Duchenne muscular dystrophy (DMD) is an X-linked genetic neuromuscular disorder caused by absence of the protein dystrophin. Dystrophin provides structural stability to the membrane of muscle cells, but is also expressed in neuronal, glia and epithelial cells in which its function is unclear[1]. Clinically, DMD patients show a high incidence of cognitive and learning disabilities as well as neurobehavioral disorders[2]. Using diffusion tensor imaging we have previously demonstrated an increased apparent diffusion coefficient (ADC) and a reduced fractional anisotropy throughout the cerebral white matter in DMD patients compared to age matched controls[3]. However, the diffusion of water in biological tissues occurs inside, outside, around, and through cellular structures, making it difficult to interpret the underlying cause of the differences detected. In diffusion weighted MRS, the ADC of metabolites is assessed, allowing for detection of diffusion changes in these metabolites which can be predominantly glial (choline), exclusively neuronal (N-acetylaspartate) or ubiquitous (creatine). This information can provide insight into the underlying disease pathophysiology[4]. The aim of this study was to assess changes in metabolite ADC in DMD patients compared to controls in a brain region where reduced FA and increased ADC of water occurs in patients.Scans were obtained with a 3 tesla scanner (Philips Achieva, Best, The Netherlands) using an 8-channel head coil in ten patients with DMD (mean age 16.2, range 10-22 years) and six controls (mean age 16.2, range 11-18 years). A 3D T1-weighted scan (TE/TR, 4.6/9.8 ms; spatial resolution 1.17x0.92x1.17 mm; 4:55 min) was obtained for anatomical reference. DWS data were acquired using an ECG triggered PRESS sequence, (TE=125 ms, TR=2 cardiac cycles, VOI 30x20x15mm, 24 signal averages, 3:20 min, b=0 and b=3765 mm/s2, three diffusion directions). An additional DWS scan was performed without water suppression as a reference with the same parameters (2 signal averages, 24 s). The VOI was positioned exclusively in the white matter (Figure 1). The DWS spectra were zero-order phased, eddy current and frequency shift corrected and quantified using LCModel with a simulated basis set. The ADCs of total N-acetylaspartate (tNAA), total creatine (tCr) and choline (Cho) were then computed.[5]An F test was used to asses variances, and an unpaired t-test (with Welch correction when variances were different) was used to assess differences in ADCs between patients and controls (p<0.05). A typical fitted spectrum demonstrating the quality of the data is shown in figure 2. One patient scan was excluded as the metabolites did not meet the inclusion criteria of a Cramer-Rao Lower-Bound <7.5. Figure 3 shows the ADCs of tNAA, tCr and Cho per group. ADCs of tNAA and Cho were significantly higher (p=0.036 and p=0.026 respectively) and differences in tCr ADC between groups approached significance (p=0.054). There was also a significantly different variance in tNAA ADC in DMD patients compared to controls (p=0.029).Within the white matter in a group of DMD patients we have found higher tNAA, Cho and tCr ADCs using DWS, complementary to the previously reported reduced fractional anisotropy and increased radial diffusivity detected with DTI. There was no cell type specificity as all three altered metabolites represent different cell types (neurons, glia or ubiquitous). Dystrophin is normally expressed in both neuronal and glia cells, but absent in DMD patients. The non-cell-type specificity of our results contributes to the body of evidence that global cerebral changes occur in DMD[3]. Combined with earlier results of increased radial diffusivity this may indicate leaky membranes which allow exchange with the extracellular space similar to that seen in muscle cells in DMD patients[6]. Alternatively, there may be structural deficits within the cells that are non-specific such as changes to mitochondria or the cytoskeleton. The larger variance in ADC values in DMD patients suggest that may not occur to a similar extent in all patients. This may be related to the type of mutation patients have, as mutations distally located within the dystrophin gene affect more than one dystrophin isoform. It may also be related to the heterogeneity of the cognitive problems known in DMD, which do not affect all patients to the same degree. For future works, we propose a larger study to explore the relationship between the ADC of metabolites, genetics and the cognitive phenotype in DMD.