Abstract 20596: A Prospective Examination of the Obesity Paradox in Heart Failure: Insights from a Population Study
Background: The association of body mass index (BMI) with heart failure (HF) is complex. While a risk factor for cardiovascular disease and HF, high BMI has been shown to be associated with lower mortality in established HF. Because this paradox has never been prospectively examined, we studied these associations in community-dwelling adults ≥65 years in the Cardiovascular Health Study.
Methods: Of the 5480 w/o prevalent HF (none had baseline BMI <18.5 kg/m2), 1139 developed centrally-adjudicated new HF over 12 years. Of these, 968 had annual BMI data (6 with BMI <18.5 preceding new HF were excluded). Using BMI 18.5-24.9 kg/m2 as reference, we examined the associations of BMI 25-29.9 (overweight) and ≥30 (obese) (1) w new HF in 5480 participants w/o HF, (2) w mortality in 5751 participants w baseline BMI, and (3) that of pre-HF BMI with mortality in 968 of those w new HF using Cox regression models. Quadratic trends were tested (using BMI and BMI2) and HRs for BMI (with death) were estimated for those w (n=968) and w/o (n=4341) new HF.
Results: Participants (n=5480) had mean age of 75 (±6) years, 50% women, and 13% African American. HR's (95% CI's) for the association of BMI w outcomes are displayed in Table. Quadratic associations of BMI w death in those w & w/o new HF are displayed in Figure. Mortality decreased with increasing BMI regardless of HF, reaching nadir at 33 and 28 for those w and w/o incident HF respectively (Figure).
Conclusions: While a high BMI is an independent risk factor for incident HF in older adults, once HF develops the survival is better in those with higher BMI, with a mortality nadir at a BMI of 33 kg/m2. This paradoxical association appears to entail more than a survivor cohort effect.
- © 2010 by American Heart Association, Inc.