Left ventricular ejection fraction (LVEF) occupies a central place within current guidelines for implantable cardioveter-defibrillator (ICD) therapy in patients with nonischemic cardiomyopathy. The 2 largest trials recruited patients with an LVEF of ≤35% that was “stable” or “irreversible.” 1,2 Patients undergoing long-term treatment for nonischemic cardiomyopathy often have substantial improvements in LVEF, although the association with reduced arrhythmia risk has not been extensively studied. The Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE) study included 458 patients with nonischemic cardiomyopathy, New York Heart Association class (NYHA) I–III symptoms, LVEF ≤35%, and ambient arrhythmias on Holter monitoring (ventricular ectopy or nonsustained ventricular tachycardia [NSVT]). All patients received optimal medical therapy and were randomized to receive an ICD or no device therapy. The trial reported a trend toward reduced mortality in the ICD arm, but did not achieve statistical significance. This may have been due to unexpectedly low mortality in the control arm. In this issue of HeartRhythm, Schliamser et al 3 investigate the relationship between changes in LVEF and clinical events in a post hoc analysis of the DEFINITE study. The protocol specified that LVEF should be assessed at yearly intervals, but only 17% patients had these data available. However, some repeated LVEF measurements were available in 197 of 449 (44%) patients and therefore these were studied. The patients were grouped according to the change in LVEF: 45% increase (group I), o5% change (group II), or 45% decrease (group III). An LVEF improvement was associated with substantial survival benefit (deaths in each group were 3.2%, 12.6%, and 33%, respectively). An improvement in NYHA class was also associated with improved survival. However, although group I had a tendency toward reduced arrhythmic events, this did not reach statistical significance. In 70 of 197 (36%) patients, LVEF increased to an absolute value of 435% but 4 of these patients went on to have an arrhythmic event after this improvement had occurred. The results of this retrospective analysis must be interpreted with caution. A large number of patients had to be excluded because follow-up LVEF assessments were not made. The patients in whom follow-up data were available had a higher incidence of ICD shocks, were more likely to be young and white, have diabetes, and have better 6-minute walk test results. Patients were categorized according to changes in LVEF of just 5%, with LVEF determination being performed according to the local hospital’s preference. Mostly, echocardiography was used, which has variable reproducibility and the assessors of LVEF were not blinded to other clinical information. While this is a technical limitation to the study, the data have been obtained within the context of standard clinical practice and, in this sense, may be more realistic. Why might recovery of LVEF not fully predict a reduction in arrhythmia risk? Nonischemic cardiomyopathy is an umbrella term that encompasses a wide range of pathological processes, with the end result being mechanical and electrical dysfunction. LVEF is just one marker of these disease processes. Electrolyte disturbances and stretch-induced arrhythmia may also contribute to arrhythmogenicity. Histological studies have shown increased fibrosis between cardiac muscle fibers and bundles that may enhance anisotropy and promote electrical uncoupling. This can provide the substrate for reentry during ventricular fibrillation. 4 In another small study of 29 patients undergoing ablation for ventricular arrhythmias in the presence of nonischemic cardiomyopathy, fibrosis was detected with magnetic resonance imaging in approximately half of the patients. 5 It is probable that fibrosis might persist even in patients whose LVEF improves. Other clinical parameters are also known to give an indication of arrhythmia risk. Left bundle branch block may be related to the presence of fibrosis.InthestudybySchliamser et al, there was a significantly lower proportion of patients with left bundle branch block in those with LVEF improvement. Other factors such as the presence of pulmonary hypertension, hyponatremia, NYHA class, and coexistent right-sided heart failure have prognostic significance. Logically, these additional pieces of information could contribute to risk stratification for
Read full abstract