Abstract
OBJECTIVE: To construct a nomogram for risk prediction of in-stent restenosis (ISR) within one year after percutaneous coronary intervention (PCI) for coronary heart disease (CHD). METHODS: This study included 842 patients with severe CHD who underwent PCI at Changzhou Affiliated Hospital of Nanjing University of Chinese Medicine between March 2016 to March 2024. Based on the occurrence of ISR within one year post-PCI, patients were stratified into two groups: the ISR group (n=112) and the non-ISR group (n=730). Potential risk factors were initially identified using LASSO regression, followed by multivariate logistic regression to determine independent predictors. A nomogram prediction model was developed using R software (version 4.2.6) and internally validated via the bootstrap resampling method (1000 iterations). Model performance was assessed through receiver operating characteristic (ROC) curve analysis, calibration curve analysis, and decision curve analysis (DCA). RESULTS: This analysis revealed seven risk factors for ISR: diabetes (OR=1.380, 95% CI: 1.090-1.747), neutrophil/lymphocyte ratio (NLR, OR=2.312, 95% CI: 1.830-2.922), low-density lipoprotein (LDL) > 1.8 mmol/L (OR=2.159, 95% CI: 1.080-4.315), calcified lesions (OR=3.780, 95% CI: 2.051-6.968), stent diameter <3 mm (OR=2.595, 95% CI: 1.404-4.796), smoking (OR=2.796, 95% CI: 1.511-5.174) and no intravascular ultrasound (IVUS) assisted (OR=2.176, 95% CI: 1.342~3.257). These seven factors were incorporated into the nomogram model. The model demonstrated excellent discriminative ability, with an area under the curve (AUC) of 0.892 (95% CI: 0.859-0.924) and a consistency index (C-index) of 0.923. Calibration analysis showed close agreement between predicted and observed outcomes, while DCA indicated strong clinical utility across a wide probability threshold range (5.0%-86.2%). The relative importance of the risk factors, ranked in descending order, was as follows: calcified lesions, stent diameter <3 mm, no IVUS assisted, LDL>1.8 mmol/L, smoking, NLR and diabetes. CONCLUSION: The study identifies several risk factors for ISR in CHD patients within one year after PCI. The constructed nomogram model has good predictive efficiency and clinical applicability.