The strong association between abdominal obesity and the risk of type 2 diabetes (T2D) and coronary heart disease (CHD) has been validated by several large-scale epidemiological prospective studies over the past 40 years. Many of these studies have also shown that abdominal obesity predicts the risk of T2D and CHD regardless of body weight and have supported the notion that the assessment of waist circumference could be a valuable addition to the use of body mass index (BMI) in order to identify individuals at high risk of T2D or CHD. These studies led several investigators to propose that waist circumference should be considered as a therapeutic target to prevent these chronic diseases. However, one major caveat of such studies is that they are limited by their observational design. Consequently, rather than the fact that they have an elevated waistline per se, confounding factors, such as diet, smoking, physical activity, sleep patterns, and many others may explain why individuals with an elevated waistline are at high risk. In other words, the causality link between abdominal obesity and these cardiometabolic diseases has not been firmly assessed.
To overcome these limitations and to investigate whether abdominal obesity should be once and for all considered a causal risk factor for T2D and CHD, investigators from the Harvard Medical School used a very innovative strategy taking advantage of recent developments in the field of genetic epidemiology. Most anthropometric and biological traits such as waist circumference, glucose levels, LDL cholesterol, and blood pressure could be explained, at least to a certain extent, by dozens of genetic variations that individually contribute to these traits. Therefore, some individuals may carry a set of genetic variants that predispose them to higher waistlines while others may carry a different set of genetic variants that predispose them to smaller waistlines. One of the most important advantages of using genetic variants for causality inference is that these variants are not influenced by environmental factors as are the confounding variables listed above which may explain to a certain extent the association between abdominal obesity and the risk of chronic cardiometabolic diseases. On the other hand, because these variants usually have a small effect, this hypothesis needs to be tested in very large cohorts. This is precisely what the authors did in this paper. They built a genetic risk score that included variants associated with abdominal obesity (defined by an elevated waist-to-hip ratio [WHR]) adjusted for BMI using data from the GIANT Consortium. This genetic risk score was associated with a higher waist circumference, while being simultaneously associated with a lower hip circumference and a lower BMI.
Interestingly, by using data from another dataset called the Global Lipids Genetic Consortium, the authors found that the genetic risk score was also associated with elevated lipid levels such as LDL cholesterol and triglyceride levels. Moreover, the genetic risk score was associated with markers of insulin resistance and with poor glycemic control in the MAGIC dataset and with a high blood pressure in participants of the UK biobank. All these genetic datasets include hundreds of thousands of participants making these associations very robust. Most importantly, the authors have tested the association between an increase in 1-standard deviation (SD) unit in their genetic risk score with the risk of T2D in two large cohorts, the DIAGRAM Consortium and the UK Biobank (which included 40,530 T2D cases and 221,277 controls) and the risk of CHD in the CARDIOGRAMplusC4D dataset and the UK Biobank (which included 66,440 CHD cases and 229,851 controls). Their results, which were published in the Journal of the American Medical Association, suggest that each 1-SD increment of the genetic risk score was associated with a 77% increase in the risk of T2D and a 46% increase in the risk of CHD.
What makes these results fascinating and very relevant to our comprehension of the pathophysiology of cardiometabolic diseases is that all the reported associations between the genetic risk score linked with the WHR, cardiometabolic risk factors, T2D and CHD were observed despite the fact that the genetic risk score associated with the WHR was simultaneously linked with a lower BMI. This study further reinforces the notion that an elevated BMI may not be so harmful in the absence of an elevated waistline and that, on the contrary, individuals with a normal body weight may be at increased risk of developing cardiometabolic diseases if they are characterized by abdominal obesity.
Although this study was not designed to identify specific thresholds at which an elevated WHR may be linked with the risk of T2D and CHD and that these results will eventually need to be confirmed in populations of non-European ancestry, this study provides an important framework for the design of intervention aimed at reducing the risk of cardiometabolic diseases. In this context, although managing cardiometabolic risk factors such as LDL cholesterol, hypertension, and glucose with drugs may reduce the risk of CHD in high-risk individuals, it is becoming increasingly clear that targeting these risk factors in isolation without tackling the most important root causes (abdominal obesity and ectopic fat deposition) with physical activity and a healthy diet may not lead to substantial and meaningful gains in the prevention of chronic cardiometabolic diseases at the population level.