0310 QEEG analysis of Sleep Spindles using Low Resolution Electromagnetic Tomographic Activity (LORETA) Computations

Abstract

Introduction: EEG analysis has improved using computer technology. Assessment with magnetoencephalography (MEG) is difficult and expensive but provides source localization of activity. Combining EEG, MEG and MRI allowed for Low Resolution Electromagnetic Tomographic Activity (LORETA) analysis, claiming to provide source localization of cerebral activity within the brain from scalpel EEG alone. LORETA results are reported as current-source-density within an approximated 3D brain image. This technique can be applied to small segments of EEG yielding results from brief data segments not possible with fMRI or PET. We applied this method of EEG analysis specifically on sleep spindles to characterize the specific features of this sleep phenomena. Methods: Overnight EEG data from patients undergoing sleep studies to evaluate parasomnias where seizures were a consideration as the cause were used for the analysis. Twenty to Thirty seconds of sleep spindles EEG data was clipped from stage 2 sleep of each patients' data, excluding V-waves and K-Complexes. The resulting segment of EEG data was used for QEEG analysis that included LORETA and Z-score sLORETA 3D images, representing a contrast of brain regional activity between resting, eyes closed awake state and Stage 2 sleep during moments of sleep spindle activity. Results: Restricted midline brain regions demonstrated the highest activity between 12 to 14 Hz that involved a large portions of the frontal lobes. Z-score mapping against awake, eyes closed, age matched controls, demonstrated at this same 12 to 14 Hz, large regions that were beyond 4 SD. These regions were portions of the frontal lobes, including the prefrontal regions and cingulate gyrus, and adjacent subcortical limbic structures and caudate. Minimal parietal and occipital regional differences were identified. Conclusion: Using LORETA computations of EEG to assess sleep physiology can provide insight into neuronal functioning on a subcortical level in a more practical fashion than more elaborate techniques such as fMRI or PET scans. After better characterizing normal function, this technique may be utilized to explore pathologic conditions and hopefully shed more insight into dysfunctional sleep and potential therapies.