Another Interpretation of These Data

We read with great interest the article by Shaikh et al regarding the accuracy of screening tests for diagnosing urinary tract infections (UTIs) in young children.1 While we commend the authors for addressing this important issue, we believe that the interpretation of their findings merits further scrutiny as their conclusions may be misleading.The authors report a sensitivity of 85% for urinalysis (UA) when considering pyuria alone. However, several aspects of their methodology and data interpretation warrant critical examination. First, the authors included approximately 20% of children without fever, despite the well-established association between fever and UTIs in this age group. The inclusion of children with nonspecific symptoms likely introduced a significant number of children with asymptomatic bacteriuria, potentially skewing their sensitivity analysis.The decision to obtain a urine sample was left to practitioner discretion, and the authors describe all included children as “symptomatic.” However, in young infants, the indications for obtaining a urine sample often involve vague symptoms. Such nonspecific clinical presentations increase the likelihood of identifying asymptomatic bacteriuria. This factor could explain the bulk of the reported false negatives, as positive urine cultures in such cases may not reflect true UTIs and will not result in pyuria.Importantly, the article emphasizes sensitivity without adequately contextualizing the negative predictive value (NPV), which is arguably more relevant for clinical practice. For instance, with a UA sensitivity of 81% (leukocyte esterase ≥1+), a specificity of 95%, and a prevalence of 9.7%, the NPV is 97.9%. Although we cannot calculate it from this paper’s data, including WBC count in the urinalysis (in addition to leukocyte esterase) could further increase the NPV. These findings support the reliability of a negative UA considering only pyuria in ruling out UTI, even while acknowledging the inherent limitations of sensitivity alone. The authors’ report of 19% negative leukocyte esterase tests among UTI cases does not detail how many of these also had negative urinary WBC counts.Lastly, in a child with a high pretest probability (for example, an inadequately treated infant with ongoing symptoms of UTI), it is important to recognize that a negative test for pyuria does not definitively rule out the diagnosis. While pyuria is a key marker, it is not universally present in all cases of UTI. There is no perfect gold standard for diagnosing UTI, and diagnostic limitations mean that some infections may be missed if pyuria alone is relied upon.In summary, the findings presented here reiterate, rather than challenge, our current understanding of UTI diagnosis in young children. We believe their data affirm the utility of UA and its high NPV (using pyuria via leukocyte esterase and WBC counts) in ruling out UTI in the majority of patients. This study should not alter our current clinical practice of requiring both an abnormal UA and a confirmatory urine culture to diagnose UTI.