Objectives

Urinary tract infection (UTI) is common among older women. However, diagnosis is challenging because of frequent chronic lower urinary tract symptoms, cognitive impairment, and a high prevalence of asymptomatic bacteriuria (ASB). Current urine diagnostics lack specificity, leading to unnecessary treatment and antimicrobial resistance. This study aimed to evaluate the diagnostic accuracy of 12 urine biomarkers for diagnosing UTI in older women.

Methods

In this case-control study, cases were women ≥65 years with ≥2 new-onset lower urinary tract symptoms, pyuria, and one uropathogen ≥104 CFU/mL. Controls were asymptomatic and classified as ASB (one uropathogen ≥105 CFU/mL), negative culture, or mixed flora. Urine biomarker concentrations were measured through liquid chromatography-mass spectrometry and ELISA. Diagnostic accuracy parameters of individual biomarkers and a biomarker model were derived from receiver operating characteristic curves.

Results

We included 162 community-dwelling and institutionalized older women. Five urine inflammatory biomarkers demonstrated high discriminative ability (area under the curve ≥0.80): interleukin 6, azurocidin, neutrophil gelatinase-associated lipocalin, tissue inhibitor of metalloproteinases 2, and C-X-C motif chemokine 9. Azurocidin exhibited the highest diagnostic accuracy (sensitivity 86% [95% CI 75%-93%] and specificity 89% [95% CI 82%-94%] at 16.7 ng/mmol creatinine). A combined biomarker and pyuria model showed improved diagnostic accuracy in patients with UTI and ASB, compared with pyuria alone.

Discussion

We identified several urine biomarkers that accurately differentiated older women with UTI from asymptomatic women, including ASB. These findings represent a potential advancement towards improved diagnostics for UTI in older women and warrant validation in a diverse population.

Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved.