Read the Complete EBM Focus and Earn CME.
Reference: N Engl J Med 2014 Jun 19;370(25):2367 (level 2 [mid-level] evidence)
Vesicoureteral reflux (VUR) is present in about 39% of children presenting with symptomatic urinary tract infection, and has been associated with an increased risk of renal scarring in some studies (N Engl J Med 2003 Jan 16;348(3):195 full-text). There has been conflicting evidence for the benefit of antibiotic prophylaxis in reducing recurrent symptomatic UTIs in this patient population. A previous systematic review demonstrated lower rates of recurrent symptomatic UTI in favor of antibiotic prophylaxis, but these results were not statistically significant (Cochrane Database Syst Rev 2011 Jun 15;(6):CD001532). In addition, a separate meta-analysis found benefit for antibiotic prophylaxis in patients with high-grade VUR, although these results also did not reach statistical significance (Pediatrics 2011 Sep;128(3):595). The largest trial included in the systematic review and meta-analysis was a high-quality randomized trial aimed at preventing recurrent urinary tract infections (also including some children without VUR). This trial did show a reduction in recurrent symptomatic UTI at 1 year in patients receiving trimethoprim-sulfamethoxazole, but the results were not statistically significant in the subgroup of children with VUR (N Engl J Med 2009 Oct 29;361(18):1748). To help resolve this uncertainty, a recent randomized trial evaluated antibiotic prophylaxis with trimethoprim-sulfamethoxazole in 607 children aged 2-72 months with grade I-IV VUR.
Children were randomized to trimethoprim 3 mg/kg/day plus sulfamethoxazole 15 mg/kg/day orally vs. placebo and followed for 2 years. VUR was identified using voiding cystourethrogram, and 46.5% had grade III-IV VUR. All children had 1-2 previous urinary tract infections with fever ≥ 38 degrees C (100.4 degrees F), or UTI symptoms within 24 hours before or after urine collection. Follow-up data were available for 509 children (84%) at 1 year and 520 children (86%) at 2 years. Renal scarring was assessed with dimercaptosuccinic acid (DMSA) renal scintigraphy at baseline, 1 year, and 2 years. At baseline, resistance to trimethoprim-sulfamethoxazole in index UTI was present in 20% of the trimethoprim-sulfamethoxazole group vs. 22% of the placebo group. Treatment failure was defined as any of 2 febrile UTIs, 1 febrile UTI plus 3 symptomatic UTIs, 4 symptomatic UTIs, or new or worsening renal scarring at 1 year.
The rate of recurrence of febrile or symptomatic UTI at 2 years was 12.9% with trimethoprim-sulfamethoxazole vs. 23.6% with placebo (p < 0.001, NNT 10). Treatment failure occurred in 5% with trimethoprim-sulfamethoxazole vs. 9.6% with placebo (p = 0.04, NNT 22). The proportion of patients with both a recurrent febrile or symptomatic UTI and resistance to trimethoprim-sulfamethoxazole was 9% in the trimethoprim-sulfamethoxazole group vs. 5% in the placebo group. The rate of renal scarring was 11.9% with trimethoprim-sulfamethoxazole vs. 10.2% with placebo (not significant).
This new trial is the first to clearly show a reduction in the recurrence of febrile or symptomatic UTI in children with VUR receiving trimethoprim-sulfamethoxazole compared to placebo. However, the trial also showed numerically higher rates of resistance to the antibiotic at 1-2 years, and the long-term effects of this need to be considered. Several previous trials have failed to show statistically significant benefit for antibiotic prophylaxis in reducing recurrent febrile or symptomatic UTI in children with VUR, but these most recent data may tilt the balance in favor of prophylaxis in this patient population.
For more information see the Vesicoureteral reflux topic in DynaMed