Avoidance of voiding cystourethrography in infants younger than 3 months with <i>Escherichia coli</i> urinary tract infection and normal renal ultrasound.

Abstract

Urinary tract infection (UTI) represents the most common bacterial infection in infants, and its prevalence increases with the presence of high-grade vesicoureteral reflux (VUR). However, voiding cystourethrography (VCUG) is invasive, and its indication in infants <3 months is not yet defined. This study aims to investigate, in infants aged 0-3 months, if the presence of <i>Escherichia coli</i> versus non-<i>E. coli</i> bacteria and/or normal or abnormal renal ultrasound (US) could avoid the use of VCUG. One hundred and twenty-two infants with a first febrile UTI were enrolled. High-grade VUR was defined by the presence of VUR grade ≥III. The presence of high-grade VUR was recorded using VCUG, and correlated with the presence of <i>E. coli</i>/non-<i>E. coli</i> UTI and with the presence of normal/abnormal renal US. The Bayes theorem was used to calculate pretest and post-test probability. The probability of high-grade VUR was 3% in the presence of urinary <i>E. coli</i> infection. Adding a normal renal US finding decreased this probability to 1%. However, in the presence of non-<i>E. coli</i> bacteria, the probability of high-grade VUR was 26%, and adding an abnormal US finding increased further this probability to 55%. In infants aged 0-3 months with a first febrile UTI, the presence of <i>E. coli</i> and normal renal US findings allow to safely avoid VCUG. Performing VCUG only in infants with UTI secondary to non-<i>E. coli</i> bacteria and/or abnormal US would save many unnecessary invasive procedures, limit radiation exposure, with a very low risk (<1%) of missing a high-grade VUR.