Competing risk analysis of ventricular arrhythmia events in heart failure patients with moderately compromised renal dysfunction

Aims: Patients with impaired renal function were shown to have an attenuated benefit from implantable cardioverter-defibrillator. However, there are limited data on the competing risk of ventricular arrhythmia events and death by renal function in patients without severe disease. Therefore, we aimed to assess the competing risk of ventricular arrhythmia events and death by renal function.

Methods and results: We analysed 1782 patients (99%) enrolled in Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy (MADIT-CRT) with glomerular filtration rate (GFR) data available. Cumulative incidence function curves were used to display the rate of ventricular tachycardia (VT), ventricular fibrillation (VF), and the competing risk of death without experiencing VT/VF. Multivariable Fine and Gray regression models and recurrent event analysis were performed. There were 355 (20%) patients with GFR < 52 and 1427 with GFR ≥ 52 (lowest quintile). The incidence of non-fatal VT/VF at 4 years was higher in patients with high GFR (26%) as compared to low GFR (16%), whereas rates of death without non-fatal VT/VF were 5% and 20% (P < 0.001). In Fine and Gray models, the low GFR group was 35% less likely to experience VT/VF compared to the high GFR [95% confidence interval (CI) 0.48-0.88, P = 0.005]. In contrast,death without experiencing VT/VF was 3.5-fold higher in the low GFR group (95% CI 2.38-5.12, P-value < 0.001). Recurrent event analysis consistently showed a lower risk of recurrent VT/VF, recurrent anti-tachycardia pacing only, and shock in the low GFR group.

Conclusions: We show, in a competing risk model, a lower risk of VT/VF events and a higher risk of mortality without an arrhythmic event in patients with moderate renal dysfunction in MADIT-CRT. These findings can be used for improved selection of patients for defibrillator therapy among CRT candidates.

Clinical trial registration: https://clinicaltrials.gov/ct2/show/NCT00180271.