Wen Bai, Manli Tao, Shaofei Li, Wei Shi, Kun Jiang, Ling Tian
LnEASIX is a simple, robust, and continuous predictor of short- and long-term mortality in critically ill patients with ischemic stroke. Integrating this zero-cost biomarker with established intensive care scoring systems provides significant incremental value for early prognostic risk stratification.
BACKGROUND: Systemic endothelial dysfunction is a critical driver of poor outcomes in ischemic stroke. We evaluated the prognostic value of the endothelial activation and stress index (EASIX) in critically ill stroke patients.
METHODS: We analyzed 3063 patients with ischemic stroke from the MIMIC-IV database. The natural log-transformed EASIX (lnEASIX) was calculated from admission LDH, creatinine, and platelets. The primary outcome was designated as 28-day mortality, with secondary outcomes including in-hospital, 90-day, and 180-day mortality. Multivariable logistic and Cox proportional hazards regression, restricted cubic splines (RCS), and advanced clinical utility analyses (NRI, IDI, and decision curve analysis) were utilized.
RESULTS: Evaluated as a continuous variable, a per-unit increase in lnEASIX was independently associated with an elevated risk for the primary endpoint of 28-day mortality (OR 1.47, 95% CI 1.36-1.59, p < 0.001), as well as all secondary mortality endpoints. RCS analysis confirmed a strictly linear, monotonic dose-response relationship (p for nonlinearity > 0.05). While lnEASIX demonstrated moderate standalone discrimination (AUC 0.64-0.67, comparable to SOFA), integrating it with the baseline SOFA score yielded significant incremental predictive value (NRI 0.208, IDI 0.011; both p < 0.01) and superior net clinical benefit. Subgroup analysis revealed consistent prognostic utility across most strata, though the impact was markedly more pronounced in patients without pre-existing cancer (p for interaction < 0.05).
CONCLUSION: LnEASIX is a simple, robust, and continuous predictor of short- and long-term mortality in critically ill patients with ischemic stroke. Integrating this zero-cost biomarker with established intensive care scoring systems provides significant incremental value for early prognostic risk stratification.