# P-418. Do We Need Broad Empirical Antibiotics for Upper Urinary Tract Infections Caused by Chromosomal AmpC-Producing Enterobacterales in Children?

## TL;DR

This study found that using initial antibiotics not effective against certain bacteria in children's urinary tract infections did not lead to worse outcomes.

## Contribution

The study shows that broad-spectrum antibiotics may not be necessary for initial treatment of AmpC-E UTIs in children.

## Key findings

- No significant difference in time to defervescence between discordant and concordant therapy groups.
- 30-day culture-confirmed recurrence was low and similar in both groups.

## Abstract

The emergence of antimicrobial resistance has complicated management of infections. We investigated whether empirical therapy discordant for chromosomal AmpC-Producing Enterobacterales (AmpC-E) had a clinical impact before switching to definitive therapy in pediatric urinary tract infections (UTIs).

A retrospective study was conducted at Tokyo Metropolitan Children’s Medical Center from July 2010 to January 2024. Inclusion criteria were patients aged < 16 years with fever ≥ 38.0℃ and ≥ 10⁴ CFU/mL of AmpC-E in urine cultures. Exclusion criteria were pre-antibiotic defervescence, alternative diagnoses, or polymicrobial growth ( ≥ 3 species). Patients were classified into two groups. The discordant group received initial antibiotics considered ineffective for AmpC-E, such as penicillins or first to third generation cephalosporins. The concordant group received antibiotics generally recommended for AmpC-E, such as aminoglycosides or fourth generation cephalosporins or carbapenems. The primary outcome was time to defervescence. The secondary outcome was 30-day culture confirmed recurrence. Time to defervescence was analyzed by Kaplan–Meier curves.

Of 135 cases identified, 76 met the inclusion criteria. The discordant and the concordant groups were 41 and 35, respectively. Median age was 6 months (IQR 4.8–16.8) in the discordant group and 8.4 months (IQR 4.8-33.6) in the concordant group. Common pathogens included Enterobacter cloacae (30%) and Klebsiella aerogenes (26%). Catheter-associated UTIs were 33%. Median time to defervescence was 16.9 hours (95% CI 5.8–28.0) in the discordant group and 20.5 hours (95% CI 17.1–23.9) in the concordant group, with no significant difference (p = 0.998) (Figure 1). 30-day culture confirmed recurrence occurred in one case each group.

In pediatric UTIs caused by AmpC-E, initial discordant therapy was not associated with adverse clinical outcomes, suggesting that unnecessary use of broad-spectrum antibiotics may be avoided in empirical treatment.

All Authors: No reported disclosures