Functional anatomy of the therapeutic effects of prism adaptation on left neglect J. Luaute ´, MD; C. Michel, PhD; G. Rode, MD; L. Pisella, PhD; S. Jacquin-Courtois, MD; N. Costes, PhD; F. Cotton, MD; D. le Bars, PhD; D. Boisson, MD; P. Halligan, PhD*; and Y. Rossetti, MD* Abstract—Objective: To investigate the anatomic substrates underlying the beneficial effect of prism adaptation in five patients with persistent left neglect following right stroke. Methods: In a functional imaging PET study, we used a covariation analysis to examine linear changes of regional cerebral blood flow over sessions as a function of left neglect improvement. Results: The network of significant brain regions associated with improvement of left neglect performance produced by prism adaptation involved the right cerebellum, the left thalamus, the left temporo-occipital cortex, the left medial temporal cortex, and the right posterior parietal cortex. Conclusion: Our results suggest that the realignment of visuomotor coordinates is processed by the cerebellum and that low level sensorimotor adaptation actively modulates cerebral areas, albeit now relying on intact cerebellocerebral connections. Hence, our data support the hypothesis that the beneficial effect of prism adaptation on the clinical presentation of left neglect derives from modulation of cortical regions implicated in spatial cognition. NEUROLOGY 2006;66:1859–1867 A brief period of right prism adaptation with left- ward compensatory aftereffects has been repeatedly shown to improve left neglect across a variety of different standard tests. 1 Two further aspects set this intervention apart from the previous attempts to alleviate unilateral neglect: (1) the observation (con- firmed in later studies) that the effects of adaptation could generalize across several different clinical measures of unilateral neglect, including wheelchair navigation, 2 postural control, 3 and neglect of mental imagery 4,5 and (2) the finding (encouraging from a rehabilitation perspective) that the effects persisted as long as 4 days after a single adaptation proce- dure. 6 Longer lasting effects (as long as 5 weeks) were even reported following an intensive twice-daily adaptation program during a 2-week period. 7 These cross-sectional observations suggest that, even after acquired brain damage, short-term exposure to visuomotor adaptation is sufficient to stimulate a long-term reorganization of the neural representa- tion of space that develops autonomously after re- moval of the prisms. The generalization and long-standing effects of prism adaptation make it a potential therapeutic treatment of choice and has revived interest in the neurocognitive mechanisms by which it has been achieved. However, the neural basis for this thera- peutic effect in unilateral neglect patients has yet to be formally established. In this study, we investi- gated the anatomic substrates underlying the benefi- cial effect of prism adaptation in five patients with persistent left neglect following right stroke using PET. Methods. Patients were selected from a neurorehabilitation center. The inclusion criterion was left neglect after right hemi- spheric ischemic or hemorrhagic stroke. In addition to a classic neurologic examination, the presence of hemianopia was assessed by means of Goldman perimetry, and unilateral neglect was for- mally assessed using the Behavioral Inattention Test (BIT). 8 The BIT provides an objective and standardized measure of unilateral neglect employing a battery of six conventional subtests (line can- cellation, letter cancellation, star cancellation, figure and shape copying, line bisection, and representational drawing). Using the cutoff suggested by the authors, a score 129 was employed as the operational diagnosis of unilateral neglect. The stability of this condition was confirmed by a second administration of the BIT 1 week later and on the day of the PET scanning. To allow for the reduction of general brain edema, a minimum delay of 1 month was employed between stroke onset and inclusion in the study. Cerebral MRI, including spin-echo T2-weighted and gradient-echo T2-weighted images, diffusion-weighted images, and high- resolution, three-dimensional T1-weighted images was performed in the intercommissural plane for each patient to confirm the type of lesion (ischemic or hemorrhagic) and rule out any other rele- vant prestroke lesions. In addition, the precise anatomic lesion locations were determined by a neuroradiologist. Patients with a *Authors contributed equally to the final paper. From the INSERM (J.L., G.R., L.P., S.J.C., D.B., Y.R.), U534, Espace et Action, Bron, France; Hospices Civils de Lyon (J.L., G.R., S.J.C., D.B., Y.R.), Ho ˆpital Henry Gabrielle, Service de Re ´e ´ducation Neurologique, Mouvement et Handicap, Saint Genis Laval, France; Universite ´ Claude Bernard Lyon I (J.L., G.R., L.P., S.J.C., F.C., D.B., Y.R.), Faculte ´ de Me ´decine, Institut Fe ´de ´ratif des Neurosciences de Lyon, IFR19, Lyon, France; INSERM (C.M.), ERIT-M 0207, Motricite ´-Plasticite ´: Performance, Dysfonctionnement, Vieillissement et Technologies d’optimisation, Dijon, France; Hospices Civils de Lyon (F.C.), Centre Hospitalier Lyon Sud, Service de Radiologie, Pierre Be ´nite, France; CERMEP (N.C., D.L.B.), Ho ˆpital Neurologique, Bron, France; INSA-502 (F.C.), CREATIS, Villeurbanne, France; and School of Psychology (P.H.), Cardiff University, Cardiff, UK. This work was supported in part by INSERM PROGRES (no. 4P012E); ACI Neuroscience (Plasticite ´), Programme Hospitalier de Recherche Clinique National (PHRC no. 30251). Disclosure: The authors report no conflicts of interest. Received August 4, 2005. Accepted in final form March 7, 2006. Address correspondence and reprint requests to Pr. Yves Rossetti, INSERM Unite ´ 534, Espace et Action, 16, Avenue du doyen Le ´pine, 69500 Bron, France; e-mail: rossetti@lyon.inserm.fr Copyright © 2006 by AAN Enterprises, Inc. 1859