Please cite this article in press as: Vargas MI, et al. Brain and spine MRI artifacts at 3Tesla. Journal of Neuroradiology (2008), doi:10.1016/j.neurad.2008.08.001 ARTICLE IN PRESS +Model NEURAD-129; No. of Pages 8 Journal of Neuroradiology (2008) xxx, xxx—xxx Disponible en ligne sur www.sciencedirect.com REVIEW Brain and spine MRI artifacts at 3Tesla Artéfacts en IRM 3Tesla du cerveau et de la moelle M.I. Vargas a,* , J. Delavelle a , R. Kohler b , C.D. Becker b , K. Lovblad a a Division of Neuroradiology, Department of Radiology, Geneva University Hospital, 24, road Micheli-du-Crest, 1211 Geneva 14, Switzerland b Department of Radiology, Geneva University Hospital, Switzerland KEYWORDS Artifacts; 3 Tesla; Brain; Spine; Magnetic resonance imaging Summary Introduction. — We illustrate here the most common MRI artifacts found on routine 3 T clinical neuroradiology that can simulate pathology and interfere with diagnosis. Materials and methods. — Our group has worked with a 3-T Magnetom Trio (Siemens, Erlangen, Germany) system for two years, with 50% of our time devoted to clinical work and 50% dedicated to research; 65% of the clinical time is dedicated to neuroradiology (2705 patients) and the remaining time to whole-body MRI. We have detected these artifacts during our case readings and have selected the most representative of each type to illustrate here. Results. — We have observed magnetic susceptibility artifacts (29%), pulsation artifacts (57%), homogeneity artifacts (3%), motion artifacts (6%), truncation artifacts (3%) and, finally, artifacts due to poor or inadequate technique in the examined region. Conclusion. — High-field imaging offers the benefit of a higher signal-to-noise ratio, thus making possible the options of a higher imaging matrix, thinner slices, the use of spectroscopy and dif- fusion tensor imaging in the routine clinical neuroradiology with a reduction in time spent. It is vital to be able to recognize these artifacts in everyday practice as they can mimic pathological appearances, thus causing diagnostic errors that could lead to unnecessary treatment. Indeed, most of these artifacts could be avoided with an adequate technique. © 2008 Elsevier Masson SAS. All rights reserved. MOTS CLÉS Artéfacts ; 3 Tesla ; Cerveau ; Rachis ; Imagerie par résonance magnétique Résumé Introduction. — Nous illustrons les artéfacts les plus fréquents en IRM à 3 T que nous avons rencontrés en pratique quotidienne et qui peuvent simuler une pathologie et donc interférer avec le diagnostic. Matériel et méthodes. — Notre groupe travaille sur un système 3 T Siemens depuis deux ans, avec une activité clinique représentant 50 % du temps machine et 50 % du temps restant dédié à la recherche fondamentale ; 65 % du temps d’activité clinique est dédié à la neuroradiologie (2705 patients au total) et le reste dédié à une activité corps entier. Nous avons identifié ces artéfacts pendant nos lectures et montrons les plus représentatifs. ∗ Corresponding author. E-mail address: Maria.i.Vargas@hcuge.ch (M.I. Vargas). 0150-9861/$ – see front matter © 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.neurad.2008.08.001