Introduction: Exercise-based approaches exist; however, it is unclear whether these approaches are based on scientific findings in the literature on trunk muscle deficits in scoliosis that could be targeted by exercises. The aims of this study were to systematically review the literature to understand the functional muscular properties of paraspinal muscles in AIS to determine: 1) differences in functional outcomes between patients with AIS and controls, 2) differences in functional outcomes between sides (concave and convex) in patients compared to controls 3) differences between concave and convex sides as well as levels in subjects with AIS, 4) differences in functional outcomes between different curve types. 5) Associations between functional outcomes and curve characteristics, and 6) associations between functional characteristics and progression. Design: Systematic review Methods: A search was conducted in EMBASE, MEDLINE, SPORTdiscus, CINAHL, SCOPUS, and Web of Science, for keywords describing functional properties of paraspinal muscles and measurement tools including: scoliosis, spinal deformity, spinal muscles, erector, rotatores, longissimus, spinalis, illiocostalis, force, strength, endurance, fatigability, and muscle fatigue. Two reviewers independently reviewed abstracts and then full-text articles to determine if they met selection criteria. Two reviewers used an extraction form to extract information and appraise the quality during the full-text review. Levels of evidence were determined for summarized results for each of the 6 objectives. Results: Our search yielded 316 unique records. Inter-reviewer agreement for abstract selection was Kappa = 0.73 and was 0.77 for full-text inclusion. Full-text review was done for 48 papers and 24 were included. A large amount of heterogeneity was observed in sample studied and assessment methodology. Quality appraisal revealed that no study met a minimum of 50 % of the relevant quality criteria. Studies recruited consistently low sample sizes and samples were largely heterogeneous. Limited evidence was noted supporting, a prolonged bilateral EMG activation during gait between AIS and controls; elevated heterolateral:homolateral activity ratios during side-bending; overall weakness in those with scoliosis compared to controls; no asymmetry in normalized muscle activity during submaximal isometric contractions; prolonged latencies on the side of the spine opposite of the curve and bilaterally in response to an unloading reflex; strength & muscle volume differences are most commonly pronounced in double curves; Axial rotation of the UEV is correlated with a high convex:concave activity ratio at the LEV; no correlation between latency and curve severity, but a correlation between latency and progression and higher convex:concave EMG ratios and progression, this is more pronounced in sitting positions. Conclusions: Evidence is limited on most of our six objectives due to low quality evidence and lack of research about muscle impairments in scoliosis. Current exerci...
