The best exercise modality and dose for reducing pain in adults with low back pain: A systematic review with model-based Bayesian network meta-analysis

OBJECTIVE: To quantify the dose-response relationship between overall and specific exercise modalities and pain, in patients with non-specific chronic low back pain (LBP). DESIGN: Systematic Review with Bayesian Network Meta-Analysis (NMA). LITERATURE SEARCH: We searched the Medline, Embase, Web of Science, Cochrane Library, Scopus, and SPORTDiscus databases from inception to June 2023. STUDY SELECTION CRITERIA: We included randomized controlled trials (RCTs) of exercise interventions in adults with non-specific chronic LBP and at least one pain outcome reported at the main trial endpoint. DATA SYNTHESIS: A random effects NMA was conducted. We assessed risk of bias using the Cochrane Risk of Bias Tool 2.0, and used the GRADE approach to judge the certainty of evidence for each outcome. RESULTS: Eighty-two trials were included (n=5,033 participants). We found a nonlinear dose-response relationship between total exercise and pain in patients with non-specific chronic LBP. The maximum significant response was observed at 920 METs-min (standardized mean difference = -1.74; 95% credible intervals: [-2.43, -1.04]). The minimum clinically-important difference (MCID) for achieving meaningful pain improvement was 520 METs-min per week. The dose to achieve MCID varied by type of exercise; Pilates was the most effective. The certainty of the evidence was very low to moderate for all outcomes. CONCLUSION: The dose-response relationship of different exercise modalities to improve pain in patients with non-specific chronic LBP had a U-shaped trajectory and low to moderate certainty evidence. The clinical effect was most pronounced with Pilates exercise.