Low-Level Pyuria and Antibiotic Prescribing for Suspected Urinary Tract Infection in Children

OBJECTIVE Our objective was to understand the impact of urinalysis results on clinician antibiotic prescribing for pediatric patients evaluated at ambulatory visits. We hypothesized that responses to low pyuria levels contribute disproportionately to overtreatment. METHODS This cross-sectional study combined electronic health record and abstracted symptom data from emergency department and urgent care visits to 4 Northern California settings. We sampled ambulatory visits by patients aged between 2 months and 17 years who underwent urinalysis between January 1, 2022, and December 31, 2023. We used multivariable logistic regression to assess relationships between urinalysis results, clinical covariates including age, and empiric antibiotic initiation. We estimated attributable impact of these relationships on overtreatment for patients in whom urinary tract infection (UTI) was excluded by culture criteria. We similarly estimated impact on treatment for those with culture-confirmed UTI. RESULTS Among 1150 patients with median age 6.3 years, 191 (16.6%) received antibiotics. Low pyuria levels (trace leukocyte esterase, 10-69 white blood cells [WBC]/mm,3 5-10 WBC/high-power field) were associated with antibiotic initiation with adjusted odds ratio 6.62 (95% CI 3.72–11.79) compared with no pyuria. The adjusted attributable fraction (aAF) for overtreatment at this level was 17.0% (95% CI 10.3%-23.2%), with aAF 1.8% (95% CI 1.1%-2.4%) for UTI treatment. Differential impacts on overtreatment and UTI treatment were consistent in age-stratified analyses. CONCLUSIONS To promote selective UTI treatment, clinicians should consider degree of pyuria in context with age, risk factors, symptoms, and other urinalysis results. Health system interventions should be designed to mitigate excess influence of low pyuria levels and promote integrated diagnostic assessments.