Effects of bisphosphonates on fracture incidence and bone metabolism in rheumatoid arthritis patients in general practice taking long-term corticosteroid therapy: a retrospective study.

Background And Objective: There is a risk that disturbances of activities of daily living (ADL) due to rheumatoid arthritis (RA) are increased by the occurrence of fractures, including vertebral compression fractures and femoral neck fractures, in RA patients receiving oral corticosteroid therapy. Bisphosphonates are most commonly used in the treatment of postmenopausal osteoporosis. In a large-scale, randomized, double-blind, placebo-controlled study that was performed to assess the prophylactic efficacy of bisphosphonates, alendronic acid decreased the incidence of vertebral fractures by approximately 50% compared with placebo in postmenopausal patients. A similar result has also been reported with risedronic acid. The present long-term retrospective study evaluated the effects of alendronic acid and risedronic acid therapy on development of new vertebral/non-vertebral fractures in RA patients receiving long-term oral prednisolone therapy at an average dose of 5 mg/day.

Methods: The subjects were 138 general practice patients aged 50-79 years with RA (alendronic acid group 80; risedronic acid group 58) who received oral prednisolone at a dose of 2-15 mg/day for at least 1 year combined with bisphosphonate therapy (alendronic acid 5 mg/day or risedronic acid 2.5 mg/day) for at least 10 months. Patients with five or more vertebral fractures at the start of bisphosphonate therapy were excluded from the study. Vertebral fractures were detected by obtaining plain x-ray films of the thoracic and lumbar spines at the start of bisphosphonate therapy and on completion of follow-up. We measured the incidence of new fractures, the speed of sound (SOS) at the calcaneus as measured by quantitative ultrasound, and levels of crosslinked N-telopeptide of type I collagen (NTX), a marker of bone resorption. The percentage change at each measuring point was tested using the paired t-test. The incidence of new fractures was compared between groups using the Cox proportional hazard model.

Results: The incidence of new vertebral fractures was 6.3% in the alendronic acid group and 13.8% in the risedronic acid group; the incidence of new non-vertebral fractures was 6.3% and 12.1%, respectively. The incidence of any fracture was significantly higher and severe fractures tended to be more common in the risedronic acid group. Analysis by the Cox proportional hazard model revealed a significant difference between the two groups with respect to the cumulative incidence of new fractures (p = 0.0386). The SOS of the calcaneus showed no appreciable difference between the two groups. NTX measurements indicated that antiresorptive activity was maintained from 6 months of treatment onwards in the alendronic acid group but not in the risedronic acid group.

Conclusion: These findings suggest that alendronic acid has a stronger prophylactic effect against fractures than risedronic acid in RA general practice patients taking long-term corticosteroid therapy.