A meta-analysis of bisphosphonates for periprosthetic bone loss after total joint arthroplasty.

BACKGROUND

Periprosthetic bone loss, which is common after joint arthroplasty, may cause bone loosening and lead to failed prosthetic fixation. Two previous meta-analyses have confirmed the mid-term effect of bisphosphonates (BPs) in preventing bone loss after arthroplasty. To determine long-term efficacy and gender bias of BPs after joint arthroplasty, we conducted a meta-analysis based on 17 RCTs involving 781 patients to evaluate the effect of BPs.

METHODS

Meta-analysis was conducted after a systematic search of Medline, Embase, the Cochrane Collaboration Central Register of Controlled Clinical Trials, CINAHL, and ISI Web of Science, and manual examination of references in selected articles and conference abstracts of key orthopedic journals. Methodological quality and abstracted relevant data were evaluated. In addition to analysis of bone mineral density (BMD), we also conducted systematic analysis of clinically relevant outcomes and bone biochemical markers.

RESULTS

Seventeen trials involving a total of 781 patients were assessed. Significantly less periprosthetic bone loss occurred in the BP-treated group than in the control group at 6 and 12 months (p < 0.0001). This protective effect was not noted at 3 months (p = 0.11) nor from 24-72 months (p = 0.14). The efficacy of BPs in the gender balance, shorter duration, and the non-nitrogenous BPs groups was no different from that for controls. Biochemical bone markers were suppressed in the BPs group. However, clinically relevant outcomes in the BPs group and controls were similar at all times.

CONCLUSIONS

The overall moderate-quality evidence from the RCTs confirmed the significant mid-term efficacy of BPs on periprosthetic bone loss after joint arthroplasty. Long-term efficacy of BPs was not observed, and the therapy was of more benefit to women, especially postmenopausal women. To achieve better efficacy, nitrogenous BPs and long duration of treatment may be recommended.