Study on Risk Factors and Development of a Predictive Model for Recurrent In-stent Restenosis in Patients With Coronary Heart Disease After Percutaneous Coronary Intervention
Objectives:To explore the risk factors for recurrent in-stent restenosis(R-ISR)in patients with coronary heart disease after percutaneous coronary intervention(PCI)and to develop a risk prediction model for R-ISR using a nomogram.Methods:All patients treated for ISR at the Fuwai Hospital,Chinese Academy of Medical Sciences from January to December 2017 were eligible for this study.A total of 1 102 ISR patients were included for analysis.Based on the recurrence of ISR after PCI,patients were divided into R-ISR group and non-R-ISR group.Univariate Cox regression analyses,LASSO regression analyses,and the combination of clinical experience were used to select predictors of R-ISR.A multivariate Cox regression model was used to analyze the independent risk factors of R-ISR and to develop a risk prediction model.Results:The median follow-up duration for participants was 1 264(1 169,1 334)days,the incidence rate of R-ISR after PCI was 10.1%.Multivariate Cox regression analysis showed that age(HR=0.98,95%CI:0.96-0.99),total bilirubin(HR=0.95,95%CI:0.91-0.99),apolipoprotein A1(HR=0.08,95%CI:0.02-0.42),high-sensitivity C-reactive protein(HR=1.05,95%CI:1.01-1.10),and reference vessel diameter(HR=0.65,95%CI:0.44-0.98)were independent determinants of R-ISR.Accordingly,the R-ISR risk prediction model was developed with a nomogram,the AUC of this model to predicto R-ISR was 0.70(95%CI:0.64-0.77).Conclusions:Coronary heart disease patients with younger age,lower levels of total bilirubin and apolipoprotein A1,smaller vessel diameter,and higher levels of high-sensitivity C-reactive protein are at higher risk of R-ISR.The developed visual risk prediction model for R-ISR shows promising predictive performance but still requires further optimization and validation.
coronary heart diseasepercutaneous coronary interventionrecurrent in-stent restenosisrisk factorprediction model
NETL
NSTL
万方数据
