Changes in normal appearing spinal cord in multiple sclerosis: another brick in the wall

Abstract

It is needless to say how magnetic resonance (MR) has changed drastically the clinical management of multiple sclerosis (MS). Research in this field has also been much influenced by advances in MR techniques. Conventional MR is invaluable for the detection of MS plaques, using T2, protondensity and FLAIR (fluid attenuated inversion recovery) images. Inflammatory activity can be observed by means of postcontrast T1-weighted images. Together, they are the basis of the present diagnostic criteria for dissemination in space and time that characterizes MS1. However, the discrepancy between clinical manifestations of the disease and conventional MR metrics for lesion burden is well known, the so-called clinico-radiological paradox2. Some of the reasons listed to explain this dissociation include: inappropriate clinical rating, lack of histopathological specificity, neglecting spinal cord involvement, underestimating damage to the normal appearing brain tissue; and masking cortical adaptation effects2. MR has once again contributed enormously in this regard. Quantitative techniques (such as magnetization transfer and diffusion-weighted MR) disclose damage not only in conventional MR-visible lesions, but also in normal appearing tissues; MR spectroscopy adds metabolic information and functional MR demonstrates how cortical activation is affected during the course of the disease3. A robust body of evidence has emerged from all these data, pointing out to our current understanding of MS as being not only a chronic inflammatory / demyelinating disease, but also a degenerative condition of the central nervous system. Another paradigm shift is related to the pathophysiology of MS: it is not only a white matter disease, characterized by focal inflammatory lesions, but it also involves more subtle and diffuse damage throughout white and gray matter3. Specifically focusing on the spinal cord, it is a highly organized and clinically eloquent structure that is often affected by MS. Due to its peculiar anatomic arrangement, the cord is a useful model to assess the relations between inflammation, demyelination, and axonal pathology4. Diffusion tensor imaging (DTI) is one of the most important quantitative techniques. It is a refinement of the conventional diffusion-weighted imaging, able to measure the random diffusional motion of water molecules and provide quantitative indices of structural and orientational features of tissues5,6. Among the most used DTI indices we can mention mean diffusivity (MD), which is a directionally averaged measure of the apparent diffusion coefficient, and fractional anisotropy (FA), which summarizes the orientational dependence of diffusivity7. More recently, new DTI metrics were introduced, probing new aspects of tissue structure and disease-related changes, such as axial (AD) and radial diffusivity (RD). AD is considered to represent the water diffusion in parallel to the main structure of the white matter fibers, while RD represents the water diffusion perpendicular to them. It has been postulated, firstly in studies involving animals, that decreased AD and increased RD could represent surrogate markers of axonal damage and demyelination, respectively8. Performing DTI of the spinal cord is much more challenging than of the brain, due to the small cross section of the cord, pulsatile cord motion, and field inhomogeneities caused by variations in susceptibility from nearby vertebrae9. Nevertheless, in spite of these limitations, Neuroradiologist, Instituto de Radiologia do Hospital das Clinicas da Faculdade de Medicina da USP. CDB – Centro de Diagnosticos Brasil, Sao Paulo SP, Brazil. Correspondence Leandro Tavares Lucato Rua Prof Pedreira de Freitas 372 / apto 101 / bloco 2 03312-052 Sao Paulo SP Brasil E-mail: ltlucato@uol.com.br Conflict of interest There is no conflict of interest to declare.