Individualized Spectral Features in First-episode and Drug-naïve Major Depressive Disorder: Insights from Periodic and Aperiodic EEG Analysis.

Background: The detection of abnormal brain activity plays an important role in the early diagnosis and treatment of major depressive disorder (MDD). Recent studies have shown that the decomposition of the electroencephalography (EEG) spectrum into periodic and aperiodic components is useful for identifying the drivers of electrophysiologic abnormalities and avoiding individual differences.

Methods: This study aimed to elucidate the pathologic changes in individualized periodic and aperiodic activities and their relationships with the symptoms of MDD. EEG data in the eyes-closed resting state were continuously recorded from 97 first-episode and drug-naïve patients with MDD and 90 healthy control (HC) participants. Both periodic oscillations and aperiodic components were obtained via the "fitting oscillations and one-over f" (FOOOF) algorithm and then used to compute individualized spectral features.

Results: MDD patients presented higher canonical alpha and beta band power but lower aperiodic-adjusted alpha and beta power. Furthermore, we found that alpha power was strongly correlated with the age of patients but not with disease symptoms. The aperiodic intercept was lower in the parietal‒occipital region and was positively correlated with the Hamilton Depression Rating Scale (HAMD) score after accounting for age and sex. In the asymmetry analysis, alpha activity appeared asymmetrical only in the HC group, whereas aperiodic activity was symmetrical in both groups.

Conclusions: The findings of this study provide insights into the role of abnormal neural spiking activity and impaired neuroplasticity in MDD progression and suggest that the aperiodic intercept in resting-state EEG may be a potential biomarker of MDD.