Assessment of Left Ventricular Relaxation by Untwisting Rate Based on Different Algorithms

Background: This study was performed in both animals and human subjects to test whether different approaches to calculate untwisting rate may lead to different results in the assessment of left ventricular (LV) relaxation. Methods: In animal experiments, congestive heart failure was successfully induced in 8 adult dogs. Transthoracic echocardiography was performed with simultaneous LV pressure recording at baseline and the stage of heart failure. In the clinical study, 72 patients undergoing right-sided heart catheterization were studied by transthoracic echocardiography. LV twist was calculated as the difference between apical and basal rotations measured using two-dimensional speckle tracking. Untwisting rate was calculated using 3 different algorithms as the peak negative time derivative of twist (UR_max) during early diastole, the slope of the linear regression of untwisting over time (UR_slope), or the average untwisting over the isovolumic relaxation period (UR_mean). Results: UR_max significantly correlated with tau and -dP/dt in dogs (r = -0.81 and 0.77, respectively, both P < .001) and was reduced at the stage of heart failure (P < .01). In 55 patients (76%) with adequate image quality, only UR_max among untwisting rates calculated by 3 different algorithms was significantly related to tau (r = -0.51, P < .001). UR_max was significantly lower in patients with tau ≥ 48 ms than in patients with tau < 48 ms (P = .004), most of whom had a depressed LV ejection fraction. Conclusion: UR_max best reflects LV relaxation in comparison with the 2 other algorithms.