摘要
目的 在急性冠脉综合征(ACS)患者中构建并验证氯吡格雷高血小板反应性(HPR)的临床预测模型.方法 单中心、回顾性研究.连续纳入 2019 年1 月1 日至2022 年11 月30 日在郑州大学人民医院 阜外华中心血管病医院接受治疗的 242 例ACS患者作为研究对象.采用简单随机抽样法按 6∶4 分为训练集 144 例和验证集 98 例.筛选出氯吡格雷HPR的独立危险因素,建立预测模型并绘制Nomogram图.在训练集和验证集中,分别采用受试者工作特征(ROC)曲线、Bootstrap自抽样法、Hosmer-Lemeshow拟合优度检验和决策曲线分析(DCA)对预测模型进行区分度、校准度和临床实用度评估.结果 Logistic 回归分析显示,CYP2C19 基因多态性、体质指数(BMI)≥28 kg/m2、平均血小板体积(MPV)≥10.85 fL和低密度脂蛋白胆固醇(LDL-C)≥2.6 mmol/L是ACS患者发生氯吡格雷 HPR 的独立危险因素.将上述 4 项指标纳入回归方程构建预测模型并绘制Nomogram图.在训练集和验证集中采用ROC曲线对模型进行区分度检验,曲线下面积(AUC)分别为 0.757(95%CI:0.676～0.840)和 0.750(95%CI:0.646～0.854);采用Bootstrap自抽样法抽样 1 000次对模型进行内部验证,矫正后的C-Statistic分别为 0.732(95%CI:0.656～0.820)和 0.725(95%CI:0.632～0.827);进行Hosmer-Lemeshow检验,P值分别为 0.954 和 0.097;进行DCA检验显示,模型的临床净获益范围分别为20.20%～97.95%和19.31%～78.92%.结论 ACS患者出现氯吡格雷HPR的独立危险因素为CYP2C19基因多态性、BMI≥28 kg/m2、MPV≥10.85 fL和LDL-C≥2.6 mmol/L.本研究建立的Nomogram图在区分度、校准度和临床实用度三个方面表现良好,可作为临床预测工具,为基层医生调整双抗方案提供便利.
Abstract
Objective To construct and validate a clinical prediction model of clopidogrel high platelet reactivity(HPR)in patients with acute coronary syndrome(ACS).Methods This was a single-center,retrospective study.This study consecutively included 242 ACS patients who were treated in Zhengzhou University People's Hospital,Fuwai Central China Cardiovascular Hospital from January 1,2019 to November 30,2022,as research subjects.The patients were divided into a training set of 144 cases and a validation set of 98 cases by simple random sampling at a ratio of 6∶4.The independent risk factors of clopidogrel HPR were screened out,the prediction model was established and the Nomogram was drawn.In the training set and validation set,receiver operating characteristic(ROC)curve,Bootstrap self-sampling method,Hosmer-Lemeshow test and decision curve analysis(DCA)were used to evaluate the discrimination,calibration and clinical utility of the prediction model,respectively.Results Logistic regression analysis revealed that CYP2C19 gene polymorphism,body mass index(BMI)≥28 kg/m2,mean platelet volume(MPV)≥10.85 fL,and low-density lipoprotein cholesterol(LDL-C)≥2.6 mmol/L were independent risk factors for the occurrence of clopidogrel HPR in patients with ACS.The four indicators above were incorporated into the regression equation to construct the prediction model,and a Nomogram was plotted.The model's discriminative power was assessed using the ROC curve in both the training and validation sets,with the area under the curve(AUC)values of 0.757(95%CI:0.676-0.840)and 0.750(95%CI:0.646-0.854),respectively.Internal validation of the model was performed through 1 000 Bootstrap resamplings,with corrected C-Statistics of 0.732(95%CI:0.656-0.820)and 0.725(95%CI:0.632-0.827),respectively.The Hosmer-Lemeshow test yielded P-values of 0.954 and 0.097,respectively.DCA demonstrated that the clinical net benefit ranges for the model were 20.20%to 97.95%and 19.31%to 78.92%,respectively.Conclusions The independent risk factors for clopidogrel HPR in ACS patients are CYP2C19 gene polymorphism,BMI≥28 kg/m2,MPV≥10.85 fL and LDL-C≥2.6 mmol/L.The Nomogram established in this study demonstrates good performance in discrimination,calibration and clinical utility,and can serve as a clinical prediction tool to facilitate the adjustment of dual antiplatelet therapy regimens by primary care physicians.
基金项目
国家重点研发计划数字诊疗装备研发重点专项2018年度定向项目(2018YFC0114500)
维普
万方数据