Recurrent Urinary Tract Infections

Document Sample
Recurrent Urinary Tract Infections Powered By Docstoc
					 Recurrent Urinary Tract Infections:
 Risk Factors and Effectiveness of
Prophylaxis in a Primary Care Cohort

  AcademyHealth Annual Meeting
         Patrick H. Conway, MD, MSc
              Avital Cnaan, PhD
         Theoklis Zaoutis, MD, MSCE
              Brandon Henry, BS
           Robert Grundmeier, MD
             Ron Keren, MD, MPH
             Epidemiology
 Urinary tract infection (UTI) is the most
  common serious bacterial infection in
  children
 Estimates of cumulative incidence in
  children 0 - 6 years suggest 70,000 to 180,000
  of the annual U.S. birth cohort will have a UTI
  by age six
 Little data on recurrent UTI rate but previous
  estimates of 20 - 48% within 6-12 months
 Traditional Conceptual Model

Figure 1 Conceptual Model
                            Prophylactic
                                                  End Stage Renal
                            antibiotics prevent
                                                      Disease
                            recurrent UTI

         UTI (s)

                             RENAL
                                                   Hypertension
                            SCARRING

          VUR
                            Surgery
                                                   Pre -eclampsia
                            corrects VUR
            Controversy
 Cochrane report summarized that
  evidence “to support widespread use
  of antibiotics to prevent recurrent UTI
  is weak”
 Two small clinical trials found
  prophylaxis had no significant effect on
  risk of recurrent UTI or renal scarring
               Specific Aims
1. To determine the factors associated with
   risk of recurrent UTI in a primary-care
   based cohort and to estimate the risk
   reduction provided by prophylactic
   antibiotics
2. To determine the risk factors for antibiotic
   resistance among recurrent UTIs
     Methods: Data Source
 Data obtained from primary care based
  network of practices who contribute to
  CHOP‟s Epic electronic health record
 27 practices from urban, suburban, and
  semi-rural areas in 3 states
 Data contains laboratory, prescription,
  and radiology data from clinic and
  emergency room settings
    Cohort Inclusion Criteria
 Identified children 0 - 6 years of age with at
  least 2 office visits between 7/1/2001 and
  5/31/2006
 From these infants, identified cohort with
  first UTI based on positive urine culture
  (>50,000 CFU/ml single organism)
 Followed infants until last documented
  contact with the network or until they
  experienced the primary outcome, a
  recurrent UTI
                        Aim 1
 Design: Cohort
 Outcome variable: Time to recurrent UTI
 Covariates:
      Age at first UTI
      Gender
      Race
      Degree of reflux
      Antibiotic prophylaxis
 Antibiotic prophylaxis was considered as a
  time varying covariate
 Analysis: Cox survival time regression
                        Aim 2
 Design: Nested case-control
 Outcome variable: Resistant versus pan-
  sensitive recurrent infections
 Covariates:
      Age at first UTI
      Gender
      Race
      Antibiotic prophylaxis exposure (yes/no)
      Degree of reflux
 Analysis: Multivariable logistic regression
      Results
   74,974 Children 0-6
   years of age with at
   least 2 clinic visits


  719 Children with any
  Urinary Tract Infection
                               First UTI
                               incidence rate:
628 Children with First UTI    0.007 per
                               person-year

611 Children with First
UTI and not Excluded          Recurrent UTI
                              incidence rate:
                              0.12 per person-
   83 Children with
                              year
   Recurrent UTI
        Observation time
 Mean observation time was 408 days
  with a median of 310 days (IQR 150 –
  584 days), range of 24 - 1600 days
                       First and Recurrent UTI
                              First UTI (Number, %)   Recurrent UTI (Number, %)
Total                                  611                    83 (13.6)
Gender
    - Male                           68 (11.1)                 8 (9.6)
    - Female                        543 (88.9)                75 (90.4)
Race
    - Caucasian                     343 (56.1)                54 (65.1)
    - Non-Caucasian                 268 (43.9)                29 (34.9)
Age
        - Less than 2 years         236 (38.6)                26 (31.3)
        - 2 - 6 years               375 (61.4)                57 (68.7)
VCUG
   - Not Performed                  400 (65.5)                52 (62.7)
   - Normal                         154 (25.2)                20 (24.1)
   - VUR Grade 1 - 3                 50 (8.2)                  8 (9.6)
   - VUR Grade 4 - 5                  7 (1.1)                  3 (3.6)
Exposure to antibiotic
prophylaxis
    - No                            483 (79.1)                64 (77.1)
    - Yes                           128 (20.9)                19 (22.9)
                       Risk of Recurrent UTI1
                                                       Univariable Hazard                      Multivariable Hazard
                                                       Ratio (95% CI)                          Ratio2 (95% CI)
Gender (Ref. Male)
  - Female                                                1.20 (0.58 – 2.50)                       1.08 (0.51 – 2.30)
Race (Ref. Non-Caucasian)
  - Caucasian                                            1.99 (1.26 – 3.16)3                       1.97 (1.22 – 3.16)3
Age (Ref. 0 - 1 year)
-1 – 2 years                                             0.99 (0.43 – 2.27)                        1.05 (1.20 – 3.37)
-2 – 3 years                                             1.22 (0.51 – 2.95)                        1.26 (0.51 – 3.07)
-3 – 4 years                                             2.55 (1.33 – 4.81)3                       2.75 (1.37 – 5.51)3
-4 – 5 years                                             2.17 (1.10 – 4.29)3                       2.47 (1.19 – 5.12)3
-5 – 6 years                                             1.36 (0.66 – 2.80)                        1.62 (0.73 – 3.62)
VCUG (Ref. Normal)
-Not Performed                                           1.00 (0.60 – 1.68)                        0.70 (0.40 – 1.21)
-VUR Grade 1-3                                           1.17 (0.52 – 2.66)                        1.05 (0.43 – 2.57)
-VUR Grade 4-5                                          4.59 (1.36 – 15.47)4                      4.38 (1.26 – 15.29)4
Antibiotic prophylaxis5                                   1.05 (0.57 – 1.94)                       1.016 (0.50 – 2.02)
1 Time-to-event performed from date of first UTI until event, recurrent UTI, or last clinic visit within the primary care network
2 Multivariable survival analysis controlling for gender, race, age, VCUG result, and prophylactic antibiotic exposure
3 p<0.01
4 p<0.05
5 Antibiotic prophylaxis exposure was modeled as time varying covariate in order to take into account total time exposed and

intermittent nature of exposure
Percent without Recurrence    Time To Recurrent UTI by Age


                             100

                              75

                              50

                              25
                                   0




                                       0        12                  24                  36
                                               Observation Time (months)

                                           Age < 2 Years                   Age ≥ 2 – 6 Years
      Antibiotic Prophylaxis
    Propensity Score Analysis
 Developed a propensity score for likelihood of
  receipt of prophylactic antibiotics
 Analyses stratified by propensity score
  quintile demonstrated no significant effect of
  antibiotic prophylaxis
 Antibiotic prophylaxis still did not decrease
  risk of recurrent UTI when controlling for:
  • Propensity quintile (HR 1.03, 0.51 – 2.08 )
  • Continuous propensity score (HR 1.02, 0.51 – 2.05 )
Risk of Antibiotic Resistance among Recurrent UTI Subjects
                                         Number of Resistant                 Odds Ratio of
                                         Infections (% of                    Recurrent UTI Being
                                         Recurrent UTI subjects)             Antibiotic Resistant1     95% CI
Gender
    - Male                                          7 (87.5)                         Ref
    - Female                                        44 (58.7)                        0.20            0.02 – 1.73
Race
    - Non-Caucasian                                 24 (82.8)                         Ref
    - Caucasian                                     27 (50.0)                        0.212           0.07 – 0.63
Age
      - Less than 2 years                           21 (80.8)                         Ref
      - 2 – 6 years                                 30 (52.6)                        0.263           0.09 – 0.80
VCUG
   - Normal                                         14 (70.0)                        Ref
   - Not Performed                                  27 (51.9)                        0.46             0.15 – 1.39
   - VUR Grade 1-3                                  7 (87.5)                         3.00            0.30 – 30.02
   - VUR Grade 4-5                                  3 (100.0)                         NA                  NA
Antibiotic prophylaxis
     - None                                         34 (53.1)                         Ref
     - Exposed to prophylaxis                       17 (89.5)                        7.502           1.60 – 35.17
 1 Odds ratio of resistant versus pan-sensitive organism as cause of recurrent UTI
 2 p ≤ 0.01
 3 p < 0.05
Probability of Recurrent UTI Being Antibiotic Resistant1

 Prophylactic                     Non-               Less than 2           VUR       Probability of
 Antibiotic exposure              Caucasian          Years of Age          Present   Resistance (%)2
                +                        +                    +                 +          98.0
                +                        +                    +                 -          94.2
                 -                       +                    +                 +          92.4
                +                        -                    +                 +          92.2
                 -                       +                    -                 +          89.6
                +                        -                    -                 +          89.3
                 -                       +                    +                 -          79.9
                +                        -                    +                 -          79.5
                 -                       -                    +                 +          74.5
                 -                       +                    -                 -          73.8
                +                        -                    -                 -          73.3
                 -                       -                    +                 -          48.9
                 -                       -                    -                 -          40.4
 1   For each exposure variable, a “+” represents that exposure being present
 2   Probability of causative organism being resistant to any antibiotic
                Summary
 Incidence rate for recurrent UTI of 12% per
  year is significantly lower than previous
  estimates
 Prophylactic antibiotics not associated with
  decreased risk of recurrent UTI but
  significantly associated with increased the
  risk of resistant infections
 Older 2-6 year old children, especially age 3-
  5, and Caucasian children had an increased
  risk of recurrent UTI
 VUR Grade 1-3 had no significant effect on
  recurrence risk
Limitations – Antibiotic exposure
 Antibiotic exposure was based on
  prescription data
 Likely overestimates the exposure in
  both subjects with and without
  recurrent UTI
 Potential confounding by indication
  and residual unobservable
  confounding
Limitations – Sparse or Missing Data

 Missing data due to VCUG not being
  performed
 Possibility of missing data from
  outside network
  • Attempted to minimize through chart
    review including correspondence
    from outside hospitals and clinics
               Strengths
 Based on primary care population
 Cohort design with large sample size that
  followed subjects for on average over 1 year
  in “natural experiment”
 Concurrently investigates potential risks
  and benefits of prophylactic antibiotics in
  same cohort
  Implications – Antibiotic Prophylaxis

 Given potential lack of prevention benefit
  and demonstrated harm due to resistant
  infections, this study in combination with
  other negative RCTs raises doubts about the
  effectiveness of prophylactic antibiotics
 Close monitoring without prophylaxis after
  first UTI may be a reasonable management
  strategy
Implications - VUR and Antibiotic Prophylaxis

 Subjects with Grade 1-3 VUR had no
  significant increased risk of recurrence and
  Grade 4-5 VUR had increased recurrence risk
 Antibiotic prophylaxis did not effect the risk
  of recurrence in either group in stratified or
  multivariable analysis
 Unclear if VUR, especially lower grade VUR,
  should be sole factor considered in
  prophylaxis recommendations
  Implications – Other Risk Factors

 Non-Caucasians had decreased risk of
  recurrence but increased risk of
  resistant infections
 Older children (age 2-6 years) had
  increased risk of recurrence; this may
  represent dysfunctional elimination
  syndromes
 Next Steps and Considerations
 RCT of antibiotic prophylaxis versus close
  monitoring
 Should UTI be considered as 2 hits
  necessary prior to long-term treatment?
  • Child with first UTI and no major urinary tract
    anomalies watched closely off treatment
 Future studies should validate whether older
  age and Caucasian race are risk factors for
  recurrence and explore mechanisms (e.g.
  dysfunctional elimination, genetic markers)
         Acknowledgments
 University of Pennsylvania CERTS grant
 Dr. Ron Keren
 Dr. Avital Cnaan
 Mr. Brandon Henry and Chris Bell, research
  assistants
 University of Pennsylvania Clinical Scholars
  Program
 Practice-Based Research Network at CHOP,
  its physicians, staff, and patients
            Males by Circumcision Status

                           First UTI (Number) Recurrent UTI
                                              (Number, %)1
Uncircumcised                         26             5 (19.2)

Circumcised                           10              0 (0)

Unknown                               32             3 (9.4)

Total                                 68             8 (11.8)




1 Differences   were not statistically significant
    Effect of Antibiotic Prophylaxis Stratified
                  by VUR Status1

                                                                         Hazard Ratio for Antibiotic
                                                                         Prophylaxis2 (95% CI)
VCUG
   - Normal                                                                     0.27 (0.04 – 2.02)
   - Not Performed                                                              1.44 (0.57 – 3.64)
   - VUR Present                                                                0.95 (0.29 – 3.13)




1 Time-to-event performed from date of first UTI until event, recurrent UTI, or last clinic visit within the primary care
network
2 Antibiotic prophylaxis exposure was modeled as time varying covariate in order to take into account total time exposed

and intermittent nature of exposure
           Risk of Recurrent UTI in Females1
                                                                    Multivariable Hazard
                                                                    Ratio2 (95% CI)
                Race (Ref. Non-Caucasian)
                  - Caucasian                                            2.12 (1.27 – 3.54)3
                Age (Ref. Less than 2 years)
                    - 2 - 6 years                                        1.94 (1.11 – 3.38)4
                VCUG (Ref. Normal)
                   - Not Performed                                       0.69 (0.39 – 1.22)
                   - VUR Grades 1 - 3                                    1.03 (0.39 – 2.66)
                   - VUR Grades 4 - 5                                     2.51 (.33 – 19.3)
                Antibiotic prophylaxis5                                  1.04 (0.49 – 2.18)



1 Time-to-event performed from date of first UTI until event, recurrent UTI, or last clinic visit within the primary care
network
2 Multivariable survival analysis controlling for gender, race, age, VCUG result, and prophylactic antibiotic exposure
3 p<0.01
4 p<0.05
5 Antibiotic prophylaxis exposure was modeled as time varying covariate in order to take into account total time exposed

and intermittent nature of exposure
               Risk of Recurrent UTI in Males1
                                                                    Multivariable Hazard
                                                                    Ratio2 (95% CI)
                Race (Ref. Non-Caucasian)
                  - Caucasian                                             0.59 (.10 – 3.74)
                Age (Ref. Less than 2 years)
                    - 2 - 6 years                                         2.30 (.48 – 11.1)
                VCUG (Ref. Normal)
                   - Not Performed                                       1.24 (0.19 – 8.21)
                   - VUR Grades 1 - 3                                    1.36 (0.48 – 11.1)
                   - VUR Grades 4 - 5                                    16.1 (1.91 – 136)
                Antibiotic prophylaxis5                                 1.73 (0.18 – 16.52)



1 Time-to-event performed from date of first UTI until event, recurrent UTI, or last clinic visit within the primary care
network
2 Multivariable survival analysis controlling for gender, race, age, VCUG result, and prophylactic antibiotic exposure
3 p<0.01
4 p<0.05
5 Antibiotic prophylaxis exposure was modeled as time varying covariate in order to take into account total time exposed

and intermittent nature of exposure
    Risk of Recurrent UTI by 1 Year Age Groups1
                                                                                Multivariable Hazard
                                                                                Ratio2 (95% CI)
              Gender (Ref. Male)
                - Female                                                              1.08 (0.51 – 1.96)
              Race (Ref. Non-Caucasian)
                - Caucasian                                                          1.97 (1.22 – 3.16)3
              Age (Ref. 0 – 1 year)
                  - 1 – 2 years                                                       1.05 (.45 – 2.47)
                  - 2 – 3 years                                                       1.26 (.51 – 3.07)
                  - 3 – 4 years                                                      2.75 (1.37 – 5.51)3
                  - 4 – 5 years                                                      2.46 (1.19 – 5.11)4
                  - 5 – 6 years                                                       1.62 (.73 – 3.62)
              VCUG (Ref. Normal)
                 - Not Performed                                                     0.68 (0.39 – 1.21)
                 - VUR Grades 1 - 3                                                  1.14 (0.47 – 2.82)
                 - VUR Grades 4 - 5                                                 4.38 (1.25 – 15.29)4
              Antibiotic prophylaxis5                                                 0.97 (0.48 – 1.96)


1 Time-to-event performed from date of first UTI until event, recurrent UTI, or last clinic visit within the primary care network
2 Multivariable survival analysis controlling for gender, race, age, VCUG result, and prophylactic antibiotic exposure
3 p<0.01
4 p<0.05
Other Recurrent UTI Studies
 Winberg studies published in „73 and „74 based
  on children 0-16 years who presented to
  Children‟s Hospital in Goteborg from 1960-66
 Proposed it was population based as “few other
  clinics” in the area
 After first UTI, children had urine tested at 13, 30,
  60, and 90 days after first UTI and then at 1, 3, and
  5 years after first UTI (not necessarily based on
  symptoms)
 Recurrence rate of 29% overall
 Decreasing “recurrence” rate with boys over time
  but no comment on circumcision status of males
   Other Recurrent UTI Studies
             (cont)
 Panaretto et al (J Paed Child Health 99)
  • 290 children 0-5 years diagnosed with UTI
    in ED, then had follow-up with 261 that
    consisted of phone call to parents at 6 and
    12 months after UTI
  • If parents reported UTI recurrence, then
    investigators attempted to confirm via
    culture
  • Found 13% recurrence rate
   Other Recurrent UTI Studies
             (cont)
 Garin et al 2006 demonstrated no significant
  different recurrence risk in prophylaxis
  group versus no prophylaxis group (17 vs
  23% overall)
 9% pyelonephritis in prophylaxis group
  versus 3% in no prophylaxis group
 Among children on prophylaxis, recurrence
  rate of 8.8% for subjects without VUR versus
  23.6% for those with VUR
       Cochrane Review
• Trials by Savage, Smellie, Stansfield in
  70‟s of prophylaxis versus placebo
• Often included children with multiple
  previous UTIs, no blinding, and testing of
  urine without symptoms
• Recurrence rate as high as 69% in control
  arm (savage)

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:273
posted:4/11/2008
language:English
pages:35