Bone marrow fat contributes to insulin sensitivity and adiponectin secretion in premenopausal women.

Abstract

Bone marrow fat is a functionally distinct adipose tissue that may contribute to systemic metabolism. This study aimed at evaluating a possible association between bone marrow fat and insulin sensitivity indices. Fifty obese (n = 23) and non-obese (n = 27) premenopausal women underwent proton magnetic resonance spectroscopy to measure vertebral bone marrow fat content and unsaturation index at L4 level. Abdominal visceral, subcutaneous fat, and epicardial fat were also measured using magnetic resonance imaging. Bone mineral density was measured by dual-energy X-ray absorptiometry. Body composition was assessed by bioelectrical impedance analysis. Fasting serum glucose, insulin, lipids, adiponectin were measured; the insulin resistance index HOMA (HOMA-IR) was calculated. Bone marrow fat content and unsaturation index were similar in obese and non-obese women (38.5 ± 0.1 vs. 38.6 ± 0.1%, p = 0.994; 0.162 ± 0.065 vs. 0.175 ± 0.048, p = 0.473, respectively). Bone marrow fat content negatively correlated with insulin and HOMA-IR (r = -0.342, r = -0.352, respectively, p = 0.01) and positively with high density lipoprotein cholesterol (r = 0.270, p = 0.043). From a multivariate regression model including lnHOMA-IR as a dependent variable and visceral, subcutaneous, epicardial fat, and bone marrow fat as independent variables, lnHOMA-IR was significantly associated with bone marrow fat (β = -0.008 ± 0.004, p = 0.04) and subcutaneous fat (β = 0.003 ± 0.001, p = 0.04). Bone marrow fat, among the other adipose depots, was a significant predictor of circulating adiponectin (β = 0.147 ± 0.060, p = 0.021). Bone marrow fat unsaturation index negatively correlated with visceral fat (r = -0.316, p = 0.026). There is a relationship between bone marrow fat content and insulin sensitivity in obese and non-obese premenopausal women, possibly mediated by adiponectin secretion. Visceral fat does not seem to regulate bone marrow fat content while it may affect bone marrow fat composition.