Influence of arthritis and non-arthritis related factors on areal bone mineral density (BMD(a)) in women with longstanding inflammatory polyarthritis: a primary care based inception cohort

BACKGROUND: The aim of this analysis was to determine the relative influence of disease and non-disease factors on areal bone mineral density (BMD(a)) in a primary care based cohort of women with inflammatory polyarthritis. METHODS: Women aged 16 years and over with recent onset inflammatory polyarthritis were recruited to the Norfolk Arthritis Register (NOAR) between 1990 and 1993. Subjects were examined at both baseline and follow up for the presence of tender, swollen and deformed joints. At the 10(th )anniversary visit, a sub-sample of women were invited to complete a bone health questionnaire and attend for BMD(a )(Hologic, QDR 4000). Linear regression was used to examine the association between BMD(a )with both (i) arthritis-related factors assessed at baseline and the 10(th )anniversary visit and (ii) standard risk factors for osteoporosis. Adjustments were made for age. RESULTS: 108 women, mean age 58.0 years were studied. Older age, decreasing weight and BMI at follow up were all associated with lower BMD(a )at both the spine and femoral neck. None of the lifestyle factors were linked. Indices of joint damage including 10(th )anniversary deformed joint count and erosive joint count were the arthritis-related variables linked with a reduction in BMD(a )at the femoral neck. By contrast, disease activity as determined by the number of tender and or swollen joints assessed both at baseline and follow up was not linked with BMD(a )at either site. CONCLUSION: Cumulative disease damage was the strongest predictor of reduced femoral bone density. Other disease and lifestyle factors have only a modest influence.