A predictive model for mortality in VA-ECMO patients based on early coagulation-immune interactions: A single-center retrospective study

OBJECTIVE: To develop and validate an early predictive model for in-hospital mortality in patients receiving veno-arterial extracorporeal membrane oxygenation (VA-ECMO) by analyzing dynamic changes in metabolic, inflammatory, immune, and coagulation parameters. RESULTS: The in-hospital mortality rate was 66.7%. Non-survivors exhibited higher baseline severity (APACHE II: 31 vs 28, p = 0.001) and significantly impaired responses to ECMO, including persistent hyperlactatemia (6.6 vs 1.9 mmol/L, P < 0.001), uncontrolled inflammation, lymphocytopenia, and platelet consumption. Using LASSO regression for variable selection, multivariate analysis identified five independent predictors: post-ECMO lactate (OR = 1.27, 95% CI: 1.15-1.42, P < 0.001), age (OR = 1.03, 95% CI: 1.00-1.06, P = 0.029), post-ECMO PCT (OR = 1.06, 95% CI: 1.02-1.11, P = 0.004), post-ECMO CRP (OR = 1.01, 95% CI: 1.00-1.02, P = 0.005), and the directional change in platelet-to-lymphocyte ratio (ΔPLR = PLR_post - PLR_pre, OR = 0.997, 95% CI: 0.994-1.00, P = 0.044). The final model demonstrated good discrimination with an AUC of 0.858 (95% CI: 0.806-0.910) and good calibration (Hosmer-Lemeshow P = 0.386). CONCLUSIONS: We developed an efficient prediction model based on early multi-system dynamic responses. Its innovation lies in capturing the directional shift in coagulation-immune status (ΔPLR) from pre-to post-ECMO, alongside metabolic and inflammatory parameters. This tool enables accurate risk stratification within 6 h of cannulation, potentially guiding individualized therapy.