|| Checking for direct PDF access through Ovid
Introduction: Patients with diabetes mellitus (DM) and heart failure with reduced ejection fraction (HFrEF) are at increased risk of sudden death. However, given the burden of competing causes of death in patients with DM, therapies to reduce the risk of sudden death may have reduced efficacy. We aimed to assess the effectiveness of implantable cardioverter-defibrillators (ICDs) in a large US registry.Methods: We linked data from the GWTG-HF registry to the Centers for Medicare/Medicaid Services claims (January 2010 to December 2014). Using Cox proportional hazard models with propensity score matching for baseline covariates, among patients with and without DM, we compared patients who received an ICD (or were prescribed an ICD at discharge) to those without an ICD. The primary outcome was all-cause mortality.Results: Among the 17,186 patients in our study cohort, 6683 (38%) had DM. For patients with DM, those with an ICD (n=663), compared to those without and ICD, were younger (74.0 vs. 78.0 years of age), more likely to be male (66.3% vs. 55.7%) and more likely to have coronary disease (67.8% vs. 60.4%) Similar trends were seen among patients without DM (n=1,031). Among patients with DM, in matched cohorts, an ICD was associated with a reduced risk of mortality (54.4% vs. 60.0%; adjusted HR [aHR] 0.74, 95% CI 0.65-0.83; figure panel A). Similar results for mortality were seen in patients without DM (47.4% vs. 57.0%; aHR 0.68, 95% CI 0.61-0.75; figure panel B). Diabetes did not modify the relationship between ICD and mortality (interaction p-value = 0.24).Conclusion: Among patients with HFrEF, primary prevention ICD was associated with reduced mortality in patients with and without diabetes. These findings support current guideline recommendation for primary prevention ICDs among eligible patients with DM.