Analyzing the Weight of Evidence on the Obesity Paradox and Heart Failure—Is There a Limit to the Madness?

Abstract

Since overweightness and obesity have adverse effects on cardiac structure and systolic and especially diastolic left ventricular (LV) function, as well as on cardiovascular (CV) diseases, including coronary heart disease (CHD) and hypertension (HTN), which are the leading causes of heart failure (HF), it really is no surprise that the prevalence of HF markedly increases with increasing severity of obesity. In a study of 5881 Framingham Heart Study participants, Kenchaiah and colleagues demonstrated that for every 1-kg/m increase in body mass index (BMI), the risk of HF increases by 5% in men and 7% in women during a 14-year follow-up, with graded increases in the risk of HF across the entire spectrum of BMI. In addition, in a study by Alpert and colleagues of 74 morbidly obese patients, nearly one third had clinical evidence of HF, and the prevalence of HF increased dramatically with the increasing duration of morbid obesity. Despite this evidence, numerous large studies and meta-analyses have demonstrated an obesity paradox in HF, ie, overweight and obese patients with HF have a better prognosis than do their leaner counterparts, with overall similar HF severity at baseline. More than a decade ago, we demonstrated this paradox with both BMI and percentage of body fat (BF); thus, in our study, for every 1% increase in BF there was a 13% reduction in major CV events. The mechanisms of the obesity paradox in HF remain unclear, and we have recently discussed the potential reasons for this puzzling paradox. Certainly, very few studies of the obesity paradox in HF have addressed the potential role of nonpurposeful weight loss prior to study entry, and in almost every condition, including HF, nonpurposeful weight loss, especially with cachexia, is associated with a poor prognosis. Many overweight and obese patients may have developed HF secondary to HTN and CHD and as a result of the abnormalities in left ventricular diastolic and systolic function that often develops in these patients. Conversely, lean patients who develop HF do so from other mechanisms that may result in a more negative prognosis. Obese patients present with HF at younger ages, have a higher LV ejection fraction (EF) and higher blood pressure, thus allowing obese patients to tolerate more cardioprotective medications and at higher doses and to have a lower expression of brain natriuretic peptides. All these factors and others may lead to an earlier presentation with less severe disease, which could lead to a better prognosis in obese patients with HF compared with lean patients. In the present and final issue of Congestive Heart Failure, Nagarajan and colleagues from the Cleveland Clinic HF program confirmed the obesity paradox in 501 patients in an advanced HF clinic, but their data suggest that there may be a “limit to this madness” by demonstrating a poor prognosis and, therefore, no obesity paradox in very obese patients (BMI ≥40 kg/m) with HF. Although their data are statistically significant, it must be recognized that this relatively “new” finding is based on a very small number of only 21 morbidly obese HF patients, as the authors acknowledge. Nevertheless, considering the progressive decline that has occurred during the past 5 decades in total energy expenditure, the marked increases in obesity prevalence during this period, and the fact that “morbid” obesity is increasing even more so than is obesity per se, these results reported herein are still noteworthy and are also in line with other recent studies demonstrating a worse prognosis among the very obese. Although substantial evidence suggests that the obesity paradox in CV diseases, especially CHD, represents as much as a “lean” paradox (the leaner patients by BMI, BF, and lean body mass doing particularly poorly) or an “overweight” paradox (the overweight patients, or those with BMI 25 to 30 kg/m, doing particularly well), the results from Nagarajan and colleagues and others support that the paradox may not apply to the most obese, where prevention of morbid obesity and weight loss strategies (nonpharmacologic, pharmacologic, and surgical) are desperately needed. It is important to point out that the major clinical guidelines from various societies have differed regarding levels where weight loss is recommended (currently none recommend weight loss for patients with BMI <30 kg/m, with various cutpoints between BMI 30–40 kg/m), but all of the guidelines advocate weight loss for BMI ≥40 kg/m. Greater efforts are needed to better define the exact role of weight Address for correspondence: Carl J. Lavie, MD, Medical Director, Cardiac Rehabilitation and Prevention, Director, Exercise Laboratories, John Ochsner Heart and Vascular Institute, Ochsner Clinical School, The University of Queensland School of Medicine, 1514 Jefferson Highway, New Orleans, LA 70121-2483 E-mail: clavie@ochsner.org