OBJECTIVE: To determine which type of exercise is best for reducing pain and disability in adults with chronic low back pain (LBP). DESIGN: Systematic review with network meta-analysis (NMA) of randomised controlled trials (RCTs). LITERATURE SEARCH: Six electronic databases were systematically searched from inception to July 2021. STUDY SELECTION CRITERIA: RCTs testing the effects of exercise on reducing self-perceived pain or disability in adults (18 to 65 years) with chronic LBP. DATA SYNTHESIS: We followed the PRISMA-NMA statement when reporting our NMA. A frequentist NMA was conducted. The probability of each intervention being the most effective was conducted according to SUCRA values. RESULTS: We included 118 trials (9710 participants). There were 28 head-to-head comparisons, 7 indirect comparisons for pain, and 8 indirect comparisons for disability. Compared with control, all types of physical exercises were effective for improving pain and disability, except for stretching exercises (for reducing pain) and McKenzie method (for reducing disability). The most effective interventions for reducing pain were: Pilates, mind-body and core-based exercises. The most effective interventions for reducing disability were: Pilates, strength and core-based exercises. On SUCRA analysis, Pilates had the highest likelihood for reducing pain (93%) and disability (98%). CONCLUSION: Although most exercise interventions had benefits for managing pain and disability in chronic LBP, the most beneficial programmes were those that included: (1) at least 1 to 2 sessions/week of Pilates or strength exercises; (2) sessions of < 60 min of core-based, strength or mind-body exercises; and, (3) training programs from 3 to 9 weeks of Pilates and core-based exercises.

Full text (sometimes free) may be available at these link(s):      help

• DOI
• PubMed
• PDF locator
• publisher