The purpose of this study was to compare antibiotic resistance patterns between first urinary tract infection (UTI) and recurrent UTI groups and to obtain information regarding empirical antibiotic selection for treating recurrent UTI.

We retrospectively reviewed 148 children treated for UTIs from January 2009 to June 2016. The patients were divided into two groups: first UTI (N = 148) and recurrent UTI (17 patients and 20 episodes).

In both groups, Escherichia coli was the most frequent causative organism, accounting for 89.9% and 75.0% in the first and recurrent UTI groups, respectively. When E. coli or Klebsiella pneumoniae was the causative organism, extended-spectrum β-lactamase (ESBL)-producing organisms were more frequent in the recurrent UTI group (17.6%) than in the first UTI group (14.0%); however, this difference was not statistically significant (P = 0.684). Cefotaxime was the most frequently used first-line empirical antibiotic in both groups. In the first UTI and recurrent UTI groups, 7.4% and 15.0% of patients were treated with intravenous antibiotics as definitive therapy, respectively (P = 0.250). Fifteen out of 17 patients having a second UTI had different causative organisms or antibiotic susceptibility patterns compared to their previous episode.

Escherichia coli was the most frequent causative organism in the recurrent UTI group. There were no differences in the proportion of ESBL-producing organisms between the first UTI and recurrent UTI groups. Therefore, when a UTI recurs in children, the antibiotics effective on the most common causative organism might be administered as empirical antibiotics.