EEG microstate and functional connectivity analyses for differentiating suicide attempt from suicidal ideation in major depressive disorder.

BACKGROUND: Suicide remains a critical public health issue, with self-report-based clinical assessments often failing to detect imminent risk. This study aimed to identify objective electroencephalography (EEG)-based neurobiological markers for differentiating a suicide attempt (SA) from suicidal ideation (SI) using EEG microstate and microstate-based functional connectivity (FC) analyses. METHODS: From 2017 to 2020, this study enrolled 130 medication-naïve major depressive disorder patients (68 SA, evaluated within 7 days of the attempt; 62 SI) at Soonchunhyang University Cheonan Hospital. Resting-state EEG data were analyzed using microstate analysis to explore temporal dynamics of brain topography and microstate-based FC to assess connectivity in theta, alpha, and beta bands. Correlations between EEG features and psychological measures (e.g., suicidal ideation, depression, emotion regulation) were examined. RESULTS: Compared with the SI group, the SA group showed a marginally lower frequency of occurrence for microstates A (auditory/language processing) and B (visual processing) (p = 0.078 for both). The SA group demonstrated significantly higher alpha-band FC during microstate E (linked to the default mode network (DMN)) for several electrode pairs (e.g., F7-C5, p = 0.009; FC5-C5, p = 0.005). The SA group also exhibited marginally higher FC in the alpha band during microstates C (DMN-related) and B, and in the theta band during microstate E. A subsequent within-group analysis revealed that in the SI group, alpha-band FC during microstate E positively correlated with scores for difficulties in emotion regulation (r = 0.433, p = 0.017). LIMITATIONS: Findings are limited by potential physiological confounds in the SA group and by the limited anatomical specificity inherent in sensor-space EEG analysis. CONCLUSION: EEG microstate dynamics and microstate-based FC differ between patients with SA and SI. Specifically, enhanced alpha-band connectivity during microstate E in the SA group potentially reflects condition-specific DMN functions. These EEG-based measures show promise as objective markers that complement clinical suicide risk assessment and inform early intervention strategies.