OBJECTIVE: This study evaluates the predictive value of the lactate/albumin ratio (LAR) for all-cause mortality in cirrhosis patients. DESIGN: Retrospective observational study. SETTING: Intensive care unit (ICU). PATIENTS OR PARTICIPANTS: 626 first-time ICU-admitted cirrhosis patients in the USA (MIMIC-IV v2.2). INTERVENTIONS: None. MAIN VARIABLES OF INTEREST: LAR index, 28-day, and 90-day all-cause mortality. RESULTS: Of 626 patients (60.86% male), 27.80% and 39.14% died within 28 and 90 days, respectively. Multivariate Cox analysis showed a significant association between higher LAR and mortality. Adjusted for confounders, elevated LAR increased the 28-day mortality risk [HR: 1.31 (1.21-1.42), P <
0.001]. A restricted cubic spline analysis revealed non-linear relationships between LAR and mortality. For 28-day mortality, the inflection point was 1.583: below this, HR was 2.29 (95% CI: 1.61-3.27, P <
0.001)
above, HR was 1.16 (95% CI: 1.02-1.31, P = 0.021
P = 0.002). For 90-day mortality, the inflection point was 1.423: below, HR was 1.60 (95% CI: 1.04-2.47, P = 0.033)
above, HR was 0.94 (95% CI: 0.75-1.16, P = 0.542
P = 0.012). CONCLUSIONS: LAR predicts 28-day and 90-day mortality with a segmented effect. An LAR ≥1.583 signals high 28-day mortality risk, necessitating intensified monitoring and potential ICU admission. For 90-day mortality, LAR near 1.423 serves as an early warning for high-risk patients and guides interventions. Continuous LAR monitoring aids management, but prospective studies are needed to confirm clinical utility.