UTI is a common and important clinical problem in infants and young children, with a prevalence of 5.3% among febrile infants seen in our Emergency Department. White females with rectal temperature > or = 39 degrees C are at particularly high risk (prevalence, 17%). Several studies have highlighted the limitations of the standard urinalysis for identifying UTI in infants and young children and have recommended performance of both urinalysis and urine culture. Alternative methods such as dipstick urinalysis, although attractive because of ease of performance, are inadequate as a screen for UTI. Hemocytometer WBC counts of an uncentrifuged urine specimen can be performed in an office or hospital-based laboratory with minimal training. Performance of Gram-stained smears, however, is most appropriate for the hospital-based laboratory. In the hospital setting where both tests can readily be performed, the positive predictive value of the combination of pyuria and bacteriuria (85%) allows prompt institution of antimicrobial therapy before culture results are available, whereas the lower positive predictive value of the single finding of either pyuria or bacteriuria (40%) justifies delaying treatment decisions until culture results are available. In the office setting where hemocytometer counts can easily be performed, culturing only specimens with pyuria and those of children presumptively treated with antimicrobials will result in the identification of almost all patients with true UTI, sparing large health care expenditures. Although the urine culture is traditionally regarded as the gold standard of UTI, positive urine cultures may occur secondary to contamination or in cases of ABU, leading to a false diagnosis of UTI. In contrast we found pyuria to be a reliable marker to discriminate infection from colonization of the urinary tract. The sustained absence of an inflammatory response, on repeat UA within 24 h, constitutes strong evidence that infection is absent. Management of ABU is controversial; many experts recommend withholding antibiotics because eradication of low virulence organisms may be followed by colonization with more virulent species that cause pyelonephritis. Preliminary results of our ongoing treatment trial suggest that management of young febrile children with UTI as outpatients receiving oral cefixime is as efficacious as inpatient management with intravenous cefotaxime. Results of renal ultrasound and DMSA scan at the time of infection have not modified management in any patient. Accordingly selective rather than routine performance of ultrasound is recommended. A voiding cystourethrogram at 1 month and a DMSA scan 6 months later have been valuable in identifying patients with vesicoureteral reflux and renal scarring, respectively. Among patients initially identified as having acute pyelonephritis, the incidence of renal scarring at 6 months has been substantially more frequent (approximately 40%) than we had expected. However, the long term implications of small scars identified with renal scintigraphy remain to be determined.