Abstracts

In order to evaluate the fMRI activation resulting from epileptic activity, it is important to know the hemodynamic response function (HRF) of this activity. We studied the effect of different HRFs on the magnitude of fMRI activation in patients with idiopathic generalized epilepsy (IGE).
We selected 23 patients with active interictal generalized spike and wave (GSW). During continuous EEG-fMRI, EEG was recorded within a 1.5T MR scanner. The artefact caused by the imaging sequence was removed from the EEG by the method of Hoffmann et al¹. Nine patients were excluded due to lack of GSW during the study in eight, and technical problems in one. The data were analyzed with five HRFs: a measured response to brief auditory stimuli which peaks at 5.4 seconds ², and four single gamma functions peaking at 3, 5, 7, and 9 s. Activations were considered significant if three contiguous voxels had a t-stat [ge] 3.0. The highest t-stat for each activation with the different models was determined.
Data from fourteen patients with IGE (13 women), a mean age of 36, and mean age at seizure onset of 10 years were analyzed. Twelve patients showed significant positive and negative cortical and thalamic activations. The highest t-stat of positive cortical activation was obtained using Glover[rsquo]s model in nine patients, peak 7 model in two, and peak 5 in one. The highest t-stat of negative cortical activation was obtained with peak 9 in four patients, peak 7 in four, peak 5 in two, peak 3 in one, and Glover[rsquo]s model in one only. The highest positive thalamic t-stat was seen with Glover[rsquo]s model in six patients, peak 3 in one, and peak 9 in one. The three patients with negative thalamic activation had the highest t-stat at peak 9 in two, and at peak 5 in one, and none showed activation using the Glover model. In the remaining patient who showed both positive and negative thalamic activations, the highest positive activation (peak 7) followed the early negative activation (peak 3). The two areas of activation did not overlap.
The commonly used Glover model is appropriate for evaluating positive thalamic and cortical activation in the majority of our patients with GSW, but some patients showed higher activation using other early (3 s) or late (7 and 9 s) models. Negative activations, however, tended to occur later (peak 7 and 9 s) and would often be missed if only the Glover model were used. It is not clear why negative activation tends to occur later than positive activation.
References:
1) Hoffmann A et al. 2000 Magn. Reson. Med. : 791
2) Glover GH. 1999 NeuroImage 416
[Supported by: Canadian Institute of Health Research (CIHR)]