Bone Abstracts

Background: DXA scan has for many years been gold standard when analyzing areal bone mineral density (aBMD) in osteoporotic patients. Risk of fractures increases as aBMD declines. However, although risk of fracture is highest in patients with osteoporosis, it is well known that the majority of fragility fractures actually occur in patients with osteopenia rather than osteoporosis. As aBMD is a projected two dimensional imaging technique it does not differentiate between cortical and trabecular bone. 3D measures of bone by quantitative computed tomography (QCT) is considered to measure true volumetric BMD (vBMD, mg/cm³). Furthermore, this technique can distinguish between cortical and trabecular bone. Therefore, we looked at correlation between aBMD by DXA and vBMD by QCT measured at the hip.

Patient and method: In a cross-sectional study, 125 postmenopausal women mean age 63 (ranges 56–82) were scanned by DXA and QCT at the hip.

Results: As expected linear regression of total hip measured by DXA and QCT showed a positive correlation between aBMD and vBMD at trabecular and integrated site varying between 0.63 and 0.75 (P<0.01). However, linear regression of aBMD and cortical vBMD showed a significant negative correlation (β=−0.56, P<0.01). The inverse relation persisted even after adjusting for confounders such as BMI.

Conclusion: Our results indicate that 2D imaging by DXA provides a true measure of trabecular bone density, whereas the inverse correlation between aBMD and cortical vBMD suggest that cortical bone strength may be overestimate by aBMD. As most fractures occur in peripheral bones which are mainly composed of cortical bone, these patients may actually have week cortical bone despite not being osteoporotic as assessed by DXA.