Although obesity is often associated with insulin resistance and a cluster of metabolic disturbances, the existence of a subgroup of healthy but obese individuals has been postulated. It is unclear why some obese individuals fail to show traditional risk factors associated with the insulin resistance syndrome despite having a very high accumulation of body fat. To address this issue, we identified and studied a subgroup of metabolically normal but obese (MNO) postmenopausal women to gain insight into potential physiological factors that may protect them against the development of obesity-related comorbidities.

We carefully examined the metabolic characteristics of 43 obese, sedentary postmenopausal women (mean ± sd, 58.0± 6.0 yr). Subjects were classified as MNO or as metabolically abnormal obese (MAO) based on an accepted cut-point for insulin sensitivity (measured by the hyperinsulinemic/euglycemic clamp technique). Thereafter, we determined 1) body composition (fat mass and lean body mass), 2) body fat distribution (abdominal visceral and sc adipose tissue areas, midthigh sc adipose tissue and muscle attenuation), 3) plasma lipid-lipoprotein levels, 4) plasma glucose and insulin concentrations, 5) resting blood pressure, 6) peak oxygen consumption, 7) physical activity energy expenditure, and 8) age-related onset of obesity with a questionnaire as potential modulators of differences in the risk profile.

We identified 17 MNO subjects who displayed high insulin sensitivity (11.2 ± 2.6 mg/min·kg lean body mass) and 26 MAO subjects with lower insulin sensitivity (5.7 ± 1.1 mg/min·kg lean body mass). Despite comparable total body fatness between groups (45.2 ± 5.3% vs. 44.8 ± 6.6%; P = NS), MNO individuals had 49% less visceral adipose tissue than MAO subjects (141 ± 53 vs. 211 ± 85 cm²; P < 0.01). No difference was noted between groups for abdominal sc adipose tissue (453 ± 126 vs. 442 ± 144 cm²; P = NS), total fat mass (38.1 ± 10.6 vs. 40.0 ± 11.8 kg), muscle attenuation (42.2± 2.6 vs. 43.6 ± 4.8 Houndsfield units), and physical activity energy expenditure (1060 ± 323 vs. 1045 ± 331 Cal/day). MNO subjects had lower fasting plasma glucose and insulin concentrations and lower insulin levels during the oral glucose tolerance test (P values ranging between 0.01–0.001). No difference was observed between groups for 2-h glucose levels and glucose area during the oral glucose tolerance test. MNO subjects showed lower plasma triglycerides and higher high density lipoprotein cholesterol concentrations than MAO individuals (P < 0.01 in both cases). Results from the questionnaire indicated that 48% of the MNO women presented an early onset of obesity (<20 yr old) compared with 29% of the MAO subjects (P = 0.09). Stepwise regression analysis showed that visceral adipose tissue and the age-related onset of obesity explained 22% and 13%, respectively, of the variance observed in insulin sensitivity (total r² = 0.35; P < 0.05 in both cases).

Our results support the existence of a subgroup of obese but metabolically normal postmenopausal women who display high levels of insulin sensitivity despite having a high accumulation of body fat. This metabolically normal profile is associated with a lower accumulation of visceral adipose tissue and an earlier age-related onset of obesity.