Fat and cardiometabolic risk: Location, location, location

Obesity is a leading cause of hypertension and diabetes which are major risk factors for cardiovascular diseases worldwide. Much of this increased risk from obesity is related to visceral rather than subcutaneous adiposity. Body mass index (BMI) is the most commonly used measure to assess adiposity and cardiometabolic disease risk; however, it has limitations when comparing across populations. Other measures such as waist circumference (WC) and waist-to-hip ratio (WHtR) may be better predictors of risk as they are better discriminators of body composition and location of fat depots which vary across racial/ethnic populations. In this manuscript, Wang et al1 examined the associations between adiposity measurements (BMI, WC, and WHtR) and the risk for prediabetes and diabetes in 925 Asian hypertensive patients. As expected, BMI was a significant independent predictor of both prediabetes and diabetes; however, this association was attenuated and no longer significant after adjusting for WC or WHtR. Both WC and WHtR were significantly associated with prediabetes and diabetes even after adjustment for BMI. Compared to BMI, WC and WHtR were better discriminators for diabetes risk, and adding them to BMI reclassified nearly 43% and 39%, respectively, of patients to the correct risk group. Of note, WHtR was not a better discriminator compared to WC. Overall, these findings suggest that WC and WHtR are better predictors of prediabetes and diabetes compared to BMI in Asian hypertensive patients. Further, WC and WHtR performed similarly in net reclassification of prediabetes or diabetes suggesting that WC, an easier measure to perform in clinic, may be the better pre-screening tool for this population. Importantly, this study examines this issue in Asian individuals who have different risks for cardiometabolic diseases compared with other racial/ethnic groups. This study has practical implications for this patient population but does have a few limitations. Overall, the relatively small number of study participants and limited number of ethnic groups (most were Chinese) are the most notable limitations. Also, the discrimination performance (area under the curve values for all measures were <0.7) was modest. Nonetheless, WC, a more practical measure to incorporate into widespread clinical use, was as good as WHtR and better than BMI for predicting prediabetes or diabetes. Also, of all of the adiposity measures, WC had the strongest correlation with both systolic and diastolic blood pressures. The study adds to the expanding knowledge that the widely clinically utilized measure BMI is not optimal and has its limitations. While BMI is correlated with adiposity, a problematic issue is that it does not distinguish between fat and lean muscle mass. Further, there is clear evidence that visceral adiposity has more adverse effects and risk for cardiometabolic diseases compared with subcutaneous fat and BMI does not capture the relative differences between fat depots.2 An extreme example of where BMI is misleading as a measure of adiposity is a wide receiver recently drafted in the National Football League who is 6 feet 3 inches in height and weighs 228 pounds. His calculated BMI is 28.5 kg/m2, which places him in the "overweight" category. His reported body fat percentage is <2%, and he runs the 40-yard dash in <4.4 seconds.3 Based upon an overweight BMI alone, he may be misclassified as being at increased risk for cardiometabolic diseases. Asian populations have a higher prevalence of diabetes compared with other ethnic groups.4 Diabetes risk increases at a lower mean BMI, approximately 23 kg/m2, in Asian populations compared with other ethnic groups.4, 5 Several multiethnic studies have demonstrated that Asians have greater visceral fat for a given BMI compared to whites.4, 6 It is this visceral adipose tissue, including intra-abdominal fat as well as hepatic and skeletal muscle lipids which have been implicated in cardiometabolic disease risk including diabetes and hypertension.2 These specific fat depots as well as fat deposited in important organs such as the heart and kidneys have been associated with target organ dysfunction and injury. Although more advanced imaging techniques such as computed tomography, magnetic resonance imaging, or dual-energy x-ray absorptiometry can quantitate total or regional adipose tissue, these are more costly and not always practical for large populations, much less in a routine clinic visit. Therefore, simple measures such as WC, WHtR, or sagittal abdominal diameter offer greater utility in the clinical setting to assess central visceral adiposity and subsequent cardiometabolic risk.7 Limitations of BMI as a predictor of adverse outcomes have been demonstrated across the spectrum of cardiometabolic diseases. Incorporation of BMI as the sole measure of adiposity has led several studies to conclude that increasing body weight (BMI) may be protective in cardiovascular diseases such as heart failure in addition to its association with reduced mortality in chronic disease.8 These relationships have been referred to as the "obesity paradox." Most of these studies suggesting a beneficial role of higher BMI have included those with chronic disease processes where weight loss and cachexia are often also characterized by loss of lean muscle and bone. It is important to note that the paradoxical beneficial associations between higher BMI and these cardiovascular outcomes occur in established disease. In a few of these studies demonstrating the obesity paradox with BMI, higher WC was associated with significantly higher mortality.9 The take-home message is that body composition and adipose tissue location, rather than body weight or BMI, are better predictors of cardiometabolic risk and long-term outcomes. If we know that WC and WHtR are better predictors of cardiometabolic risk and overall outcomes in cardiovascular disease and that BMI has inherent limitations, then why do we not see these measures incorporated into clinical practice? One reason is that BMI is an easy measure that can be obtained in virtually any clinical setting with good reproducibility. Although it has its limitations, BMI is usually correlated well with risk for cardiometabolic diseases such as hypertension and diabetes.10 WC measurement may be incremental for overall adiposity, but importantly, it is a better measure of fat distribution compared to BMI. WC measurement has been recommended by several organizations to assess cardiometabolic risk; however, there has been some variability in WC measurement protocols.11 Some have suggested WC measurement directly above the superior border of the iliac crest, while others have recommended measurement at the umbilicus, at the minimal waist, or the midpoint between the last rib and iliac crest. The choice of measurement site does influence the reliability of WC measurements, but overall WC measurement is highly reproducible regardless of the site chosen.11 In this particular study by Wang et al, the World Health Organization protocol was utilized, measuring at the approximate midpoint between the lower margin of the last palpable rib and the top of the iliac crest. Although jackets or "tight clothing" were removed, there is likely a chance for measurement error depending on what clothing is removed. Other factors which may influence WC measurement include respiratory phase, posture, and timing of last meal.12 While subtle differences may not be important clinically when comparing across a consistent measurement site, these differences are more relevant when deciding on thresholds for risk. As such, racial/ethnic thresholds have been suggested for WC similarly to BMI. There are also likely sex differences whereby certain WC measurement sites are better indicators of visceral adiposity. Irrespective of these limitations, WC can provide important information related to visceral adiposity and should be considered in the risk assessment for cardiometabolic diseases. In cases of severe obesity, additional measures including WC and WHtR may not add additional predictive ability. These measurements are seemingly more helpful in establishing cardiometabolic risk in those in the overweight or class I obesity categories. As with any measurement that physicians collect in the clinical setting, these data should be considered in the context of overall patient health. Recognition of cardiometabolic risk is only part of this responsibility. Unfortunately, many physicians fail to provide weight loss counseling in obese patients. With the growing epidemic of overweight and obesity worldwide, this will be more and more important from a global health perspective. Weight management including diet and physical activity is a clear choice that should be recommended regardless of body weight or cardiometabolic risk. For patients with higher risk or more cardiometabolic disease burden, other therapies such as weight loss medications and bariatric surgery may be considered. Currently, most recommendations for bariatric surgery are based on BMI thresholds and the presence of comorbid conditions. However, some have suggested the need for integration of other predictors of metabolic disease such as WC when considering patients for bariatric surgery for the treatment of diseases such as diabetes.13 In conclusion, although thresholds for adiposity may vary across racial/ethnic populations, central visceral obesity is the driving force behind most cardiometabolic diseases including hypertension and diabetes. Although increasing BMI generally correlates with increasing blood pressure and diabetes risk, WC is a better predictor for these diseases and a better discriminator for both prediabetes and diabetes. WC is a reliable, inexpensive, and fairly easy way to assess cardiometabolic risk irrespective of racial/ethnic classification, and further implementation into widespread clinical practice is warranted. Authors acknowledge grant funding from the National Institutes of Health (NIDDK grant 1K08DK099415 and NIGMS grants P20GM104357 and 5U54GM115428) and American Heart Association (AHA 19CDA34760232). All authors have no significant conflicts of interest.