: Cardiac resynchronization therapy CRT is a well-established, ef cient strategy for medically-refractory congestive heart failure HF with ventricular conduction disturbances. However, about 30 of patients who undergo CRT do not receive any bene t. Therefore, we investigated the usefulness of the QRS-left ventricle LV interval for predicting responders during CRT implantation. This study included 66 patients who underwent CRT implantation. The de nition of responder was a ≥ 15 reduction in LV end-systolic volume or ≥ 20 increase in LV ejection fraction. The QRS-LV interval was measured from the beginning of the body surface electrocardiogram QRS complex to the LV potential recorded by LV leads. We analyzed the correlations between the QRS-LV intervals and CRT responders, admission for HF and mortality. The patients were 67 12 years old, and their mean LV ejection fraction was 26.3 8.3 . During follow-up 27.2 19.9 months , 27 patients were admitted for HF 40.1 , and 17 died 25.7 ; the median QRS-LV interval was 103 33 msec. Patients were divided into 2 groups: wide QRS-LV 103 msec , and narrow QRS-LV 103 msec . The wide QRS-LV group had a lower mortality rate than the narrow QRS-LV group 77 vs. 53 , P 0.05 . In patients with dilated cardiomyopathy, the QRS-LV interval was signi cantly wider in responders, compared to non-responders 112 9.2 vs. 80.0 10 msec, P 0.05 . The QRS-LV interval did not correlate with CRT responders or admission for HF. The mortality rate was lower in patients with wide QRS-LV intervals, compared to narrow QRS-LV intervals. Furthermore, a wide QRS-LV interval might be a predictor for CRT responders in patients with dilated cardiomyopathy.