The BMI: Is It Time to Scratch for a More Accurate Assessment of Metabolic Dysfunction?
The body mass index (BMI = Kg/M2) is not a valid measure for clinical decisions, especially whether a patient will benefit from bariatric surgery. The measure, as used, discriminates against the muscular, the aged, women, and racial groups such as Asians and African Americans. The requirement must be eliminated since it denies many patients the only currently available therapy. This chapter provides the bibliographic data to support this argument and should prove useful in convincing carriers that the BMI is an inaccurate and, too often, cruel guideline.
KeywordsBMIObesityBariatric surgeryMetabolic syndrome
In the nineteenth century, Belgian social scientist, Adolph Quetelet (1796 – 1874), developed a formula to describe the “average man.” This formula was a simple measurement: weight (kilograms) divided by height (meters2). Today we now know this equation as body mass index (BMI = Kg/M2) and, since the 1991 National Institutes of Health Consensus it has served as the main criterion in determining the eligibility of patients for bariatric surgery .
An index for the measurement of obesity is certainly needed. The last 20 years have seen an explosion in the obesity pandemic, insulin resistance diabetes, and the constellation of criteria known as metabolic syndrome. Today, one out of every four adults over 65 in the US is afflicted with type 2 diabetes (T2DM). The World Health Organization estimates that by the year 2030 close to 336 million people will have T2DM, which is associated with higher cardiovascular and metabolic diseases [2•]. Rough estimates suggest that patients with health care utilization costs for metabolic syndrome increase by an average of 24 % per additional risk factor .
The cost, morbidity, and mortality that these diseases inflict on society are tremendous. Unfortunately, in spite of decades of research, development of new and expensive medical therapies, T2DM still remains the primary cause of blindness, amputation, and renal failure in the US today. With health care costs exploding over the last 30 years safer and more durable treatments are badly needed. Short of a “magic pill” that can reverse these disease processes, metabolic surgery offers a useful option.
Three prospective randomized studies have proven the benefits of surgery compared to intense medical management [4••, 5••, 6]. With the advent of Centers of Excellence, metabolic surgery has proven to be as safe as a cholecystectomy, with a 90 day mortality of 0.6 % (LAGB: 0 %, RYGB: 0.3 %, Gastric sleeve: 0.1 %, Duodenal switch: 1.0 % .
World Health Organization (WHO) 1999 criteria for metabolic syndrome
Triglycerides ≥ 1.695 mmol/L
HDL ≤ 0.9 mmol/L (male)
HDL ≤ 1.0 mmol/L (female)
Waist:Hip Ratio > 0.90 (male)
BMI ≥ 30 kg/m2
Albumin excretion ratio ≤ 20 μg/min
Albumin:Creatinine ratio ≥ 30 mg/g
Assessment of Obesity
The current indices for obesity, the body mass index (BMI = Kg/M2) and the WHO Classification, based on the BMI, fail to distinguish weight from adiposity. Body mass index (BMI) is a simple index of weight-for-height that is commonly used to classify underweight, overweight and obesity in adults. It is defined as the weight in kilograms divided by the square of the height in meters (kg/m2). For example, an adult who weighs 70 kg and whose height is 1.75 m will have a BMI of 22.9. (BMI = 70 kg / (1.75 m2) = 70 / 3.06 = 22.9).
The International Classification of adult underweight, overweight, and obesity according to BMI
Principal cut-off points
Additional cut-off points
Obese class I
Obese class II
Obese class III
It is time for a more accurate tool beyond BMI for assessment and treatment of people with metabolic syndrome. BMI has proven to be useful in measuring epidemiologic population-based relationships between obesity and mortality rates. The BMI fails to encompass the complex clinical nature of obesity and metabolic syndrome. Factors that are known to be important in the development of metabolic syndrome, such as age, race, muscle mass, adipose distribution, and gender are not part of the equation. In addition, the BMI does not account for racial and ethnic disparities. Current BMI classifications for obesity do not translate to populations outside European decent [11–13]. Razak et al. studied four ethnic groups (South Asian, Chinese, Aboriginals, and Europeans) from four regions of Canada and found new BMI cutoff points derived from cardiometabolic risk factors to define obesity. These findings suggest that defined obesity was up to ~6 Kg/M2 lower for patients of non-European decent . These results have significant clinical relevance given the global emergence of obesity.
BMI fails to account for the percentage of a person’s weight that is attributed to lean muscle mass vs. adipose mass. Human skeletal muscle density is approximately 18 % greater than that of fat . The classically used example is the BMI range of an American football player. In 2012 the average BMI in the National Football League was 31.35 kg/m2 ± (23.01-45.64). Five separate positions, running back, tight end, defensive end , linebacker, and lineman had an average BMI >30 kg/m2 which, under that current criteria, would be considered at least class I obesity . The fastest running back at East Carolina University was 5′7″ tall and weighed 307 lbs with a BMI of 47. Those data would certainly have made him eligible for bariatric surgery with every major insurance carrier, but our surgical staff could not catch him! This example proves that BMI as strict criterion measurement of obesity is flawed. These athletes are often in peak condition and are often not obese.
Over the last decade, research has started to elucidate the true metabolic nature of adipose tissue. A person’s metabolic status based on the distribution of adipose tissue has a significant impact over most outcomes. There is mounting evidence that there may be a preferential distribution of adipose that is protective of insulin resistance. Subcutaneous adipose tissue with central and thigh distribution seem to be protective in nature. Conversely, the visceral abdominal fat is implicated in a higher rate of insulin resistance. The implication of different body compositions such as, android “apple” or gynoid “pear,” are being elucidated. Studies are now associating higher volumes of visceral abdominal adipose tissue with increased insulin resistance, inflammatory markers, and overall risk of cardiovascular events. It has been noted, however, that total visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) are both correlated with metabolic risk, but VAT is more strongly associated with insulin resistance and cardiovascular outcomes. The differences in metabolic behavior may be due, in part, to the fact that visceral fat drains through the portal system directly from the gut to the liver while subcutaneous, retroperitoneal, and brown fat drain into the systemic circulation.
In a subset analysis of the Framingham Heart Study, participants that were free of cardiovascular disease were scanned for SAT and VAT volumes. The analysis showed that VAT is more strongly associated with an adverse metabolic syndrome risk profile for both men and woman (woman: SAT (OR 3.0); VAT (OR 4.7) men: SAT (OR 2.5); VAT (OR 4.2) p < 0.0001). Interestingly, the average BMI (woman 26.7 ± 5.4 and men 28.3 ± 4.4) for this population was well below the observed criteria for metabolic syndrome, yet there was a definite correlation observed in the distribution of adipose and metabolic syndrome beyond BMI . These observations are especially true of Asians. This and multiple other studies suggest that BMI measurement fails to account for the disparity and impact of this adipose distribution.
Age and gender also play a significant role in the distribution of adipose tissue. Humans lose muscle and bone mass as they age. This is often replaced by a higher percentage of adipose tissue. Males and females have different percentages of fat even with the same calculated BMI. BMI as a uni-metric measurement fails to capture these differences.
Despite these major drawbacks, BMI is still used as a screening tool for metabolic syndrome. BMI has been adopted as the standard for most major private and federally funded insurance carriers for access to metabolic surgery. This automatically eliminates a patient with class I obesity even if they have diabetes, hypertension and dyslipidemia, i.e., metabolic syndrome. Today, surgery is safe and provides the only potential cure for metabolic syndrome. In a meta-analysis, Cohen et al. reported that metabolic surgery outcomes in patients with diabetes and low BMI found resolution of hypertension in 58 %, hypercholesterolemia in 64 %, and hypertriglyceridemia on 58 % of patients. They estimated a ten-fold relative risk reduction in fatal and nonfatal stroke (57 % and 50 % respectively) and fatal and non-fatal coronary heart disease (84 % and 73 % respectively) [18, 19].
Metabolic surgery also provides protection from the development of diabetes. In 1987, the Swedish Obese Subjects (SOS) study was initiated. At this time, there were no strict BMI eligibility criteria for metabolic surgery, thus early in the study patients with BMI <35 kg/M2 were eligible and received surgery. Sjostrom et al. evaluated patients enrolled in the SOS study with low BMIs that would not be eligible under the current 1991 NIH consensus criteria and found a similar reduction in cardiovascular risk factors. This report also concluded that metabolic surgery provides a durable remission of diabetes in less obese patients. The only weakness of this study was that a majority of the patients underwent a vertical band gastroplasty (VBG), a standard operation at the time, but a surgery that is rarely performed in the United States today. Regardless, the implication that current BMI cutoff levels do not correlate with the benefits of diabetes prevention, decreased coronary events, cancer, and mortality hold true .
There have been many alternative measurements proposed to replace the BMI in clinical practice. Unfortunately, a quantitative measure that surpasses BMI in its ability to predict cardiovascular and metabolic diseases has not been found. This measure would have to fulfill three main criteria. First, it must be easily accessible, i.e., not requiring expensive equipment. Second, measurements must be standardized so that clinicians can reproduce the results. Finally, the measure must be validated based on age, gender, and ethnicity. Only then would physician have the elusive holy grail of metabolic syndrome.
In a recent meta-analysis study, Ashwell et al. reported that waist-to-height ratio (WHtR) and waist to circumference (WC) in multiple ethnic populations were better predictors of cardiometabolic risk factors when compared to BMI. WHtR proved to be a better anthropometric measurement for diabetes, hypertension, dyslipidemia, cardiovascular disease, and overall health. Remarkably, in an analysis of a subset of men, WHtR did not show the same statistical significance for metabolic syndrome (men AUC 0.003, p = 0.662 and woman AUC 0.009, p = 0.04) . This suggests that there are still alternative methods that need to be taken into consideration beyond simple measurement techniques. Bergman and his colleagues offered another equation to assess adiposity: BAI = (HC)/((height)1.5)-18, but at this time, it has not passed validation [22, 23].
Adjusted body mass index (ABMI)
ABMI = BMI + calculated points
ABMI > 35 kg/M2 = metabolic surgery
Obstructive Sleep Apnea
In conclusion, no longer should adipose tissue be looked at as a simple storage cell for fat. Instead, current research has reshaped our understanding of adipose as a metabolically active system that exerts great influence on the body as a whole. It is clear that adipose tissue is an important mediator for both cardiovascular risk and metabolic syndrome. It is time to look beyond BMI. The scientific community must take into account the complex interplay of hormones and cytokines that leads to the constellation of diseases known as metabolic syndrome. It is time to adopt a new quantitative standard for determination of metabolic syndrome that encompasses age, ethnicity, adipose distribution, fitness, and gender, a metric that would also lead to a far more accessible fair therapeutic approach to the obese, most of whom are poor and disadvantaged.
Compliance with Ethics Guidelines
Conflict of Interest
E. Charles Moore declares that he has no conflict of interest.
Walter J. Pories has received grant support from Johnson & Johnson, Ethicon, Golden Leaf, GlaxoSmithKline, and National Institutes of Health.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.