Left atrial reservoir strain combined with E/E' as a better single measure to predict elevated LV filling pressures in patients with coronary artery disease

• Main Authors: Jingru Lin, Hong Ma, Lijian Gao, Yang Wang, Jiangtao Wang, Zhenhui Zhu, Kunjing Pang, Hao Wang, Weichun Wu
• Format: Article
• Language: English
• Published: BMC 2020-04-01
• Series: Cardiovascular Ultrasound
• Subjects: Left atrial strain; Left ventricle; Diastolic function; Filling pressure; Coronary artery disease
• Online Access: http://link.springer.com/article/10.1186/s12947-020-00192-4

Abstract Background The 2016 guidelines for left ventricular diastolic dysfunction diagnosis has been simplified from previous versions; however, multiparametric diagnosis approach still exists indeterminate left ventricular diastolic dysfunction category. Left atrial (LA) strain was recently found useful to predict elevated left ventricular (LV) filling pressures noninvasively. This study aimed to (1) analyze the diagnostic value of LA strain for noninvasive assessment of LV filling pressures in patients with stable coronary artery disease (CAD) with preserved LV ejection fraction (LVEF), using invasive hemodynamic assessment as the gold standard, and (2) explore whether LA strain combined with conventional diastolic parameters could detect elevated LV filling pressures alone. Methods Sixty-four patients with stable CAD having LVEF > 50% and 30 healthy controls were enrolled. Two-dimensional speckle-tracking echocardiography was used to measure LA strain during the reservoir (LASr), conduit, and contraction phases. LV end-diastolic pressure (LVEDP), as a surrogate for LV filling pressures, was invasively obtained by left heart catheterization. Logistic regression was used to calculate the odds ratio to predict LV filling pressures. Pearson’s correlation was used to analyze associations between echocardiographic parameters and LVEDP. The area under the receiver-operating characteristic curve was calculated to determine the capability of the echocardiographic parameters to detect elevated LVEDP. Inter-technique agreement was analyzed by contingency tables and tested by kappa statistics. Results LASr and the ratio of early-diastolic transmitral flow velocity (E) to tissue Doppler early-diastolic septal mitral annular velocity (E/E′septal) significantly predicted elevated LV filling pressures. LASr was combined with E/E′septal to generate a novel parameter (LASr/E/E′septal). LASr/E/E′septal had the best predictive ability of elevated LV filling pressures. LVEDP was negatively correlated with LASr and LASr/E/E′septal but positively correlated with E/E′septal. The area under the receiver-operating characteristic curve of LASr/E/E′septal was higher than that of LASr alone (0.83 vs. 0.75), better than all conventional LV diastolic parameters. Inter-technique agreement analysis showed that LASr/E/E′septal had good agreement with the invasive LVEDP measurement, better than the 2016 guideline (kappa = 0.63 vs. 0.25). Conclusions LASr provided additive diagnostic value for the noninvasive assessment of LV filling pressures. LASr/E/E′septal had the potential to be a better single noninvasive index to predict elevated LV filling pressures in patients with stable CAD and preserved LVEF.