构建模型预测STEMI患者PPCI后1年内发生MACE的风险

A prediction model was constructed to predict the risk of MACE in STEMI patients received PPCI at one year after surgery

目的:建立与验证一个可预测接受直接经皮冠状动脉介入治疗(PPCI)的急性ST段抬高型心肌梗死(STEMI)患者术后1年内发生主要不良心血管事件(MACE)风险的列线图预测模型。方法:本研究是一项回顾性队列研究,选取2019年10月—2022年8月来自新疆维吾尔自治区人民医院490例行PPCI的STEMI患者为研究对象。应用LASSO回归和logistic多因素回归分析接受PPCI的STEMI患者1年内发生MACE风险的独立危险因素。构建接受PPCI的STEMI患者1年内发生MACE的列线图预测模型,采用ROC曲线、校准曲线进行模型表现评估。并用Bootstrap法进行内部验证,绘制校准曲线、计算C统计量评估模型内部验证表现,绘制DCA临床决策曲线评价模型临床净收益。结果:LASSO回归与多因素logistic回归分析结果显示,入院72 h内肌钙蛋白T峰值升高、术后24 h内血红蛋白水平降低、术中发生低血压、术中发生无复流、既往有高血压病史是接受PPCI的STEMI患者1年内发生MACE的危险因素。模型在本研究人群中的AUC值为0.730(95%CI:0.679~0.782),Hosmer-Lemeshow检验结果提示该模型在本研究人群中有较好的预测准确性。使用Bootstrap法重抽样500次进行内部验证,模型的C指数为0.719,且模型内部验证的校准曲线和理想模型接近,表明模型在内部验证中的区分度及预测准确性较好。DCA临床决策曲线显示该模型在阈值概率范围为0.14~0.66时能够带来临床净获益。结论:本研究构建的简单易用的列线图预测模型可以有效地预测接受PPCI的STEMI患者术后1年内发生MACE的个体化风险。

Objective To establish and validate a nomogram prediction model that can predict the risk of major adverse cardiovascular events within 1 year after primary percutaneous coronary intervention in patients with acute ST-segment elevation myocardial infarction.Methods This study is a retrospective cohort study.The 490 patients with STEMI who underwent PPCI from the People's Hospital of Xinjiang Autonomous Region from October 2019 to August 2022 were selected as the research objects.LASSO and logistic regression analysis of independent risk factors for MACE in patients with STEMI undergoing PPCI within 1 year.A nomogram prediction model of MACE within 1 year in STEMI patients undergoing PPCI was constructed,and the model performance was evaluated using ROC curve and calibration curve.Bootstrap method was used for internal verification.The calibration curve was drawn,the C statistic was calculated to evaluate the model's internal validation performance,and the DCA clinical decision curve was drawn to evaluate the model's clinical net benefit.Results LASSO regression and multivariate logistic regression analysis showed that the increase of peak troponin T within 72 hours of admission,the decrease of hemoglobin level within 24 hours after surgery,intraoperative hypotension,intraoperative no reflow,and previous history of hypertension were risk factors for MACE within 1 year in STEMI patients receiving PPCI.The AUC value of the model in the study population was 0.730(95%CI:0.679-0.782),and the Hosmer-Lemeshow test results suggested that the model had good prediction accuracy in the study population.The Bootstrap method was used to re-sample 500 times for internal verification.The C-index of the model was 0.719,and the calibration curve of the internal verification of the model was close to the ideal model,indicating that the model had good differentiation and prediction accuracy in the internal verification.The DCA clinical decision curve showed that the model could provide a net clinical benefit at a threshold probability range of 0.14-0.66.Conclusion The simple and easy-to-use nomogram prediction model constructed in this study can effectively predict the individualized risk of MACE in STEMI patients receiving PPCI at 1 year after surgery.