UTICalc tool validated for diagnosing pediatric UTI

The UTICalc tool has strong diagnostic performance for detecting urinary tract infection (UTI) in febrile children aged 2 to 24 months, according to a recent study published in JAMA Network Open.¹

Fourteen percent of child emergency department (ED) visits have been linked to febrile illness, with up to 7% of these cases identified as UTIs.² However, nonspecific symptoms and difficulty obtaining clean urine samples have been reported as challenges toward diagnosing these conditions in preverbal and pre–toilet-trained children.¹

“Accurate diagnosis is critical; missed UTIs risk potential progression to urosepsis and may have long-term consequences such as kidney scarring, while overuse of antibiotics contributes to antimicrobial resistance,” wrote investigators.

Assessing the UTICalc tool

The UTICalc tool was developed to improve UTI diagnosis in this population. Investigators conducted a multicenter prospective diagnostic study to validate version 3.0 of this tool in children aged 2 to 24 months. The study was conducted at 2 tertiary care pediatric EDs in Canada.

Participants included children aged 2 to 24 months presenting with fever, defined as a temperature of 38.0 °C or greater. Those with known congenital kidney or urinary tract abnormalities, oral or intravenous antibiotic use, immunosuppression, insurmountable language barrier, or prior study enrollment were excluded.

Treating clinicians and caregivers were blinded to final UTI status. The clinical team made decisions about urinalysis, empiric antibiotic treatment, urine culture, and ED disposition.

Data was obtained through a caregiver-reported electronic data collection form and included sex, age, circumcision status, fever duration, and maximum recorded temperature. Prior UTI history was also reported. The fever source was determined through a standardized form completed by the treating team.

Defining UTI and participant demographics

UTI-positive cases were defined by both a urinalysis showing a leukocyte esterase greater than trace on dipstick or greater than 5 white blood cells per high-power field on microscopy and the growth of a uropathogen at 10 × 10⁷ colony-forming units/L or greater as displayed in urine culture. UTI status was also determined in the follow-up questionnaire.

There were 2561 participants included in the final analysis, 47% of whom were female, 64% younger than 12 months, and 4% classified as having a UTI. Of the 111 UTI cases, 105 were diagnosed based on the UTI-positive case definition, 2 were reported in the follow-up questionnaire, and 4 were adjudicated as UTI-positive cases.

Children with UTI were significantly more likely to be younger than 12 months than those without UTI, with rates of 64% and 36%, respectively. Of patients, 47% overall and 62% with UTI were female. For uncircumcised male patients, these rates were 34% and 35%, respectively, vs 19% and 3%, respectively, for circumcised male patients.

Model performance and results

Less than 1% of the cohort had missing data for a predictor variable in the clinical model. Urine dipstick results were available for 52% of patients with a predicted UTI risk above 2% based on clinical predictors.

Model performance was evaluated at 2% for the clinical model and 5% for both models. The clinical model displayed a high sensitivity of 96.4% and low specificity of 34.1% at the 2% threshold, alongside a positive predictive value and negative predictive value of 6.2% and 99.5%, respectively.

For the 5% threshold, these rates were 82%, 73.8%, 12.5%, and 98.9%, respectively, for the clinical model alone vs 94%, 86.9%, 47.5%, and 99.1%, respectively, when assessing the clinical and dipstick model together.

Both models showed strong discrimination, with an area under the receiver operating characteristic curve of 0.84 for the clinical model and 0.95 for the clinical and dipstick model. Clinical utility was also observed, with the greatest clinical advantage observed between 0.02 and 0.06.

Implications

This indicated a greater net benefit from the clinical model vs either the treat-all or treat-none strategies. Overall, the data supported the use of UTICalc when evaluating and managing UTI in young children.

“Future research should assess UTICalc performance across varying levels of clinician pretest suspicion and evaluate earlier integration into ED workflows,” wrote investigators.

References

1. Kinlin C, Gravel J, Barrowman N, et al. Diagnosing urinary tract infection in young febrile children in the emergency department. JAMA Netw Open. 2026;9(3):e261741. doi:10.1001/jamanetworkopen.2026.1741
2. Shaikh N, Morone NE, Bost JE, Farrell MH. Prevalence of urinary tract infection in childhood: a meta-analysis. Pediatr Infect Dis J. 2008;27(4):302-308. doi:10.1097/INF.0b013e31815e4122