Evaluating left atrial strain and left ventricular diastolic strain rate as markers for diastolic dysfunction in patients with mitral annular calcification

Abstract

Abstract Background Mitral annular calcification (MAC) poses many challenges to the evaluation of diastolic function using standard echocardiography. Left atrial (LA) strain and left ventricular early diastolic strain rate (DSr) measured by speckle-tracking echocardiography (STE) are emerging techniques in the noninvasive evaluation of diastolic function. We aim to evaluate the utility of LA strain and early DSr in predicting elevated left ventricular filling pressures (LVFP) in patients with MAC and compare their effectiveness to ratio of mitral inflow velocity in early and late diastole (E/A). Methods We included adult patients with MAC who presented between January 1 and December 31, 2014 and received a transthoracic echocardiogram (TTE) and cardiac catheterization with measurement of LVFP within a 24-h period. We used Spearman’s rank correlation coefficient to assess associations of LA reservoir strain and average early DSr with LVFP. Receiver operating characteristic (ROC) curves were computed to assess the effectiveness of LA strain and DSr in discriminating elevated LVFP as a dichotomized variable and to compare their effectiveness with E/A ratio categorized according to grade of diastolic dysfunction. Results Fifty-five patients were included. LA reservoir strain demonstrated poor correlation with LVFP (Spearman’s rho = 0.03, p = 0.81) and poor discriminatory ability for detecting elevated LVFP (AUC = 0.54, 95% CI 0.38–0.69). Categorical E/A ratio alone also demonstrated poor discriminatory ability (AUC = 0.53, 95% CI 0.39–0.67), and addition of LA reservoir strain did not significantly improve effectiveness (AUC = 0.58, 95% CI 0.42–0.74, p = 0.56). Average early DSr also demonstrated poor correlation with LVFP (Spearman’s rho = −0.19, p = 0.16) and poor discriminatory ability for detecting elevated LVFP (AUC = 0.59, 95% CI 0.44–0.75). Addition of average early DSr to categorical E/A ratio failed to improve effectiveness (AUC = 0.62, 95% CI 0.46–0.77 vs. AUC = 0.54, 95% CI 0.39–0.69, p = 0.38). Conclusions In our sample, LA reservoir strain and DSr do not accurately predict diastolic filling pressure. Further research is required before LA strain and early DSr can be routinely used in clinical practice to assess filling pressure in patients with MAC.