Introducing ICD-resistant mortality as an end point to evaluate the clinical efficacy of an implantable cardioverter-defibrillator in ischaemic cardiomyopathy.

OBJECTIVE: A new end point called ICD-resistant mortality was evaluated to assess the clinical efficacy of ICD implantations. METHODS AND RESULTS: In 302 ICD patients with ischaemic cardiomyopathy, we investigated which clinical parameters predicted useful ICD implantations using cumulative incidence competing risk analysis. Implantation was deemed clinically useful when the ICD provided appropriate therapy and the patient survived implantation by 1 year and the first shock by 30 days. ICD-resistant mortality (ICDRM) was defined as death within 30 days after the first shock, within 1 year of implantation or without previous appropriate ICD therapy. After 5 years, ICDRM occurred in 23% of implantations, while 36% were clinically useful. After multivariable analysis, ICDRM was associated with LVEF <35% (HR: 2.63; p = .005), beta-blocker dose <50% (HR: 2.0; p = .01) and anterior or diffuse infarct location (HR: 3.61; p = .001 and HR: 2.89; p = .02). Useful ICD implantations were associated with beta-blocker dose <50% (HR: 1.64; p = .02) and non-anterior infarct location (HR: 3.22 vs anterior and 1.59 vs diffuse; combined p<.001). CONCLUSIONS: Five years after implantation, an ICD could be classified as useful in 1 out of 3, while ICDRM occurred in one out of four patients. At 10 years, up to 80% of implantations could be categorized. Lower LVEF was related with significantly higher incidence of ICDRM. Anterior infarcts were associated with more ICDRM and less useful implantations than non-anterior infarcts. Future risk stratification for ICD should focus more on the discrimination between arrhythmic risk, probably preventable by ICDs and ICD-resistant mortality risk.