Electroencephalogram resting state frequency power characteristics of suicidal behavior in female patients with major depressive disorder

Background: Difficulties in predicting suicidal behavior hamper effective suicide prevention. Therefore, there is a great need for reliable biomarkers, and neuroimaging may help to identify such markers. Methods: Electroencephalography (EEG) was used to investigate resting state spatial-frequency power characteristics of female patients with major depressive disorder (MDD); 19 were recent suicide attempters (within the previous 30 days), 36 were suicide ideators, and 23 were nonsuicidal. Patients were enrolled at neuroCare Clinic Nijmegen (Nijmegen, the Netherlands) between May 2007 and November 2016, and the primary diagnosis of nonpsychotic MDD was confirmed using the Mini-International Neuropsychiatric Interview, DSM-IV criteria, and a score of ≥ 14 on the 21-item Beck Depression Inventory. Nonparametric, cluster-based permutation tests were applied to detect robust power differences between the study groups on the EEG broadband signal (2-100 Hz). Furthermore, a nonadaptive distributed source imaging method (eLORETA) was utilized to examine if these suicide-based frequency characteristics are localized in brain areas previously reported in the neuroimaging literature. Results: When compared to nonsuicidal depressed patients, attempters and ideators displayed both decreased beta and low gamma activity in the frontal regions. Moreover, ideators had increased alpha activity over the posterior regions and increased high beta, low gamma activity over the left occipital region when compared to psychiatric controls. Attempters had reduced beta and low gamma activity over the right temporal region when compared to ideators. In addition, eLORETA localized attempter and ideator reduced frontal activity within the orbito-, medial-, middle-, superior-, and inferior-frontal areas and the anterior cingulate cortex. In attempters, reduced right temporal activity was localized within the right inferior-, middle-, and superior-temporal cortices and the fusiform gyrus. Conclusions: Frequency power characteristics of attempters and ideators are consistent with findings from the neuroimaging literature concerning suicide, implying EEG resting state assessment could become a potential biomarker to predict suicide risk.