Role of left ventricle deformation analysis in stress echocardiography for significant coronary artery disease detection: A diagnostic study meta-analysis
Role of left ventricle deformation analysis in stress echocardiography for significant coronary artery disease detection: A diagnostic study meta-analysis
复制标题
DOI：
10.1111/echo.14365
复制
发表时间：
2019-06-01
影响因子：
1.5
通讯作者：
Bajaj, Navkaranbir S.
中科院分区：
文献类型：
--
作者：
Gupta, Kartik;Kakar, Tanya S.;Bajaj, Navkaranbir S.
背景：我们比较了负荷超声心动图中纵向应变（LS）成像与室壁运动（WM）视觉评估检测显著冠状动脉疾病（CAD）的诊断准确性。方法我们的系统检索包括报告负荷超声心动图期间LS成像和WM视觉评估诊断措施检测显著CAD的研究。评估了诊断准确性指标，包括曲线下面积（AUC）、灵敏度、特异性、诊断比值比（DOR）和似然比（LR）。结果13篇研究共978例患者，其中10篇研究以有创冠状动脉造影作为参考标准。LS成像诊断显著CAD的合并AUC为0.92（95%置信区间[CI] 0.89-0.94），而WM视觉评估为0.83（95% CI 0.80-0.86），P < 0.001。LS成像的敏感性更高（88% [95%CI 84-92] vs 74% [95%CI 68-80]，P < 0.001），特异性与WM视觉评估相当（80% [95%CI 72-87] vs 83% [95%CI 74-90]，P = 0.592）。LS成像和视觉评估WM的DOR分别为31和15，P = 0.254。LS成像和WM视觉评估的阳性LR均为4.5;阴性LR分别为0.14和0.31，P = 0.002。结论负荷超声心动图纵向应变成像对冠心病的诊断准确性优于视觉评价。需要使用更大样本量和标准化应变测量技术进行研究，以进一步确定应变测量相对于单独目视评估的额外优势。
Background We compared the diagnostic accuracy of longitudinal strain (LS) imaging during stress echocardiography with visual assessment of wall motion (WM) for detecting significant coronary artery disease (CAD). Methods Our systematic search included studies reporting diagnostic measures for LS imaging and visual assessment of WM for detecting significant CAD during stress echocardiography. Summary diagnostic accuracy measures including area under the curve (AUC), sensitivity, specificity, diagnostic odds ratio (DOR), and likelihood ratios (LRs) were estimated. Results In thirteen studies with 978 patients, ten studies used invasive coronary angiography as the reference standard. Pooled AUC for diagnosing significant CAD was 0.92 (95% confidence interval [CI] 0.89-0.94) for LS imaging as compared to 0.83 (95% CI 0.80-0.86), P < 0.001 for visual assessment of WM. LS imaging had higher sensitivity (88% [95% CI 84-92] vs 74% [95% CI 68-80], P < 0.001) and comparable specificity to visual assessment of WM (80% [95% CI 72-87] vs 83% [95% CI 74-90], P = 0.592). The DOR for LS imaging and visual assessment of WM was 31 and 15, P = 0.254, respectively. The positive LR was 4.5 for both; negative LR was 0.14 and 0.31, P = 0.002 for LS imaging and visual assessment of WM, respectively. Conclusions Longitudinal strain imaging during stress echocardiography has better diagnostic accuracy for detecting significant CAD as compared to visual assessment of WM. Studies using larger sample size and standardized techniques of strain measurement are required to further ascertain the added advantage of strain measurement over visual assessment alone.