Urine biomarkers individually and as a consensus model show high sensitivity and specificity for detecting UTIs
Marzieh Akhlaghpour,Emery Haley,Laura Parnell,Natalie Luke,Mohit Mathur,Richard A Festa,Michael Percaccio,Jesus Magallon,Mariana Remedios-Chan,Alain Rosas,Jimin Wang,Yan Jiang,Lori Anderson,David Baunoch
DOI: https://doi.org/10.1186/s12879-024-09044-2
2024-01-31
Abstract:Background: Current diagnoses of urinary tract infection (UTI) by standard urine culture (SUC) has significant limitations in sensitivity, especially for fastidious organisms, and the ability to identify organisms in polymicrobial infections. The significant rate of both SUC "negative" or "mixed flora/contamination" results in UTI cases and the high prevalence of asymptomatic bacteriuria indicate the need for an accurate diagnostic test to help identify true UTI cases. This study aimed to determine if infection-associated urinary biomarkers can differentiate definitive UTI cases from non-UTI controls. Methods: Midstream clean-catch voided urine samples were collected from asymptomatic volunteers and symptomatic subjects ≥ 60 years old diagnosed with a UTI in a urology specialty setting. Microbial identification and density were assessed using a multiplex PCR/pooled antibiotic susceptibility test (M-PCR/P-AST) and SUC. Three biomarkers [neutrophil gelatinase-associated lipocalin (NGAL), and Interleukins 8 and 1β (IL-8, and IL-1β)] were also measured via enzyme-linked immunosorbent assay (ELISA). Definitive UTI cases were defined as symptomatic subjects with a UTI diagnosis and positive microorganism detection by SUC and M-PCR, while definitive non-UTI cases were defined as asymptomatic volunteers. Results: We observed a strong positive correlation (R2 > 0.90; p < 0.0001) between microbial density and the biomarkers NGAL, IL-8, and IL-1β for symptomatic subjects. Biomarker consensus criteria of two or more positive biomarkers had sensitivity 84.0%, specificity 91.2%, positive predictive value 93.7%, negative predictive value 78.8%, accuracy 86.9%, positive likelihood ratio of 9.58, and negative likelihood ratio of 0.17 in differentiating definitive UTI from non-UTI cases, regardless of non-zero microbial density. NGAL, IL-8, and IL-1β showed a significant elevation in symptomatic cases with positive microbe identification compared to asymptomatic cases with or without microbe identification. Biomarker consensus exhibited high accuracy in distinguishing UTI from non-UTI cases. Conclusion: We demonstrated that positive infection-associated urinary biomarkers NGAL, IL-8, and IL-1β, in symptomatic subjects with positive SUC and/or M-PCR results was associated with definitive UTI cases. A consensus criterion with ≥ 2 of the biomarkers meeting the positivity thresholds showed a good balance of sensitivity (84.0%), specificity (91.2%), and accuracy (86.9%). Therefore, this biomarker consensus is an excellent supportive diagnostic tool for resolving the presence of active UTI, particularly if SUC and M-PCR results disagree.