Abstract
BACKGROUND: The structure and function of human tissue, such as bone, muscle and tendon, can be improved with targeted exercise training. However, the effects of exercise training on intervertebral disc tissue remain unclear. OBJECTIVE: We aimed to examine the impact of physical loading exposure (exercise training, sport and physical activity) on intervertebral disc (IVD) health compared to non-physical loading (or lower volume of the same physical loading) controls. METHODS: We conducted a systematic review and meta-analysis. Seven electronic databases (PubMed, CINAHL, SPORTDiscus, EMBASE, CENTRAL, Web of Science and Scopus) and two trial registries (World Health Organization International Clinical Trials Registry Platform and National Institutes of Health) were searched from inception to 3 June, 2025. Forward and backward citation tracking was conducted for included reports. Certainty of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation criteria (GRADE). Risk of bias was assessed using Cochrane RoB2 and Johanna Briggs Institute critical appraisal checklists. We included randomised controlled trials (n = 2), cohort (n = 9) and cross-sectional studies (n = 28) of any physical loading compared to controls with continuous measurements of IVD health (e.g. T2-relaxation, IVD height) via magnetic resonance imaging and/or categorically graded measures of IVD degeneration (e.g. Pfirrmann grade). RESULTS: Forty-five reports of 39 studies (participants: 4152) were included. The pairwise random-effects meta-analysis estimated the standardised mean difference (Hedges' g) of continuous outcomes for combined physical loading and independent subgroups and odds ratios (Paule-Mandel estimator) of categorical outcomes for combined physical loading only. A meta-analysis revealed upright bipedal loading (mostly running; Hedges' g [95% confidence interval] 0.31 [0.12, 0.50]; P = 0.002; n = 7, GRADE: very low) was associated with better IVD health; however, no other subgroup of physical loading was associated with better or worse IVD health. Combined physical loading revealed greater odds of IVD degeneration via reduced signal intensity (odds ratio [95% confidence interval] 2.80 [1.53, 5.11], P = 0.001; n = 5, GRADE: low); however, no other measure of IVD degeneration was significant. CONCLUSIONS: Running was the only physical loading exposure associated with better IVD health. The mixed results from the combined physical loading analyses suggest that the type of physical loading plays a role in IVD health. As our estimates rely on observational data, prospective running interventions that examine the causal effect on IVD health appear warranted. CLINICAL TRIAL REGISTRATION: The review was prospectively registered with International Prospective Register of Systematic Reviews (PROSPERO) [CRD42022366391]. See Supplement P of the Electronic Supplementary Material for protocol deviations since registration.