Statistics from Altmetric.com
There has been a lot of controversy concerning the prognostic role of obesity on cardiovascular morbidity and mortality. Several studies with cardiovascular patients (heart failure, coronary, cerebrovascular and peripheral artery diseases, hypertension and atrial fibrillation) have reported better or similar risk of cardiovascular events and mortality among overweight and obese individuals as compared to lean ones.1 Since obesity is an independent risk factor for cardiovascular diseases, these findings initially surprised researchers, who coined the term ‘obesity paradox’. The increasingly debated ‘metabolically healthy’ or ‘metabolically benign’ obese phenotype might be an explanation to the ‘obesity paradox’ in cardiovascular diseases. In their article, van der Leeuw and colleagues add more fuel to the discussion by investigating the influence of adiposity and cardiometabolic dysfunction on recurrent major cardiovascular events (stroke, myocardial infarction, retinal infarction and vascular mortality) and mortality in a sample of the Secondary Manifestations of ARTerial disease (SMART) study population without diabetes.2
In a population of 5232 patients with previously diagnosed cardiovascular disease (58.5% coronary artery disease), the authors created six groups by the combination of adiposity, as defined by Body Mass Index (BMI) strata (normal weight, BMI 20–24.9 kg/m2; overweight, BMI 25–29.9 kg/m2; obesity, BMI≥30 kg/m2), and cardiometabolic dysfunction, which was based on the presence of at least three criteria of a modified National Cholesterol Education Program (NCEP) metabolic syndrome definition (waist circumference replaced by C-reactive protein). Over a mean follow-up time of 6 years, they found that, as compared to the normal weight without cardiometabolic dysfunction group, overweight and obesity without cardiometabolic dysfunction were not associated with a higher risk of recurrent major cardiovascular events and mortality, whereas, all groups with cardiometabolic dysfunction showed similarly increased risks of both outcomes. As expected, cardiometabolic dysfunction was highest within the obese category (75%), but the prevalence was high (40%) in the normal weight group was well.
These findings support the idea that cardiometabolic dysfunction is a better predictor of recurrent cardiovascular events than BMI per se. Additionally, the study suggests that, in regard to cardiometabolic dysfunction, patients within the same BMI category can be highly heterogeneous. Having in mind the limitations of BMI as a marker of adiposity and the complexity of the adipose tissue and its interplay with the cardiovascular system helps us to critically evaluate the results.
Despite being a practical measure of total body mass, the performance of BMI as a marker of body composition varies widely among different populations.3 At a given BMI, there is substantial variability not only in the amount of fat and lean muscle mass, but also in the proportion of lean/fat mass. A high proportion of lean/fat mass is associated with high cardiovascular fitness and good nutritional and functional status, which might reduce the potential detrimental effects of high adiposity on the cardiovascular health of individuals with a high BMI.1 BMI does not reflect fat distribution or the proportion of fat between the subcutaneous (SAT) and visceral adipose tissue (VAT) compartments either. A more central distribution of fat (low SAT:VAT ratio) leads to fat accumulation in body organs and muscles and presents a proinflammatory profile.4 In this regard, van der Leeuw and colleagues report that higher amounts of VAT, which was only moderately correlated to BMI in the SMART study, were associated with an increased risk of major cardiovascular events and mortality regardless of the presence of cardiometabolic dysfunction.
Reports of better cardiovascular outcomes in individuals with high body fat mass than in those with low body fat support the idea that the ‘obesity paradox’ is not only related to the limitations of BMI as a marker of adiposity.4 Moreover, it implies that differences in the amount and distribution of fat do not fully explain the diverse clinical phenotypes and outcomes within obesity, suggesting that functionality should be considered as well. The complex biology of the adipose tissue and its interplay with the cardiovascular system has been described for several years by now, but are not fully elucidated.
The adipose tissue responds to genetic and environmental factors (excessive caloric intake, physical inactivity, medications, pollutants, toxins, gut flora) in diverse ways according to individual susceptibility. The type of response determines the characteristics of the cross-talk between the adipose tissue and the cardiovascular system (figure 1). The interaction between genetic and environmental factors might lead to a wide range of histological and functional alterations in the adipose tissue. Ectopic fat infiltration in cardiomyocytes, pericardium and perivascular tissue, insulin resistance and a low-grade inflammatory state ensue and might, then, affect the cross-talk between the adipose tissue and the cardiovascular system. It is important to notice that the fat (dys)function is dynamic, as suggested by a recent meta-analysis, which found an increased risk of death in the ‘metabolically healthy’ obese group only after exclusion of studies with less than 10 years of follow-up.5
In fact, there is no consensus regarding the definition of metabolic health. The downside of applying the metabolic syndrome criteria to define cardiometabolic health is that, due to its categorical nature, it prevents the recognition of the cardiometabolic state as a continuum. In an attempt to overcome this limitation (though, partially), van der Leeuw et al performed additional analyses considering the effect of the number of risk factors across the BMI strata. The results showed that normal weight and overweight/obesity were associated with an increased risk of major cardiovascular events only in the presence of at least four risk factors. Similar findings were observed for overall mortality, but only in the group with normal weight.
The influence of selection issues and residual confounding on the results needs to be recognised as well. By selecting patients with a previous cardiovascular event, the study may have inadvertently selected subjects with unknown risk factors, particularly among individuals with normal weight. This might have influenced the distribution of risk factors in patients included in the study and reduced the magnitude of risk differences between the ‘normal weight’ control group and the overweight/obese subgroups. Additionally, neither temporal variations of weight and cardiometabolic dysfunction nor the influence of fitness were appreciated in this study. Fitness, in particular, can alter the relationship between obesity and mortality, since high-fit obese individuals have risks that are similar to the ones with normal weight.1 All this points out to the need of standardising the definition of metabolic health among studies, as well as of performing studies with longer duration and repeated and alternate measures of both adiposity and cardiometabolic dysfunction.
Then, how does this study impact on the clinical care of patients with cardiovascular disease? First, it corroborates recommendations of risk stratification based on the cardiometabolic state. It is still debateable, however, whether measures of insulin resistance and inflammation add prognostic information for recurrent events. Second, we can infer from the results that anthropometric measures more closely correlated with visceral fat than BMI, such as waist circumference, might be considered for prognostic purposes.6 Third, as subjects with cardiometabolic dysfunction had increased risk of event regardless of BMI strata, the evaluation of the cardiometabolic status of patients with normal weight should be given similar attention as that of overweight/obese patients.
Unfortunately, the present study does not shed light on the scarce evidence on the long-term effects of weight control among overweight/obese patients with cardiovascular disease. Except for bariatric surgery, which was associated with a decreased risk of fatal and non-fatal cardiovascular events,7 interventions focused on weight loss failed to demonstrate a reduction of cardiovascular events risk in overweight/obese patients in the long term.8 Since there is no definite evidence of the long-term benefits of weight loss in patients with cardiovascular disease, reports of the ‘obesity paradox’ and the ‘metabolically healthy’ obese phenotype continue to concern clinicians. First, because they might convey the message that weight control is harmful and second, because they may eventually stimulate patients to maintain unhealthy lifestyle habits. On the other hand, labelling every person with cardiovascular disease and a high BMI as ‘unhealthy’ might lead to overtreatment. This may have consequences at the individual level (loss of muscle and bone mass, weight cycling, appetite dysfunctions, anxiety, frustration) and public health level (diverting resources from prevention of childhood/adolescence obesity). Thus, a person-centred approach is indisputable in the care of patients with cardiovascular disease. More importantly, a more comprehensive approach to health should be undertaken. Evaluating the cardiometabolic profile, and also the patient functional status and the presence of other conditions (osteoarthritis, sleep apnoea, gastro-oesophageal reflux disease, psychological issues) helps physicians understand the patient's personal experience of illness, and balance between the benefits and risks of weight loss.
Contributors Both authors contributed to this editorial.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.