                           Depression
                                                                           Scale)        All measures on the SF-
                                                                                         36 showed trends in
                                                                                         favour of the
                                                                                         intervention group but
                                                                                         not all were significant

                                                                                         Depression
                                                                                         At three months
                                                                                         postpartum fewer
                                                                                         women in the
                                                                                         intervention group had a
                                                                                         score on the Edinburgh
                                                                                         postnatal depression
                                                                                         scale suggestive of
                                                                                         depression 8% vs. 17%)

                                                                                         Anxiety
                                                                                         Similar in both groups.

                                                                                         Birthweight - g
                                                                                         Intervention: 335±551
                                                                                         Routine care: 3482±660,
                                                                                         adjusted P<0.001


                                                                                         Secondary outcomes:
                                                                                         Neonatal:
                                                                                         LGA
                                                                                         Intervention: 68 (13%)
                                                                                         Routine care: 115
                                                                                         (22%), adjusted
                                                                                         P<0.001

                                                                                         Macrosomia ≥4kg
                                                                                         Intervention:49 (10%)
                                                                                         Routine care: 110 (21),
                                                                                         adjusted p<0.001

                                                                                         Neonatal




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 126 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         Hypoglycaemia requiring
                                                                                         IV therapy:
                                                                                         No significant difference

                                                                                         Respiratory distress
                                                                                         syndrome:
                                                                                         No significant difference
                                                                                         Maternal:
                                                                                         Antenatal preeclampsia:
                                                                                         No significant difference

                                                                                         Any perineal trauma:
                                                                                         No significant difference

                                                                                         Length of postnatal stay
                                                                                         Intervention:4 (3-5)
                                                                                         Routine care:4 (3.5)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 127 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                      Follow-up &              Effect Size             Study Summary                 Reviewer
 Information & Evidence   Patients Characteristi Comparison                           Outcome                                                                      Comments
                 Level                 cs                                             Measures
Jensen             Study Type:       3260   Pregnant women   Diagnostic             Outcome              Results from multivariate   The frequency of         IGT not treated
DM;Damm            Cohort                   who underwent    thresholds based on    Measures:            logistic regression:        macrosomia showed        Multivariate logistic
P;Sorensen                                  OGTT at 31.3     a 75g 2-h OGTT.        Pregnancy            Odds Ratio (OR) vs.         >50% increase in women   regression controlled for
B;Molsted-         Evidence level:          weeks.           Group 1: 2-h value     induced              group 1 i.e.: 2-h blood     with 2-h OGTT 7.8        maternal prepregnancy
Pedersen           2++                                       <7.8 mmol/l            hypertension         glucose <7.8 mmol/l         mmol/l - 8.9 mmol/l      BMI, maternal age, parity,
L;Westergaard                                                (n=2596) not treated   (PIH)/preeclampsi                                compared with women      smoking, weight gain
JG;Korsholm                                                                         a                 Preeclampsia/PIH:              with 2-h OGTT <7.8       during pregnancy,
L;Ovesen P;Beck-                                             Group 2: 2-h value                       Group 2: OR 0.9 (0.5-          mmol.                    gestational age,
Nielsen H;                                                   7.8-8.9 mmol/l         Shoulder dystocia 1.8)                                                    anamnestic risk indicators
                                                             (n=289) not treated                      Group 3: OR 1.6 (0.9-                                   for GDM, ethnic
2003 Jan                                                                            Spontaneous       2.7)                                                    background and clinical
                                                             Group 3. 2-h value     preterm birth     Group 4: 2.9 (1.3-6.2)                                  centre
145
                                                             9.0-11.0 mmol/l
                                                             (n=278) treated        Caesarean            Spontaneous preterm
                                                                                    section              delivery
                                                             Group 4. 2-h value                          Group 2: OR 1.3 (0.4-
                                                             ≥11.1 mmol/l (n=97) LGA                     3.9)
                                                             treated                                     Group 3: OR 2.0 (1.0-
                                                                                 Macrosomia              3.6)
                                                                                 (birthweight            Group 4: OR 5.1 (2.4-
                                                                                 ≥4000g)                 11.0) p<0.001

                                                                                    Hypoglycaemia     LGA
                                                                                    (IV glucose duringGroup 2: OR 1.7 (1.1-
                                                                                    the first 48 hours2.4)
                                                                                    of life)          Group 3: OR 2.0 (1.4-
                                                                                                      2.8)
                                                                                    Jaundice (treated Group 4: 3.0 (1.8-5.1)
                                                                                    with              p<0.001
                                                                                    phototherapy)
                                                                                                      Birthweight ≥4000g
                                                                                    Respiratory       Group 2: 1.5 (1.1-2.2)
                                                                                    distress          Group 3. 1.2 (0.8-1.7)
                                                                                                      Group 4. 2.1 (1.1-3.8)

                                                                                                         Hypoglycaemia
                                                                                                         Group 2: 0.7 (0.2-2.2)
                                                                                                         Group 3: 3.4 (2.0-5.9)
                                                                                                         Group 4: 8.1 (3.9-16.7)
                                                                                                         p<0.001

                                                                                                         Shoulder dystocia was




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                      Page 128 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                                                         significantly increased in
                                                                                         univariate analysis
                                                                                         (group 1: 1.5%; group 2:
                                                                                         3.4%; group 3: 4.4%;
                                                                                         group 4: 5.1%;
                                                                                         p=0.004).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 129 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &            Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                                   Comments
                 Level                 cs                                                 Measures
Langer          Study Type:       126 GDM        All women         Intervention: GDM    Outcome           No significant difference    The incidence of adverse IGT untreated
O;Brustman      Case-control      126 one        referred for an   by NDDG criteria vs. Measures:         in blood glucose value       neonatal outcome
L;Anyaegbunam                     abnormal value OGTT              1 abnormal value on LGA (≥90%)         (mean and SD)                among the one abnormal Matched by age, weight,
A;Mazze R;      Evidence level:   126 normal                       OGTT                                   between GDM group            value group was           parity and race
                2++               values                                                Macrosomia        before treatment and the     significantly higher than
1987                                                               Comparison: Normal (≥4000g)            group with 1 abnormal        for other groups.
                                                                   values on OGTT                         OGTT value. Significant
144
                                                                                        Neonatal          difference between           If treated the patients
                                                                                        hypoglycaemia     GDM group after              with one abnormal value
                                                                                        (blood glucose    treatment and the group      would have most
                                                                                        <2.2 mmol/l at    with one abnormal            probably experienced a
                                                                                        birth reconfirmed OGTT value:                  significantly lower
                                                                                        by laboratory     Fasting: p<0.002             incidence of adverse
                                                                                        analysis)         After breakfast:             perinatal outcome
                                                                                                          p<0.0001                     consistent with that found
                                                                                        Hyperbilirubinemi Before lunch: p<0.0001       among individuals
                                                                                        a (blood level    After lunch: p<0.0001        treated for GDM.
                                                                                        >12m/dl)          Before dinner: p<0.01
                                                                                                          After dinner: p<0.001
                                                                                        Polycythemia      Bedtime: p<0.0001
                                                                                        (central venous
                                                                                        hematocrit >65%) Ambulatory glucose
                                                                                                          profile:
                                                                                                          The profiles of the GDM
                                                                                                          group before treatment
                                                                                                          were indistinguishable
                                                                                                          from the profiles of the
                                                                                                          group with 1 abnormal
                                                                                                          value. The profiles of the
                                                                                                          GDM group and the
                                                                                                          group with 1 abnormal
                                                                                                          value were significantly
                                                                                                          different from those of
                                                                                                          the normal group. After
                                                                                                          treatment the profiles of
                                                                                                          the GDM group were
                                                                                                          significantly different
                                                                                                          from the group with one
                                                                                                          abnormal value. There
                                                                                                          was no significant
                                                                                                          difference between the
                                                                                                          GDM group after




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 130 of 441
                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size              Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                           Comments
                 Level                 cs                                   Measures
                                                                                         treatment and the
                                                                                         normal group.

                                                                                         Neonatal outcomes: A
                                                                                         significant difference in
                                                                                         the incidence of LGA
                                                                                         and macrosomic infants
                                                                                         between the group with
                                                                                         1 abnormal value and
                                                                                         the normal group (34%
                                                                                         vs. 9%;p<0.01) and the
                                                                                         GDM group (34% vs.
                                                                                         12%, p<0.01). No
                                                                                         significant difference in
                                                                                         LGA and macrosomic
                                                                                         infants was found
                                                                                         between the GDM group
                                                                                         and the normal group.
                                                                                         Neonatal metabolic
                                                                                         disorders were
                                                                                         significantly higher in the
                                                                                         group with 1 abnormal
                                                                                         value (15%) when
                                                                                         compared with group
                                                                                         the GDM and normal
                                                                                         groups (3%, p<0.01).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                    Page 131 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &            Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                            Comments
                 Level                 cs                                                 Measures
Mello G;Parretti   Study Type:       829 women        No history of    Glucose levels   Outcome            Logistic regression                        There is a significant
E;Cioni            Cohort            who did not      pregestational                    Measures:                                                     association between 75g
R;Lucchetti                          have GDM         diabetes or GDM.                  Neonatal:           Early period (16-20                       load values and abnormal
R;Carignani        Evidence level:   according to                                       Cranial/thoracic    weeks)                                    neonatal anthropometric
L;Martini          2++               Carpenter and                                      circumference       1. Neonatal CC/TC                         features.
E;Mecacci                            Coustan criteria                                   (CC/TC) ratio ≤     ratio≤10th percentile:
F;Lagazio                                                                               10th percentile     Fasting 75g glucose                       The results suggest the
C;Pratesi M;                                                                                                load values; adjusted                     possibility of using 1-h
                                                                                        Ponderal index      OR 1.05 (95% CI 1.02-                     75g glucose load as a
2003 Apr                                                                                (birthweight/lengt 1.07)                                      single test for the
                                                                                        h3 x 100)           1-h 75g glucose load                      diagnosis of GDM
143
                                                                                                            value: adjusted OR 1.81                   adopting a threshold
                                                                                        Macrosomia (≥       (95% CI 1.15-2.83)                        value of 8.3 mmol/l at 16-
                                                                                        90th percentile for 2-h 75g glucose load:                     20 weeks and 8.8 mmol/l
                                                                                        gestational age) adjusted OR 1.03 (1.02-                      at 26-30 weeks.
                                                                                                            1.04)

                                                                                                           2.Neonatal ponderal
                                                                                                           index ≥90th percentile
                                                                                                           for gestational age:
                                                                                                           Fasting 75g glucose
                                                                                                           load value: adjusted OR
                                                                                                           1.02 (0.99-1.05)
                                                                                                           1-h 75g glucose load
                                                                                                           value: adjusted OR 1.03
                                                                                                           (1.02-1.03)
                                                                                                           2-h 75g glucose load
                                                                                                           value: adjusted OR 1.03
                                                                                                           (1.02-1.04)

                                                                                                           3. Macrosomia
                                                                                                           Fasting 75g glucose
                                                                                                           load value: adjusted OR
                                                                                                           1.03 (1.00-1.06)
                                                                                                           1-h 75g glucose load
                                                                                                           value: adjusted OR 1.02
                                                                                                           (1.02-1.03)
                                                                                                           2-h 75g glucose load
                                                                                                           value: adjusted OR 1.02
                                                                                                           (1.00-1.03)

                                                                                                           Late period (26-30




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 132 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                        Comments
                 Level                 cs                                   Measures
                                                                                         weeks)
                                                                                         1. Neonatal CC/TC
                                                                                         ratio≤10th percentile
                                                                                         Fasting 75g glucose
                                                                                         load values: adjusted
                                                                                         OR 1.07 (1.05-1.10)
                                                                                         1-h 75g glucose load
                                                                                         value: adjusted OR
                                                                                         1.86 (1.19-2.91)
                                                                                         2-h 75g glucose load
                                                                                         adjusted OR: 1.04 (1.03-
                                                                                         1.05)

                                                                                         2.Neonatal ponderal
                                                                                         index ≥90th percentile
                                                                                         for gestational age
                                                                                         Fasting 75g glucose
                                                                                         load value: adjusted OR
                                                                                         1.04 (1.01-1.06)
                                                                                         1-h 75g glucose load
                                                                                         value: adjusted OR 1.03
                                                                                         (1.02-1.03)
                                                                                         2-h 75g glucose load
                                                                                         value: adjusted OR
                                                                                         1.03 (1.02-1.04)

                                                                                         3. Macrosomia
                                                                                         Fasting 75g glucose
                                                                                         load value: adjusted OR
                                                                                         1.03 (1.00-1.06)
                                                                                         1-h 75g glucose load
                                                                                         value: adjusted OR 1.02
                                                                                         (1.01-1.03)
                                                                                         2-h 75g glucose load
                                                                                         value: adjusted OR 1.02
                                                                                         (1.00-1.03)

                                                                                         There were significant
                                                                                         differences between
                                                                                         pregnancies with a
                                                                                         neonatal CC/TC ratio
                                                                                         ≤10thpercentile or >10th




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 133 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                                                         percentile in rates of
                                                                                         caesarean section
                                                                                         (38/112(33.9%) vs.
                                                                                         119/717 (166%);
                                                                                         p<0.001; OR 3.93, 95%
                                                                                         CI 2.6-5.9) and shoulder
                                                                                         dystocia (6/112 (5.4%)
                                                                                         v. 1/717 (0.1%);
                                                                                         p<0.0003; OR 6.65,
                                                                                         95% CI 4.6-9.4)

                                                                                         ROC curves for the
                                                                                         prediction of neonatal
                                                                                         CC/TC ratio ≤10th
                                                                                         percentile in both early
                                                                                         and late periods were
                                                                                         constructed for 1-hr 75g
                                                                                         glucose load values.
                                                                                         Inflection points were
                                                                                         identified for a threshold
                                                                                         value of 8.3 mmol/l in
                                                                                         the early period and 8.8
                                                                                         mmol/l in the late period.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 134 of 441
                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                Follow-up &            Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                     Outcome                                                                    Comments
                 Level                 cs                                       Measures
Ostlund I;Hanson    Study Type:       213 women         GDM defined as        Outcome            Caesarean section           IGT is independently and Higher caesarean rate not
U;Bjorklund         Case-control      with              fasting blood         Measures:          adjusted OR (adjusted       significantly associated  linked to diagnosis of
A;Hjertberg R;Eva                     undiagnosed       glucose ≥ 6.7         Caesarean          for parity, LGA infant,     with an increased         GDM
N;Nordlander        Evidence level:   and untreated     mmol/l and/or 2-h     section            PIH/preeclampsia,           incidence of caesarean
E;Swahn             2++               IGT during        blood glucose ≥11.1                      ethnicity, BMI) 1.9 (1.2-   section and prematurity
ML;Wager J;                           pregnancy         mmol/l                Prematurity (<37   2.9).                       and a markedly
                                                                              weeks)             Emergency caesarean         increased proportion of
2003 Jul                                                IGT fasting blood                        section adjusted OR 2.1     LGA and macrosomic
                                                        glucose <6.7 mmol/l Birthweight          (1.3-3.5 mmol/l)            infants and admission to
146
                                                        and 2-hr level 9.0-                                                  NICU for 2 days or longer
                                                        11.0 mmol/          Macrosomia (≥        Prematurity adjusted OR
                                                                            4000g, ≥ 4500g, ≥    2.0 (95% CI 1.0-3.9)
                                                        Comparison: 815     5000g)
                                                        controls                                 LGA adjusted OR 7.3
                                                                            LGA                  (95% CI 4.1-12.7).

                                                                              SGA                Admission to NICU
                                                                                                 adjusted OR 2.3 (95%
                                                                              Hypoglycaemia      CI 1.3-4.0)

                                                                              Hyperbilirubinemi 71.3% of infants in the
                                                                              a requiring       IGT group had no
                                                                              treatment         neonatal complications
                                                                                                compared with 87.3% of
                                                                              Admission to      the control group.
                                                                              NICU 2 days or
                                                                              longer




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                               Page 135 of 441
                                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                           Follow-up &            Effect Size              Study Summary                  Reviewer
 Information & Evidence   Patients Characteristi Comparison                                Outcome                                                                      Comments
                 Level                 cs                                                  Measures
Sermer M;Naylor   Study Type:       3637             Women who were Plasma glucose       Outcome            Macrosomia:                 Progressively increasing   Untreated and
CD;Farine         Cohort            nondiabetc       24 years or over values following   Measures:          Increased with each         plasma glucose was         undiagnosed IGT.
D;Kenshole                          patients (by     at time of       OGTT               Preeclampsia       quartile of OGTT fasting    associated with an         Physicians and
AB;Ritchie        Evidence level:   NDDG criteria)   delivery.                                              level (p<0.001) 1-h level   increased incidence of a   researchers blinded to
JW;Gare           2++                                                                    Macrosomia         (p=0.004) 2-h level         variety of adverse         GCT and OGTT results.
DJ;Cohen                                                                                 (birthweight       (p<0.001) and with each     outcomes
HR;McArthur                                                                              ≥4000g)            quartile of GCT                                       Multivariate analysis
K;Holzapfel                                                                                                 (=0.001).                  However after adjusting controlled for obesity
S;Biringer A;                                                                            Caesarean                                     for other risk factors the
                                                                                         section            Preeclampsia:              independent incidence of
1998 Aug                                                                                                    Increased with each        glucose values on
                                                                                         Phototherapy       quartile of OGTT 1-h       maternal and fetal
141
                                                                                                            level, (p=0.004) 2-h level outcomes was modest.
                                                                                         Maternal length of (p=0.001) and GCT
                                                                                         stay (LOS)         (p=0.001).                 The usual treatment of
                                                                                                                                       GDM lowers macrosomia
                                                                                         Neonatal nursery Caesarean section:           but the ate of caesarean
                                                                                         LOS                Increased with each        delivery remains
                                                                                                            quartile of OGTT 1-h       inexplicably high
                                                                                         Fetal trauma       level (p=0.005) 2-h level
                                                                                                            (p=0.003) 3-h level
                                                                                                            (p=0.001) and each
                                                                                                            quartile of GCT
                                                                                                            (p=0.001).

                                                                                                            The need for
                                                                                                            phototherapy, maternal
                                                                                                            LOS and neonatal
                                                                                                            nursery LOS all
                                                                                                            significantly increased
                                                                                                            with worsening
                                                                                                            carbohydrate
                                                                                                            intolerance. There was
                                                                                                            no relationship between
                                                                                                            OGTT and GCT values
                                                                                                            and congenital
                                                                                                            malformations, IV
                                                                                                            therapy for
                                                                                                            hypoglycaemia and
                                                                                                            RDS. For fetal trauma
                                                                                                            the higher quartiles are
                                                                                                            associated with the




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                          Page 136 of 441
                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size         Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                      Comments
                 Level                 cs                                   Measures
                                                                                         highest percentage of
                                                                                         fetal trauma (NS).

                                                                                         Multivariate analysis:
                                                                                         Caesarean section:
                                                                                         BMI (OR 1.10, 95% CI
                                                                                         1.07-1.16, p<0.05)
                                                                                         OGTT 1-h (OR 1.10,
                                                                                         95% CI 1.01-1.22)
                                                                                         OGTT 3-h (OR 1.10,
                                                                                         95% CI 1.06-1.22,
                                                                                         p<005)

                                                                                         Birthweight>4500:
                                                                                         BMI (OR 1.2, 95% CI
                                                                                         1.06-1.25, p<0.05)
                                                                                         OGTT Fasting level (OR
                                                                                         2.6, 95% CI 1.69-4.11,
                                                                                         p<0.05)

                                                                                         Preeclampsia:
                                                                                         BMI (OR 1.30, 95% CI
                                                                                         1.19-1.32, p<0.05)
                                                                                         OGTT 2-h (OR 1.10,
                                                                                         95% CI 1.00-1.27).

                                                                                         Pregnancy outcomes
                                                                                         based on glucose test
                                                                                         results:
                                                                                         Macrosomia (>4000g)
                                                                                         Negative screenees:
                                                                                         13.7%
                                                                                         Untreated GDM by C&C
                                                                                         criteria: 28.7%
                                                                                         Treated GDM by NDDG
                                                                                         criteria: 10.5%
                                                                                         (p<0.001)

                                                                                         Caesarean delivery:
                                                                                         Negative screenees:
                                                                                         20.2%
                                                                                         Untreated GDM by C&C




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 137 of 441
                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &       Effect Size       Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                   Comments
                 Level                 cs                                   Measures
                                                                                         criteria: 29.6%
                                                                                         Treated GDM by NDDG
                                                                                         criteria: 33.6%
                                                                                         P<0.001




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                            Page 138 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                           Follow-up &             Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                Outcome                                                                   Comments
                 Level                 cs                                                  Measures
Weiss          Study Type:       220     OGTT at 28             Capillary blood          Sensitivity for the Sensitivity and             In women with elevated
PAM;Haeusler   Case-control              weeks and              glucose levels at 0,     diagnosis of fetal specificity of fasting and   amniotic fluid insulin
M;Tamussino                              elevated amniotic      1 and 2-h following      hyperinsulinism     postload glucose values     concentration 97% had
K;Haas J;      Evidence level:           fluid insulin levels   glucose load of                              for the diagnosis of        1-h postload glucose
               2++                       at 31 weeks            1g/kg (n=161) or                             elevated amniotic fluid     levels >8.9mmol/l.
2000                                     (≥42pmol/l).           75g(n=59)                                    insulin concentration
                                         Excluded:                                                                                       The 1-hr value had the
405
                                         Women with             Comparison: 220                              Fasting ≥5.0 mmol/l         highest sensitivity for
                                         factors other than     nondiabetic (i.e. with                       Sensitivity: 58%            measuring fetal
                                         GDM that could         normal OGTT)                                 Specificity: 97%            hyperinsulinism.
                                         influence amniotic     pregnant women
                                         fluid insulin levels   with amniotic insulin                        1-h ≥8.9 mmol/l
                                         such as type 1         fluid levels                                 Sensitivity: 97%
                                         diabetes, insulin      <42pmol/l.                                   Specificity: 99%
                                         treatment, rhesus
                                         disease and fetal                                                   2-h ≥7.8mmol/l
                                         malformation.                                                       Sensitivity: 54%
                                                                                                             Specificity: 97%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 139 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &            Effect Size            Study Summary             Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                               Comments
                 Level                 cs                                                Measures
Tallarigo, L.,     Study Type:       249   Normal results on     Group A (n=151)       Outcome            Macrosomia              There is a positive
Giampietro, O.,    Cohort                  the 100g OGTT         women with 2-h        Measures:          Group A: 15/151 (9.9%)  correlation between the
Penno, G., et al                           using O'Sullivan's    plasma glucose        Macrosomia         Group B: 9/58 (15.5%)   2-h plasma glucose level
                   Evidence level:         criteria (0=5.8, 1-   levels <5.6 mmol/l.   (≥4000g)           Group C: 11/40 (27.5%)  and the incidence of
1986               2++                     h=10.5, 2-h=9.2,                                               P<0.01 (test for linear complications of
                                           3-h=8.0 mmol/l.)      Group B (n=58)        Congenital         trend)                  pregnancy even though
140
                                           No history of         women with 2-h        abnormalities                              all values were in the
                                           diabetes.             levels of 5.6-6.6                        Maternal complications accepted normal range
                                                                 mmol/l                Toxaemia           (toxemia/ceasarean)     (<9.1mmol/l). In addition
                                                                                       (presence of       Group A: 30/151         there is a correlation
                                                                 Group C (n=40)        convulsive         (19.9%)                 between frequency of
                                                                 women with 2-h        seizures or any    Group B: 15/58 (25.9%) macrosomia, toxemia,
                                                                 values 6.6-9.1        two of the         Group C: 16/40 (40%)    caesarean section or
                                                                 mmol/l (IGT           following signs:   P<0.01 (test for linear both and the 2-h plasma
                                                                 according to          proteinuria,       trend)                  glucose level in
                                                                 NDDG).                hypertension,                              nonobese women
                                                                                       edema).            No correlation was
                                                                                                          found between the two-
                                                                                       Caesarean          hour plasma glucose
                                                                                       section            level and maternal age
                                                                                                          or maternal weight.

                                                                                                          A significant association
                                                                                                          between 2-h plasma
                                                                                                          glucose level and both
                                                                                                          macrosomia and
                                                                                                          complications persisted
                                                                                                          after controlling for
                                                                                                          maternal age and
                                                                                                          weight.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                     Page 140 of 441
                                                                                                                                        DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &             Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                   Comments
                 Level                 cs                                                Measures
Sacks, D., et al   Study Type:       3352 women   Fasting blood    2-h post 75g plasma Birthweight         Mulitvariate analysis    Although there was an
                   Cohort                         glucose <5.8     glucose levels                          adjusted for maternal    overall progressive
1995                                              mmol/l and 2-h                       Macrosomia          age, ethnicity, parity,  increase in the
                   Evidence level:                values <11                           (≥90th percentile   BMI, mean weight gain,   prevalence of
142
                   2++                            mmol/l.                              for gestational                              macrosomia with
                                                                                                           family history of diabetes
                                                  Untreated.                           age)                and obstetric history.   incremental GTT
                                                                                                                                    threshold values no
                                                                                       Ponderal index      A significant positive   isolated infection point for
                                                                                                           relatationship was found fasting or 2-h values
                                                                                                           between fasting, 1 and could be identified.
                                                                                                           2-h values and
                                                                                                           percentiles of
                                                                                                           birthweight.

                                                                                                           Only the fasting and 2-h
                                                                                                           levels were found to be
                                                                                                           significantly and
                                                                                                           independently
                                                                                                           associated with
                                                                                                           macrosomia.

                                                                                                           ROC curves were
                                                                                                           constructed for fasting
                                                                                                           and 2 values for the
                                                                                                           prediction of
                                                                                                           macrosomia. No clear
                                                                                                           inflection point was
                                                                                                           identified for any
                                                                                                           threshold value. .




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 141 of 441
                                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                            Follow-up &        Effect Size          Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                                 Outcome                                                       Comments
                 Level                 cs                                                   Measures
Saldana, T. et al   Study Type:       1190 white   Women between     GDM defined as two   Birthweight    The overall prevalence                    Small numbers of women
                    Cohort            women        24 and 29 weeks   abnormal values on                  of GDM was 5.0% and                       with IGT
2003                                  865 Black    pregnancy         100g OGTT            Macrosomia     IGT was 2.6%.
                    Evidence level:   women                          (Carpenter and       (≥4000g)
147
                    2+                                               Coustan criteria)-   LGA            Among white women 70
                                                                     treated with                        had GDM, 40 had IGT
                                                                     diet/insulin                        and 1,080 had NGT.
                                                                     IGT defined as one                  Among black women 32
                                                                     abnormal value on                   had GDM, 13 had IGT
                                                                     100g OGTT - not                     and 820 had NGT.
                                                                     treated
                                                                     NGT no high values                  Among Black women
                                                                     on OGTT                             logistic regression
                                                                                                         adjusted for pre-
                                                                                                         pregnancy BMI,
                                                                                                         maternal age and height
                                                                                                         and gestational age
                                                                                                         continued to show a
                                                                                                         significant increase in
                                                                                                         birthweight compared
                                                                                                         with women with NGT.

                                                                                                         In contrast among white
                                                                                                         women there was no
                                                                                                         significant increase in
                                                                                                         birthweight associated
                                                                                                         with IGT compared with
                                                                                                         NGT.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                Page 142 of 441
                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &   Effect Size   Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                           Comments
                 Level                 cs                                   Measures




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                    Page 143 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                  Comments
                 Level                 cs                                                      Measures
Kremer CJ;Duff P; Study Type:       197 women        Average age 28       Intervention:     Follow-up period: 124 women (63%)              Approximately 80% of
                  Cohort            with gestational (range 18-40).       Treatment with                      achieved satisfactory        patients who fail to
2004                                diabetes         The mean             glyburide .       Outcome           blood glucose control        respond to diet can be
                  Evidence level:                    gestational age at                     Measures:         with diet alone; 73          treated effectively with
181
                  3                                  diagnosis was 24 The starting dose     Achievement of    (37%) required               glyburide.
                                                     weeks and the    was 2.5mg,            satisfactory      glyburide. Women in the
                                                     mean gestational administered in the glucose control     glyburide group were
                                                     age at start of  morning before        (mean plasma      significantly more likely
                                                     glyburide        breakfast. The daily fasting glucose ≤ to be morbidly obese
                                                     treatment was 30 dose was increased 5 mmol/l, mean 1- than the group
                                                     weeks            in increments as      hour value ≤ 7.4 controlled by diet
                                                                      needed to a           mmol/l)           (p<0.05).
                                                     82% had BMI>30 maximum of 20 mg.
                                                     33% had BMI>40                                           Of the 73 patients
                                                                      The goal was to                         treated with glyburide,
                                                     (cf Langer)      maintain fasting                        59 (81%; 95% CI 76.4-
                                                                      glucose ≤ 5 mmol/l                      86.6) had acceptable
                                                     Country: Florida and 1-hour value ≤                      glucose control on
                                                                      7.4 mmol/l                              medical therapy. 23
                                                                                                              patients required a dose
                                                                      Patients were                           of only 2.5mg, 20
                                                                      prescribed an                           responded to 5mg, 8 to
                                                                      American Diabetes                       7.5mg and 8 required
                                                                      Association Diet,                       doses greater than
                                                                      calculated as 30                        7.5mg. 21 women
                                                                      kcal/kg of ideal body                   required morning and
                                                                      weight, and were                        evening doses of
                                                                      asked to monitor                        glyburide.
                                                                      their blood sugar 4
                                                                      times per day                           9 women experienced
                                                                      (before breakfast                       side effects (4 malaise
                                                                      and 1 hour after                        and weakness, 2
                                                                      each meal) 4 to 7                       nausea, 1 light headed,
                                                                      days per week.                          2 symptoms of
                                                                                                              hypoglycaemia). Only 1
                                                                      The dose was                            patient discontinued due
                                                                      adjusted if weekly                      to side effects.
                                                                      mean blood glucose
                                                                      values exceeded the                     In the 59 women
                                                                      threshold or if 20%                     controlled with glyburide,
                                                                      of values exceeded                      11 (19%) had infants
                                                                      the threshold.                          who weighed >4000g.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 144 of 441
                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                     Outcome                                                         Comments
                 Level                 cs                                       Measures
                                                                                             Overall 36/59 (49%)
                                                        Patients who failed                  required caesarean
                                                        to achieve                           delivery. 11 (19%) had
                                                        satisfactory control                 caesarean deliveries
                                                        were treated with                    because of labour
                                                        twice daily dose of                  abnormalities or
                                                        insulin.                             suspected macrosomia.
                                                                                             (Cf 14-17% primary CS
                                                        Comparison:                          rate in general obstetric
                                                                                             population).

                                                                                             No fetal abnormalities




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                      Page 145 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                            Follow-up &              Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                 Outcome                                                                  Comments
                 Level                 cs                                                   Measures
Langer O;Conway Study Type:      404 women          Women were        Intervention:       Follow-up period:    Blood glucose (mmol/l):  We found that among
DL;Berkus       RCT              with gestational   screened for      Glyburide                                Fasting                  women with gestational
MD;Xenakis                       diabetes           diabetes with a                       Outcome              Glyburide: 5.7±1.4       diabetes, the degree of
EM;Gonzales O; Evidence level:   201 assigned to    one-hour, 50g     Comparison: Insulin Measures: Blood      Insulin:5.9±1.4 (p=0.12) glycemic control and
                1++              receive            oral glucose                          glucose                                       perinatal outcomes were
2000 Oct 19                      glyburide and      challenge. The                        Fasting (mmol/l)     Preprandial:             essentially the same for
                                 203 to receive     women ranged in                       Preprandial          Glyburide: 5.7± 1.1      those treated with
406
                                 insulin.           age from 18 to 40                     (mmol/l)             Insulin:5.9±1.3 (p=0.16) glyburide and those
                                                    and were all                          Postprandial                                  treated with insulin.
                                                    medicaid                              (mmol/l)             Postprandial:
                                                    recipients. 83%                       Mean (mmol/l)        Glyburide: 7.2±1.4       Glyburide is an effective
                                                    Hispanic, 12%                         Glycosated           Insulin:7.1±1.5 (p=0.69) alternative to insulin in
                                                    non Hispanic                          haemoglobin (%)      Mean:                    women with gestational
                                                    white, 5 % black.                                          Glyburide: 6.3±1.05      diabetes.
                                                                                          Neonatal             Insulin:6.4±0.1.2
                                                    Country: USA                          outcomes:            (p=0.33)
                                                                                          Birth weight (g)
                                                                                          LGA (no,%)           Glycosated
                                                                                          Macrosomia           haemoglobin (%):
                                                                                          (no,%)               Glyburide: 5.7±1.3
                                                                                          Cord-serum           Insulin: 5.6±1.2 (p=0.42)
                                                                                          insulin
                                                                                          Intravenous          Birth weight (g)
                                                                                          glucose therapy      Glyburide:3256±543
                                                                                          (no,%)               Insulin:3194±598
                                                                                          Hypoglycaemia        (p=0.28)
                                                                                          Hypocalcaemia
                                                                                          Hyperbilirubinemi    LGA (no,%)
                                                                                          a                    Glyburide:25 (12)
                                                                                          Polycythemia         Insulin:26 (13) p=0.76
                                                                                          Lung
                                                                                          complications (no,   Macrosomia (no,%)
                                                                                          %)                   Glyburide:14 (7)
                                                                                          Respiratory          Insulin:9 (4) p=0.26
                                                                                          support (no,%)
                                                                                          Admission to         Cord-serum insulin
                                                                                          neonatal intensive   Glyburide:15±13
                                                                                          care                 Insulin:15±21 (p=0.84)
                                                                                          Congenital
                                                                                          anomaly              Intravenous glucose
                                                                                          Stillbirth           therapy (no,%)
                                                                                          Neonatal death       Glyburide:28 (14)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 146 of 441
                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size          Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         Insulin:22 (11)p=0.36

                                                                                         Neonatal hypoglycaemia
                                                                                         Glyburide:18 (9)
                                                                                         Insulin:12 (6) p=0.25

                                                                                         Hypocalcaemia
                                                                                         Glyburide:2 (1)
                                                                                         Insulin:2 (1) p=0.99

                                                                                         Hyperbilirubinemia
                                                                                         Glyburide:12 (6)
                                                                                         Insulin:8 (4) p=0.36

                                                                                         Polycythemia
                                                                                         Glyburide:4 (2)
                                                                                         Insulin:6 (3) p=0.52

                                                                                         Lung complications (no,
                                                                                         %)
                                                                                         Glyburide:16 (8)
                                                                                         Insulin:12 (6) p=0.43

                                                                                         Respiratory support
                                                                                         (no,%)
                                                                                         Glyburide:4 (2)
                                                                                         Insulin:6 (3)p=0.52

                                                                                         Admission to neonatal
                                                                                         intensive care
                                                                                         Glyburide:12 (6)
                                                                                         Insulin:14 (7) p=0.68

                                                                                         Congenital anomaly:
                                                                                         Glyburide: 5 (2)
                                                                                         Insulin:4 (2) p=0.74

                                                                                         Stillbirth
                                                                                         Glyburide:1 (0.5)
                                                                                         Insulin:1 (0.5) p=0.99

                                                                                         Neonatal death




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                Page 147 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                        Comments
                 Level                 cs                                   Measures
                                                                                         Glyburide:1 (0.05)
                                                                                         Insulin: 1 (0.05) p=0.99

                                                                                         Glyburide was not
                                                                                         detected in the cord
                                                                                         serum of any infant.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 148 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                  Follow-up &           Effect Size           Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                       Outcome                                                                Comments
                 Level                 cs                                                         Measures
Bertini AM;Silva   Study Type:       27 women           Women with          Intervention: OHA    Follow-up period: Glucose control was not Glibenclamide controlled Not blinded
JC;Taborda         RCT               assigned insulin   GDM (WHO            (glibenclamide                          achieved in 5 patients in blood glucose levels in
W;Becker                             therapy            criteria) who       acarbose)            Outcome            the Glibenclamide group 79% of patients
F;Lemos Bebber     Evidence level:   24 women           needed therapy in                        Measures: Birth (20.8%) and 8 patients
FR;Zucco Viesi     1+                assigned           addition to diet    Whenever maximum weight                 in the acarbose group
JM;Aquim                             glibenclamide      and exercise.       dose was reached     Macrosomia         (42.1%). No maternal
G;Engel RT;                          therapy            Brazil. Patients    without reaching     (>4000g) LGA       hypoglycaemia requiring
                                     19 assigned to     exclusively from    glucose control this (>90th percentile) hospital admission was
2005                                 acarbose           the public health   therapy was          Neonatal           reported.
                                     therapy            system. Patients    replaced by insulin hypoglycaemia
178
                                                        with concomitant    therapy.             (glucose level <40 GA at birth: mean (SD)
                                                        pathologies that                         mg/dl)             Insulin: 38.5 (1.2)
                                                        could affect        Comparison: Insulin                     Glibenclamide: 38.1
                                                        therapy or                                                  (1.2)
                                                        perinatal results                                           Acarbose:38.2 (1.2)
                                                        were excluded.                                              P=0.42

                                                        Country: Brazil                                           Apgar 1: mean (SD)
                                                                                                                  Insulin: 8.1 (0.97)
                                                                                                                  Glibenclamide:8 (0.7)
                                                                                                                  Acarbose 8.4 (0.9)
                                                                                                                  P=0.19

                                                                                                                  Apgar 5: mean (SD)
                                                                                                                  Insulin: 9.4 (0.7)
                                                                                                                  Glibenclamide: 9.0 (0.6)
                                                                                                                  Acarbose: 9.3 (.4)
                                                                                                                  P=0.10

                                                                                                                  Birth weight
                                                                                                                  Insulin: 3151.2 (407.2)
                                                                                                                  Glibenclamide: 3395.6
                                                                                                                  (524.4)
                                                                                                                  Acarbose: 3242.6
                                                                                                                  (400.6)
                                                                                                                  P=0.15

                                                                                                                  LGA
                                                                                                                  Insulin: 1 (3.7%)
                                                                                                                  Glibenclamide: 6 (25%)
                                                                                                                  Acarbose: 2 (10.5%)
                                                                                                                  P=0.073




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 149 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         SGA
                                                                                         Insulin: 2 (7.4%)
                                                                                         Glibenclamide: 0
                                                                                         Acarbose: 0

                                                                                         >4000g
                                                                                         Insulin: 0
                                                                                         Glibenclamide: 4 (16%)
                                                                                         Acarbose: 0

                                                                                         Neonatal
                                                                                         hypoglycaemia:
                                                                                         Insulin: 1 (3.7%)
                                                                                         Glibenclamide: 8
                                                                                         (33.3%)
                                                                                         Acarbose: 1 (5.3%)
                                                                                         P=0.006
                                                                                         1 infant from
                                                                                         Glibenclamide group
                                                                                         (born at 36 weeks)
                                                                                         required special care for
                                                                                         2 days. The other cases
                                                                                         were of mild
                                                                                         hypoglycaemia and
                                                                                         managed with maternal
                                                                                         milk.

                                                                                         There were 2 SGA
                                                                                         infants in the insulin
                                                                                         group (7.4%).

                                                                                         C section
                                                                                         Insulin:44.4%
                                                                                         Gylburide:50%
                                                                                         Acarbose: 52.2%

                                                                                         There were no perinatal
                                                                                         deaths




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 150 of 441
                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &              Follow-up &           Effect Size              Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                   Outcome                                                              Comments
                 Level                 cs                                     Measures
Glueck          Study Type:       42 pregnancies            Intervention:   Follow-up period: Of the 42 pregnancies,
CJ;Goldenberg   Cohort            in 39 women               Metformin                         gestational diabetes was
N;Wang                            with PCOS      Country:                   Outcome           diagnosed in three
P;Loftspring    Evidence level:                             Comparison:     Measures:         (7.1%)
M;Sherman A;
                                                                                              Weight
2004                                                                                          The median percentage
                                                                                              reduction in weight was
407
                                                                                              5.8% (p<0.0001).

                                                                                              Insulin
                                                                                              The median reduction in
                                                                                              insulin from pre-
                                                                                              treatment baseline to
                                                                                              the last preconception
                                                                                              visit was 40%
                                                                                              (p<0.0001). Insulin did
                                                                                              not increase in the first
                                                                                              or second trimester
                                                                                              (p>0.05) and rose 10%
                                                                                              in the third trimester.

                                                                                              Insulin resistance
                                                                                              The median percentage
                                                                                              reduction in HOMA
                                                                                              Insulin resistance from
                                                                                              pre-treatment baseline
                                                                                              to the last preconception
                                                                                              visit was 46%
                                                                                              (p<0.0001). Median
                                                                                              insulin resistance at the
                                                                                              last pre-conception visit
                                                                                              did not differ from insulin
                                                                                              resistance in the third
                                                                                              trimester.

                                                                                              Insulin secretion
                                                                                              The median percentage
                                                                                              reduction in insulin
                                                                                              secretion from pre-
                                                                                              treatment baseline to
                                                                                              last preconception visit




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                         Page 151 of 441
                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &       Effect Size       Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                   Comments
                 Level                 cs                                   Measures
                                                                                         was 45% (p=0.0004).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                            Page 152 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size              Study Summary                    Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                        Comments
                 Level                 cs                                                      Measures
Jacobson       Study Type:       268 women       Diagnosed with     Intervention:            Follow-up period: During 1999 through to       Our retrospective study      Demographic changes in
GF;Ramos       Cohort            with GDM        GDM (50g GCT ≥     Glibenclamide                               2000 268 women were         suggests that                obstetric population
GA;Ching                         treated with    7.7 mmol/l) OGTT                            Outcome            diagnosed with GDM          glibenclamide can be         overtime.
JY;Kirby       Evidence level:   insulin         NDDG criteria).    Both groups:             Measures:          and treated with insulin    effectively introduced for
RS;Ferrara     2+                                between 12 and     nutritional              Preterm delivery                               the treatment of women       Dosing protocol varied
A;Field DR;                      236 women       34 weeks where     counselling on a         (<37 weeks)        During 2001 through         with GDM                     between centres
                                 with GDM        diet therapy       diabetic diet for        LGA (above 90th 2002, after the
2005                             treated with    insufficient.      pregnancy with 3         percentile)        introduction of             Larger randomised trials
                                 glibenclamide                      meals and 3 snacks;      Macrosomia         glibenclamide, 236          are needed to investigate
179
                                                 Exclusions fasting and instructions on      (≥4000g)           women were diagnosed        less frequent
                                                 plasma glucose: SBGM (4 times per           Hyperbilirubinemi with GDM and treated         complications such as
                                                 >7.7 mmol/l        day: fasting and         a (total bilirubin with glibenclamide.         preeclampsia, NICU
                                                                    either 1 or 2 hours      12mg/dL or                                     admission, need for
                                                 Population of      postprandial).           greater within the Women in the insulin        phototherapy and birth
                                                 managed care       Targets for fasting,     first 7 days of    group had a higher          injury.
                                                 organisation       1-hr and 2-hr            birth)             mean BMI (31.9 vs 30.6,
                                                 (does not include postprandial were         Polycythemia       p=0.04) a greater
                                                 extremes of        <5.5, <8.5 and <7        (hematocrit        proportion identified
                                                 socio-economic     respectively.            >60%)              themselves as white
                                                 distribution).                              Hypocalcaemia      (43%, 28%, p<0.001)
                                                                    Patients were            (calcium           and fewer as Asian
                                                 Country: USA       expected to have         7.0mg/dL or less (24%, 37%, p=0.001)
                                                                    BS evaluation for a      within 3 days of   OGTT performed at a
                                                                    minimum of 3 days        birth).            later gestational age
                                                                    on an appropriate        NICU admission, (26.3±5.3 weeks vs
                                                                    diet before initiating   length of stay,    25.5±5.3weeks, p=0.02)
                                                                    glibenclamide.           need for           and a significantly
                                                                                             oxygen/assisted higher fasting OGTT
                                                                    Glibenclamide was        ventilation        (p=0.005) compared
                                                                    begun with an initial    Maternal           with the glibenclamide
                                                                    daily dose of 2.5 mg     hypoglycaemia      group.
                                                                    with the morning         (<3.3mmol/l)
                                                                    meal. If targets were                       After logistic regression
                                                                    not met on a                                (BMI, ethnicity,
                                                                    maximum dose of                             gestational age at
                                                                    20mg insulin was                            diagnosis, fasting
                                                                    initiated.                                  plasma glucose on
                                                                                                                OGTT) glibenclamide
                                                                    Comparison: Insulin                         treatment was
                                                                                                                significantly more likely
                                                                                                                to be associated with
                                                                                                                achieving mean glucose




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 153 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                                                         goals (adjusted OR
                                                                                         0.27, 95% CI 0.13-0.52).

                                                                                         Adjusted OR (insulin
                                                                                         group s reference):
                                                                                         Preeclampsia: 2.32
                                                                                         (95% CI 1.17-4.63)
                                                                                         NICU admission: 0.7
                                                                                         (0.34-0.93).
                                                                                         Phototherapy: 2.20
                                                                                         (0.99-4.87)
                                                                                         No other differences in
                                                                                         maternal or neonatal
                                                                                         outcomes.


                                                                                         28 (12%) women failed
                                                                                         glibenclamide and were
                                                                                         switched to insulin: 8 for
                                                                                         side effects primarily
                                                                                         attributed to
                                                                                         hypoglycaemia, 14 for
                                                                                         poor control and for 6
                                                                                         the reason was not
                                                                                         clear. For these women
                                                                                         the mean BMI was 31.6,
                                                                                         mean fasting level on
                                                                                         OGTT 5.7 mmol/l and
                                                                                         mean pregnancy eight
                                                                                         gain 25.6 lb. Only 3
                                                                                         women who switched for
                                                                                         poor control were on
                                                                                         maximum dose of
                                                                                         20mg/d.
                                                                                         An additional 11 (5%)
                                                                                         discontinued
                                                                                         glibenclamide, most for
                                                                                         side effects attributed to
                                                                                         hypoglycaemia, and
                                                                                         never started insulin.

                                                                                         3 infants in the insulin




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 154 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                                                         group had birth injuries
                                                                                         (1 clavicle fracture, 1
                                                                                         brachial plexus injury, 1
                                                                                         bone injury). 8 infants in
                                                                                         the glibenclamide group
                                                                                         had birth injuries (4
                                                                                         clavicle fractures, 1
                                                                                         brachial plexus injury, 3
                                                                                         bone injuries)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 155 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &            Effect Size             Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                                   Comments
                 Level                 cs                                                    Measures
Ostlund I;Hanson    Study Type:       213 women                       Intervention: Is     Follow-up period: Caesarean section           IGT is independently and Higher caesarean rate not
U;Bjorklund         Case-control      with                            increased maternal                     Adjusted OR (adjusted       significantly associated  linked to diagnosis of
A;Hjertberg R;Eva                     undiagnosed     Country: Sweden or neonatal          Outcome           for parity, LGA infant,     with an increased         GDM
N;Nordlander        Evidence level:   and untreated                   morbidity in         Measures:         PIH/preeclampsia,           incidence of caesarean
E;Swahn             2++               IGT during                      connection with      Caesarean         ethnicity, BMI) 1.9 (1.2-   section and prematurity
ML;Wager J;                           pregnancy                       Impaired glucose     section           2.9).                       and a markedly
                                                                      tolerance (IGT)      Prematurity (<37 Emergency caesarean          increased proportion of
2003 Jul                                                              during pregnancy     weeks)            section adjusted OR 2.1     LGA and macrsomic
                                                                      when the condition   Birthweight       (1.3-3.5 mmol/l)            infants and admission to
146
                                                                      is not treated.      Macrosomia (≥                                 NICU for 2 days or longer
                                                                                           4000g, ≥ 4500g, ≥ Prematurity adjusted OR
                                                                       GDM defined as      5000g)            2.0 (95% CI 1.0-3.9)
                                                                       fasting blood       LGA
                                                                       glucose ≥ 6.7       SGA               LGA adjusted OR 7.3
                                                                       mmol/l and/or 2-h   Hypoglycaemia     (95% CI 4.1-12.7).
                                                                       blood glucose ≥11.1 Hyperbilirubinemi
                                                                       mmol/l              a requiring       Admission to NICU
                                                                                           treatment         adjusted OR adjusted
                                                                       IGT fasting blood   Admission to      OR 2.3 (95% CI 1.3-4.0)
                                                                       glucose <6.7 mmol/l NICU 2 days or
                                                                       and 2-hr level 9.0- longer            71.3% of infants in the
                                                                       11.0 mmol/                            IGT group had no
                                                                                                             neonatal complications
                                                                       Comparison: 815                       compared with 87.3% of
                                                                       controls                              the control group.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 156 of 441
                                                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                     Follow-up &             Effect Size              Study Summary                    Reviewer
 Information & Evidence   Patients Characteristi Comparison                                          Outcome                                                                         Comments
                 Level                 cs                                                            Measures
Langer, O.,        Study Type:       555 women        Exclusion criteria    Intervention:        Follow-up period: Composite variable              Untreated gestational        The sample size was
Yogev, Y., Most,   Case-control      with untreated   included              Treatment for                             Untreated: 327/555 (59)      diabetes carries             caculated to detect
O., Xenakis, E.                      gestational      established           gestational diabetes Outcome              Treated: 197/1110 (18)       significant risks for        reductions of 30% in the
                   Evidence level:   diabetes         pregestational        (SMBG, diet alone    Measures:            Nondiabetic: 126/1110        perinatal morbidity in all   primary outcome rate for
2005               2++               1110 women       diabetes,             or insulin +diet).   Primary outcome: (11)                             disease severity levels.     the treated and
                                     treated for      substance abuse,      Patients remained in a composite          OR untreated GDM vs                                       nondiabetic control
408
                                     gestational      mulitfetal            diet group if OGTT   variable             nondiabetic 11.20                                         groups with a type 1 error
                                     diabetes         gestation.            at fasting <5.3      composed of          (8.71-14.39)                                              of 5% (2-sided) and
                                     1110             GDM in untreated      mmol/l, 2-hr         stillbirth, neonatal OR treated GDM vs                                         power of 80%.
                                     nondiabetic      group diagnosed       postprandial levels  macrosomia/LGA, nondiabetic1.69 (1.33-
                                     women            after week 37.        <6.6 mmol/l and      neonatal             2.15)
                                                      Groups were           mean blood glucose   hypoglycaemia,
                                                      matched on            levels <5.5 mmol/l.  erythrocytosis       Stratification by BMI:
                                                      obesity, parity,      Diet involved caloric                     In the untreated group
                                                      gestational age at    restriction using 25 Total pregnancy there was a significantly
                                                      delivery, ethnicity   (overweight/obese)   weight gain          higher rate of adverse
                                                      and number of         to 35 (normal        Fetal growth and outcome in the
                                                      prenatal visits.      weight) kcal/kg for  wellbeing            overweight group
                                                                            actual pregnancy     assessments          compared to normal
                                                      Country: USA          weight with a        LGA (>90th           weight group.
                                                                            recommended diet     percentile)          Comparable rates of
                                                                            of 3 meals and 4     Macrosomia           adverse outcome were
                                                                            snacks. Patients     (birthweight>4000 found between
                                                                            who did not achieve  g)                   overweight and normal
                                                                            established levels ofNeonatal             weight groups in the
                                                                            glycaemic control    ponderal index       treated and nondiabetic
                                                                            after 2 weeks on     >2.85                groups.
                                                                            diet therapy         Respiratory
                                                                            received insulin     distress             Stratification by severity
                                                                            therapy (both        Neonatal             category:
                                                                            intermediate and/or  hypoglycaemia (2 The untreated group
                                                                            short-acting human   consecutive          had a higher level of
                                                                            insulin).            values <2.2          adverse outcomes in
                                                                                                 mmol/l)              comparison with the
                                                                            Comparison: No       Hyperbilirubineam treatment group in all
                                                                            treatment for        ia                   severity categories.
                                                                            gestational diabetes Erythrocytosis
                                                                                                 Hypocalcaemia




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                      Page 157 of 441
                                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &         Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                             Comments
                 Level                 cs                                                    Measures
Jovanovic, L.,   Study Type:       27 women with Age 29.7±6.9      Intervention: Insulin                  Mean reductions in
                                                                                           Follow-up period:                         The safety and
Howard, C.,      RCT               GDM           years, HbA1c      Aspart                  From the       HbA1c:                     effectiveness of Insulin
Pettitt, D.                                      <7% at diagnosis.                                        Insulin aspart; 0.3±0.5%
                                                                                           diagnosis of GDM                          Aspart was comparable
                 Evidence level:                                   5 minutes before                       Human
                                                                                           (18-28 weeks) to                          to Human Insulin in
2005             1+                              Country: USA      meal using              6 weeks        insulin:0.1±0.4%, NS       pregnant women. The
                                                                   NovoPen-3               postpartum     Upper respiratoy tract     ease of use of Insulin
79
                                                                                                          infection:                 Aspart injected just
                                                                     Both groups also    Outcome          Insulin aspart: 14%        before meals rather than
                                                                     received basal      Measures: Mean Human insulin:23%            30 minutes prior to meals
                                                                     insulin (Novolin N) reductions in    Hypoglycaemic events:      may offer a more
                                                                                         HbA1c            Insulin aspart: 0          convenient therapy for
                                                                     Comparison: Human Hypoglyaemic       Human insulin: 0           management of diabetes
                                                                     Insulin             events           Average insulin-specific   for patients with GDM
                                                                                         Insulin specific antibodies
                                                                     30 minutes before   antibodies       Insulin aspart:0.97%
                                                                     meal                Birthweight      Human insulin: 0.07%
                                                                                                          Average percentile
                                                                                                          birthweights:
                                                                                                          Insulin aspart:40%
                                                                                                          Human insulin:44%
                                                                                                          Overall safety profiles
                                                                                                          were similar for Insulin
                                                                                                          aspart and Human
                                                                                                          insulin treatment groups




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                       Page 158 of 441
                                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                      Follow-up &             Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                           Outcome                                                                     Comments
                 Level                 cs                                                             Measures
Mecacci, F., et al   Study Type:       25 women with    Diagnosed           Intervention: Insulin   Follow-up period: 1 hr postprandial blood       In women with GDM the
                     RCT               GDM received     according to C&C    lispro                                     glucose values (Total)       use of insulin lispro
2003                                   insulin lispro   criteria. Three                             Outcome            Insulin lispro:108.4±10.7    enabled the attainment of
                     Evidence level:   24 women with    main meals at       Comparison:             Measures: Blood Regular                         near-normal glucose
84
                     1++               GDM received     8.00am, 12.00,      Regular insulin         glucose profiles   insulin:121.0±13.2           levels at the 1-hr
                                       normal insulin   8.00pm.Insulin                              performed using a Controls:105.6±4.7,           postprandial time point
                                       50 non GDM       given before each                           reflectance meter p<0.01                        and was associated with
                                       controls         meal. No NPH                                and consisted of                                normal anthropometric
                                                        insulin.                                    9 determinations: Cranial-thoracic              characteristics; the use of
                                                                                                    fasting/preprandia circumference (CC/CT)        regular insulin was not
                                                        Country: Italy                              l, 1 and 2 hour    ratio (percentile rank, n,   able to blunt the 1 hr
                                                                                                    post-prandial.     %)                           postprandial response to
                                                                                                    HbA1c              <10                          a near-normal extent and
                                                                                                    Birthweight        Lispro:1(4)                  resulted in infants with a
                                                                                                    Ponderal index     Regular:2(8.3)               tendency toward
                                                                                                    Cranio-thoracic    Control:1(2)                 disproportionate growth.
                                                                                                    circumference      10-25                        Insulin lispro can be
                                                                                                                       Lispro:3 (12)                considered as a valuable
                                                                                                                       Regular:9 (37.5)             option for the treatment
                                                                                                                       Control:7 (14), L,C vs R     of gestational diabetes.
                                                                                                                       <0.05
                                                                                                                       26-50
                                                                                                                       Lispro:9 (36)
                                                                                                                       Regular:6 (25)
                                                                                                                       Control:19 (38)
                                                                                                                       51-75
                                                                                                                       Lispro:8 (32)
                                                                                                                       Regular:4 (16.7)
                                                                                                                       Control:17 (34)
                                                                                                                       76-90
                                                                                                                       Lispro:4 (16)
                                                                                                                       Regular:3 (12.5)
                                                                                                                       Control:5 (10)
                                                                                                                       >90
                                                                                                                       Lispro:-
                                                                                                                       Regular:-
                                                                                                                       Control:1 (2)



                                                                                                                       No other significant
                                                                                                                       differences between




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                       Page 159 of 441
                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size     Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                  Comments
                 Level                 cs                                   Measures
                                                                                         groups in neonatal
                                                                                         outcome.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                           Page 160 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                     Follow-up &           Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                          Outcome                                                               Comments
                 Level                 cs                                                            Measures
Algert S;Shragg   Study Type:       22 obese            GDM (O’Sullivan      Intervention:         Follow-up period: Average calorie intake:   Modest calorie reduction
P;Hollingsworth   Cohort            women with          and Mahan            Moderate calorie                        Control: 2282±524         to 25kcal/kg per day
DR;                                 GDM                 criteria)            restriction           Outcome           Lean women with GDM:      (1700-1800 kal/day) is a
                  Evidence level:   31 lean women       Obese group                                Measures:         1822±224                  reasonable and safe
1985 Apr          2+                with GDM            were significantly   Obese women with                        Obese women with          approach for obese
                                    (BMI<27)            older than other     GDM: 1700-1800                          GDM: 1750±188             women with GDM.
162
                                    10 nonobese         groups.              kcal/day                                kcal/day
                                    normal glucose                           (2 women received
                                    tolerant controls   Country:             small amounts of                        Mean weight gain during
                                                                             regular insulin                         pregnancy:
                                                                             before each meal)                       Control:13.4±3.5kg
                                                                                                                     Lean GDM:13.3±4.1kg
                                                                             Lean women with                         Obese
                                                                             GDM: 2000-                              GDM:10.6±7.7kg,
                                                                             3000kcal/day                            p<0.03

                                                                             Comparison: Control                     Ketonuria was not
                                                                             group:                                  observed in any patient
                                                                             recommended diet                        even when reported
                                                                             for pregnancy                           caloric intakes were as
                                                                                                                     low as 1500 to 1600 per
                                                                                                                     day.

                                                                                                                     Mean birthweight was
                                                                                                                     significantly higher in
                                                                                                                     obese women with GDM
                                                                                                                     than other two groups
                                                                                                                     despite a lower calric
                                                                                                                     intake and a smaller
                                                                                                                     weight gain during
                                                                                                                     pregnancy.

                                                                                                                     Mean birthweight
                                                                                                                     Control: 3448±303g
                                                                                                                     Lean GDM:3544±598g
                                                                                                                     Obese GDM:
                                                                                                                     3922±662g, p<0.03

                                                                                                                     All women had excellent
                                                                                                                     glycaemic control (2-hr
                                                                                                                     postbreakfast plasma
                                                                                                                     glucose levels <7).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 161 of 441
                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &       Effect Size     Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                 Comments
                 Level                 cs                                   Measures
                                                                                         HbA1c at 35 weeks
                                                                                         Control: 6.4±1.2%
                                                                                         Lean:6.7±0.7%
                                                                                         Obese:6.8±0.4%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                          Page 162 of 441
                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                   Follow-up &             Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                        Outcome                                                                    Comments
                 Level                 cs                                          Measures
Clapp III JE;   Study Type:       A cross-over              Intervention: Low GI Follow-up period:   The blood glucose           The average increase in
                RCT               study of 14               diet                                     values were significantly   blood sugar was
1998                              nonpregnant    Country:                          Outcome           higher on the High GI       significantly lower on the
                Evidence level:   nondiabetic               After 7-10 days on     Measures:         diet at each time point     low GI diet. Insulin
154
                1+                women                     the diet postprandial Glycaemic          through 120 minutes. As     response was also
                                                            glucose was            response          a result the average        significantly lower.
                                                            measured after a                         increase in blood sugar
                                                            540 kcal test meal.                      was significantly lower
                                                            Each meal                                (p<0.001) on the low GI
                                                            contained 17%                            diet ( 0.33 mmol/min vs
                                                            protein, 28% fat and                     0.72 mmol/min). The
                                                            55% carbohydrate                         insulin response was
                                                            but differed in the GI                   also significantly lower
                                                            of the carbohydrates                     (p<0.001) on the low GI
                                                            (54 s 92).                               diet (102 nmol/min vs
                                                                                                     168 nmol/min)
                                                            Comparison: Hi GI
                                                            diet




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                    Page 163 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                Follow-up &             Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                     Outcome                                                                    Comments
                 Level                 cs                                                       Measures
Dornhorst      Study Type:       35 women with   GDM diagnosed       Intervention: GDM        Follow-up period: 8/43 (18%) women with          Calorie restriction
A;Nicholls     Cohort            GDM             after O'Sullivan    women prescribed a                         GDM diagnosed during           reduced infant birth
JS;Probst                        2337            screen with 3-h     calorie-restricted       Outcome           the study required             weights to levels
F;Paterson     Evidence level:   nondiabetic     OGTT. 2337          diet with complex        Measures:         insulin in addition to diet.   comparable with the
CM;Hollier     2++               women           controls included   carbohydrates            Maternal weight                                  general prenatal
KL;Elkeles                                       two control         replacing refined        gain              Weight gain                    population with no
RS;Beard RW;                                     groups of 35        carbohydrates and        Birth weight      Booking to delivery (kg):      adverse maternal or
                                                 women each          fats. The total          Birth weight      GDM:4.6±4.9                    neonatal outcomes.
1991 Dec                                         matched for race,   calorie intake was       percentile        General prenatal
                                                 age, BMI, and       prescribed to be         Macrosomia        population: 9.3±5.3            Only18% required insulin
160
                                                 parity - Group A    30% less than            Admission to      Group A:9.7±5.3                in addition to diet.
                                                 (normal screen)     consumed before          special care baby Group B: 9.7±5.4,
                                                 and Group B         pregnancy. The           unit              p<0.005 (GDM vs.
                                                 (abnormal screen,   composition of the                         controls groups).
                                                 normal OGTT).       diet was 50%                               28 wk to delivery
                                                                     carbohydrate, 15%                          GDM:1.7±1.6
                                                 Country:            protein and 35% fat,                       Group A:4.1±3.1
                                                                     with most                                  Group B:4.6±2.9
                                                                     carbohydrate taken                         p<0.005 (GDM vs.
                                                                     as unrefined                               controls groups).
                                                                     carbohydrate,
                                                                     protein as fish, white                       Macrosomia (≥4000g)
                                                                     meat, beans and                              GDM:2 (6%)
                                                                     dairy products; and                          General postnatal
                                                                     fat as                                       population:178 (7.5%)
                                                                     polyunsaturated fat.                         Group A: 3(9%)
                                                                                                                  Group B:8 (23%)
                                                                     The criteria for                             P<0.005 vs. general
                                                                     remaining on dietary                         population
                                                                     treatment alone was
                                                                     a diurnal glucose                            Birthweight, birth weight
                                                                     profile of averaged                          percentile, gestational
                                                                     over six                                     age and admission to
                                                                     measurements of                              baby care unit were
                                                                     ≤6mM.                                        similar across all groups.

                                                                     Comparison: Control
                                                                     group given no
                                                                     specific dietary
                                                                     advice.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 164 of 441
                                                                                                                                       DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &        Effect Size           Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                           Comments
                 Level                 cs                                                        Measures
Gillen L;Tapsell   Study Type:       16 women with                                          Follow-up period: The GDM group
                                                                        Intervention: Diet in                                          Compared with glucose
LC;Martin          Case-control      newly                              women with newly                      consumed significantly   tolerant women with
GS;Daniells                          diagnosed     Country: Australia   diagnosed GDM       Outcome           less total energy and    GDM reported less
S;Knights          Evidence level:   GDM                                                    Measures:         carbohydrate from low    carbohydrate rich foods
S;Moses RG;        2+                24 healthy                         Comparison: Health                    glycaemic index (GI)     with low GI values and a
                                     pregnant                           pregnant women                        foods. (p=0.01).         reduced spread of
2002                                 women                              matched for age,                                               consumption of foods
                                                                        prepregnancy                                                   with higher GI values.
156
                                                                        weight at gestation
                                                                        age.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 165 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                         Follow-up &               Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                              Outcome                                                                     Comments
                 Level                 cs                                                Measures
Nolan CJ;      Study Type:       5 women with   mean gestation   Intervention: Low fat Follow-up period:     A significant (p<0.05)      A diet rich in unrefined
               RCT               gestational    33.4±1.4 weeks   high unrefined        4 days                improvement in glucose      carbohydrate was
1984 Aug                         diabetes       gestation        carbohydrate diet                           tolerance was observed      associated with
               Evidence level:                                                         Outcome               in the high unrefined       improvement in glucose
157
               1+                               Country: NZ      Comparison: Low       Measures:             carbohydrate (HUC) diet     tolerance, reductions in
                                                                 carbohydrate diet     Fasting plasma        compared to the low         urinary glucose output,
                                                                                       glucose               carbohydrate (LC) diet.     fasting FFA and
                                                                                       Urinary glucose       Urinary glucose output      cholesterol
                                                                                       output                was 50% lower during        concentrations.
                                                                                       Fasting plasma        the HUC diet than
                                                                                       triglyceride levels   during the LC diet.
                                                                                       Fasting plasma        Fasting plasma
                                                                                       cholesterol           triglyceride levels were
                                                                                       Concentrations        generally lower.
                                                                                       Fasting plasma        Fasting plasma
                                                                                       free fatty acid       cholesterol
                                                                                       levels                concentrations were 6%
                                                                                       Glucose tolerance     lower on the NUC diet
                                                                                                             (p<0.01)
                                                                                                             Fasting plasma free fatty
                                                                                                             acids were 14% lower
                                                                                                             (p<0.02)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 166 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                                  Comments
                 Level                 cs                                                        Measures
Ostman EM;Frid   Study Type:       7 women with   Mean age 32          Intervention: Low       Follow-up period: All women lowered their     Replacing high GI/Low
AH;Groop         RCT               impaired       (range 27-           GI/high dietary fibre                       insulin response to the   dietary fibre bread with
LC;Bjorck IME;                     glucose        41years). Mean       bread                   Outcome             intravenous glucose       low GI/high dietary fibre
                 Evidence level:   tolerance      BMI 28.3±4.9.        Cross over design:      Measures:           challenge (mean 35%).     bread improved insulin
2006             1+                               Previous             3 weeks each arm        Fasting blood       No changes were found     economy in women with
                                                  diagnosis of         with 3 weeks            glucose             in fasting levels of      impaired glucose
158
                                                  gestational          washout period.         insulin             glucose, insulin, HDL     tolerance.
                                                  diabetes in last 3                           serum lipids        cholesterol or TG.
                                                  years. Diagnosed     Comparison: High        Intravenous
                                                  with IGT at follow   GI/low dietary fibre    glucose tolerance
                                                  up (WHO criteria).   bread                   test followed by a
                                                                                               euglycaemic-
                                                  Country:                                     hyperinsulinaemic
                                                                                               clamp was
                                                                                               performed to
                                                                                               assess potential
                                                                                               effects of the two
                                                                                               dietary regimens
                                                                                               on insulin
                                                                                               requirement and
                                                                                               insulin resistance.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                Page 167 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                    Follow-up &        Effect Size          Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                         Outcome                                                          Comments
                 Level                 cs                                                           Measures
Wang Y;Storlien     Study Type:       56 women with    Chinese             Intervention: Diet Follow-up period: Weight                   Gluocse intolerance and
LH;Jenkins          Case-control      GDM              GDM diagnosed       (24-hr diet recall at                GDM: 68.6±1.2kg          gestational diabetes were
AB;Tapsell LC;Jin                     38 women with    using NDDG          time of screening).Outcome           IGT: 66.1±1.4kg          associated independently
Y;Pan JF;Shao       Evidence level:   glucose          criteria for OGTT                      Measures:         Controls: 61.2±0.8kg     with a reduced intake of
YF;Calvert          2+                intolerance      performed           Comparison: IGT (1                   (p<0.05 GDM and IGT      polyunsaturated fats.
GD;Moses                              77 women with    between 24 and      value met on OGTT                    compared to controls).
RG;Shi HL;Zhu                         normal glucose   28 weeks of         by NDDG criteria)                    BMI
XX;                                   tolerance        pregnancy.          Normal glucose                       GDM: 26.4±0.4
                                                                           tolerance                            IGT: 25.7±0.5
2000 Apr                                               Country:                                                 Controls: 24.2 ±0.31
                                                                                                                (p<0.05 GDM and IGT
409
                                                                                                                compared to controls).

                                                                                                                 Total energy (kcal)
                                                                                                                 GDM:2134±61
                                                                                                                 IGT: 2223±75
                                                                                                                 Controls: 2133±42

                                                                                                                 Fat (%kcal)
                                                                                                                 GDM:30.0±1.0
                                                                                                                 IGT: 30.6±1.0
                                                                                                                 Controls: 32.4±0.8

                                                                                                                 Protein (%kcal)
                                                                                                                 GDM:16.2±0.4
                                                                                                                 IGT: 17.2±0.5
                                                                                                                 Controls:16.3±0.3

                                                                                                                 Carbohydrate (%kcal)
                                                                                                                 GDM: 53.8±1.2
                                                                                                                 IGT: 52.3±1.3
                                                                                                                 Controls:51.7±0.9

                                                                                                                 Fat profile (%fat)
                                                                                                                 Polyunsaturated:
                                                                                                                 GDM:28.2±0.9
                                                                                                                 IGT: 29.5±1.0
                                                                                                                 Controls: 31.6±0.7,
                                                                                                                 p<0.05
                                                                                                                 Monosaturated
                                                                                                                 GDM: 25.7±0.8
                                                                                                                 IGT: 28.8±1.0




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 168 of 441
                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size      Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                   Comments
                 Level                 cs                                   Measures
                                                                                         Controls: 26.3±0.6
                                                                                         Saturated:
                                                                                         GDM: 46.1±1.4
                                                                                         IGT: 41.8±1.6
                                                                                         Controls: 42.1±1.0,
                                                                                         p<0.05
                                                                                         P:S ratio
                                                                                         GDM: 0.67±0.04
                                                                                         IGT: 0.77±0.05
                                                                                         Controls: 0.81±0.04
                                                                                         Fiber (g)
                                                                                         GDM: 13.2±0.7
                                                                                         IGT: 14.4±1.3
                                                                                         Controls: 14.1±0.8




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                            Page 169 of 441
                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                 Follow-up &            Effect Size             Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                      Outcome                                                                 Comments
                 Level                 cs                                        Measures
Clapp, J.      Study Type:       10 pregnant                Intervention: Low   Follow-up period: Fasting glucose - Mean     Dietary carbohydrates
               RCT               women on low               glycaemic diet in                       (SE)                     which elevate
2002                             glycaemic index Country:   pregnancy           Outcome             Mid pregnancy            postprandial glucose
               Evidence level:   diet                                           Measures:           Low GI: 3.83 (0.11)      levels in mid and late
155
               1+                10 women on                Comparison: High    Postprandial        High GI:4.00 (0.17)      pregnancy markedly
                                 high glycaemic             glycaemic diet      glucose response Late pregnancy              increase feto-placental
                                 index diet                                     Postprandial        Low GI: 3.78 (0.11)      growth rate. The type of
                                                                                insulin response High GI: 4.28 (0.22)        carbohydrate eaten
                                                                                Insulin resistance p<0.05                    during pregnancy
                                                                                index                                        influences maternal fat
                                                                                Insulin sensitivity Fasting insulin          deposition.
                                                                                index (after        Mid pregnancy
                                                                                Matsuda and         Low GI: 67 (9)
                                                                                DeFronzo 1999) High GI: 83(11)
                                                                                                    Late pregnancy
                                                                                Maternal weight     Low GI: 63 (9)
                                                                                gain                High GI: 131 (15),
                                                                                Birthweight         p<0.05

                                                                                                  Average postprandial
                                                                                                  glucose increase
                                                                                                  Mid pregnancy
                                                                                                  Low GI: 0.61 (0.18)
                                                                                                  High GI: 1.44 (0.17),
                                                                                                  p<0.05
                                                                                                  Late pregnancy
                                                                                                  Low GI: 0.60 (0.23)
                                                                                                  High GI: 1.56 (0.22),
                                                                                                  p<0.05

                                                                                                  Average insulin increase
                                                                                                  Mid pregnancy
                                                                                                  Low GI: 138 (300
                                                                                                  High GI: 234 (36),
                                                                                                  p<0.05
                                                                                                  Late pregnancy
                                                                                                  Low GI: 186 (30)
                                                                                                  High GI: 324 (48)
                                                                                                  p<0.05

                                                                                                  Insulin resistance index
                                                                                                  Mid pregnancy




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                               Page 170 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         Low GI: 257
                                                                                         High GI: 336, p<0.05
                                                                                         Late pregnancy
                                                                                         Low GI: 238
                                                                                         High GI: 536, p<0.05

                                                                                         Insulin sensitivity index
                                                                                         Mid pregnancy
                                                                                         Low GI: 0.53
                                                                                         High GI: 0.40, p<0.05
                                                                                         Late pregnancy
                                                                                         Low GI: 0.54
                                                                                         High GI: 0.24, p<0.05

                                                                                         Birthweight (kg) - Mean
                                                                                         (SE)
                                                                                         Low GI:3.33 (0.11)
                                                                                         High GI:4.17 (0.12),
                                                                                         p<0.01

                                                                                         Ponderal index:
                                                                                         Low GI:2.47 (0.08)
                                                                                         High GI:2.74 (0.04),
                                                                                         p<0.01

                                                                                         Fat mass (g)
                                                                                         Low GI:266 (44)
                                                                                         High GI:427 (61),
                                                                                         p<0.01

                                                                                         Maternal weight gain
                                                                                         (kg)
                                                                                         Low GI:10.4 (1.1)
                                                                                         High GI:18.6 (1.1),
                                                                                         p<0.01




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 171 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &              Effect Size              Study Summary                 Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                                       Comments
                 Level                 cs                                                 Measures
Fraser, R., Ford,   Study Type:       15 non                      Intervention: Diet   Follow-up period: On diet 1 there was a           The loss of insulin
F., Lawrence, G.    RCT               pregnant                                                             loss of insulin sensitivity   sensitivity which is typical
                                      women         Country: UK   Comparison: Diet 1: Outcome              in pregnancy but not on       of western women in the
1988                Evidence level:   14 pregnant                 40% energy as        Measures:           diets 2 or 3.                 third trimester of
                    1++               women                       carbohydrate         Insulin sensitivity                               pregnancy and which is
153
                                                                  10% as dietary fibre                                                   considered to be
                                                                  Diet 2: 40% energy                                                     physiological may be a
                                                                  as carbohydrate                                                        diet induced artefact.
                                                                  52g dietary fibre
                                                                  Diet 3. 60% energy
                                                                  as carbohydrate and
                                                                  84g dietary fibre




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 172 of 441
                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                      Follow-up &            Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                           Outcome                                                                Comments
                 Level                 cs                                             Measures
Fraser, B.,    Study Type:       64 nonpregnant African          Intervention:       Follow-up period: Mean (1 SD) blood         Normal pregnant women
               Cohort            women                           Glucose tolerance                     glucose levels            in Nairobi have a lower
1981                             27 women at 30 Country: Kenya   testing             Outcome           Fasting                   fasting level of blood
               Evidence level:   weeks                                               Measures: Blood Non pregnant: 73.5          glucose than non prgnant
152
               2++               gestation                       Comparison:         glucose levels    (13.1)                    women. The
                                 31 women after                                                        Week 30: 67.0 (7.4)       diabetogenic influence of
                                 37 weeks                                                              Week 37+:66.5 (9.6),      pregnancy is not
                                 gestation                                                             p=0.05                    demonstrated in this
                                                                                                                                 series; in fact the
                                                                                                       Peak value                opposite is the case as
                                                                                                       Non pregnant:137.4        pregnant women have
                                                                                                       (22.01)                   improved homeostasis
                                                                                                       Week 30:123.6 (25.2)      with advancing
                                                                                                       Week 37+:118.3 (23.3),    pregnancy.
                                                                                                       p=0.05

                                                                                                       One hour:
                                                                                                       Non pregnant:129 (23.8)
                                                                                                       Week 30:109.7 (26.8)
                                                                                                       Week 37+:108.1 (25.0),
                                                                                                       p=0.05

                                                                                                       Mean +2SD (upper
                                                                                                       criteria for normality)
                                                                                                       Fasting
                                                                                                       Non pregnant:100
                                                                                                       Week 30:82
                                                                                                       Week 37+:86

                                                                                                       Peak value
                                                                                                       Non pregnant:181
                                                                                                       Week 30:174
                                                                                                       Week 37+:165

                                                                                                       One hour:
                                                                                                       Non pregnant:177
                                                                                                       Week 30:163
                                                                                                       Week 37+:158




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 173 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                    Follow-up &           Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                                         Outcome                                                           Comments
                 Level                 cs                                                           Measures
Bonomo, M., et al. Study Type:       151 women         Women               Intervention: US       Follow-up period: Insulin treatment was
                   RCT               with GDM          diagnosed with      exams were                                required in 16.7% of
2004                                 managed           GDM between 24      scheduled every two    Outcome            control group and 30.5%
                   Evidence level:   according to US   and 28 weeks        weeks in the           Measures:          of intervention group
176
                   1+                assessment        gestation.          intervention group     Macrosomia         (p=0.024)
                                     78 GDM            Diagnosed using                            (birthweight
                                     controls          50g GCT screen                             ≥4000g             The intervention group
                                                       followed by 100g                           LGA                had significantly lower
                                                       OGTT (C&C           BG targets for         Neonatal           incidence of LGA and
                                                       criteria).          intervention group     hypoglycaemia      macrosomia
                                                       Exclusions: Other   were 4.4 mmol/l        Admission to
                                                       medical             fasting and 5.5        neonatal intensive
                                                       complications       mmol/l postprandial    care unit
                                                       affecting fetal     if ≥ 75th abdominal
                                                       growth. All         circumference
                                                       women on SMBG       centile and
                                                       and diet or diet    5.5mmol/l fasting
                                                       and insulin if bg   and 7.7 mmol/l
                                                       exceeded target     postprandial if <
                                                       values.             75th abdominal
                                                                           circumference
                                                       Country: Italy      centile

                                                                           Comparison: US
                                                                           exams were
                                                                           scheduled at 34 and
                                                                           38 weeks in the
                                                                           control group.

                                                                           BG targets for
                                                                           control group were 5
                                                                           mmol/l fasting and
                                                                           6.6 mmol/l
                                                                           postprandial.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 174 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                  Follow-up &         Effect Size             Study Summary                   Reviewer
 Information & Evidence   Patients Characteristi Comparison                                       Outcome                                                                   Comments
                 Level                 cs                                                         Measures
Brankston, G.,      Study Type:       32 women with GDM diagnosed      Intervention: Diet + Follow-up period:   32 women completed          Resistance exercise may   Diagnostic fasting and 1-
Mitchell, B., Ryan, RCT               gestational   after 50g screen   Resistance exercise                      the study                   help to avoid insulin     hour values were
E.                                    diabetes      (GDM diagnosed                          Outcome                                         therapy for overweight    significantly higher in the
                    Evidence level:                 if ≥10.3 mmol/l)   Diet: 40%            Measures:           Required insulin            women with gestational    diet alone group
2004                1++                             and 75g OGTT       carbohydrate, 20% Requirement for        Diet: 9/16 (56.3%)          diabetes.
                                                    (Canadian          protein, 40% fat,    insulin             Diet +exercise: 7/16                                  A larger sample size
165
                                                    Diabetic           calculated at 24 to  Gestational age     (43.8%), p=0.48                                       (n=56) would be needed
                                                    Association        30 cal/kg per day on at delivery                                                               to show that the effect
                                                    Criteria).         basis of subject’s   Birthweight         Amount of insulin                                     that was shown (12.5%)
                                                    Gestational age    ideal prepregnant                        required:                                             was significant
                                                    between 26 and     body weight, and                         Diet:0.48±0.3
                                                    32 weeks, BMI <4   divided into 3 meals                     Diet+insulin: 0.22±0.2,
                                                                       and 3 snacks.                            p<0.05
                                                    Country: Canada
                                                                       Circuit type                             Latency to insulin
                                                                       resistance exercise                      requirement:
                                                                       3xweek. Exercise                         Diet:1.11±0.8
                                                                       intensity                                Diet+insulin:371±3.1,
                                                                       progressively                            p<0.05
                                                                       increased.

                                                                       Glucose monitoring                       Within the diet+exercise
                                                                       in both groups                           group 30% of the
                                                                       included daily                           women who exercised 2
                                                                       fasting and 1 or 2                       to 3 times per week
                                                                       hour postprandial                        were prescribed insulin
                                                                       measurements.                            therapy compared to
                                                                                                                67% of those who
                                                                       Insulin therapy                          exercised 0 to 1.9 times.
                                                                       initiated if any of the
                                                                       following were                           A subgroup analysis that
                                                                       exceeded                                 looked only at women
                                                                       consistently:                            with a prepregnancy
                                                                       Mean fasting ≥5.3                        BMI>25
                                                                       mmol/l                                   found that women in the
                                                                       Mean 1-hour                              diet+exercise group
                                                                       postprandial ≥7.8                        showed a significantly
                                                                       Mean 2-hour                              lower incidence of
                                                                       postprandial ≥6.7                        insulin use (3/10 vs
                                                                       mmol/l.                                  8/10, p<0.05).
                                                                       These values were
                                                                       also goals for insulin                   Home blood glucose




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 175 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                        therapy.                         levels (4 women did not
                                                                                         record adequately)
                                                        Comparison: Diet                 Fasting
                                                                                         Diet (n=12): 5.1±0.65
                                                                                         Diet+exercise
                                                                                         (n=12):4.7±0.39, p=0.07
                                                                                         Pooled 2-hour
                                                                                         postprandial:
                                                                                         Diet: 6.4±0.81
                                                                                         Diet+exercise:6.0±0.29,
                                                                                         p<0.05

                                                                                         There were no
                                                                                         significant differences in
                                                                                         gestational age at
                                                                                         delivery, rate of
                                                                                         caesarean deliveries or
                                                                                         birthweight.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 176 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                Follow-up &           Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                     Outcome                                                                 Comments
                 Level                 cs                                                       Measures
Pettitt, D.,   Study Type:       15 women with       Women with        Intervention: Insulin  Follow-up period: Glucose                     Insulin aspart was
               RCT               gestational         GDM unable to     Aspart (5 minutes                        60 minutes:                 effective in decreasing
2003                             diabetes            control glucose   before a meal)         Outcome           No exogenous                the postprandial glucose
               Evidence level:   randomised to       concentrations                           Measures:         insulin:122.8±3.0           concentration when
185
               1+                insulin aspart of   with diet and     Randomised cross Plasma glucose          Regular                     administered 5 minutes
                                 regular human       exercise          over study:            concentration     insulin:115.8±2.9,          before a meal to women
                                 insulin                               Day 1: Test meal       area under the    p<0.05                      with GDM.
                                                     Country: USA      with no insulin        curve and         Insulin aspart:111.6±3.9,
                                                                       Day 2: Test meal       Serum Insulin     p=0.01                      Regular human insulin
                                                                       with either insulin or area under the    90 minutes:                 failed to make a
                                                                       aspart                 curve calculated No exogenous                 significant impact in
                                                                       Day 3: Test meal       at 120, 180 and   insulin:112.8±2.8           lowering the postprandial
                                                                       with the other insulin 240 minutes.      Regular                     glucose concentration.
                                                                       preparation                              insulin:108.4±2.6,
                                                                       Test meal consisted                      p<0.05
                                                                       of yoghurt and                           Insulin aspart:103.3±4.3,
                                                                       banana (40%                              p<0.05
                                                                       carbohydrate, 20%                        120 minutes:
                                                                       protein and 40%                          No exogenous
                                                                       fat).                                    insulin:102.5±2.5
                                                                                                                Regular
                                                                       Comparison:                              insulin:100.6±2.5
                                                                       Regular insulin (30                      Insulin aspart:93.1±3.5,
                                                                       minutes before a                         p=0.01
                                                                       meal)                                    180 minutes:
                                                                                                                No exogenous
                                                                                                                insulin:85.1±2.1
                                                                                                                Regular
                                                                                                                insulin:79.9±2.6, p<0.05
                                                                                                                Insulin aspart:72.5±3.4,
                                                                                                                p=0.01, (p<0.05 apart vs
                                                                                                                regular insulin)
                                                                                                                240 minutes:
                                                                                                                No exogenous
                                                                                                                insulin:74.7±1.7
                                                                                                                Regular
                                                                                                                insulin:66.7±2.0, p=0.01
                                                                                                                Insulin aspart:65.5±2.0,
                                                                                                                p=0.01

                                                                                                                Insulin:
                                                                                                                60 minutes:




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 177 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         No exogenous
                                                                                         insulin:72.3±9.7
                                                                                         Regular
                                                                                         insulin:84.7±10.8,
                                                                                         p<0.05
                                                                                         Insulin aspart:95.9±10.9,
                                                                                         p=0.01
                                                                                         90 minutes:
                                                                                         No exogenous
                                                                                         insulin:53.9±8.6
                                                                                         Regular
                                                                                         insulin:68.2±10.4,
                                                                                         p<0.05
                                                                                         Insulin aspart:77.4±8.8,
                                                                                         p=0.01
                                                                                         120 minutes:
                                                                                         No exogenous
                                                                                         insulin:35.6±6.2
                                                                                         Regular
                                                                                         insulin:55.2±9.4, p=0.01
                                                                                         Insulin aspart:51.7±9.8,
                                                                                         p=0.01
                                                                                         180 minutes:
                                                                                         No exogenous
                                                                                         insulin:20.0±4.1
                                                                                         Regular
                                                                                         insulin:26.9±4.5, p<0.05
                                                                                         Insulin aspart:24.6±4.6
                                                                                         240 minutes:
                                                                                         No exogenous
                                                                                         insulin:11.2±3.0
                                                                                         Regular
                                                                                         insulin:18.8±3.3, p=0.01
                                                                                         Insulin aspart:15.4±1.9,
                                                                                         p<0.05

                                                                                         C-peptide
                                                                                         60 minutes:
                                                                                         No exogenous
                                                                                         insulin:6.8±0.7
                                                                                         Regular insulin:5.8±0.4
                                                                                         Insulin aspart:5.5±0.3,




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 178 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         p<0.05
                                                                                         90 minutes:
                                                                                         No exogenous
                                                                                         insulin:6.2±0.6
                                                                                         Regular insulin:5.8±0.5
                                                                                         Insulin aspart:5.4±0.5
                                                                                         120 minutes:
                                                                                         No exogenous
                                                                                         insulin:5.4±0.7
                                                                                         Regular insulin:5.4±0.6
                                                                                         Insulin aspart:4.8±0.7
                                                                                         180 minutes:
                                                                                         No exogenous
                                                                                         insulin:3.9±0.5
                                                                                         Regular insulin:3.1±0.4
                                                                                         Insulin aspart:2.7±0.4,
                                                                                         p=0.01 (p<0.05 aspart
                                                                                         vs regular insulin)
                                                                                         240 minutes:
                                                                                         No exogenous
                                                                                         insulin:2.6±0.5
                                                                                         Regular insulin:1.8±0.4,
                                                                                         p=0.01
                                                                                         Insulin aspart:1.5±0.2,
                                                                                         p=0.01 (p<0.05 aspart
                                                                                         vs regular human
                                                                                         insulin)

                                                                                         Glucose areas under the
                                                                                         curve were significantly
                                                                                         lower with insulin aspart
                                                                                         but not with regular
                                                                                         human insulin.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 179 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                 Follow-up &            Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                      Outcome                                                                    Comments
                 Level                 cs                                                        Measures
Conway, D.,     Study Type:       75 women with     GDM (ADA              Intervention:        Follow-up period: 63/75 women (84%)             It appears that women
Gonzales, O.,   Cohort            GDM treated       criteria).            Glibenclamide                          treated with                  with GDM can be
Skiver, D.                        with glyburide    All women on diet                          Outcome           glibenclamide were able       successfully treated with
                Evidence level:   (glibenclamide)   (40%                  Patiets started on   Measures:         to achieve good               glibenclamide when diet
2004            2+                                  carbohydrate,         2.5mg/day.                             glycemic control.             therapy alone fails to
                                                    35% protein, 25%                                             12(16%) switched to           reduce blood glucose
180
                                                    fat) and SBGM         Comparison:                            insulin prior to delivery.    values sufficiently.
                                                    7x/day (before
                                                    and 2-h after                                                 The two groups were          Women who are farther
                                                    each meal and                                                 similar in terms of          along the continuum to
                                                    before bed).                                                  baseline characteristics     type 2 diabetes, as
                                                    Targets mean                                                  and diabetes risk factors    evidenced by an
                                                    level ≤5.8, fasting                                           but women who were           accumulation of risk
                                                    ≤5.2, 2-hr                                                    successfully treated with    factors, or fasting
                                                    postprandial ≤6.3.                                            glibenclamide were had       hyperglycaemia, may be
                                                    If good control                                               significantly lower values   better served by
                                                    was not acieved                                               on the OGTT.                 treatment with insulin
                                                    within 2 weeks of                                                                          rather than glyburide.
                                                    diet patients                                                 Fasting on OGTT:
                                                    offered glyburide                                             Glibenclamide:5.61±0.7
                                                    as alternative to                                             7
                                                    insulin (if                                                   Insulin:6.3±1.32, p=0.02
                                                    gestational ages                                              1-h value
                                                    11-33weeks and                                                Glibenclamide:11.28±1.
                                                    OGTT fasting                                                  27
                                                    level <7.7).                                                  Insulin:12.65±1.98,
                                                                                                                  p<0.01
                                                    Country:                                                      2-h value
                                                                                                                  Glibenclamide:9.30±1.8
                                                                                                                  7
                                                                                                                  Insulin:11.22±3.63,
                                                                                                                  p<0.01
                                                                                                                  3-h value
                                                                                                                  Glibenclamide:7.32±1.8
                                                                                                                  2, p<0.01
                                                                                                                  Insulin:9.68±3.58

                                                                                                                  Fasting levels of ≥5.2
                                                                                                                  mmol/l detected 92% of
                                                                                                                  women who converted
                                                                                                                  to insulin but had a false
                                                                                                                  positive rate of 70%.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 180 of 441
                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &       Effect Size        Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                    Comments
                 Level                 cs                                   Measures
                                                                                         Macrosomia
                                                                                         Glibenclamide: 11.1%
                                                                                         Insulin:8.3% (NS)

                                                                                         Neonatal IV glucose
                                                                                         infusion:
                                                                                         Glibenclamide: 25%
                                                                                         Insulin:12.7% (NS)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                             Page 181 of 441
                                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                    Follow-up &            Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                         Outcome                                                                   Comments
                 Level                 cs                                                           Measures
Schaefer-Graf, U., Study Type:      99 women with   All women             Intervention: Insulin   Follow-up period: In the control group 30      GDM management
et al              RCT              GDM managed     prescribed diet       started when the                           women (30%) met the         based on fetal growth
                                    according to    (30kcalkg/day)        abdominal               Outcome            criteria for insulin. Of    combined with high
2004              Evidence level:   Ultrasound      with caloric          circumference (AC)      Measures: SGA these 27 received                glycemic criteria provides
                  1+                assessment of   retriction for        > 75th percentile       (≤10th percentile) insulin, 2 refused and 1    outcomes equivalent to
175
                                    abdominal       overweight            before 36 completed     LGA (≥90th         left care.                  management based on
                                    circumference   women                 weeks. Gluoose          percentile)                                    strict glycemic criteria
                                    100 controls    (25kcal/kg/day).      targets were not                           In the US group 40          alone.
                                    (managed by     Women advised         discussed with                             women (40%) met the
                                    glycaemic       to exercise after     patients and glucose                       criteria for insulin, all
                                    levels)         meals and taught      values were not                            with AC>75th percentile.
                                                    to self monitor       used as a guide to                         Of these, 3 refused
                                                    blood glucose         management,                                insulin and 1 did not
                                                    with a memory         unless any fasting                         receive therapy due to
                                                    reflectance meter.    value >6.6mmol/l or                        an erroneous AC
                                                    After one week        any 2-hr value                             calculation.
                                                    women assessed        greater than 11.0
                                                    for eligibility to    mmol/l. Insulin was                        Neonatal outcomes (no
                                                    protocol. Criteria:   not prescribed,                            significant differences)
                                                    GDM (2 abnormal       regardless of AC,                          C-section (%):
                                                    values on 75g         when FCG <4.4 and                          US group: 18
                                                    OGTT). All            2-h value was <5.5.                        Control group:19
                                                    capillary fasting     US was perfrmed at                         Birth weight (g)
                                                    values <6.6 and       entry and then 4-                          US group:3306.1±558
                                                    2-h <11.0mmol/l.      week intervals at 20,                      Control
                                                    16-34 weeks           4, 28, 32 and 36                           goup:3371.2±500
                                                    gestation. No         weeks.                                     SGA (%)
                                                    drug abuse.                                                      US group:12
                                                                          Insulin was titrated                       Control group:13
                                                    Country:              to achieve FCG                             LGA (%)
                                                                          <4.4mmol/l and 2-h                         US group: 12
                                                                          <6.05 mmol/l                               Control group: 10
                                                                                                                     Hypoglycaemia
                                                                          Comparison: Insulin                        US group: 17
                                                                          prescribed before 36                       Control group: 16
                                                                          weeks if 2 glucose                         Transfer to NICU
                                                                          profiles had ≥2                            US group:14
                                                                          elevated values                            Control group: 15
                                                                          (FCG>4.95 mmol/l,
                                                                          2-hr >6.6 mmol/l) or
                                                                          4 profiles had at
                                                                          least 1 elevated




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                    Page 182 of 441
                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &               Follow-up &   Effect Size   Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                    Outcome                                           Comments
                 Level                 cs                                      Measures
                                                        value during a 2
                                                        week period.Insulin
                                                        was titrated to
                                                        achieve FCG <4.95
                                                        & 2-h pp <6.6 mmol/




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                       Page 183 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                Follow-up &            Effect Size              Study Summary                Reviewer
 Information & Evidence   Patients Characteristi Comparison                                     Outcome                                                                    Comments
                 Level                 cs                                                       Measures
Kjos, S.       Study Type:       48 women with   Inclusion criteria   Intervention:           Follow-up period: Mean BMI significantly >     US measurement of fetal Control group: Insulin
               RCT               GDM managed     1)GDM,               Intervention group:                        in control group.           AC identified fetuses at prescribed to all women
2001                             on relaxed      2) FPG               Prescribed insulin if   Outcome                                        low risk of neonatal     immediately
               Evidence level:   gylcaemic       concentrations       fetal abdominal         Measures: SGA In the intervention group        macrosomia.
174
               1+                criteria and    >5.8 and <6.6        circumference (AC)      (≤10th percentile) 22 women (45%) had a
                                 ultrasound      3)gestational age    was ≥70th percentile    LGA (≥90th         fetal AC of ≥70th           The difference in C-
                                 measurement     >14 and <34          or FPG >6.6.            percentile)        percentile at enrolment.    section rate was not
                                 of fetal AC.    weeks at time of     mmol/l. Insulin was                        21 of these women were      explained by
                                 48 women with   entry to study 5)    titrated to achieve                        immediately started on      birthweights.
                                 GDM managed     No drug abuse.       preprandial capillary                      insulin. An additional 9
                                 on maternal     All women            blood glucose                              women in the
                                 glycaemic       received             ≤4.4mmol/l and 2-hr                        intervention group
                                 criteria        instruction in       postprandial values                        began insulin
                                                 SBGM, diet and       of ≤6.05 mmol/l.                           subsequently because
                                                 exercise.                                                       AC≥70th percentile
                                                                      After the baseline                         (n=6) or FSG >6.6
                                                 Country: USA         measurements, AC                           mmol/l (n=2) or because
                                                                      measurements                               of inadequate SBGM
                                                                      made at 20, 24, 28,                        (n=1)
                                                                      32 and 36 weeks.
                                                                                                                Neonatal outcomes:
                                                                      Comparison: Control                       Birthweight:
                                                                      group: Insulin                            Intervention: 3369 (461)
                                                                      prescribed to all                         Control:3271 (458)
                                                                      women immediately.
                                                                      Insulin adjusted to                       Birthweight≥4000g
                                                                      achieve preprandial                       Intervention: 3 (6.3%)
                                                                      capillary blood                           Control:2 (4.2%)
                                                                      glucose
                                                                      concentrations                            C-section rate was
                                                                      ≤4.95 mmol/l and 2-                       significantly lower in the
                                                                      hr postprandial                           control group (33.3% vs
                                                                      values ≤ 6.6 mmol/l                       14.6%, p=0.03).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 184 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                     Outcome                                                                  Comments
                 Level                 cs                                                       Measures
Botta, R., et al   Study Type:       75 women with GDM according to   Intervention: What      Follow-up period: 52 pateints were treated    Early diagnosis of GDM,
                   Cohort            GDM            NDDG & C&C        factors are                                 with diet alone and 23    early rise in triglyceride
                                                    criteria. All     predictive of insulin   Outcome             with diet and insulin.    levels and elevated blood
                   Evidence level:   75 women with women instructed   treatment in women      Measures:                                     glucose levels 180
171
                   2+                normal glucose on SBGM and       with GDM?               Gestational week Week of diagnosis:           minutes after glucose
                                     tolerance      diet. Insulin                             at diagnosis        Insulin group: 19.4±6.9   load in women with
                                                    started when      Comparison:             Prepregnancy        weeks                     gestational diabetes with
                                                    fasting bg >5.2                           BMI                 Diet group: 24.5±6.2      a pregestational BMI>28
                                                    mmol/l and 2-hr                           Blood glucose       weeks, p<0.003            and normal weight gain
                                                    pp > 6.9 mmol/l                           levels during                                 can be predictive of
                                                                                              OGTT                Pregestational BMI        insulin treatment.
                                                   Country:                                   Area under the      Insulin: 28.8±5.1
                                                                                              curve (AUC)         Diet: 27.7±5.4            In these women insulin
                                                                                              Triglycerdie levels Normal glucose            treatment was initiated
                                                                                              Weight gain         tolerance                 when blood glucose
                                                                                                                  (NGT):25.7±4.5            levels were higher than
                                                                                                                  p<0.05 insulin group vs   5.2 (fasting) and 6.9
                                                                                                                  NGT                       mmol/l (2-hr
                                                                                                                                            postprandial). This
                                                                                                                 Weight gain:               treatment strategy was
                                                                                                                 Insulin: 5.3±4.7           insufficient to normalise
                                                                                                                 Diet:8.0±5.6               birthweights.
                                                                                                                 NGT: 5.8±5.4
                                                                                                                 P <0.05 diet vs insulin
                                                                                                                 and NGT

                                                                                                                 Triglycerdie levels
                                                                                                                 (mg/dl)
                                                                                                                 Insulin: 232.2±110.7
                                                                                                                 Diet:185.9±81.8
                                                                                                                 NGT:147.4±70.1
                                                                                                                 P<0.001 insulin group vs
                                                                                                                 NGT

                                                                                                                 3-hr bg levels on OGTT
                                                                                                                 Insulin:9.2 ±2.7 mmol/l
                                                                                                                 Diet:7.04±2.2
                                                                                                                 P<0.001

                                                                                                                 AUC
                                                                                                                 Insulin:10,97±1,949
                                                                                                                 Diet: 9,199±1,277,




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 185 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                        Comments
                 Level                 cs                                   Measures
                                                                                         p<0.01

                                                                                         Infant body weight index
                                                                                         Insulin:1.24±0.16
                                                                                         Diet:1.14±0.14
                                                                                         NGT:1.09±0.7
                                                                                         P<0.001 insulin group vs
                                                                                         NGT and diet group

                                                                                         LGA (>90th percentile)
                                                                                         Insulin group: 39%
                                                                                         Diet:32%
                                                                                         NGT: 21%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 186 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &             Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                                    Comments
                 Level                 cs                                                    Measures
Bochner, C., et al Study Type:       201 women   GDM (according      Intervention: 30-33   Follow-up period: Infants with a birthweight     Patients with fetal
                   Cohort            with GDM    to O’Sullivan and   week Ultrasound                           of >90th percentile had      abdominal circumference
1987                                             Mahan)              (US)                  Outcome             significantly > incidence    >90th percentile at 30 to
                   Evidence level:               diagnosed before                          Measures:           (p<0.001) of C-section       33 weeks had a
172
                   2+                            34 weeks.           Comparison:           Sensitivity to      for failure to progress in   significantly increased
                                                 Maintained                                predict             labour (41.7% vs             incidence of failure to
                                                 weekly fasting                            macrosomia or       14.3%), shoulder             progress, shoulder
                                                 glucose levels                            labour dystocia at dystocia (16.7% vs            dystocia and birth
                                                 <5.5 mmol/l and                           term (cut off point 0.7%) and birth trauma       trauma, whereas patients
                                                 2-hr postprandial                         90th percentile)    (3.6% vs 20.8%)              with AC measurement
                                                 glucose levels                            Labour dystocia     compared with infants        ≤90th percentile had no
                                                 <6.6 mmol/l with                          (failure to         with a birth weight <90th    greater risk than general
                                                 dietary                                   progress in labour percentile.                   population.
                                                 management                                for >2 hours once
                                                 alone.                                    the active phase Predicting macrosomia
                                                                                           of labour had       at term:
                                                 Country: USA                              begun)              AC >90th percentile:
                                                                                           Shoulder dystocia 36/64
                                                                                                               AC≤ 90th percentile:
                                                                                                               5/137
                                                                                                               Sensitivity: 87.8
                                                                                                               Specificity: 82.5
                                                                                                               PPV : 56.3
                                                                                                               NPV: 96.4




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 187 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &              Effect Size             Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                                   Comments
                 Level                 cs                                                    Measures
Catalano, P., et al Study Type:       195 women   GDM diagnosed       Intervention:         Follow-up period:   67/195 women (34%)          Infants of women with
                    Cohort            with GDM    with 3-h 100g       Women with GDM                            required insulin.           GDM have increased
2003                                  220 women   OGTT (NDDG).        performed home        Outcome             Women treated with          body fat compared with
                    Evidence level:   with NGT    Exclusions:Birthw   glucose monitoring Measures:              insulin had greater         women with NGT. This is
410
                    2++                           eight <2000g,       (7 times/day).                            pregestational weight       independent of birth
                                                  multifetal          Targets <5.5 mmol/l                       (87.9 ±25.5 kg vs           weight.
                                                  gestations and      (fasting) and <6.6                        77.4±21.1 kg, p<0.0002)
                                                  neonates with       mmol/l (2-h pp). All                      and parity (49% .1 vs
                                                  abnormalities.      women encouraged                          32% >1.0, p=0.02) than
                                                                      to walk for 30                            diet group.
                                                  Country: USA        minutes after meals.                      There was no significant
                                                                      Insulin intiated when                     difference in birthweight
                                                                      women had                                 between groups
                                                                      persistent glucose                        (3497±556 vs 3346±542
                                                                      concentrations                            g, p=0.07).
                                                                      >targets.
                                                                                                                The skinfold thickness
                                                                      Comparison: 220                           was significantly greater
                                                                      women with NGT                            in the insulin group than
                                                                                                                in the NGT group.
                                                                                                                Subscapular:
                                                                                                                5.9±1.6mm vs 5.1±1.2
                                                                                                                mm, p=0.0001
                                                                                                                Triceps: 5.0±1.1 mm vs
                                                                                                                4.5 ±1.0 mm, p<0.003
                                                                                                                Flank: 4.5 ± 1.2 mm vs
                                                                                                                4.0 ±1.1 mm, p=0.01
                                                                                                                Thigh: 6.3 ±1.5 mm vs
                                                                                                                5.8 ± 1.3 mm, p=0.02
                                                                                                                Abdomin: 4.0 ±1.4 mm
                                                                                                                vs 3.3 ±0.9 mm,
                                                                                                                p=0.0001

                                                                                                                There was significantly
                                                                                                                greater fat mass
                                                                                                                (492±215g vs 407 ±196
                                                                                                                g, p=0.006) and % body
                                                                                                                fat (13.6% ±4.6% vs
                                                                                                                11.7% ± 4.5%, p=0.007)
                                                                                                                in the insulin group
                                                                                                                compared with the diet
                                                                                                                group.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 188 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         Regression:
                                                                                         In the GDM group
                                                                                         gestational age had the
                                                                                         strongest correlation
                                                                                         with birthweight
                                                                                         although fasting blood
                                                                                         glucose had the
                                                                                         stongest correlation with
                                                                                         fat mass and percent
                                                                                         body fat.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 189 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                               Follow-up &            Effect Size             Study Summary                 Reviewer
 Information & Evidence   Patients Characteristi Comparison                                    Outcome                                                                    Comments
                 Level                 cs                                                      Measures
Jovanovic, L.   Study Type:       23 women with    >18 years of age, Intervention: Insulin   Follow-up period: Women randomised to          The postprandial glucose Not blinded
                RCT               GDM taking       diagnosed with    lispro                                    the lispro group were        rise was significantly less
1999                              regular human    OGTT between                              Outcome           older (34.2 years) than      after a standardized dose
                Evidence level:   insulin          14 and 32 weeks. Comparison:              Measures:         in the regular insulin       of insulin lispro than after
183
                1+                19 women with                      Regular human           Serum insulin     group (29.8 years,           regular human insulin. It
                                  GDM taking       Country: USA      insulin                 Blood glucose     P<0.01)                      was also effective in
                                  insulin lispro                                             C peptide                                      lowering overall
                                                                                             concentrations    No insulin lispro was        glycaemia. The women in
                                                                                                               detected in umbilical        the lispro group had less
                                                                                             C-sections        cord blood.                  hypoglycaemia.
                                                                                             Gestational age
                                                                                             at delivery       The areas under the
                                                                                             Neonatal          curve for glucose, insulin
                                                                                             birthweight       and C peptide were
                                                                                             1-min and 5-min lower for insulin lispro
                                                                                             apgar scores      group, a difference that
                                                                                             Fetal abnormality remained significant
                                                                                             Neonatal          after multiple regression
                                                                                             hypoglycaemia     (age, BMI, fasting
                                                                                             Neonatal          plasma glucose
                                                                                             hypocalcaemia     concentration).

                                                                                                                HbA1c(%)
                                                                                                                Regular human insulin
                                                                                                                At enrollment:5.24±0.09
                                                                                                                After 6 weeks:5.16±0.12
                                                                                                                Difference from
                                                                                                                baseline: 0.07 (2.8%)
                                                                                                                Lispro
                                                                                                                At enrolment: 5.47±0.09
                                                                                                                After 6 weeks:5.12±0.11
                                                                                                                Difference from
                                                                                                                baseline: 0.0018 (5.7%)


                                                                                                                For the remainder of
                                                                                                                pregnancy
                                                                                                                hypoglycaemic episodes
                                                                                                                before breakfast were
                                                                                                                lower in the lispro group
                                                                                                                (p=0.025) but not before
                                                                                                                lunch or dinner. The




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                              Page 190 of 441
                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size        Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                     Comments
                 Level                 cs                                   Measures
                                                                                         number of
                                                                                         hyperglycaemic
                                                                                         episodes was lower in
                                                                                         the lispro group
                                                                                         (p=0.019).

                                                                                         No difference in
                                                                                         obstetric or fetal
                                                                                         outcomes.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                              Page 191 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &            Effect Size             Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                                 Comments
                 Level                 cs                                                 Measures
Jovanovic-          Study Type:      9 women with  GDM            Intervention: Diet    Follow-up period: HbA1c:                      Arm ergometer training is
Peterson, M., et al RCT              GDM in diet                  plus cardiovascular                      Diet:4.7%±0.2%             feasible n women with
                                     group         Country: USA   conditioning          Outcome            Diet +                     gestational diabetes
1989               Evidence level:   10 women with                                      Measures:          exercise:4.2%±0.2%,        mellitus and results in
                   1++               GDM in diet                  Diet: 6 weeks, 24-    HbA1c              p<0.001                    lower HbA1c fasting and
164
                                     plus exercise                30kcal/kg/24 hours;   50g GCT                                       1-hr plasma glucose
                                     group                        20% protein, 40%      Self monitoring    GCT:                       concentrations than diet
                                                                  carbohydrate, 40%     (fasting and 1     Fasting:                   alone.
                                                                  fat.                  hour after meals). Diet: 4.8±0.3mmol/l
                                                                                                           Diet +
                                                                  Diet plus exercise:                      exercise:3.9±0.4mmol/l,
                                                                  As above + 20                            p<0.001
                                                                  minutes exercise 3                       1-hour value:
                                                                  times/week on arm                        Diet:10.3±0.7mmol/l
                                                                  ergometer.                               Diet + exercise:5.8±1.0,
                                                                                                           <0.001
                                                                  Comparison: Diet




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                        Page 192 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &           Effect Size              Study Summary                     Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                                        Comments
                 Level                 cs                                                 Measures
Buchanan, et al   Study Type:       303 women   GDM (fasting     Intervention: 30       Follow-up period: 205/303 (68%) women         Fetal ultrasound             Rate of LGA infants in
                  RCT               with GDM    <5.8 mmol/l).    women with AC                            with GDM had AC <75th       identified women with        group <75th percentile
1994                                            Hispanic women   ≥75th percentile in    Outcome           percentile. Of these, 15    mild GDM at risk for fetal   does not include women
                  Evidence level:               between 29 and   diet (and SBGM)        Measures:         (7.3%) were placed on       macrosomia. Insulin          removed from analysis
173
                  1+                            33 weeks         and insulin group.     Birthweight       insulin therapy because     therapy reduced the          due to initiation of insulin
                                                gestation.       Insulin dose           Body length       of persistent fasting       macrosomia indicating        (n=15).
                                                                 adjusted to maintain   Tricep thickness glycaemia ≥5.8 mol/l. 11     that fetal ultrasound can
                                                Country: USA     premeal glycaemia      Subscapular       women delivered before      be used to guide             In total this strategy would
                                                                 between 3.3 and 4.4    thickness         36 weeks and 8              metabolic therapy in         give insulin to 37% of
                                                                 mmol/l and 2-h         Iliac crest skin  delivered at another        pregnancies complicated      GDM population (98
                                                                 postmeal glycaemia     fold thickness    location. Analysis was      by mild GDM.                 women with AC≥ 75th
                                                                 <6.1 mmol/l.           Birth trauma      undertaken on data from                                  percentile + 15 women
                                                                                                          remaining 171 women.                                     with persistent fasting
                                                                 Comparison: 29                           Birthweight: 3,444±38g                                   glycaemia >5.8mmol/l)
                                                                 women with AC                            LGA: 24/171 (14%)
                                                                 ≥75th percentile in
                                                                 diet (+ SBGM) group                      Women with AC ≥75th
                                                                                                          percentile
                                                                 205 women with                           1. Birthweight
                                                                 AC<75th percentile                       Diet+insulin: 3,647±67g
                                                                 prescribed diet.                         Diet:3,878±84g
                                                                 Insulin administered                     P<0.02
                                                                 if fasting glucose                       2. LGA
                                                                 ≥5.8 mmol/l.                             Diet+insulin:4/30 (13%)
                                                                                                          Diet: 13/29 (45%)
                                                                                                          P<0.02

                                                                                                          Neonatal skinfold
                                                                                                          measurements at 3 sites
                                                                                                          were reduced in insulin
                                                                                                          treatment group
                                                                                                          (p<0.005)

                                                                                                          The insulin treated
                                                                                                          group had much higher
                                                                                                          c-section rates (43%,
                                                                                                          P<0.05 vs other groups)

                                                                                                          No difference in rates of
                                                                                                          neonatal hypoglycaemia
                                                                                                          Glucose levels in both
                                                                                                          groups were lower at the




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 193 of 441
                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size             Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                          Comments
                 Level                 cs                                   Measures
                                                                                         follow-up breakfast test
                                                                                         than at baseline (p<0.01
                                                                                         for diet only , p<0.0001
                                                                                         for insulin group).
                                                                                         Weekly capillary blood
                                                                                         glucose readings also
                                                                                         reduced significantly in
                                                                                         both treatment groups,
                                                                                         although they were 0.3-
                                                                                         0.6 mmol/l lower in the
                                                                                         insulin group throughout
                                                                                         the treatment period
                                                                                         (P<0.005).
                                                                                         The frequency of
                                                                                         symptomatic
                                                                                         hypoglycaemia (<3.3
                                                                                         mmol/l) was 0.3
                                                                                         episodes per week. Half
                                                                                         of the insulin treated
                                                                                         women experienced no
                                                                                         symptomatic
                                                                                         hypoglycaemia. 5
                                                                                         women had >5 episodes
                                                                                         during 5-10
                                                                                         Weeks of treatment.
                                                                                         There were no episodes
                                                                                         requiring assistance.

                                                                                         3/29 women in the diet
                                                                                         treated group would
                                                                                         have qualified for insulin
                                                                                         based on glycaemic
                                                                                         control (fasting <5.5 and
                                                                                         1-h pp <7). 1 of these
                                                                                         women had a LGA
                                                                                         infant. 26/29 women in
                                                                                         diet treated group would
                                                                                         not have qualified for
                                                                                         insulin treatment on the
                                                                                         basis of glycaemic
                                                                                         control. 12 of these
                                                                                         women had a LGA




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                   Page 194 of 441
                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &         Effect Size   Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                 Comments
                 Level                 cs                                   Measures
                                                                                         infant.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                          Page 195 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                           Follow-up &               Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                Outcome                                                                       Comments
                 Level                 cs                                                  Measures
Simmons, D.,   Study Type:       Infants of 20   GDM. Maori and       Intervention:      Follow-up period:     Follow-up data was           A larger RCT is required
               Cohort            mothers with    pacific islander     Maternal insulin   Follow up (mean       obtained from 11/20          to answer the question:
1997                             gestational     population.          treatment          2 years 8 months      women treated with diet      Does insulin therapy
               Evidence level:   diabetes        Exclusions:                             ± 1 month) of         and 9/15 women treated       reduce the risk of obesity
411
               2+                                (Maternal)           Comparison:        infants of mothers    with insulin.                in offspring of mothers
                                                 Smoked ≥10                              with GDM.             Women treated with           with GDM?
                                                 cigarettes/day,                                               insulin were older, had
                                                 late booking, poor                      Outcome               >BMI and a greater
                                                 clinic attendance,                      Measures:             proportion of pacific
                                                 antenatal or                            Crown-rump            islanders.
                                                 perinatal                               length
                                                 complications                           Biceps skinfold       Subscapular skinfold
                                                                                         thickness             thickness (mm) median
                                                 Country: NZ                             Triceps skinfold      and interquartile range:
                                                                                         thickness:            Diet group: 7..9 (7.0-9.4)
                                                                                         Subscapular           Insulin group: 5.9 (5.4-
                                                                                         skinfold              7.9) p<0.05
                                                                                         thickness:            Bicep skinfold thickness:
                                                                                         Suprailiac skinfold   Diet: 6.3 (6.0-9.0)
                                                                                         thickness:            Insulin: 5.1 (4.3-6.6)
                                                                                         Waist, hip and        p=0.01
                                                                                         head                  Triceps skinfold
                                                                                         circumference.        thickness (mm)
                                                                                                               Diet:10.8 (9.0-11.3)
                                                                                                               Insulin:8.4 (7.1-10.7)
                                                                                                               p=0.06
                                                                                                               Sum (mm)
                                                                                                               Diet:34.2 (29.9-36.1)
                                                                                                               Insulin:26.2 (23.7-34.9)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 196 of 441
                                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                      Follow-up &           Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                           Outcome                                                               Comments
                 Level                 cs                                             Measures
Drexel, H.,    Study Type:       102 women                  Intervention: GDM       Follow-up period: By the end of gestation   Excess mortality and
               Cohort            with GDM                   to maintain blood       Austria           88/102 women received     morbidity can be
1988                             102 nondiabetic Country:   glucose                                   insulin at a mean dose    prevented by early
               Evidence level:   women                      concentration <7.15     Outcome           of 18 U/day.              institution of tight
412
               2++               25 untreated               mmol/l 1-h after        Measures:         Fasting blood glucose     metabolic control.
                                 women due to               breakfast.              Birthweight       levels on OGTT did not
                                 late                       Treatment consisted     Macrosomia        predict insulin use.
                                 presentation               of a diet low in        Dystropathy
                                                            oligosaccharides        hypoglycaemia     LGA
                                                            and fat and, if         Hypocalcaemia     Control: 9/103
                                                            necessary, once         Hyperbilirubinemi GDM:7/103
                                                            daily insulin.          a                 Untreated GDM:5/25
                                                                                    Fetal acidosis
                                                            Comparison: GDM         Apgar scores      >4000g
                                                            (O’Sullivan criteria)                     Contol:9/103
                                                            Diet (low fat, low                        GDM:6/103
                                                            oligosaccharide )                         Untreated GDM:4/25

                                                                                                      >4500g
                                                                                                      Contol:2/103
                                                                                                      GDM:1/103
                                                                                                      Untreated GDM:1/25

                                                                                                      There were no
                                                                                                      differences between the
                                                                                                      control group and the
                                                                                                      insulin group in any
                                                                                                      other neonatal outcome.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                  Page 197 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                             Follow-up &            Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                  Outcome                                                                  Comments
                 Level                 cs                                                    Measures
Cheng                Study Type:       14,771   Inclusion criteria:   Intervention: 50g    Follow-up period: Estimated risks in          Although a markedly
YW;Esakoff           Cohort                     women had a 50-       glucose load test                       women with GLT ≥200        elevated GLT may not be
TF;Block-                                       g glucose loading     (GLT).               Outcome            mg/dL but no GDM:          universally diagnostic for
Kurbisch I;Ustinov   Evidence level:            test during the       Women with GLT       Measures:          caesarean section OR       GDM, a GLT value of
A;Shafer             2-                         study period          results ≥200 mg/dL   Maternal           4.18 (95% CI 1.15-15.2);   ≥200 mg/dL is associated
S;Caughey AB;                                   (including the                             outcomes:          preterm delivery <32       with unfavourable
                                                ones who              Comparison: GLT      Mode of delivery weeks’ gestation OR          perinatal outcomes and
2006 Nov                                        underwent early       <200 mg/dL           (as well as        8.05 (95% CI 1.02-63.6),   may warrant further
                                                screening for         stratified by GDM    indication for     shoulder dystocia OR       follow-up regardless of
148
                                                GDM and               diagnosis.           caesarean          15.14 (95% CI 64-140),     the subsequent GTT
                                                screened at 24-28                          delivery),         5-minute Apgar score <7    results.
                                                weeks’ gestation                           preeclampsia,      for neonates OR 6.41
                                                and beyond).                               third- or fourth   (95% CI 1.23-33.3).
                                                Exclusion criteria:                        degree perineal
                                                Intrauterine fetal                         lacerations,       Estimated risks in
                                                demise, neonates                           postpartum         women with a GLT ≥200
                                                with known lethal                          hemorrhage and mg/dL and diagnosed
                                                congenital                                 peripartum         with GDM: caesarean
                                                anomalies,                                 infections.        section OR was also
                                                multiple                                   Neonatal           elevated compared to
                                                gestations and                             outcomes:          those with a GLT <200
                                                maternal diabetes                          Birth weight,      mg/dL (OR 2.24, 95% CI
                                                diagnosed before                           preterm delivery, 1.19-4.21).
                                                the study                                  shoulder dystocia,
                                                pregnancy                                  5-min Apgar
                                                                                           score <7 and birth
                                                Country: USA                               trauma




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 198 of 441
                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type           Aim of Study        Number of      Population          Outcome            Results & Comments                Study             Reviewer
 Information & Evidence                                Patients &    Characteristic       measures                                            Summary            Comment
                 Level                                   Patient          s
                                                      Characteristic
                                                            s
Downs             Study Type:     Intervention: A mail 28           Postpartum women Behavioural            -Strongest perceived             To increase
DS;Ulbrecht JS;   OtherRetrospec survey assessing                   with gestational (exercise              advantage of exercise during     exercise
                  tive            their self-reported               diabetes         advantages),           pregnancy was controlling        behaviour and
2006 Feb                          exercise beliefs                                   normative (important   blood glucose and during         reduce the risk
                  Evidence Level: (advantages,                      Country: USA     others with a strong   postpartum it was controlling    of type 2
167
                  2-              barriers, and                                      influence), and        weight                           diabetes in
                                  important social                                   control (factors       -Most common barrier to          women with
                                  influences) and                                    preventing exercise    exercise during pregnancy was    gestational
                                  behaviours.                                                               fatigue and postpartum it was    diabetes,
                                  Leisure-Time                                                              lack of time                     researchers
                                  Exercise                                                                  -Women’s husband/partner         and health care
                                  Questionnaire                                                             most strongly influenced their   professionals
                                  (LTEQ), assessing                                                         exercise during pregnancy and    are encouraged
                                  mild, moderate, and                                                       postpartum                       to use women’s
                                  strenuous exercise                                                        -Women exercised more            exercise
                                  done for at least 15                                                      during the postpartum period     beliefs, that is,
                                  min                                                                       than before or during            advantages,
                                                                                                            pregnancy                        social
                                  Comparison:                                                               -Number of exercise              influences, and
                                                                                                            advantages was positively        perceived
                                                                                                            associated with women’s          barriers to
                                                                                                            pregnancy and postpartum         exercise, as a
                                                                                                            exercise behavior.               framework for
                                                                                                                                             designing
                                                                                                                                             effective
                                                                                                                                             diabetes
                                                                                                                                             treatment and
                                                                                                                                             prevention
                                                                                                                                             programmes




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                               Page 199 of 441
                                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                              Follow-up &             Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                                   Outcome                                                                    Comments
                 Level                 cs                                                     Measures
Lesser         Study Type:       5 normal and 6   Women with        Intervention: A         Follow-up period: Mean values with vs           Study results indicate
KB;Gruppuso    RCT               women with       normal glucose    mixed nutrient meal                         without exercise did not    that a single bout of
PA;Terry                         gestational      levels (NL) and   with exercise stress    Outcome             differ for                  exercise did not blunt the
RB;Carpenter   Evidence level:   diabetes         women with        test.                   Measures:           Fasting glucose (NL         glycaemic response
MW;            1+                                 gestational       Exercise consisted      Fasting glucose     78.9 +/- 2.6 vs. 80.0 +/-   observed following a
                                                  diabetes (GDM)    of upright stationary   and insulin levels, 2.6 mg/dl; GDM 86.4 +/-     mixed nutrient meal.
1996 Jul                                                            cycling for 30 min at   peak glucose and 2.0 vs. 82.1 +/- 3.5 mg/
                                                  Country: USA      a heart rate            insulin levels and dl),
166
                                                                    consistent with 60%     incremental area Peak glucose (NL 132.3
                                                                    VO2max                  of the glycaemic +/- 10.4 vs. 139.1 +/-
                                                                                            and insulin curves 15.6 mg/dl; GDM 165.8
                                                                    Comparison: A                               +/- 5.5 vs. 160.3 +/- 7.8
                                                                    mixed nutrient meal                         mg/dl), Area under the
                                                                    without an exercise                         glycaemic curve (NL
                                                                    stress that took                            5758 +/- 1038 vs. 6393
                                                                    place 14 h earlier.                         +/- 1281 mg/dl.min;
                                                                                                                GDM 8,178 +/- 890 vs.
                                                                                                                8,331 +/- 563
                                                                                                                mg/dl.min).
                                                                                                                Plasma insulin levels did
                                                                                                                not differ between the
                                                                                                                two groups.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 200 of 441
                                                                                                                                         DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                          Follow-up &             Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                               Outcome                                                                    Comments
                 Level                 cs                                                 Measures
Peterson        Study Type:       25     Women aged 21-        Intervention: Each       Follow-up period:    (Group 1) Women with        A weight loss regimen
CM;Jovanovic-   RCT                      50 yrs                woman received           1st follow up at     previous gestational        consisting of 40%
Peterson L;                              Exclusion criteria:   nutritional              week 6 and 2nd       diabetes mellitus were      carbohydrate results in
                Evidence level:          Any medical           supplement bars for      follow up at week    comparable to (Group 2)     lower triglyceride levels
1995 Aug        1+                       condition for         breakfast, lunch,        12                   obese women without a       than those achieved with
                                         which a calorie-      and snacks and a                              history of previous         a 55% carbohydrate
163
                                         restricted diet       meal plan for dinner     Outcome              gestational diabetes        content diet in obese
                                         may be harmful        which comprised 1/3      Measures:            except that the former      women.
                                         including             of total calories. The   Women’s              had higher maximum
                                         pregnancy or          nutritional              adherence to the levels of glucose on a
                                         planned               supplement bars          program,             glucose tolerance test
                                         pregnancy during      were identically         Body weight,         and higher fasting
                                         the trial,            wrapped, each            triglyceride levels, insulin levels consistent
                                         hypertension,         containing 180 kcal      cholesterol levels with greater insulin
                                         diuretic use,         with 20% protein.        and fasting insulin resistance. Weight loss
                                         thyroid disease,      Group 1 received         levels.              was comparable for all
                                         or frank diabetes     bar containing 40%                            groups during the first 6
                                         by National           carbohydrate                                  weeks but attenuated in
                                         Diabetes Data                                                       all groups during the
                                         Group Criteria.       Comparison: Group                             second 6 weeks of the
                                         Group 1: obese        2 received bar                                trial regardless of
                                         women with            containing 55%                                diabetes history or
                                         previous              carbohydrate                                  treatment group
                                         gestational                                                         allocation. Women with
                                         diabetes                                                            or without a previous
                                         Group 2 : obese                                                     history of gestational
                                         women who had                                                       diabetes had higher
                                         not had                                                             triglycerides while on a
                                         gestational                                                         55% carbohydrate diet
                                         diabetes                                                            than while on a 40%
                                                                                                             carbohydrate diet.
                                         Country: USA




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                           Page 201 of 441
                                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type           Aim of Study        Number of      Population               Outcome              Results & Comments               Study            Reviewer
 Information & Evidence                                Patients &    Characteristic            measures                                             Summary           Comment
                 Level                                   Patient          s
                                                      Characteristic
                                                            s
Smits MW;Paulk   Study Type:     Intervention: A       263          Women diagnosed        Outcomes in women:      Overall incidence of abnormal   No significant
TH;Kee CC;       Other           hospital, outpatient-              with gestational       Polyhydramnios,         outcomes for women with         reduction in the
                 Descriptive     based, nursing                     diabetes delivered     preeclampsia,           gestational diabetes was        risk of
1995                             intervention and the               at the hospital        premature               54.1% and 72.7% for their       abnormal
                 Evidence Level: traditional, office-               during the year        contractions, vaginal   infants.                        outcomes for
187
                 3               based care provided                following the          bleeding due to         Nurse-led intervention had      woman or
                                 by obstetricians by                initiation of the      placenta praevia.       53.7% incidence of abnormal     infant. A
                                 using a research                   Nursing intervention   Delivery at <37         maternal outcomes and 68.3%     significantly
                                 model. The research                                       weeks or > 42           incidence of abnormal infant    greater risk of
                                 model used three                   Country: USA           weeks.                  outcomes versus an incidence    having an infant
                                 types of variables as                                     Delivery                of 50% and 75% respectively     with one or
                                 suggested by the                                          complications such      for obstetrician-led            more abnormal
                                 literature: input                                         as induction,           intervention.                   outcomes in
                                 variables (risk                                           caesarean section,                                      first-time
                                 factors prior to                                          mid-forceps, mid                                        mothers,
                                 gestation),                                               cavity vacuum                                           women with
                                 moderating variables                                      extraction.                                             gestational
                                 (conditions that                                          Postpartum length of                                    diabetes on
                                 occur during                                              stay 5 days or                                          medications,
                                 pregnancy), and                                           longer.                                                 and women
                                 outcome variables                                         Infant outcomes:                                        with gestational
                                 (normal vs abnormal                                       APGAR 7 or less at 1                                    diabetes
                                 outcomes for mother                                       min and 5 min, SGA                                      experiencing
                                 and infant).                                              < 10th percentile or                                    complications.
                                                                                           LGA > 90th
                                 Comparison:                                               percentile,
                                                                                           postpartum length of
                                                                                           stay ≥5 days,
                                                                                           hypoglycaemia,
                                                                                           respiratory distress
                                                                                           syndrome,
                                                                                           polycythemia,
                                                                                           hyperbilirubinemia,
                                                                                           hypocalcemia, birth
                                                                                           trauma.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                      Page 202 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                   Follow-up &           Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                        Outcome                                                                 Comments
                 Level                 cs                                                          Measures
Rossi G;              Study Type:       141 women       Women with           Intervention:       Follow-up period: Twenty-nine women           Early fetal ultrasound
Somigliana E;         RCT                               uncomplicated        Group A, fetal      41 months         whose fetal abdominal       assessment at 28 weeks’
Moschetta M;                            1. Group A (n = singleton            ultrasound was                        circumference exceeded      gestation should be
Bottani B; Barbieri   Evidence level:   73)             pregnancies, who     assessed at 28      Outcome           the 75th percentile were    encouraged since insulin
M; Vignali M;         1++                               planned to deliver   weeks’ and 32       Measures:         considered eligible for     administration after 32
                                        2. Group B (n = vaginally and        weeks’ gestation.   1. HbA1c values insulin therapy.              weeks’ gestation is less
2000 Aug                                68)             whose diagnosis                          2. Gestational                                effective in lowering the
                                                        of gestational       Comparison:         age at delivery   15 out of 20 macrosomic     rate of macrosomic
177
                                                        diabetes was         Group B, fetal      3 Mode of         infants had an              infants.
                                                        established          ultrasound was      delivery          abdominal
                                                        before 28 weeks’     assessed at 32      4. Hypoglycaemia circumference ≥ 75th
                                                        gestation.           weeks’ gestation    5. Hypocalcemia percentile at ultrasound
                                                                             only.               6. Hyperbilirubin assessment performed
                                                        In all cases                                               before 32 weeks’
                                                        gestational age                                            gestation.
                                                        had been
                                                        determined                                                 There was a statistically
                                                        previously and                                             significant increase in
                                                        verified by the                                            the percentage of
                                                        first-trimester                                            macrosomic infants born
                                                        ultrasound.                                                to women whose
                                                                                                                   ultrasound abdominal
                                                        Insulin plus                                               circumference
                                                        dietary therapy                                            assessment was
                                                        was initiated as                                           performed only at 32
                                                        soon as fetal                                              weeks’ gestation (Group
                                                        abdominal                                                  B) when compared to
                                                        circumference                                              women evaluated at
                                                        exceeded the                                               both 28 weeks’ and 32
                                                        75th percentile.                                           weeks’ gestation (Group
                                                        Women whose                                                A) [71.43% vs 33.33%,
                                                        abdominal                                                  p<0.05].
                                                        circumference
                                                        was less than                                              The rate of macrosomia
                                                        75th percentile                                            was reduced
                                                        were maintained                                            significantly to 11.1% in
                                                        on diet only.                                              women in Group A
                                                                                                                   where insulin was
                                                        Country:                                                   initiated at 28 weeks’
                                                        Italy                                                      gestation.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 203 of 441
                                                                                                                                      DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                        Follow-up &             Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                             Outcome                                                                    Comments
                 Level                 cs                                               Measures
Sacks DA;Chen    Study Type:       30    Inclusion criteria:   Intervention:          Follow-up period: Glucose concentrations        Glucose concentrations Very small sample size
W;Wolde-Tsadik   Cohort                  -Pregnant women       Compare glucose                          after morning meal vs         were significantly higher
G;Buchanan TA;                           between 28-38         values for women       Outcome           evening meal were             3-9 hours after an
                 Evidence level:         weeks’ gestation      with diet-controlled   Measures:         significantly greater at 1    evening meal, whereas
1999 Oct         2-                      -singleton            gestational diabetes   Plasma glucose    hour, not different at 2      suppression of free fatty
                                         pregnancies           during two 9-hr        Insulin/glucose   hours, and significantly      acids and beta-
159
                                         -gestational          fasts, 1 during the    ratios            lower from 3-9 hours          hydroxybutyrate was less
                                         diabetes being        day and 1 overnight    Free fatty acid   postprandially.               sustained after a morning
                                         managed by diet       Analyses of            Serum cortisol    Plasma beta-                  meal among women with
                                         alone                 hormones and           β-hydroxybutyrate hydroxybutyrate and           gestational diabetes
                                         -fasting plasma       metabolites involved                     free fatty acid               treated with diet alone
                                         glucose levels <      in intermediary                          concentrations higher 5-
                                         105 mg/Dl and         metabolism                               9 hours after morning         The mechanisms
                                         the 1h                                                         meal vs evening meal.         underlying these may
                                         postprandial          Comparison:                              Total and free cortisol       involve diurnal influences
                                         glucose level                                                  levels higher for the first   of counter-regulatory
                                         <140 mg/Dl                                                     7 hours after the             hormones.
                                         Exclusion criteria:                                            morning feeding.
                                         -cigarette                                                     Overweight patients'
                                         smoking                                                        glucose values
                                         -recreational drug                                             significantly greater than
                                         use                                                            those of lean subjects
                                         -chronic                                                       during the last 4 hours
                                         hypertension                                                   of the overnight fast.
                                         -cardiac,
                                         pulmonary,
                                         gastrointestinal or
                                         renal disease (if
                                         they were on
                                         medication for
                                         any of these
                                         conditions)

                                         Country: USA




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                         Page 204 of 441
                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type           Aim of Study         Number of      Population             Outcome       Results & Comments               Study            Reviewer
 Information & Evidence                                 Patients &    Characteristic          measures                                      Summary           Comment
                 Level                                    Patient          s
                                                       Characteristic
                                                             s
Sameshima        Study Type:     Intervention: 2       11            Inclusion criteria:   Blood glucose   Daily glucose profiles of the    In women with
H;Kamitomo       Other Paired    insulin-meal                        Japanese women                        two groups showed that their     tightly
M;Kajiya S;Kai   Prospective     intervals                           with gestational                      glycaemic control on the days    controlled
M;Ikenoue T;     study           15 min                              diabetes requiring                    of observation were good and     gestational
                                 30 min                              insulin for good                      that the two glucose profile     diabetes during
2001 Aug         Evidence Level: Both regimens                       glycaemic control                     curves were superimposable.      3rd trimester,
                 2-              examined in each                    (fasting level                                                         insulin-meal
186
                                 patient in random                   <100mg/dl and 2h                      A transient decrease in          intervals of 15
                                 order, 2 days apart                 postprandial level of                 glucose (nadir 62 +/- 6 mg/dl) min are
                                                                     <120 mg/dl                            was observed at 6-10 min of      beneficial when
                                 Comparison: 15 min                  Exclusion criteria:                   meal ingestion in the 30-min     compared to
                                 vs 30 min                           toxemia                               regimen, which was               30-min
                                                                     chronic                               significantly different from the intervals,
                                                                     hypertension                          glucose fluctuations during the because they
                                                                     diabetic                              15-min regimen.                  avoid
                                                                     complications                         2-h postprandial glucose levels preprandial
                                                                     thyroid disease                       were similar in both             hypoglycemia
                                                                     collagen disease                      experiments.                     without
                                                                     multiple                                                               increasing 2-h
                                                                     pregnancies                                                            postprandial
                                                                     cigarette smoking                                                      hyperglycaemia
                                                                                                                                            .
                                                                     Country: Japan




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                               Page 205 of 441
                                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                   Follow-up &            Effect Size               Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                        Outcome                                                                   Comments
                 Level                 cs                                                          Measures
Poyhonen-Alho   Study Type:       23 consecutive      Women with          Intervention: Short-   Follow-up period: The two groups were           The results suggest that
M;Teramo        RCT               women with          insulin-requiring   acting insulin                           similar in terms of age,      gestational diabetes can
K;Kaaja R;                        insulin-requiring   gestational                                Outcome           BMI and parity.               be treated with short-
                Evidence level:   gestational         diabetes            Comparison: Long-      Measures:                                       acting insulin.
2002 Mar        1+                diabetes.                               acting insulin         HbA1c             Insulin dose was raised
                                                      Country: Finland                                             in 6 of the 11 pregnant
184
                                                                                                                   women on short-acting
                                                                                                                   insulin.

                                                                                                                    Birth weight was higher
                                                                                                                    in the long-acting insulin
                                                                                                                    group (3943g ± 492g
                                                                                                                    and 3079g ± 722g,
                                                                                                                    respectively; p = 0.005).

                                                                                                                    There were no
                                                                                                                    differences in the mode
                                                                                                                    of delivery, neonatal
                                                                                                                    complications,
                                                                                                                    malformations, number
                                                                                                                    of infants with Apgar
                                                                                                                    score less than 7 at 1
                                                                                                                    min of age, or maternal
                                                                                                                    HbA1c between the
                                                                                                                    groups.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                   Page 206 of 441
                                                                                                                                           DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                            Follow-up &            Effect Size                Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                 Outcome                                                                    Comments
                 Level                 cs                                                   Measures
Thompson          Study Type:       108 women   68 women with        Intervention: Diet   Follow-up period: Among 68 women                 Maternal obesity or
DJ;Porter         RCT                           gestational          alone                                  successfully treated for       failure to achieve
KB;Gunnells                                     diabetes who                              Outcome           a minimum of 6 weeks,          glycaemic control should
DJ;Wagner         Evidence level:               received either      Comparison: Diet     Measures:         the mean birth weight          alert the clinician to a
PC;Spinnato JA;   1+                            diet alone or diet   plus insulin.        1. Mean birth     and macrosomia rate            substantially increased
                                                plus insulin (20                          weigh             were reduced                   risk of macrosomia.
1990 Jun                                        units NPH and 10                          2. Macrosomia     significantly in the
                                                units regular) for                        rate              insulin-treated group.
169
                                                glycaemic control
                                                were included in                                             Insulin reduced birth
                                                this study.                                                  weights significantly.
                                                                                                             (4060 ± 342 versus
                                                Country: USA                                                 3397 ± 640 g).

                                                                                                             No woman with good
                                                                                                             glucose control had a
                                                                                                             newborn over 4000 g.
                                                                                                             Women with poor
                                                                                                             glycaemic control were
                                                                                                             at greatest risk (30%) for
                                                                                                             fetal overgrowth whether
                                                                                                             initially receiving insulin
                                                                                                             or not.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                             Page 207 of 441
                                                                                                DRAFT FOR CONSULTATION




Q14 – What is cost effective treatment for gestational diabetes?

No specific searches were conducted for this clinical question and so there are no evidence tables (this is a health
economics question)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                               Page 208 of 441
                                                                                                                                              DRAFT FOR CONSULTATION




Q15 - What are the target ranges for blood glucose during pregnancy?

Bibliographic Study Type Number of    Patient      Intervention                                   Follow-up &           Effect Size             Study Summary                    Reviewer
 Information & Evidence   Patients Characteristics       &                                         Outcome                                                                      Comments
                 Level                             Comparison                                      Measures
Jovanovic        Study Type:       52 women with     Mean age 28.6           Intervention:       Follow-up period: All 52 patients            Problems of morbidity        The original study design
L;Druzin         Cohort            type 1 diabetes   years and had type      Euglycaemia                           maintained glucose         and mortality may be         was to randomise patients
M;Peterson CM;                     52 nondiabetic    1 diabetes for mean     program instituted  Outcome           control within the         eliminated if patients are   with insulin-dependent
                 Evidence level:   women             duration of 7.1 years   prior to the 12th   Measures: Fetal desired range                maintained euglycemic        diabetes into two groups.
1981 Dec         2+                                  (range 1 to 23).        week of             loss              throughout gestation in    from the 12th week.          Group I would receive
                                                                             pregnancy           Malformations     the outpatient                                          standard care and group
194
                                                                                                 Birthweight       department, except for                                  2 would receive intensive
                                                                             Euglycemia in the Hypoglycaemia       two patients who were                                   care. However after 4
                                                                             diabetic group      Respiratory       hospitalised for 53 and                                 deliveries in the standard
                                                                             was documented distress               45 days respectively to                                 care group resulted in two
                                                                             by daily home-      1 and 5 min       ensure glucose control.                                 infants with macrosomia,
                                                                             monitored           Apgar scores                                                              three infants with severe
                                                                             glucose             Hypocalcaemia     During gestation there                                  neonatal hypoglycaemia,
                                                                             determinations                        was no deterioration in                                 one delivery complicated
                                                                             and bimonthly                         renal or ocular status                                  by shoulder dystocia, and
                                                                             HbA1c levels and                                                                              one pregnancy with
                                                                             25 hr glucose                         No fetal loss.                                          premature rupture of
                                                                             profiles and                          Mean body weight                                        membranes and fetal
                                                                             urinary collections                   2910g, no infants above                                 death, it was decided that
                                                                             during periods of                     75th percentile or below                                to continue with the
                                                                             hospitalisation, at                   48th percentile.                                        original protocol would be
                                                                             presentation, 20                      No major or minor                                       unethical. Therefore all
                                                                             or 26 weeks, and                      malformations                                           patients were offered the
                                                                             at term.                              (compared to 2 minor                                    intensive program.
                                                                                                                   and 2 major
                                                                             Therapy was                           malformations in the
                                                                             initiated during a                    nondiabetic group,
                                                                             5 to 7 day                            p<0.03).
                                                                             hospitalisation in                     All infants maintained
                                                                             which diet and                        blood glucose levels
                                                                             insulin regulation                    between 3.0 and 4.4
                                                                             were calculated                       mmol/l except for twin B
                                                                             according to body                     who had asymptomatic
                                                                             weight and                            blood glucose 2.2
                                                                             adjusted to                           mmol/l at 2 hours.
                                                                             achieve 'normal'                      No respiratory distress
                                                                             bloodglucose                          compared to 5 cases in
                                                                             levels, defined as                    nondiabetic group
                                                                             a fasting level of                    (p<0.03)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                 Page 209 of 441
                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                Follow-up &        Effect Size      Study Summary           Reviewer
 Information & Evidence   Patients Characteristics       &                      Outcome                                                   Comments
                 Level                             Comparison                   Measures
                                                          3.0-3.6 mmol/l, a                  No other morbidity.
                                                          mean blood
                                                          glucose level of
                                                          4.4 to 4.8 mmol/l
                                                          and a one-hour
                                                          postprandial level
                                                          not exceeding
                                                          7.7mmol/l.

                                                          Comparison: 52
                                                          nondiabetic
                                                          women matched
                                                          for maternal age,
                                                          mode of delivery,
                                                          gestational age,
                                                          and
                                                          socioeconomic
                                                          status and ethnic
                                                          background.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                Page 210 of 441
                                                                                                                                 DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                    Follow-up &             Effect Size           Study Summary           Reviewer
 Information & Evidence   Patients Characteristics       &                          Outcome                                                             Comments
                 Level                             Comparison                       Measures
Wyse LJ;Jones   Study Type:       66     22 type 1, 4 type 2,   Intervention:     Follow-up period:   19/66 (29%) had
M;Mandel F;     Cohort                   41 gestational         Glycaemic control                     birthweight macrosomia.
                                                                                  Outcome
1994 Jul        Evidence level:                                 Comparison:       Measures:           13/66 (20%) had at least
                2+                                                                HbA1c (the one      one marker. Of these
192
                                                                                  taken closest to    13, 10 had HbA1c>6.3%
                                                                                  the ultrasound)     and also had birthweight
                                                                                                      macrosomia.

                                                                                  Fetal               There was a significant
                                                                                  complications       difference in HbA1c
                                                                                  based on            between those with
                                                                                  ultrasound          markers and those
                                                                                  assessment:         without (P<0.03)
                                                                                  Macrosomia
                                                                                  (estimated fetal    To reduce the incidence
                                                                                  weight >90th        of birthweight
                                                                                  percentile for      macrosomia glycaemic
                                                                                  gestational age     control needs to be
                                                                                  uing using the      geared to maintaining
                                                                                  Williams growth     long term euglycaemia
                                                                                  curve;              or a HbA1c valus <6.3%
                                                                                  polyhydramnios
                                                                                  (amniotic fluid
                                                                                  index ≥25cm or
                                                                                  one fluid pocket
                                                                                  ≥8cm, or fat line
                                                                                  (≥ 5mm
                                                                                  subcutaneous
                                                                                  tissue near the
                                                                                  level of the
                                                                                  umbilical vein
                                                                                  insertion).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                              Page 211 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                                   Follow-up &           Effect Size              Study Summary               Reviewer
 Information & Evidence   Patients Characteristics       &                                         Outcome                                                                  Comments
                 Level                             Comparison                                      Measures
Karlsson        Study Type:       180 infants in     Gestational diabetes   Intervention:        Follow-up period: Perinatal deaths:           There is a significant
K;Kjellmer I;   Cohort            179 women, 20      20,White class B 49,   Maternal blood                         <5.5 mmol/l: 2/52 (3.8%)    reduction of the perinatal
                                  with gestational   White Class C 60,      sugar                Outcome           5.5-8.25 mmol/l: 12/77      mortality rate to about
1972 Jan 15     Evidence level:   diabetes , 160     White class D                               Measures:         (16%)                       4% when the maternal
                2+                with pre-          42,White Class F+R     Blood sugar was      Perinatal deaths >8.25 mmol/l 9/38:           blood glucose level is
200
                                  existing type 1    9.                     routinely                              (24%)                       reduced below
                                  diabetes.                                 measured at least    Hypoglycaemia                                 5.5mmol/l.
                                                     84 patients treated    three times/day.                       Neonatal hypoglycaemia
                                                     between 1961 and       For each patient a   Jaundice          <5.5 mmol/l: 6/52 (12%)     The low frequency of
                                                     1965                   mean blood sugar                       5.5-8.25 mmol/l: 8/77       neonatal hypoglycaemia
                                                     95 mothers treated     level was            Respiratory       (10%)                       may be due to the earlier
                                                     between 1966 and       calculated for the   distress          >8.25 mmol/l: 4/38          provision of fluid and
                                                     1970 (more             time period                            (11%)                       nutrients compared to
                                                     intensive therapy)     between 30 and       Malformations                                 other studies
                                                                            32 weeks until                         Jaundice
                                                                            delivery.                              <5.5 mmol/l: 0/52 (0%)
                                                                                                                   5.5-8.25 mmol/l:6/77
                                                                            Women divided                          (8%)
                                                                            into three groups:                     >8.25 mmol/l: 3/38 (8%)
                                                                            <5.5mmol/l
                                                                            5.5-8.25mmol/l                         Respiratory distress
                                                                            >8.25mmol/l                            <5.5 mmol/l: 9/52 (17%)
                                                                                                                   5.5-8.25
                                                                            Comparison:                            mmol/l:24/77(31%)
                                                                                                                   >8.25 mmol/l:9/38 (24%)

                                                                                                                   Malformations
                                                                                                                   <5.5 mmol/l: 2/52 (3.8%)
                                                                                                                   5.5-8.25 mmol/l:12/77
                                                                                                                   (16%)
                                                                                                                   >8.25 mmol/l:5/38 (13%)

                                                                                                                   There was a statistically
                                                                                                                   significant difference
                                                                                                                   (p<0.05) between
                                                                                                                   women with blood
                                                                                                                   glucose values <5.5
                                                                                                                   mmol/l and those above
                                                                                                                   for perinatal death,
                                                                                                                   respiratory distress,
                                                                                                                   jaundice and
                                                                                                                   malformations.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 212 of 441
                                                                                                                  DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention            Follow-up &       Effect Size          Study Summary           Reviewer
 Information & Evidence   Patients Characteristics       &                  Outcome                                                      Comments
                 Level                             Comparison               Measures
                                                                                         No association was
                                                                                         found between maternal
                                                                                         hyperglycaemia and
                                                                                         overweight neonate at
                                                                                         birth.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                               Page 213 of 441
                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                    Follow-up &           Effect Size             Study Summary               Reviewer
 Information & Evidence   Patients Characteristics       &                          Outcome                                                                 Comments
                 Level                             Comparison                       Measures
Landon         Study Type:       75 women with White class B    Intervention:      Follow-up period: Group1 (<6.0 mmol/l):     These data suggest that
MB;Gabbe       Cohort            type 1 diabetes through to D   Maternal blood                       n= 43                     maintaining mean
SG;Piana                                                        glucose levels     Outcome           Group 2(>6.0mmol/l):      capillary blood glucose
R;Mennuti      Evidence level:                                                     Measures:         n=32                      values <6.0 mmol/l may
MT;Main EK;    2++                                              Optimal glucose Hypoglycaemia                                  serve to reduce several
                                                                control was        (<1.65 mmol/l)    Gestational age (wk)      major forms of morbidity
1987 May                                                        considered                           Group 1: 38.42 ± 1.4      in the infant of the
                                                                present if fasting Macrosomia        Group 2: 38.15 ± 2.2,     diabetic mother. This
191
                                                                capillary glucose                    NS                        information is helpful in
                                                                values were ≤ 5.5 Respiratory                                  establishing objectives
                                                                mmol/l and         distress          Morbidity                 for glycaemic control in
                                                                preprandial                          Group 1: 14 (32%)         pregnant women using
                                                                values were        LGA (birth weight Group 2 17 (53.1%), NS    self-monitoring
                                                                ≤6.6mmol/l.        ≥ 90th percentile                           techniques.
                                                                                                     Nursery days
                                                                Patients           SGA (≤ 10th       Group 1: 4.72 ± 1.4
                                                                performed a daily percentile)        Group 2: 6.56 ± 2.0,
                                                                fasting and three                    p<0.001.
                                                                preprandial
                                                                determinations                       Birth weight >4000g
                                                                throughout the                       Group 1: 7%
                                                                second and third                     Group2: 28%,p<0.05
                                                                trimesters. These
                                                                values were used                     LGA
                                                                to calculate a                       Group 1:4 (9.3%)
                                                                mean capillary                       Group 2: 11 (34.3%),
                                                                blood glucose.                       p<0.05
                                                                Mean capillary
                                                                blood glucose                        Neonatal hypoglycaemia
                                                                was calculated                       Group 1: 8 (18.6%)
                                                                from a minimum                       Group 2: 13
                                                                of 16 weeks of                       (40.6%),p<0.05
                                                                determinations.
                                                                                                     Respiratory distress
                                                                The study                            syndrome
                                                                population was                       Group 1:1 (2.3%)
                                                                divided into two                     Group 2: 7 (21.8%),
                                                                groups based on                      p<0.01
                                                                mean capillary
                                                                blood glucose                        Hyperbilirubinemia
                                                                values. Optimal                      Group 1:10 (23.2%)
                                                                control was                          Group 2: 13 (40.6%), NS




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                  Page 214 of 441
                                                                                                                    DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention               Follow-up &        Effect Size        Study Summary           Reviewer
 Information & Evidence   Patients Characteristics       &                     Outcome                                                     Comments
                 Level                             Comparison                  Measures
                                                          defined as a
                                                          mean capillary                    Polycythemia
                                                          blood glucose                     Group 1:1(2.3%)
                                                          level <6.0 mmol/l                 Group 2: 3 (9.3%)

                                                          Comparison:                       Hypocalcaemia
                                                                                            Group1: 0
                                                                                            Group 2: 2 (6.2%)

                                                                                            There was only one
                                                                                            perinatal death which
                                                                                            occurred in an infant
                                                                                            with a malformation
                                                                                            whose mother
                                                                                            presented at 12 weeks
                                                                                            with a HbA1c of 10.2%




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                 Page 215 of 441
                                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                        Follow-up &           Effect Size             Study Summary                   Reviewer
 Information & Evidence   Patients Characteristics       &                              Outcome                                                                     Comments
                 Level                             Comparison                           Measures
Miodovnik M;   Study Type:       145 women   Type 1 diabetes.   Intervention:         Follow-up period: Excluding 17               History of premature        Tocolytic agents
               Cohort                        Recruited during   Glucose control                          pregnancies with          labour in a previous        (isoxsuprine
1987                                         first trimester.   Goals of glucose      Outcome            premature delivery due    pregnancy, premature        hydrochloride until 1980,
               Evidence level:                                  control were          Measures:          to fetal distress the     rupture of membranes,       then ritrodrine
201
               2++                                              fasting blood         HbA1c every four premature labour rate       poor second trimester       hydrochloride) were used
                                                                glucose <5.5          weeks (normal      was 51/164 (31.1%) vs     glycaemic control, and      in all cases except for
                                                                mmol/l and a 90       range 5.5-8.5%). 828/4109 (20.2%) in the     urogenital infection were   patients with ruptured
                                                                minute                Pre and            control population.       significantly associated    chorioamniotic
                                                                postprandial          postprandial                                 wit premature labour.       membranes of cervical
                                                                glucose level <7.7    capillary glucose HbA1 at 26 wks (%)                                     dilation of 4cm or more.
                                                                mmol/l.               were measured      No premature labour:      Polyhydramnios and
                                                                                      alternately at     7.9 ± 1.5                 hypogmagnesemia,
                                                                Glycaemic control     each weekly visit. Premature labour:         contrary to hypothesis,
                                                                was obtained with                        8.7 ± 1.8, p<0.01         were not factors
                                                                a split dosage        Premature labour:                            significantly associated
                                                                regiment of           Gestational age    Previous premature        with premature labour.
                                                                insulin, using both   20-37 weeks with delivery:                   (May reflect sample size
                                                                short and long        evidence of        No premature labour:31    and socio-economic
                                                                lasting insulin,      regular uterine    (27.4)                    profile of sample ).
                                                                and dietary           contractions       Premature labour: 20
                                                                regulation.           documented by      (39.2), p=0.001           Improvement of
                                                                Patients were         tocodynometry,                               glycaemic control and
                                                                hospitalised          and cervical       Premature rupture of      eradication of urogenital
                                                                where necessary       dilation of 3cm or membranes:                infection will decrease
                                                                to maintain these     more and/or        No premature labour: 2    the spontaneous
                                                                goals.                progressive        (1.8)                     premature labour rate in
                                                                                      effacement of the Premature labour:          these patients.
                                                                Comparison:           cervix.            14 (27.4), p<0.001
                                                                Control                                  Candida vaginitis:
                                                                population 4109                          No premature labour: 28
                                                                nondiabetic                              (24.5)
                                                                pregnant women                           Premature labour:
                                                                who received                             21 (41.2%), p<0.03
                                                                prenatal and
                                                                delivery care from                      Urogenital infection:
                                                                the same                                No premature labour: 49
                                                                obstetricians.                          (43.4)
                                                                                                        Premature labour:
                                                                                                        34 (66.7), p<0.005

                                                                                                        Logistic regression,
                                                                                                        once the contribution of




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                      Page 216 of 441
                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention            Follow-up &        Effect Size          Study Summary           Reviewer
 Information & Evidence   Patients Characteristics       &                  Outcome                                                       Comments
                 Level                             Comparison               Measures
                                                                                         pregnancy history and
                                                                                         premature rupture of
                                                                                         membranes were
                                                                                         excluded, confirmed
                                                                                         HbA1c (R2=0.105) and
                                                                                         urogenital infection
                                                                                         (R2=0.093) as
                                                                                         significant covariates
                                                                                         predicting the onset of
                                                                                         premature labour.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                Page 217 of 441
                                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                   Follow-up &             Effect Size             Study Summary                Reviewer
 Information & Evidence   Patients Characteristics       &                         Outcome                                                                    Comments
                 Level                             Comparison                      Measures
Rosenn B;      Study Type:       171     Women with type 1    Intervention:      Follow-up period:   39 minor congenital        poor glycaemic
               Cohort                    diabetes. 86% had    Glycaemic control                      malformations              controlduring the latter
1990                                     entered study by                        Outcome             diagnosed in 32 infants    part of the first trimester
               Evidence level:           13th week            HbA1c measured Measures: Minor         (18.7%)                    is associated with an
202
               2+                        gestation.           every 4 weeks      congenital                                     increased risk of
                                                              (normal range      malformations       HbA1c week 8               minormalformations.
                                         Goals of glycaemic 5.5-8.5%).Pre or                         Malformation group         Improving glycaemic
                                         control were a       postprandial                           (n=32): 10.6 ± 1.7         control during this period
                                         fasting blood        capillary glucose                      No malformation group      should decrease the risk.
                                         glucose              measured                               (n=139): 9.8 ± 2.2, NS
                                         concentration of     alternately at
                                         ≤5.5 mmol/l and a    each weekly visit.                     HbA1c week 12
                                         90-minute            After 1982                             Malformation group:
                                         postprandial glucose patients routinely                     10.0 ± 1.6
                                         concentration <7.7 recorded self-                           No malformation:9.0 ±
                                         mmol/l.              monitoring of                          1.8 (p=0.02)
                                                              glucose with
                                         Glycaemic control    reflectance                            HbA1c week 16
                                         obtained with split  meters.                                Malformation group: 9.1
                                         dosage regimen of                                           ± 1.5
                                         short and            Comparison:                            No malformation: 8.3 ±
                                         intermediate acting                                         1.6, (p=0.02)
                                         insulin + dietary
                                         control                                                     HbA1c week 20
                                                                                                     Malformation group: 8.9
                                                                                                     ± 1.3
                                                                                                     No malformation: 7.9 ±
                                                                                                     1.5, p=0.001

                                                                                                     HbA1c week 24
                                                                                                     Malformation group: 8.1
                                                                                                     ± 0.9
                                                                                                     No malformation group:
                                                                                                     7.8 ± 1.5, NS

                                                                                                     Mulitvariate analyis
                                                                                                     indicated that HbA1c at
                                                                                                     20 weeks was the
                                                                                                     strongest independent
                                                                                                     variable correlated with
                                                                                                     the presence of minor
                                                                                                     malformations (p=0.02).




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                   Page 218 of 441
                                                                                                                                               DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention                                 Follow-up &              Effect Size             Study Summary                      Reviewer
 Information & Evidence   Patients Characteristics       &                                       Outcome                                                                           Comments
                 Level                             Comparison                                    Measures
Jovanovic-          Study Type:       323 women        Recruited before 21 Intervention:      Follow-up period:   LGA (>90th percentile)       The data indicate that        Elevated blood glucose
Peterson            Cohort            with diabetes,   days gestation      Glycaemic control                      Diabetic: 92 (28.5%)         glycaemic control in late     variables in the first
L;Peterson                            361 control                                             Outcome             Control: 47 (13.1%),         pregnancy is the              trimester correlated
CM;Reed             Evidence level:                                        No fixed goals for Measures:           p<0.001                      strongest metabolic           significantly with elevate
GF;Metzger          2++                                                    glycaemic control Fasting and          The relationship             determinant of birth          variables in the third
BE;Mills JL;Knopp                                                                             nonfasting          between metabolic            weight, but there is a        trimester.
RH;Aarons JH;                                                              Comparison:        venous plasma       parameters and birth         major contribution from
                                                                                              glucose were        weight when each             the nonfasting and
1991 Jan                                                                                      measured on         trimester analyzed           glycosated haemoglobin
                                                                                              alternate weeks in  independently:               from both prior
196
                                                                                              the first trimester (slope ± SE)                 trimesters.
                                                                                              and monthly         Fasting glucose:
                                                                                              thereafter.         First trimester:             Our study shows that
                                                                                                                  0.002±0.002, p=0.248         peak glucose levels
                                                                                               HbA1c measured Second trimester: 0.006          during the day are more
                                                                                               week; in the first ± 0.002, p=0.024             predictive of the infant
                                                                                               trimester and      Third trimester:             birth weight than the
                                                                                               monthly thereafter 0.005 ± 0.003, p=0.068       lowest blood glucose
                                                                                                                                               level of the day (fasting).
                                                                                                                    Nonfasting glucose:
                                                                                                                    First trimester: 0.006 ±   Our data also indicate
                                                                                                                    0.002, p<0.001             that monitoring only
                                                                                                                    Second trimester:          fasting and premeal
                                                                                                                    0.006 ± 0.002, p=0.02      glucose concentrations,
                                                                                                                    Third trimester:           which reflect the lowest
                                                                                                                    0.013 ± 0.004, p<0.001     glucose levels of the day,
                                                                                                                    HbA1c                      does not provide an
                                                                                                                    First trimester:           adequate indication of
                                                                                                                    0.25 ± 0.009,              overall metabolic control
                                                                                                                    p=0.008                    and risk of macrosomia.
                                                                                                                    Second trimester: 0.032    If women who are at risk
                                                                                                                    ± 0.012, p=0.009           of macrosomia are to be
                                                                                                                    Third trimester:           identified early and have
                                                                                                                    0.047 ± 0.013, p<0.001     their treatment altered to
                                                                                                                                               improve metabolic
                                                                                                                    After adjusting for        control, optimum
                                                                                                                    variables in the prior     surveillance should
                                                                                                                    trimester or trimesters,   include measurements of
                                                                                                                    second trimester mean HbA1c and nonfasting
                                                                                                                    fasting plasma glucose (postprandial) blood
                                                                                                                    (p=0.035), third trimester glucose concentrations.
                                                                                                                    mean nonfasting plasma




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 219 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristics       &                  Outcome                                                         Comments
                 Level                             Comparison               Measures
                                                                                         glucose (p=0.009) and
                                                                                         third trimester HbA1c
                                                                                         (p=0.018) correlated
                                                                                         with infant birth weight.

                                                                                         After adjusting for
                                                                                         maternal hypertension;
                                                                                         renal dysfunction,
                                                                                         smoking, ponderal
                                                                                         index, and weight gain,
                                                                                         the above relationships
                                                                                         remained.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 220 of 441
                                                                                                                            DRAFT FOR CONSULTATION




Bibliographic Study Type Number of    Patient      Intervention              Follow-up &            Effect Size               Study Summary               Reviewer
 Information & Evidence   Patients Characteristics       &                    Outcome                                                                    Comments
                 Level                             Comparison                 Measures
Evers IM;De Valk   Study Type:       289   Women with type   Intervention:   Follow-up period: Macrosomia: 49% when         Glucose variability and
HW;Mol BWJ;Ter     Cohort                  1diabetes         HbA1c                             compared to growth           postprandial peaks,
Braak                                                                        Outcome           charts based on data         which are not reflected in
EWMT;Visser        Evidence level:                           Comparison:     Measures:         collected in 1998.           the HbA1c value, may be
GHA;               2++                                                       Macrosomia (birth 73/289 had a birth           involved
                                                                             weight >90th      weight > 4000 gms.
2002                                                                         percentile)                                    Future research should
                                                                                               Overall blood glucose        focus on new more
195
                                                                                               control was good ( mean      detailed glucose
                                                                                               HbA1c ≤ 7.0%) in 84%         monitoring (such as
                                                                                               of mothers.                  continuous glucose
                                                                                                                            monitoring)
                                                                                               HbA1c first trimester
                                                                                               Macrosomia: 6.7 ± 1.1
                                                                                               No Macrosomia: 6.4 ±
                                                                                               1.0, p=0.07

                                                                                               HbA1c second trimester
                                                                                               Macrosomia: 6.1 ± 1.0
                                                                                               No Macrosomia: 5.8 ±
                                                                                               0.8, p=0.007

                                                                                               HbA1c third trimester
                                                                                               Macrosomia: 6.5 ± 1.1
                                                                                               No Macrosomia: 6.0 ±
                                                                                               1.1, p=0.0001

                                                                                               Third trimester HbA1c
                                                                                               >6.5: RR 2.7 95% CI
                                                                                               1.3-5.4

                                                                                               In logistic regression 3rd
                                                                                               trimester HbA1c was the
                                                                                               most powerful predictor
                                                                                               for the occurrence of
                                                                                               macrosomia but only
                                                                                               explained 4.7% of
                                                                                               variance in macrosomia.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                               Page 221 of 441
                                                                                                                                              DRAFT FOR CONSULTATION




Bibliographic Study Type          Aim of Study            Number of      Population                     Outcome              Results & Comments               Study             Reviewer
 Information & Evidence                                   Patients &    Characteristic                  measures                                             Summary            Comment
                 Level                                      Patient          s
                                                         Characteristic
                                                               s
Langer          Study Type:     Intervention:            2461 women with     Diagnosed with         Blood glucose (mean     Blood glucose (mean per day) Memory               No differences
O;Rodriguez     OtherNon        Intensive                gestational         gestational diabetes   per day)                Intensified group: 5 ± 2     reflectance          between the
DA;Xenakis      randomised      management:              diabetes            (1 hour glucose        Labour augmentation                                  meters enabled       groups with
EM;McFarland    intervention    patients assigned        1316 conventional   challenge test         (%)                     Labour augmentation (%)      us to                regard to parity,
MB;Berkus       study           memory reflectance       management          (50gm) followed by     Induction (%)           Conventional: 28             demonstrate a        ethnicity, obesity,
MD;Arrendondo                   meters and               1145 intensified    100gm OGTT if          No of days of           Intensified: 15, p<0.01      significant          past obstetric
F;              Evidence Level: instructed by nurse      management          plasma glucose ≥       hospital stay (mean ±   Non diabetic control: 28,    correlation          history,family
                2++             educators in self                            7.2 mmol/l)            SD)                     p<0.01                       between rates        history,
1994 Apr                        monitoring blood                             between week 24        Primary caesarean                                    of LGA and           gestational age at
                                glucose technique (7                         and week 28 of         section (%)             Induction (%)                macrosomia           entry, degree of
190
                                times/day: fasting,                          pregnancy.             LGA (≥90th              Conventional:27              and                  abnormality of
                                preprandial, 2-hr                            83% Hispanic, 12%      percentile)             Intensified: 22, p<0.01      glycaemiccontr       the screening or
                                postprandial and at                          white, 5% black        Macrosomia (birth       Non diabetic control: 13     ol. Verified self-   GTTs.
                                bedtime).                                                           weight ≥ 4000gm)                                     monitored
                                                                                                    Neonatal intensive      Failed induction (%)         blood glucose
                                Treatment (diet or                                                  care unit (%            Conventional: 10             measurement
                                diet+insulin) for both                                              admission)              Intensified: 7, p<0.01       enhanced
                                groups the same.                                                    Respiratory             Non diabetic control: 10     glycaemia
                                                                                                    complications (%):                                   control and
                                Goals for both                                                      Shoulder dystocia:      No of days of hospital stay  significantly
                                groups: Mean blood                                                                          (mean ± SD)                  reduced
                                glucose (overall                                                                            Conventional 4.3±4           macrosomia.
                                mean from diagnosis                                                                         Intensified 3.7± 4, p<0.01
                                to delivery) between                                                                        Non diabetic control:3.3 ± 3
                                4.95 and 5.5 mmol/l,
                                fasting blood                                                                               Primary caesarean section (%)
                                glucose between                                                                             Conventional: 19
                                3.3and 4.95 mmol/l                                                                          Intensified: 13.0, p<0.01
                                and postprandial                                                                            Non diabetic control: 11
                                blood glucose <6.6
                                mmol/l.                                                                                     LGA (%)
                                                                                                                            Conventional:20.1
                                Comparison:                                                                                 Intensified: 13.1, p<0.0001
                                Conventional                                                                                Nondiabetic control: 11.9
                                management:
                                Patients instructed                                                                         Macrosomia
                                by a nurse educator                                                                         Conventional: 13.6
                                and assessed                                                                                Intensified: 7.1, p<0.0001




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 222 of 441
                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type      Aim of Study             Number of      Population      Outcome     Results & Comments                Study         Reviewer
 Information & Evidence                                Patients &    Characteristic   measures                                     Summary        Comment
                 Level                                   Patient          s
                                                      Characteristic
                                                            s
                            weekly for fasting                                                   RR 2.07, 95% CI 1.6-2.8
                            and 2-hour
                            postprandial venous                                                  Neonatal intensive care unit
                            plasma glucose                                                       (% admission)
                            during clinic visits. 4                                              Conventional: 25.0
                            daily self-monitored                                                 Intensified: 6.3, P<0.0001
                            blood glucose                                                        Non diabetic control: 4.7
                            determinations with
                            strips.                                                              Respiratory complications (%):
                                                                                                 Conventional: 6.2
                                                                                                 Intensified: 2.3, p<0.0001
                                                                                                 Non diabetic control: 2.1

                                                                                                 Shoulder dystocia:
                                                                                                 Conventional: 1.4
                                                                                                 Intensified: 0.4, p<0.0001
                                                                                                 Nondiabetic control: 0.5

                                                                                                 Stillbirth:
                                                                                                 Conventional: 4/1000
                                                                                                 Intensified:1/1000, NS
                                                                                                 Nondiabetic control: 4/1000

                                                                                                 Neonatal death:
                                                                                                 Conventional: 2/1000
                                                                                                 Intensified:3/1000, NS
                                                                                                 Nondiabetic control:4.7/1000

                                                                                                 Logistic regression: only
                                                                                                 overall mean blood glucose
                                                                                                 (seven determinations per day)
                                                                                                 previous LGA infants and
                                                                                                 gestational age at delivery
                                                                                                 predicted fetal macrosomia.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                        Page 223 of 441
                                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                  Follow-up &         Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                       Outcome                                                             Comments
                 Level                 cs                                                         Measures
Nielsen GL,      Study Type:       573               Women with type     Intervention:      Follow-up period: Of 573 pregnancies, 165 First-trimester HbA1c
Møller M,        Cohort            pregnancies in    1 diabetes          Routine ultrasound First through third (29%) terminated with     levels slightly >7% show
Sørensen HT.                       women with        residing in North                      gestational week. adverse outcomes.
                                                                         scan (11, 14, 20, 21                                             a dose-dependent
                 Evidence level:   type 1 diabetes   Jutland County,     and 24 gestational                                               association with risk of
2006             2++                                 Denmark whose       weeks) of women    Outcome             The prevalence of         adverse pregnancy
                                                     pregnancies                            Measures:
                                                                         with type 1 diabetes.                  adverse outcomes          outcome, but HbA1c
{Nielsen, 2006                                       ended between                          1. Congenital       varied six-fold from 12% levels have limited value
37159 /id}                                           1985 and 2003.    Comparison:          anomalies           (95% CI 7.2 to 17) in the in predicting outcome in
                                                                       Routine ultrasound                       lowest to 79% (95% CI individual pregnancies.
                                                     Women with        scan (11, 14, 20, 21 2. Morbidity of the 60% to 91%) in the
                                                     gestational or    and 24 gestational   infant diagnosed highest quintile of
                                                     type 2 diabetes   weeks) of women      within the first    HbA1c measurements.
                                                     were not included without diabetes.    month of life.
                                                     in the study.                                              From HbA1c levels
                                                                                            3. Mortality        >7%, there was almost
                                                     Country:                                                   linear association
                                                     Denmark                                                    between with risk of
                                                                                                                adverse outcome: a 1%
                                                                                                                increase in HbA1c
                                                                                                                corresponded to 5.5%
                                                                                                                (95% CI 3.8% to 7.3%)
                                                                                                                increased risk of
                                                                                                                adverse outcome.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                            Page 224 of 441
                                                                                                                                                     DRAFT FOR CONSULTATION




Q16 - How should blood glucose and ketones be monitored during pregnancy?

Bibliographic Study Type Number of   Patient     Intervention &                                       Follow-up &             Effect Size              Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                            Outcome                                                                   Comments
                 Level                 cs                                                              Measures
Crowther            Study Type:       490 women in   Women between         Intervention:             Follow-up period: The rate of serious           Treatment of gestational
CA;Hiller JE;Moss   RCT               intervention   16-30 weeks           Treatment                                    perinatal outcomes           diabetes in the form of
JR;McPhee                             group          gestation. One or     (institution's clinical   Outcome            among infants was            dietary advice, blood
AJ;Jeffries         Evidence level:   510 women in   more risk factors     practice for              Measures: A        significantly lower in the   glucose monitoring and
WS;Robinson         1++               routine care   for gestational       gestational               composite          intervention group than      insulin therapy as
JS;Australian                         group          diabetes on           diabetes) including       measure of         the routine care group       required for glycaemic
Carbohydrate                                         selective             individualised            serious perinatal (1% vs. 4%; p=0.01,           control reduces the rate
Intolerance Study                                    screening or a        dietary advice,           complications      adjusted for maternal        of serious perinatal
in Pregnant                                          +ve 50g GCT and       instructions on self-     (death, shoulder age, ethnic group and          complications without
Women                                                had a 75g OGTT        monitoring (4xday),       dystocia, bone     parity).                     increasing the rate of
(ACHOIS) Trial                                       at 24-34 weeks        target levels for         fracture, nerve                                 cesarean section.
Group.;                                              gestation in which    blood glucose of          palsy)             The number needed to
                                                     venous plasma         between 3.5 and 5.5       Admission to       treat to prevent a
2005 Jun 16                                          glucose               mmol/l (fasting) <5.5     neonatal nursery serious outcome in an
                                                     <7.8mmol/l after      mmol/l (preprandial)      Jaundice           infant was 34 (95% CI
{Crowther, 2005                                      an overnight fast     and <7.0 mmol/l           requiring          20-103).
28420 /id}                                           and 7.8-11.0          (postprandial) and        phototherapy
                                                     mmol/l at 2 hours.    insulin therapy.          Induction of       Admission to neonatal
                                                     Exclusions:                                     labour             nursery
                                                     Women                 Comparison:               Caesarean          Intervention: 71%
                                                     previously treated    Routine care              section            Routine care: 61%,
                                                     for gestational                                 Maternal health    adjusted p=0.01
                                                     diabetes or active                              (SF-36)
                                                     chronic systemic                                Anxiety            Jaundice requiring
                                                     disease (except                                 (Spielberger       phototherapy:
                                                     essential                                       State-Trait        No significant difference
                                                     hypertension).                                  Anxiety Inventory) (adjusted p=0.72)
                                                                                                     Depression
                                                     Country: Multi-                                 (Edinburgh         Induction of labour:
                                                     centre (Australia ,                             Postnatal          Intervention: 39%
                                                     UK)                                             Depression         Routine care: 29%,
                                                                                                     Scale)             adjusted p<0.001

                                                                                                                         Caesarean section:
                                                                                                                         No significant difference
                                                                                                                         (adjusted p=0.73)

                                                                                                                         All measures on the SF-
                                                                                                                         36 showed trends in




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                       Page 225 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         favour of the
                                                                                         intervention group but
                                                                                         not all were significant

                                                                                         Depression
                                                                                         At 3 months postpartum
                                                                                         fewer women in the
                                                                                         intervention group had a
                                                                                         score on the Edinburgh
                                                                                         postnatal depression
                                                                                         scale suggestive of
                                                                                         depression (8% vs 17%)

                                                                                         Anxiety
                                                                                         Similar in both groups.

                                                                                         Birthweight (g)
                                                                                         Intervention: 335±551
                                                                                         Routine care: 3482±660,
                                                                                         adjusted p<0.001


                                                                                         Secondary outcomes:
                                                                                         Neonatal:
                                                                                         LGA
                                                                                         Intervention: 68 (13%)
                                                                                         Routine care: 115
                                                                                         (22%), adjusted p<0.001

                                                                                         Macrosomia ≥4kg
                                                                                         Intervention:49 (10%)
                                                                                         Routine care: 110
                                                                                         (21%), adjusted p<0.001

                                                                                         Neonatal
                                                                                         Hypoglycaemia requiring
                                                                                         IV therapy:
                                                                                         No significant difference
                                                                                         (adjusted p=0.16)

                                                                                         Respiratory distress
                                                                                         syndrome:




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 226 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         No significant difference
                                                                                         (adjusted p=0.15)

                                                                                         Maternal:
                                                                                         Antenatal preeclampsia:
                                                                                         Intervention group: 58
                                                                                         (12)
                                                                                         Routine care group:
                                                                                         93 (18)
                                                                                         Adjusted p=0.02

                                                                                         Any perineal trauma:
                                                                                         No significant difference
                                                                                         (adjusted p=0.42)

                                                                                         Length of postnatal stay
                                                                                         (days)
                                                                                         No significant difference
                                                                                         (adjusted p=0.80)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 227 of 441
                                                                                                                                             DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                  Follow-up &           Effect Size             Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                       Outcome                                                                Comments
                 Level                 cs                                                         Measures
Manderson, J.,      Study Type:       31 women with                       Intervention:        Follow-up period: Preeclampsia                Postprandial capillary
Patterson, C.,      RCT               type 1 diabetes                     Postprandial                            Preprandial: 6/38 (21%)    blood glucose monitoring
Hadden, D. et al.                     assigned to     Country: Northern   monitoring: Before   Outcome            Postprandial: 1/30 (3%),   during pregnancy in type
                    Evidence level:   preprandial     Ireland             breakfast and 1 hour Measures:          p=0.048                    1 diabetes may lead to
2003                1++               monitoring, 30                      after the            Success in                                    significant improvements
                                      assigned to                         commencement of      glycaemic control: Triceps skinfold           in obstetric and neonatal
{Manderson, 2003                      postprandial                        each meal.           the number of      thickness (mm) mean        outcomes.
28829 /id}                            monitoring.                                              readings within    (SD)
                                                                          Insulin doses were target ranges/total Preprandial: 5.1 (1.3)
                                                                          adjusted to achieve number of           Postprandial: 4.5 (0.9),
                                                                          fasting glucose      readings taken     p=0.05
                                                                          levels between3.3    Compliance with
                                                                          and 5.0 mmol/l,      monitoring         Respiratory problem
                                                                          preprandial values   schedule:          Preprandial: 10/31
                                                                          between 3.3 and 5.9 Number of           (32%)
                                                                          mmol/l and           readings           Postprandial: 5/30
                                                                          postprandial values taken/expected      (16%), p=0.07
                                                                          less than 7.8 mmol/l number of
                                                                                               readings           Birthweight >4000g
                                                                          The glucose                             Preprandial: 9/31 (29%)
                                                                          reflectance meters   Preeclampsia       Postprandial: 4/30
                                                                          were downloaded                         (3.3%), p=0.35
                                                                          every 2 weeks.       Neonatal
                                                                                               hypoglycaemia      Success in glycaemic
                                                                          Comparison:          (blood glucose     control (%)
                                                                          Preprandial          <1.7 mmol/l)       trimester 2:
                                                                          monitoring                              Preprandial: 29.4
                                                                                               Hyperbilirubinemi Postprandial: 51.6,
                                                                                               a                  p<0.001
                                                                                                                  trimester 3:
                                                                                                                  Preprandial: 30.3
                                                                                                                  Postprandial: 55.5,
                                                                                                                  p<0.001

                                                                                                                  Compliance with
                                                                                                                  schedule:
                                                                                                                  Trimester 2
                                                                                                                  Preprandial: 47.6
                                                                                                                  Postprandial: 39.7,
                                                                                                                  p=0.13
                                                                                                                  Trimester 3:
                                                                                                                  Preprandial: 30.2




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                               Page 228 of 441
                                                                                                                     DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size            Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                         Comments
                 Level                 cs                                   Measures
                                                                                         Postprandial: 35.7, p-
                                                                                         0.35

                                                                                         One unexplained
                                                                                         stillbirth occurred at 35
                                                                                         weeks in the preprandial
                                                                                         group.




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                  Page 229 of 441
                                                                                                                                                DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                                      Follow-up &           Effect Size            Study Summary              Reviewer
 Information & Evidence   Patients Characteristi Comparison                                           Outcome                                                               Comments
                 Level                 cs                                                             Measures
De Veciana, M.,     Study Type:       33 preprandial    85% Hispanic.        Intervention:          Follow-up period: Glycosylated              The results of this study
Major, C.,          RCT               monitoring        BMI 29.0 ±3.2        Postprandial                             haemoglobin (%)           support the hypothesis
Morgan, M. et al                      33 postprandial   (preprandial         monitoring: daily      Outcome           Initial                   that postprandial glucose
                    Evidence level:   monitoring        group) 28.4±3.8      monitoring before      Measures:         Preprandial: 8.6±2.3      monitoring can
1995                1++                                 (postprandial        breakfast (fasting)    Glycosylated      Postprandial:8.9±3.2      significantly improve the
                                                        group). Age 31±6     and one hour after     haemoglobin       Final                     outcomes of pregnancy
{de Veciana, 1995                                       years (preprandial   each meal.             Mode of delivery Preprandial: 8.1±2.2       in women with
28823 /id}                                              group) 29±5                                 Preterm labour    Postprandial: 6.5±1.4,    gestational diabetes who
                                                        years                Insulin dose was       Preeclampsia      p=0.006                   require insulin therapy.
                                                        (postprandial        adjusted to achieve    Birth weight      Change
                                                        group)               a fasting blood        Neonatal          Preprandial: -0.6±1.6
                                                                             glucose value of       hypoglycaemia     Postprandial: -3.0±2.2,
                                                        Country: USA         3.3-5.0 mmol/l,        (blood glucose    p<0.001
                                                                             preprandial values     <1.7 mmol/l).
                                                                             of 3.3-5.9 mmol/l or   Hyperbilirubinemi Caesarean sections
                                                                             postprandial values    a                 performed for
                                                                             <7.7mmol/l             Respiratory       cephalopelvic
                                                                                                    complications     disproportion (%)
                                                                             Comparison:                              Preprandial: 36
                                                                             Preprandial                              Postprandial: 12
                                                                             monitoring: Daily                        RR 3.0 (1.1-8.3) p=0.04
                                                                             monitoring of
                                                                             fasting, preprandial                     LGA
                                                                             and bedtime                              Preprandial:14 (42)
                                                                             capillary-blood                          Postprandial:4(12)
                                                                             glucose                                  RR 3.5 (1.3-9.5) p=0.01
                                                                             concentrations.
                                                                                                                      Birthweight >4000g
                                                                                                                      Preprandial: 12(36)
                                                                                                                      Postprandial: 3 (9)
                                                                                                                      RR 4.1 (1.3-13.2)
                                                                                                                      p=0.01

                                                                                                                      Shoulder dystocia
                                                                                                                      Preprandial:6(18)
                                                                                                                      Postprandial: 1(3)
                                                                                                                      RR 6.0 (0.8-47.1)
                                                                                                                      p=0.10

                                                                                                                      Neonatal hypoglycaemia
                                                                                                                      preprandial: 7(21)
                                                                                                                      postprandial: 1(3)




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                                                  Page 230 of 441
                                                                                                                   DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &            Follow-up &        Effect Size          Study Summary           Reviewer
 Information & Evidence   Patients Characteristi Comparison                 Outcome                                                       Comments
                 Level                 cs                                   Measures
                                                                                         RR7.0 (0.9-53.8) p=0.05

                                                                                         Stillbirth
                                                                                         preprandial: 1(3)
                                                                                         Postprandial: 0




Diabetes in pregnancy: full guideline, Appendix D DRAFT (September 2007)                                                Page 231 of 441
                                                                                                                          DRAFT FOR CONSULTATION




Bibliographic Study Type Number of   Patient     Intervention &                  Follow-up &            Effect Size        Study Summary               Reviewer
 Information & Evidence   Patients Characteristi Comparison                       Outcome                                                             Comments
                 Level                 cs                                         Measures
Kerssen, A., de     Study Type:       43   Pregnant women   Intervention:      Follow-up period:   Self blood glucose     Hyperglycaemia and      Does not look at
valk, H., Visser,   Cohort                 with type 1      Continuous glucose                     monitoring             hypoglycaemia detection pregnancy outcome
G.                                         diabetes         monitoring         Outcome             Detection of           rates increase when
                    Evidence level:                                            Measures: