摘要
目的探讨影响冠状动脉慢性完全闭塞病变PCI术后(CTO-PCI)患者发生主要不良心血管事件(MACE)的独立危险因素,为辅助制定临床决策提供参考依据。方法回顾性分析1415例符合冠状动脉慢性完全闭塞病变的PCI术后患者的临床和随访资料。运用R3.6.2和易侕软件对数据建立MFP模型、全因素Cox回归模型(full model)以及逐步回归模型(stepwise model),获得影响冠状动脉慢性完全闭塞病变PCI术后患者发生MACE的独立危险因素。根据ROC曲线下面积(AUC)、一致性指数(C-index)及校准曲线(calibration curve)来评价3种模型的预测能力,最终建立列线图预测模型。结果MFP模型、全因素Cox回归模型和逐步回归模型均显示心率、高血压、吸烟、lg(脑钠肽)[lg(BNP)]是冠状动脉慢性完全闭塞病变患者PCI术后出现MACE的独立预后因素(P<0.05)。三种预测模型的AUC分别为0.77、0.77和0.78,一致性指数分别为0.72、0.73和0.73。不同随访时间点对应的AUC值均在0.7~0.8之间波动,最终确定逐步回归模型是最佳预测模型,具有良好的预测效能。基于逐步回归模型中的吸烟(HR=1.96,95%CI:1.09~3.45,P=0.0215)、心率(HR=1.03,95%CI:1.01~1.04,P=0.0023)、高血压(HR=1.92,95%CI:1.21~3.03,P=0.0054)和lg(BNP)(HR=3.01,95%CI:1.72~5.26,P=0.0001)这些变量,最终构建了预后列线图预测模型。结论心率、高血压、吸烟和lg(BNP)是冠状动脉慢性完全闭塞病变PCI术后患者MACE发生的独立预后因素,根据这些预后因素构建的列线图模型可量化评估患者的预后。
Aim To explore the independent risk factors for major adverse cardiovascular events(MACE)in patients with chronic total occlusion of coronary artery after PCI(CTO-PCI),and to provide a reference for assisting in clinical decision-making.Methods Clinical and follow-up data of 1415 patients with chronic total occlusion of coronary artery after PCI were retrospectively analyzed.R3.6.2 and YIER software were used to establish the MFP model,the all-factor Cox regression model and the stepwise regression model on the data,so as to obtain the independent risk factors affecting the occurrence of MACE events in patients with chronic total occlusion of coronary artery after PCI.The prediction ability of the three models was evaluated according to the area under the ROC curve(AUC),the C-index and the calibration curve,and finally the nomogram prediction model was established.Results MFP model,all-factor Cox regression model and stepwise regression model all showed that heart rate,hypertension,smoking and brain natriuretic peptide(lg(BNP))were independent prognostic factors related to MACE in patients with chronic total occlusion of coronary artery after PCI(P<0.05).AUC of the three prediction models was 0.77,0.77 and 0.78,and C-index was 0.72,0.73 and 0.73,respectively.The AUC values corresponding to different follow-up time points fluctuated between 0.7 and 0.8,and the stepwise regression model was finally determined to be the best prediction model with good prediction performance.Based on the variables of smoking(HR=1.96,95%CI:1.09~3.45,P=0.0215),heart rate(HR=1.03,95%CI:1.01~1.04,P=0.0023),hypertension(HR=1.92,95%CI:1.21~3.03,P=0.0054)and lg(BNP)(HR=3.01,95%CI:1.72~5.26,P=0.0001)in the stepwise regression model,the prognosis nomogram prediction model was finally constructed.Conclusion Heart rates,hypertension,smoking,and lg(BNP)were independent prognostic factors for MACE events in patients with chronic total occlusion of coronary artery after PCI,and the nomogram prediction model based on these prognostic factors could be used to quantitatively evaluate the prognosis of patients.
作者
王凯阳
吴婷婷
郑颖颖
马艳艳
王万荣
潘颖
马依彤
谢翔
WANG Kaiyang;WU Tingting;ZHENG Yingying;MA Yanyan;WANG Wanrong;PAN Ying;MA Yitong;XIE Xiang(Department of Heart Center,the First Affiliated Hospital of Xinjiang Medical University,Urumqi,Xinjiang 830054,China)
出处
《中国动脉硬化杂志》
CAS
2021年第6期475-482,共8页
Chinese Journal of Arteriosclerosis
基金
新疆维吾尔自治区创新环境(人才、基地)建设专项(人才专项计划天山雪松计划)(2018XS17)。
