Eur Radiol (2005) 15: 1686–1693 DOI 10.1007/s00330-005-2775-0 NEURO Paul E. Sijens Roy Irwan Jan Hendrik Potze Jop P. Mostert Jacques De Keyser Matthijs Oudkerk Received: 6 January 2005 Revised: 15 March 2005 Accepted: 5 April 2005 Published online: 22 April 2005 # Springer-Verlag 2005 Analysis of the human brain in primary progressive multiple sclerosis with mapping of the spatial distributions using 1 H MR spectroscopy and diffusion tensor imaging Abstract Primary progressive multi- ple sclerosis (ppMS; n=4) patients and controls (n=4) were examined by 1H magnetic resonance spectroscopy (MRS) and diffusion tensor imaging (DTI) in order to map choline (Cho), creatine and N-acetylaspartate (NAA), the fractional anisotropy (FA) and the apparent diffusion constant (ADC). After chemical shift imaging (point- resolved spectroscopy, repetition time/ echo time 1,500 ms/135 ms) of a supraventricular volume of interest of 8×8×2 cm 3 (64 voxels) MRS peak areas were matched to the results of DTI for the corresponding volume elements. Mean FA and NAA values were reduced in the ppMS patients (P<0.01, both) and the ADC increased (P<0.02). The spatial distribution of NAA showed strong correlation to ADC in both ppMS patients and controls (r =-0.74 and r= -0.70; P<0.00001, both), and weaker corre- lations to FA (r=0.49 and r=0.41; P<0.00001, all). FA and ADC also correlated significantly with Cho in patients and controls (P<0.00001, all). The relationship of Cho and NAA to the ADC and the FA and thus to the content of neuronal structures suggests that these metabolite signals essen- tially originate from axons (NAA) and the myelin sheath (Cho). This is of interest in view of previous reports in which Cho increases were associated with demyelination and the subse- quent breakdown of neurons. Keywords Magnetic resonance spectroscopy . Diffusion tensor imaging . Metabolism . Multiple sclerosis P. E. Sijens (*) . R. Irwan . J. H. Potze . M. Oudkerk Department of Radiology, University Medical Center Groningen, Hanzeplein 1, Groningen, The Netherlands e-mail: p.e.sijens@rad.azg.nl Tel.: +31-50-3613534 Fax: +31-50-3611798 J. P. Mostert . J. De Keyser Department of Neurology, University Medical Center Groningen, Hanzeplein 1, 9713 GZGroningen, The Netherlands Introduction Diffusion tensor imaging (DTI) is a recent application of MRI based on measurement of the Brownian motion of water molecules [1]. In DTI, a series of diffusion-weighted images with diffusion-encoding gradients applied in var- ious directions are acquired and the tensor is computed via linear or nonlinear regression [2, 3]. The first clinical ap- plications of DTI in multiple sclerosis (MS) have been reported [2, 4–8]. The main parameters measured are the fractional anisotropy (FA), which reflects the preferential direction of water diffusion along the white matter tracks, and the apparent diffusion coefficient (ADC), which re- flects overall diffusivity. The FA ranges from 1 to 0, with 0 being completely isotropic diffusion and 1 being diffusion constrained to occur in one direction only. ADCs are usu- ally expressed in square millimeters per second or square centimeters per second, and are typically observed for 20– 50 ms on clinical scanners. A normal ADC in the human brain is 10 -5 cm 2 /s, using an observation time of 40 ms. The ADC decreases when the protons are slowed down in their random motion by the presence of membranes and other cell structures. In general, the more unrestricted the water molecules are in a given tissue, the higher the ADC will be and the lower the FA. 1 H magnetic resonance spectroscopy (MRS) has been used for decades as a noninvasive technique for assessing the biochemistry of the human brain. This is reflected by the increasing number of clinical applications of spectroscopy to investigate neurological disorders. 1 H-MRS of the brain