Diffusion MRI of the spinal cord: from structural studies to pathology Yoram Cohen a,b *, Debbie Anaby a and Darya Morozov a Diffusion MRI is extensively used to study brain microarchitecture and pathologies, and water diffusion appears highly anisotropic in the white matter (WM) of the spinal cord (SC). Despite these facts, the use of diffusion MRI to study the SC, which has increased in recent years, is much less common than that in the brain. In the present re- view, after a brief outline of early studies of diffusion MRI (DWI) and diffusion tensor MRI (DTI) of the SC, we provide a short survey on DTI and on diffusion MRI methods beyond the tensor that have been used to study SC microstruc- ture and pathologies. After introducing the porous view of WM and describing the q-space approach and q-space diffusion MRI (QSI), we describe other methodologies that can be applied to study the SC. Selected applications of the use of DTI, QSI, and other more advanced diffusion MRI methods to study SC microstructure and pathologies are presented, with some emphasis on the use of less conventional diffusion methodologies. Because of length con- straints, we concentrate on structural studies and on a few selected pathologies. Examples of the use of diffusion MRI to study dysmyelination, demyelination as in experimental autoimmune encephalomyelitis and multiple sclero- sis, amyotrophic lateral sclerosis, and traumatic SC injury are presented. We conclude with a brief summary and a discussion of challenges and future directions for diffusion MRI of the SC. Copyright © 2016 John Wiley & Sons, Ltd. Keywords: diffusion MRI (DWI); diffusion tensor imaging (DTI); q-space diffusion MRI (QSI); spinal cord; microstructure; spinal cord pathology INTRODUCTION MRI is currently by far the most important imaging modality of the central nervous system (CNS), and diffusion is one of the ma- jor contrast mechanisms used to non-invasively image and study brain architecture and pathologies (1,2). Diffusion MRI is an established method for studying brain microstructure and for characterizing neurological disorders and diseases both pre- clinically and in the clinic (1,2). In contrast to the enormous num- ber of diffusion MRI studies of the brain in both animals and human subjects, the number of diffusion MRI studies of the spi- nal cord (SC), which has increased in recent years, is still dramat- ically lower than that of the brain. The SC is an integral part of the CNS that interconnects the brain with the peripheral nervous system. Water diffusion, at least in the white matter (WM) of the SC, appears highly anisotropic. The fact that diffusion MRI has not been more widely used to study the SC is partially due to the fact that in vivo diffusion MRI of the SC is significantly more difficult than that in the brain (3,4). The SC, which runs deep in the body, is a much smaller organ in the axial plane (about 15 mm in human and about 3 mm in rat, for example) than the brain. In addition, the bony structures that surround the SC con- tribute to significant susceptibility artifacts (3), and respiratory motion, cardiac pulsation, and other physiological and physical motions that cause artifacts are significantly more pronounced in the SC than in the brain (3,4). Furthermore, many of the diffu- sion tensor imaging (DTI) studies of the brain have been devoted to diffusion tensor tractography (DTT) (1,2,5), which appears, at least at first glance, less relevant to the SC than to the brain. In addition, many of the early diffusion-weighted MRI (DWI) studies in the CNS have focused on cerebral ischemia, a common pathol- ogy of the brain. * Correspondence to: Y. Cohen, The Sackler School of Chemistry, The Raymond and Beverly Sackler Faculty of Exact Sciences, Tel Aviv University, Ramat Aviv, Tel Aviv, Israel. E-mail: ycohen@post.tau.ac.il a Y. Cohen, D. Anaby, D. Morozov The Sackler School of Chemistry, The Raymond and Beverly Sackler Faculty of Exact Sciences, Tel Aviv University, Tel Aviv, Israel b Y. Cohen The Sagol School of Neuroscience, Tel Aviv University, Tel Aviv, Israel Dedicated to the memory of Professor Sir Paul T. Callaghan – a superb scientist and a great man. Abbreviations used: AD, axial diffusivity; ADC, apparent diffusion coefficient; ALS, amyotrophic lateral sclerosis; ASIA, American Spinal Injury Association; AxD, axon diameter; BBBS, Basso–Beattie–Bresnahan score; BDNF, brain-de- rived neurotrophic factor; BMS, Basso mouse scale; CNS, central nervous sys- tem; CSa, cross-sectional area; CST, cortico-spinal tract; DAPI, 4′,6-diamidino- 2-phenylindol; DBSI, diffusion basis spectrum imaging; DC, dorsal column; DFER, DTI-based fiber estimate ratio; DKI, diffusion kurtosis imaging; d-PFG, double-pulsed field gradient; DTI, diffusion tensor imaging; DTT, diffusion ten- sor tractography; DWI, diffusion-weighted imaging; EAE, experimental autoim- mune encephalomyelitis; EDSS, Expanded Disability Status Scale; EPI, echo planar imaging; FA, fractional anisotropy; FWHM, full width at half maximum; GFA, generalized FA; GM, gray matter; HARDI, high-angular-resolution diffu- sion imaging; H&E, hematoxylin and eosin; lADC, longitudinal diffusivity; LDK, Lenaldekar; LFB, Luxol fast blue; MAD, mean axon diameter; MBP, myelin basic protein; MD, mean diffusivity; Md, myelin deficient; MRM, magnetic res- onance microscopy; MS, multiple sclerosis; MT, magnetization transfer; MTR, MT ratio; NAA, N-acetyl aspartate; NAWM, normal-appearing white matter; NF, neurofilament; NODDI, neurite orientation dispersion and density imaging; NOGSE, non-uniform oscillating-gradient spin echo; OGSE, oscillating-gradient spin echo; PPMS, primary progressive MS; QBI, q-ball imaging; QSI, q-space dif- fusion MRI; RA, relative anisotropy; RD, radial diffusivity; Rmsd, root-mean- square displacement; ROI, region of interest; SC, spinal cord; SCI, SC injury; SGP, short gradient pulse; SNR, signal-to-noise ratio; STE, stimulated echo; tADC, transverse diffusivity; VLWM, ventrolateral white matter; WM, white matter. Special issue review article Received: 30 September 2015, Revised: 1 June 2016, Accepted: 5 July 2016, Published online in Wiley Online Library: 6 September 2016 (wileyonlinelibrary.com) DOI: 10.1002/nbm.3592 NMR Biomed. 2017;30:e3592. Copyright © 2016 John Wiley & Sons, Ltd. 1 of 33