Comparative Effectiveness of Orthoses for Thumb Osteoarthritis: A Systematic Review and Network Meta-analysis

Objective: To evaluate the best option among orthoses for carpometacarpal (CMC) osteoarthritis (OA) of the thumb, using a network meta-analysis.

Data sources: Medline, Embase, Cochrane, and ClinicalTrials.gov registry databases were used. PubMed, Embase, Cochrane Controlled Trials Register, Cochrane, and other databases were used without language restrictions.

Study selection: We searched randomized controlled trials (RCTs) on adults with OA of the thumb by studying any orthosis from the beginning to March 10, 2020.

Data extraction: Data were extracted and checked for accuracy and completeness by pairs of reviewers. Outcomes were pain and function. Comparative treatment effects were analyzed by random-effects model for direct pairwise comparisons and Bayesian network meta-analyses to integrate direct and indirect evidence.

Data synthesis: Eleven RCTs involving 619 patients were included. We evaluated 5 groups, for 4 different orthoses: short thermoplastic CMC splint (rigid CMC) (n=5), long thermoplastic carpometacarpal-metacarpophalangeal splint (rigid CMC-MCP) (n=7), short neoprene CMC splint (soft CMC) (n=1), long neoprene CMC-MCP splint (soft CMC-MCP) (n=5), and one as a control group (n=5). Our results show that all splints were superior to placebo to reduce pain intensity and the top-ranked intervention was the rigid CMC-MCP (surface under the cumulative ranking curve analysis [SUCRA], score: 65.4). In function evaluation, we report a 71.6 SUCRA for rigid CMC.

Conclusions: Although the current evidence is unclear on the use of the splint in OA of the thumb, it is not known which orthosis is more effective and whether the orthosis is more effective than other interventions. The network meta-analysis shows that a long thermoplastic splint it is the best choice for pain relief and the short thermoplastic CMC splint is the best treatment to increase function. These results may suggest initial treatment with a long rigid orthosis and then a short rigid orthosis.