BoneKEy-Osteovision | Commentary
Prevention of hip fractures with risedronate
DOI:10.1138/2001015
The recent publication of the risedronate hip fracture study by McClung et al () represents a further landmark in the development of the therapeutics of osteoporosis. This is a randomized controlled trial comparing the effects of two doses of risedronate with placebo in the prevention of hip fractures in older postmenopausal women. Almost 10,000 women were enrolled and the study had a duration of 3 years. In the overall analysis, both risedronate 2.5 mg and 5 mg per day reduced hip fracture risk by approximately 30%. Thus, risedronate joins alendronate () and calcium-calciferol () in the small group of agents which have been demonstrated to decrease the incidence of hip fractures. Together with the observational data which suggest a similar action for long term use of hormone replacement therapy, we now have a number of proven strategies for dealing with this most devastating of osteoporotic fractures. It should be noted that all women in the McClung study received a daily 1 g calcium supplement, together with calciferol if their baseline serum 25-hydroxyvitamin D level was low, so the benefit of the bisphosphonate is in addition to the calcium-vitamin D benefit previously demonstrated by Chapuy (). The intervention was well tolerated with no difference in adverse events between the risedronate and placebo groups, strengthening the position of the potent bisphosphonates as a first-line intervention for the management of osteoporosis.
Sub-analyses of the data provide important insights into the therapeutic limitations of bone-active drugs in fracture prevention. There were two broad sets of entry criteria to the McClung study. Slightly more than half of the study subjects were aged 70-79 years and had moderately severe osteoporosis, as judged by femoral neck T-score. The balance of the cohort were aged greater than 80 years and were selected predominantly on the basis of clinical risk factors for hip fracture. Bone density was measured in about one third of this older group - the majority were found to be osteoporotic and as expected, had a relatively high incidence of hip fracture (9.7% over 3 years in those randomized to placebo). However the balance of this older group who did not have bone density measurements had a much lower fracture rate (3.6% in the placebo group), suggesting that most were not osteoporotic and/or that the clinical risk factors used in recruitment did not have the clinical utility in identifying fracture risk that has been suggested by other studies.
The anti-fracture efficacy of risedronate in the younger osteoporotic cohort was quite different from that in the older predominantly non-osteoporotic group. In the younger group, fracture risk was reduced by 40% and in those with both low bone density and a pre-existing vertebral fracture, fracture risk was reduced by 60%. In contrast, in the older group, the relative risk of fracture was 0.8 (P= 0.35). There are parallels between these results and those of the FIT study, in which hip and other fractures were reduced in those with low bone density and pre-existing vertebral fractures, but in whom anti-fracture efficacy could not be demonstrated in patients with proximal femoral bone densities within the young normal range (). At one level, these results could be said to be unsurprising, in that treatment for osteoporosis in patients who do not have the condition does not produce a detectable beneficial effect on fracture incidence. However, fracture risk is continuously related to bone density so it could also be argued that some benefit should accrue from treatment of individuals in the ‘osteopenic’ range. Neither the present study nor the FIT study really rule out this possibility since much larger studies would be necessary to assess anti-fracture efficacy in the subjects whose fracture risk is relatively low. However, it is reasonable to conclude that, at present, the use of potent bisphosphonates is not cost-effective in non-osteoporotic subjects.
These results contrast with those of two other studies of hip fracture prevention, in which efficacy does not appear to be dependent on the presence of low bone density. The Chapuy study (), using calcium and calciferol, did not stratify patients according to bone density or fracture risk yet still produced a significant reduction in hip fracture. Of course, it is unknown whether partitioning subjects in this study into those with and without osteoporosis might have led to differences in anti-fracture efficacy being demonstrated, similar to those found in the McClung study. However, it does raise the possibility that calcium and vitamin D supplements might prevent hip fractures by mechanisms other than just improving bone density. The deleterious effect of vitamin D deficiency on muscle function raises the possibility that calcium-calciferol supplementation might partly act by preventing falls. The second study is that of Kannus et al () who have recently shown that the use of hip protectors produces substantial reductions in hip fracture incidence. These devices attenuate the energy imparted to the greater trochanter during a fall and, unlike bisphosphonates, do not act via effects on bone density. Other non-pharmacological interventions may also be important in falls prevention and thus in decreasing fracture numbers. The PROFET study has shown that simple attention to household safety, optimal correction of vision and rationalisation of medications, can reduce the number of recurrent falls by more than 50% ().
The key message to emerge from the literature reviewed above is that we should no longer be nihilistic regarding the prevention of hip fractures. We have powerful and effective pharmacological tools and clear evidence that falls prevention strategies and hip protectors are of value. This gives all those working in the osteoporosis field a responsibility to communicate these new therapeutic possibilities to our colleagues, our patients and to the wider community, such that these strategies can be selectively used to maximize their impact on hip fracture incidence.
This work is licensed under a
Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.