63 - Effect of Urine Concentration in Identifying Urinary Tract Infection in Toddlers
Friday, April 22, 2022
6:15 PM – 8:45 PM US MT
Poster Number: 63 Publication Number: 63.105
Jiwoong J. Shin, University of Texas Southwestern Medical School, Dallas, TX, United States; Shahid Nadeem, PM Pediatrics Urgent Care, Richardson, TX, United States; Halim Hennes, University of Texas Southwestern Medical School, Frisco, TX, United States; Jason Park, University of Texas Southwestern Medical School, Dallas, TX, United States
Peds ER Fellow University of Texas Southwestern Medical School Dallas, Texas, United States
Background: A screening urinalysis is often used to predict the results of the urine culture and to direct presumptive treatment. Recent studies have shown that the accuracy of pyuria in predicting positive urine culture changes with urine concentration. However, these studies have all been limited to children under 24 months of age.
Objective: The objective of this study was to investigate the effect of urine concentration on pyuria in identifying positive urine cultures in children 2-5 years of age. The study also looked to determine the fraction of positive infection in children 2-5 years of age with suspected urinary tract infection (UTI).
Design/Methods: This was a retrospective cross-sectional study of children between 2-5 years of age who were evaluated in the emergency department for suspected UTI with a paired urinalysis and urine culture. Urine specific gravity was used as a surrogate marker for urine concentration. Patients were assigned to 3 different urine specific gravity (SG) groups (low < 1.011, moderate 1.011 to 1.020, and high >1.020). We calculated the test characteristics including likelihood ratios for pyuria cutoffs at 3 urine SG concentrations. We also performed a subgroups analysis for males and females. The primary outcome is positive urine culture defined as growth of a single urinary pathogen of >50,000 CFU/ml on a catheterized specimen or > 100,000 CFU/ml on a clean catch specimen.
Results: A total of 6,586 patients were analyzed, with a median age of 3.02 years (2.44-3.88 years). Urine culture was positive in 577 (8.8%) patients. Of the 577 positive cases, 81 (14%) were males, while 496 (86%) were females. E. coli made up 85.9% of cases in females, and 46.7% of cases in males. In males, LR+ (positive likelihood ratio) to predict positive urine culture decreased from 13.7 at low SG, 11.9 at moderate SG, and 10.4 and high specific gravity for > 5 WBCs per HPF. In females, LR+ decreased from 6.8 at low SG, 6.5 at moderate, and 6.3 for > 5 WBCs per HPF. Conclusion(s): The incidence of UTI in our study population was 8.8%. Pyuria cut points in predicting positive urine culture changed with urine concentration. The study also demonstrated that in the 2-5 year age group, females need a higher WBCs per HPF cutoff to have the same predictive value as males in identifying urinary tract infections. Patient Characteristics Sensitivty, Specificty, LR+ and LR- of Microscopic Pyuria, and LE by Urine SG Group