The use of quantitative ultrasound to diagnose osteoporosis in patients with rheumatoid arthritis

Cryer, J. R., Otter, S. J. and Bowen, Catherine Jane (2004) The use of quantitative ultrasound to diagnose osteoporosis in patients with rheumatoid arthritis. Rheumatology, 43 (Supplement 2), ii151 - ii152.

Abstract

Background: Patients with rheumatoid arthritis (RA) are recognised as being at risk of osteoporosis both as part of the disease process and due to the drugs used to treat it.1,2 This study aimed to consider the use of calcaneal scanning using quantitative ultrasound - Contact Ultrasound Bone Analysis (CUBA) to demonstrate osteoporosis in patients with RA.
Methods: Forty-six patients (35 female,11 male) with established RA had a DEXA scan of the non-dominant wrist and a CUBA scan of the non-dominant heel. Sensitivity, specificity, positive and negative predictive values were used to determine the correlation between osteoporosis as determined by the CUBA heel scan compared to a DEXA wrist scan.
Results: The CUBA heel scan revealed a sensitivity of 90% with a specificity of 44% for osteoporosis compared to a DEXA scan of the wrist. The positive predictive value of the CUBA was 31% and the negative predictive value was 94%.
If a normal bone density is achieved with the CUBA there is a 94% certainty this is correct. However, if osteoporosis is demonstrated using CUBA there is only a 31% certainty this is correct. In such instances a secondary scan utilising a different methodology e.g. DEXA would needed. Future work should consider the effect of minor alterations to the equipment and/or scanning protocol in order to yield greater accuracy.
Conclusions: The clinical implications of this work are that the CUBA unit could be used as a primary screening device. Given some of the cost and accessibility issues associated with DEXA scans, quantitative ultrasound may have a role in screening for osteoporosis in the Primary Care setting.

This record has no associated files available for download.

Contributors

Download statistics