BMI or Waist Circumference for Assessing Risk in Obese Patients?

From the American Diabetes Association

NEW YORK CITY—At a recent meeting of the American Diabetes Association, Michael Jensen, MD, professor of medicine at the Mayo College of Medicine in Rochester, Minn, elaborated on the current debate in the medical community about the best measurement for assessing excess morbidity and mortality in overweight persons.

Excess body mass index (BMI) is a well-known predictor of morbidity and mortality, Dr Jensen says, but many now suggest that waist circumference is a more beneficial measure. “There has been a lot of hype about body-fat distribution, especially central obesity, as another independent predictor.” The question is, Dr Jensen says, whether we should be using waist circumference when making decisions about patient care.

When added to BMI, waist circumference can provide additional information, but for practical purposes, in the primary care setting, he emphasizes, “BMI—height and weight—is the easiest, because it is already being done routinely….The utility of waist circumference is that it is supposed to be an added value, that is, while not an independent predictor of obesity-related disease, at least a partial predictor.” Although both BMI and waist circumference correlate well with total body fat, they do not correlate as well “with the percent body fat.”

The advantage of waist circumference is that it better predicts the amount of intraabdominal or visceral fat, which is “important when determining the patient’s risk for obesity-related disease.” Excess intraabdominal adipose tissue has been correlated with higher risk of cardiovascular disease and all-cause mortality.

Waist circumference is “about twice as good at predicting future coronary heart disease as BMI alone,” Dr Jensen says. And the evidence shows that using both BMI and waist circumference “is a better predictor of the risk for type 2 diabetes than BMI alone.” Thus using the 2 measurements together is better at predicting morbidity and mortality risk than BMI alone, but measuring waist circumference is not that simple and is not practical in primary care.

Measuring waist circumference is fine, he says, “if you’ve got a guy who hasn’t had any kids.” In contrast, women often appear to have a normal or low waist circumference when they stand up, but when you measure their waist circumference while they lie down, “it’s pretty obvious that they’ve got severe abdominal obesity.”

Furthermore, most physicians can easily get their patients’ weight and height, so calculating their BMI is quite simple. In contrast, there are different ways of measuring waist circumference. “Some measure it midway between the lowest rib and the iliac crest; others will measure it at the umbilicus. Some people measure it at the narrowest, others at the widest,” he says.

A decision to use waist circumference as the criterion for risk assessment would require a positive reply to the question—“Would waist circumference measurements identify a nontrivial number of patients who would be at increased cardiometabolic risk and who you wouldn’t otherwise suspect on the basis of BMI alone?”

Based on the evidence from clinical studies, only about 3% to 4% of women and less than 1% of men with a normal BMI will have a high waist circumference. “So if you’re measuring waist circumference in people with a BMI of 25 or less, you’re going to be measuring a lot of waist circumferences to pick up very few high-risk people,” Dr Jensen says.

At BMIs between 25 and 30 kg/m², research suggests that about 15% to 20% of women and 4% to 10% of men are going to meet the high-risk waist circumference criteria of >88 cm in women and >102 cm in men. Even so, more than 99% of men and 98% of women would be treated the same way, if you follow the National Heart, Lung, and Blood expert panel recommendations for obesity management, regardless of their waist circumference.

“Except for a BMI between about 27 and 35, waist circumference is of little value. Even in these ranges, a waist circumference measurement will not help, unless the value is on the chart for the physician’s quick review and is linked with prompts to check on risk factors, such as lipids and glucose level,” Dr Jensen says.

Adding a low waist circumference could provide reassurance if patients have a BMI in the “risk range,” he says, since this may suggest that the patient “is not somebody who you need to be aggressively intervening for to reduce cardiovascular risk.” But this is not useful for those who are already obese, in whom waist circumference is clearly high.

Finally, new findings from an Australian study that is expected to be published later this year indicate that measuring waist-to-hip ratio may be a better predictor of future mortality than waist circumference alone.

ONLINE EXTRA

Additional Comments from Dr Jensen

Waist circumference should be measured while the patient is lying down. You should imagine what a computed tomography (CT) scan would show at the area that you are measuring. “When was the last time you saw someone get a CT scan standing up? You don’t,” Dr Jensen points out. For some patients with a large waist circumference, the CT scan would show a lot of subcutaneous fat; for others, a lot of visceral fat. Measuring waist circumference adds information about the amount of intraabdominal adipose tissue, which is important for determining the patient’s risk for obesity-related comorbidities.

Most studies have shown that, when waist circumference is added to BMI, you can better predict a patient’s risk for type 2 diabetes (Figures 1A and 1B) and for all-cause mortality. Waist circumference is also about twice as good at predicting future coronary heart disease than BMI alone.

Some research suggests that lower body fat is associated with better overall health. The more lower body fat—irrespective of the amount of upper body fat—the less likely a patient is to develop cardiovascular (CV) disease or diabetes. “So I think it may be more complex than just saying ‘upper body fat is bad,’” says Dr Jensen. What would be really bad is if someone has upper body fat but almost no lower body fat. People are probably better off if they have both upper body fat and a generous amount of lower body fat.

Just like BMI, the meaning of a particular waist circumference value can vary according to gender, age, and race or ethnicity. “Some ethnic groups seem to have a lot higher cardiometabolic risk at a lower BMI and a lower waist circumference,” Dr Jensen points out. Several questions are involved. Did you measure it right? What’s the age? What’s the sex? What’s the ethnicity? “It gets really complex to put this into a hard clinical tool,” he says.

The current waist circumference cut points of 102 cm in men and 88 cm in women are used to determine health risks.

Reductions in waist circumference appear to be a fairly good predictor of the health response to diet and exercise. People whose waist circumference is decreasing in response to a lifestyle intervention are those most likely to be enjoying improvements in their metabolic risk profile.

“The message I would take home is you probably don’t need waist circumference in people with a BMI under 25. You don’t need it if they’re over 35,” Dr Jensen says. “If you’ve already got all their lab values and their blood pressure sitting there in front of you, you probably don’t need it.”

3 Hypotheses Linking Waist Circumference and Metabolic Risk

There are currently 3 main hypotheses for the biologic mechanism underlying the relationship between a large waist circumference and increased metabolic risk.

Free fatty acid hypothesis

Dietary fat is turned into triglycerides, which are broken down into fatty acids, which are then taken up by fat cells, where they are stored until needed as fuel sources for most body tissues, including muscle, liver, and heart. People who gain weight predominantly in the lower part of the body tend to have normal fatty acid metabolism, whereas those who gain weight in the apple shape and who have a big waist circumference have high levels of free fatty acids.

Evidence was shown that women with upper body obesity had high levels of free fatty acids in the blood, which predisposes to insulin resistance.

However, people with the same amount of fat but with a small waist circumference tend to have normal free fatty acid levels in their bloodstream and more normal health.

Adipokine hypothesis

In the past decade, adipose tissue has been found to produce the 3 hormones leptin, resistin (which appears to cause insulin resistance), and visfatin (an insulin-sensitizing hormone). The adipokine hypothesis suggests that dysregulation in the release of some of these adipokines might be contributing to the adverse health consequences associated with a large waist circumference. And, except for leptin, levels of these hormones are more abnormal in people with large waists than in those people with large hips.

However, no clear link has been established between any of these hormones and CV or metabolic abnormalities.

Impaired fat storage (ectopic fat) hypothesis

The third hypothesis centers around the idea that people with a large waist circumference store fat in their upper body, because they cannot store it normally in “safer depots” such as in the legs. Research suggests that leg fat may be somewhat protective against the complications of obesity. If fat cannot be stored in the leg compartments, the default storage location appears to be the upper body subcutaneous region. And if the fat cannot be stored there (as may be the case in obese individuals), then the fat may be stored in the islet cells of the pancreas, the liver, and the muscle, which would contribute to the well-known complications of obesity.