Results of Presurgical fMRI Testing of Language on a 1 Tesla Clinical MR Scanner in Patients with Temporal Lobe Epilepsy Compare Favourably with the Wada-Test.
Abstract number :
1.242
Submission category :
Year :
2001
Submission ID :
2667
Source :
www.aesnet.org
Presentation date :
12/1/2001 12:00:00 AM
Published date :
Dec 1, 2001, 06:00 AM
Authors :
P. Boon, MD, PhD, Neurology, Ghent University Hospital, Gent, Belgium; K. Deblaere, MD, MR Department, Ghent University Hospital, Gent, Belgium; P. Vandemaele, MSc, MR Department, Ghent University Hospital, Gent, Belgium; V. De Herdt, Neurology, Ghent Uni
RATIONALE: The golden standard for presurgical language and memory localization in temporal lobe epilepsy (TLE) is the Wada-test. This study tests the feasibility and reliability of fMRI language lateralization at 1 Tesla as compared with the Wada-test.
METHODS: We prospectively examined 16 TLE patients. All underwent a Wada-procedure with consecutive bilateral carotid injection of 2 mg/kg sodium amobarbital. fMRI was performed on a 1 Tesla MR system (Siemens Magnetom, whole head, 32 axial EPI images, TR=4 sec). A block design with 2 conditions was used. In condition 1 patients were given a word through headphones, instructed to create new words starting with the last letter of the previous word (covert word chain task - CWCT) and avoid repetitions. In condition 2 patients were instructed to start counting covertly. The duration of each test epoch was optimal at 10 scans (60 seconds: counting; 60 seconds: word generation; 60 seconds: counting; etc...). Language lateralization was calculated by contrasting condition 1 with condition 2. SPM99 was used for statistical analysis and image post-processing.
RESULTS: One patient failed the Wada-test because of a seizure; in another patient fMRI data were corrupted with motion artifacts. In 13/14 remaining patients, including 3 left-handers, there was significant activation in language related areas (p[lt]0.05, corrected). In the only ambidexter patient activation was less significant (p[lt]0.05 uncorrected). In 13/13 patients, fMRI and Wada-test congruently showed unequivocal left-hemisphere language dominance. One right-handed patient, who showed bilateral speech during the Wada-test had bilateral (R[gt]L) fMRI activation.
CONCLUSIONS: The covert word chain task is an easy to perform language task despite the long duration of the test epoch, due to the low field MR system that was used. Reliable fMRI language lateralization at 1 Tesla is feasible when the paradigm is adapted to the possibilities of the clinical MR unit.
Support: Grants BOZF-01104495 and 011A0996 from Ghent University, by a grant from the Fund for Scientific Research - Flanders (F.W.O.) and by the Clinical Epilepsy Grant 1999-2001.