Abstract
Patients with septic myocardial injury have a relatively high risk of death. Early identification of patients with a high risk of death is crucial for improving prognosis. The ratio of red blood cell distribution width to albumin (RAR) has been found to be associated with poor prognosis in other diseases, but its role in septic myocardial injury remains unclear. This study aims to explore the relationship between RAR and short-term all-cause death in patients with septic myocardial injury. This study adopted a retrospective cohort study design, using two publicly available datasets, MIMIC-IV and eICU-CRD, as well as data from 107 patients who visited the First Affiliated Hospital of Xinjiang Medical University from January 2021 to October 2024. Pearson analysis was used to examine the association between RAR and SOFA and APS III scores. The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the ability of RAR to predict short-term mortality, and the cutoff values and other indicators of RAR in predicting outcome events in the public databases were compared. Then, according to the median of RAR, it was divided into Low and High groups. RAR was used as a continuous and categorical variable for regression analysis, and the original model, a model adjusted for factors such as age, and a model further adjusted for multiple confounding factors were constructed. Restricted cubic splines were used to study the association between RAR and short-term mortality. The Kaplan-Meier survival curve and COX risk regression model were combined to calculate the hazard ratio (HR) and 90% confidence interval (CI), and subgroup analysis was performed in multiple subgroups. This study included a total of three datasets. There were differences in the death status of patients in the three cohorts, but the data volume and trends were similar. High RAR patients showed differences in many aspects and had a higher short-term mortality rate. ROC curve analysis showed that RAR had a certain predictive value for the mortality rate of the three cohorts. Among them, in the MIMIC-IV and eICU-CRD cohorts, RAR was significantly positively correlated with SOFA and APS III scores (p < 0.001). In the three datasets, the Kaplan-Meier survival curve showed that the short-term mortality rate of patients in the high RAR group was higher than that in the low RAR group. The multivariate COX regression model indicated that RAR was associated with short-term death risk to varying degrees. After adjusting for factors, RAR in the MIMIC-IV and eICU-CRD cohorts had a nonlinear relationship with short-term all-cause mortality. Subgroup analysis showed that in specific cohorts, the relationship between RAR and different mortality types was stable and consistent in each subgroup. RAR can be used as a predictor of mortality in patients with septic myocardial injury, which is helpful for risk stratification and prognosis prediction. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1038/s41598-025-32791-4.