Predictive Model for Early Neurological Deterioration in Acute Ischemic Stroke Utilizing Novel Thrombotic Biomarkers.

backgroundNovel thrombotic molecular markers are significantly linked to acute ischemic stroke (AIS). However, the relationship between thrombin-antithrombin complex (TAT), tissue plasminogen activator-inhibitor complex (t-PAIC), plasmin-α2 plasmin inhibitor complex (PIC), thrombomodulin (TM), and early neurological deterioration (END) remains unclear. Therefore, we developed a prediction model for END based on these markers and evaluated its accuracy and clinical utility.

methodsRetrospective analysis of patients diagnosed with AIS in our hospital from 2023-2024. The above patients were divided into a training set (N = 577) and a test set (N = 246) in a 7:3 ratio. Least absolute shrinkage and selection of operator regression (LASSO) valid predictors were used. The coefficients of the predictors in logistic regression were used to develop a nomogram and to validate its differentiation, calibration, and clinical utility.

resultsThe prevalence of END in AIS patients was 24.3%. Predictors screened according to LASSO regression analysis included age, the National Institutes of Health Stroke Scale (NIHSS) score, t-PAIC, PIC, lymphocyte, and platelet. The resulting nomograms had the area under the curve (AUC) of 0.867 (95% CI, 0.834-0.9) and 0.825 (95% CI, 0.757-0.892) in the training and test sets, respectively, which had good differentiation. In addition, the calibration curve and decision curve analysis (DCA) showed that the model had good calibration and clinical utility.

conclusionA predictive model for END was developed using the serological markers t-PAIC (male >17.13 ng/mL;female >10.52 ng/mL), PIC >0.85 µg/mL, Lymph ≤ 3.2×10^9/L, NIHSS, age, and platelet. The model has significant predictive value for END occurrence in patients with AIS.