Body Mass Index and Heart Failure Mortality: More Is Less?^∗Free Access

Introduction

Excess adiposity has been associated with structural and functional cardiovascular abnormalities and incident heart failure (HF) in several populations (1). Although HF subtypes including reduced ejection fraction (HFrEF) and preserved ejection fraction (HFpEF) may have similar mortality rates, there is significant heterogeneity in their pathophysiology. Obesity is most closely associated with an increased risk of HFpEF and many HFpEF risk factors including hypertension, left ventricular hypertrophy, inflammation, hyperglycemia, sleep-disordered breathing, and renal dysfunction among others. Racial/ethnic differences in the prevalence of HFpEF and HFrEF have also been noted. Regardless of the etiologic differences between HF subtypes, obesity increases the risk of developing HF and HF risk factors. However, several studies have demonstrated that increased body mass index (BMI) may be associated with reduced mortality in patients who develop HF (1). This paradoxical relationship has not been extensively studied in blacks and Hispanics who have different rates of obesity and HF compared with whites.

In this issue of JACC: Heart Failure, Powell-Wiley et al. (2) assessed the relationships between BMI and 30-day and 1-year mortality rates using a nationally representative cohort: the Get With The Guidelines–Heart Failure (GWTG-HF) registry. This cohort had a mixture of HFrEF (∼40%) and HFpEF (∼60%) patients and different racial/ethnic groups, including whites, blacks, Hispanics, and Asians, albeit there was a much higher enrollment of white patients. The enrolled population was also older, with a mean age of nearly 80 years. Notably, HFpEF was more common in women and was associated with higher mean BMI than HFrEF (∼29 kg/m² vs. 27 kg/m², respectively). In both HFpEF and HFrEF patients, blacks were younger and had higher BMI with the highest rates of class II (BMI, 35 to 39.9 kg/m²) and Class III (BMI ≥40 kg/m²) obesity compared with other groups.

The associations of BMI with in-hospital, 30-day, and 1-year all-cause mortalities were evaluated separately for HFpEF and HFrEF patients. The authors observed 37% lower 30-day all-cause mortality in HFpEF patients with higher BMI (BMI, 30 kg/m² vs. 18.5 kg/m²), and lower 1-year mortality with increasing BMI from 18.5 kg/m² up to 30 kg/m² (4% lower mortality per 1-U BMI increase up to 30 kg/m²). However, there was a slight increase in 30-day mortality in those with a BMI >30 kg/m², which remained relatively flat at all higher BMI levels. In HFrEF patients, 30-day all-cause mortality was 27% lower with higher BMI, and each 1-U increase of BMI up to 30 kg/m² was associated with lower mortality. However, for every unit increase in BMI >30 kg/m², there was a 1% increased rate of 1-year mortality in patients with HFrEF. Notably, in patients with HFpEF or HFrEF, increased BMI was associated with higher in-hospital mortality. No significant race/ethnicity interactions with BMI were noted in the relationships with 30-day mortality for HFpEF or HFrEF patients, 1-year mortality among HFpEF patients, or HFrEF patients with BMI >30 kg/m². The rate of 1-year mortality in patients with BMI <30 kg/m² was different among racial/ethnic groups with HFrEF.

The study by Powell-Wiley et al. (2) adds to the growing body of literature suggesting a “protective” effect of increased BMI against mortality after patients develop major chronic debilitating diseases, such as HF, cancer, or end-stage renal disease. This study has important strengths including use of a large cohort that has been linked to standardized outcomes data (Centers for Medicare and Medicaid). Furthermore, they examined HFpEF and HFrEF outcomes across multiple racial/ethnic groups. However, there are important limitations to consider when interpreting these data.

Although BMI is highly correlated with adiposity and easily obtained in large population studies, it has well-described shortcomings as a measure of obesity. Specifically, BMI does not directly account for body composition, including amount of fat or fat distribution, bone structure, muscle mass, or other changes associated with aging or sex differences (3). To illustrate this point, a running back in the National Football League who weighs 225 lbs at 6 ft tall would have a BMI of 30.5 kg/m². Based on this simple measure, he would be considered “obese,” although he can run a 40-yard dash in 4.5 s and likely has excellent cardiometabolic performance. In contrast, subjects with normal BMI and excess visceral adiposity or ectopic fat in organs, such as the liver, heart, and kidneys, often have cardiometabolic derangements and other adverse health consequences. Although dual-energy x-ray absorptiometry, magnetic resonance imaging, or computed tomography offer methods to quantitate total or regional adipose volumes or intraorgan steatosis, simple measures, such as waist circumference, waist-to-hip ratio, or sagittal abdominal diameter can be easily obtained in the clinical setting and are more reflective of central adiposity than BMI.

Many studies supporting the obesity paradox in HF have only used BMI as a measure of obesity. A recent article published by Tsujimoto and Kajio (4) in the Journal of the American College of Cardiology illustrates that this approach may lead to paradoxical results compared with more direct assessment of adiposity. Using data from the TOPCAT Study, they evaluated HFpEF patients with abdominal obesity (defined by waist circumference ≥102 cm in men and ≥88 cm in women) and observed a 52% increased risk of all-cause mortality over a mean follow-up of 3.4 years compared with those without abdominal obesity. In contrast, these authors observed an obesity “paradox” when they used BMI and compared mortality rates in normal weight to overweight and obese participants. This study suggests that body composition, rather than body weight or BMI, may be a more reliable way to assess the impact of obesity on outcomes in patients with HF.

Morbidly obese HF patients tend to be younger and present for health care earlier. Some hypothesize that because they have symptoms earlier, therapies with known survival benefits (at least in HFrEF) can be initiated earlier in the disease process. This may be true but, anecdotally, morbidly obese 80-year-old HF patients are rare. In the GWTG-HF Registry analysis by Powell-Wiley et al. (2), < 10% of the studied population had Class II or III obesity suggesting that severe obesity is rare in the elderly HF population.

One unknown variable in this relationship is the duration of obesity, and how BMI before the onset or early in the HF disease process affects outcomes. In the analysis by Powell-Wiley et al. (2), BMI was assessed at the time of admission for acute HF but the chronicity of HF was not specified. Reduced body weight associated with cachexia is a hallmark of most end-stage diseases so comparing patients with a BMI of 18.5 kg/m² (the reference group) may include patients who have already lost weight because of advanced disease. Atherosclerosis Risk in Communities (ARIC) Study investigators examined this issue in participants by evaluating the association of BMI measured ≥6 months before an incident HF event with mortality over a 10-year follow-up period (5). They found that being overweight or obese had a protective effect on survival compared with participants with normal BMI suggesting that patients with excess body weight may have higher metabolic reserve providing a survival advantage in HF.

The possible survival advantage of increased body weight in HF may also be age-dependent. As people age, weight (and BMI) may remain relatively stable while muscle mass decreases and body fat increases. In the data presented by Powell-Wiley et al. (2) using the GWTG-HF registry, the mean age was nearly 80 years. In a large worldwide analysis of acute HF, lower BMI (normal weight range) was associated with increased age, worsened HF severity, and increased risk of death (6). Perhaps excess body weight in older individuals helps counteract the catabolic effects of HF and provides a “metabolic cushion” to attenuate cachexia. A more plausible explanation is that age-related reductions in skeletal muscle mass and redistribution of body fat to visceral organs may contribute to elevated mortality risk in aged subjects. Sarcopenic obesity (reduced skeletal muscle mass and increased abdominal obesity) has been associated with increased mortality in elderly individuals. In the Health, Aging and Body Composition Study, participants who developed HF had higher total body mass, and subsequent incident HF in older adults was associated with a disproportionate loss of muscle mass (7).

Intentional weight loss in obese patients with bariatric surgery is associated with profound reductions in the risk of incident HF and reduces HF exacerbations (8,9). However, significant unintentional weight loss of ≥5% in patients with chronic HF is associated with increased 1-year mortality, particularly in obese HF patients (10). Instead of using a single measurement at a given point in time, changes in body weight and composition as well as duration of overweight or obesity (“obesity-years”) may be a more precise way to examine risks associated with chronic conditions, such as HF.

Although some studies have demonstrated a U-shaped relationship with mortality in HF with worsened outcomes for those with low or high BMI, the study by Powell-Wiley et al. (2) showed worse outcomes for those with lower BMI and a fairly flat (but slightly increased) mortality rate greater than a BMI of 30 kg/m². What is the take-home message from this analysis? Powell-Wiley et al. (2) studied a large, diverse, and well-validated HF registry and demonstrated that the obesity paradox may exist in both older HFpEF and HFrEF patients, and that race/ethnicity did not seem to modify the relationship between BMI and 30-day mortality. Furthermore, longer-term studies including more detailed metrics of body composition and cardiorespiratory fitness may help clarify this paradoxical phenomenon. Regardless, lifestyle modifications that include increased physical activity, a healthy diet, and weight control should be encouraged in overweight and obese individuals, particularly in those with central adiposity, to prevent cardiovascular disease.

Footnotes