OBJECTIVES: Heart failure (HF) patients with diabetes mellitus experience poor prognosis. We assessed the independent predictive effect of prevalent diabetes mellitus on all-cause mortality in HF outpatients. Furthermore, we investigated if optimized HF medication differed in diabetic versus nondiabetic patients. METHODS: From 6,289 patients included in the Norwegian HF registry during 2000-2012, 724 diabetic HF outpatients were propensity-score-matched with nondiabetic HF outpatients (1:1), based on 21 measured baseline variables. Baseline characteristics, measured comorbidities and medication were balanced in the matched sample. RESULTS: Diabetes was not an independent predictor of all-cause mortality in the propensity-matched analyses (hazard ratio 1.041; 95% confidence interval 0.875-1.240). No interactions were found between the prognostic impact of diabetes and the strata renal function, systolic function or etiology of chronic HF. Diabetic HF outpatients were independently prescribed higher doses of β-blockers and loop diuretics (both p < 0.001) and were more prone to receive statins (p = 0.003) than nondiabetics. CONCLUSIONS: Prevalent diabetes mellitus was not an independent predictor of all-cause mortality in HF outpatients. Explanations other than tight glycemic control should be assessed to improve the prognosis of diabetic HF outpatients. The more intensive, optimized HF medication for diabetic HF outpatients may, to a certain degree, explain our results.
OBJECTIVES:Heart failure (HF) patients with diabetes mellitus experience poor prognosis. We assessed the independent predictive effect of prevalent diabetes mellitus on all-cause mortality in HF outpatients. Furthermore, we investigated if optimized HF medication differed in diabetic versus nondiabeticpatients. METHODS: From 6,289 patients included in the Norwegian HF registry during 2000-2012, 724 diabetic HF outpatients were propensity-score-matched with nondiabetic HF outpatients (1:1), based on 21 measured baseline variables. Baseline characteristics, measured comorbidities and medication were balanced in the matched sample. RESULTS:Diabetes was not an independent predictor of all-cause mortality in the propensity-matched analyses (hazard ratio 1.041; 95% confidence interval 0.875-1.240). No interactions were found between the prognostic impact of diabetes and the strata renal function, systolic function or etiology of chronic HF. Diabetic HF outpatients were independently prescribed higher doses of β-blockers and loop diuretics (both p < 0.001) and were more prone to receive statins (p = 0.003) than nondiabetics. CONCLUSIONS: Prevalent diabetes mellitus was not an independent predictor of all-cause mortality in HF outpatients. Explanations other than tight glycemic control should be assessed to improve the prognosis of diabetic HF outpatients. The more intensive, optimized HF medication for diabetic HF outpatients may, to a certain degree, explain our results.