Redefining the Classifications of Response to Cardiac Resynchronization Therapy: Results From the REVERSE Study

Objectives: This study sought to assess the impact of a more detailed classification of response on survival.

Background: Cardiac resynchronization therapy (CRT) improves functional status and outcomes in selected populations with heart failure (HF). However, approximately 30% of patients do not improve with CRT by various metrics, and they are traditionally classified as nonresponders.

Methods: REVERSE (Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction) was a randomized trial of CRT among patients with mild HF. Patients were classified as Improved, Stabilized, or Worsened using prespecified criteria based on the clinical composite score (CCS) and change in left ventricular end-systolic volume index (LVESVi). All-cause mortality across CRT ON subgroups at 5 years was compared.

Results: Of the 406 subjects surviving 1 year, 5-year survival differed between CCS subgroups (p = 0.03), with increased mortality in the Worsened response group. Of the 353 subjects with adequate echocardiograms, survival differed significantly between response groups (p < 0.001), also due to increased mortality in the Worsened group. When combining CCS and LVESVi results, the lowest survival was observed among subjects who worsened for both measures, whereas the highest survival occurred in subjects who did not worsen by either endpoint. Multivariate analysis showed that LVESVi worsening with CRT at 6 months, baseline LVESVi, and gender were independent predictors of survival.

Conclusions: For both CCS and reverse remodeling, patients who worsen with CRT have a high mortality, although remodeling was the more important endpoint. Patients who stabilize early with CRT have a much better prognosis than previously recognized, suggesting that the current convention of nonresponder classification should be modified. (REVERSE [Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction]; NCT00271154).