Heightened effective connectivity of DLPFC-mPFC and DLPFC-ACC circuits in major depressive disorder with suicidal ideation: evidence from a TMS-EEG study

Abstract

Major depressive disorder (MDD) with suicidal ideation (SI) significantly impacts global health. Suicidal ideation is associated with alterations in brain network connectivity, yet the effective connectivity from the dorsolateral prefrontal cortex (DLPFC) to functional network nodes remains poorly understood. This study utilizes transcranial magnetic stimulation-electroencephalography (TMS-EEG) to investigate DLPFC connectivity and cortical excitability changes, providing insights into the neurobiological mechanisms and potential treatments for MDD with SI. This study recruited 166 patients with MDD and 61 healthy controls. The TMS-EEG technique was used to assess effective connectivity based on abnormal time-locked TMS evoked potentials (TEPs). Suicidal ideation was assessed using the suicidality module of the Mini International Neuropsychiatric Interview (MINI), and participants were classified into suicidal ideation (SI) and non-SI (NSI) groups based on the presence of active ideation. Subgroup analysis evaluated significant current scattering (SCS) in DLPFC-related circuits through source localization, with multiple functional networks defined as downstream regions of interest. TEP analysis at the F3 electrode revealed no significant differences between the MDD and HC groups across components. However, the SI group exhibited increased N100 amplitudes compared to the NSI group (uncorrected) and healthy controls. Source-level brain network analysis showed that the SCS of the DLPFC-mPFC and DLPFC-ACC circuits was significantly greater in the SI group than in the NSI and control groups. After controlling for age, logistic regression analysis indicated a significant association between these connectivity patterns and the presence of suicidal ideation. MDD patients with suicidal ideation exhibit altered cortical inhibition and enhanced effective connectivity between the DLPFC and key brain regions, such as the ACC and mPFC. These exploratory findings contribute to a deeper understanding of the neural circuitry involved in suicidal ideation.

Fig. 1. Overview of experimental design and data analysis.

A Schematic diagram of the experimental procedure. All participants underwent clinical assessment (anxiety, depression) and TMS-EEG testing. After data collection was completed, patients were categorised into two subgroups with or without suicidal ideation based on interview. HC, health controls; NSI, Major depressive disorder with no suicidal ideation; SI, Major depressive disorder with suicidal ideation. B Process framework for TMS-EEG data analysis.

Fig. 2. Sensor-level analysis of subgroups.

A TMS evoked potentials (TEPs) recorded at the F3. B Comparison of the mean amplitude of TEPs between SI, NSI, and healthy controls groups.

Fig. 3

Comparison of SCS between stimulation target (F3/LDLPFC) and ROI nodes of each brain network, and then the inclusion of variables that differed significantly between the SI and NSI groups in the regression model for prediction of the presence or absence of suicidal ideation.