Remission of left ventricular systolic dysfunction and of heart failure symptoms after cardiac resynchronization therapy: Temporal pattern and clinical predictors

Background The aim of the study was to determine whether cardiac resynchronization therapy (CRT) may induce a heart failure (HF) remission phase (recovery to New York Heart Association functional class I-II and regression of left ventricular [LV] dysfunction: LV ejection fraction [EF] >= 50%) and to define the incidence and predictors of such a process. Methods Cardiac resynchronization therapy devices were successfully implanted in 520 consecutive HF patients from 1999 to 2006 (mean age 66 years, 82% male sex, New York Heart Association class >= II, LVEF 28%, QRS 164 milliseconds, 6-minute hall walk distance 302 m) at our institution. Follow-up data were prospectively collected every 3 to 6 months. Continuous variables were stratified in tertiles. Results Over a median follow-up of 28 months, 26% of patients achieved LV remission (rate: 16 per 100 person-years). At univariate analysis, female sex (P = .032), non-coronary artery disease (CAD) etiology (P < .001), mitral regurgitation <2/4 (P = .022), higher EF tertile (P < .001), lower diameter and volume tertiles (both P < .001), previous conventional right ventricle pacing (P = .029), and post-CRT-paced QRS (P = .008) predicted remission. At multivariate analysis, non-CAD etiology, LVEF 30% to 35%, and LV end-diastolic volume <180 mL were strongly associated with HF remission phase (all P < .001). Concomitance of these 3 factors yielded a significantly higher remission rate compared with either no or only 1 factor (respectively, 60 vs 7 and 11 per 100 person-years, P < .001). Conclusions Cardiac resynchronization therapy induces HF remission phase in 26% of patients, even after 3 years. Non-CAD etiology and moderately compromised LV function at baseline may easily predict this process.