Objective:

To assess if the symmetry of bilateral sleep spindles can predict surgical outcomes in corpus callosotomy.

Background:

Corpus callosotomy is a palliative surgical treatment that could be offered to persons with medication-resistant epilepsy that reduces the severity and frequency of generalized seizures by disconnecting the two cerebral hemispheres. To this date pre-surgical predictors of response to corpus callosotomy remain poorly understood.

Design/Methods:

EEG was obtained using the international 10-20 system from 10 persons with medication-resistant epilepsy prior to their corpus callosotomy as part of routine pre-surgical evaluation. Only the initial 24 hours of EEG was included in the analysis to minimize the confounding effects of medication withdrawal. The EEG was down sampled to 100 Hz, and bandpass filtered (third order Butterworth filter, zero phase shift) to low-frequency sleep spindle (10-12 Hz) and high-frequency sleep spindle (12-16 Hz). The symmetry of sleep spindles was assessed using cross-correlation at 0-time lag for each homologous electrode pairs in the frontal, central, and parietal scalp EEG electrodes. Inter-group differences were assessed using 2-tailed t-test.

Results:

The super responders to corpus callosotomy showed a higher cross-correlation in the low-frequency sleep spindles compared to the poor responders in the central area (P = 0.04), but not in the frontal (P = 0.86) or the parietal (P = 0.97) regions. There was no inter-group difference in the high-frequency sleep spindles.

Conclusions:

The symmetry of the sleep spindles in the pre-surgical scalp EEG may serve as a predictor of post-surgical outcome in corpus callosotomy.