The Impact of TMS Intensity and Location on Neural Activation: An iTBS-fMRI Study

1329 

Abstract Submission 

Kai-Yen Chang¹, Martin Tik², Yuki Mizutani-Tiebel¹, Lucia Bulubas¹, Frank Padberg¹, Daniel Keeser¹

¹Department of Psychiatry and Psychotherapy, University Hospital LMU, Munich, Germany, ²Center for Medical Physics and Biomedical Engineering, Medical University of Vienna, Vienna, Austria

Kai-Yen Chang  
Department of Psychiatry and Psychotherapy, University Hospital LMU
Munich, Germany

Martin Tik  
Center for Medical Physics and Biomedical Engineering, Medical University of Vienna
Vienna, Austria

Yuki Mizutani-Tiebel  
Department of Psychiatry and Psychotherapy, University Hospital LMU
Munich, Germany

Lucia Bulubas  
Department of Psychiatry and Psychotherapy, University Hospital LMU
Munich, Germany

Frank Padberg  
Department of Psychiatry and Psychotherapy, University Hospital LMU
Munich, Germany

Daniel Keeser  
Department of Psychiatry and Psychotherapy, University Hospital LMU
Munich, Germany

The determination of intermittent theta burst stimulation (iTBS) intensity relies on motor threshold (MT) measured in the primary motor cortex (M1). Despite the common practice of using 80% to 120% of M1 resting MT for prefrontal cortex stimulation, the transferability and adequacy of this approach remain unclear. Utilizing concurrent TMS-fMRI, we investigate the impact of subthreshold intensities (80% rMT) and low stimulation intensity (40% rMT) on the left DLPFC. We also include 80% rMT over the left M1 as a reference. By comparing these conditions within the same subjects, we seek to understand how varying TMS intensities influence neural activation in the prefrontal region and whether left M1 intensity induces similar patterns in the left DLPFC. This exploration is crucial for establishing the appropriate stimulation intensity for clinical treatment, shedding light on the differentiated relationship between stimulation intensity and neuronal activation in the prefrontal brain region.

All participants signed informed consent approved by the local LMU ethical committee. 18 healthy participants (10 males, mean age = 26.4, SD = 3.1) were included. We randomized three conditions for each participant: 80% or 40% rMT intensity over the left DLPFC (-38, 44, 26) (1) and 80% rMT over the left M1. In these sessions, we acquired a structural and interleaved iTBS-fMRI sequence. Apart from the standard MRI data preprocessing (2), we used independent component analysis (ICA) to remove general noise such as ventricular and motion artifacts in addition to potential coil artifacts in the fMRI data as suggested in (3). To establish a correlation between simulated electric field (E-field) and fMRI BOLD activation in each subject under the conditions of 80% rMT in the DLPFC and M1, we adopted predefined targets for the left DLPFC and M1 (-37, -21, 58) (4), and calculated the mean E-field magnitude on SimNIBS 4.0 and the mean beta value from fMRI data within a 10mm radius of region of interest.

At 80% rMT over left DLPFC, significantly increased BOLD was observed under the stimulated location, M1, anterior cingulate gyrus (ACC), anterior insula, and bilateral auditory cortices. It is noteworthy that the activation of the bilateral ACC and anterior insula represents a classic marker for the so-called salience network. While lower iTBS intensity resulted only in the presence of BOLD activity on the contralteral right DLPFC, bilateral auditory cortices and ACC. In M1 condition, activations were observed in the bilateral primary motor cortices, right DLPFC, auditory cortex, and primary somatosensory cortex. In the subcortical regions, the BOLD activations were found in putamen region and left caudate region. The correlation analysis examined E-field magnitude and fMRI beta values in two conditions: DLPFC 80% rMT and M1 80% rMT. A weak positive correlation (r = 0.31) was found in the DLPFC condition, while the M1 condition showed a strong positive correlation (r = 0.65).

The results of our study revealed that focal BOLD activations were found under stimulated region in the left DLPFC after ICA denoising was applied. This observation suggests that artifacts from the TMS coil or TMS-MRI hardware coupling were removed. Moreover, it revealed activation in the salience network, containing the anterior insula and anterior cingulate gyrus, aligning with Hawco et al.'s findings (5). Conversely, the 40% rMT DLPFC condition did not exhibit this specific pattern; only the right DLPFC showed activity, potentially indicating interhemispheric compensation. The correlation analysis suggests that higher E-field magnitudes in the cortex generally correlate with higher BOLD signals in fMRI. Due to the higher complexity of the DLPFC, the prediction of simulation results becomes more difficult, highlighting the challenge in predicting outcomes under the same TMS intensity for DLPFC and M1.

TMS ²

Psychiatric (eg. Depression, Anxiety, Schizophrenia)

Activation (eg. BOLD task-fMRI) ¹

BOLD fMRI

MRI

Psychiatric Disorders

Transcranial Magnetic Stimulation (TMS)

Other - Concurrent TMS-fMRI