Feasibility of diffusion tensor imaging in cervical spondylotic myelopathy using MUSE sequence

Abstract

Background context: The most frequent type of spinal cord injury is cervical spondylotic myelopathy (CSM). Conventional structural magnetic resonance imaging (MRI) is the gold diagnosis standard for CSM. Diffusion tensor imaging (DTI) could reflect microstructural changes in the spinal cord by tracing water molecular diffusion in early stages of CSM. However, due to the complex local anatomical structure and small field of view of the spinal cord, the imaging effect of traditional DTI imaging on the spinal cord is limited. MUSE (MUltiplexed Sensitivity-Encoding) -DTI is a novel diffusion-weighted imaging (DWI) sequence that achieves higher signal intensity through multiple excitation acquisition. MUSE sequence may improve the quality of spinal cord DTI imaging.

Study design: Prospective study.

Purpose: This study aimed to investigate the clinical diagnosis value of a novel protocol of MUSE-DTI in patients with cervical spondylotic myelopathy (CSM).

Patient sample: From August 2021 to March 2022, a total of 60 subjects (22-71 years) were enrolled, including 51 CSM patients (22 males, 29 females) and 9 healthy subjects (4 males and 5 females). Each subject underwent a MUSE-DTI examination and a clinical Japanese Orthopedic Association (JOA) scale.

Outcome measures: We measured values of FA (Fractional Anisotropy), MD (Mean Diffusivity), AD (Axial Diffusivity), and RD (Radial Diffusivity), and collected the clinical JOA scores of each subject before the MR examination.

Methods: A 3.0T MR scanner (Signa Architect, GE Healthcare) performed the MUSE-DTI sequence on each subject. The cervical canal stenosis of subjects was classified from grade 0 to grade Ⅲ according to the method of an MRI grading system. FA, MD, AD, and RD maps were generated by postprocessing MUSE-DTI data on the GE workstation. Regions of interest (ROIs) were manually drawn at the C2 vertebral body level and C2/3-C6/7 intervertebral disc levels by covering the whole spinal cord. The clinical severity of myelopathy of subjects was assessed by the clinical Japanese Orthopedic Association scale (JOA).

Results: MUSE-DTI can acquire a high-resolution diffusion image compared to traditional DTI. The FA_MCL values showed a decreasing trend from grade 0 to grade Ⅲ, while the MD_MCL, AD_MCL, and RD_MCL values showed an overall increasing trend. Significant differences in MD_MCL, AD_MCL, and RD_MCL values were found between adjacent groups among grades Ⅰ-Ⅲ (p<.05). The AD_C2 values in CSM patients (grade I-Ⅲ) were significantly lower than in healthy individuals (grade 0) (p=.019). The clinical JOA score has a significant correlation with FA_MCL (p=.035), MD_MCL (p<.001), AD_MCL (p<.001), and RD_MCL (p<.001) values.

Conclusions: MUSE-DTI displayed a better image quality compared to traditional DTI. MUSE-DTI parameters displayed a grade-dependent trend. All the MUSE-DTI parameters at MCL were correlated with the clinical JOA scores. The AD_C2 values can reflect the secondary damage of distal spinal cord. Therefore, MUSE-DTI could be a reliable biomarker for clinical auxiliary diagnosis of spinal cord injury severity in cervical spondylotic myelopathy.

Keywords: Axial diffusivity; Cervical spondylotic myelopathy; Fractional anisotropy; Japanese Orthopaedic Association scale, Mean diffusivity; MUltiplexed sensitivity-encoding diffusion tensor imaging; Radial diffusivity.