Network meta-analysis of the effect of exercise therapy on chronic neck pain

The purpose of this paper is to evaluate the effect of different exercise therapy on patients with chronic neck pain. By searching PubMed, EMbase, The Cochrane Library, Web of Science, CNKI, WanFang Data and other databases, we screened the literatures that met the inclusion criteria until June 2022. RCT bias risk assessment tool recommended in Cochrane manual was used to evaluate the literature quality, and Stata16.0 software was used for mesh meta-analysis. A total of 76 studies, 12 exercise therapies and 5219 participants were included. In terms of pain, Pilates [MD ​= ​−2.71,95%CI (−3.87, −1.55), P ​= ​0.000] had the most significant effect. Second training for the stability of [MD ​= ​-2.39, 95 ​% CI (−3.14, −1.64), P ​= ​0.000) and VR [MD ​= ​−2.35, 95 ​% CI (−3.6, −1.11), P ​= ​0.000). In terms of functional indexes, respiratory training [MD ​= ​−16.31,95%CI (−28.52, −4.1), P ​= ​0.009] had the most significant effect. Secondly for muscle energy technology [MD ​= ​−7.47, 95 ​% CI (−12.37, −2.58), P ​= ​0.003) and aerobic training [MD ​= ​−7.36, 95 ​% CI (−12.56, −2.16), P ​= ​0.006]. Current evidence shows that Pilates, stability training and virtual reality training are the most effective in pain intervention, muscle energy technology and comprehensive exercise training are outstanding in helping functional recovery, suggesting that appropriate exercise therapy should be implemented according to the symptoms of patients to achieve the best treatment effect.