Over the past few years, we have published studies trying to determine the accuracy of pyuria for diagnosing a urinary tract infection (UTI) based on urine concentration (10.1542/peds.2016-2370, 10.1542/peds.2019-0467). Yet in both cases the sample sizes were either somewhat small or limited to infants. Nadeem et al studied a larger sample size of infants and toddlers <24 months of age (N=24,171 patients) over a 6-year period in a retrospective study of children who had both a urinalysis and a urine culture (10.1542/peds.2020-014068). The authors divided urine specific gravity into three categories: low (<1.011), moderate (1.011-1.020), and high (>1.020).
8.3% of those with urinalysis and culture had a culture-positive UTI, and the authors provide cut-offs of 3 white blood cells per high powered field for the low, 6 for the moderate, and 8 at high urine concentrations with strong likelihood ratios associated with each cutoff (10.5, 12, and 11.1 accordingly). Similarly, leukocyte esterase turns out to be a good indicator of UTI with likelihood ratios for even a small positive leukocyte esterase going from 25.2 to 33.1 to 37.6 across the three categories of concentration. So why is the positive leukocyte esterase an even stronger predictor of UTI than the pyuria thresholds identified (which are not shabby either)? The authors compare their work to prior studies in an interesting discussion section of their article. Go with the flow of results and analysis and take a peek at this interesting study that may help you make even more accurate diagnoses of UTIs in young children.
