Predictors and long-term outcome of super-responders to cardiac resynchronization therapy

BackgroundThe level of improvement in left ventricular ejection fraction (LVEF) in super-responders to cardiac resynchronization therapy (CRT) is exceptional. However, the long-term prognosis remains unknown in a large population. HypothesisWhether super-responders haven good long-term outcomes. MethodsWe registered 347 patients with primary CRT-D indication. Super-response was defined by LVEF >50% at follow-up echocardiogram. Best-subset regression analysis identified predictors of super-response. Endpoints were major adverse cardiac events (MACE; eg, all-cause mortality or heart failure hospitalization, cardiac death, and appropriate ICD therapy). ResultsFifty-six (16%) patients with LVEF >50% were classified as super-responders. Female sex (OR: 3.06, 95% CI: 1.54-6.05), nonischemic etiology (OR: 2.70, 95% CI: 1.29-5.68), higher LVEF at baseline (OR: 1.07, 95% CI: 1.02-1.13), and wider QRS duration (OR: 1.17, 95% CI: 1.04-1.32) were predictors of super-response. Cumulative incidence of MACE at a median of 5.3 years was 18% in super-responders, 22% in responders, and 51% in nonresponders (P<0.001). None of super responders died from cardiac death, compared to 9% of responders and 25% of non-responders (P <0.001). None of super-responders experienced appropriate ICD therapy, compared with 10% of responders and 21% of non-responders (P<0.001). In super-responders, the adjusted hazard ratio was 0.37 (95% CI: 0.19-0.73) for MACE and 0.44 (95% CI: 0.20-0.95) for total mortality, compared with non-responders. ConclusionsFemale sex, non-ischemic etiology, higher baseline LVEF, and wider QRS duration were independently associated with super-response. Super-response was associated with persistent excellent prognosis regarding survival and appropriate ICD therapy during long-term follow-up.