ST段抬高急性心肌梗死患者PCI术后双抗治疗严重出血预测模型构建与评价

A prediction model for severe bleeding after dual antiplatelet therapy in patients with ST-segment elevation acute myocardial infarction after PCI

目的 构建ST段抬高型急性心肌梗死患者经皮冠状动脉介入(percutaneous coronary intervention,PCI)术后口服双抗(阿司匹林+氯吡格雷)严重出血的列线图模型,并验证该模型的准确性。方法 回顾性分析2016年1月1日-2021年1月1日在葫芦岛市中心医院行PCI介入治疗的ST段抬高型急性心肌梗死患者临床资料,并对所有患者随访12个月,剔除随访丢失的患者,最终纳入1 139例,根据随访期间是否发生严重出血分为严重出血组(n=113)和非严重出血组(n=1 026),利用多因素Logistic回归分析患者发生出血的危险因素,基于R语言构建ST段抬高型急性心肌梗死患者PCI术后口服双抗导致严重出血的列线图模型,并利用Bootstrap方法以及临床决策曲线验证该模型的准确性和临床决策的获益性。结果 多因素Logistic分析显示,围手术期使用Ⅱb/Ⅲa受体拮抗剂、既往消化道疾病病史、年龄>60岁、肌酐清除率较低、血小板计数较低是导致ST段抬高型急性心肌梗死患者PCI术后口服双抗导致严重出血的独立危险因素(OR=1.012、6.042、1.205、1.215、1.322,P均=0.001);ROC曲线验证列线图模型显示:列线图预测患者PCI术后发生严重出血的风险能力较强(AUC=0.84,95%CI0.72~0.91);采用Bootstrap方法重复抽样1000次验证列线图,发现校准曲线的平均绝对误差为0.018,说明校准曲线与理想曲线贴合良好;临床决策曲线显示,当PCI术后口服双抗发生严重出血的发生阈值为0.06~0.88时该模型图的适用性最佳。结论 构建的ST段抬高型急性心肌梗死患者PCI术后口服双抗严重出血的列线图模型具有较好的临床预测价值。

Objective To construct a nomogram model for severe bleeding caused by oral dual-antibodies(aspirin and clopidogrel)in patients with ST-segment elevation acute myocardial infarction after PCI,and to verify the accuracy of the model.Methods The clinical data of patients with ST-segment elevation acute myocardial infarction who had undergone PCI intervention therapy in our hospital between January 1,2016 and January 1,2021 was retrospectively analyzed.All the patients were followed up for 12 months,and those who were out of contact were excluded.A total of 1139 patients were finally included.According to the severity of serious bleeding during the follow-up,these patients were divided into the serious bleeding group(n=l13 cases)and the non-serious bleeding group(n=1026 cases).Multivariate logistical analysis was conducted to analyze the risk of bleeding.A nomogram model for severe bleeding caused by oral dual-antibodies in patients with ST-segment elevation acute myocardial infarction after PCI was established based on R language,and the Bootstrap method and clinical decision curve were used to verify the accuracy of the model and its benefits to clinical decision-making.Results Multivariate logistic analysis showed that perioperative use ofⅡb/Ⅲa receptor antagonists,a history of gastrointestinal diseases,age above 60,low creatinine clearance,and low platelet count were independent risk factors for severe bleeding caused by oral dual-antibodies after PCI in patients with ST-segment elevation acute myocardial infarction(OR=1.012,6.042,1.205,1.215,1.322,all P=0.001).ROC curve validation of the nomogram model suggested that the nomogram could well predict the risk of severe bleeding after PCI,with an AUC of 0.84,95%CI 0.72-0.91.The Bootstrap method led to the finding that the mean absolute error of the calibration curve was 0.018,indicating that the calibration curve fitted in well with the ideal curve.The best fit of the model was when the threshold for severe bleeding was between 0.06 and 0.88.Conclusion This nomogram model for severe bleeding induced by oral dual-antibodies in patients with ST-segment elevation acute myocardial infarction after PCI is of good clinical predictive value.