Are bisphosphonates efficacious in knee osteoarthritis? A meta-analysis of randomized controlled trials

To clarify the effects of bisphosphonates in knee osteoarthritis (OA) using an up-to-date meta-analysis of randomized controlled trials (RCTs).

The protocol is registered in PROSPERO (CRD42017073449). We searched MEDLINE, EMBASE, Google Scholar, Web of Science, and Cochrane Database from inception until August 2017. We included only RCTs comparing any bisphosphonates vs placebo in knee OA patients and reporting validated pain and function scales, radiographic progression, and adverse events (AEs) outcomes. We excluded studies using active comparators or concomitant medications besides non-steroidal anti-inflammatory drugs (NSAIDs) and acetaminophen. We calculated standardized mean differences (SMDs) to account for variation in outcome scales. Random effects meta-analyses were performed.

We included seven RCTs (3013 patients, 69% female); most patients (N = 2767) received oral risedronate. No pain or function outcomes, regardless of dose, route, time point or measuring instrument, revealed statistically significant results (end of trial pain SMD = -0.16 [95% confidence interval (CI): -0.34, 0.02]). Similarly, we found no statistically significant effect on radiographic progression (risk ratio = 0.98 [95% CI: 0.77, 1.26]). One small RCT in patients with bone marrow lesions (BMLs) suggested a reduction in BML size at 6 months. Bisphosphonates displayed good tolerability, with no statistically significant differences in AE outcomes vs placebo.

Contrary to prior reviews, our analysis showed that bisphosphonates neither provide symptomatic relief nor defer radiographic progression in knee OA. However, these agents may still be beneficial in certain subsets of patients who display high rates of subchondral bone turnover. Future studies should be directed at defining such OA subsets and investigating the effects of bisphosphonates in those patients.