斑点追踪显像联合Logistic回归评价慢性肺心病患者左心室扭转的改变

Measurement of left ventricular torsion by 2-dimensional speckle-tracking imaging in patients with chronic cor pulmonale

目的 应用Logistic回归筛选超声二维斑点追踪显像技术（STI）检测慢性肺心病患者左心室旋转/扭转改变的敏感指标,并探讨其评价慢性肺心病患者左心室功能异常的临床价值. 方法 选取36例慢性肺心病患者（慢性肺心病组）和38例健康体检者（对照组）,采集胸骨旁二尖瓣口水平及心尖水平短轴切面图像,应用Echo PAC软件测定心底旋转峰值、心尖旋转峰值、左室扭转峰值、收缩末期心底旋转值、收缩末期心尖旋转值、收缩末期左室扭转值,应用Logistic回归对左室旋转/扭转的相关指标进行分析筛选,并建立回归方程,应用受试者工作特征（ROC）曲线明确以扭转特异参数（左室扭转峰值、收缩末期左室扭转值）判定左室功能减退的最佳诊断界值. 结果 与对照组比较,慢性肺心病组左心室旋转/扭转各特异参数降低（均P＜0.01）.Logistic回归分析结果显示,收缩末期左心室扭转值和左室扭转峰值与慢性肺心病患病有关（OR=0.473、0.706,P=0.007、0.011）.ROC曲线示以左心室扭转峰值判定左室功能减退的曲线下面积为0.819（95％CI：0.683-0.956）,临界值为12.070°,灵敏度84.6％,特异度73.9％;以收缩末期左室扭转值判定左室功能减退的曲线下面积为0.875（95％ CI：0.744-1.000）,临界值为10.680°,灵敏度84.6％,特异度91.3％.结论 STI技术能敏感地检测慢性肺心病患者左室心肌扭转的改变,并评估其左心室功能的状态.

Objective To evaluate the sensitive indicator of left ventricular rotation/torsion assessed by two-dimensional speckle tracking imaging （STI） and Logistic regression analysis,and to investigate the clinical value of the sensitive indicator for assessment of left ventricular dysfunction in patients with chronic cor pulmonale （CCP）.Methods 36 patients with CCP （CCP group） and 38 healthy controls （control group） were included in this study.Imaging in parasternal short-axis view （in basal and apical level） were selected.Parasternal short-axis views at mitral valve and apical levels were collected.Basal peak rotation,apical peak rotation,peak torsion,basal end-systolic rotation,apical end-systolic rotation and end systolic torsion were measured with Echo PAC software.Relevant indicators of left ventricular rotation/torsion were selected by using logistic regression analysis and the regression equation was established.Optimal values of specific parameters （Peak torsion and end-systolic torsion） were calculated with receiver-operating characteristic （ROC） curve.Results Specific parameters of rotation/torsion were significantly reduced in patients with CCP as compared with controls （all P＜0.01）.Logistic regression analysis showed that end-systolic torsion and peak torsion were correlated with CCP （OR=0.473 and 0.706,P=0.007 and 0.011）.Cut-off value of peak torsion for predicting left ventricular dysfunction was 12.070°,the area under the ROC curve （AUC） was 0.819 （95％ CI：0.683-0.956）,the sensitivity was 84.6％,and the specificity was 73.9 ％.Cut-off value of end-systolic torsion for predicting left ventricular dysfunction was 10.680°,AUC was 0.875（95％CI：0.744 1.000）,the sensitivity was 84.6％,and the specificity was 91.3％.Conclusions Two-dimensional speckle tracking imaging can sensitively assess left ventricular torsion and evaluate the left ventricular dysfunction in patients with CCP.