Temporal Dynamics of EEG microstates in Elderly Depression and the Comorbidity of Parkinsonism

620 

Abstract Submission 

Yen-Liang Liu¹, Chiu-Jung Huang¹, Wei-Chung Mao², Tung-Ping Su², Li-Fen Chen^1,3

¹Institute of Brain Science, National Yang Ming Chiao Tung University, Taipei, Taiwan, ²Department of Psychiatry, Cheng Hsin General Hospital, Taipei, Taiwan, ³Brain Research Center, National Yang Ming Chiao Tung University, Taipei, Taiwan

Yen-Liang Liu  
Institute of Brain Science, National Yang Ming Chiao Tung University
Taipei, Taiwan

Chiu-Jung Huang  
Institute of Brain Science, National Yang Ming Chiao Tung University
Taipei, Taiwan

Wei-Chung Mao  
Department of Psychiatry, Cheng Hsin General Hospital
Taipei, Taiwan

Tung-Ping Su  
Department of Psychiatry, Cheng Hsin General Hospital
Taipei, Taiwan

Li-Fen Chen  
Institute of Brain Science, National Yang Ming Chiao Tung University|Brain Research Center, National Yang Ming Chiao Tung University
Taipei, Taiwan|Taipei, Taiwan

Major depressive disorder (MDD) is a common psychiatric disorder among elderly individuals, with a global prevalence of 31.74% (Zenebe et al., 2021). Late-life depression increases the likelihood of developing neurodegenerative disorders such as Parkinson's disease (PD) (Leentjens et al., 2003). Notably, a recent study indicates that diagnosing prodromal PD in elderly individuals with MDD is challenging because of the resemblance of explicit syndromes between "pseudo-parkinsonism" of depression and PD (Weiss & Pontone, 2019). EEG microstates have been proposed to reveal the spatiotemporal dynamics of large-scale brain networks and applied to identify abnormalities in patients with MDD (Murphy et al., 2020) and PD (Pal et al., 2021). In this study, we utilized EEG microstate analysis to investigate alternation of neural networks in patients with comorbidity of MDD and neurodegenerative disorders. Our objective is to explore the discrimination of dynamic patterns of neural networks between MDD and MDD with Parkinsonism (MDD-PD) patients.

We enrolled 14 patients with MDD, 7 patients with MDD-PD, and 14 healthy elderly individuals. The specific uptake ratio of TRODAT in the striatum of MDD-PD patients was below 0.8. Four-minute resting-state EEG signals with eyes-closed were recorded using a 32-channel cap (Easycap, Herrsching, Germany) and preprocessed by noise removal of eye blinks and electrocardiogram artifacts. The noise-free EEG data were then analyzed using MICROSTATELAB toolbox in EEGLAB (Nagabhushan Kalburgi et al., 2023) to extract five microstate template components for each group. Four parameters of each microstate, including mean duration, frequency of occurrence, time coverage, and transition probability (Murphy et al., 2020), were estimated for each individual. To examine the group differences in these microstate parameters, we conducted a one-way ANOVA with Bonferroni correction to adjust the p-value in the post hoc testing.

The topographies of the microstates template components were similar between groups (Figure 1a). When compared to the healthy controls, the MDD group exhibited a significantly shorter duration in the microstates A (p = 0.002), C (p = 0.008), D (p = 0.01), and E (p = 0.014), as well as higher occurrence in microstates A (p = 0.017), B (p = 0.013), D (p = 0.005), and E (p = 0.006) (Figure 1b,1c). There were no significant differences in microstate coverage between groups (Figure 1d). Figure 2 depicts the probability of microstate transition within and between groups. Specific transition preferences in the MDD group (A to E, B to E, and D to C) were significantly different from the control group (Figure 2b). Furthermore, MDD-PD patients also demonstrated a higher transition probability from microstate D to C (p = 0.005) than MDD patients. There were no significant differences between the MDD-PD and control groups.

Our results reveal increased temporal dynamics in MDD group, which is consistent with previous report that excessive temporal variations in MDD reflect abnormal communications between large-scale brain networks (Long et al., 2020). Within the context of rapid network alterations, transition from microstate D to C could not only represent potential features for late-life depression but also provide insights for early detection of MDD patients with comorbidity of Parkinsonism. There was no significant difference in microstate between the MDD-PD and control groups due to the limited sample size in the MDD-PD population. In conclusion, microstate analysis could serve as an adjunctive tool to uncover different dynamic pattern in MDD and PD.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

EEG/MEG Modeling and Analysis ²

Affective Disorders

Electroencephaolography (EEG)

Other - Parkinson's disease; Microstate