Does continuous antibiotic prophylaxis for infants with vesicoureteral reflux grade III to V provide more benefits than harms?
Antibiotic prophylaxis for infants with grade III to V vesicoureteral reflux and no history of UTI reduced the likelihood of a first UTI (NNT = 7) but had no impact on any renal outcomes and increased the likelihood of antibiotic-resistant and non–E. coli isolates. The authors suggest (sensibly) that we should probably not give prophylaxis to all comers. The benefit-risk ratio is more favorable for girls and for those infants who have already had one infection.
Randomized controlled trial (nonblinded)
Previous trials have shown a reduction in urinary tract infections (UTIs) for infants with vesicoureteral reflux and a previous UTI, there was no apparent effect on kidney scarring or other important renal outcomes. In this study, researchers identified 292 infants 1 month to 5 months old with vesicoureteral reflux grade III to V and no previous UTI. They were randomized to receive daily antibiotic prophylaxis with the antibiotic of the clinician's choosing (typically nitrofurantoin, cefixime, amoxicillin clavulanate, or trimethoprim-sulfamethoxazole) or no treatment (care as usual). The population was 78% men and 96% White, and 80% had grade IV or V vesicoureteral reflux. Groups were balanced at baseline. Allocation concealment is not described, and the trial was open label, which would tend to bias results toward the active intervention. After 24 months, there were fewer first symptomatic UTIs in the antibiotic group (21.2% vs 35.6%; P = .008; number needed to treat (NNT) = 7 over 2 years). There were also slightly fewer total UTIs (60 vs 79; relative risk 0.76; 95% CI 0.59 - 0.97), but no difference in the percentage of UTIs resulting in hospitalization or the use of intravenous antibiotics. The number of new kidney scars was almost identical between groups, and there were no differences in the glomerular filtration rate or serum creatinine level between groups. The antibiotic group was more likely to have non–Escherichia coli infections such as pseudomonas (6 vs 0 in the usual care group) and were more likely to have antibiotic-resistant isolates. In the usual care group, 64.4% did not have a single UTI during the study period.
Mark H. Ebell, MD, MS
University of Georgia