Endothelial dysfunction assessed by the endothelial activation and stress index (EASIX) predicts risk of mortality in chronic heart failure patients

Endothelial dysfunction accompanies chronic heart failure (CHF) but is hard to quantify in routine practice. The Endothelial Activation and Stress Index (EASIX), calculated from creatinine, lactate dehydrogenase, and platelets, predicts mortality and complications in conditions like allogeneic stem cell transplantation, COVID-19, and coronary artery disease. Aim To evaluate if EASIX is a prognostic biomarker in patients with CHF. Training (n = 1796) and validation cohorts (n = 1796) included all patients from the outpatients' CHF registry of the University of Heidelberg, Germany, with available laboratory parameters for EASIX calculations at first clinical presentation. Five-year overall survival was assessed by multivariable Cox regression adjusting for age, sex, New York Heart Association (NYHA) score, etiology of heart failure, heart failure category, and NT- proBNP as covariates. Fractional polynomials modeled non-linear relationships, and prognostic performance was evaluated using the Brier score and concordance index. EASIX moderately correlated with NT- proBNP, NYHA stage and left ventricular ejection fraction (LVEF), and associated with increased hazard of death in both cohorts (hazard ratio (HR) per log2 increase: training 1.51 (1.14-2.01), p < 0.01; validation 1.59 (1.25-2.01, p < 0.001). The effect was consistent across all etiologies, classes and stages of heart failure as assessed by fractional polynomials. Pre-established EASIX cut-offs independently predicted increased risk of mortality (cut-off 2.32: training: HR 2.14 (1.52-3.02), p < 0.001; validation: HR 3.34 (1.88-5.93), p < 0.0001). Both models (continuous and discrete EASIX values) were validated in the validation cohort using integrated Brier score and C-index. EASIX is a novel biomarker to predict risk of mortality in patients with CHF.