Context:Accumulating evidence indicates that vascular and bone mineralization may be related, though the exact mechanism remains unknown.

Objective: To investigate whether an observed inverse association between bone mineral density (BMD) and coronary artery calcification (CAC) in postmenopausal women currently taking estrogen therapy (ET) is mediated by osteoprotegerin (OPG) or RANK ligand (RANKL).

Design: Participants were 92 postmenopausal women (aged 58–81 years) taking ET who had hip and spine BMD assessed by dual-energy x-ray absorptiometry and CAC measured by electron-beam computed tomography in 1998–2002, and serum RANKL and OPG levels measured in samples collected in 1997–1999. Total CAC score was dichotomized as "none/minimal" ( 10) versus "some" (>10).

Results: OPG serum levels were higher in women who had some CAC compared to those who had none/minimal (126.8±1.08 vs. 102.9±1.07 pg/ml, respectively, p=0.03); these differences became non-significant after adjustment for age and other risk factors (p=0.51). A one standard deviation increase in hip BMD was associated with significantly lower odds of having CAC > 10 (OR=0.52; 95% CI: 0.29–0.93) independent of age, fat-free mass, HDL cholesterol, current smoking, and use of cholesterol-lowering medications. Other skeletal sites demonstrated a similar pattern. Addition of RANKL and/or OPG to the model had minimal effect on the magnitude or statistical significance of the BMD-CAC association. Additionally, a test of interaction indicated that RANKL and OPG are not significant effect modifiers.

Conclusions: Serum OPG and RANKL do not account for the observed association between bone and coronary artery calcification among postmenopausal women using hormone therapy.