Asymmetry in the Onset of Paraspinal Muscles Activity Differs in Adolescents With Idiopathic Scoliosis Compared With Those With a Symmetrical Spine

Abstract

Background: Adolescent idiopathic scoliosis (AIS) is characterized by an asymmetrical formation of the spine and ribcage. Recent work provides evidence of asymmetrical (right versus left side) paraspinal muscle size, composition, and activation amplitude in adolescents with AIS. Each of these factors influences muscle force generation. The timing of paraspinal muscle activation may also contribute to an asymmetry in the timing of forces applied to the spine.

Questions/purposes: The main objectives were to determine (1) whether the timing and asymmetry of erector spinae muscle activation during a rapid bilateral arm raise task differs between adolescents with AIS and those without AIS and (2) whether the magnitude of erector spinae activation asymmetry in AIS is associated with scoliosis curve severity (Cobb angle) or skeletal development level (Risser stage). Finally, (3) we investigated potential kinematic confounders to determine whether symmetry of bilateral rapid arm movements differed between those with and without AIS, and whether any asymmetry in arm movement was associated with erector spinae activation asymmetry.

Methods: All patients were made aware of the project through flyers at one outpatient spine clinic and a scoliosis rehabilitation clinic in Brisbane, Australia. They were invited between August 2022 and September 2023 to contribute if they met the selection criteria. This cross-sectional study included females with AIS who agreed to participate (n = 24, mean ± SD age of 14 ± 2 years). They all had a primary right-thoracic curve, diagnosed by an orthopaedic specialist. Twenty age- and sex-matched controls (age 13 ± 2 years) who did not have AIS were recruited from the local community. Volunteers (from either group) were excluded if they had any history of spinal surgery, neurological disorders, or musculoskeletal disorders (other than AIS). The experimental task required participants to perform a bilateral rapid arm flexion in response to a visual cue. Muscle activation was recorded using surface electrodes, placed bilaterally on the anterior deltoid and erector spinae adjacent to the C7, T9 (the curve apex for AIS), T12, and L5 vertebrae. Muscle activation onsets were determined from 6 of 10 trials with the quickest deltoid onset for each participant. A linear mixed model (with fixed factors) was used to determine whether activation asymmetry (left-right onset difference) differed between groups (AIS, control) and vertebral level (C7, T9/apex, T12, and L5). Where a group difference in onset asymmetry was identified, the relation of the Cobb angle and Risser stage with the magnitude of asymmetry was evaluated in the AIS cohort using a linear mixed model. Task kinematics, including peak angular arm movement velocity and deltoid onset relative to the light signal, were analyzed using a linear mixed model with group and side as fixed factors.

Results: Erector spinae activation timing asymmetry differed between groups at the T9/apex (mean difference 14 ± 23 ms; p < 0.01). In the AIS group, muscle activation was 6 ± 17 ms earlier on the right (convex) relative to the left side of the spine, whereas in controls, activation was 8 ± 19 ms earlier on the left relative to the right side. This difference in activation timing asymmetry between groups was explained by later activation of the T9 level erector spinae muscles on the left (concave) side of the spine in AIS compared with controls (mean group difference of left T9/apex erector spinae onset 13 ± 26 ms; p = 0.01). There were no between-group differences at other vertebral levels. Within the AIS group, no association was observed between the magnitude of the erector spinae activation asymmetry measured at T9/apex and Cobb angle or Risser stage. There were no differences between groups in either the bilateral deltoid onset relative to light or arm peak velocity.

Conclusion: Erector spinae muscle activation is asymmetrical at the T9/apex vertebral level during a rapid bilateral arm raise task. This asymmetry was opposite between the AIS and control cohorts, with left-side activation delayed in AIS.

Clinical relevance: It is well established in conditions such as cerebral palsy that muscles forces can influence bone development in children. In children with AIS, there is growing evidence of asymmetrical paraspinal muscle size, composition, and activation amplitude. Each of these factors contribute to paraspinal muscle force generation. Our findings add to what we know by identifying an asymmetry in the timing of erector spinae activation during a well-controlled, bilateral movement task. Combined with previous research, these results support further investigation into whether asymmetrical paraspinal muscle forces might contribute to the curve progression and asymmetrical bony development in AIS. This is important as muscle forces are modifiable through targeted rehabilitation.