Guidelines for Investigating Stillbirths:
An Update of a Systematic Review
Paula Corabian, BSc, MPH,1 N. Ann Scott, BSc, PhD,1 Carolyn Lane, MD, CCFP, FCFP,2
Grace Guyon, RN, BScN3
Institute of Health Economics, Edmonton AB
University of Calgary, Calgary AB
Alberta Perinatal Health Program, Edmonton AB
Objective: To identify formal, publicly available guidelines for Objectif : Identifier les directives cliniques officielles et accessibles
stillbirth investigation and to identify the most appropriate clinical au public portant sur la façon de soumettre la mortinaissance à
practice guideline (or component of a guideline) for use in Alberta. une enquête, et identifier la directive clinique (ou la partie d’une
Methods: A systematic literature search was conducted to identify directive clinique) convenant le mieux à la situation en Alberta.
primary and secondary research studies published between
January 1985 and August 2006 and formal, publicly available Méthodes : Une recherche documentaire systématique a été menée
guidelines on the subject of stillbirth investigation. The Cochrane afin de recenser les études primaires et secondaires, publiées
Library, PubMed, EMBASE, CINAHL, HealthSTAR, Science entre janvier 1985 et août 2006, ainsi que les directives cliniques
Citation Index, BIOSIS, and the NHS and CRD databases were officielles et accessibles au public portant sur la façon de
searched. The methodological quality of the selected primary soumettre la mortinaissance à une enquête. Les bases de
research studies was assessed according to specific criteria. données Cochrane Library, PubMed, EMBASE, CINAHL,
HealthSTAR, Science Citation Index, BIOSIS, NHS et CRD ont fait
Results: All six of the publicly available clinical practice guidelines l’objet de recherches. La qualité méthodologique des études
selected for this review outlined similar steps in the stillbirth primaires retenues a été évaluée en fonction de critères
investigation but differed about which tests to include and which particuliers.
components should be core or additional investigations. They
agreed on including several elements for routine investigation, Résultats : Les six directives cliniques accessibles au public
such as complete autopsy and detailed examination of the cord retenues aux fins de la présente analyse décrivaient des étapes
and placenta. d’enquête similaires en présence d’une mortinaissance, mais
Of 61 retrieved primary research studies, only seven met the présentaient des divergences quant aux tests à inclure et aux
inclusion criteria. No studies compared the value of specific composantes devant faire partie des enquêtes de base ou des
guidelines. Although reviewed evidence highlights the value of enquêtes supplémentaires. Elles s’entendaient toutefois sur
fetal autopsy and placental examinations as integral components l’inclusion de plusieurs éléments à l’enquête systématique, tels
of stillbirth investigation, the value of other components is still not qu’une autopsie exhaustive et un examen détaillé du cordon et du
Conclusions: No firm scientific judgement could be made about Seulement sept des 61 études primaires retenues ont satisfait aux
which clinical practice guideline for stillbirth investigation is the critères d’inclusion. Aucune étude n’a comparé la valeur de
most appropriate or which components are essential. Currently directives cliniques particulières. Bien que les résultats analysés
there is no generally accepted reference guideline for stillbirth soulignent la valeur de l’autopsie fœtale et des examens
investigation. Fetal autopsy and placental examination remain placentaires à titre de composantes intégrales de l’enquête
important components, assuming the postmortem examination is portant sur une mortinaissance, la valeur des autres composantes
of high quality. These data may be helpful in counselling parents n’est toujours pas manifeste.
who are considering whether or not to consent to a postmortem
examination. Conclusions : Aucun jugement scientifique ferme n’a pu être rendu
quant à la détermination de la directive clinique la plus appropriée
sur la façon de soumettre la mortinaissance à une enquête ou
quant à la détermination des composantes essentielles d’une telle
directive clinique. À l’heure actuelle, aucune directive clinique
n’est généralement acceptée à titre de référence sur la façon de
soumettre la mortinaissance à une enquête. L’autopsie fœtale et
Key Words: Autopsy, cause of death, fetal death, review l’examen placentaire demeurent des composantes importantes,
en présumant que l’examen post mortem est de grande qualité.
Competing Interests: None declared. Ces données peuvent s’avérer utiles pour conseiller les parents
qui se doivent de prendre une décision quant à la tenue ou non
Received on October 16, 2006 d’un examen post mortem.
Accepted on December 22, 2007
J Obstet Gynaecol Can 2007;29(7):560–567
560 l JULY JOGC JUILLET 2007
Guidelines for Investigating Stillbirths: An Update of a Systematic Review
n 2004, the rate of stillbirth in Alberta was 7.3 per 1000 All original studies published between January 1985 and
I total births.1 Stillbirths accounted for 65% of all perinatal August 2006 reporting on the use of a specific clinical prac-
tice guideline, test, or examination conducted to determine
mortality in that same year. Each stillbirth may involve
unique circumstances and problems, and it is always a dev- the cause of stillbirth were identified by searching the
Cochrane Library, PubMed, EMBASE, CINAHL,
astating experience for families and caregivers. The ques-
HealthSTAR, Science Citation Index, BIOSIS, the NHS
tion most ask in this situation is “Why did this happen?”
and CRD databases, and the websites of various health
Although risk factors such as low birth weight or multiple technology assessment agencies, clinical trial and research
pregnancies are associated with stillbirths,2 the cause registers, evidence-based medicine resources, and practice
remains unexplained in a significant proportion of cases. guideline sites. A systematic search of the literature pub-
lished between 1985 and June 2005 was conducted for the
By defining as clearly as possible the cause of death, infor- initial systematic review5 and an updating search was per-
mation obtained from investigating a stillbirth can help alle- formed in August 2006 to capture any recently published
viate concerns over prenatal events and may have implica- literature. Internet search engines and professional library
tions for future pregnancies. Clinical practice guidelines listservs were also used to locate “grey literature” (material
have been proposed to standardize the investigation in not published in the peer reviewed literature or indexed by
order to ensure that details are not missed and that there is major databases such as PubMed). The bibliographies of all
an orderly systematic search for the cause of death. publications retrieved in full were manually searched for rel-
evant references that may have been missed in the database
Recognizing that clinicians and families could benefit from searches.
knowing the cause of stillbirth, the Alberta Medical Associ-
ation (AMA) Committee on Reproductive Care imple- We also searched the literature to identify any publicly avail-
mented a formal clinical practice guideline for stillbirth able clinical practice guidelines that detailed a specific meth-
investigation in 1998.3 The Alberta guideline was designed odology for investigating the cause of stillbirth and that
to obtain the greatest amount of information without sig- were recommended by a professional society, health
nificantly increasing the budget and workload of laboratory authority, or government-funded organization in Canada or
and pathology services. An analysis of data collected over worldwide. Published guidelines used by individual hospi-
five years before and after the introduction of this guideline tals or recommended by privately funded special interest
showed no significant change between the two time periods groups or organizations were not included.
in the proportion of clinically unexplained stillbirths For the purpose of this review, a stillbirth was defined as the
(30% vs. 28%, P = 0.220) and in the number of clinically birth of a fetus at or after 20 weeks of pregnancy (or, if ges-
unexplained stillbirths undergoing autopsy (63% vs. 64%, tational age was not known, with fetal weight = 500 g) that
P = 0.825).4 However, in the five years following the intro- had died prior to delivery; that is, a fetus with no evidence of
duction of the guideline, more clinically unexplained still- respiration, heartbeat, umbilical cord pulse, or voluntary
births were diagnosed after autopsy (rising from 18% to muscle movement after expulsion from the uterus.2 The
23%, P = 0.071) and the proportion of placentas showing value of stillbirth investigation was assessed by selecting pri-
pathology increased from 80.8% to 88% (P < 0.001). mary and secondary research studies (systematic reviews
and cross-sectional analytic studies with at least 10 subjects)
The Alberta Perinatal Health Program, of which AMA is a that compared guidelines, tests, or examinations used to
partner, intends to update the protocol to ensure that determine the cause of stillbirth and reported on their diag-
women living in Alberta continue to receive perinatal care nostic contribution.
according to international best practices. To support that
For studies to be included in the review, we required that
intention, a systematic review of publications on the value
they (1) use a specific clinical practice guideline, test, or
of clinical practice guidelines for stillbirth investigation was
examination to determine the cause of stillbirth; (2) perform
undertaken.5 The aims of this review were (1) to collate and
a reference or comparison test in all of the study population
compare guidelines currently recommended by health
or sample undergoing the index test; and (3) report the pro-
authorities and obstetric and gynaecologic associations in
portion of stillbirths with an attributable cause for both the
Canada and other countries, and (2) to describe current
reference and index test.
published evidence on the use of guidelines for investigat-
ing the cause of stillbirth. This report summarizes the major Other reported outcomes considered were technical fail-
points from the systematic review. ures and the proportion of results that were inconclusive or
JULY JOGC JUILLET 2007 l 561
uninterpretable. Audits of the accuracy of prenatal tests, Each of these clinical practice guidelines recommends an
such as prenatal ultrasound diagnosis of fetal anomalies, extensive and comprehensive stillbirth investigation,
were excluded. Studies that pooled information on still- including a wide range of diagnostic modalities and tech-
births and perinatal deaths or stillbirths and elective preg- niques. Some test elements are recommended for routine
nancy terminations were excluded unless the stillbirth rate investigation (as core procedures, performed in all cases) by
could be separated from the aggregate data. all or most (at least four of six) guidelines, suggesting a list
Each of the included studies was assessed for the degree to of minimal core procedures (Table 1).
which the findings obtained from the index investigation The reviewed guidelines provide varying recommendations
(clinical practice guideline, test, or examination) confirmed about including different test elements in the stillbirth
the cause of death established by the reference standard, investigation.3,7–11 They also differ in terms of the compo-
changed the cause of death established by the reference nents recommended as core and/or additional investiga-
standard, provided additional information that did not tions of the mother, the stillborn, the cord, and the placenta.
change the established cause of death, or provided an expla-
The Alberta guideline3 compared favourably with the other
nation for an otherwise unexplained stillbirth.
five guidelines,7–11 and has served as a model for the guide-
The initial study selection, based on study titles and lines developed in British Columbia and Manitoba.7,8 How-
abstracts only, was conducted by one reviewer; copies of ever, modifications were necessary to accommodate local
the full text of potentially eligible studies were then differences that may exist in causes of stillbirth, population
retrieved and assessed independently by two reviewers. Any preferences and cultures, available technology and exper-
disagreements were resolved by discussion. Closer exami- tise, and financial resources.
nation of the full text articles revealed whether the retrieved
studies met the inclusion criteria. Reviewed Research Studies
The selected primary research studies were independently No systematic reviews assessing the value of specific clinical
screened and critically appraised by the same two reviewers, practice guidelines, tests, or examinations for stillbirth
using predetermined methodological quality criteria. The investigations were identified.
included trials were assessed with respect to various meth- Of 62 primary research studies identified as potentially
odological aspects, using a modified version of the meeting the inclusion criteria of the review, only seven met
QUADAS tool.6 Conciliation of disagreements was the criteria when the full text articles were more closely
achieved through discussion until consensus was reached. examined.12–18 The main features of these studies and
Study profile information and outcome data were extracted results of interest are summarized in Table 2. The selected
by one reviewer using standardized data extraction forms studies were non-randomized comparative studies that
developed a priori. used a cross-sectional analytic design to assess the value of
The full methodology for this review is detailed elsewhere.5 different types of investigations in terms of their diagnostic
contribution in the same stillborn groups. These studies
RESULTS compared diagnoses based on clinical data (alone or com-
bined with autopsy data) with autopsy diagnoses or placen-
No primary or secondary research studies comparing the
tal examination diagnoses (with knowledge of clinical data
value of specific clinical practice guidelines were identified.
alone or combined clinical and autopsy data) in the same
Publicly Available Clinical Practice Guidelines for study populations (Table 2). Two studies were prospec-
Stillbirth Investigation tive12,17 and five were retrospective.13–16,18 None of the
The literature search yielded six publicly available clinical studies compared formal or standard guidelines designed
practice guidelines developed specifically for stillbirth specifically for the comprehensive investigation of stillbirths.
investigation.3,7–11 Four of these guidelines were developed A meta-analysis was not performed because the studies
in Canada by the AMA,3 the British Columbia Reproductive were heterogeneous with respect to their study design, pop-
Care Program,7 the College of Physicians and Surgeons of ulation, and investigations (reference and index) and out-
Manitoba,8 and the Maternal Fetal Medicine Committee of come measures used. A detailed description of the refer-
the Society of Obstetricians and Gynaecologists of Canada ence and index tests that were used was not provided in all
(SOGC).9 The fifth selected guideline was developed in the studies. The studies also used different definitions for still-
United States by the Wisconsin Stillbirth Service Program,10 birth, and the method of determining intrauterine death was
and the sixth guideline was developed by the Perinatal Mor- never stated. Information about other factors that might
tality Special Interest Group of the Perinatal Society of influence the accuracy of the evaluated guidelines, tests, or
Australia and New Zealand.11 examinations (such as the stillborn infant’s state of
562 l JULY JOGC JUILLET 2007
Guidelines for Investigating Stillbirths: An Update of a Systematic Review
Table 1. Agreement on routine stillbirth investigation3,7–11
Steps Test elements recommended by the reviewed clinical practice guidelines
Maternal and family history All guidelines recommend as core procedures:
previous obstetric history;
history of current pregnancy;
review of antenatal investigations; and
maternal/paternal family history.
Most guidelines (4/6) agree on including the review of labour and delivery events as a core
Maternal investigations All guidelines recommend the Kleihauer-Betke screen as a core procedure.
Most guidelines (at least 4/6) recommend as core procedures:
complete blood count including platelet count;
toxoplasmosis serology; and
hemoglobin A1C test.
Stillborn examination All guidelines recommend as core procedures:
radiological examination; and
limited tissue or needle biopsy (for DNA analysis, cytogenetic studies and cultures).
Most guidelines (4/6) recommend internal examination as a core procedure.
Cord examination and cord blood (or All guidelines recommend as core procedures:
infant cardiac blood) investigation
gross/macroscopic/clinical examination; and
Most guidelines (4/6) recommend the following test elements as core procedures:
blood group; and
complete blood count.
Placenta examination and placental tissue All guidelines recommend as core procedures:
gross/macroscopic/clinical examination; and
JULY JOGC JUILLET 2007 l 563
preservation and the cause of death classification used) was This systematic review was undertaken in an effort to deter-
not provided in all reviewed studies. Information about the mine which parts of a clinical practice guideline for stillbirth
expertise of the investigator(s) was provided in three stud- investigation were of greatest value in providing additional,
ies,15,17,18 but it referred only to those who performed or confirmative, or diagnostic information with respect to the
supervised autopsies and placental examinations. cause of stillbirth. However, the reviewed evidence did not
allow a firm scientific judgement about the most appropri-
The information summarized from the reviewed studies
ate guideline for stillbirth investigation or which compo-
(Table 2) illustrates the difficulties in comparing their
nents should be considered for a relevant and efficient
results and using them to address the questions that initi-
investigative guideline. Of the six guidelines examined,3,7–11
ated this project. Several methodological problems and
all recommended that reviews of past obstetrical history,
inadequate data presentation limited interpretation and gen-
current pregnancy history, antenatal investigations, and
eralization of the results.
family history should be core elements of the stillbirth
According to the results reported by five retrospective investigation.
cross-sectional analytic studies,13–16,18 autopsy findings
Fetal autopsy and placental examination were considered
agreed with the clinical diagnosis in 28.6% to 89.8% of
essential in determining the cause of stillbirth and were rec-
cases and were diagnostic (revealed a new diagnosis or
ommended by all reviewed guidelines as integral compo-
made a change in the clinical diagnosis) in 10.2% to 38% of
nents of a relevant and efficient investigative guideline. 3,7–11
cases. Additional information (which did not change the
The reviewed research evidence supports these recommen-
clinical diagnosis) was obtained at autopsy in 3.9% to 24.3%
of cases. The cause of death remained unexplained in up to
40% of cases. Autopsy includes external examination, photography and
radiography of the fetus, examination of internal organs
Autopsy findings were more likely to be useful when no
with detailed macroscopic and histological examination,
clear clinical diagnosis was available or when there was a
and appropriate laboratory investigations including
fetal malformation,13–16,18 and they were more likely to con-
microbiologic, cytogenetic, and metabolic testing. Post-
tribute additional information when fresh stillbirth cases
mortem examination is of most value when there is a fetal
were available for examination.13
malformation or no obvious clinical cause of still-
Evidence reported from two prospective cross-sectional birth.13–16,18 A comprehensive autopsy performed as close
analytic studies12,17 suggested that routine macroscopic and as possible to the time of the fetal demise is most likely to
histologic examination of the placenta following a stillbirth yield clinically significant information.13,20,27,29,30
was a necessary complement to autopsy. Findings from pla-
The reviewed evidence demonstrates that placental pathol-
cental examination confirmed clinical or autopsy findings,
ogy, which includes gross and histological evaluation of the
or both, in up to 75% of cases and were diagnostic in 22.7%
cord and placenta, is also an essential component of any
to 46.3% of cases. The cause of death remained unex-
stillbirth investigation.12,17 Placental pathology appears to be
plained in up to 12% of cases.
readily accepted as an investigative option by the grieving
The probability of autopsy and placental examinations pro- parents of a stillborn infant.
viding clinically significant information was directly related On the basis of the findings of this systematic review, it is
to the quality of the postmortem report, as assessed by its not clear which of the other components should be
details and the additional investigations undertaken.13 included as part of a relevant and efficient stillbirth investi-
gation. The results from the studies that provide details
about their investigations could be evaluated only with
The cause of stillbirth is not always obvious at the time of respect to the investigation as performed in each examina-
delivery. Although multiple risk factors for stillbirth have tion setting. It was not possible to evaluate individual com-
been identified, they are not necessarily the direct cause. ponents of the assessed investigations. Some clinical prac-
Theoretically, determining the cause of a stillbirth could tice guidelines recommend the routine performance of
alleviate anxiety in a future pregnancy if a non-recurring ancillary investigations, such as maternal serology, testing
cause were found; it could assuage guilt in mothers if an for evidence of fetal-maternal transfusion, and
uncontrollable cause were determined, provide benefits to microbiologic and cytogenetic studies; others recommend
the extended family and opportunities for prenatal testing if they be performed only when clinically indicated. The indi-
a familial cause were evident, allow for modification of risk cation for the selection of each test was not provided, and
factors when identified, and amend the perinatal practices many of the examinations were clustered in the reviewed
of providers when appropriate.19–29 studies.12–18 Whether or not the performance of bacterial
564 l JULY JOGC JUILLET 2007
Guidelines for Investigating Stillbirths: An Update of a Systematic Review
Table 2. Summary of reviewed primary research studies
Study Study population Type of investigation Reported results
Agapitos et al. N = 164 [Ref]: Autopsy data + Clinical Cause of death explained in 42.1% of SBs
1996 SB definition: death occurring in data Cause of death unexplained in 57.9% of SBs
utero, > 22 wk of gestation [Index]: Placental examination Cause of death explained in 88.4% of SBs
+ [Ref] Cause of death unexplained in 11.6% of SBs
Findings were diagnostic in 46.3% of SBs*
Cartlidge et al.13 N = 104 (33 fresh; [Ref]: Clinical data —
1995 71 macerated)
SB definition: death occurring in
[Index[: Autopsy data + [Ref] Findings were diagnostic in 11.5% of all included SB
utero, ³ 20 wk of gestation (12.1% fresh; 11.3% macerated)
Additional information provided (but no change in
diagnosis) in 21.2% to 22% of all included SB (2.8%
macerated SB; 12.1% fresh SB)
Killeen et al.14 N = 130 SB (87 early SB; [Ref]: Clinical data —
2004 43 late SB)
SB definition: birth weight [Index]: Autopsy data + [Ref] Cause of death explained in 89% of early SB** and
> 500 g (regardless of 62% of late SB**
gestational age) Cause of death unexplained in 11% of early SB**
and 38% of late SB**
Findings were confirmative in 51% of early SBs**
and 33% of late SBs**
Findings were diagnostic in 38% of early SBs** and
29% of late SBs**
Kock et al.15 N = 177 [Ref]: Clinical data Cause of death explained in 53.7% of SB
2003 SB definition: intrauterine (as in death certificate) Cause of death unexplained in 46.3% of SB
death or death during delivery, [Index]: Autopsy data Cause of death explained in 60.5% of SB
³ 28 wk of gestation (as in autopsy report) Cause of death unexplained in 39.5% of SB
Findings were confirmative in 89.8% of SB
Findings were diagnostic in 10.2% of SB
Cause of death changed from unexplained to
explained in 6.8% of SB
Additional information provided in 24.3% of SB
Rasmussen et N = 325 [Ref]: Clinical data Cause of death explained in 60% of SB
al.16 2003 SB definition: antepartum fetal Cause of death unexplained in 40% of SB
death, ³ 28 wk of gestation or [Index]: Autopsy data + [Ref] Cause of death explained in 66% of SB
birth weight ³ 1000 g Cause of death unexplained in 34% of SB
Cause of death changed from unexplained to
explained in 6% of SB
Rayburn et al.17 N = 88 [Ref]: Clinical data and/or —
1985 SB definition: stillborn infants, Autopsy data
> 20 wk of gestation [Index]: Placental histologic Cause of death explained in 98% of SB
examination + [Ref] Cause of death unexplained in 1.1% of SB
(Clinical data) Findings were confirmative in 75% of SB
Findings were diagnostic in 22.5% of SB
Cause of death changed from unexplained to
explained in 11.2% of SB
Saller et al.18 N = 77 [Ref]: Clinical data —
1995 SB definition: fetal death, ³ 20
wk of gestation [Index]: Autopsy data + [Ref] Cause of death explained in 62.3% of SB
Findings were confirmative in 28.6% of SB
Findings were diagnostic in 29.9% of SB
Additional information provided in 3.9% of SB
Additional information: the index investigation revealed additional findings, which were not suspected or determined by the reference investigation, but did not
change the main diagnosis; confirmative: cause of death confirmed or agreement between the reference and index investigation in the case of unexplained death;
diagnostic: cause of death found, or cause of death changed, or definitive diagnosis made; g: grams; [Index]: index investigation (clinical practice guideline, test, or
examination); N: total number of stillbirths included in the study; [Ref]: reference investigation (clinical practice guideline, test, or examination); SB: stillbirth(s);
*Statistically significant contribution (P value not stated); ** P < 0.1
JULY JOGC JUILLET 2007 l 565
and viral cultures or cytogenetic studies would increase the The selected primary research studies were assessed using a
diagnostic contribution of autopsy is a matter of specula- quality tool, with the expectation that this would aid in
tion. At present, it seems the best approach is to be aware of identifying studies that should be given more weight in the
the additional tests available and to employ appropriate overall synthesis. However, none of the selected studies
investigations as clinically indicated. compared the same investigations, and all of them had
methodological weaknesses. Although the original aim of
Also, the usefulness of routine photographic documenta- the quality assessment became redundant because of these
tion or X-ray examinations could not be ascertained from factors, it still has value in highlighting the study design and
the reviewed data. Neither was it clear whether postmortem execution flaws.
ultrasound examination or magnetic resonance imaging
(MRI) of the stillborn infant could substitute for full CONCLUSIONS
autopsy and placental examination.
There is value in investigating unexplained stillbirths, but
In Alberta, the fetal autopsy rate increased from 54.0% in investigations need to be applied in a manner that results in
2000 to 74.5% in 2001 after the introduction of the Alberta a reasonable cost–benefit ratio. Currently there is no vali-
guideline,3 but it subsequently dropped to 48.8% in 2004.1 dated clinical practice guideline that can be applied in the
It is likely that the introduction of the Alberta guideline investigation of all unexplained stillbirths. The available
resulted in increased uptake of autopsy initially but declined guidelines should be used as a guide to ancillary investiga-
when the protocol was no longer a novel tool. The findings tions that might yield added value when applied selectively.
of this systematic review highlight the value of fetal autopsy According to the reviewed published evidence, all unex-
and placental examination as integral components of still- plained stillbirths warrant a good clinical history and physi-
birth investigation in terms of their diagnostic contributions cal examination, with subsequent autopsy and placental
to refining and defining the cause of death and providing pathology performed by a skilled pathologist in a timely
additional valuable information. Thus, it is essential that manner.
health care providers and parents are well informed about
It is the responsibility of the physicians, midwives, and
their value and encouraged to use them routinely in
nurses present at the delivery of a stillborn infant to explain
the value and potential benefit of postmortem investiga-
Clinical practice guidelines for stillbirth investigation need tions and to obtain informed consent from parents for
to evolve to take into account the possibility of including these procedures. Increasing the rate of autopsy and placen-
postmortem MRI of the stillborn infant, which has been tal examinations is likely to lead to a better understanding of
increasingly evaluated during the last decade as a non- the causes of stillbirth and an improvement in perinatal
invasive adjunct and potential substitute for full outcomes overall.
autopsy.31–39 Further investigation is needed to determine
whether this diagnostic technique can provide a minimally ACKNOWLEDGEMENTS
invasive approach for examination of the stillborn with We are grateful to Ms Liza Chan, Alberta Heritage Founda-
detection rates for anomalies and determination of the tion for Medical Research, who provided information ser-
cause of death similar to those of full autopsy. vices support for the preparation of the manuscript.
The present review has several limitations. The literature REFERENCES
review was confined to papers written in English and pub-
1. Alberta Perinatal Health Program. Provincial Perinatal Report 2000–2004.
lished after 1985. It was assumed that the inclusion of data Edmonton AB: Alberta Perinatal Health; 2006.
published before 1985 would not be valid because of the 2. Alberta Health and Wellness. Alberta Reproductive Health: Pregnancies and
very rapid evolution of diagnostic techniques, perinatal Births. Edmonton AB: Alberta Health and Wellness; 2002.
therapeutic procedures, and practice patterns in fetal- 3. Alberta Medical Association. Investigation of stillborn protocol. 2004.
Available at: http://www.albertadoctors.org/bcm/ama/ama-website.nsf/
maternal medicine, obstetrics, and neonatal care in the last e9f00d4a47929ecc87256b0a005c4e8b/f487f7a3ddd4473787256e2a005ed92
20 years. d!OpenDocument. Accessed April 3, 2007.
4. Alberta Perinatal Health Program. Perinatal mortality database 1994–2004.
The tests and examinations recommended for stillbirth Edmonton AB: Alberta Perinatal Health; 2006.
investigation by the reviewed clinical practice guidelines 5. Corabian P, Scott A. Protocols for stillbirth investigation. Edmonton AB:
have only been summarized. No attempt was made to Alberta Heritage Foundation for Medical Research; May 2005 (HTA36:
Series A). Available at: http://www.ihe.ca/documents/hta/
appraise their scientific foundations. The available versions HTA_Report_36.pdf. Accessed April 3, 2007.
of these guidelines did not describe the methodological 6. Whiting P, Rutjes AWS, Reitsma JB, Bossuyt PM, Kleijnen J. The
approaches used in their development. development of QUADAS: a tool for the quality assessment of studies of
566 l JULY JOGC JUILLET 2007
Guidelines for Investigating Stillbirths: An Update of a Systematic Review
diagnostic accuracy included in systematic reviews. BMC Med Res 23. Chambers HM. The perinatal autopsy: a contemporary approach. Pathology
Methodol 2003;3:25. 1992;24:45–55.
7. British Columbia Reproductive Care Program. Perinatal Mortality Guideline
24. Diagnosis and management of fetal death. ACOG Technical Bulletin
5. Investigation and assessment of stillbirths. Vancouver BC: British
Columbia Reproductive Care Program; 2005. Number 176-January 1993. Int J Gynaecol Obstet 1993;42:291–9.
8. College of Physicians and Surgeons of Manitoba. Pregnancy, perinatal and 25. Pauli RM, Reiser CA, Lebovitz RM, Kirkpatrick SJ. Wisconsin Stillbirth
neonatal care: investigation of stillbirths-guideline 1687. In: Guidelines and Service Program: I. Establishment and assessment of a community-based
statements, 1998. Accessed August 5, 2004 at: http://www.umanitoba.ca/ program for etiologic investigation of intrauterine deaths. Am J Med Genet
9. Leduc L. Stillbirth and bereavement: guidelines for stillbirth investigation.
SOGC Clinical Practice Guidelines, No. 178, May 2006. J Obstet Gynaecol 26. Pauli RM, Reiser CA. Wisconsin Stillbirth Service Program: II. Analysis of
Can 2006;28:540–5. diagnoses and diagnostic categories in the first 1,000 referrals. Am J Med
10. Wisconsin Stillbirth Service Program, University of Wisconsin. Guide to
etiologic evaluation of the stillborn infant: the WISSP Protocol. Available at: 27. Royal College of Obstetricians and Gynaecologists and Royal College of
http://www.wisc.edu/wissp/guidetoe.htm. Accessed April 3, 2007. Pathologists. Fetal and perinatal pathology. London, UK: RCOG
11. Perinatal Mortality Special Interest Group of the Perinatal Society of
Australia and New Zealand. Clinical practice guideline for perinatal 28. SOS Forum. SIDS focussing on stillbirth. Investigation and prevention of
mortality audit. The Perinatal Society of Australia and New Zealand, editor. stillbirth. Setting the policy and research agenda. November 29–30, 2001;
Brisbane, Australia: The Perinatal Society of Australia and New Zealand; Sydney, Australia.
29. Bove KE, Autopsy Committee of the College of American Pathologists.
12. Agapitos E, Papadopoulou C, Kavantzas N, Papoulias J, Antonaki V, Practice guidelines for autopsy pathology: the perinatal and pediatric
Davaris P. The contribution of pathological examination of the placenta in autopsy. Arch Pathol Lab Med 1997;121:368–76.
the investigation of the causes of foetal mortality. Arch Anat Cytol Pathol
1996;44:5–11. 30. Maternal and Child Health Research Consortium. CESDI 8th Annual
Report. Australia: CESDI; 2002. Project 27/28.
13. Cartlidge PH, Dawson AT, Stewart JH, Vujanic GM. Value and quality of
perinatal and infant postmortem examinations: cohort analysis of 400 31. Brookes JS, Hall-Craggs MA. Postmortem perinatal examination: the role of
consecutive deaths. BMJ 1995;310:155–8. magnetic resonance imaging. Ultrasound Obstet Gynecol 1997;9:145–7.
14. Killeen OG, Burke C, Devaney D, Clarke TA. The value of the perinatal 32. Wright C, Lee RE. Investigating perinatal death: a review of the options
and neonatal autopsy. Ir Med J 2004;97:241–4. when autopsy consent is refused. Arch Dis Child Fetal Neonatal Ed
15. Kock KF, Vestergaard V, Hardt-Madsen M, Garne E. Declining autopsy 2004;89:F285–F288.
rates in stillbirths and infant deaths: results from Funen County, Denmark, 33. Griffiths PD, Paley MN, Whitby EH. Post-mortem MRI as an adjunct to
1986–96. J Matern Fetal Neonatal Med 2003;13:403–7. fetal or neonatal autopsy. Lancet 2005;365:1271–3.
16. Rasmussen S, Albrechtsen S, Irgens LM, Dalaker K, Maartmann-Moe H,
34. Woodward PJ. Postmortem fetal magnetic resonance imaging. Contemp
Vlatkovic L, et al. Unexplained antepartum fetal death in Norway, 1985–97:
Rev Obstet Gynaecol 1998;10:195–9.
diagnostic validation and some epidemiologic aspects. Acta Obstet Gynecol
Scand 2003;82:109–15. 35. Brookes JA, Hall-Craggs MA, Sams VR, Lees WR. Non-invasive perinatal
17. Rayburn W, Sander C, Barr M, Jr, Rygiel R. The stillborn fetus: placental necropsy by magnetic resonance imaging. Lancet 1996;348:1139–41.
histologic examination in determining a cause. Obstet Gynecol 36. Alderliesten ME, Peringa J, van der Hulst V, Blaauwgeers HL, van Lith JM.
1985;65:637–41. Perinatal mortality: clinical value of postmortem magnetic resonance
18. Saller DN, Jr, Lesser KB, Harrel U, Rogers BB, Oyer CE. The clinical utility imaging compared with autopsy in routine obstetric practice. BJOG
of the perinatal autopsy. JAMA 1995;273:663–5. 2003;110:378–82.
19. Lindsey JL. Evaluation of fetal death. Web MD September 17, 2004. 37. Whitby EH, Paley MN, Cohen M, Griffiths PD. Postmortem MR imaging
Available at: http://www.emedicine.com/med/topic3235.htm. Accessed of the fetus: an adjunct or a replacement for conventional autopsy? Semin
April 3, 2007. Fetal Neonatal Med 2005;10:475–83.
20. Magee JF. Investigation of stillbirth. Pediatr Dev Pathol 2001;4:1–22. 38. Whitby EH, Paley MN, Cohen M, Griffiths PD. Post-mortem fetal MRI:
21. Carey JC. Diagnostic evaluation of the stillborn infant. Clin Obstet Gynecol what do we learn from it? Eur J Radiol 2006;57:250–5.
1987;30:342–51. 39. Hagmann CF, Robertson NJ, Sams VR, Brookes JA. Postmortem MRI as
22. Rushton DI. Prognostic role of the perinatal postmortem. Br J Hosp Med an adjunct to perinatal autopsy for renal tract abnormalities. Arch Dis Child
1994;52:450–4. Fetal Neonatal Ed 2006.
JULY JOGC JUILLET 2007 l 567