Cortisol metabolism in human obesity : Impaired cortisone→cortisol conversion in subjects with central adiposity

For a given body mass index (BMI), mortality is higher in patients with central compared to generalized obesity. Glucocorticoids play an important role in determining body fat distribution, but circulating cortisol concentrations are reported to be normal in obese patients. Our recent studies show enhanced conversion of inactive cortisone (E) to active cortisol (F) through the expression of 11beta-hydroxysteroid dehydrogenase type 1 (11betaHSD1) in cultured omental adipose stromal cells; the autocrine production of F may be a crucial factor in the pathogenesis of central obesity. We have now analyzed F metabolism in subjects with BMIs between 20-25 kg/m2 (group A), 25-30 kg/m2 (group B), and more than 30 kg/m2 (group C; n 12 in each group; six males and six premenopausal females; aged 23-44 yr). Glucose/insulin were measured using a 75-g oral glucose tolerance test, and each subject had total body and regional fat (scapular, waist, hip, and thigh) quantified using dual energy x-ray absorptiometry. Urinary total F metabolites (measured by gas chromatography/mass spectrometry) were increased in subjects with obesity [group A, 11,176 +/- 1,530 microg/24 h (mean +/- SE); group C, 13,661 +/- 1,444], although not significantly so (P = 0.08). There was a significant reduction in the urinary tetrahydrocortisol (THF) +/- 5alpha-THF/tetrahydrocortisone (THE) and the cortol/cortolone ratio in obesity (group A vs. C, 1.06 +/- 0.08 vs. 0.84 +/- 0.04 and 0.41 +/- 0.03 vs. 0.34 +/- 0.03, respectively; both P < 0.05). Urinary free F (UFF) excretion was similar in all three groups, as was the UFF/urinary free E (UFE) ratio. The 0900 h circulating F, E, and ACTH pre- and postovernight 1-mg dexamethasone suppression values were similar in all three groups, but a reduction in the generation of serum F from dexamethasone-suppressed values after oral cortisone acetate (25 mg) was evident in both obese groups [e.g. 546 +/- 37 nmol/L in group A vs. 412 +/- 40 in group B (P < 0.05) and 388 +/- 38 in group C (P < 0.01) 180 min post-E]. Insulin resistance was present in groups B and C, but regression analysis revealed no relationship between F metabolites or the THF +/- 5alpha-THF/THE ratio and insulin action (homeostasis model assessment analysis and insulin values in the oral glucose tolerance test). There was, however, a highly significant relationship between the THF +/- 5alpha-THF/THE ratio and BMI (t = -3.44; P < 0.01) and total body fat (t = -2.27; P < 0.05). Stepwise regress