Abstract
BACKGROUND: To find out, based on the available recent randomized controlled trials (RCTs), if the nonsurgical interventions commonly used for knee osteoarthritis patients are valid and quantify their efficiency. METHODS: The database of MEDLINE and EMBASE were searched for RCTs evaluating nonsurgical treatment strategies on patients with mild to moderate knee osteoarthritis. A Bayesian random-effects network meta-analysis was performed. The primary outcome was the mean change from baseline in the Western Ontario and McMaster university (WOMAC) total score at 12 months. Raw mean differences with 95% credibility intervals were calculated. Treatments were ranked by probabilities of each treatment to be the best. RESULTS: Thirteen trials assessed 7 strategies with WOMAC at 12 months: injection of platelet rich plasma (PRP), corticosteroids, mesenchymal stem cells (MSCs), hyaluronic acid, ozone, administration of nonsteroidal anti-inflammatory drugs with or without the association of physiotherapy. For treatment-specific effect size, a greater association with WOMAC decrease was found significantly for MSCs (mean difference, -28.0 [95% CrI, -32.9 to -22.4]) and PRP (mean difference, -19.9 [95% CrI, -24.1 to -15.8]). Rank probabilities among the treatments indicated that MSCs had a much higher probability (P = .91) of being the best treatment compared with other treatments, while PRP ranked as the second-best treatment (P = .89). CONCLUSION: In this systematic review and network meta-analysis, the outcomes of treatments using MSCs and PRP for the management of knee osteoarthritis were associated with long-term improvements in pain and function. More high quality RCTs would be needed to confirm the efficiency of MSCs and PRP for the treatment of patients with knee osteoarthritis.