Complete Left-Sided Reverse Remodeling With CRT

Study Questions:

What is the rate and predictive value of complete left-sided reverse remodeling (reduction in both left ventricular end-systolic volume [LVESV] and left atrial volume [LAV]) on heart failure (HF) and death events in cardiac resynchronization therapy with a defibrillator (CRT-D) patients with left bundle branch block (LBBB)?


This substudy of MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial–Cardiac Resynchronization Therapy) comprised 533 CRT-D patients with LBBB, 212 (40%) with complete left-sided remodeling (above-median change in both LAV and LVESV), 115 (22%) with discordant remodeling (above-median change in only LAV or LVESV), and 206 (38%) of them with lesser remodeling (below-median LAV and LVESV change). The primary endpoint was HF or death; secondary endpoints included HF alone and death alone during long-term follow-up.


Patients with complete left-sided remodeling had a significantly lower rate of HF or death than those with discordant remodeling or lesser remodeling (p < 0.001). Greater than 20% of CRT-D patients exhibited discordant reverse remodeling in the left ventricle and the left atrium. There was a decreased risk for HF and death in patients with complete remodeling compared with discordant remodeling or lesser remodeling (hazard ratio, 0.66; p = 0.002). This finding was similar for HF alone and death alone.


Patients with LBBB and complete left-sided remodeling had a significantly lower risk of HF and death, HF alone, and death alone during long-term follow-up, compared with patients with discordant or lesser remodeling.


CRT results in functional improvement and increased survival in select patients. Despite the fact that CRT is a mature field, it has struggled to find the optimal surrogate outcome parameter, which would be easy to obtain and highly reproducible, yet succinctly reflect reverse remodeling of the left ventricle. LVESV, LV end-diastolic volume, change in LVESV, and more recently LAV have been most consistently reported, and probably most meaningful. The present study offers an insight on how combining indexes of LV and LA remodeling may enhance the prediction of clinical outcomes. It is unclear why some patients show improvement in only one of the two parameters, and this deserves further study. The degree of fibrosis of the ventricle and the atrium, potentially quantifiable with cardiac magnetic resonance imaging, is possibly related to the remodeling potential.

< Back to Listings