Abstract 4131: ICD and Shock Experience In Dialysis Patients
BACKGROUND: The occurrence of ventricular arrhythmias in end stage renal disease (ESRD) patients on hemodialysis (HD) is common. There are limited data on ICD patients with ESRD on HD.
METHODS: We analyzed a prospectively collected database of patients with ICD implants at our institute. The patients with and without ESRD were compared for the baseline demographic variables, co-morbidities and outcomes, especially the rates of VT storm, total ICD therapies and all-cause mortality.
RESULTS: There are 1438 patients with ICD implantation and follow up at our institution from April 1995 to August 2007. Of these 40 (3.3%) were on hemodialysis secondary to ESRD. The mean follow up was shorter in ESRD patients (536 ± 455 vs. 886 ± 881, p − 0.05). The ESRD group was comparable to the general ICD population for age, gender distribution, ejection fraction (mean EF 28.8 ± 15 vs. 28.9 ± 14), presence of HTN, ischemic heart disease, use of Amiodarone, beta-blockers, Digoxin and incidence of implants for primary prevention (85% vs. 74%, p - NS). The ESRD group had higher incidence of Diabetes (45% vs. 26%, p - 0.01) and lower use of Ace-inhibitors (32.5% vs. 50%, p - 0.03), biventricular pacemakers (10% vs. 24.8%). The ESRD patients had significantly worse overall mortality (52.5% vs. 23.8%, p - 0.01), one year mortality (17.5% vs. 7.4%, p - 0.029)), appropriate ICD therapies (30% vs. 17%, p - 0.03) and VT storms (17.5% vs. 5.2%, p < 0.001). On Multivariate analysis, presence of ESRD was the strongest predictor of overall mortality (see table⇓) but not an independent predictor of one year mortality.
Conclusion: In our single center experience, the dialysis patients had significantly worse outcomes than the general ICD patients, in terms of higher mortality, ICD therapies and cluster shocks. Presence of ESRD was an independent predictor of overall survival but does not predict one year mortality.