For personal use. Only reproduce with permission from The Lancet Publishing Group. Conventional MRI can improve accuracy in the diagnosis of multiple sclerosis (MS) and monitor the efficacy of experimental treatments. However, conventional MRI provides only gross estimates of the extent and nature of tissue damage associated with this disease. Other quantitative magnetic-resonance-based techniques have the potential to overcome the limitations of conventional MRI and, as a consequence, to improve our understanding of the natural history of MS. Magnetisation-transfer, diffusion- weighted, and functional MRI—as well as proton magnetic- resonance spectroscopy—are helping us to elucidate the mechanisms that underlie injury, repair, and functional adaptation in patients with MS. These techniques are substantially changing our understanding of how MS causes irreversible disability and should be used more extensively in clinical trials and in studies of disease progression. Lancet Neurology 2003; 2: 337–46 Conventional MRI plays an important part in the diagnosis of multiple sclerosis (MS; figure 1) and in the monitoring of the efficacy of experimental treatments. 1 However, this technique provides limited information on important pathological features of MS—such as the extent and nature of tissue damage in lesions visible on T2-weighted imaging, in normal appearing white matter, and in grey matter, the effectiveness of repair and cortical adaptation after tissue injury, and the factors that lead to irreversible disability in this disease. 2,3 Other quantitative magnetic-resonance-based techniques could improve our understanding of how MS develops. Magnetisation-transfer MRI, 4 diffusion-weighted MRI, 5 proton magnetic-resonance spectroscopy, 6 and functional MRI (fMRI) 7 have been used extensively to investigate injury, repair, and functional adaptation in patients with MS. In this article we review studies in which structural, metabolic, and functional magnetic-resonance-based techniques have been used to assess the pathology of MS in vivo. Magnetisation-transfer MRI The technique of magnetisation-transfer MRI is based on the comparison of proton interactions in a “free” environment (eg, water) with proton interactions in an environment where motion is restricted (eg, in tissue). When off-resonance irradiation is applied, the magnetisation of protons in tissue becomes saturated. Magnetisation is then transferred from these protons to more mobile protons, which causes a reduction of the tissue signal and a high magnetisation-transfer ratio (the exact ratio will depend on the scanner and sequence used). Thus, a low magnetisation-transfer ratio indicates a reduced ability of macromolecules in tissue to exchange magnetisation with surrounding water molecules, and is interpreted as an indication of damage to myelin and to other cellular structures (such as the axonal membrane; figure 2). A postmortem study that showed a strong correlation of the magnetisation-transfer ratio of lesions and normal appearing white matter with the extent of axonal loss and demyelination strongly suggests that a substantial reduction of magnetisation-transfer ratio in MS lesions indicates severe tissue damage. 9 By use of magnetisation-transfer MRI with variable scanning frequencies, several authors have investigated structural changes in new enhancing lesions in patients with MS. 10–17 In general, the magnetisation-transfer ratio falls substantially when the lesions begin to enhance, but can recover partially, or completely, during the subsequent 1–6 months. Rapid and pronounced changes in the magnetisation-transfer ratio suggest that demyelination and remyelination are the most likely underlying pathological mechanisms. Nevertheless, oedema, and its subsequent THE LANCET Neurology Vol 2 June 2003 http://neurology.thelancet.com 337 The use of quantitative magnetic-resonance- based techniques to monitor the evolution of multiple sclerosis Massimo Filippi, Maria A Rocca, and Giancarlo Comi MF, MAR, and GC are all at the Department of Neurology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy. MF and MAR are also at the Neuroimaging Research Unit, Scientific Institute and University Ospedale San Raffaele, Milan, Italy. Correspondence: Dr Massimo Filippi, Neuroimaging Research Unit, Department of Neurology, Scientific Institute and University Ospedale San Raffaele, via Olgettina 60, 20132 Milan, Italy. Tel +39 02 2643 3032; fax +39 02 2643 3054; email filippi.massimo@hsr.it Reviews Figure 1. Axial proton density-weighted (left) and post-contrast (gadolinium DTPA, 0·1 mmol/Kg) T1-weighted (right) spin-echo magnetic-resonance images of the brain of a patient with clinically definite MS. Left: multiple hyperintense lesions are visible, which suggest multifocal white-matter pathology. Right: some of these lesions are contrast-enhanced, which indicates the presence of local disruption of the blood–brain barrier.