Interpreting cortical bone adaptation and load history by quantifying osteon morphotypes in circularly polarized light images

Birefringence variations in circularly polarized light (CPL) images of thin plane-parallel sections of cortical bone can be used to quantify regional differences in predominant collagen fiber orientation (CFO). Using CPL images of equine third metacarpals (MC3s), R.B. Martin, V.A. Gibson, S.M. Stover, J.C. Gibeling, and L.V. Griffin. (40) described six secondary osteon variants (‘morphotypes’) and suggested that differences in their regional prevalence affect fatigue resistance and toughness. They devised a numerical osteon morphotype score (MTS) for quantifying regional differences in osteon morphotypes. We have observed that a modification of this score could significantly improve its use for interpreting load history. We hypothesized that our modified osteon MTS would more accurately reveal differences in osteon MTSs between opposing “tension” and “compression” cortices of diaphyses of habitually bent bones. This was tested using CPL images in transverse sections of calcanei from sheep, deer, and horses, and radii from sheep and horses. Equine MC3s and sheep tibiae were examined as controls because they experience comparatively greater load complexity that, because of increased prevalence of torsion/shear, would not require regional mechanical enhancements provided by different osteon morphotypes. Predominant CFO, which can reliably reflect adaptation for a regionally prevalent strain mode, was quantified as mean gray levels from birefringence of entire images (excluding pore spaces) in anterior, posterior, medial, and lateral cortices. Results showed that, in contrast to the original scoring scheme of Martin et al., the modified scheme revealed significant anterior/posterior differences in osteon MTSs in nearly all “tension/compression” bones (p < 0.0001), but not in equine MC3s (p = 0.30) and sheep tibiae (p = 0.35). Among habitually bent bones, sheep radii were the exception; relatively lower osteon populations and the birefringence of the primary bone contributed to this result. Correlations between osteon MTSs using the scoring scheme of Martin et al. with CFO data from all regions of each bone invariably demonstrated weak-to-moderate negative correlations. This contrasts with typically high positive correlations between modified osteon MTSs and regional CFO. These results show that the modified osteon MTS can be a strong correlate of predominant CFO and of the non-uniform strain distribution produced by habitual bending.