Relationship between bone structural variables from bone biopsy and bone mineral density (BMD) in patients on long term bisphosphonate (BP) Therapy

Document Type

Conference Proceeding

Publication Date


Publication Title

Journal of Bone and Mineral Research


BMD is a 2D measurement of a 3D structure that has been the cornerstone to diagnose osteoporosis, monitor progression, and response to therapy. Bone volume in biopsy is also a 3D estimate of a 2D measurement. Intuitively it is expected to correlate with each other. A few studies showed significant correlations between bone biopsy and BMD variables in a group of patients with various metabolic bone diseases. However, the strengths of correlations were modest or not significant in a more homogenous subset. Our aim was to understand if there were correlations between the bone structural variables and BMD in a group of post-menopausal women on long-term BP therapy. Methods: Trans-Iliac bone biopsies were obtained from 25 postmenopausal women (mean age 66.8 ± 6.6y) treated with BP for 6.1 ± 4.3 y. Using standard methods cortical bone volume, mean cortical thickness, trabecular bone volume, and trabecular thickness were measured. BMD of both proximal femurs (femoral neck, total hip, and trochanter) was measured by DEXA about 1-2 months before biopsy. Correlations among proximal femur measures and among biopsy variables, as well as between BMD and biopsy structural variables were performed using SigmaPlot. Results: Mean right and left femoral neck BMDs were 0.636 ± 0.068 and 0.637 ± 0.072 with highly significant correlation between the sides (r=0.77; P<0.0001). As expected, there were statistically significant correlations among various BMD parameters (r=0.46 to 0.87; p=0.02 to <.0001) with much stronger relationship between the sides (left Vs. right; r=0.62 to 0.84; p<0.001 for all). In contrast, bone structural measurements showed variable correlations ranging from strongest correlation between trabecular thickness and volume (r=0.83; p<0.001) to lack of correlation between cortical and trabecular bone volumes. There was a trend for a relationship between cortical thickness and trabecular bone volume (r=0.36; p=0.06). Finally, there were very few, if any, correlations between BMD and bone structural measurements. Conclusion: We confirmed strong correlation between the right and left femoral neck, trochanter, and total hip BMDs. However, weak to poor correlation between BMD and bone structural measurements implies that the two methods assess different aspects of bone biology. Alternatively, BPs may have differential effects on BMD and bone structure. Further research is needed to elucidate underlying pathophysiologic mechanism for these observations.




Supplement 1

This document is currently not available here.