RESUMO
BACKGROUND: Urinary tract infections (UTI) are the most common bacterial infections among infants and young children with fever without a source. Extended-spectrum ß-lactamases (ESBLs) have emerged as emerging cause of UTI globally; however, data about risk factors and clinical features of children with ESBL-UTI have been scarce. OBJECTIVE: To describe the predisposing risk factors, clinical and microbiologic features associated with pediatric UTIs caused by ESBL-producing bacteria (ESBL-PB). METHODS: Our nested case-control study ran from January 1, 2012 to December 31, 2016. Pediatric patients with ESBL-PB UTI were compared with patients with non-ESBL-PB UTI matched for age and year of diagnosis. RESULTS: A total of 720 children were enrolled (240 cases and 480 controls). Patients with ESBL-PB UTI were more likely to have a history of prior intensive care unit (ICU) admission (22.5% vs. 12.3%, P < 0.001), at least one underlying comorbidity (19.2% vs. 5.8%, P < 0.001), prior hospitalization (47.1% vs. 32.9%, P < 0.001), exposure to a cephalosporin antibiotic within 30 days before culture (7.5% vs. 4.2%, P = 0.035), and to have cystostomy (7.9% vs. 1.5%, P < 0.001) compared with those with non-ESBL-PB UTI. Patients with ESBL-PB UTI were more likely to present with hypothermia (48.8% vs. 38.5%, P = 0.009); had significantly longer average hospital stays {8.7 days [95% confidence interval (CI): 3.2-14.3] vs. 4.0 days (95% CI: 2.5-5.5)} and were more likely to be admitted to the ICU [odds ratio (OR) 1.8; 95% CI: 1.1-2.9). Multivariate analysis determined that only having cystostomy (OR 3.7; 95% CI: 1.4-9.4] and at least one underlying comorbidity (OR 2.4; 95% CI: 1.3-4.3) were the independent risk factors for ESBL-PB UTI. All ESBL-PB isolates tested against meropenem were susceptible, majority were resistant to multiple non-beta-lactam antibiotics. CONCLUSIONS: Children with underlying comorbidities and cystostomy are at higher risk for ESBL-PB UTI, but majority of ESBL cases were patients without any known risk factors. Clinical signs/symptoms and commonly used biochemical markers were unreliable to differentiate cases caused by ESBL-PB from those caused by non-ESBL-PB. Further research is needed to elucidate the conditions most associated with ESBL-PB UTIs among children to properly guide empirical therapy in patients at-risk for these infections, to improve the outcomes, and finally, to determine strategies for rational antimicrobial use.