The "obesity paradox": does it persist among Israeli patients with decompensated heart failure? A subanalysis of the Heart Failure Survey in Israel (HFSIS).

Research paper by Shmuel S Schwartzenberg, Michal M Benderly, Stephen S Malnick, Jacob J George, Sorel S Goland

Indexed on: 27 Dec '11Published on: 27 Dec '11Published in: Journal of Cardiac Failure


Earlier studies among heart failure (HF) patients reported a paradox of reduced mortality rates in those with increased body mass index (BMI). Recently, however, it has been shown that obesity was not associated with better prognosis in certain groups. The aim of this study was to evaluate the "obesity paradox" among patients included in the Heart Failure Survey in Israel (HFSIS).Clinical, demographic, and laboratory characteristics of 2,323 patients hospitalized with a diagnosis of acute or decompensated chronic HF in 25 public Israeli hospitals between March 1 and April 30, 2003, were categorized by BMI as: normal weight (18.5-24.9 kg/m(2); n = 837), overweight (25.0-29.9 kg/m(2); n = 877), or obese (≥30.0 kg/m(2); n = 574), excluding 35 patients with BMI <18.5 kg/m(2). Survival over 15 months was inversely related to BMI category. Age-adjusted mortality hazard ratio (HR) was 0.95 (95% confidence interval [CI] 0.79-1.14) for overweight patients and 0.70 (95% CI 0.55-0.88) for obese patients compared with normal-weight patients. After further adjustment for gender, ejection fraction, New York Heart Association functional class, ischemic heart disease, diabetes, hypertension, dyslipidemia, renal function, and medications (angiotensin-converting enzyme inhibitor/angiotensin receptor blocker, β-blocker, spironolactone), obesity was associated with a nonsignificant HR of 0.79 (95% CI 0.59-1.05). Hypertension and dyslipidemia were also paradoxically associated with better survival in our model (HR 0.74, CI 0.59-0.92; and HR 0.77, CI 0.63-0.94; respectively; both P < .05).Our study falls in line with the obesity paradox observation (in obese but not overweight patients) in a large survey of HF patients, although this finding was not statistically significant on multivariate adjustment analysis.