The aim of this study was to evaluate the association between improvement in left ventricular end-systolic volume (LVESV) with cardiac resynchronization therapy (CRT) and mortality and whether this relationship was modified by the presence of a left bundle branch block (LBBB) electrocardiographic pattern.Left ventricular reverse remodeling in patients receiving CRT has been shown to predict outcomes. However, the extent to which reverse remodeling contributes to long-term survival is not well understood.Changes in LVESV were assessed in MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy) patients receiving CRT with a defibrillator (CRT-D) and echocardiograms available at 1 year (n = 752), stratified by LBBB, relative to long-term all-cause mortality, compared with those with implantable cardioverter-defibrillators (ICDs) only (n = 684).In patients with LBBB, a reduction in LVESV of >35% (median) translated into significantly lower risk for long-term mortality (hazard ratio [HR]: 0.34; p < 0.001), heart failure (HF) events (HR: 0.21; p < 0.001), and HF or death (HR: 0.27; p < 0.001) compared with patients with ICDs only. Patients with reductions in LVESV ≤35% had a significantly lower risk for HF, and HF or death, and a nonsignificantly lower rate of death compared with those with ICDs only (HR: 0.74; p = 0.13). Risk reduction in HF events was uniform across all LVESV quartiles. In patients without LBBB, there was no survival benefit (HR: 0.68; p = 0.271) despite an LVESV reduction greater than the median (>27.6%). CRT-D patients without LBBB with the least reverse remodeling (quartile 1) had a more than 3-fold increased risk for death compared with those with ICDs only (HR: 3.11; p < 0.001).In patients with LBBB, CRT-D-induced reduction in LVESV at 1 year is associated with long-term survival benefit. Despite left ventricular reverse remodeling with CRT-D, there is no survival benefit and potential harm in patients without LBBB.