冠心病病人PCI术后1年出血危险因素的模型构建和验证

Construction and validation of risk factors model related to hemorrhage in patients with coronary heart disease after PCI 1 year

目的:基于LASSO-Logistic回归分析冠心病病人经皮冠状动脉介入(PCI)术后1年出血的危险因素,并建立Nomogram预测模型,以便临床采取及时、有效的干预方案。方法:选取2021年1月—12月在我院行PCI术的226例冠心病病人为研究对象,依据PCI术后1年是否出血分为出血组、未出血组。采用LASSO-Logistic回归筛选PCI术后1年出血的危险因素,进一步构建Nomogram预测模型。利用受试者工作特征(ROC)曲线、校准曲线验证Nomogram预测模型的价值。结果:PCI术后1年226例冠心病病人出血发生率为11.95%;年龄≥60岁、有消化道病史、慢性肾功能不全、使用血管闭合器、球囊扩张次数、支架术后抗血小板药物停药模式(PARIS)评分、支架术后双重抗血小板治疗病人出血并发症预测(PRECISE-DAPT)评分为PCI术后1年出血的危险因素(P<0.05);Nomogram预测模型预测PCI术后1年出血的ROC曲线下面积为0.959。结论:年龄≥60岁、有消化道病史、慢性肾功能不全、使用血管闭合器、球囊扩张次数、PARIS评分、PRECISE-DAPT评分为冠心病病人PCI术后1年出血的危险因素,基于LASSO-Logistic回归分析建立的Nomogram预测模型对PCI术后1年出血具有一定预测价值,临床应筛查高危人群并实施严格观察,制定合理治疗措施,以降低出血风险。

Objective:Based on LASSO-Logistic regression analysis of hemorrhage related factors in patients with coronary atherosclerotic heart disease(CHD)1 year after percutaneous coronary intervention(PCI),and to establish a Nomogram prediction model for timely and effective clinical intervention.Methods:226 patients with coronary heart disease who underwent PCI in our hospital from January to December 2021 were selected as the study objects,and were divided into hemorrhage group and non-hemorrhage group according to whether they had hemorrhage one year after PCI.LASSO-Logistic regression was used to screen the risk factors of hemorrhage 1 year after PCI,and further construct a Nomogram prediction model.Receiver operating characteristic(ROC)curve and calibration curve were used to verify the value of Nomogram prediction model.Results:One year after PCI,the incidence of hemorrhage in 226 patients with coronary heart disease was 11.95%.Age≥60 years old,gastrointestinal history,chronic renal insufficiency,use of vessel closure,balloon dilation times,PARIS score,precision-DAPT score were the risk factors for bleeding 1 year after PCI(P<0.05).The Nomogram prediction model predicted the AUC of hemorrhage at 1 year after PCI to be 0.959.Conclusion:Age≥60 years,history of digestive tract,chronic renal insufficiency,use of vessel closure,balloon dilation times,PARIS score and precision-DAPT score are risk factors for hemorrhage 1 year after PCI in patients with coronary heart disease.The establishment of a Nomogram prediction model based on LASSO-Logistic regression analysis has a certain value in predicting hemorrhage 1 year after PCI.Clinical screening and strict observation of high-risk groups should be carried out,and reasonable treatment measures should be formulated to reduce the risk of hemorrhage.