Association between Endothelial Activation and Stress Index and 28-day mortality among critically III patients with infective endocarditis : a retrospective cohort analysis

Objective The present study aimed to assess the relationship of the Endothelial Activation and Stress Index (EASIX) and 28-day mortality in critically ill patients with infective endocarditis (IE). Methods We designed a retrospective cohort study using data of patients admitted to intensive care units and identified as IE from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Routine laboratory biomarkers including lactate dehydrogenase (LDH), serum creatinine (Cre) and platelet counts were collected to measure EASIX scores. Participants were divided into tertiles on the basis of log2 transformed EASIX, and the relationship of which with 28-day mortality was explored using multivariable Cox regression models, Kaplan–Meier curves, smooth curve fitting, as well as subgroup analyses. Results A total of 508 participants were enrolled for final analyses, and the primary endpoint of 28-day mortality was reached in 112 patients (22.05%). As a continuous variable, a higher Log2(EASIX) level was significantly related to an increased 28-day mortality risk (HR, 1.21; 95% CI, 1.06, 1.37, P  = 0.004). As a categorical variable, the higher risks of 28-day mortality among those with higher Log2(EASIX) levels in Tertile 2 (HR, 1.21; 95% CI, 1.06, 1.37, P  = 0.004) and Tertile 3 (HR, 1.21; 95% CI, 1.06, 1.37, P  = 0.004) were observed compared with participants with lower Log2(EASIX) level in Tertile 1. Kaplan–Meier curves revealed that participants with higher Log2(EASIX) score were with lower survival rates. Smooth curve fitting showed a linear association between Log2(EASIX) and 28-day mortality. No interaction between Log2(EASIX) and any potential variables was found through subgroup analyses. Conclusion Higher EASIX scores were significantly associated with increased risk of 28-day mortality in ICU patients with IE. Clinical trial number: Not applicable.