The cholesterol paradox may be attenuated in heart failure patients with diabetes

BACKGROUND: In heart failure (HF) patients, a lower total cholesterol (TC) appears to portend an ominous prognosis. We studied if the prognostic impact of TC was different according to diabetes mellitus (DM) status in a chronic HF population. METHODS: Patients with systolic HF under optimized and stable evidence-based therapy were prospectively recruited from our HF clinic. We excluded patients on renal replacement therapy and those hospitalized in the previous 2 months. A venous blood sample was collected. Patients were followed for up to 5 years and all-cause mortality was the endpoint under analysis. The prognostic impact of TC was analyzed using a Cox-regression analysis. Analysis was stratified according to coexistence of DM. RESULTS: We studied 262 chronic HF patients, 182 males, mean age 69 years, 98 (37.4%) diabetic and 62.2% with severe left ventricular systolic dysfunction. Median B-type natriuretic peptide: 237.8 pg/mL; median TC: 169 mg/dL. During follow-up 121 (46.2%) patients died. Patients with TC>200 mg/dL had better survival than those with lower TC; however, this protective effect was mostly observed in non-diabetic HF patients. In non-diabetics the multivariate adjusted 5-year mortality hazard ratio (HR) was 0.36 (95% CI: 0.16-0.79) for those with TC>200 mg/dL. In diabetic HF patients, there was a non-significant survival benefit of TC>200 mg/dL; HR 0.51 (95% CI: 0.20-1.30). CONCLUSIONS: Non-diabetic chronic HF patients with TC>200 mg/dL have a 64% lower risk of 5-year death. In diabetics, there is a non-significant 49% protective effect of elevated TC. The cholesterol paradox may be attenuated in diabetic HF patients.