Pyuria: Still a Valuable Component of Urinalysis

Shaikh and colleagues have published a study assessing the accuracy of screening tests for the diagnosis of urinary tract infections (UTI) infections in young children.1 They evaluated the urine of 4188 children for bacteriologic culture and a variety of measures of inflammation. They concluded that for febrile children <36 months of age who undergo bladder catheterization for suspected UTI, pyuria will be absent in ∼20% of children shown to have pure growth of a pathogen (>50 000 cfu/ml) on culture. They extrapolate from this observation that currently available methods for detecting inflammation in urine lead to a misdiagnosis in 20% of children and therefore challenge the appropriateness of requiring pyuria for the diagnosis of UTI.We question the number of 20%. Other sources project that the number of children with a positive urine culture and a bland UA is between 9 and 12%.2 Even a close look at the data presented in supplemental information Table 3, shows that trace leukocyte esterase is positive in 88% of all children subjected to evaluation (specificity 92%) and is positive in 89% of febrile children (specificity 92%) which characterized most children in this cohort.The authors acknowledge that two potential explanations for a positive urine culture in the absence of inflammation are asymptomatic bacteriuria and external contamination of the urine sample during acquisition. Although they estimate that both are unlikely contributors this is worthy of further discussion.Although the best way to determine the frequency of contamination of urines obtained by catheterization is when they are compared to suprapubic aspiration in the same patient, these studies are infrequent.3 However, especially in girls <6 months of age and in uncircumcised boys contamination of urines obtained by catheter occur 14% of the time.4 The complexity in assigning contamination as an explanation for a positive culture of 50 000 cfu/ml, is the assumption that contamination has occurred only when there is a low-density positive culture or a culture with mixed organisms. This strategy inappropriately dismisses any high colony count, single organism culture which we know may result from contamination. We can safely estimate the rate of contamination as at least 5% rather than 0.5%.The microbiology of UTI also influences the likelihood of pyuria. Most UTIs are caused by Escherichia coli and these infections can be expected to be accompanied by significant pyuria. Less often (10%-15%), UTIs are caused by enterococcus, Klebsiella or Pseudomonas species, and then, expectations must be adjusted as we now know these infections may generate less inflammation.5The decision to obtain a urine culture is usually made on clinical grounds – unexplained fever in a child <3 years. The decision for empirical therapy may be governed by the initial clinical appearance of the child with or without laboratory data. When empirical treatment has been initiated the clinician can interpret the results of the urinalysis and culture together the following day. The absence of pyuria may be explained by the infecting organism, by information related to acquisition of the urine or other clinical features that are now present explaining the fever.If empirical therapy has not been started, the child may now be improved,2 repeat urine samples may be helpful in clarifying the issue or there may be alternate explanations for the fever. We agree with the authors that the bulk of information to date, including their study, shows that although the tests are not perfect, the presence or absence of pyuria is very useful.