Quantitative Magnetic Resonance Imaging Analysis of the Cervical Spine Extensor Muscles: A Pilot Study

Abstract

Introduction Variations in cervical muscle cross-sectional area (CSA) and composition, particularly of the multifidus muscle, have been reported in patients with chronic neck pain. However, few studies have reported on the reliability of such muscle measurements and there remains no standard protocol for tissue segmentation. Therefore, the purpose of this pilot study was to provide a detailed muscle measurement protocol and determine the reliability of associated cervical muscle size and composition measurements using an open-source image analysis software (ImageJ). Material and Methods Cervical magnetic resonance images of 10 individuals with spinal stenosis were selected from an internal database. Muscle CSA and functional cross-sectional area (FCSA, fat-free area) measurements of the multifidus, semispinalis cervicis, semispinalis capitis, and splenius capitis were acquired bilaterally from axial T2-weighted magnetic resonance image from C2–C3 to C6–C7 levels. All measurements were repeated twice, at least 5 days apart and the assessor was blinded to all earlier measurements. Results The reliability for the upper (C2–C3 and C3–C4) and lower cervical levels (C4–C5, C5–C6, and C6–C7) was assessed separately. The intrarater reliability measurements were comparable between muscles and spinal levels. The intraclass correlation coefficient (ICC) for the CSA measurements varied between 0.79 to 0.97 at C2–C4 and 0.75 to 0.91 at C4–C7. The reliability was similar for the FCSA measurements and varied between 0.73 to 0.93 at C2–C4 and 0.78 to 0.90 at C4–C7. Conclusion The results of this pilot study suggest that the proposed method to investigate cervical muscle size and composition is reliable, with moderate-to-excellent reliability across cervical muscles and vertebral levels.