The Role of MR Diffusion Weighted Imaging in the Differential Diagnosis of Spinal Cord Lesions

Introduction: Diffusion weighted imaging (DWI) and diffusion tensor imaging (DTI) are magnetic resonance imaging methods, nowadays commonly used to depict the brain. Application of these methods for the spinal cord imaging is technically more demanding and less frequent. The aim of this pilot study was to summarize authors' current experience with DWI and DTI of the spinal cord, considering the potential value for the differential diagnosis of the spinal cord lesions. Methods: We retrospectively evaluated DWI/DTI findings in a group of 11 patients with pathological findings of the spinal cord on conventional MRI examination. The diagnosis comprised spinal cord ischemia, multiple sclerosis, myelitis, radiation myelopathy and arteriovenous malformations. We measured apparent diffusion coefficient (ADC) in all patients and, of the DTI data, we also measured fractional anisotropy (FA) values and evaluated the deterministic tractography reconstructions. Results: In four patients with spinal cord ischemia, we observed a decrease of the ADC values of the spinal cord lesions compared to the normal-appearing segment within a range 36-61%. Small areas of restricted diffusion of the spinal cord were found also in a patient with radiation myelopathy. In two patients with spinal cord tumors, DTI tractography showed displacement and/or disruption of the spinal cord tracts and marked decrease of the FA values (0.247 and 0.299). No abnormalities were observed on tractography in the rest of the patients, moderate decrease of the FA values was found in patients within demyelinating lesions of the spinal cord (0.494 and 0.471). Conclusion: DWI/DTI of the spinal cord may contribute to the correct direction of the differential diagnostic considerations through depiction of restricted diffusion within the spinal cord ischemia and evaluation of tractography in patients with spinal cord tumors. Further research with larger numbers of patients might enable differentiation of the spinal cord lesions based on quantification of DTI parameters.