Diabetes may modify the efficacy of primary prevention implantable cardioverter-defibrillators in non-ischemic systolic heart failure: an extended follow-up of the DANISH trial

Abstract

Abstract Background Since patients with both diabetes and heart failure (HF) have a substantially worse prognosis than those with each of these alone, the efficacy of implantable cardioverter defibrillator (ICD) implantation in patients with systolic HF may be modified by diabetes status. Purpose In this extended follow-up analysis of the Danish Study to Assess the Efficacy of primary prevention of Implantable Cardioverter Defibrillators in Patients with Non-ischemic Systolic Heart failure on Mortality (DANISH), adding four years of additional follow-up, we examined the long-term effects of primary prophylactic ICD implantation, compared with usual clinical care, in patients with non-ischemic systolic HF according to diabetes status. Methods In DANISH, 1,116 patients with non-ischemic systolic HF were randomized to receive an ICD (N=556) or usual clinical care (N=550). Outcomes were analyzed according to diabetes status at baseline. The primary outcome was death from any cause. Secondary outcomes were cardiovascular death and sudden cardiovascular death. Results Of the 1,116 patients randomized in DANISH, 211 (18.9%) had diabetes at baseline (median age 64 year-old, 159 males, median left ventricular ejection fraction 25%). There was a statistically significant interaction between diabetes and the effect of ICD implantation on death from any cause and cardiovascular death. There was a trend toward a reduction in both outcomes with ICD implantation, compared with usual clinical care; in patients without diabetes (death from any cause, HR 0.81 [95% CI, 0.65–1.01]; cardiovascular death, HR 0.77 [0.59–1.01]), but not in those with diabetes (death from any cause, HR 1.19 [0.81–1.74]; cardiovascular death, HR 1.20 [0.79–1.82]) (P for interaction = 0.04 and 0.04 for death from any cause and cardiovascular death, respectively). Although there was no statistically significant interaction between diabetes and the effect of ICD implantation on sudden cardiovascular death, ICD implantation significantly reduced the rate of sudden cardiovascular death in patients without diabetes (HR 0.51 [0.31–0.85]), but not in those with diabetes (HR 0.91 [0.41–1.99]) (P for interaction = 0.15). Conclusions In patients with non-ischemic systolic HF, there was a significant interaction between diabetes and the effect of ICD implantation on death from any cause and cardiovascular death, suggesting that individuals without diabetes may derive greater benefit from ICD implantation. Funding Acknowledgement Type of funding sources: None.