To understand modes of death and factors associated with the risk for cardiac and non-cardiac deaths in patients with cardiac resynchronization therapy with implantable cardioverter-defibrillator (CRT-D) vs. implantable cardioverter-defibrillator (ICD) therapy, which may help clarify the action and limitations of cardiac resynchronization therapy (CRT) in relieving myocardial dysfunction.Methods and results
In Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy (MADIT-CRT), during 4 years of follow-up, 169 (9.3%) of 1820 patients died of known causes, 108 (63.9%) deemed cardiac, and 61 (36.1%) non-cardiac. In multivariate analysis, increased baseline creatinine was significantly associated with both cardiac and non-cardiac deaths [hazard ratio (HR) 2.97, P < 0.001; HR 1.80, P = 0.035, respectively], as was diabetes (HR 1.79, P = 0.006; HR 1.73, P = 0.038, respectively), and the worst New York Heart Association Class > II more than 3 months prior to enrolment (HR 1.90, P = 0.012; HR 2.46, P = 0.010, respectively). Baseline left atrial volume index was significantly associated only with cardiac mortality (HR 1.28 per 5 unit increase, P < 0.001). Ischaemic cardiomyopathy was associated only with non-cardiac death (HR 3.54, P = 0.001). CRT-D vs. an ICD-only was associated with a reduced risk for cardiac death in patients with left bundle branch block (LBBB) (HR 0.56, P = 0.029) but was associated with an increased risk for non-cardiac death in non-LBBB patients (HR 3.48, P = 0.048).Conclusions
In MADIT-CRT, two-thirds of the deaths were cardiac and one-third non-cardiac. Many of the same risk factors were associated with both cardiac and non-cardiac mortalities. CRT-D was associated with a reduced risk for cardiac death in LBBB but an increased risk for non-cardiac death in non-LBBB.Clinical Trial Registration
Information for the MADIT-CRT main study http://www.clinicaltrials.gov, NCT00180271.