Antibiotic Prophylaxis for Urinary Tract Infections in Children

With Spina Bifida on Intermittent Catheterization

Bas Zegers,* Cuno Uiterwaal, Jan Kimpen, Jan van Gool, Tom de Jong,Pauline Winkler-Seinstra, Saskia Houterman, Carla Verpoortenand Catharine de Jong-de Vos van SteenwijkFrom the Department of Pediatrics, Wilhelmina Children’s Hospital (BZ), and Julius Center for Health Sciences and Primary Care (CU),University Medical Center (JK, TdJ, PW-S, CdJ-dVvS), Utrecht and Department of Pediatrics (BZ), Maxima Medical Center (SH), Veldhoven,The Netherlands, Division of Pediatric Urology, University Essen-Duisburg, Essen, Germany (JvG), and Department of Pediatrics,Gasthuisberg University Hospital, Leuven, Belgium (CV)

Purpose: Antibiotic prophylaxis (low dose chemoprophylaxis) has been pre-scribed since the introduction of clean intermittent catheterization in childrenwith spina bifida. We hypothesized that stopping low dose chemoprophylaxis doesnot increase the number of urinary tract infections in these patients.Materials and Methods: A total of 176 patients with spina bifida participated ina randomized controlled trial (ISRCTN trial number 56278131) of either contin-uation or discontinuation of low dose chemoprophylaxis. During the 18-monthstudy period biweekly urine samples were evaluated for leukocyturia and bacte-riuria with dipsticks and cultures. Asymptomatic significant bacteriuria (positiveculture results without clinical symptoms) and urinary tract infections (signifi-cant bacteriuria with clinical symptoms and leukocyturia) were analyzed.Results: Discontinuation of low dose chemoprophylaxis resulted in higher rates ofasymptomatic significant bacteriuria (incidence rate ratio 1.23, 95% CI 1.08–1.40,p � 0.002) and urinary tract infection (IRR 1.44, 95% CI 1.13–1.83, p � 0.003). Forurinary tract infection the number needed to harm was 2.2, that is if 2 patientsdiscontinued low dose chemoprophylaxis for a year, 1 extra urinary tract infectionwould result. Febrile urinary tract infection occurred once in every 30 patient-yearsand slightly more often in the discontinuation group (relative risk 2.0, 95% CI0.38–10.6, p � 0.4). Of 88 patients allocated to discontinuation of low dose chemo-prophylaxis 38 (43%) switched back to chemoprophylaxis. The urinary tract infectionrate was nonsignificantly higher in the presence of vesicoureteral reflux. Male genderand a low pre-study rate of urinary tract infection predicted successful discontinuation.Conclusions: Patients with spina bifida on clean intermittent catheterizationand antibiotic prophylaxis for urinary tract infections can safely discontinue thisprophylaxis, in particular males, patients with low urinary tract infection ratesand patients without vesicoureteral reflux.

Key Words: antibiotic prophylaxis, bacteriuria, intermittent urethral

Abbreviations

and Acronyms

ABU � asymptomatic significantbacteriuria

CIC � clean intermittentcatheterization

LDCP � low dosechemoprophylaxis

UTI � urinary tract infection

VUR � vesicoureteral reflux

Submitted for publication April 28, 2011.Funded by grants from the Wilhelmina Chil-

dren’s Hospital Fund, Utrecht, The Netherlandsand Astra Tech, Brussels, Belgium.

* Correspondence: Department of Pediatrics,Maxima Medical Center, P. O. Box 7777, 5500MB, Veldhoven, The Netherlands (telephone: 00-31-40-888-8270; FAX: 00-31-40-888-9609; e-mail:b.zegers@mmc.nl).

catheterization, spina bifida cystica, urinary tract infections

WITH an incidence of 1 in every 2,000live births, spina bifida is still amongthe most common and serious congeni-tal birth defects.1,2 Most children withspina bifida survive with fairly stable

medical and social problems, except for

0022-5347/11/1866-2365/0THE JOURNAL OF UROLOGY®

© 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RES

the neuropathic bladder sphincter dys-function. About half of all patients suf-fer from detrusor-sphincter dyssyner-gia, a functional obstruction of thebladder outlet.3,4 This condition predis-

poses to incomplete bladder emptying,

Vol. 186, 2365-2371, December 2011Printed in U.S.A.

EARCH, INC. DOI:10.1016/j.juro.2011.07.108www.jurology.com 2365

CHEMOPROPHYLAXIS IN CHILDREN WITH SPINA BIFIDA ON CATHETERIZATION2366

high bladder pressures due to high sphincter pressureand increased risk of urinary tract infections, whichcan lead to vesicoureteral reflux, upper urinary tractdilatation, renal scarring and renal failure.5,6

Clean intermittent catheterization, often combinedwith anticholinergic medication to decrease detrusoractivity, has revolutionized the management of detru-sor-sphincter dyssynergia since its introduction in1972.7,8 Low dose chemoprophylaxis is commonly usedalong with CIC and anticholinergics to prevent recurrentUTIs due to detrusor-sphincter dyssynergia.9,10 Periodiccystometry and urodynamic investigations, along withthe aforementioned treatments, have decreased renalscarring, end-stage renal failure and the need for kidneyreplacement therapy or transplantation.11

Despite these therapeutic adjustments and prophy-laxis, children with spina bifida on CIC still manifestbacteriuria and UTIs, necessitating antibiotic treat-ment.12 These findings are also reported in childrenwithout spina bifida, demonstrating a variable benefit oflong-term LDCP in recurrent UTI, vesicoureteral reflux,perinatal hydronephrosis and functional anomalies.13,14

As a result, there is ongoing international debate on, andlocally variable use of, LDCP.

Differentiation between asymptomatic significantbacteriuria and UTI without fever can be difficult. CICper se can result in low grade leukocyturia and varyingdegrees of bacteriuria,15,16 and specific symptoms of UTI,such as pollakiuria, urgency and dysuria, will be absentin most children with spina bifida. However, patientswith spina bifida on CIC and their parents often describespecific symptoms during UTI, such as increased incon-tinence and foul smell of urine. Fever is a strong indicatorof pyelonephritis and, as reported in several studies, fe-brile UTIs specifically cause renal scarring.17,18 UTIswithout fever have proved to be less detrimental for renalsurvival.19

The long-term use of LDCP induces multiresistanturopathogens in the fecal flora—a high price to pay forprevention of afebrile UTIs, which are perceived tohave little or no renal consequences. We hypothesizedthat LDCP does not decrease the number of febrileUTIs in children with spina bifida on CIC. To ourknowledge there has not been experimental researchto prove the effectiveness of LDCP, although its usehas become common practice. We present a randomizedstudy aimed at determining whether LDCP in childrenwith spina bifida on CIC can be discontinued safely, asmeasured by number of febrile urinary infections.

PATIENTS AND METHODS

ParticipantsAll patients with known spina bifida seen at the outpa-tient clinics of Wilhelmina Children’s Hospital, Utrechtand Gasthuisberg University Hospital, Leuven in 2005

were candidates for inclusion in the study, provided they

were performing CIC and using LDCP during the preced-ing 6 months. Combining Utrecht (210 patients) and Leu-ven (252) populations, 462 patients with spina bifida onCIC and LDCP were identified to be eligible for the study.Eligible patients were invited by personal invitation andwritten explanation to participate in a randomized trial ofdiscontinuing LDCP. The study ran from February 2005through March 2009.

Baseline MeasurementsPatient age, gender, type and dose of antibiotic prophy-laxis, and type of bladder sphincter dysfunction wererecorded. UTIs proved in the year preceding randomiza-tion were recorded, and renal ultrasound was performedto determine renal length and presence of pyelum dila-tation as a marker for VUR. Urine culture was per-formed to exclude a current UTI at the start of the studyperiod.

InterventionsPatients were randomly allocated to continue or discon-tinue LDCP. The Julius Center for Health Sciences andPrimary Care, Utrecht performed computer based, ran-dom, concealed allocation of patients to either continue ordiscontinue LDCP. Stratification was performed for ageyounger and older than 3 years, gender and participatingcenter.

Continuation of LDCP meant that the individually pre-scribed type and dose of antibiotics were continued. Thedosages and types were allowed to differ between patientsaccording to antibiotic resistance patterns in pre-studycultures.

Followup, Outcome

Measurements and Primary Outcome DefinitionDuring 18-month followup biweekly dipstick tests andurine cultures were performed by the patients, their par-ents or their primary caretakers. The dipstick for urinaryleukocytes and nitrite (Combur2LN®) was rated as eithernegative (no color change) or positive (any color change) bythe primary caretakers.

Urine culture was performed using a Uricult® test withMacConkey and cysteine lactose electrolyte deficient me-dia. In Utrecht the culture was sent to the laboratory ofclinical microbiology of the university medical center for24-hour incubation at 37C (98F). If rated positive, coloniesfrom the culture were plated on sheep blood agar, incu-bated for 72 hours, and evaluated for significant growthand bacterial resistance patterns. When negative, the cul-ture was not incubated on sheep blood agar. In Leuven the24-hour incubation period was performed at home by theprimary caretakers (parents or nurses), using a feedingbottle warmer (Philips Healthcare, Andover, Massachu-setts) at 37C. If rated positive, the culture was sent to thelaboratory of clinical microbiology of the university hospi-tal, where a sheep blood agar was performed and evalu-ated after 72 hours of incubation at 37C for significantgrowth and bacterial resistance patterns. When rated neg-ative at home, the culture was sent to the trained researchnurse for professional review. When the nurse secondarilyrated the culture as positive after all, it was also sent to

the laboratory for incubation.

CHEMOPROPHYLAXIS IN CHILDREN WITH SPINA BIFIDA ON CATHETERIZATION 2367

Significant bacteriuria was defined as more than10,000 cfu/ml for a single specimen in a catheterizedsample, according to previous reviews.13,16 UTI wasdefined as 1) significant bacteriuria, 2) positive readingof leukocyturia on dipstick and 3) clinical symptoms,such as increasing incontinence and foul smell or cloud-iness of the catheterized urine with or without fevergreater than 38.5C (101F), and treated as such. Morethan 1 specimen of bacteria with or without clinicalsymptoms in any culture was considered a contamina-tion and, therefore, not treated. The primary outcomewas the incidence of ABU and afebrile and febrile UTIper patient-year.

BlindingA priori we did not use placebo LDCP, aiming to mea-sure pharmaceutical effects and ceasing induced exter-nal effects to mimic the anticipated effects in currentclinical practice. Use of placebo would have been diffi-cult since the dosages and types of LDCP differed be-tween participants due to antibiotic resistance patternsin pre-study cultures. We performed blinded outcome

Total excluded n=Choice parent/patient 52% No LDCP 17% Social circumstances 10% Refusal of GP / care taker 9Language barrier 9% LDCP < 6 months 3%

Continue LDCP n

Lost to follow-up n=0 Treatment change during

follow-up n=0

Total analyzed after up n=88

Study enrollment and outcomes flow

evaluation since graders of Uricults and laboratory cul-tures were unaware of patient treatment allocations.

Statistical MethodsGiven a 17% annual pre-study UTI rate in our patientswith spina bifida on LDCP, randomization of 150 pa-tients was calculated to be sufficient for showing equiv-alence of continuing or discontinuing LDCP regardingprimary outcome, with a power of 80% (alpha � 0.05),accepting a difference of 15% between the groups asbioequivalent. After 6 months the episodes of (a)febrileUTIs were reviewed to determine the need for continu-ation of the study and the risk of discontinuing LDCP.

Main treatment effect analyses were done accordingto the intent to treat principle. First, important prog-nostic characteristics regarding UTIs were tabulatedagainst allocated treatment. Log binomial regressionwas used with first febrile UTI (yes/no) as dependentvariable and a group indicator (LDCP yes/no) as inde-pendent variable. Subsequently differences in UTI ratesbetween the 2 treatment groups were analyzed usingPoisson regression, with a group indicator (LDCP yes or

l eligible n=462 t n=210 / Leuven

n=252

andomized n=176

Discontinue LDCP n=88

to follow-up n=0ent change during llow-up n=38

Total analyzed after follow-up n=88

TotaUtrech

Total r

286

%

=88

follow-

LostTreatm

fo

chart. GP, general practitioner.

CHEMOPROPHYLAXIS IN CHILDREN WITH SPINA BIFIDA ON CATHETERIZATION2368

no) as independent variable. Results are expressed asrelative risks (febrile UTI) and incidence rate ratios(UTIs, ABUs) with 95% confidence intervals and p val-ues. As a post hoc analysis, we assessed which baselinecharacteristics could predict successful compliance withstopping LDCP. We used Poisson regression with offsetset at 1 and with successful compliance with discontin-uing (yes/no) as dependent variable, and age, gender,and history of VUR and UTIs as independent variables.

Results are expressed as relative risks with 95%confidence intervals and p values. Data were analyzedusing the generalized linear regression module ofSPSS®, version 17 for Windows.

RESULTS

Of 462 patients with spina bifida 286 were excludedfrom the study due to patient and/or parental re-fusal to participate, disapproval of the treating phy-sician to participate, lack of administration of LDCP6 months before the study and language barrier (seefigure). The remaining 176 eligible patients allagreed to randomization. A total of 88 patients wereallocated to discontinue LDCP and 88 to continueLDCP. The figure illustrates the study flowchart forintake and 18 months of followup. The LDCP regi-men in participants was diverse, consisting of eithertrimethoprim, nitrofurantoin, cefuroxim, co-tri-moxazole or a combination of prophylactics.

Table 1 outlines baseline characteristics thatwere considered prognostic indicators of primaryoutcome. None of these characteristics differed be-tween the treatment groups. Table 2 reveals thestudy results, with febrile UTIs occurring 7 timesin 6 patients, 2 in the continuation group and 4 inthe discontinuation group. The discontinuationgroup had a twofold higher risk of febrile UTI thanthe continuation group but this difference was not

Table 1. Baseline characteristics by treatment group

Discontinued Continued

Mean � SD age (yrs) 9.4 � 5.9 8.7 � 6.7No. age:

Younger than 3 yrs 14 16Older than 3 yrs 74 72

No. males (%) 38 (45) 38 (42)Mean � SD renal length (mm):

Lt side 75.4 � 21.3 75.2 � 19.2Rt side 74.8 � 23.7 76.6 � 16.2

% Overactive detrusor 35.3 31.4% Overactive sphincter 25.0 22.1% Reflux:

Rt side 10.3 4.8Lt side 10.3 9.5Rt and/or lt side 16.9 17.0

Mean � SD bladder capacity (ml) 269.8 � 142.7 280.1 � 161.3No. UTI/pt-yr before study 1.88 1.82

statistically significant. One participant from the

stop group suffered 2 febrile UTIs, which occurredafter switching back to LDCP.

Afebrile UTIs occurred in 84 (47%) of the par-ticipants, leaving 92 (53%) without UTIs, with a44% higher rate in the discontinuation group (1.5UTIs per year) vs the continuation group (1.1).This finding resulted in a number needed to harmof 2.2, meaning that if 2.2 patients stopped LDCPfor a year, 1 extra UTI would result. Overall rateof UTIs in patients allocated to discontinue LDCPdecreased from 1.8 per year before the study to 1.5per year during followup.

Regarding VUR, the number of UTIs per pa-tient-year was similar in the continuation group(1.2 vs 1.0 per patient-year in those with andwithout VUR, respectively, p � 0.58) and nonsig-nificantly higher in the discontinuation group (2.3vs 1.3, p � 0.22). VUR grade, bladder compliance,bladder capacity, type of LDCP and age did notsignificantly influence number of UTIs per pa-tient-year. Finally, ABU occurred at a 23% higherrate in the discontinuation group (incidence rate4.6 per year) than in the continuation group (3.6).

Of the 88 patients allocated to the discontinua-tion group 50 (57%) fully complied with the proto-col during the study and 38 (43%) switched back toLDCP. Table 3 outlines baseline characteristics ofchildren as predictors of compliance with discon-tinuing LDCP. Age and VUR did not clearly pre-dict compliance. However, males had a 1.4 timesincreased chance of compliance, and for every ex-tra previous UTI per patient-year there was a 20%decreased chance of compliance.

DISCUSSION

Discontinuation of low dose chemoprophylaxis inpatients with spina bifida on CIC resulted in sig-nificantly more UTIs overall compared to contin-uation of chemoprophylaxis. However, this in-crease was clinically small. If 2.2 patients stoppedLDCP for 1 year, only 1 extra, most probably

Table 2. Effects of antibiotic prophylaxis for urinary tractinfections and bacteriuria episodes

Absolute No.Febrile UTIs* No. ABUs/Pt-Yr†

No. AfebrileUTIs/Pt-Yr†

LDCP:Continued 2 3.64 1.07Discontinued 4 4.58 1.52

Relative risk/incidencerate ratio (95% CI)

2.0 (0.38–10.6) 1.25 (1.08–1.40) 1.44 (1.13–1.83)

p Value 0.42 0.002 0.003

* Log binomial analysis.

† Poisson regression analysis.

CHEMOPROPHYLAXIS IN CHILDREN WITH SPINA BIFIDA ON CATHETERIZATION 2369

harmless, afebrile UTI would result. The numberof febrile UTIs between the discontinuation andcontinuation groups did not differ significantly.

To our knowledge this is the first randomizedtrial regarding the necessity of prescribing LDCPin children with spina bifida on CIC. PreviouslyClarke et al studied 81 patients with spina bifidaon CIC who were randomized to discontinue orcontinue LDCP with a relatively short followup of4 months.9 However, analysis of the data wasperformed after excluding 28 patients for noncom-pliance with randomization and by analysis perprotocol instead of the intent to treat principle.Therefore, these results are less applicable in clin-ical practice. The pragmatic character of our trialdesign resulted in substantial crossovers in theLDCP discontinuation group, although this resultmimics the daily practice of many attending phy-sicians.

Importantly the UTIs in our participants wereafebrile in all but 7 patients in both groups. Theseinfections are less likely to cause renal damageaccording to previous studies.12,19 Our study dem-onstrated that patients with VUR or younger than3 years did not suffer from a significantly greaternumber of UTIs per patient-year, either with orwithout LDCP. These results are different fromstudies of regular pediatric cohorts from past de-cades,13,20 and may be due to the predominantlylow grades of VUR in our patient population.

ABU is a well-known and logical consequence ofdaily CIC. In earlier studies colonization of thebladder with bacteria was seen in almost 50% ofchildren with spina bifida on CIC.15,21 In our se-ries the number of positive urine cultures withoutclinical signs was significantly lower, ie 16% and20% in the continuation and discontinuationgroups, respectively. These lower percentages canbe explained by an evident decrease in positivecultures in the latter months, most likely due toincreased hygiene and awareness in participants.

In our study many of the patients who discon-tinued LDCP had multiple periods of asymptom-

Table 3. Predictors of compliant discontinuation of low dose ch

Compliance Un

Yes No Relative Risk (

Mean age (yrs) 9.0 9.8 0.99 (0.96–No. gender: 1.48 (1.03–

M 27 23F 12 26

No. VUR: 0.86 (0.49–Present 7 43Absent 7 31

Mean UTIs/yr 1.1 2.7 0.80 (0.67–

Analysis of 88 children allocated to no LDCP, of whom 38 switched back to LDCP after

atic significant bacteriuria and urinary tract in-fection, causing 43% to restart LDCP during thetrial. However, these children already had a highnumber of pre-study UTIs per patient-year, andrestarting LDCP did not decrease the number ofUTIs per patient-year in the remaining study pe-riod. This finding confirms previous results thatthe minor influence of LDCP on the number ofUTIs does not outweigh the disadvantages of dailyantibiotic prophylaxis, such as gastrointestinalside effects and, most importantly, increased bac-terial resistance patterns.14

Limitations of our study involve the Uricultmethod. Analysis of the cultures is prone to interob-server variability, which was overcome by using just2 research nurses to evaluate the results. We chosegreater than 10,000 cfu/ml as the cutoff for UTI,according to previous studies and reviews.13,16 If wehad chosen more than 100,000 cfu/ml, lower rates ofUTI would have been noted in both groups, althoughwithout different conclusions regarding LDCP.Transportation was not cooled, which is in accor-dance with the manufacturer statement of adequatereading of bacterial growth at room temperatureafter 48 hours. The different approach to cultureevaluation at both centers due to logistic circum-stances was overcome by a second expert inspectionof every sample in the Leuven branch of the study.The objective evaluation of cultures and dipsticks byexperts blinded to the prior use of antibioticsstrengthens the validity of our conclusions.

Although the difference in UTIs per patient-year was statistically significant, in clinical prac-tice patients with spina bifida have to endure anextensive period of daily antibiotic prophylaxis toprevent a single afebrile nonscarring UTI. VUR,gender, age, and bladder compliance and capacitydo not influence the risk of UTI. Therefore, anti-biotic prophylaxis should be continued only inchildren with spina bifida with a high rate of fe-brile UTIs.

rophylaxis

Analysis Multivariable Analysis

p Value Relative Risk (95% CI) p Value

0.54 0.98 (0.95–1.01) 0.110.04 1.40 (1.00–1.96) 0.05

0.60 0.92 (0.57–1.49) 0.92

0.02 0.79 (0.65–0.96) 0.02

emop

ivariable

95% CI)

1.02)2.12)

1.50)

0.96)

randomization.

CHEMOPROPHYLAXIS IN CHILDREN WITH SPINA BIFIDA ON CATHETERIZATION2370

CONCLUSIONSDiscontinuing low dose chemoprophylaxis in pa-tients with spina bifida on clean intermittent cath-

eterization does not significantly increase the

REFERENCES

EDITORIAL COMMENTS

REFERENCE

1. Lapides J, Diokno AC, Silber SJ et al: Clean, intermittent self-catheterization in the tr

number of febrile urinary tract infections. There-fore, prophylaxis may be discontinued once neuro-genic bladder sphincter dyssynergia has been sta-

bilized.

1. Rader JI and Schneeman BO: Prevalence of neu-ral tube defects, folate status, and folate fortifi-cation of enriched cereal-grain products in theUnited States. Pediatrics 2006; 117: 1394.

2. Shaer CM, Chescheir N and Schulkin J: Myelo-meningocele: a review of the epidemiology, ge-netics, risk factors for conception, prenatal diag-nosis, and prognosis for affected individuals.Obstet Gynecol Surv 2007; 62: 471.

3. van Gool JD, Dik P and De Jong TP: Bladder-sphincter dysfunction in myelomeningocele. EurJ Pediatr 2001; 160: 414.

4. Bauer SB: Neurogenic bladder: etiology and as-sessment. Pediatr Nephrol 2008; 23: 541.

5. Lawrenson R, Wyndaele JJ, Vlachonikolis I et al:Renal failure in patients with neurogenic lowerurinary tract dysfunction. Neuroepidemiology2001; 20: 138.

6. Singhal B and Mathew KM: Factors affectingmortality and morbidity in adult spina bifida. EurJ Pediatr Surg, suppl., 1999; 9: 31.

7. Dik P, Klijn AJ, van Gool JD et al: Early start totherapy preserves kidney function in spina bifidapatients. Eur Urol 2006; 49: 908.

8. Lapides J: Urinary diversion. Surgery 1971; 69:

9. Clarke SA, Samuel M and Boddy SA: Are prophy-lactic antibiotics necessary with clean intermit-tent catheterization? A randomized controlledtrial. J Pediatr Surg 2005; 40: 568.

10. Schlager TA, Anderson S, Trudell J et al: Nitro-furantoin prophylaxis for bacteriuria and urinarytract infection in children with neurogenic blad-der on intermittent catheterization. J Pediatr1998; 132: 704.

11. de Jong TP, Chrzan R, Klijn AJ et al: Treatment ofthe neurogenic bladder in spina bifida. PediatrNephrol 2008; 23: 889.

12. Sauerwein D: Urinary tract infection in patientswith neurogenic bladder dysfunction. Int J Anti-microb Agents 2002; 19: 592.

13. Craig JC, Simpson JM, Williams GJ et al: Anti-biotic prophylaxis and recurrent urinary tract in-fection in children. N Engl J Med 2009; 361:1748.

14. Williams GJ, Wei L, Lee A et al: Long-termantibiotics for preventing recurrent urinary tractinfection in children. Cochrane Database Syst Rev2006; 3: CD001534.

15. Schlager TA, Dilks S, Trudell J et al: Bacteriuria in

eatment of urinary tract dis

mittent catheterization: natural history. J Pediatr1995; 126: 490.

16. Ottolini MC, Shaer CM, Rushton HG et al: Rela-tionship of asymptomatic bacteriuria and renalscarring in children with neuropathic bladderswho are practicing clean intermittent catheteriza-tion. J Pediatr 1995; 127: 368.

17. Wennerstrom M, Hansson S, Jodal U et al: Renalfunction 16 to 26 years after the first urinary tractinfection in childhood. Arch Pediatr Adolesc Med2000; 154: 339.

18. Hiraoka M, Hashimoto G, Tsuchida S et al: Earlytreatment of urinary infection prevents renaldamage on cortical scintigraphy. Pediatr Nephrol2003; 18: 115.

19. Jahnukainen T, Chen M and Celsi G: Mechanismsof renal damage owing to infection. PediatrNephrol 2005; 20: 1043.

20. Montini G and Hewitt I: Urinary tract infections:to prophylaxis or not to prophylaxis? PediatrNephrol 2009; 24: 1605.

21. Dilks SA, Schlager T, Kopco JA et al: Frequencyand correlates of bacteriuria among childrenwith neurogenic bladder. South Med J 1993;

142. children with neurogenic bladder treated with inter- 86: 1372.

The introduction by Lapides et al of clean intermit-tent catheterization revolutionized urology.1 Itmade augmentation cystoplasty and neobladderspractical. It also freed many patients, particularlythose with neurogenic bladders, from the dangers ofchronic urethral and suprapubic catheters. Whilemost patients could safely perform CIC without riskof infection, some were observed to be more vulner-able. The current study confirms that CIC can besafely conducted in most patients with neurogenicbladders without antibiotic prophylaxis—but not

continued prophylaxis later restarted it. The au-thors note that these children seemed to have a highnumber of pre-study UTIs per patient-year. Doesthis finding suggest that these patients are actuallydifferent than the typical patients followed by theauthors? Further investigation of these patientsmay help to determine better who should need pro-phylaxis and why.

Julian Wan

Department of UrologyUniversity of Michigan

all. It is intriguing that 43% of those who had dis- Ann Arbor, Michigan

ease. J Urol 1972; 107: 458.

CHEMOPROPHYLAXIS IN CHILDREN WITH SPINA BIFIDA ON CATHETERIZATION 2371

I congratulate the authors for performing a random-ized controlled trial. The use of antibiotic prophy-laxis (LDCP) for children with neuropathic bladderis less common in the United States but is still acontentious issue. With the ongoing controversy re-garding LDCP in children with reflux, it is unset-tling that we have not determined its usefulness inchildren with a much higher risk of renal damage.Complicating the issue, and not discussed or con-trolled in the present study, is the well-known ex-tremely low compliance with taking long-term anti-biotics.1 If one were to factor in an expectedcompliance of less than 40%, then the extrapolationis that LDCP may decrease bacteriuria and UTIeven more significantly.

Although the present study is limited by space,further evaluation and discussion should be affordedto nonantibiotic treatments and preventive factorsregarding UTI. We have been successful having ourpatients avoid antibiotics by treating cloudy urine ormild symptoms (increased incontinence or suprapu-

REFERENCES

eterization. Aggressive bowel management and reliefof constipation are also associated with a decreasedincidence of UTI.

The primary focus of urological management forchildren with spina bifida is preservation of renalfunction. Secondary goals are achieving urinary con-tinence and preventing symptomatic infections.Multivariate analysis in a large, long-term patientcohort showed that girls with reflux have a 55 timesincreased risk of renal cortical loss.2 The presentstudy suggests that children with spina bifida andvesicoureteral reflux should be maintained on LDCP,whereas males without preexisting UTI problems donot benefit from LDCP. Although this finding seemslike common sense, this prospective analysis is appre-ciated and substantiates our practice patterns withobjective evidence.

Eric A. Kurzrock

Division of Pediatric UrologyU. C. Davis Children’s Hospital

Shriners Hospital for Children–Northern California

bic pain) with copious hydration and frequent cath- Sacramento, California

1. Copp HL, Nelson CP, Shortliffe LD et al: Compliance with antibiotic prophylaxis in children with vesicoureteral reflux: results from a national pharmacy claims database.J Urol 2010; 183: 1994.

2. DeLair SM, Eandi J, White MJ et al: Renal cortical deterioration in children with spinal dysraphism: analysis of risk factors. J Spinal Cord Med, suppl., 2007; 30: S30.