Relation of Body Fat Distribution to Metabolic Complications of Obesity

Studies of body fat topography indicate that in men, adiposity is largely confined to the upper parts of the body. In women, the adipose mass may predominate in the upper body segment or in the lower segment (buttocks and thighs). In an epidemiological survey, there was a continuous rise in the frequency of clinical diabetes as the waist-to-hip girth ratio increased. It was found by relative risk analysis that obesity alone accounted for a diabetic risk of 3.17, and localization of fat in the upper body segment increased the risk to 10.34. The authors investigated two age- and weight-matched subgroups of obese women with either upper or lower body segment obesity. They determined the morphology and metabolic activity of fat cells obtained from subcutaneous adipose tissue of upper and lower body segments and their relationship to the metabolic aberrations of obesity. The results suggested that a classification of obesity based upon the pattern of body fat distribution may provide a means of assessing the susceptibility of obese women to glucose intolerance, hyperinsulinemia, and hypertriglyceridemia. Twenty-five moderately obese (percentage of ideal body weight, 130 to 190 per cent) females between the ages of 20 and 40 years were separated into two subgroups: 16 subjects with predominantly upper body segment obesity and nine subjects with predominantly lower body segment obesity. The two obese subgroups were matched for age (32 ± 1 vs. 32 ± 2 years), body weight (106 ± 4 vs. 90 ± 5 kg), and percentage of ideal body weight (162 ± 5 per cent vs. 163 ± 7 per cent). Nine nonobese (percentage of ideal body weight, 90 to 105 per cent) women of a comparable age group served as controls. Subjects known to suffer from diabetes, hypertension, or clinically recognizable heart disease were excluded. Subjects were admitted to the Clinical Research Center and stabilized on a weight maintenance diet providing 2500 to 3000 calories per day with a constant composition (40 per cent carbohydrate, 20 per cent protein, and 40 per cent fat). Oral glucose tolerance tests were performed, after an overnight fast, with 40 g of glucose/ m2 of surface. Blood samples were analyzed for glucose and insulin. Glucose and insulin areas were calculated as follows: area (units) = sum of half the fasting value plus the values at 30, 60, 90, and 120 minutes and half the 180-minute value. Glucose intolerance was considered present when the subject exhibited all of the following abnormalities: 60-minute value of 200 or more mg/ dl; 90-minute value of 165 or more mg/dl; and 120-minute value of 140 or more mg/dl. In 16 upper and nine lower body segment obese and nine control subjects, adipose tissue biopsies from both abdomen and thigh were obtained at 0800 hours after a 10-hour overnight fast. The interval between the two biopsy procedures did not exceed 20 minutes, and care was taken not to infiltrate the adipose tissue with local anesthetic. Fat cell size and adipocyte adrenergic receptor activity were determined by standard methods. Comparisons between groups were made by means of a nonparametric method for unpaired measurements (Mann-Whitney U test). Student's t-test for paired data was employed to evaluate the effects of catecholamines on adipose tissue activity. The mean fasting glucose concentrations in both obese subgroups were not significantly different from those in nonobese control subjects. The upper body segment obese glucose curve, however, showed higher levels compared to the lower body segment obese curve. This difference was significant at all points except the 180-minute value. The glucose area in the former group was significantly higher than that in the latter (mean ± SE, 801 ± 33 vs. 628 ± 23 U; P < 0.005). Ten of 16 upper body segment obese women had glucose tolerance results falling within the diabetic range. The total plasma glucose area of the lower body segment obese women was only 18 per cent above that of nonobese controls