Abstract

Many brain regions show metabolic and perfusion abnormalities in major depressive disorder (MDD), including anterior cingulate and prefrontal cortices. Some of these same areas also show abnormal function with low resolution electromagnetic tomography (LORETA). However, LORETA results are not always consistent across studies, nor with findings from other imaging modalities. These discrepancies may be due, among other factors, to the sensitivity of EEG source localization to different electrode montages. Thirty-six channel EEG was collected from healthy controls and age- and gender-matched unmedicated subjects with MDD (n = 74). EEGs were analyzed with LORETA to assess resting state current density at each of 2,394 cortical voxels. For comparison to previous studies, LORETA was performed using all electrodes or with specific prefrontal electrodes removed. Voxel-by-voxel differences between the depressed and healthy groups were calculated using non-parametric statistics. MDD subjects showed significantly elevated current density in delta, theta, alpha, beta1, and beta2 frequency bands relative to controls in anterior cingulate and prefrontal cortices. Removal of certain prefrontal electrodes from input to LORETA decreased or eliminated significant differences between groups. LORETA detects differences in brain activity between MDD subjects and healthy controls that are consistent with previous findings using other imaging modalities. Inconsistent findings among LORETA studies, and between LORETA studies and those using other functional imaging techniques, may result from differences in electrode montages.