Obesity Paradox in Heart Failure Revisited: Etiology as Effect Modifier.

The prognostic value of overweight/obesity in heart failure (HF) may vary according to HF etiologies. We aim to determine whether body mass index has differential impacts on survival among hospitalized HF patients with varying etiologies. Consecutive hospitalized HF patients between December 2006 and December 2017 were included. Multivariable analyses, including Cox proportional hazard models and restricted cubic splines, were used to investigate the impact of body mass index on mortality by HF etiology. Among the 3,836 patients included (mean age 57.1 years, 28.4% women), 1,475 (38.5%) were identified as having ischemic etiology. Of the remaining 2,361 patients with non-ischemic etiologies, dilated cardiomyopathy (DCM) accounted for 45.6% (n = 1,077). The rest of the patients were uniformly classified as having non-ischemic-non-DCM HF. The unadjusted data demonstrated an adiposity-related survival paradox in HF across all etiologies. However, the paradox holds only among non-ischemic-non-DCM HF patients after multivariate adjustment, wherein overweight patients exhibit the lowest mortality compared with their normal-weight counterparts (adjusted hazard ratio [aHR] 0.69, 95% confidence interval [CI] 0.52 to 0.91), with a nadir in mortality risk at 28.18 kg/m2. Similar survival benefits of overweight were not demonstrated in ischemic or DCM HF patients (ischemic etiology: aHR 1.07, 95% CI 0.84 to 1.36; DCM etiology: aHR 0.97, 95% CI 0.74 to 1.28). In conclusion, being overweight or obese does not confer better survival in HF patients of ischemic or DCM etiology, and the prognostic benefit of being overweight is maintained only in non-ischemic-non-DCM HF patients. Pathophysiologic interpretations are warranted, and whether patients of certain etiologies would benefit from weight reduction needs to be explored.