Adiposity and Fat Distribution Outcome Measures: Assessment and Clinical Implications

Obesity is the excessive accumulation of adipose tissue to an extent that health is impaired. Obesity is usually determined using body mass index (BMI). A BMI ≥30 kg/m2 correlates strongly with obesity-related comorbid conditions and mortality. In 2000, nearly 39 million American adults met the criterion for obesity; namely, a BMI ≥30 kg/m2. Nearly 61% of the U.S. adult population has a BMI ≥25 kg/m2. Thus, overweight and obesity has reached epidemic proportions in the United States, as well as worldwide. Obesity is associated with significantly increased risk of diabetes mellitus, hypertension, dyslipidemia, certain forms of cancer, sleep apnea, and osteoarthritis. In addition, the increasing prevalence of obesity and its associated complications places a tremendous burden on health care use and costs. This epidemic of obesity mandates prompt attention from the health care and preventive health services to minimize the rise in the incidence of new cases of diabetes, heart disease, and other obesity-related complications. The National Heart, Lung, and Blood Institute Obesity Education Initiative Expert Panel recently developed guidelines for identification, assessment, and treatment of obesity ((1)). The Clinical Guidelines recommend measurement of BMI and waist circumference as "vital signs" for evaluating the obese patient. This paper reviews these recommended assessment tools including the rationale and evidence supporting their use in evaluating patients, relationships to health-related variables, and concomitant changes brought about by intentional weight loss. BMI is a measure of the degree of overweight, in relation to the individuals' height, but not a surrogate measure of metabolic body size, such as fat-free mass or body surface area. The National Institutes of Health Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults define normal weight as a BMI of 18.5 to 24.9 kg/m2, overweight as a BMI of 25 to 29.9 kg/m2, and obesity as a BMI of at least 30 kg/m2 ((1)). In 1998, roughly 97 million American adults met the criterion for overweight or obese, defined as BMI >25 kg/m2 ((1)). Data collected by the National Center for Health Statistics indicate that the prevalence of obesity, defined as a BMI >30 kg/m2, has increased from 12.8% between 1976 and 1980 to 22.5% from 1988 to 1994 ((2)). More recent information from a preliminary analysis of 1999 National Health and Nutrition Examination Survey (NHANES) data indicates that the prevalence of overweight has increased to 34.0% and the prevalence of obesity has risen to 27.0% ((3)). Thus, in the United States, 110 million adults are overweight or obese ((3)). Defining overweight and obesity according to cut-points in BMI is consistent with World Health Organization guidelines ((4)), and thus provides a systematic and consistent way to evaluate the worldwide obesity epidemic. It is estimated that worldwide there are roughly 250 million obese adults and 500 million overweight adults. The use of BMI for the classification of weight status is based on epidemiological associations of BMI with morbidity and mortality ((5)). The classification of normal weight, overweight, and obese provides a mechanism to identify those individuals who are at increased risk of being affected (either presently or in the future) by obesity-related health problems. The rationale for the use of the BMI is the demonstrated strong relationship to risk of hypertension, diabetes, dyslipidemia, and overall mortality. As a clinical tool, BMI is simple to calculate, inexpensive, and a well-defined criterion for diagnosis of the severity of obesity. BMI is not a measure of body composition, although it is correlated with body fatness. Recently, Gallagher et al. ((6)) explored the relationship between BMI and percent fat and developed prediction equations for estimating total body fatness from its relationship to BMI. The relationship between BMI and body fat is age- and sex-specific but seems to be consistent across ethnic groups. The limitations of the BMI as an assessment tool are that it does not distinguish between lean and fat tissue and thus could lead to misclassification of overweight among muscular individuals, and it does not delineate the pattern of fat distribution, which has important health implications (see below). The relationship between increasing body weight and health risks is curvilinear and is often described as J-shaped ((7)). In a prospective study of more than 1 million adults in the United States (457,785 men and 588,369 women), Calle et al. ((8)) studied the relation between BMI and the risk of death from all causes. In healthy people who had never smoked, the nadir of the mortality curve occurred at a BMI of 23.5 to 24.9 kg/m2 in men and 22.0 to 23.4 kg/m2 in women. Relative risk of death was greatly increased in subjects with the highest BMIs compared with those who had a BMI of 23.5 to 24.9 kg/m2. A high BMI was most predictive of death from cardiovascular disease, especially in men (relative risk, 2.90). The risk of death from all causes, cardiovascular disease, cancer, or other diseases increases throughout the range of moderate and severe overweight for both men and women in all age groups. Blood pressure is clearly and strongly correlated with BMI. In the INTERSALT study ((9)), the relationship between BMI and blood pressure was studied in over 10,000 men and women between the ages of 20 and 59 years, sampled from 52 centers around the world. BMI was significantly associated with systolic and diastolic blood pressure, independent of age, alcohol intake, smoking habit, and sodium and potassium excretion. The precise prevalence of obesity-related hypertension associated with obesity varies with age, race, and sex of the population studied and with the criteria used for the definition of hypertension and obesity. Roughly 30% of cases of hypertension may be attributable to obesity, and in men under the age of 45 years, this figure may be as high as 60% ((10), (11)). In the Framingham Offspring Study, 78% of cases of hypertension in men and 64% of cases in women were attributable to obesity ((12)), and obesity was noted to be the single best predictor of hypertension incidence. Moreover, even after controlling for initial obesity, at 8 years follow-up, obesity remained a major controllable contributor to hypertension ((12)). Prospective studies have shown that obesity increases the risk of developing hypertension ((13), (14)). Moreover, weight gain in adulthood is in itself an important risk factor for the development of hypertension ((15)). In the long-term Nurses Health Study ((13), (15)), BMI at 18 years of age and midlife were positively associated with the occurrence of hypertension. Weight gain after 18 years of age significantly increased the risk for hypertension (compared with weight gain ≤2 kg, multivariate relative risks were 1.74 for a gain of 5.0 to 9.9 kg, and 5.21 for a gain ≥25.0 kg). Excess weight, and even modest adult weight gain, substantially increase risk for hypertension: each 1-kg increase in weight after age 18 was associated with a 5% increase in risk for hypertension. It is likely that the 25% increase in the prevalence of diabetes in the last 20 years in the United States ((16)) is because of the marked increase in the prevalence of obesity. BMI, abdominal fat distribution, and weight gain are important risk factors for type 2 diabetes. Data from NHANES III indicated that two-thirds of adult men and women in the United States diagnosed with type 2 diabetes have a BMI of 27 kg/m2 or greater ((5)). Moreover, the risk of diabetes increased in a linear relationship with BMI; diabetes prevalence was 2%, 8%, and 13% in those with BMI 25 to 29.9 kg/m2 (overweight), 30 to 34.9 kg/m2 (class I obesity), and ≥35 kg/m2 (class II/III obesity), respectively ((16)). Data from the Nurses Health Study demonstrated that the risk of diabetes begins to increase in "normal"-weight women, when BMI exceeds 22 kg/m2 ((17), (18)). Weight gain and increases in BMI during adulthood also increase the risk of diabetes, even at relatively low levels of BMI in initially normal-weight individuals ((18), (19)). Apart from degree of overweight, the distribution of adipose tissue is also strongly associated with diabetes risk; increased abdominal fat mass increases the risk of diabetes at any BMI value ((20)). It has been argued that the relationship of BMI to mortality from all causes and from cardiovascular diseases is actually a reflection of the association of BMI to individual cardiovascular disease risk factors rather than a direct relationship to mortality. Obesity is associated with increased risk for developing coronary heart disease (CHD), presumably by way of its impact on risk factors, including hypertension, dyslipidemia, impaired glucose tolerance, and type 2 diabetes ((21)). There has been considerable controversy as to whether obesity is simply a correlate of these risk factors or an independent contributing factor. Evidence from long-term observational studies demonstrates that overweight and obesity are predictive of cardiovascular disease, independent of its effects on traditional risk factors ((22)). This increase in relative risk occurs at relatively low levels of overweight (e.g., 72% increased risk for a BMI of 25 to 29 kg/m2 compared with a BMI of <23 kg/m2) and with weight gain after the age of 18 per se, even in individuals with initial normal weight ((7), (23), (24)). Willett et al. ((24)) noted a 2-fold increase in the risk of coronary heart disease in women who had a BMI of 25 to 28.9 kg/m2 and 3.6 relative risk in women whose BMI was ≥29 kg/m2. Weight gain of 8 to 10.9 kg in women after age 18 increased the CHD risk 1.6-fold, and a gain of 11 to 19 kg increased CHD risk by 1.9-fold. In men, a 10% weight gain was shown to be associated with a 38% increase in the risk of coronary heart disease and a 20% weight gain was associated with an 86% increased CHD risk ((25)). In view of this body of evidence and the accepted contribution of obesity to the development of CHD ((26)), the American Heart Association now considers obesity to be a major risk factor for CHD ((21)). Overweight and obesity increase the risk for ischemic stroke but not hemorrhagic stroke. In the Nurses' Health Study, the risk of ischemic stroke was 75% greater in women with a BMI >27 kg/m2 and 137% higher in those with a BMI >32 kg/m2, compared with women who had a BMI <21 kg/m2 ((27)). BMI is positively correlated with associated plasma triglyceride levels, total and low-density lipoprotein (LDL)-cholesterol levels, and inversely related to high-density lipoprotein (HDL) cholesterol levels ((28)). Data from the National Health and Nutrition Examination Survey were used to investigate the relationships between BMI and serum lipid levels in men. Increases in BMI in men ages 20 to 44 years were associated with an increase in LDL-cholesterol level. For middle-aged (44 to 59 years) and older (60 to 74 years) men, increases in BMI were associated with smaller yet significant increases in total cholesterol, whereas the LDL-cholesterol levels were unchanged. Although increasing age per se seems to blunt the deleterious impact of BMI on total cholesterol and LDL-cholesterol levels, the relationship between increases in BMI and increased triglyceride and reduced HDL-cholesterol levels was consistent across all age groups ((28)). In the Framingham Offspring Study, BMI was strongly associated with changes in serum lipids known to increase coronary heart disease risk; subjects with higher BMI had smaller LDL particles and reduced HDL-cholesterol levels. In an analysis of NHANES III data, Brown et al. ((14)) demonstrated that the prevalence of high blood cholesterol and low HDL-cholesterol was significantly increased in relation to increased BMI levels. In recent years there has been considerable interest in the original observations of Vague regarding the health consequences of central obesity, also referred to as abdominal, android, truncal, or upper-body obesity ((29)). Fat distribution, independent of absolute fatness, has been related to a number of health risks and conditions. Measures of abdominal obesity (such as the waist-to-hip ratio or waist circumference) are related to a number of metabolic variables, including cardiovascular risk factors and abnormal glucose metabolism ((30), (31), (32), (33), (34)). Furthermore, abdominal obesity is associated with increased risk of developing diabetes, ischemic heart disease, and stroke ((33), (34), (35)). The risks associated with obesity itself are also exacerbated by the presence of increased levels of visceral fat and central obesity. The National Heart, Lung, and Blood Institute Guidelines suggest a waist circumference above 102 cm (40 in) in men and above 88 cm (35 in) in women indicates a high risk ((1)). While methods for measuring visceral fat (such as computerized tomography and magnetic resonance imaging) are expensive and not suitable for routine clinical use, waist circumference is an easily obtained surrogate measure (see Figure 1). The most recent guidelines of the National Cholesterol Education Program, Adult Treatment Panel III, also include waist circumference as a metabolic risk factor using the same cut-points of 102 cm in men and 88 cm in women ((36)). Moreover, excessive waist circumference is now used as one of the diagnostic abnormalities of the metabolic syndrome. According to the Adult Treatment Panel III, the presence of three of the following abnormalities is diagnostic of the metabolic syndrome: waist circumference >102 cm; serum triglycerides level of at least 150 mg/dL (1.69 mM); HDL-cholesterol level <40 mg/dL (1.04 mM) in men and 50 mg/dL (1.29 mM) in women; blood pressure of at least 130/85 mm Hg; or serum glucose level of at least 110 mg/dL (6.1 mM) ((36)). Technique for measuring waist circumference. Locate the upper-hip bone and the top of the right iliac crest. Place a measuring tape in a horizontal plane around the abdomen at the level of the iliac crest. Before reading the tape measure, ensure that the tape is snug, but does not compress the skin, and is parallel to the floor. The measurement is made at the end of a normal expiration. Dotted line, measuring-tape position for waist (abdominal) circumference in adults. A considerable body of evidence has established the relationship between regional fat distribution and obesity-related morbidity and mortality. At least three body composition components are associated with obesity-related adverse health outcomes: (1) the total amount of body fat (expressed as a percentage of body weight), (2) the amount of subcutaneous truncal or abdominal fat (upper-body fat), and (3) the amount of visceral fat located in the abdominal cavity. These three compartments are highly intercorrelated ((37)), which makes it difficult to determine the relative contribution of each individual compartment to metabolic variables. Each of these components of body fat is associated with varying degrees of metabolic abnormalities and independently predicts adverse health outcomes ((38)). The study of abdominal adipose tissue distribution by computed tomography and magnetic resonance imaging has advanced our understanding of the health consequences of visceral adiposity ((30), (39), (40)). There is a strong association between visceral adiposity and insulin resistance ((41), (42)). Visceral adipose tissue is also related to essential hypertension, dyslipidemia, and other factors, such as impaired fibrinolysis, that contribute to heightened risk of cardiovascular disease ((39), (43), (44), (45), (46)) A number of prospective epidemiological studies have shown that increased abdominal obesity is associated with increased cardiovascular risk and total mortality. Research by Després and colleagues ((31), (42), (47)) argues strongly in favor of a primary role of visceral fat in both the pathogenesis of obesity-related comorbidities and their reduction with weight loss. The existing literature, however, is inconsistent and inconclusive with regard to whether individuals who have a high BMI without abdominal obesity are also at an increased risk for various adverse health consequences. Other investigations have challenged the clinical usefulness of waist circumference by noting the redundancy with BMI in the identification of patients at risk ((48)). It has been proposed that an algorithm that uses BMI and the presence of CHD risk factors may be sufficient to identify high-risk patients. The relationship of abdominal obesity to the insulin resistance syndrome, including an atherogenic serum lipoprotein profile that substantially increases coronary heart disease risk, has been well-studied ((30)). However, some investigators ((49)) have challenged the significance of visceral fat as the major determinant of insulin resistance. Goodpaster et al. ((50)) found that central subcutaneous fat made an additional significant contribution to the prediction of insulin sensitivity beyond that of visceral fat, suggesting that visceral fat may not be the most powerful modulator of insulin resistance. Furthermore, these investigators and others also observed a significant correlation between intramuscular fat and insulin sensitivity, and concluded that increased fat content in muscle may also be an important independent marker of obesity-related insulin resistance ((51)). Data from the Framingham offspring have shed light on the clustering and interrelationships among risk factors, including fat distribution; the insulin resistance syndrome seems to reflect the summation of three distinct physiological domains linked through mutual associations with obesity and hyperinsulinemia: central metabolic syndrome, impaired glucose tolerance, and hypertension ((45)). According to this analysis, BMI, but not fat distribution, links blood pressure to the lipid components of the insulin resistance syndrome, whereas hyperinsulinemia seems to link impaired glucose tolerance to the syndrome. The mechanisms by which visceral or abdominal adiposity is linked to the insulin resistance syndrome include higher responsiveness to lipolytic stimulation (such as catecholamines) and greater resistance to antilipolytic stimuli (such as insulin) in visceral compared with subcutaneous adipocytes, and the consequent increased free fatty acids draining into the portal circulation ((52)). The link between abdominal obesity and heart disease is partly modulated by the relationship of abdominal obesity to lipid abnormalities: excess visceral or abdominal fat has also been shown to be independently associated with larger very low-density lipoprotein (VLDL); small, dense LDL; and lower levels of HDL-2 ((53)). Increased abdominal fat mass as well as waist circumference have been shown to be associated with an increased risk of type 2 diabetes across the whole range of BMI values ((54), (55)). In summary, epidemiological evidence suggests that obesity (BMI) and body fat distribution (waist circumference) are each independently associated with an increased risk of diabetes and coronary heart disease, as well as deleterious effects on the metabolic risk variables which predispose an individual to develop these diseases. Both increased BMI and excessive abdominal fat are correlated with a number of cardiovascular and metabolic risk factors that improve or reverse with weight loss. Modest weight loss produces regional fat loss as well as reductions in visceral fat content ((56)). Whereas loss of visceral fat is strongly associated with metabolic improvements ((57)), it has not been proven that loss of visceral fat independently and selectively accounts for weight-loss related improvements in metabolic variables and risk factors. It is noteworthy that total weight loss of as little as 5% would translate into a >5% loss of visceral fat ((50), (57)). Thus, the subset of obese patients with predominantly abdominal obesity and marked metabolic disturbances may benefit more from modest weight loss than patients at lower risk (see Figure 2). Relationships among waist circumference, accumulation of visceral adipose tissue, and the metabolic risk profile, as well as the effect of 5% to 10% weight loss. Figure appears from an article by Després et al. ((78)), courtesy of BMJ. A substantial body of evidence indicates that intentional weight loss is beneficial, further strengthening the argument that aggressive efforts to facilitate sustained body-weight reduction are needed. The Swedish Obese Subjects study of obesity surgery demonstrated graded improvements in blood pressure, HDL-cholesterol, triglycerides, insulin, and glucose with weight losses of up to 40 kg maintained for 2 years ((58)). This intentional weight loss also produced a 5-fold decrease in the risk of developing diabetes over an 8-year follow-up period. However, more recent results from this study did not confirm the durability of the beneficial effect of weight loss on blood pressure that was initially observed; after the third year of follow-up, both systolic and diastolic blood pressure gradually increased over the subsequent 5 years ((59)). of the relationship of intentional weight loss to mortality risk indicate that intentional weight loss in men and women with type 2 diabetes was associated with a 25% reduction in mortality and a decrease in death by cardiovascular diseases and diabetes The relationship of intentional weight loss to mortality in individuals is there seems to be a beneficial effect of intentional weight loss on in those overweight or obese individuals who are affected by obesity-related health complications A to evaluate the effect of weight loss on plasma lipid levels indicated based on the results of clinical for each 1 kg of weight total cholesterol was reduced by LDL-cholesterol by triglyceride levels by and HDL-cholesterol increased by at a body weight, but by during weight loss et al. whether 5% to 10% weight loss is associated with sustained long-term changes in lipid levels HDL-cholesterol levels significantly during the 8 of weight but to values by This study the of in the of the plasma lipid to weight loss as well as the of weight and BMI changes from changes in the lipid et al. that a reduction in central abdominal fat not other or was correlated with a atherogenic lipid et al. found that waist reduction related to weight loss in women and that the reduction in waist circumference for age, alcohol and correlated strongly with reductions in total cholesterol and LDL-cholesterol but not with HDL-cholesterol or triglyceride levels. BMI to changes in lipid levels, whereas changes in the waist-to-hip ratio were not correlated with lipid For those women whose waist measurement by at least 5 10% or greater improvements occurred in at least one risk factor. These results suggest that reduction of waist circumference of 5 to cm in women considerable health and the use of waist circumference as an assessment a amount of weight loss in overweight individuals is associated with a decrease in pressure over the last most studies have demonstrated that treatment of hypertension with weight loss in lower blood Although the in blood pressure has been in to a concomitant reduction in sodium intake, which occurs when total is weight loss has an independent effect even in the of sodium Weight loss also the risk for developing the risks were reduced by for weight loss of 5.0 to 9.9 kg and by for a weight loss of kg or more Weight loss clearly is one of the most of blood pressure and is the of treatment for hypertension Modest weight loss also or the for studies suggest that central fat distribution is a predictor for hypertension than overall fat mass circumference above 88 cm in women and above 102 cm in men correlates with blood pressure levels both in and et al. that blood pressure reduction by weight loss correlated with changes in visceral fat and not with changes in body weight or BMI. This suggests that a decrease in fat blood pressure in patients and the use of waist circumference, a surrogate marker for abdominal fat, as a tool to and obesity-related Weight loss clearly beneficial metabolic changes in with abnormal glucose tolerance and type 2 diabetes. Weight loss in obese patients with diabetes and in obese patients both and insulin to reduced glucose and glucose Weight loss and and in patients with severe and may even increase insulin although this is not consistent In the long-term United Prospective Study, treatment of diagnosed type 2 diabetes patients for with a in a weight loss of kg, which was associated with reduction in plasma glucose and a decrease in from a level of These are consistent with substantial evidence that a amount of weight loss Weight loss of 5% over 1 year reduced insulin, and levels and diabetes in obese, patients, and weight loss of or more the for Weight loss of as little as 5% to 10% has been shown to or the development of type 2 diabetes the Program, a by the National Institutes of demonstrated that compared with treatment with an which produced a weight loss of kg reduced the risk of from impaired glucose tolerance to diabetes by over a follow-up of years Although weight loss clearly the insulin resistance that is of obesity, it whether this effect is related to changes in regional fat visceral or abdominal or to total body weight loss. of some investigations the that improvements in insulin sensitivity are correlated with the reduction in visceral fat whereas others not As it is difficult to distinguish the individual of total weight loss and regional fat losses to changes in metabolic variables because these total weight losses and visceral fat losses are strongly evidence indicates that both increased body weight and body fat are associated with an increased risk for adverse health consequences. BMI and waist circumference are simple tools to for assessment of their patients to identify patients who would benefit from further risk assessment and who are in of obesity These tools are in the of patients over during