Areal bone mineral density (BMD) by DXA, although an important index, does not accurately assess risk of fragility fracture. Another bone structural parameter, the orientation of type I collagen, is known to add to risk determination, independently of BMD. Accordingly, we investigated the Haversian systems of transiliac crest biopsies from non-osteoporotic women with low-trauma fractures, matched to healthy women without fracture by age and BMD. We employed circularly polarized light (CPL) microscopy because 1) each of the extinct and bright birefringent signals of CPL corresponds to a specific collagen arrangement; and 2) CPL can employ magnification suitable to provide data, of manageable size, from the whole cortical component of a section of biopsy. Under CPL, the coaxial layers of osteons, called lamellae, appear either birefringent extinct or bright. On a section transverse to the Haversian system, the extinct lamella comprises mainly collagen forming small angles, and the bright lamella comprises mainly collagen forming large angles, relative to the general orientation of the Haversian system. We performed semi-automatic morphometry for birefringent and structural parameters for which we computed intra- and inter-observer errors. The statistical analysis used a linear mixed model to compare fracturing and non-fracturing groups while addressing pairing of fracturing and non-fracturing subjects, and linear regression to assess differences between matched subjects. We found significant reduction in 1) lamellar width and area for extinct lamella and bright lamella; 2) percentage of extinct birefringence in osteons, and 3) single osteon area; in the fracturing group; and in lamellar width in the fracturing subject of all pairs. Our results evidence the need to investigate, in a larger sample of subjects, the distribution of collagen orientation as a parameter diagnostic of increased fracture risk.
All Science Journal Classification (ASJC) codes
- Endocrinology, Diabetes and Metabolism