超声心动图分层应变技术评价左心室射血分数正常的肥厚型心肌病患者室壁收缩功能的改变

Assessment of left ventricular systolic function in hypertrophic cardiomyopathy patients with normal left ventricular ejection fraction by using echocardiography layer strain

目的探讨二维斑点追踪成像(2D-STI)技术定量分析左心室射血分数(LVEF)正常的肥厚型心肌病(HCM)患者左心室心肌分层应变的临床应用价值。方法选取2015年1月至2016年1月在大连医科大学附属第一医院就诊的经临床与超声心动图确诊的HCM患者28例(均为非梗阻型HCM),作为HCM组;选取20名健康志愿者作为健康对照组。采用STE技术获得左心室整体及各层各段收缩期峰值纵向应变(LPS),计算跨壁阶差(△LS)及跨壁阶差百分比(△LS%)等指标进行分析。HCM组与健康对照组心内膜整体纵向应变(LPSE_(Endo))、中层心肌整体纵向应变(LPS_(Mid))、心外膜整体纵向应变(LPSE_(Epi))、基底段心肌纵向应变(LPS_b);中间段心肌纵向应变(LPS_m);心尖段心肌纵向应变(LPS_a);左心室整体纵向应变(GLPS)等左心室心肌应变值、△LS、△LS%等采用独立样本t检验比较。HCM组、健康对照组左心室各层各段及整体心肌间纵向应变,组间比较采用两独立样本t检验,组内比较采用LSD-t检验。结果 HCM组与健康对照组组内左心室各层各段LPS均存在梯度特征LPS_(Endo)与LPS_(Mid)比较[(18.36±4.97)%vs(13.80±4.23)%,(26.41±2.93)%vs(22.19±2.49)%],差异均有统计学意义(t=3.690、4.913,P均<0.05);LPS_(Endo)与LPS_(Epi)比较[(18.36±4.97)%vs(11.91±3.63)%,(26.41±2.93)%vs(19.43±2.20)%],差异均有统计学意义(t=5.550、8.529,P均<0.05);健康对照组LPS_(Mid)与LPS_(Epi)比较[(22.19±2.49)%vs(19.43±2.20)%],差异有统计学意义(t=3.709,P<0.05),即LPS_(Endo)>LPS_(Mid)>LPS_(Epi);LPS_a与LPS_m比较,差异有统计学意义(t=4.029、6.839,P均<0.05);LPS_a与LPS_b比较,差异有统计学意义(t=5.304、9.887,P均<0.05);健康对照组LPS_m与LPS_b比较,差异有统计学意义(t=4.170,P<0.05);即LPS_a>LPS_m>LPS_b。与健康对照组比较,HCM组左心室整体及各层各段心肌的LPS降低,差异均有统计学意义[GLPS:(14.63±3.75)%vs(22.68±2.51)%,t=-8.347;心内膜至心外膜LPS:t=-6.477、-7.909、-8.242;心尖段至基底段LPS:t=-6.647、-8.790、-7.267;P均<0.05]。HCM组左心室各节段及整体△LS较健康对照组降低,但差异均无统计学意义(P均>0.05)。HCM组与健康对照组组内左心室各节段心肌△LS%由心尖段至基底段依次减低,差异均有统计学意义(HCM组:t=9.985、5.969;健康对照组:t=17.513、7.043;P均<0.05)。HCM组心尖段、中间段的△LS%均较健康对照组高[(58.86±11.32)%vs(43.70±4.73)%,(28.43±11.48)%vs(20.30±3.66)%],差异均有统计学意义(t=5.634、3.049,P均<0.05)。结论 (1)心肌分层应变技术可以准确判断HCM患者左心室局部或整体的收缩功能。(2)△LS%可能在反映△LS变化方面敏感性更好,尚需更多的研究以探讨其应用价值。

Objective To investigate the value of quantitative analysis of the left ventricular longitudinal strain in patients with hypertrophic cardiomyopathy （HCM） and with normal left ventricular ejection fraction （LVEF） by using two-dimensional speckle tracking imaging. Methods Twenty-eight HCM patients with normal LVEF （all of the cases were non obstructive HCM）, who were diagnosed by clinical and ultrasound echocardiography between January 1, 2015 and January 1, 2016 in the First Affiliated Hospital of Dalian Medical University, served as the experimental group. And twenty healthy volunteers served as the healthy control group. The peak longitudinal strain （LPS） of the left ventricle and the systolic peak of the left ventricle were calculated by the STE technique. The indexes such as the transmural gradient （ △LS=LPSEndo-LPSEpi） and the transmural gradient percentage （△LS%= △LS/LPSEnao） were calculated. The Peak systolic longitudinal strain of endocardium （LPSEndo）, the peak systolic longitudinal strain of mid-cardium （LPSMid）, the peak systolic longitudinal strain of epicardium （LPSEpi）, the peak systolic longitudinal strain of basal segment （LPSb）, the peak systolic longitudinal strain of middle segment （LPSm）, the peak systolic longitudinal strain of apical segment （LPSa）, the global peak systolic longitudinal strain （GLPS） and other left ventricular myocardial strain, such as /k LS, /- LS%, in both the HCM group and the healthy control group, were analyzed by using independent samples t test comparison. For each layer of the left ventricle and the overall myocardial longitudinal strain, two independent sample t test was used for comparison between groups, and LSD-t test was used for intra-group comparison. Results （1） There was a gradient of LPS among the three layers and the three segments in both of the two groups： LPSEndo and LPSMid [（18.36±4.97）% vs （13.80±4.23）%, （26.41±2.93）% vs （22.19±2.49）%], the difference was statistically significant （t=5.550, 8.529, P 〈 0.05）; LPSEndo and LPSEpi [（18.36±4.97） % vs （11.91 ± 3.63）%, （26.41 ±2.93）% vs （19.43 ±2.20）%], the difference was statistically significant （t=5.550, 8.529, P 〈 0.05）; There was significant difference between LPSMid and LPSEpi in the healthy control group [（22.19±2.49）% vs （19.43±2.20）%, t=3.709, P 〈 0.05）], that was, LPSEnao 〉 LPSMid 2〉 LPSEpi. LPSa and LPSm, the difference was statistically significant （t=4.029, 6.839, P 〈 0.05）; LPSa and LPSb, the difference was statistically significant （t=5.304, 9.887, P 〈 0.05）; There was significant difference between LPSm and LPSb in the healthy control group （t=4.170, P 〈 0.05）; that was, LPSa 〉 LPSm 〉 LPSb. In the HCM group, LPS in the 3 layers, 3 segments, and the whole left ventricular wall were lower than that of the the healthy control group, the differences were statistically significant [GLPS： （14.63±3.75）% vs （22.68± 2.51）%, t=-8.347; LPSEndo to LPSEpi： t=-6.477, -7.909, -8.242; LPSa to LPSb： t=-6.647, -8.790, -7.267; all P 〈 0.05）. （2） Compared with the healthy control group, both the segmental gradient and global transmural gradient in the HCM group were found reduced, but the difference had no statistical significance （all P 〉 0.05）. （3） The transrnural gradient percentage both in the healthy control group and the HCM group were reduced from the apical segment to the basal segment, the difference were statistically significant （HCM group： t=9.985, 5.969; healthy control group： t=17.513, 7.043; all P 〈 0.05）. Compared with the healthy control group, the △LS%a and the △LS%m of HCM group were significantly higher [（58.86±11.32）% vs （43.70±4.73）%, （28.43±11.48）% vs （20.30±3.66）%], and the difference was statistically significant （t=5.634, 3.049, all P 〈 0.05）. Conclusions （1） Using 2D-STI could accurately determine the regional or the global left ventricular systolic function in patients with HCM. （2） The transmural gradient percentage can be more sensitive to reflect the change of the transmural gradient, and more research needed to explore its value for clinical application.