基于CMR-LGE图像的左心室熵对肥厚型心肌病患者MACE的预测研究  

Prediction of MACE in Patients with Hypertrophic Cardiomyopathy by Left Ventricular Entropy Based on CMR-LGE Image

作　　者：郑月 马运婷 沙立辉 赵新湘[1] ZHENG Yue;MA Yunting;SHA Lihui(Department of Radiology,The Second Affiliated Hospital of Kunming Medical University,Kunming,Yunnan Province 650101,P.R.China)

机构地区：[1]昆明医科大学第二附属医院放射科,650101

出　　处：《临床放射学杂志》2024年第2期217-222,共6页Journal of Clinical Radiology

基　　金：国家自然科学基金项目(编号:82160332);北京医学奖励基金会课题研究项目(编号:YXJL-2022-0665-0214);云南省重点项目(编号:202203AC10007,202201AY070001-097);昆医附二院院内临床研究项目(编号:ynllT2021010)。

摘　　要：目的探讨基于心脏磁共振成像(CMR)-钆延迟强化(LGE)图像获得的左心室熵能否预测肥厚型心肌病(HCM)患者主要心脏不良事件(MACE)的发生风险。方法选取2018年1月至2021年12月符合HCM诊断标准且具有CMR检查的患者,以MACE为终点事件。根据随访结果将患者分为MACE组和无MACE组,采用CVI42心脏后处理软件测得左心室射血分数(LVEF),将LGE图像导入Python 3.8软件获得左心室熵。结果最终纳入107例HCM患者,中位随访为23个月,其中MACE组34例,无MACE组73例。MACE组左心房内径(LAD)[(39.14±6.85)mm vs.(35.24±7.70)mm]、左心室最大室壁厚度(LVMWT)[(21.18±4.37)mm vs.(19.23±3.11)mm]、左心室熵[5.12(3.94,6.18)vs.4.04(2.64,5.45)]均大于无MACE组(P均<0.05);MACE组LVEF[(53.54±14.94)%vs.(59.64±10.95)%]小于无MACE组(P=0.038)。单因素Logistic回归显示,LAD、LVMWT、LVEF、左心室熵是发生MACE的危险因素,进一步多因素回归显示,LVEF(OR=0.960,95%CI:0.927~0.995,P=0.026)、左心室熵(OR=1.600,95%CI:1.148~2.229,P=0.006)是HCM发生MACE的独立危险因素。受试者工作特征(ROC)曲线显示,LVEF单变量预测模型的曲线下面积(AUC)为0.637,最佳截断值为52.41%,敏感度为41.18%,特异度为84.93%;左心室熵单变量预测模型的AUC为0.692,最佳截断值为4.975,敏感度为61.76%,特异度为71.23%。左心室熵和LVEF联合预测模型的AUC为0.721。结论基于CMR-LGE图像获得的左心室熵可反映心肌的异质性,是HCM患者预后的有效预测因子,HCM患者发生MACE的风险随左心室熵的增大而增加,有助于为HCM患者提供更全面的危险分层,提高患者预后。Objective To investigate whether left ventricular entropy based on MR-LGE image can predict the risk of major cardiac adverse events(MACE)in patients with hypertrophic cardiomyopathy(HCM).Methods Patients who met the diagnostic criteria of HCM and had cardiac magnetic resonance(CMR)examination from January 2018 to December 2021 were selected,with MACE as the endpoint event.According to the follow-up results,the patients were divided into the MACE group and the non-MACE group.The left ventricular ejection fraction(LVEF)was measured by CVI42 cardiac postprocessing software,and the LGE image was imported into Python 3.8 software to obtain the left ventricular entropy.Results A total of 107 HCM patients were included,with a median follow-up of 23 months,including 34 patients in the MACE group and 73 patients without MACE.The inner diameter of left atrium(LAD)in MACE group[(39.14±6.85)mm vs.(35.24±7.70)mm],maximum left ventricular Maximum wall thickness(LVMWT)[(21.18±4.37)vs.(19.23±3.11)],and left ventricular entropy[5.12(3.94,6.18)vs.4.04(2.64,5.45)]were higher than those without MACE(all P<0.05);The LVEF of MACE group[(53.54±14.94)%vs.(59.64±10.95)%]was lower than that of non-MACE group(P=0.038).Single logistic regression showed that LAD,LVMWT,LVEF and left ventricular entropy were the risk factors for MACE.Further multi-factor regression showed that LVEF(OR=0.960,95%CI:0.927-0.995,P=0.026)and left ventricular entropy(OR=1.600,95%CI:1.148-2.229,P=0.006)were independent risk factors for MACE in HCM.Receiver Operating characteristic(ROC)curve showed that the area under the curve(AUC)of LVEF univariate prediction model was 0.637,the optimal cut-off value was 52.41%,the sensitivity was 41.18%,and the specificity was 84.93%.The AUC of the single variable prediction model of left ventricular entropy was 0.692,the optimal cut-off value was 4.975,the sensitivity was 61.76%,and the specificity was 71.23%.The AUC of the combined prediction model of left ventricular entropy and LVEF was 0.721.Conclusion Left ventricular

关 键 词：肥厚型心肌病 磁共振成像 纹理分析 熵 左心室 主要心脏不良事件 

分 类 号：R542.2[医药卫生—心血管疾病]