Objective: African Americans (AAs) have less visceral and more subcutaneous fat than whites, thus the relationship of adiponectin and leptin to body fat and insulin sensitivity in AA may be different from that in whites.
Methods and procedures: Sixty-nine non-diabetic AA (37 men and 32 women), aged 33 +/- 1 year participated. The percent fat was determined by dual-energy X-ray absorptiometry, abdominal visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) volume by computerized tomography (CT), and insulin sensitivity by homeostasis model assessment (HOMA).
Results: VAT was greater in men (1,619 +/- 177 cm(3) vs. 1,022 +/- 149 cm(3); P = 0.01); women had a higher percentage of body fat (34.1 +/- 1.4 vs. 24.0 +/- 1.2; P < 0.0001), adiponectin (15.8 +/- 1.2 microg/ml vs. 10.4 +/- 0.8 microg/ml; P = 0.0004) and leptin (23.2 +/- 15.8 ng/ml vs. 9.2 +/- 7.2 ng/ml; P < 0.0001). SAT and HOMA did not differ because of the sex. Adiponectin negatively correlated with VAT (r = -0.41, P < 0.05) in men, and with VAT (r = -0.55, P < 0.01), and SAT (r = -0.35, P < 0.05) in women. Adiponectin negatively correlated with HOMA in men (r = -0.38, P < 0.05) and women (r = -0.44, P < 0.05). In multiple regression, sex (P = 0.02), HOMA (P = 0.03) and VAT (P = 0.003) were significant predictors of adiponectin (adj R (2) = 0.38, P < 0.0001). Leptin positively correlated with VAT, SAT, percent fat and HOMA in men (r = 0.79, r = 0.86, r = 0.89, and r = 0.53; P < 0.001) and women (r = 0.62, r = 0.75, r = 0.83, and r = 0.55; P < 0.01). In multiple regression VAT (P = 0.04), percent body fat (P < 0.0001) and sex (P = 0.01), but not HOMA were significant predictors of serum leptin (adj R (2)= 0.82, P < 0.0001).
Discussion: The relationship of adiponectin and leptin to body fat content and distribution in AA is dependent on sex. Although VAT and insulin sensitivity are significant determinants of adiponectin, VAT and percent body fat determine leptin.