Purpose: Chronic low back pain (CLBP) is an aging and public health issue that is a leading cause of disability worldwide and has a significant economic impact on a global scale. Treatments for CLBP are varied, and there is currently no study with high-quality evidence to show which treatment works best. Exercise therapy has the characteristics of minor harm, low cost, and convenient implementation. It has become a mainstream treatment method in clinics for chronic low back pain. However, there is insufficient evidence on which specific exercise regimen is more effective for chronic non-specific low back pain. This network meta-analysis aimed to evaluate the effects of different exercise therapies on chronic low back pain and provide a reference for exercise regimens in CLBP patients.

Methods: We searched PubMed, EMBASE, Cochrane Library, and Web of Science from inception to 10 May 2022. Inclusion and exclusion criteria were used for selection. We collected information from studies to compare the effects of 20 exercise interventions on patients with chronic low back pain.

Results: This study included 75 randomized controlled trials (RCTs) with 5,254 participants. Network meta-analysis results showed that tai chi [standardized mean difference (SMD), -2.11; 95% CI, -3.62 to -0.61], yoga (SMD, -1.76; 95% CI -2.72 to -0.81), Pilates exercise (SMD, -1.52; 95% CI, -2.68, to -0.36), and sling exercise (SMD, -1.19; 95% CI, -2.07 to -0.30) showed a better pain improvement than conventional rehabilitation. Tai chi (SMD, -2.42; 95% CI, -3.81 to -1.03) and yoga (SMD, -2.07; 95% CI, -2.80 to -1.34) showed a better pain improvement than no intervention provided. Yoga (SMD, -1.72; 95% CI, -2.91 to -0.53) and core or stabilization exercises (SMD, -1.04; 95% CI, -1.80 to -0.28) showed a better physical function improvement than conventional rehabilitation. Yoga (SMD, -1.81; 95% CI, -2.78 to -0.83) and core or stabilization exercises (SMD, -1.13; 95% CI, -1.66 to -0.59) showed a better physical function improvement than no intervention provided.

Conclusion: Compared with conventional rehabilitation and no intervention provided, tai chi, toga, Pilates exercise, sling exercise, motor control exercise, and core or stabilization exercises significantly improved CLBP in patients. Compared with conventional rehabilitation and no intervention provided, yoga and core or stabilization exercises were statistically significant in improving physical function in patients with CLBP. Due to the limitations of the quality and quantity of the included studies, it is difficult to make a definitive recommendation before more large-scale and high-quality RCTs are conducted.