A diagnostic model for non-invasive localization of culprit lesions in NSTE-ACS by integrating CCTA geometric and plaque parameters.

ObjectiveThis study aimed to identify valuable quantitative plaque parameters and adverse vascular geometric features for constructing a diagnostic model to improve culprit vessel identification in Non-ST segment elevation acute coronary syndrome (NSTE-ACS).MethodsData from 144 consecutive NSTE-ACS patients who underwent both Coronary CT angiography (CCTA) and invasive coronary angiography (ICA) were retrospectively analyzed. Coronary geometric characteristics and quantitative plaque parameters were derived from CCTA data. Patients were randomly split into training and validation sets at a 7:3 ratio. Using logistic regression and stepwise regression, three models were developed: (I) diameter stenosis + high-risk plaque (HRP); (II) Model I + quantitative plaque parameters; (III) Model II + adverse geometric characteristics (AGCs). The optimal model was selected by comparing the areas under the receiver operating characteristic curves (AUC), and nomograms with calibration and decision curves were generated.ResultsThe results demonstrated that the multivariable logistic regression model integrating diameter stenosis, HRP, quantitative plaque parameters, and AGCs exhibited excellent diagnostic performance, with AUCs of 0.95 and 0.92 in the training and validation sets, respectively. Decision curve analysis confirmed the model's net benefit across threshold probabilities, demonstrating its clinical utility.ConclusionsThis CCTA-based model showed high predictive capability, further enhanced by incorporating vascular geometric features, offering a clinically valuable tool for the precise identification of culprit lesions in NSTE-ACS.