Are nonsustained ventricular tachycardias predictive of major arrhythmias in patients with dilated cardiomyopathy on optimal medical treatment?

BACKGROUND: To evaluate the role of nonsustained ventricular tachycardias (NSVT) for the prediction of major ventricular arrhythmias (MVA) in patients with idiopathic dilated cardiomyopathy (DCM) after optimization of medical treatment. METHODS AND RESULTS: Three hundred nineteen consecutive DCM patients were evaluated after adequate stabilization on optimal angiotensin-converting enzyme (ACE) inhibitor (88%) and beta-blocker (82%) therapy. Frequency, length, and rate of NSVT at 24-hour Holter monitoring were analyzed to assess their values in predicting MVA (unexpected sudden death, SVT, ventricular fibrillation, and appropriate implantable cardioverter defibrillator interventions). During follow-up (median 96 months, 1(st)-3(rd) interquartile range 52-130), MVA incidence was low, and not statistically different between patients with and without NSVT (3 and 2 per 100 patient-years, respectively, P = nonsignificant [NS] at log-rank analysis). At multivariable analysis, the number of NSVT was predictive of MVA only if left ventricular ejection fraction (LVEF) was > 0.35 (two NSVT/day vs no NSVT/day: hazard ratio [HR] 5.3, 95% confidence interval [CI] 1.59-17.85 in LVEF > 0.35 vs HR 0.93, 95% CI 0.3-2.81 in LVEF < or = 0.35). Consequently, in patients with LVEF < or = 0.35, MVA incidence rates were similar regardless of NSVT (3.6 and 4.1 patient-years, respectively, in those with and without NSVT, P = NS), while in patients with LVEF > 0.35, MVA incidence (3.1 per 100 patient-years vs 0.9 per 100 patient-years, P = 0.003) was significantly higher when NSVT were present. CONCLUSIONS: After medical stabilization, NSVT did not increase the risk of MVA in patients with DCM and LVEF < or = 0.35. Conversely, the number and length of NSVT runs were significantly related to the occurrence of MVA in the patients with LVEF > 0.35.