DOI: 10.1542/peds.2012-1870; originally published online December 17, 2012; 2013;131;e251Pediatrics

and Armando J. LorenzoLuis H. Braga, Hana Mijovic, Forough Farrokhyar, Julia Pemberton, Jorge DeMaria

HydronephrosisAntibiotic Prophylaxis for Urinary Tract Infections in Antenatal

  

  http://pediatrics.aappublications.org/content/131/1/e251.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

Antibiotic Prophylaxis for Urinary Tract Infections inAntenatal Hydronephrosis

abstractBACKGROUND AND OBJECTIVE: Continuous antibiotic prophylaxis(CAP) is recommended to prevent urinary tract infections (UTIs) innewborns with antenatal hydronephrosis (HN). However, there isa paucity of high-level evidence supporting this practice. The goalof this study was to conduct a systematic evaluation to determinethe value of CAP in reducing the rate of UTIs in this patient population.

METHODS: Pertinent articles and abstracts from 4 electronic data-bases and gray literature, spanning publication dates between 1990and 2010, were included. Eligibility criteria included studies of children,2 years old with antenatal HN, receiving either CAP or not, andreporting on development of UTIs, capturing information on voidingcystourethrogram (VCUG) result and HN grade. Full-text screening andquality appraisal were conducted by 2 independent reviewers.

RESULTS: Of 1681 citations, 21 were included in the final analysis (N =3876 infants). Of these, 76% were of moderate or low quality. PooledUTI rates in patients with low-grade HN were similar regardless ofCAP status: 2.2% on prophylaxis versus 2.8% not receiving prophylaxis.In children with high-grade HN, patients receiving CAP hada significantly lower UTI rate versus those not receiving CAP (14.6%[95% confidence interval: 9.3–22.0] vs 28.9% [95% confidence interval:24.6–33.6], P, .01). The estimated number needed to treat to prevent1 UTI in patients with high-grade HN was 7.

CONCLUSIONS: This systematic review suggests value in offering CAPto infants with high-grade HN, however the impact of important var-iables (eg, gender, reflux, circumcision status) could not be assessed.The overall level of evidence of available data is unfortunately moder-ate to low. Pediatrics 2013;131:e251–e261

AUTHORS: Luis H. Braga, MD, MSc, PhD,a Hana Mijovic,MD,a Forough Farrokhyar, MPhil, PhD,a Julia Pemberton,MSc,a Jorge DeMaria, MD,a and Armando J. Lorenzo, MD,MScb

aDepartment of Surgery, McMaster University, Hamilton, Ontario,Canada; and bDivision of Urology, Hospital for Sick Children andUniversity of Toronto, Toronto, Ontario, Canada

KEY WORDSantibiotic, hydronephrosis, prophylaxis or prevention, urinarytract infection

ABBREVIATIONSAPD—anteroposterior diameterCAP—continuous antibiotic prophylaxisCI—confidence intervalHN—hydronephrosisOR—odds ratioSFU—Society for Fetal UrologyUTI—urinary tract infectionVCUG—voiding cystourethrogramVUR—vesicoureteral reflux

Dr Braga conceptualized and designed the study, interpreted thedata, and drafted and revised the initial manuscript; Dr Mijoviccarried out initial analyses and drafted the initial manuscript;Dr Farrokhyar carried out final analyses and critically reviewedthe manuscript; Ms Pemberton designed the data collectioninstruments, coordinated and supervised data collection, andrevised the manuscript; Dr DeMaria conceptualized the studyand critically reviewed the manuscript; and Dr Lorenzointerpreted the data and critically reviewed the manuscript. Allauthors approved the final manuscript as submitted.

www.pediatrics.org/cgi/doi/10.1542/peds.2012-1870

doi:10.1542/peds.2012-1870

Accepted for publication Sep 17, 2012

Address correspondence to Luis H. Braga, MD, MSc, PhD,McMaster University, 1280 Main St West, HSC 4E19, Hamilton, ON,L8S 4K1 Canada. E-mail: braga@mcmaster.ca

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2013 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they haveno financial relationships relevant to this article to disclose.

FUNDING: No external funding.

PEDIATRICS Volume 131, Number 1, January 2013 e251

REVIEW ARTICLE

at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

Antenatal hydronephrosis (HN) is one ofthemostcommoncongenital anomalies,occurring in 1% to 5% of all pregnan-cies.1 Continuous antibiotic prophylaxis(CAP) has been empirically recom-mended for newborns who have ante-natal HN in an attempt to reduce therate of urinary tract infections (UTIs)during the first 2 years of life.2,3 Giventhe 2009 American Urological Associa-tion update and the 2010 Society forFetal Urology (SFU) consensus state-ment on HN, it seems prudent to con-sider use of CAP in “high-risk”populations, such as those with highergrades of HN.1,3,4 Nevertheless, thispractice is admittedly based on limiteddata and heavily reflects expert opinion.

Previous reports have suggested thatinfants who have moderate to severe HNare at a higher risk of developing UTIs,thus supporting the use of prophylacticmeasures.4–6 However, recent studieshave provided contradictory informa-tion, suggesting that infants with allgrades of HN can be safely managedwithout CAP, with UTI rates as low as4%.7–9 With increasing concerns aboutbacterial antibiotic resistance and un-known long-term effects, a growingnumber of physicians and parents arechallenging the necessity and effective-ness of CAP in preventing UTIs. Ulti-mately, the issue becomes a risk/benefitquestion, which heavily relies onwhether daily antibiotic administrationdecreases the risk of infections and ifthis risk reduction is enough to translateinto a clinically significant intervention.

To evaluate the current evidence on CAPeffectiveness and address the contempo-rary status of clinical equipoise, we con-ductedasystematicreviewofthepertinentliterature to assess the impact of CAP onUTI rates in infantswhohave antenatal HN.

METHODS

Eligibility Criteria

Studies fulfilling the following eligibilitycriteria were included: (1) primary

diagnosisofantenatalHN; (2)all subjectsaged ,2 years; (3) intervention armsinclude CAP, no treatment, or both; (4)reported rate of UTI; (5) reportednumber of patients who underwentvoiding cystourethrogram (VCUG); (6)HN grade according to the SFU classi-fication and/or anteroposterior di-ameter (APD) of the renal pelvis; and(7) publication date between 1990 and2010. Exclusion criteria included casereports, case series with,5 subjects,and review articles. No languagerestrictions were imposed.

Identification of Studies

A comprehensive and systematic searchof 4 databases (Medline, Embase, CINAHL,and CENTRAL) was conducted in dupli-cate. When possible, the search usedMeSH terms, and the strategy was de-vised with an information specialist atour institution. Gray literature wassearched by using key words in Paper-sFirst and ProceedingsFirst databases,along with a hand search of Dialogues inPediatric Urology. The reference list ofprominent review articles was cross-referenced to minimize omissions.Lastly, a content expert review of the finallist of included studies was conducted.

Screening and Assessing forEligibility

Two reviewers screened titles andabstracts of the identified studies fromthe electronic search to select allcitations that might contain the com-parisons of interest. Subsequently, 2independent reviewers conducted full-text screening of identified pertinentarticles as well as those obtained byhand-search review of reference lists.All screening resultswere reviewed forconcordance, and all disagreementswere resolved by a third reviewer andcontent expert. The following lan-guageswere included: English, French,Italian, Dutch, Portuguese, Spanish,German, Chinese, and Arabic. Studyselection was not blinded; conceal-ment has been shown to have no sig-nificant effect on the final results ofsystematic reviews.10

Assessment of MethodologicQuality

Critical appraisal and assignment ofa level of evidence for the includedstudieswasconductedby 2 independentreviewers byusing a quality assessmentinstrument adopted from Elyas et al,11

modified to fit the objectives of this

TABLE 1 Quality Appraisal of Included Studies

Variable Yes (%) No (%) Unclear (%) NA (%)

Nature of the research question (was question regarding CAP onantenatal HN, and/or possible impact on UTI prevention clearlystated?)

13 (62) 5 (24) 3 (14) 0 (0)

Standard quality itemsWas the study design reported? 18 (86) 0 (0) 3 (14) 0 (0)Was a follow-up schedule reported? 14 (67) 7 (33) 0 (0) 0 (0)Was the number of patients who were lost to follow-up

adequately described?10 (48) 11 (52) 0 (0) 0 (0)

Were the reasons for loss to follow-up adequatelydescribed?

4 (19) 15 (71) 0 (0) 2 (10)

Were statistical tests reported? 15 (71) 6 (29) 0 (0) 0 (0)Was P value reported for primary research study question? 11 (52) 9 (43) 1 (5) 0 (0)Were 95% CIs reported? 7 (33) 13 (62) 1 (5) 0 (0)Were eligibility criteria reported (inclusion and exclusion)? 20 (95) 0 (0) 1 (5) 0 (0)

Specific quality itemsWas antibiotic administration adequately described? 11 (52) 8 (38) 1 (5) 1 (5)Was there a comparative group (control)? 5 (24) 16 (76) 0 (0) 0 (0)Were the criteria for febrile UTI reported? 11 (52) 10 (48) 0 (0) 0 (0)Was the method of urine collection reported? 9 (43) 12 (57) 0 (0) 0 (0)Were any complications reported? 10 (47) 9 (43) 1 (5) 1 (5)

NA, not applicable.

e252 BRAGA et al at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

study (Table 1). All discrepancies wereresolved through third-party review ofthe methodology.

Study rating was determined by di-viding the number of criteria met (in-dicated as “Yes”) by the total number ofquality assurance criteria in the in-strument (n = 13), obtaining a per-centage result. For scoring purposes,we have assumed equal weight for allquestions. The data were then segre-gated into studies that met ,25% ofthe quality criteria (very low meth-odologic quality), between 25% and49% (low quality), between 50% and74% (moderate quality), and .75% ofthe criteria (high quality).

Extraction of Data

Data extraction was completed in du-plicate and reviewed for accuracy andquality.Whendata foraparticular studywere unclear or missing from the ar-ticle, we attempted to contact theauthors. Unfortunately, in cases ofmissing information, wewere unable toobtain any additional information.

Assessment of Agreement

The k (k) statistic was used to examinethe extent of agreement between indi-viduals who determined study eligibilityand to evaluate interobserver agree-ment in methodologic quality scores.A priori, we chose a criterion of k$ .65to indicate adequate agreement.12

Outcome Measures

The SFU system was used to gradepostnatally confirmed HN. For the pur-pose of this review, we considered SFUgrades I and II and/or correspondingtransverse renal pelvis APD rangingbetween 4.0 and 14.9 mm on postnatalultrasound as low-grade HN. Wheneverpossible, we excluded patients with noHN when these were reported withina category (ie, SFU 0–II). If not enoughinformation was provided to conductthis segregation, these children were

analyzed with the lowest grade of HN(SFU I). SFU grades III and IV and/ora transverse APD of the renal pelvis$15.0 mm on postnatal ultrasoundwere grouped into high-grade HN.

The primary outcome was developmentof UTI in infants with low-grade HNcompared with those with high-gradeHN, as previously defined and basedon the information extracted from theincluded studies. This was analyzed inthe context of prophylaxis administra-tion, with comparisons made betweeninfants receiving CAP versus those notreceiving CAP at the time of the UTI. Weaccepted that the protocol for adminis-tration, typeofantibiotic, anddurationofprophylaxiswould varybetweenstudies.Nevertheless, CAP does not refer toantibiotics provided for surgical pro-phylaxis or before completing imagingstudies (VCUG). For the purpose of thisreview, we focused on administration ofantibiotics beyond that period.

Secondary analyses chosen a prioriwere as follows: occurrence of a UTI in

patientswithconcomitantvesicoureteralreflux (VUR) versus no VUR, and oc-currence of a UTI in females comparedwith males. Although we initially plan-ned on assessing UTI rates in circum-cised versus uncircumcised boys, datafrom the available studies were in-sufficient to allow for this particularcomparison.

Statistical Analyses

We anticipated that the search resultswould yield case series (single groupretrospective or prospective studies)reporting the overall proportion of UTIand some comparative data reportingthe proportion of UTI for CAP status (yesversus no) or HN grade (high versuslow). Therefore, due to the inherentheterogeneity of the case series andobservational designs, we plannedapriori to use randomeffectsmodels toaccount for between-study heteroge-neity.

Odds ratios (OR) and 95% confidenceintervals (CIs) for the primary outcome

FIGURE 1Selection process for study inclusion (PRISMA flow diagram).

REVIEW ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 e253 at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

(UTI) were calculated whenever fre-quency data were available. We in-cluded in our primary meta-analysesany study that reported a rate of UTI,obtaining pooled estimates for infantswith low-andhigh-gradeHN,patientsonCAPversusnot, infantswithandwithoutVUR, and girls compared with boys. Weapplied the random effects model byusing the DerSimonian-Laird methoddue to the inherent heterogeneity ofcase series to estimate the pooledweighted UTI proportion of the includedstudies in these meta-analyses. Thepooled weighted proportion was cal-culated as the back-transformation ofthe weighted mean of the transformedproportions, using DerSimonian-Laird(1986) weights for the random effectsmodel. StatsDirect software version2.7.8 (www.statsdirect.com) was usedto calculate the pooled weighted pro-portions.

When deemed appropriate, we con-ducted a secondary meta-analysis ofany study that reported a directcomparison of UTI rates for infantswith low- and high-grade HN, patientson CAP versus not, and girls comparedwith boys. We performed a randomeffects model by using an inversevariance method to calculate the ORwith 95% CIs for the primary outcome(UTI) for CAP status (yes versus no), HNgrade (high versus low), and gender(girls versusboys). RevMan version5.1(Review Manager, Copenhagen: TheNordic Cochrane Centre, The CochraneCollaboration, 2011) was used forcalculating ORs. A P value of .05 wasset for statistical significance. Thenumber needed to treat was also es-timated to prevent the primary out-come (UTI) by receiving CAP.

Evaluation of Heterogeneity

Heterogeneity between studies wasquantified by using the I2 statistic,which represents the percentage oftotal variation across included studies TA

BLE2

Characteristicsof

Included

Studies

Author

Year

Country

StudyType

NGirls

Boys

Boys

Circ

SFUI–IIPatientsN(%

)SFUIII–IVPatientsN(%

)CAPN(%

)VURN(%

)UTI

N(%

)

Dacher

etal13

1992

USProspective

413

310

3NS

NSNS

47(11%

)13

(3.1%)

Blacharetal14

1994

Israel

Prospective

99NS

NSNS

101a

15a

NS14

(14.1%

)7(7.1%)

Misra

etal15

1999

UKProspective

4217

25NS

22a

18*

12(28.6%

)5(11.9%

)NS

Herndonetal16

1999

USRetrospective

7115

5637

NSNS

NS71

(100%)

18(25.4%

)Yerkes

etal17

1999

USProspective

6017

43NS

60(100%)

NS6(10%

)NS

Farhatetal18

2000

Canada

Retrospective

317

24NS

31a

9a31

(100%)

31(100%)

8(25.8%

)McIlroy

etal19

2000

NewZealand

Retrospective

6937

32NS

NSNS

NS69

(100%)

8(11.6%

)Brophy

etal20

2002

USRetrospective

234

60174

NS141(60.3%

)80

(34.2%

)97

(41.5%

)40

(17.1%

)10

(4.3%)

McLellanetal21

2002

USRetrospective

5418

36NS

NSNS

54(100%)

00

Shuklaetal22

2005

USRetrospective

408

32NS

NSNS

12(30.0%

)0

2Wollenbergetal23

2005

Switzerland

Retrospective

78NS

NSNS

42(53.8%

)36

(46.2%

)41

(52.6%

)9(11.5%

)15

(19.2%

)Mears

etal24

2007

UKProspective

5516

39NS

NSNS

NS8(14.5%

)NS

Song

etal4

2007

SouthKorea

Retrospective

105

2382

00

105(100%)

00

38(36.2%

)Coelho

etal5

2008

Brazil

Prospective

192

52140

0139(72.4%

)53

(27.6%

)119(62.0%

)16

(8.3%)

27(14.1%

)deKortetal25

2008

Netherlands

Retrospective

125

2699

NS106(84.8%

)19

(15.2%

)125(100%)

11(8.8%)

9(7.2%)

Leeetal6

2008

SouthKorea

Retrospective

430

79351

0255(59.3%

)175(40.7%

)0

083

(19.3%

)Lidefeltetal26

2008

Sweden

Retrospective

103

NSNS

NSNS

NS50

(48.5%

)9(8.7%)

9(8.7%)

Estradaetal28

2009

USRetrospective

1514

NSNS

NS1514

(100%)

322(21.3%

)322(21.3%

)21

(1.4%)

Mam

ietal29

2009

Italy

Prospective

223

NSNS

NS223(100%)

04(1.8%)

8(3.6%)

Roth

etal7

2009

USRetrospective

9220

72NS

092

(100%)

00

4(4.3%)

Yavascan

etal27

2010

Turkey

Prospective

246

63183

NSNS

NS246(100%)

32(13.0%

)NS

Circ,circumcised;NS,not

stated;UK,UnitedKingdom;US,UnitedStates.

aInrenalunits.Som

estudiesdidnotspecify

criteria.

e254 BRAGA et al at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

that is due to heterogeneity rather thanchance. Publication bias was graphi-cally assessed by constructing funnelplots depicting precision (measuredaccording to sample size) against thelogit of the event rate.

Secondary meta-analyses were selec-tively conducted for studies thatreported a direct comparison betweenUTI rates according to the grade of HN(low versus high) and CAP use (yesversus no) to reduce clinical hetero-geneity. Methodologic heterogeneitywas explored by conducting sensitivityanalyses comparing low- versus high-quality articles according to ourquality appraisal criteria.

RESULTS

Included Studies

Our literature search identified 1681potentially relevant citations. Afterscreening, a total of 309 articles un-derwent full-text review, and 21 of thesecitations4–7,13–29 proved eligible for in-clusion and final analysis (Fig 1). Theweighted k for overall agreement be-tween reviewers for the final eligibilitydecision was 0.85 (95% CI: 0.57–1.00).None of the included studies was a ran-domized clinical trial, and thus all wereobservational. Thirteen studies hada retrospective design4,6,7,16,18–23,25,26,28

and 8 were prospective.5,13–15,17,24,27,29

Patient Characteristics

Patient characteristics for the includedstudiesaresummarized in Tables2and3.The total number of infants in thesestudies was 3876. Given the incon-sistencies in available data provided byeach study, we conducted analyses in-cluding only complete relevant in-formation to obtain pooled estimatesthat would specifically address theobjectives of our review.

Study Quality

Five studies were judged to be of highmethodologic quality,5,6,19,23,29 8 to be of

moderate quality,4,7,15,20,25–28 5 studiesto be of low quality,14,17,18,21,22 and 3studies to be of very low quality13,16,24

(Table 4). Agreement between reviewersin assessment of study quality was high(k ..9).

Overall Rates of UTI

UTI rates in low-grade HN5,6,20,23,25,28,29

and high-grade HN4–7,20,23,25 werereported in 7 of 21 studies, for 2420 and560 infants, respectively. The pooled UTIrate was significantly lower for low-grade HN when compared with high-grade HN (4.7% [95% CI: 2.3–7.9] vs23.3% [95% CI: 12.0–37.0], P , .01)(Figs 2 A and B, respectively).

CAP and Rates of UTI

UTI rates for infants receiving CAP werereported in 95,18,20–23,25,26,28 of 21 studies,for a total of 851 patients; UTI rates forthose not receiving CAP were reportedin 84,6,7,20,23,25,28,29 of 21 studies, for a totalof 2370 infants. The pooled UTI rate forantenatal HN patients receiving CAPwassimilar to that of infants not receivingCAP (9.9% [95% CI: 4.6–17.1] vs 8.3%[95% CI: 2.9–16.0], P = .21) (Figs 3 A andB, respectively).

High-Grade Versus Low-Grade HNand Rates of UTI

A stratified analysis was conducted byusing individual patient data availablefrom eligible studies to determine UTIrates in infants with low-grade andhigh-grade HN according to their CAPstatus. The pooled UTI rates in patientswith low-grade HN were similar re-gardless of CAP status (2.2% [95% CI:1.3–4.0] for infants on CAP vs 2.8% [95%CI: 2.1–3.7] for those not on CAP, P = .52)(Table 5, n = 2181 patients6,20,23,25,28,29).In the high-grade HN group, infants onCAP had a significantly lower UTI rateversus those not on CAP (14.6% [95% CI:9.3–22.0] vs 28.9% [95% CI: 24.6–33.6],P, .01) (Table 6,n=507 infants4,6,7,20,23,25).The estimated number needed to

treat to prevent UTIs in infants withhigh-grade HN receiving CAP was 7,which represents a measure of ben-efit (ie, that prescribing CAP to 7infants with high-grade HN translatesinto preventing 1 UTI).

VUR and Rates of UTI

The pooled UTI rate for infants with VURwas 22.8% (95% CI: 9.7–39.4) comparedwith the pooled rate of 9.3% (95% CI:3.1–18.5) for patients without VUR (P =.31) (Figs 4 A and B). We could not ex-tract sufficient data to examine theassociation between VUR and UTIstratified according to HN grade or useof CAP.

Gender and Rates of UTI

A total of 8 studies4–7,16,18,20,22 reportedon UTI rates according to gender. UTIrates were similar in boys and girlswith antenatal HN (15% vs 18.9%; P =.21) (Fig 5). We could not extract

TABLE 3 Patient Characteristics AvailableFrom Included Studies

Characteristic Stated Not Stated

N % N %

GenderPublications 16 76 5 24Patients 1859 48 2017 52

CAP statusPublications 16 76 5 24Patients 3522 91 354 9

SFU gradePublications 13 62 8 38Patients 3225 83 651 17

UTIPublications 17 81 4 19Patients 3473 90 403 10

CircumcisionPublications 5 24 16 76Boys 639 46 759 54

TABLE 4 Overall Methodologic QualityRating of Included Studies

Methodologic Quality No. (%) of Studies

High (.75%) 5 (24)Moderate (50%–74%) 8 (38)Low (25%–49%) 5 (24)Very low (,25%%) 3 (14)

REVIEW ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 e255 at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

sufficient data to compare UTI ratesbetween girls and boys based on HNgrade or addressing circumcision sta-tus.

Heterogeneity and Publication Bias

To reduce heterogeneity, a randomeffects model was followed when con-ducting the current meta-analysis,limited to studies reporting a directcomparison between low- versus high-grade HN (Fig 6). The pooled OR was 5.8(95% CI: 3.9–8.7), suggesting that high-grade HN was significantly associatedwith a higher rate of UTI. To adjust forpotential confounding effects of VUR

and CAP on UTI, we have chosen todisplay the percentage of VUR and CAPpatients in Fig 6. To further validatethese findings, a sensitivity analysiswas conducted, comparing moderateversus high methodologic quality arti-cles. The pooled OR for high-qualitystudies (6.3 [95% CI: 3.9–9.5]) alsosuggests that high-grade HN was sig-nificantly associated with higher UTIrates (Fig 7).

When we conducted a separate meta-analysis including only studies thatcompared CAPuse versus no use of CAP,we obtained a pooled OR of 1.7 (95% CI:1.1–2.8), indicating that patients re-

ceiving CAP had significantly lower UTIrates (Fig 8).

A funnel plot was created, involv-ing 17 studies that reported UTIrates.5–7,13,14,16,18–23,25,26,28,29 Thedistribution of studies is scatteredover the 4 quadrants of the graph,suggesting a lack of publication bias(Fig 9).

DISCUSSION

With thewidespread practice of routineprenatal ultrasound, the number offetuses with detected HN has increasedcommensurately. Management of themany patients who have persistent HN

FIGURE 2(A) UTI rates in infants with low-grade HN. Pooled UTI rate is 4.7% (95% CI: 2.3–7.9), I2 (inconsistency) = 85.4% (95% CI: 69.7–91.2). (B) UTI rates in infants withhigh-grade HN. Pooled UTI rate is 23.3% (95% CI: 12.0–3.7), I2 (inconsistency) = 91.6% (95% CI: 85.7–94.4).

FIGURE 3(A) UTI rates in infants with antenatal HN receiving CAP. Pooled UTI rate is 9.9% (95% CI: 4.6–17.1), I2 (inconsistency) = 87.8% (95% CI: 78.8–91.9). (B) UTI rates ininfants with antenatal HN not receiving CAP. Pooled UTI rate is 8.3% (95% CI: 2.9–16.0), I2 (inconsistency) = 96.2% (95% CI: 95–97).

e256 BRAGA et al at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

after birth remains controversial anddevoid of guidance based on high levelsof evidence. Among contentious topics,the routine prescription of CAP hasgained attention due to the perceivedlack of clinically significant benefit inVUR trials, as well as concerns forbacterial resistance and long-termadverse effects. Highlighting theseissues, the 2009 Canadian UrologicalAssociation guidelines on antenatal HNstated that the role of CAP is indeedcontroversial and provided a grade Drecommendation (based on case seriesand clinical experts’ opinion) for thisspecific topic.30 More recently, the 2009American Urological Association Up-date Series on Prenatal Diagnosis ofUrological Disease stated that althoughCAP is generally recommended for se-vere (grades III and IV) HN, the practiceis not evidence based.1 Not surpris-ingly, the lack of consensus has trans-lated into noticeable differencesbetween individual health care pro-

viders and specialties.1,31–33 Ismailiet al31 conducted a study amongFrench-speaking pediatric nephrolo-gists and urologists and found a sig-nificant variation in CAP prescribingpatterns for antenatal HN within andacross the 2 groups. Similarly, a surveyof pediatric urologists from Europeand the United States concluded thatthere is also a considerable variationin CAP use for antenatal HN betweenthe 2 continents, as well as amongpediatric urologists from the samecountry.32

In general, physicians prescribe CAP toreduce the risk of UTI in infants whohave antenatal HN and to prevent UTI-associated morbidity as well as long-term complications of renal damage.34

Nevertheless, the ultimate decision tostart an infant on CAP currently cen-ters on parental acceptance aligningwith individual health care providerand/or institution protocols. Due tothe paucity of published data, treat-ment philosophy plays an importantrole. Thus, the decision is based onpartially defined risks and benefits,heavily affected by the perception ofwhat is more valuable: potentiallypreventing pyelonephritis and sepsisor accepting the administration ofa drug that may provide little benefitbut can be associated with adverseeffects and bacterial resistance. Con-ceptually, it would make sense to tai-lor the prescription of CAP based onspecific patient characteristics thataffect the odds of experiencing theoutcome of interest. However, con-sensus on what constitutes a riskfactor for UTI warranting CAP in thispopulation has yet to be determined.To date, there is limited informationon the topic.

All of the reviewed publications haverecognized important shortcomings interms of study design, number ofpatients evaluated, inclusion and ex-clusioncriteria,definitionofevents,and

monitoring and assessment of com-pliance. Thus, the information gatheredfromeachof themonan individualbasisis difficult to bring into clinical practice.In addition, the findings from relevantseries are at times inconsistent. Coelhoet al5 found that girls with VUR or uri-nary tract obstruction had a higherrisk of UTI during follow-up and rec-ommended CAP and close clinicalmonitoring. Estrada et al28 observedthat in patients with postnatal persis-tent grade II HN, identification of VURthrough a VCUG and use of CAP signif-icantly reduced the risk of febrile UTI.Conversely, some authors have arguedthat CAP does not change the risk of UTIin specific patient populations. Rothet al7 conducted a retrospective studyof patients with SFU grade III/IV HN and,based on an overall low baseline rate ofUTI (4%), recommended against wide-spread use of CAP for all infants withantenatal HN. Similarly, in a recentprospective study, Islek et al9 con-cluded that CAP use is not indicated ininfants with ureteropelvic junction ob-struction, regardless of its severity, asthe risk of UTI was minimal in thisgroup.

Given thesediscrepancies, ourgoalwasto conduct a systematic review, aimingto summarize the current evidenceregarding theuseof CAP inantenatalHNpatients. In addition, this exercisesought to further analyze direction andmagnitude of potential treatmenteffects while highlighting deficienciesin knowledge that shouldbe consideredfor future studies. To our knowledge, nosimilar reviews have been published onthis particular topic thus far. Our stricteligibility criteria for included studieshave allowed us to gather a ratherhomogeneous patient population,summarizing information on a rela-tively large number of infants identifiedantenatally and followed up for a mini-mum of 12 months after birth.

TABLE 5 Rates of UTI in Infants With SFUGrades I and II HN

CAP Status UTI No UTI Total

N % N %

Yes 10 2.2a 449 97.8 459No 49 2.8b 1673 97.2 1722Total 59 2.7 2122 97.3 2181

Data were stratified according to CAP status based onavailable data from the following studies included in theanalysis: 6, 20, 23, 25, 28, and 29 (VUR was an exclusioncriterion in studies 6 and 23). P = .52.a 95% CI for CAP use: 1.3 to 4.0.b 95% CI for no CAP use: 2.1 to 3.7; 348 (16%) of 2181patients had VUR.

TABLE 6 Rates of UTI in Infants With SFUGrades III and IV HN

CAP Status UTI No UTI Total

N % N %

Yes 18 14.6a 105 85.4 123No 111 28.9b 273 71.1 384Total 129 25.4 378 74.6 507

Data were stratified according to CAP status based onavailable data from the following studies included in theanalysis: 4, 6, 7, 20, 23, and 25 (VUR was an exclusioncriterion in studies: 4, 6, 7, and 23). P # .01.a 95% CI for CAP use: 9.3 to 22.b 95% CI for no CAP use: 24.6 to 33.6; 28 (5.5%) of 507patients had VUR.

REVIEW ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 e257 at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

Aggregate evaluation by using meta-analysis demonstrated that UTI ratesare significantly higher in infants withhigh-grade HN even when restricted toselected articles that reported a directcomparison of UTI rates betweenpatientswith low- versus high-gradeHNwithin the same study. CAP in the high-

grade HN group was associated withsignificant reduction of UTI rates whencompared with no treatment (14.6% vs28.9%; P, .01). In contrast, UTI rates ininfants with low-grade HN were similarregardless of CAP use. This core findingsupports the notion of selective useof CAP for patients with high-grade

HN. Nevertheless, it should also benoted that due to limitations of thereported data in the studies included,the effect of important variables (eg,VUR, gender) could not be properlyevaluated.

Aswithmostmeta-analyses, thecurrentreport has important limitations. The

FIGURE 4(A) UTI rates in infants with VUR. Pooled UTI rate is 22.8% (95% CI: 9.7–39.4), I2 (inconsistency) = 92.6% (95% CI: 88.4–94.8). (B) UTI rates in infants without VUR.Pooled UTI rate is 9.3% (95% CI: 3.1–18.5), I2 (inconsistency) = 96.6% (95% CI: 95.4–97.3).

FIGURE 5UTI rates in infants with antenatal HN grouped according to gender. IV, inverse variance.

FIGURE 6Meta-analysis on UTI rate in high-grade versus low-grade HN. AHN, antenatal HN. M-H, Mantel-Haenszel.

e258 BRAGA et al at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

most notable is the unavoidable re-striction to observational studies be-cause we could not identify anyrandomized controlled trials examin-ing the relationship between CAP andUTIs in infants with antenatal HN. Thisfinding reflects the state of currentevidence on the topic of CAP and an-tenatal HN. Methodologic and analyti-cal issues related to observationalstudies of different quality have to becarefully considered and kept presentwhen aggregating data and placingthese outputs in perspective. Even if allthe included nonexperimental studieswere of high methodologic quality, theresultant level of evidence would bemoderate. In addition, it is possiblethat our estimates of UTI rates werebiased due to restricted analysis of the

6 studies with data on direct com-parison of UTI rates between infantswith low- and high-grade HN. Never-theless, the followed strategy allowedfor optimizing statistical power (byvirtue of a larger sample size) withina more homogeneous patient pop-ulation, ultimately combining resultsand providing a common measure ofeffect size. Moreover, the format inwhich data were presented variedwidely between studies. We elected togroup infants based on SFU grade to beable to make meaningful comparisonsbetween the studies. However, in-consistent grading of HN (a situationthat was more notable in older studies)made it difficult to compare all patientpopulations. As previously indicated,some studies were excluded from the

final analysis because the providedmeasurements could not be translatedinto SFU grading. Similarly, althoughmost articles reported on occurrenceof antenatal HN in individual patients,some articles reported their results inrenal units without correlating thesedata to the number of patients affected(as shown in Table 2).

Due to the paucity of data, we wereunable to comment on the associationbetween antenatal HN concurrent withVUR and UTI, and gender effect on UTI,adjusted according to HN grade.Associations between the primaryoutcome and isolated HN (ureter-opelvic junction obstruction–like) orhydroureteronephrosis (primary mega-ureter) were often difficult to analyzebecause the included articles had not

FIGURE 7Sensitivity analysis comparing high-quality versus moderate-quality studies. AHN, antenatal HN.

FIGURE 8Meta-analysis on UTI rate in CAP use versus no CAP use.

REVIEW ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 e259 at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

consistently provided UTI rates dis-criminated according to etiology. In anattempt to assess UTI rates solely inpatients with hydroureteronephrosiswithout VUR (primary megaureter),we obtained data from 4 pertinentstudies4–7 as displayed in Fig 10. The

pooled UTI rate was 34% (95% CI: 14–58), significantly higher than the rate ofall patients with high-grade HN (23.3%[95% CI: 3.7–12]), suggesting that thesepatients might be at an even higher riskof developing UTI and could benefit fromadministration of CAP.

Finally, in the sameway,wewere unabletodrawconclusionsabout the impact ofcircumcision on UTI rates; UTI data oncircumcised males were only availablefor 10 boys in 2 studies.

Despite those limitations, we believethere is value in the current report.We present compelling evidence tosuggest that infants who have high-grade HN who are not receiving CAPare at a significantly higher risk ofUTI when compared with infantswho have high-grade HN who are re-ceiving prophylaxis. This informationis a first step toward tailoring ourapproach for patients with antenatalHN, which could be selectively fa-vored in cases of high-grade HN. Thissystematic review corroborates thepresence of clinical equipoise re-garding the postnatal management ofinfants with antenatal HN due to thelack of sound evidence supportingor disproving the effectiveness of CAPto prevent UTIs in this patient pop-ulation.

CONCLUSIONS

This systematic review implies thatoffering CAP to 7 infants who havehigh-grade HN would prevent 1 UTI,suggesting value in this subgroupof patients. Given the existing data,the effect of VUR, gender, and cir-cumcision on UTI could not be de-termined. Current lack of bestpractice guidelines in the pediatricurology community, coupled with thecontroversial literature surroundingthis topic, calls for prospective, ran-domized placebo-controlled trials toinvestigate the effect of CAP in re-ducing UTI rates in infants with an-tenatal HN.

ACKNOWLEDGMENTSWe thank Ms Veronica Chan and DrSoojin Kim for their contributionswiththe search strategy and abstractscreening.

FIGURE 9Funnel plot portraying precision of included studies4–7,13,14,16,18–23,25,26,28,29 with regard to the UTIrates.

FIGURE 10UTIRates in infantswithhydroureterandnoVUR.PooledUTI rate is34%(95%CI: 14–58), I2 (inconsistency)= 87% (95% CI: 61.7–93.2).

e260 BRAGA et al at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

REFERENCES

1. Nguyen HT, Herndon CD, Cooper C, et al. TheSociety for Fetal Urology consensus state-ment on the evaluation and managementof antenatal hydronephrosis. J PediatrUrol. 2010;6(3):212–231

2. Woodward M, Frank D. Postnatal manage-ment of antenatal hydronephrosis. BJU Int.2002;89(2):149–156

3. Herndon ACD, Cain MP. Prenatal Diagnosisof Urological Disease. Vol. 28. Linthicum,MD: American Urological Association, Edu-cation and Research Inc; 2009:21–32

4. Song SH, Lee SB, Park YS, Kim KS. Is anti-biotic prophylaxis necessary in infants withobstructive hydronephrosis? J Urol. 2007;177(3):1098–1101, discussion 1101

5. Coelho GM, Bouzada MC, Lemos GS, PereiraAK, Lima BP, Oliveira EA. Risk factors forurinary tract infection in children withprenatal renal pelvic dilatation. J Urol.2008;179(1):284–289

6. Lee JH, Choi HS, Kim JK, et al. Nonrefluxingneonatal hydronephrosis and the risk ofurinary tract infection. J Urol. 2008;179(4):1524–1528

7. Roth CC, Hubanks JM, Bright BC, et al. Oc-currence of urinary tract infection in chil-dren with significant upper urinary tractobstruction. Urology. 2009;73(1):74–78

8. Greenfield SP. Antibiotic prophylaxis in pe-diatric urology: an update. Curr Urol Rep.2011;12(2):126–131

9. Islek A, Güven AG, Koyun M, Akman S,Alimoglu E. Probability of urinary tractinfection in infants with ureteropelvicjunction obstruction: is antibacterial pro-phylaxis really needed? Pediatr Nephrol.2011;26(10):1837–1841

10. Berlin JA. Does blinding of readers affectthe results of meta-analyses? University ofPennsylvania Meta-analysis Blinding StudyGroup. Lancet. 1997;350(9072):185–186

11. Elyas R, Guerra LA, Pike J, et al. Is stagingbeneficial for Fowler-Stephens orchiopexy?A systematic review. J Urol. 2010;183(5):2012–2018

12. Cohen J. Weighted kappa: nominal scaleagreement with provision for scaled dis-agreement or partial credit. Psychol Bull.1968;70(4):213–220

13. Dacher JN, Mandell J, Lebowitz RL. Urinarytract infection in infants in spite of prenataldiagnosis of hydronephrosis. Pediatr Radiol.1992;22(6):401–404, discussion 404–405

14. Blachar A, Blachar Y, Livne PM, Zurkowski L,Pelet D, Mogilner B. Clinical outcome andfollow-up of prenatal hydronephrosis.Pediatr Nephrol. 1994;8(1):30–35

15. Misra D, Kempley ST, Hird MF. Are patientswith antenatally diagnosed hydronephrosisbeing over-investigated and overtreated?Eur J Pediatr Surg. 1999;9(5):303–306

16. Herndon CD, McKenna PH, Kolon TF, GonzalesET, Baker LA, Docimo SG. A multicenter out-comes analysis of patients with neonatalreflux presenting with prenatal hydro-nephrosis. J Urol. 1999;162(3 pt 2):1203–1208

17. Yerkes EB, Adams MC, Pope JC 4th, Brock JW3rd. Does every patient with prenatalhydronephrosis need voiding cystourethrog-raphy? J Urol. 1999;162(3 pt 2):1218–1220

18. Farhat W, McLorie G, Geary D, et al. Thenatural history of neonatal vesicoureteralreflux associated with antenatal hydro-nephrosis. J Urol. 2000;164(3 pt 2):1057–1060

19. McIlroy PJ, Abbott GD, Anderson NG,Turner JG, Mogridge N, Wells JE. Outcomeof primary vesicoureteric reflux detectedfollowing fetal renal pelvic dilatation. JPaediatr Child Health. 2000;36(6):569–573

20. Brophy MM, Austin PF, Yan Y, Coplen DE.Vesicoureteral reflux and clinical outcomesin infants with prenatally detected hydro-nephrosis. J Urol. 2002;168(4 pt 2):1716–1719; discussion 1719

21. McLellan DL, Retik AB, Bauer SB, et al. Rateand predictors of spontaneous resolutionof prenatally diagnosed primary non-refluxing megaureter. J Urol. 2002;168(5):2177–2180, discussion 2180

22. Shukla AR, Cooper J, Patel RP, et al. Pre-natally detected primary megaureter:a role for extended followup. J Urol. 2005;173(4):1353–1356

23. Wollenberg A, Neuhaus TJ, Willi UV, Wisser J.Outcome of fetal renal pelvic dilatation di-agnosed during the third trimester. Ultra-sound Obstet Gynecol. 2005;25(5):483–488

24. Mears AL, Raza SA, Sinha AK, Misra D.Micturating cystourethrograms are notnecessary for all cases of antenatally di-agnosed hydronephrosis. J Pediatr Urol.2007;3(4):264–267

25. de Kort EH, Bambang Oetomo S, Zegers SH.The long-term outcome of antenatalhydronephrosis up to 15 millimetres jus-

tifies a noninvasive postnatal follow-up.Acta Paediatr. 2008;97(6):708–713

26. Lidefelt KJ, Herthelius M. Antenatal hydro-nephrosis: infants with minor postnataldilatation do not need prophylaxis. PediatrNephrol. 2008;23(11):2021–2024

27. Yavascan O, Aksu N, Anil M, et al. Postnatalassessment of growth, nutrition, and uri-nary tract infections of infants with ante-natally detected hydronephrosis. Int UrolNephrol. 2010;42(3):781–788

28. Estrada CR, Peters CA, Retik AB, Nguyen HT.Vesicoureteral reflux and urinary tract in-fection in children with a history of pre-natal hydronephrosis—should voidingcystourethrography be performed in casesof postnatally persistent grade II hydro-nephrosis? J Urol. 2009;181(2):801–806,discussion 806–807

29. Mamì C, Paolata A, Palmara A, et al. Out-come and management of isolated mod-erate renal pelvis dilatation detected atpostnatal screening. Pediatr Nephrol. 2009;24(10):2005–2008

30. Psooy K, Pike J. Investigation and man-agement of antenatally detected hydro-nephrosis. Can Urol Assoc J. 2009;3(1):69–72

31. Ismaili K, Avni FE, Piepsz A, et al. Currentmanagement of infants with fetal renalpelvis dilation: a survey by French-speakingpediatric nephrologists and urologists.Pediatr Nephrol. 2004;19(9):966–971

32. Merguerian PA, Herz D, McQuiston L, VanBibber M. Variation among pediatricurologists and across 2 continents inantibiotic prophylaxis and evaluation forprenatally detected hydronephrosis:a survey of American and European pe-diatric urologists. J Urol. 2010;184(suppl4):1710–1715

33. Yiee JH, Tasian GE, Copp HL. Managementtrends in prenatally detected hydro-nephrosis: national survey of pediatricianpractice patterns and antibiotic use. Urol-ogy. 2011;78(4):895–901

34. Lin KY, Chiu NT, Chen MJ, et al. Acute py-elonephritis and sequelae of renal scar inpediatric first febrile urinary tract in-fection. Pediatr Nephrol. 2003;18(4):362–365

REVIEW ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 e261 at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2012-1870; originally published online December 17, 2012; 2013;131;e251Pediatrics

and Armando J. LorenzoLuis H. Braga, Hana Mijovic, Forough Farrokhyar, Julia Pemberton, Jorge DeMaria

HydronephrosisAntibiotic Prophylaxis for Urinary Tract Infections in Antenatal

  

ServicesUpdated Information &

tmlhttp://pediatrics.aappublications.org/content/131/1/e251.full.hincluding high resolution figures, can be found at:

Subspecialty Collections

_diseasehttp://pediatrics.aappublications.org/cgi/collection/infectiousInfectious Disease & Immunitythe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

tmlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhtables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at UNIV OF CHICAGO on May 16, 2013pediatrics.aappublications.orgDownloaded from