Back muscles biometry in adolescent idiopathic scoliosis

BACKGROUND CONTEXT: Many studies have been devoted to the role of back muscle activity in the development of scoliosis. While an imbalance in the electromyographic (EMG) activity has often been detected at the skin surface, very little information is available on the mechanisms by which such an imbalance could take place. To gain insight into those mechanisms, an important step could be the collection of anatomical data on the volume of the erector spinae muscle on both sides of the spine as well as on the skin and subcutaneous fat (skinfold) thickness separating those muscles from the body surface. For this purpose, the use of magnetic resonance (MR) imaging is appropriate. PURPOSE: To collect anatomical information on the erector spinae muscles and skinfold thickness along the spinal deviations of scoliotic patients. STUDY DESIGN: In an observational retrospective study, MR images of scoliotic patients treated in a pediatric hospital in the last 5 years were analyzed. PATIENT SAMPLE: Images were obtained from adolescent idiopathic scoliosis patients. METHODS: For 15 patients (Group 1), three clinical acquisition protocols were used. Five investigators were asked to grade the contrast of the images obtained with each protocol. All the assessments were carried on the same monitor without any change in its settings. For the MR sequence providing the best contrast, 25 fully imaged scoliotic deviations were obtained from 17 patients (Group 11). A manual segmentation with an image processing software package was done on the erector spinae muscle on both sides of the spine on each of the available images in order to determine their volume. Skinfold was also measured; first at regular intervals from C7 to L3 over the erector spinae muscle and then at sites centered over the apex of each curve. RESULTS: For Group 1, the spin echo (SE-Tl) was found to provide the best contrast to identify the contour of individual muscle. With this sequence, the analysis of the fully imaged scoliotic curves (Group 11) revealed that back muscle volume was found larger 14 times on the concave side and I I times on the convex one. When the length of each curve was normalized and then divided into three equal regions, muscle volume was larger I I times at the apex (6 times on concave side), 7 times above and 7 times below (4 times on the concave side for both positions). From C7 to L3, the mean skinfold thickness of each patient ranged from 7.3 mm to 16.3 mm. On average, this thickness was < 10 mm between T3 and T 12 but became larger at L3 level. At the apex of each scoliotic deviation, skinfold thickness was always larger on the concave side, and the difference decreased progressively as the distance from the apex increased. CONCLUSION: A larger back muscle volume in adolescent idiopathic scoliosis patients was slightly more frequent on the concave than on the convex side. The differences were more frequent at the apex of the curve. Skinfold thickness was always greater on the concave side at the apex region. (c) 2007 Elsevier Inc. All rights reserved.