P0523 - Urinary Tract Infections in Children: Accuracy of a Novel Dipstick Biomarker Test via Bagged Urine Sampling

Background: Absolute levels of urinary neutrophil gelatinase-associated lipocalin (uNGAL) appear highly accurate to identify urinary tract infections (UTIs) in young children when obtained via catheterization. Prior studies have neither evaluated the accuracy of uNGAL levels in bag (likely contaminated) specimens nor assessed the accuracy of a dipstick (potential point-of-care (POC)) test to detect UTIs.

Objective: To compare the overall accuracy and test characteristics of laboratory-based uNGAL, dipstick uNGAL, and POC urinalysis (UA) from catheter and bag specimens

Design/

Methods: Prospective cross-sectional study of otherwise healthy children less than 24 months of age evaluated for fever (≥ 38°C) and for whom catheterized urine studies were being obtained to evaluate for UTIs. We analyzed the overall accuracy and test characteristics of laboratory-based absolute uNGAL levels, dipstick uNGAL results (semi-quantitative), and dipstick UA results (semi-quantitative). The UTI reference standard was defined as >100,000 CFUs/mL of a pathogen in culture from catheterized specimens. We previously identified a potentially optimal cutoff level for absolute uNGAL of 39.1 ng/mL for the diagnosis of UTI in this population. The dipstick uNGAL has built-in color-change thresholds; we used the >50 ng/mL dipstick mark as it was closest to our previously established 39.1 ng/mL threshold.

Results: 172 patients with urine obtained via catheterization were included in the analysis; 101 patients had both bag and catheter samples obtained. 18/172 (10.5%) had UTIs. Table 1 notes patient characteristics. Median uNGAL levels from catheterized specimens were 206.8 (IQR 91.6, 1000.0) and 0.01 ng/mL (0.01, 11.8) in the UTI and culture-negative groups, respectively. The area under the receiver-operating characteristic curve for uNGAL from a catheterized sample was 0.95 (95% CI: 0.87-1.0) as compared to 0.93 (95% CI: 0.87, 1.0) from a bagged sample. Table 2 notes the test characteristics for absolute uNGAL levels, dipstick uNGAL results and POC UA results. The dipstick uNGAL test and absolute uNGAL levels from catheterized samples had higher sensitivities compared to POC UA at typically used LE thresholds (e.g. ≥1+). uNGAL from bag specimens had lower specificities than from catheterized samples, similar to POC UA.

Conclusion(s): uNGAL had excellent test characteristics from catheterized samples. However, absolute and dipstick uNGAL test characteristics did not show clear benefit compared to POC UA.