Abstract
Acute respiratory distress syndrome (ARDS) carries high in-hospital mortality. We evaluated whether a laboratory-based frailty index (FI-Lab) at ICU admission is associated with in-hospital death in ARDS. We conducted a retrospective cohort study using MIMIC-IV v3.1. FI-Lab was calculated from routine laboratory tests and vital signs at ICU admission. We applied 1:1 propensity score matching and assessed associations using multivariable logistic regression; survival across FI-Lab tertiles was compared using Kaplan-Meier and Cox models. Among 859 ARDS patients, 332 (38.6%) died in hospital. Non-survivors had higher FI-Lab than survivors (0.59 [0.51-0.68] vs 0.54 [0.48-0.62], P < .001). After propensity score matching, 262 matched pairs were well balanced. FI-Lab remained independently associated with in-hospital mortality in adjusted models (odds ratio 1.26 per 0.1 FI-Lab increase; 95% CI 1.14-1.40; P = .003). Compared with the lowest FI-Lab tertile, the highest tertile had increased mortality risk (hazard ratio 1.98; 95% CI 1.52-2.57; P < .001). Higher FI-Lab at ICU admission was associated with increased in-hospital mortality in ARDS. FI-Lab may be a promising routinely available marker for early risk stratification; external validation and prospective studies are warranted before clinical implementation.