Lipid profile and bone paradox: Higher serum lipids are associated with higher bone mineral density in postmenopausal women

Background: Previous studies suggest a relationship between cardiovascular disease (CVD) and osteoporosis; however, the mechanism of the relationship and whether serum lipids are positively or negatively associated with bone mineral density (BMD) are unclear. Methods: We investigated the relationship among serum lipids, dietary saturated fat, BMD of various skeletal sites, and markers of bone turnover. This was a cross-sectional analysis in 136 Caucasian, healthy, postmenopausal women, who were not taking lipid-lowering medications or drugs affecting bone metabolism. BMD at multiple skeletal sites was assessed by DXA. Concentration of serum triglycerides, cholesterol, osteocalcin (OC), and undercarboxylated osteocalcin (UOC) and urinary cross-linked N-telopeptides were analyzed by routine methods. Saturated fat, total calcium (food and supplements), total vitamin K, alcohol, and energy intake were estimated using 3-day dietary records. Physical activity was assessed and used as a confounder with other anthropometric measurements. Results: Serum triglycerides were positively related to femoral shaft BMD and serum cholesterol to total body BMD (p < 0.05). Also, subjects with serum triglycerides above the median had significantly higher BMD in femoral Ward's triangle than those below the median (p = 0.037, by ANCOVA). Subjects with a serum cholesterol level of >= 240 mg/dL (cutoff for increased risk for CVD) had significantly higher BMD at the total body and at all sites of the femur (except neck). There was no relationship between serum lipids and markers of bone turnover. Saturated fat intake was not associated with BMD of any skeletal site. Conclusions: These findings indicate that higher levels of serum triglycerides and cholesterol are positively associated with BMD of various skeletal sites. The mechanism of this association is not clear, and studies are needed to clarify this relationship.