Psychophysiology, 38:S30, 2001.

Pretreatment differences in ERPs between SSRI antidepressant responders and non-responders

Gerard Bruder, Craig Tenke, Jürgen Kayser, Paul Leite, Patrick McGrath, Frederic Quitkin

Department of Biopsychology, New York State Psychiatric Institute, 1051 Riverside Drive, New York, NY 10032-2695, USA

Abstract

Differences among depressed patients on electrophysiologic and dichotic listening measures have been found to be related to therapeutic responsiveness to antidepressants. This report compares the pretreatment ERPs of 38 depressed patients who responded to the selective serotonin reuptake inhibitor (SSRI) fluoxetine, 17 patients who were non-responders, and 49 healthy controls. EEG was recorded from 30 sites during oddball tasks using binaural complex tones or syllables. A principal components analysis with unscaled Varimax rotation measured five components identified in our prior study (Kayser et al., 1998): N1 (N100), N2 (N215), early P3 (P315), late P3 (P400) and a late positive slow wave. Both responders and non-responders showed the same task-dependent hemispheric asymmetries of N2 and P3 components as seen for controls. There was a significant difference among groups in early P3 amplitude, but not in late P3 amplitude. Non-responders showed reduced early P3 amplitude, whereas responders had the same amplitude as controls. While the late P3 had the expected parietal maximum scalp distribution, early P3 extended into frontocentral regions. The frontally-oriented topography of early P3 resembled that typically seen for other early P3 subcomponents, i.e., P3a or the novelty P3. The reduction of early P3 in fluoxetine non-responders may therefore be consistent with reports of prefrontal dysfunction in patients who fail to respond to antidepressants.

Keywords: ERP, Depression, Antidepressant