[ECG score to predict ICD therapies in patients with nonischemic cardiomyopathy and primary prophylactic CRT-D].

BACKGROUND AND PROBLEM: Recently published results of the DANISH study raise concerns, if primary prophylactic ICD implantations in patients with nonischemic cardiomyopathy (NICM) and severe reduced left ventricular ejection fraction (LVEF) should be performed without further risk stratification. There was no significant difference in the overall mortality of patients with or without ICD and CRT defibrillator (CRT-D) or CRT pacemaker (CRT-P), respectively. Clinical risk scores to identify patients with ischemic cardiomyopathy (ICM) who benefit most from an ICD have been recommended. The need for risk stratification systems concerning patients with NICM has been emphasized.

STUDY DESIGN AND METHODS: A retrospective study of 434 consecutive patients with CRT-D implantation was performed. Patients with no regular follow-up at our institution (n = 132), secondary prophylactic ICD indication (n = 61), and upgrade to CRT (n = 95) were excluded. The occurrence of an adequate ICD therapy was defined as the endpoint. Left ventricular ejection fraction (LVEF), genesis of the cardiomyopathy as well as the modified Selvester ECG score (MSES) for evaluation of the left ventricular scar burden were documented among other characteristics.

RESULTS: Within a median follow-up of 605 days, 24% of the patients experienced an adequate ICD therapy. These patients had significantly lower LVEF (20% vs. 23%), and the MSES was higher (7 vs. 3 points). There was no significant difference in patients suffering from ICM vs NICM. A receiver-operating-characteristic (ROC) analysis revealed a sensitivity of 0.914 and a specifity of 0.586 for MSES ≥4 to predict the occurrence of an ICD therapy. None of 35 patients suffering from NICM with MSES <4 experienced an ICD therapy.

INTERPRETATION: The evaluation of the left ventricular scar burden using MSES can be useful for the decision between CRT-D and CRT-P in patients suffering from NICM.