Clinical therapeutic efficacy of mesenchymal stem cells derived from adipose or bone marrow for knee osteoarthritis: a meta-analysis of randomized controlled trials

Aim: This meta-analysis, only including randomized controlled trials (RCTs), was conducted to assess separately and compare the therapeutic efficacy of adipose-derived mesenchymal stem cells (ADMSCs) and bone marrow-derived mesenchymal stem cells (BMSCs) for knee osteoarthritis (OA) at the same follow-up time. Methods: Potential relevant researches were identified from PubMed, Web of Science, Embase, Cochrane Library and clinicaltrials.gov. The data, from clinical trials concentrating on knee OA treated with ADMSCs or BMSCs, were extracted and pooled for meta-analysis to compare the clinical outcomes of patients with knee OA in visual analog scale (VAS), Western Ontario McMaster Universities Osteoarthritis Index (WOMAC), Lysholm knee scale (Lysholm) and Tegner activity scale (Tegner). Results: Nine randomized controlled trials including a total of 377 patients met the inclusion criteria. This meta-analysis obtained the following results. First, the improvement of VAS scores was statistically significant after BMSCs treatment at 6-, 12- and 24-month follow-up compared with control groups (p < 0.01). In contrast, the improvement of WOMAC scores was of no statistical significance, but showed a positive trend with the prolongation of the follow-up time (6 months: mean difference [MD] = 6.51; 95% CI: -2.38 to 15.40; p = 0.15; 12 months: MD = -6.81; 95% CI: -13.94 to 0.33; p = 0.06). Lysholm scores presented a similar pattern (12 months: MD = 1.93; 95% CI: -11.52 to 15.38; p = 0.78; 24 months: MD = 8.94; 95% CI: 1.45 to 16.43; p = 0.02). Second, VAS and WOMAC scores of patients after ADMSCs treatment were significantly improved at any follow-up time (p 0.05). The improvement of Lysholm scores was of no statistical significance compared with control groups, although treatment outcome at 12-month follow-up was better than that at 24-month follow-up, which was debatable because only data of one clinical trial were pooled in the analysis (12 months: MD = 7.50; 95% CI: -1.94 to 16.94; p = 0.12; 24 months: MD = 5.10; 95% CI: -3.02 to 13.22; p = 0.22). Finally, by comparing the statistical results of VAS and WOMAC scores, it could be concluded that the therapeutic effect of ADMSCs on knee OA was more effective than that of BMSCs. Conclusion: This meta-analysis showed that regeneration with BMSCs or ADMSCs had a great application potential in the treatment of patients with knee OA, and ADMSCs tended to be superior to BMSCs according to the limited clinical evidences available.