Association between red blood cell distribution width-to-albumin ratio and all-cause mortality in intracerebral hemorrhage.

Background: Intracerebral hemorrhage (ICH) remains a devastating cerebrovascular condition, marked by high fatality and limited availability of prognostic tools. The red blood cell distribution width-to-albumin ratio (RAR) has recently gained attention as a composite biomarker of systemic inflammation and nutritional condition, but its prognostic value in ICH remains unclear. We aim to examine how RAR relates to mortality risk among individuals with ICH.

Methods: We performed a retrospective cohort analysis using the Medical Information Mart for Intensive Care-IV database. A total of 1,410 ICH individuals hospitalized in the intensive care unit were included and categorized into quartiles according to their RAR levels. The primary endpoint was all-cause mortality at 365 days, while 90-day all-cause mortality served as a secondary endpoint. Kaplan-Meier survival analysis, time-varying Cox regression model, and restricted cubic spline analyses (RCS)analysis were performed to assess the link between RAR and mortality risk. The predictive utility of RAR was further assessed through receiver operating characteristic (ROC)curve. Subgroup analyses explored potential effect modifications.

Results: Among the 1,410 ICH patients analyzed, the median age was 69 years. The all-cause mortality rates at 90-day and 365-day were 35.53 and 42.62%, respectively. Individuals with the highest RAR levels experienced significantly greater 90 days (54.34% vs. 21.97%,  < 0.001) and 365 days (62.18% vs. 29.77%,  < 0.001) than those with the lowest levels. Time-varying Cox regression model revealed that increased RAR levels were significantly and independently linked to greater mortality risk (hazard ratios [HR] for 365-day mortality:1.07, 95% CI:1.02-1.13,  = 0.005; HR for 90-day mortality: 1.14, 95%CI: 1.05-1.12,  = 0.001). ROC curve analysis demonstrated that combining RAR with the SOFA score improved predictive accuracy for 90-day and 365-day. RCS analyses indicated a nonlinear connection between higher RAR values and mortality rates. Subgroup analyses revealed that a largely uniform effect of RAR across different subpopulations except for age, gender, and race.

Conclusion: An elevated RAR is independently and significantly associated with increased all-cause mortality in ICH patients, regardless of established risk predictors. Its combination with the SOFA score enhances prognostic accuracy. These results suggest its potential clinical utility for early risk stratification.