Reverse epidemiology in different stages of heart failure

In heart failure (HF), traditional cardiovascular risk factors (RF) as body mass index (BMI), total cholesterol (TC) and systolic blood pressure (SBP) are associated with better survival. It is unknown at which time point along the disease continuum the adverse impact of these RF ceases and may 'start to reverse'. We analyzed the distribution of RF and their association with survival across HF stages.

We pooled data from four cohort studies from the German Competence Network HF. Employing ACC/AHA-criteria, patients were allocated to stage A (n=218), B (n=1324), C1 (i.e., New York Heart Association [NYHA] classes I & II; n=1134), and C2+D (NYHA III & IV; n=639).

With increasing HF severity median age increased (63/67/67/70 years), whereas the proportion of females (56/52/37/35%), median BMI (26.1/28.8/27.7/26.6 kg/m(2)), TC (212/204/191/172 mg/dl), and SBP (140/148/130/120 mmHg) decreased (P<0.001 for trend for all). In the total cohort, higher levels of all RF were associated with better survival, even after extensive adjustment for multiple confounders. If analyses were stratified, however, a higher RF burden predicted better survival only in clinically symptomatic patients: hazard ratio (HR) per +2 kg/m(2) BMI 0.91 (95% confidence interval 0.88; 0.95); per +10 mg/dl TC 0.93 (0.92; 0.95); per +5 mmHg SBP 0.94 (0.92; 0.95).

In this well-characterized sample of patients representing the entire HF continuum, reverse associations were only consistently observed in symptomatic HF stages. Our data indicate that the phenomenon of a "reverse epidemiology" in HF is subject to significant selection bias in less advanced disease.