Abstracts

Rationale: Event-related spectral amplitude changes on electrocorticography (ECoG) have been used for functional brain mapping in humans. Both short-lasting event-related gamma augmentation and long-lasting alpha-beta attenuation have been considered as surrogate markers of task-related cortical activation. We have recently reported short-latency (up to 40 msec) median-nerve somatosensory-induced gamma (30-250 Hz) augmentation recorded on ECoG (Fukuda et al, Brain 2008). In the present study, we described the spatial and temporal characteristics of long-latency (after 40 msec) median-nerve somatosensory-induced spectral amplitude changes. Methods: Ten children with focal epilepsy were studied (age: 4-17 yrs; 7F & 3M). All patients underwent preoperative MRI and extraoperative ECoG recording. All patients were given 200 median-nerve stimuli at the wrist contralateral to the epileptogenic hemisphere. Somatosensory-induced spectral frequency change was measured in comparison to the baseline immediately preceding stimulus. The frequency bands of interest included: high-frequency gamma (100-250 Hz), low-frequency gamma (30-100 Hz), beta (12-30 Hz), and alpha (8-12 Hz) bands. Results: Visual assessment revealed high-frequency gamma-augmentation in the post-central gyrus at 13.6 to 17.5 msec, gradually slowing down in frequency around and below 100 Hz, progressively involving the neighboring areas and lasting up to 60 msec. Quantitative assessment revealed that either post- or pre-central gyrus showed significant amplitude augmentation in high-frequency gamma (N=10 pts), low-frequency gamma (N=10), beta (N=9) and alpha (N=8) bands. The maximal augmentation of high-frequency gamma-oscillations occurred earlier than that of low-frequency gamma-oscillations (median: 20.0 ms vs. 27.5 ms, respectively). Similarly, the maximal augmentation of low-frequency gamma-oscillations occurred earlier than beta-oscillations, and that of beta-oscillations occurred earlier than alpha-oscillations (median: 25.0 ms vs. 50.0 ms vs. 75.0 ms). Following such beta and alpha augmentation, sustained beta and alpha attenuation was noted in the post- or pre-central gyrus in 6 and 6 subjects, respectively. The maximal attenuation of beta amplitudes occurred earlier than that of alpha amplitudes (median: 200 ms vs. 325 ms). Beta-attenuation subsided to the baseline earlier than alpha attenuation (median: 463 ms vs. 513 ms). Conclusions: The common median-nerve somatosensory-induced spectral changes included: (i) high-frequency gamma augmentation gradually evolving into low-frequency gamma augmentation, (ii) subsequent beta and alpha augmentation and (iii) subsequent long-lasting beta and alpha attenuation. The short-latency somatosensory-induced high-frequency gamma-augmentation in the post-central gyrus probably represents the initial neural processing for somatosensory stimuli, but the mechanisms for long-latency amplitude modulation involving beta and alpha bands remain hypothetical. (NIH Grant NS47550 to E. A)