Abstract
BackgroundPost-intervention coagulation abnormalities are potentially modifiable factors influencing ischemic stroke outcomes, but comprehensive assessment tools are lacking. This study developed a novel coagulation score integrating platelet count, INR, and APTT to evaluate its association with mortality in ischemic stroke patients undergoing cerebrovascular intervention.MethodsA historical cohort study was conducted using data from the MIMIC-IV v3.0 database. 1711 adult ischemic stroke patients receiving intervention with documented baseline platelet, INR, and APTT levels were included. A coagulation score (range 0-6) was calculated by summing points assigned for abnormal APTT, INR, and platelet values. Patients were stratified into quartiles (Q1-Q4) based on score. Primary outcomes were 28-day and 60-day all-cause mortality. Associations were analyzed using Kaplan-Meier curves, multivariable Cox regression (adjusting for demographics, illness severity scores, and comorbidities), and restricted cubic spline (RCS) models. Discriminative performance was assessed via ROC analysis.ResultsIncreasing coagulation score quartiles were significantly associated with progressively worse survival (Log-rank P = 0.006 for 28-day; P = 0.0032 for 60-day). Multivariable Cox regression revealed a significant dose-response relationship: each point increase in the continuous score was associated with higher 28-day (HR = 1.075, 95%CI:1.025-1.127, P = 0.003) and 60-day mortality risk (HR = 1.086, 95%CI:1.038-1.136, P < 0.001) in base models. Q4 patients had significantly higher mortality risk than Q1. RCS confirmed a linear positive association without threshold effects (P-nonlinearity>0.05). Subgroup analysis showed stronger associations in patients without heart failure (P-interaction<0.001). While the coagulation score had the highest AUC among individual parameters (28-day: 0.5494; 60-day: 0.5494), overall discriminative performance was modest.ConclusionAn integrated coagulation score demonstrates a linear dose-response relationship with increased mortality risk in ischemic stroke patients post-intervention, particularly among those without heart failure, in partially adjusted analyses. However, its association was attenuated after full adjustment for comorbidities and illness severity. This score may have value as a component of a comprehensive assessment rather than as a standalone predictor, though its predictive ability is limited.