Longitudinal urinary 8-oxoGuo as a potential biomarker of persistent risk during ICU and post-ICU hospitalization.

Abstract

Persistent biological injury may remain undetected despite apparent clinical stabilization during critical illness. We evaluated whether serial urinary markers of RNA and DNA oxidation were associated with 90-day mortality, transition from ICU to post-ICU phase, and mortality after ICU discharge, and whether they provided prognostic information beyond baseline illness severity. This single-center longitudinal cohort included 438 critically ill adults who provided 2,516 urine samples during ICU and post-ICU hospitalization. Urinary 8-oxo-7,8-dihydroguanosine (8-oxoGuo) and 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodGuo) were quantified by isotope-dilution ultra-performance liquid chromatography-tandem mass spectrometry and normalized to creatinine. Bayesian joint models were adjusted for age, sex, Charlson Comorbidity Index, and baseline Sequential Organ Failure Assessment (SOFA) score. Each doubling of longitudinal urinary 8-oxoGuo was associated with higher 90-day mortality (hazard ratio [HR], 1.47; 95% credible interval [CrI], 1.17-1.85), higher mortality after ICU discharge (HR, 1.52; 95% CrI, 1.18-1.97), and a lower transition rate to post-ICU care (HR, 0.89; 95% CrI, 0.79-0.99). Associations were similar when APACHE II replaced SOFA, whereas longitudinal 8-oxodGuo showed no consistent associations. Adding baseline 8-oxoGuo to SOFA increased the area under the curve for 90-day mortality from 0.669 to 0.719 (p = 0.046), while incorporating longitudinal trajectory information increased the area under the curve for mortality after ICU discharge from 0.682 to 0.741 (p = 0.029). Overall, these findings indicate that higher longitudinal urinary 8-oxoGuo is associated with adverse outcomes across ICU and post-ICU hospitalization and provides incremental prognostic discrimination beyond baseline SOFA.