Application of HARDI to assess optimal coil orientation in neuronavigated TMS of the motor cortex

50 

Abstract Submission 

Constanze Ramschütz¹, Andrey Zhylka², Silas Preis³, Sandro Krieg⁴, Haosu Zhang⁴, Claus Zimmer³, Bernhard Meyer⁵, Nico Sollmann⁶, Severin Schramm⁷

¹Department of Diagnostic an Internventional Neuroradiology, Technical University of Munich, Munich, Germany, ²Surgical Department, Eindhoven University of Technology, Eindhoven, Netherlands, ³Department of Neuroradiology, Klinikum Rechts der Isar, Technical University of Munich, Munich, Germany, ⁴Department of Neurosurgery, Universitätsklinikum Heidelberg, Heidelberg, Germany, ⁵Department of Neurosurgery, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany, ⁶Dep. of Diagnostic and Interventional Radiology, University Hospital Ulm, Munich, Germany, ⁷Department of Diagnostic and Interventional Neuroradiology, Technical University of Munich, Munich, Germany

Constanze Ramschütz  
Department of Diagnostic an Internventional Neuroradiology, Technical University of Munich
Munich, Germany

Andrey Zhylka  
Surgical Department, Eindhoven University of Technology
Eindhoven, Netherlands

Silas Preis  
Department of Neuroradiology, Klinikum Rechts der Isar, Technical University of Munich
Munich, Germany

Sandro Krieg  
Department of Neurosurgery, Universitätsklinikum Heidelberg
Heidelberg, Germany

Haosu Zhang  
Department of Neurosurgery, Universitätsklinikum Heidelberg
Heidelberg, Germany

Claus Zimmer, Prof. Dr.  
Department of Neuroradiology, Klinikum Rechts der Isar, Technical University of Munich
Munich, Germany

Bernhard Meyer, Prof. Dr.  
Department of Neurosurgery, Klinikum rechts der Isar, Technical University of Munich
Munich, Germany

Nico Sollmann, PD Dr. Dr. med  
Dep. of Diagnostic and Interventional Radiology, University Hospital Ulm
Munich, Germany

Severin Schramm, Dr. med. sci  
Department of Diagnostic and Interventional Neuroradiology, Technical University of Munich
Munich, Germany

Transcranial magnetic stimulation (TMS) is a modality for noninvasive brain stimulation with rapidly growing diagnostic (1) and therapeutic (2) applications. While used in the treatment of conditions such as depression or neuropathic pain, results of neuromodulation protocols are known to be heterogeneous between individuals and centers (3). Additionally, the neurophysiological processes underlying TMS effects are still insufficiently understood (3). One factor that might contribute to optimized results of TMS is the orientation of the stimulating coil in relation to cortical and subcortical anatomy. We present preliminary results from healthy participants in whom we investigated the role of intragyral fiber orientations in terms of TMS effects on the motor cortex.

20 healthy participants (average age: 27 years, 10 females) underwent high angular resolution diffusion imaging (HARDI; 60 gradient directions modeled on a sphere [b-value=1500 s/mm2] and 6 interleaved b0 volumes) and T1-weighted (T1w) imaging at 3 Tesla. The T1w images were used for neuronavigated TMS (nTMS). Up to three nTMS sessions separated by at least 14 days were conducted per subject to assess the robustness of the optimal coil orientation. In each nTMS session, we elicited 140 motor-evoked potentials (MEPs) from the abductor pollicis brevis muscle hotspot using 7 different coil orientations (30° - 150° relative to the longitudinal axis of the precentral gyrus, in 20° steps) on the dominant hemisphere. The MEPs were analyzed regarding the influence of coil orientation on MEP amplitudes. Additionally, HARDI data were corrected for signal drift, motion, echo planar imaging /EPI), and eddy current (EC) distortions and coregistered to the T1w imaging used in the navigation of TMS via ExploreDTI (4). Herein, constrained spherical deconvolution (CSD) truncated at maximum harmonic order L-max=8 was used to model fiber orientation distributions (FODs). These were visualized in relation to the observed optimal stimulation direction (Figure 1).

In total, 6720 individual MEPs were analyzed. The MEP amplitude correlated significantly with coil orientation (rho = -0.39, p << 0.0001; Figure 2). A random effects regression model predicted a deviation from the optimal orientation (defined by highest mean MEP amplitude) by 20 degrees to lead to MEP decreases of -168 μV (β=-168, t=-28.97). On a qualitative level, when comparing the optimal stimulation direction to HARDI-derived fiber orientations, we noticed a notable parallel alignment of FODs to the e-field direction yielding maximal MEPs (Figure 1).

Coil orientation during TMS in relation to gyral anatomy significantly modulates motor responses. HARDI-derived fiber orientation imaging could aid in predicting optimal coil orientation in non-motor areas, e.g. for therapeutic TMS applications.

Non-invasive Magnetic/TMS ¹

Cortical Anatomy and Brain Mapping

White Matter Anatomy, Fiber Pathways and Connectivity

Diffusion MRI ²

Cortex

ELECTROPHYSIOLOGY

Motor

Tractography

Transcranial Magnetic Stimulation (TMS)

WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC