Prognostic significance of the estimated pulse wave velocity in critically ill patients with coronary heart disease: analysis from the MIMIC‑IV database
Yingzhen Gu,Xiaorong Han,Jinxing Liu,Yifan Li,Zuozhi Li,Wei Zhang,Naqiang Lv,Aimin Dang
DOI: https://doi.org/10.1093/ehjqcco/qcae076
2024-09-14
Abstract:Background: There are currently no specialized risk-scoring systems for critically ill patients with coronary heart disease (CHD). Arterial stiffness, as measured by estimated pulse wave velocity (ePWV), has emerged as a potential indicator of mortality or adverse cardiovascular events in individuals with CHD. This study aimed to evaluate the association between ePWV and all-cause mortality among critically ill patients with CHD beyond traditional risk scores. Methods: This study included 11 001 participants with CHD from the Medical Information Mart for Intensive Care IV, with a one-year follow-up. The primary endpoint was one-year all-cause mortality, and the secondary endpoint was in-hospital mortality. Results: Elevated ePWV was significantly associated with higher risks of in-hospital (OR 1.15, 95% CI 1.12-1.17, p < 0.001) and one-year (HR 1.21, 95% CI 1.20-1.23, p < 0.001) mortality. These associations remained consistent when adjusted for traditional risk scores and potential confounders. When ePWV was integrated into traditional risk scoring models (OASIS, SOFA score, APSIII, SIRS score, SAPS II, and LODS score), the predictive accuracy (area under the curve: 64.55 to 70.56, 64.32 to 72.51, 72.35 to 75.80, 55.58 to 67.68, 71.27 to 73.53, 67.24 to 73.40, p < 0.001) and reclassification (net reclassification index: 0.230, 0.268, 0.257, 0.255, 0.221, 0.254; integrated discrimination improvement: 0.049, 0.072, 0.054, 0.068, 0.037, 0.061, p < 0.001) of these models significantly improved for one-year mortality. Similar results were also found for in-hospital mortality. Conclusions: ePWV is a strong independent predictor of both short- and long-term mortality in critically ill patients with CHD. Importantly, integrating ePWV into traditional risk scores significantly boosts the predictive accuracy for one-year and in-hospital all-cause mortality.