Determinants of Mortality in Patients Undergoing Cardiac Resynchronization Therapy: Baseline Clinical, Echocardiographic, and Angioscintigraphic Evaluation Prior to Resynchronization

were NYHA Class IV (P < 0.001), and need for dobutamine the month preceding CRT (P < 0.008), while use of β-blocking agents (P < 0.08) and left ventricular EF (P < 0.09) were negative ones. NYHA Class IV was the only independent predictor at multivariate analysis (P < 0.01). Survival at 24 months was 85% in Class II, 80% in Class III, and 37% in Class IV (II vs III, P = ns; III vs IV, P < 0.001). When using a composite endpoint of death from any cause and unplanned rehospitalization for a major cardiovascular event, there were 48 events (14 HF deaths, 3 noncardiac deaths, 26 HF rehospitalizations, 2 paroxysmal atrial fibrillation, 2 sustained ventricular tachycardia, 1 nonfatal pulmonary embolism). Predictors of death from any cause/unplanned rehospitalization for a major cardiovascular event in the follow-up were NYHA Class IV (P < 0.001), need for dobutamine during the month preceding CRT (P < 0.002), and PAP (<0.02). NYHA Class IV was the only independent predictor at multivariate analysis (P < 0.05). Event-free proportion at 24 months was 70% in Class II, 64% in Class III, and 37% in Class IV (II vs III, P = ns; III vs IV, P < 0.01). When considering determinants of mortality only in NYHA Class IV patients, no variable was significantly correlated to mortality. Need for dobutamine during the last month preceding CRT did not add an adjunctive mortality risk. Conclusion: Baseline NYHA Class IV at implantation appears as the most important determinant of a poor clinical outcome in terms of both mortality and morbidity. No predictive criteria seem available for NYHA Class IV patients, in order to discriminate who will die after CRT and who will not. NYHA Class IV strongly influences the clinical outcome, suggesting that, in future studies planned on mortality and rehospitalization as major endpoints, baseline NYHA Class IV should be separately taken into account. (PACE 2005; 28:1260‐1270)