Osteoporosis in rheumatoid arthritis: findings in the metacarpal, spine, and hip and a study of the determinants of both localized and generalized osteopenia.
OBJECTIVE: To determine the relationship between axial and peripheral bone density in patients with rheumatoid arthritis (RA) compared to controls. To study risk factors for axial and peripheral bone loss in RA. METHODS: Twenty women with RA and 20 age matched controls underwent dual energy x-ray absorptiometry (DXA) of the spine and hip and microradioscopy of the hands in order to quantify the combined cortical thickness (CCT) of the 2nd metacarpal. Sharp's method was used to assess the degree of joint space narrowing and erosions. RESULTS: Compared to controls, the RA group had significantly decreased bone density at the hip (0.761 g/cm2 vs 0.852 g/cm2, p < 0.05) and at the 2nd metacarpal (3.77 vs 4.39 mm, p < 0.05), but not at the spine (0.960 vs 0.947 g/cm2, NS). In patients with RA, measurement of the CCT did not predict bone density at the spine or hip (r = 0.07, NS, and r = 0.095, NS, respectively), whereas in controls, the CCT was significantly correlated with bone density at both these axial sites (r = 0.47, p < 0.05, and r = 0.64, p < 0.01, respectively). Grip strength in RA was correlated with the CCT (r = 0.62, p < 0.01), implying a relationship between hand function and local bone density. Sharp's method did not correlate with axial bone density, but it was strongly correlated with the CCT (r = -0.64, p < 0.01). CONCLUSION: In RA, the CCT is not a useful screen for osteoporosis at the spine or hip. This suggests that localized and generalized osteopenia in RA are best viewed as independent processes.