126 Inter-ventricular delay at peak exercise is independently correlated with left ventricular remodeling at three months in heart failure patients selected for cardiac resynchronisation therapy

Abstract

Cardiac resynchronisation therapy (CRT) is known to improve survival and to induce beneficial left ventricle (LV) remodeling in heart failure (HF) patients presenting with severe LV systolic dysfunction and prolonged QRS interval. Using those selection criteria, about 35 to 40% of patients remain non-responders to CRT. We assessed whether exercise inter-ventricular dyssynchrony could be a relevant index for the prediction of response to CRT. Eighteen HF patients performed a semi-supine symptom-limited exercise echocardiography before device implantation. Inter-ventricular dyssynchrony, assessed by inter-ventricular mechanical delay (IVMD) in pulsed wave doppler, was recorded both at rest and peak exercise. We determined correlations between IVMD at exercise and cardiac output, mitral regurgitation, assessed by the effective regurgitant orifice area (ERO), and right ventricular (RV) longitudinal systolic function, assessed by tissue doppler imaging. We looked for correlations between those parameters and the degree of LV reverse remodeling at three months follow-up assessed by the percentage of change in LV end-sytolic volume (%ΔESV) and absolute value of change in ejection fraction (ΔLVEF). 39% of patients were non-responders according to LV remodeling. Exercise-induced changes in IVMD only significantly correlated with RV longitudinal systolic function (r= - 0.602; p = 0.01) at peak exercise. In a multivariate analysis by stepwise regression, IVMD at peak exercise was independently correlated with %ΔESV (β = 0.418; p = 0.01) and ΔLVEF (β = -0.563, p = 0.03) at follow-up. Peak exercise ERO was the only other independent factor correlated with %ΔESV (β = 0.808, p < 0.001) and ΔLVEF (β = -0.404, p = 0.02). In HF patients, IVMD at peak exercise seems to be a relevant independent predictor of mid-term LV remodeling after CRT. Its relation with RV longitudinal systolic function during exercise could provide a pathophysiological explanation.