External Validation of the EARL-UTI Score and Exploratory Evaluation of a Procalcitonin-Modified Score for In-Hospital Mortality in Older Adults with Urinary Tract Infection

Background and Objectives: The EARL-UTI score was developed to estimate in-hospital mortality risk in older adults with urinary tract infection (UTI), but independent validation remains limited. This study aimed to externally validate the EARL-UTI score and to explore a procalcitonin-modified version (PEARL-UTI) in which procalcitonin replaced albumin. Materials and Methods: This single-center retrospective cohort study included hospitalized patients aged ≥65 years who presented to a tertiary emergency department with UTI between January 2021 and December 2023. The primary outcome was all-cause in-hospital mortality. Missing predictor values were addressed using multiple imputation. Discrimination was quantified using the area under the receiver operating characteristic curve (AUC) with 95% confidence intervals (CIs). Calibration, performance at the ≥4-point threshold, bootstrap-based optimism correction, and exploratory decision curve analysis (DCA) were also assessed. Results: Among 158 patients, 37 (23.4%) died in hospital. EARL-UTI showed good discrimination (AUC, 0.819; 95% CI, 0.721–0.888). At ≥4 points, sensitivity was 53.5%, specificity 89.8%, and accuracy 81.3%. PEARL-UTI, incorporating a three-point procalcitonin component at ≥2.72 ng/mL, showed numerically higher discrimination (AUC, 0.910; 95% CI, 0.823–0.957), with a sensitivity of 81.1%, a specificity of 86.7%, and an accuracy of 85.4%. Its optimism-corrected AUC was 0.901. Exploratory bootstrap-corrected DCA showed higher estimated net benefit for PEARL-UTI than EARL-UTI and both reference strategies across threshold probabilities of 5–97%. Conclusions: EARL-UTI demonstrated good discrimination in this independent cohort of hospitalized older adults and a high-specificity threshold for in-hospital mortality. Replacing albumin with procalcitonin yielded the exploratory PEARL-UTI score, which showed numerically higher discrimination and greater sensitivity in this cohort. PEARL-UTI should nevertheless be considered exploratory until independently validated in larger, multicenter populations.