To systematically evaluate the difference of therapeutic effects of different Glucocorticoid (GC) combined with or without sodium hyaluronate (HA) on knee osteoarthritis (KOA).

PubMed, Cochrane Library, Web of Science, Sciencedirect, China National Knowledge Infrastructure (CNKI), Wanfang , China Science and Technology Journal Database(VIP), and China Biology Medicine disc (CBM) were searched by computer, and the literatures about randomized controlled trial (RCT) of intra-articular injection of glucocorticoid in the treatment of knee arthritis were searched. The retrieval time ranged from establishment of the database to May 2023. The repetitive publication, lack of access to the full text and incomplete or incorrect research data were excluded, and two researchers screened the literature independently from the literature type, research object, intervention measures and outcome indicators.The data were extracted and the risk of bias was evaluated. Software ReviewManager5.4 and Stata17 were used for network meta-analysis.

A total of 2098 related literatures were searched, and 49 RCTs were finally included, including 13 drug regimens, specifically HA, compound betamethasone (CB), dexamethasone (DXM), methylprednisolone acetate (MPA), triamcinolone acetonide (TA), placebo (PLA) with water for injection or normal saline as the main component, HA+CB, HA+PLA, HA+DXM, HA+MPA, HA+TA, CB+PLA, TA+PLA. In the process of the treatment of KOA, network meta-analysis showed results as follows :(1)for improving the effective rate, the top three drug regimens were HA+CB ［compared with TA, odds ratio (OR)=22.13, 95% (confidence interval) CI (6.32, 77.47)］, HA+TA ［compared with TA, OR=20.39,95%CI (5.49, 75.73)］, HA+DXM ［compared with TA, OR=19.30,95%CI (4.77, 78.20)］; 95%CI does not contain 1, indicating that the difference is statistically significant.(2)For reducing the scores of Western Ontario and McMaster Universities osteoarthritis index (WOMAC), the top three drug regimens were DXM ［compared with CB, the mean difference (MD) =-22.46% CI (- 38.01, -6.90)］, HA+TA ［compared with CB, MD=-13.87,95%CI (- 27.38, -0.36)］, HA+CB ［compared with CB, MD=-11.03,95%CI (- 21.47, -0.60)］; 95%CI does not contain 0 to indicate that the difference is statistically significant.(3)In terms of reducing the score of visual analogue scale (VAS), the top three medication schemes were HA+TA［compared with MPA, MD=-4.65, 95%CI (-6.78, -2.51)］, CB+PLA［compared with MPA, MD=-2.60, 95%CI (-4.69, -0.51)］, HA+CB［compared with MPA, MD=-2.50, 95%CI (-4.06, -0.94)］.(4)In terms of reducing Lequesne index, the top three drug regimens were HA+TA［compared with PLA, MD=-6.72, 95%CI (-9.33, -4.12)］, DXM［compared with PLA, MD=-5.60, 95%CI (-8.80, -2.40)］, HA［compared with PLA, MD=-4.63,95%CI (-6.77,-2.49)］.

HA combined with GC is preferred in the effectiveness of intra-articular injection of glucocorticoid in the treatment of knee osteoarthritis, among which HA+TA has a higher curative effect in the overall evaluation of KOA. Limited by the quantity and quality of RCTs, this conclusion still needs further verification by more high-quality clinical RCTs.