Beyond the Curve: The Muscle‐Specific Asymmetries of Adolescent Idiopathic Scoliosis

ABSTRACT Background Adolescent idiopathic scoliosis (AIS) describes an asymmetrical formation of the spine that develops rapidly during adolescence. It is well known that the forces applied to bones, such as from paraspinal muscles, are substantial moderators of growth and adaptation. Despite the intimate relationship between the paraspinal muscles and the spine, very little is known about these muscles, particularly during adolescence when there is the greatest risk of the development and progression of AIS. Here, we aimed to quantify paraspinal muscle volume, intramuscular fat, and fat‐free muscle asymmetry in female adolescents with AIS and a healthy adolescent control cohort. Methods Twenty‐nine female adolescents with primary‐right‐thoracic scoliosis (range: Cobb angle mean [SD]: 39 [15]°; age mean [SD]: 13.8 [1.5] years) and 19 age‐matched female control participants without scoliosis (age mean [SD]: 13.1 [1.8] years). Participants underwent T1‐weighted and coronal mDixon magnetic resonance image scans. An Asymmetry index of muscle volume, intramuscular‐fat, and fat‐free muscle were determined for five paraspinal muscles across 11 vertebral levels (T4‐L4/5) (Asymmetry index = Ln [right‐side/left‐side]). Results AIS participants have greater asymmetry in paraspinal muscle volume, intramuscular fat, and fat‐free muscle compared to controls ( p < 0.05, linear mixed effects analysis). Across numerous vertebral levels adjacent to the primary thoracic curve apex in AIS, multifidus volume and multifidus, longissimus, and spinalis intramuscular fat asymmetries were greater in AIS (13%–57% larger on the left side) than in the control group (1%–20%), p < 0.05. In the lumbar spine, multifidus volume and multifidus, longissimus, and psoas intramuscular fat were greater on the right side of the lumbar curve in AIS (18%–54%) than in controls (1%–14%), p < 0.05. Scoliosis curve severity was moderately correlated with asymmetries in muscle volume, intramuscular fat, and fat‐free muscle (range: R = 0.40–0.64, p < 0.05). Conclusions These findings provide evidence that asymmetries in paraspinal muscle size and composition exist along the length of the scoliotic spine. The asymmetries are associated with curve severity; therefore, supporting the need to further consider muscle in the pathogenesis of AIS.