King's College London

Research portal

Adjustment of DXA BMD measurements for anthropometric factors and its impact on the diagnosis of osteoporosis

Research output: Contribution to journalArticlepeer-review

Ling Wang, Limei Ran, Xiaojuan Zha, Kaiping Zhao, Yingying Yang, Qing Shuang, Yandong Liu, Karen Hind, Xiaoguang Cheng, Glen M. Blake

Original languageEnglish
Article number155
JournalArchives of osteoporosis
Issue number1
Published1 Dec 2020

King's Authors


Summary: We compared the effect of anthropometric factors on osteoporosis diagnosis by quantitative computed tomography (QCT) and dual energy X-ray absorptiometry (DXA) and found QCT spine volumetric bone mineral density (vBMD) was not associated with body weight, body mass index (BMI) or DXA anteroposterior spine thickness. In contrast, DXA spine and hip areal bone mineral density (aBMD) were strongly associated with all three factors. Adjustment of DXA aBMD measurements improved consistency with QCT vBMD. Purpose: Although the diagnosis of osteoporosis using DXA T scores preferentially targets patients with BMI, there is evidence that obesity is not protective against fractures. The aim of this study was to compare the effect of anthropometric factors on osteoporosis diagnosis by QCT and DXA and investigate whether adjustment of DXA aBMD can achieve a more even distribution of diagnoses between slimmer and heavier individuals consistent with QCT. Methods: The participants were 964 men and 682 women referred for low-dose chest CT and DXA examinations as part of their employers’ health check-up programs. QCT vBMD was measured in the L1–L2 vertebrae and DXA aBMD in the spine and hip. The prevalence of osteoporosis in each tertile of BMI in participants aged > 50 years was evaluated based on their QCT and DXA findings, and then re-evaluated after adjustment to the mean BMI in each sex. Similar investigations were performed for body weight and DXA anteroposterior (AP) spine thickness. The effect of the adjustment of DXA aBMD for anthropometric factors on the correlation with QCT vBMD was also examined. Results: For spine QCT, correlations of age adjusted vBMD residuals against BMI were not statistically significant in men (P = 0.44) or women (P = 0.32). In contrast, slopes for aBMD residuals were all highly statistically significant (P < 0.001). There were similar findings for weight and AP spine thickness. Adjustment of DXA aBMD for anthropometric factors resulted in a more equal spread of diagnoses of osteoporosis and greater consistency with QCT. Conclusion: Our study highlights differences between DXA and QCT in their correlation with anthropometric factors and its effect on the diagnosis of osteoporosis. Adjustment of DXA T scores for anthropometric factors gave greater consistency with QCT vBMD. Further studies are required into whether adjusting DXA aBMD for anthropometric factors has a beneficial impact on the discriminative or predictive power for vertebral fracture.

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454