Background .– Mechanical dyssynchrony along with its role in cardiac resynchronization therapy (CRT) has been studied for over 10 years now. Still, there are no recommendations for the use of imaging techniques to best select the patients who are the most likely to positively answer to CRT. We sought, then, to test up-to-date echocardiographic tools at rest and during a standardized exercise.

Patients and results .– Thirty out of 197 patients were prospectively recruited. All were chronic heart failure patients in sinus rhythm, optimally pharmacologically treated but still in NYHA III and able to perform some degree of exercise. They were followed at 6-month. Patients were responders according to a decrease in left ventricular end-systolic volume > 15%. Responders and non-responders were significantly different at baseline according to QRS-width (162 ± 18 vs. 144 ± 22 ms, P = 0.01), left atrial volume (33 ± 10 vs. 53 ± 12 mL/m ² , P < 0.001), mitral inflow duration/RR at rest and exercise (0.32 ± 0.05 vs. 0.37 ± 0.08, P = 0.03), left pre-ejection time delay (129 ± 36 vs. 104 ± 34, P = 0.03) right atrial area, right ventricular annulus s’ and TAPSE at rest and during exercise (TAPSE exercise 21.2 ± 2.9 vs. 16 ± 6.5, P = 0.004). Using a stepwise multivariate logistic regression, the exercise mitral inflow duration and the degree of deformation in the LV lateral wall during exercise were the two best independent predictors of response.

Conclusion .– Prediction of response to CRT has to be based on a multivariable analysis including RV function, LA size, but also dyssynchrony and regional LV function.