The Urinalysis in Infant UTI: How Accurate Is It?

Diagnostic Accuracy of the Urinalysis for Urinary Tract Infection in Infants <3 Months of Age

Schroeder AR, Chang PW, Shen MW, Biondi EA, Greenhow TL
Pediatrics. 2015;135:965-971

Study Summary

The sensitivity of a urinalysis for diagnosing urinary tract infection (UTI) is not optimal—75%-85%, depending on the study. Although the American Academy of Pediatrics (AAP) recommended in 2011 that practitioners diagnose UTIs only when both abnormal urinalysis and a positive urine culture are present, in order to avoid overtreating asymptomatic bacteriuria, it is not clear that practitioners hold to that standard.^[1]

This was a study of infants with UTI and bacteremia from the same organism, collected from 20 participating hospitals from 1998 to 2013. The goal was to calculate the sensitivity and specificity of the urinalysis in these patients, who were considered to have definite UTIs on the basis of bacteremia with the same organism. The infants younger than 3 months of age. The study excluded children with major comorbid conditions that would alter the outcome, those cared for in the intensive care unit, and those with genitourinary or venous catheters at the time of the urine and blood samples.

To standardize measures across the many hospitals, the investigators grouped urinalysis findings in terms of white blood cells per high-powered field (WBC/hpf) into groups of 0-3, 4-10, 11-20, 21-50, and > 50 WBC/hpf. Consistent with AAP guidelines, they considered a urinalysis positive for pyuria if there were > 3 WBC/hpf along with bacteria visible in the urine specimen or any urine sample with > 10 WBC/hpf. To calculate specificity, the investigators used a comparison group of 115 infants who had negative urine cultures but who had urinalyses performed.

A total of 245 infants had both bacteremia and UTI. The median age of the infants was 37 days, and 60% of these infants were boys compared with 50.4% of the infants with negative cultures, a difference that did not reach statistical significance. Escherichia coli comprised 91% of the isolates, followed by Enterobacter species (3.7%) and Klebsiella species (2.5%). All other isolates were found < 1% of the time. Leukocyte esterase was positive in large or at least moderate amounts in 89% of these infants, demonstrating trace positivity in 8.6% of samples. By comparison, nitrites were positive in only 39.5% of the infants with both bacteremia and UTI. Almost all (96%) of the infants with bacteremia and UTI had > 3 WBC/hpf, with approximately 81% having > 10 WBC/hpf.

With respect to sensitivity of single urinalysis measures, leukocyte esterase had the highest sensitivity at 97.6%, but nitrates alone had poor sensitivity (39.5%). Pyuria had a sensitivity of 98.3%, but this dropped to 81% if pyuria was defined as > 10 WBC/hpf. When looking at the sensitivity and specificity of aggregate components, the highest were found with any measure of pyuria plus leukocyte esterase or bacteria, all of which demonstrated sensitivities above 97%. However, specificities were only good for the aggregate measures that included leukocyte esterase, with specificities of < 65% for pyuria plus evidence of bacteriuria on microscopy. The investigators concluded that in young infants with UTI and bacteremia, the urinalysis is highly sensitive and specific.

Viewpoint

These investigators conceded that part of the reason for the high sensitivity and specificity findings could be the narrow spectrum of more severely ill infants included in this study (children with UTI and bacteremia, not just UTI). However, they also make a compelling case that previous studies may have included many children with asymptomatic bacteria, providing lower sensitivity measures for any urinalysis findings alone or in aggregate. Multiple sensitivities and specificities are quoted in the article, but the bottom line is this: Combinations on the urinalysis that include measures of pyuria plus at least trace leukocyte esterase generally produce highly sensitive and specific predictions. This can best be applied in situations where patients might have a positive culture but don't otherwise seem to have a true UTI, being either afebrile or not ill-appearing. These data would suggest that for some of those patients, it may be appropriate to not treat on the basis of urine culture results alone.

Abstract

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.