Urine Dipstick for the Diagnosis of Urinary Tract Infection in Febrile Infants Aged 2 to 6 Months.

Abstract

Objective: Urine dipsticks can be performed at the point of care, whereas urinalysis requires laboratory analysis. We compared the accuracy of urine dipstick with urinalysis for the diagnosis of urinary tract infection (UTI) in febrile infants aged 2 to 6 months.

Methods: We performed a cross-sectional study of previously healthy infants aged 2 to 6 months who presented to one of 5 emergency departments with a temperature greater than or equal to 38.0 °C and had a catheterized urine culture obtained. We defined a UTI with a urine culture growing greater than or equal to 50 000 colony-forming units (CFUs) per milliliter of a single bacterial uropathogen. Using receiver operator characteristic (ROC) curve analysis to select the optimal urine white blood cell (WBC) cut point, we compared positive urine dipstick (≥1+ leukocyte esterase or positive nitrite) to dichotomized urine WBC count for the diagnosis of UTI.

Results: Of 9387 febrile infants who had a urine culture performed, 1044 (11%) had a UTI. Escherichia coli was the most common pathogen identified (923; 88.4%). The optimal urine WBC cut point was greater than or equal to 7 cells per high-power field (HPF). When compared with urine WBC count of greater than or equal to 7 cells per HPF, urine dipstick had a higher sensitivity (831/921 [90.2%] dipstick vs 738/880 [83.9%] urine WBC; difference 6.4%, 95% CI 3.8%-8.9%) and specificity (6352/6862 [92.6%] dipstick vs 3679/4231 [87.0%] urine WBC; difference 5.6%, 95% CI 4.7%-6.6%).

Conclusion: Urine dipstick is an accurate diagnostic test for UTI in febrile infants aged 2 to 6 months. Laboratory urinalysis may not be required to guide initial treatment decisions.