Construction and validation of a prediction model for in-stent restenosis following coronary stent implantation during dual antiplatelet therapy

ObjectivesIn-stent restenosis (ISR) is a serious complication that occurs after a percutaneous coronary intervention (PCI). This study aims to integrate thromboelastography (TEG) parameters, coagulation function, and clinical indicators to develop and validate a prediction model for ISR during antiplatelet therapy after PCI.MethodA total of 401 consecutive patients with coronary artery disease who underwent a PCI with drug-eluting stent implantation between January 2018 and January 2024 were enrolled in this study. Clinical baseline characteristics were collected. The Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to screen 6 key predictors from 34 candidate clinical features. Multivariate logistic regression identified independent risk factors for ISR. Model performance was evaluated using the receiver operating characteristic curve for discrimination, bootstrap internal validation for stability, the Hosmer–Lemeshow test for goodness of fit, and decision curve analysis for clinical utility.ResultsAmong the 401 enrolled patients, 137 (34.2%) developed ISR. LASSO regression selected 6 key predictors from 34 candidate variables. Multivariable logistic regression ultimately identified a low adenosine diphosphate (ADP) inhibition rate, a low arachidonic acid (AA) inhibition rate, reaction time (R-value), and age as independent risk factors for ISR. The constructed nomogram demonstrated good discrimination (area under the curve = 0.752, 95% confidence interval: 0.702–0.803), stable performance upon bootstrap internal validation, and satisfactory calibration (Hosmer–Lemeshow test, P = 0.320). Furthermore, decision curve analysis indicated a favorable clinical net benefit across a broad range of threshold probabilities.ConclusionsThe prediction model, constructed based on TEG parameters (ADP inhibition rate, AA inhibition rate, R-value), age, and N-terminal pro-B-type natriuretic peptide, effectively identifies patients at high risk for ISR after a PCI. It provides a basis for individualized antiplatelet therapy and postoperative management decisions.