Bisphosphonates and implants

The initial stability of joint replacements can be improved both by systemic and local bisphosphonates.

The initial fixation of joint prostheses appears to be crucial for long-term success (Kärrholm et al. 1994, Ryd et al. 1995). The surfaces of joint replacements are mostly adjacent to cancellous bone. There is often a gradual change in position of the prosthesis during the first weeks, which can be explained by microfracturing, until the prosthesis has settled into a position in which the load is evenly distributed to the bone. This is not the only explanation, however. By use of radiostereometry, we have shown that both systemically and locally applied bisphosphonates reduce this early migration (Hilding et al. 2000, Hilding and Aspenberg 2006, 2007). This means that osteoclasts are involved. The bisphosphonates may inhibit resorption of the bone next to the prosthesis, which can be nothing but necrotic and prone to resorption. Another possibility is that the response to the trauma gets a more positive balance between bone formation and resorption, as previously described for screws. The first study used a rather high dose of clodronate given systemically, starting 3 weeks before surgery (Hilding et al. 2000). The second used ibandronate applied locally to the cut bone surface immediately before cementation (Hilding and Aspenberg 2007).

A third knee study failed to show any effect of a bisphosphonate upon radiostereometric migration. In contrast to the other 2 studies, this study used an uncemented prosthesis and a conventional osteoporosis-based dose of alendronate taken orally (Hansson et al. 2009). The reason that it did not work may be that uncemented prostheses migrate much more than cemented ones. This probably represents settling due to unevenness of the cut bone surfaces. At first, only small cancellous areas have to take all the load, and they are likely to be compacted. Any effect upon resorption might be hidden in the variation associated with this larger migration. Another possibility might be that the dose of alendronate was too low.