Cardiac Resynchronisation Therapy (CRT): with or without ICD back-up?The incidence of heart failure is increasing and patients are threatened to die of pump failure as well as of sudden cardiac death (SCD). Lately, cardiac resynchronisation therapy (CRT) has been introduced to treat patients with persistent heart failure and dyssynchronous left ventricular activation evidenced by complete left bundle branch block in the ECG. The COMPANION trial demonstrated, that in patients with an EF <0.35, NYHA class III or IV, and complete left bundle branch block, CRT with and without implantable cardioverter defibrillator (ICD) back-up reduced the combined endpoint of mortality and hospitalisation. A reduction in total mortality could only be achieved by the use of CRT-ICD. The CARE-HF trial randomised similar patients to CRT or medical treatment. The secondary endpoint of death from all causes was reached by significantly fewer patients in the CRT group. However, approximately one third of the deaths was attributable to SCD.To date, there is no trial available comparing the effects of CRT and CRT-ICD on total mortality as a primary endpoint. Apart from the results of the COMPANION trial, several conclusions can be drawn from primary prevention trials in the general ICD population. The SCD-HeFT trial showed a significant mortality reduction in patients with an LVEF 90.35 who received an ICD for primary prevention as compared to placebo and amiodarone. After successful treatment with CRT, the mean LVEF improved from 0.25 to 0.32 in the CARE-HF trial. The majority of patients would therefore continue to fulfil the criteria for ICD-implantation according to the SCDHeFT trial. SCD-HeFT also showed a mortality reduction in patients with NYHA class II but not in those with NYHA class III. It is conceivable, that CRT can change the mode of death from pump failure to SCD. To date, the available data indicate, that an ICD-back-up should at least be considered in every patient eligible for CRT therapy.