Abstract
OBJECTIVE: The direct impact on survival of lactate-to-albumin ratio (LAR) in acute aortic dissection (AAD) remains uncertain. METHODS: Medical electric records of 919 patients in the intensive care unit (ICU) who diagnosed with AAD were extracted from the MIMIC-IV database. These patients were subsequently divided into quartiles according to their LAR. Multivariable Cox regression incorporating restricted cubic spline terms was applied to examine how LAR relates to all-cause death at 30-, 90-, and 365-day intervals after ICU admission. To probe possible nonlinearities, we used smoothed curve analyses and fitted segmented (2-piece) linear models where appropriate. RESULTS: Kaplan-Meier curves showed progressively poorer survival among patients in the highest LAR quartile. Multivariable Cox proportional-hazards analysis shown the greatest quartile exhibited adjusted hazard ratios of 1.329 (95% confidence interval [CI], 1.118-1.579) at 30 days, 1.314 (95% CI, 1.130-1.527) at 90 days and 1.263 (95% CI, 1.096-1.455) at 365 days compared with the lowest quartile. Restricted cubic spline modeling revealed a positive association between increasing LAR and mortality risk in critically ill patients with AAD. Consistent patterns emerged across all clinically relevant subgroups. Smooth-curve fitting suggested a nonlinear relationship between LAR and mortality risk; subsequent piecewise linear regression identified inflection points for all-cause death at 30, 90, and 365 days of 0.769, 0.385, and 0.381, respectively. CONCLUSIONS: Even after multivariable adjustment, the LAR demonstrated a robust association with all-cause mortality at 30, 90, and 365 days in patients with AAD in the ICU.