Uropathogens and antibiotic resistance in culture-positive urine samples from children: a six-year retrospective analysis of resistance trends at a tertiary hospital

Abstract Background Urinary tract infections (UTIs) are among the most common bacterial infections of childhood, and rising antibiotic resistance complicates empirical therapy. Using culture-positive urine specimens, we evaluated the distribution of uropathogens, their phenotypic antibiotic resistance, and six-year changes in resistance at a tertiary hospital. Methods In this retrospective cross-sectional study, records of children younger than 18 years with a positive urine culture between January 2019 and December 2024 were analysed. The analysis was specimen (isolate)-based; a proxy patient-level sensitivity analysis (patients approximated by date of birth and sex, first isolate retained) was also performed. Identification and susceptibility testing used an automated system (Vitek 2 Compact) interpreted by the CLSI criteria in force each year; intermediate isolates were classified as resistant. Following a data audit, duplicate antibiogram entries were resolved to a single interpretation per specimen-antibiotic combination. Extended-spectrum beta-lactamase (ESBL) production was determined phenotypically and analysed only among isolates with a valid recorded result. Yearly trends were examined with the Cochran–Armitage test with false-discovery-rate adjustment and were interpreted only for antibiotics with consistently high testing coverage; associations with ESBL positivity and multidrug resistance (MDR) were assessed by logistic regression adjusted for sex, age, year and care setting. Results A total of 10,348 isolates were analysed; the specimens originated predominantly from inpatient wards (57.5%) and the emergency department (32.9%). Escherichia coli was the most frequent pathogen (58.3%), followed by Klebsiella spp. (18.1%) and Proteus spp. (6.7%). In E. coli, resistance was highest to ampicillin (62.5%, 95% CI 61.3–63.7) and lowest to imipenem (0.3%), meropenem (0.4%) and amikacin (1.6%). Phenotypic ESBL positivity was 30.7% among the 3,879 E. coli isolates (64.3% of all E. coli) with a valid recorded ESBL result. Among antibiotics with consistently high testing coverage, resistance rose from 2019 to 2024 for ciprofloxacin (14.7% to 44.5%) and ampicillin (58.8% to 68.9%) (adjusted p < 0.001), whereas amikacin and carbapenem resistance remained low and stable. Because ESBL testing coverage increased from 4.5% to 98.7% of E. coli isolates over the period, the temporal course of ESBL positivity could not be evaluated reliably. The MDR rate in E. coli was 42.1% (52.9% in 2024); male sex, the 12–60-month age group, calendar year and inpatient setting showed adjusted associations with ESBL positivity and MDR. Results were materially unchanged in the proxy patient-level analysis. Conclusions In this predominantly hospitalized and referred pediatric population, resistance to ampicillin and trimethoprim-sulfamethoxazole was high, whereas carbapenems, amikacin and nitrofurantoin retained low resistance, and ciprofloxacin resistance increased markedly. These laboratory surveillance findings are not generalisable to community settings; treatment decisions should follow clinical guidelines and local antimicrobial stewardship policies informed by regularly updated, setting-specific antibiograms.