Prediction of Coronary Artery Restenosis in Patients Undergoing Angioplasty

Abstract

BACKGROUND AND OBJECTIVE: Coronary artery restenosis following angioplasty may lead to major unwanted cardiac events. This is one of the most important concerns of cardiologists. This study was performed to evaluate the predictors of coronary artery restenosis in patients undergoing angioplasty. METHODS: In this cohort study, all 2159 patients who referred to Madani Hospital in Khorramabad, Iran for angioplasty between 2004 and 2015 were followed up regarding the incidence of coronary artery restenosis for one or two years (at least 50% stenosis in the place of the previous angioplasty) depending on the type of stent, and clinical variables related to the lesion and procedure were recorded in an information form. Then, agents for predicting coronary artery restenosis were identified based on CART (Classification and Regression Tree). FINDINGS: Cumulative incidence of coronary artery restenosis was 12.9% (279 patients) in subcategories: non-drug-eluting stent (25.8%, 114 patients), comorbidity of diabetes and hyperlipidemia (under the subcategory of drug-eluting stents, 19%, 32 patients), multiple drug-eluting stents (under the subcategory of non-diabetic patients, 14.6%, 25 patients), comorbidity of hyperlipidemia and drug abuse (under the subcategory of non-diabetic group who use one drug-eluting stent, 12.8%, 16 patients). CONCLUSION: According to the results of this study, stent type, history of diabetes, number of stents, hyperlipidemia and drug abuse are among the most important predictors of coronary artery restenosis.

Keywords

• Coronary Artery Restenosis
• Coronary Angioplasty
• Coronary Stent
• Data Mining
• Classification Tree