Abstract
BACKGROUND AND OBJECTIVE: Guidelines recommend combining physical and psychological programmes for chronic low back pain (CLBP); however, exercise therapy (ET) and manual therapy (MT) are often delivered separately. This systematic review with meta-analysis and meta-regression of randomised controlled trials (RCTs) aimed to compare the efficacy of ET with MT in terms of pain intensity, disability and physical function in people with CLBP. DATABASES AND DATA TREATMENT: MEDLINE, Web of Science, PEDro, Cochrane Library and Scopus were searched July-August 2024 for RCTs comparing ET with MT in participants aged 18-54 years. Outcomes were extracted for the short-, medium- and long-term follow-up periods. Risk of bias (RoB 2.0 Cochrane Tool) and certainty of evidence (GRADE) were appraised. RESULTS: Six RCTs (743 patients) were included. Meta-analyses showed, albeit non-clinically relevant, a significant difference for long-term in favour of ET for disability (SMD = -0.25, 95% CI [-0.43, -0.07], p = 0.007). Meta-regression showed that the female-male ratio, treatment duration and mean age explain variability in pain intensity and disability. CONCLUSIONS: ET had a small beneficial effect on long-term disability in people with CLBP. Nevertheless, evidence does not provide conclusive differences between both the treatments overall, influenced by heterogeneity and the number of studies included. Biopsychosocial factors may moderate the differences in outcomes. The GRADE assessment revealed very low certainty across all outcomes, highlighting the lack of high-quality research. SIGNIFICANCE STATEMENT: ET may offer small long-term benefits over MT for disability in people with CLBP. Differences seem to be influenced by sex, age and treatment duration. The choice of ET over MT, or vice versa, as a stand-alone treatment does not appear to be supported by current evidence. PROSPERO REGISTRATION NUMBER: CRD42024569120.