RESUMO
BACKGROUND: There are conflicting data on the effect of cardiac resynchronization therapy with a defibrillator (CRT-D) on the risk of life-threatening ventricular tachyarrhythmia in heart failure patients. OBJECTIVES: The authors aimed to assess whether QRS morphology is associated with risk of ventricular arrhythmias in CRT recipients. METHODS: The study population comprised 2,862 patients implanted with implantable cardioverter defibrillator (ICD)/CRT-D for primary prevention who were enrolled in 5 landmark primary prevention ICD trials (MADIT-II [Multicenter Automated Defibrillator Implantation Trial], MADIT-CRT [Multicenter Automated Defibrillator Implantation Trial-Cardiac Resynchronization Therapy], MADIT-RIT [Multicenter Automated Defibrillator Implantation Trial-Reduction in Inappropriate Therapy], MADIT-RISK [Multicenter Automated Defibrillator Implantation Trial-RISK], and RAID [Ranolazine in High-Risk Patients With Implanted Cardioverter Defibrillators]). Patients with QRS duration ≥130 ms were divided into 2 groups: those implanted with an ICD only vs CRT-D. The primary endpoint was fast ventricular tachycardia (VT)/ventricular fibrillation (VF) (defined as VT ≥200 beats/min or VF), accounting for the competing risk of death. Secondary endpoints included appropriate shocks, any sustained VT or VF, and the burden of fast VT/VF, assessed in a recurrent event analysis. RESULTS: Among patients with left bundle branch block (n = 1,792), those with CRT-D (n = 1,112) experienced a significant 44% (P < 0.001) reduction in the risk of fast VT/VF compared with ICD-only patients (n = 680), a significantly lower burden of fast VT/VF (HR: 0.55; P = 0.001), with a reduced burden of appropriate shocks (HR: 0.44; P < 0.001). In contrast, among patients with non-left bundle branch block (NLBBB) (N = 1,070), CRT-D was not associated with reduction in fast VT/VF (HR: 1.33; P = 0.195). Furthermore, NLBBB patients with CRT-D experienced a statistically significant increase in the burden of fast VT/VF events compared with ICD-only patients (HR: 1.90; P = 0.013). CONCLUSIONS: Our data suggest a potential proarrhythmic effect of CRT among patients with NLBBB. These data should be considered in patient selection for treatment with CRT.