心肌纵向分层应变参数评价冠心病患者左心室局部及整体收缩功能

Assessment for regional and global systolic function of left ventricle in patients with coronary arterial disease by longitudinal layer-specific strain

目的应用二维斑点追踪显像（2D-STI）技术获取冠心病患者左室心内膜下、中层及心外膜下心肌的节段及整体纵向应变，评价冠心病患者左室局部及整体收缩功能，探讨该参数预测冠脉病变程度及范围的临床价值。方法临床疑诊冠心病患者78例，根据选择性冠脉造影结果分为冠心病组46例和对照组32例；其中冠心病组再根据冠脉狭窄程度或累及支数分为轻度狭窄组（冠脉狭窄50％～75％）和重度狭窄组（冠脉狭窄≥75％）、单支病变组和多支病变组；分别采集受试者心尖四腔、两腔、左室长轴观动态图像，应用EchoPAC工作站分别获取左室各节段心内膜下心肌、中层心肌、心外膜下心肌各节段纵向应变（TLSendo，TLSmid，TLSepi）及左室心内膜下心肌、中层心肌、心外膜下心肌整体纵向应变（GLSendo，GLSmid，GLSepi）。结果无论有无冠脉狭窄及冠脉病变范围大小，TLSendo、TLSmid、TLSepi及GLSendo、GLSmid、GLSepi均呈梯度递减（P〈0．05）；与对照组比较，冠心病轻度狭窄组及重度狭窄组TLSendo、TLSepi均显著减低（P〈0．05），而TLSmid与对照组间比较差异无统计学意义（P〉0．05）；与冠心病轻度狭窄组比较，重度狭窄组TLSendo、TLSepi均显著减低（P〈0．05），但TLSmid差异无统计学意义（P〉0．05）；与对照组比较，冠心病多支病变组及单支病变组GLSendo、GLSmid、GLSepi均显著减低（P〈0．05）；与冠心病单支病变组比较，多支病变组GLSendo、GLSmid、GLSepi均显著减低（P〈0．05）。TLSendo、TLSepi截断值分别为20．5％和15．5％时，预测冠脉狭窄≥75％的敏感性、特异性分别为55．6％、72．6％和57．8％、76．4％；GLSendo、GLSmid、GLSepi截断值分别为21．5％、17．5％、16．5％时，预测冠状动脉多支病变的敏感性、特异性分别为54．5％、71．4％；50．0Y00、87．5％和63．6％、87．5％。结论冠心病患者左室各层心肌收缩功能减低，2D-STI获取的纵向分层应变参数能准确评价冠心病患者左室心肌局部及整体收缩功能，并能有效预测冠脉狭窄程度及病变范围。

Objective To analyze longitudinal layer-specific strain in endocardial, midcardial and epicardial layers of left ventricle （LV） by two-dimensional speckle tracking imaging（2D-STI） in patients with coronary heart disease （CHD） and to evaluate the regional or global systolic function of LV. Methods Seventy-eight patients with suspected CHD were divided into CHD group （46 subjects） and control group （32 subjects） according to the results of selective coronary arteriography. According to the stenosis degree or number of coronary arteries, the CHD group was divided into mild stenosis group （corresponding blood- supply vessel stenosis 50% -75 %）, severe stenosis group （corresponding blood-supply vessel stenosis 75 %） or single-vessel stenosis group, multi-vessel stenosis group. Two-dimensional images with high frame rate were recorded in apical four-chamber, long-axis, two-chamber of LV in all subjects. The strain parameters by EchoPAC analysis software included territorial longitudinal strain （TLS） of endocardial, midcardial and epicardial layers （TLSendo, TLSmid, TLSepi） and global longitudinal strain （GLS） ofendocardial, midcardial and epicardial layers （ GLSendo, GLSmid, GLSepi）. Results Regardless of with or without corresponding blood-supply coronary artery stenosis and range of coronary artery stenosis, TLSendo, TLSmid, TLSepi and GLSendo, GLSmid, GLSepi all showed a gradient decrease （ P 〈 0.05）. Compared with the control group, TLSendo, TLSepi of mild stenosis group and severe stenosis group decreased significantly（ P 〈0. 05）, while there was no significant difference about TLSmid between the two groups（ P 〉0.05）. Compared with the mild stenosis group, TLSendo, TLSepi of severe stenosis group decreased significantly（ P 〈0.05 ）, while there was no significant difference about TLSmid（ P 〉 0.05）. Moreover, compared with the control group, GLSendo, GLSmid, GLSepi of single-vessel stenosis group and multi-vessel stenosis group decreased significantly（ P 〈 0.05）. Compared with the single-vessel stenosis group, GLSendo, GLSmid, GLSepi of multi-vessel stenosis group decreased significantly（ P 〈0.05）. For predicting corresponding blood-supply vessel stenosis ≥75 %, the sensitivity and specificity were 55.6 %, 72.6% and 57.8%, 76.4%, when the cut-off of TLSendo and TLSepi were 20.5% and 15.5M respectively. For predicting multi-vessel stenosis, the sensitivity and specificity were 54.5 %, 71.4% ;50.0%, 87.5 % and 63.6%,87.5%; when the cut-off of GLSendo, GLSmid, GLSepi were 21.5%, 17.5% and 16.5% respectively. Conclusions Systolic dysfunction of LV happened through all layers in CHD patients. The longitudinal strain of layer-specific by 2D-STI can accurately evaluate the regional and global systolic function of LV in patients with CHD, which can also be used to predict the degree or range of coronary artery stenosis.