Language Lateralization in Epilepsy Patients: Comparison between Picture Naming Functional MRI and Wada Test
Abstract number :
1.246
Submission category :
Year :
2001
Submission ID :
3045
Source :
www.aesnet.org
Presentation date :
12/1/2001 12:00:00 AM
Published date :
Dec 1, 2001, 06:00 AM
Authors :
M.V. Spanaki, MD, PhD, Neurology, Medical College of Wisconsin, Milwaukee, WI; S.J. Swanson, PhD, Neurology, Medical College of Wisconsin, Milwaukee, WI; T.A. Hammeke, PhD, Neurology, Medical College of Wisconsin, Milwaukee, WI; E.T. Possing, MS, Neurolog
RATIONALE: Functional MRI (fMRI) during a semantic decision task has been successfully used to determine language localization in normal and epilepsy subjects. In this study, we compared an alternate fMRI language activation paradigm, a picture naming task, with Wada language testing in epilepsy patients.
METHODS: Fifty-eight patients with refractory epilepsy (age range 20-59 years) underwent Wada testing and fMRI at 1.5T during a picture naming task. Patients were shown a picture and a letter and were to decide if the name of the picture began with the letter. Fifty-two of these patients also performed a semantic decision task (Binder et al; Neurology 1996;46:978-984) during fMRI. We correlated MRI activation data with a reference vector on a voxel-by-voxel basis. The number of voxels exceeding the threshold was calculated in each patient for each hemisphere. We calculated the fMRI laterality index (LI) for each patient as the normalized difference between left (L) and right (R) hemisphere total voxel counts using the formula (L-R)/(L+R)X100. LIs between 0 and +100 indicate left and LIs between 0 and -100 indicate right hemispheric dominance. LIs from the picture naming fMRI were correlated with LIs from Wada and semantic fMRI.
RESULTS: Twenty-eight patients had right and 22 had left temporal epilepsy; 8 were undetermined or multifocal. Forty-eight patients were right handed (RH), 6 left (LH) and 4 were ambidextrous. Picture naming fMRI and Wada were in agreement with respect to language representation in 53/58 patients (91%). In 3 of the discordant cases (2 RH, 1 LH) Wada showed bilateral language representation and fMRI revealed weak left lateralization. In another 2 LH patients, Wada indicated right lateralization where fMRI showed weak left lateralization. Correlation between Wada and picture naming LIs was significant (r = .60, 95%CI; p [lt] .001).
Correlation between picture naming and semantic decision task fMRI was also significant (r = .74, 95%CI; p [lt] .001)
CONCLUSIONS: We found significant concordance between fMRI picture naming and Wada tests with respect to language lateralization in epilepsy patients. Use of multiple language task paradigms during fMRI may be important for maximizing accuracy and reliability of language localization. FMRI may become a reasonable alternative in determining language lateralization in the presurgical evaluation of epilepsy patients.
Support: NINDS RO1 35929 and NINDS NS33576