Four electric field modeling methods of Dosing Prefrontal Transcranial Magnetic Stimulation (TMS): Introducing APEX MT dosimetry

Abstract

•Electric field (E-field) TMS dosing is an emerging approach, but is not yet optimized or widely adopted.•Here we propose four methods of E-field TMS dosing, including APEX MT.•Motor threshold and fixed threshold E-field dosing had large motor vs. prefrontal E-field variances of 16.9-20.7%.•APEX MT combines motor threshold and E-field modeling to more optimally dose prefrontal TMS.•APEX MT reduced the within-subject motor vs. prefrontal E-field variance to 0%. We recently performed electric field (E-field) modeling on 38 structural magnetic resonance imaging (MRI) scans in the motor and prefrontal cortices to reexamine the widely used 120% resting motor threshold (rMT) dosing technique(1). Using the finite element method (FEM) to simulate electromagnetic currents passing through tissues with different conductivities (i.e. skin, bone, cerebrospinal fluid, grey matter, and white matter) [[2]Saturnino G.B. Puonti O. Nielsen J.D. Antonenko D. Madsen K.H. Thielscher A. SimNIBS 2.1: a comprehensive pipeline for individualized electric field modelling for transcranial brain stimulation.in: Makarov S. Horner M. Noetscher G. Brain and hum bod model 2018. Springer Copyright, Cham (CH)2019: 3-25Crossref Google Scholar], we found that TMS would need to be applied at an average of 133.5% of resting motor threshold (rMT) over the left prefrontal cortex to produce equivalent E-fields as 100% rMT stimulation over the left primary motor cortex, with large interindividual variability (range = 79.9–247.5% rMT) [[1]Caulfield K.A. Li X. George M.S. A reexamination of motor and prefrontal TMS dosing in tobacco use disorder: time for personalized electric field TMS dosing?.Clin Neurophysiol. 2021; Crossref Scopus (1) Google Scholar]. E-field dosing may be a useful method of reducing cortical E-field variations([3Caulfield K.A. Badran B.W. DeVries W.H. Summers P.M. Kofmehl E. Li X. et al.Transcranial electrical stimulation motor threshold can estimate individualized tDCS dosage from reverse-calculation electric-field modeling.Brain Stimul. 2020; 13: 961-969Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar, 4Caulfield K.A. Badran B.W. Li X. Bikson M. George M.S. Can transcranial electrical stimulation motor threshold estimate individualized tDCS doses over the prefrontal cortex? Evidence from reverse-calculation electric field modeling.Brain Stimul. 2020; 13: 1150-1152Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar, 5Caulfield K.A. Indahlastari A. Nissim N.R. Lopez J.W. Fleischmann H.H. Woods A.J. et al.Electric field strength from prefrontal transcranial direct current stimulation determines degree of working memory response: a potential application of reverse-calculation modeling?.Neuromodulation. 2020; Crossref Scopus (7) Google Scholar, 6Gomez L.J. Dannhauer M. Koponen L.M. Peterchev A.V. Conditions for numerically accurate TMS electric field simulation.Brain Stimul. 2020; 13: 157-166Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar, 7Deng Z.-D. Liston C. Gunning F.M. Dubin M.J. Fridgeirsson E.A. Lilien J. et al.Electric field modeling for transcranial magnetic stimulation and electroconvulsive therapy. Brain and hum bod model. Springer, Cham2019: 75-84Google Scholar]). However, there is currently no widely adopted E-field dosing approach or consensus on the optimal E-field dosing threshold. Here we investigate four E-field modeling methods that could be used to prospectively dose prefrontal TMS, demonstrating the benefits and drawbacks of each approach through modeling on our 38 previously acquired scans [[8]Li X. Hartwell K.J. Henderson S. Badran B.W. Brady K.T. George M.S. Two weeks of image-guided left dorsolateral prefrontal cortex repetitive transcranial magnetic stimulation improves smoking cessation: a double-blind, sham-controlled, randomized clinical trial.Brain Stimul. 2020; 13: 1271-1279Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar](Fig. 1A–D; Supplementary Section 1 for E-field modeling methods). We propose a new method of prefrontal TMS dosing that combines measured rMT and E-field modeling, which we call “A Personalized E-field X Motor Threshold” (APEX MT) dosing(Method 4; Fig. 1D). As detailed below, APEX MT has the advantages of determining a personalized E-field threshold and producing identical prefrontal and motor E-fields. Thus, APEX MT reduces the possibility of under- or over-dosing compared to the widely utilized 120% rMT approach or fixed intensity E-field dosing. Method 1 (Application of 133.5% rMT; Fig. 1A) represents the prospective application of our previous E-field modeling data over the motor and prefrontal cortices(1). As a blunt instrument based on prior group level data, Method 1 relies on our previously determined group average of 133.5% rMT for equivalent prefrontal E-fields and could be applied prospectively following a rMT acquisition. This approach does not require any MRI scanning, additional resources, or cost. However, it simply updates the 120% rMT dosing schema without any further consideration for interindividual variability. As such, the average E-field deviation between 100% rMT motor stimulation and 133.5% rMT prefrontal stimulation remains relatively large; in these 38 participants, there was an average within-individual difference between the motor and prefrontal cortices of 20.6% (SD = 17.5%, range = 0.1–85.4%). Method 2 (Regression of rMT x Prefrontal E-Field; Fig. 1B) builds upon our prior results, using a regression to determine the relationship between the measured rMT over the motor cortex and prefrontal E-field at the cortical target. These values significantly correlated, R2 = 0.34, p < 0.001. The regression formula could be used prospectively, by plugging in a new participant's rMT value, to estimate the prefrontal E-field using Equation 1: PrefrontalElectricField=1.43∗ rMT + 38.3Since E-field models are linearly scalable, the TMS intensity could then be adjusted based on the desired prefrontal E-field at the cortical level. For instance, if the participant's motor threshold estimated a prefrontal E-field of 90V/m, this could be scaled up to 120V/m by stimulating at 133.3% of the rMT stimulation intensity. This regression approach somewhat but not fully accounts for interindividual differences between the regression-predicted E-field and actual prefrontal E-field in the 38 participants, with an average within-individual difference between the motor and prefrontal cortices of 16.9% (SD = 13.2%, range = 0.8–53.8%). Method 3 (Fixed Prefrontal E-field Intensity; Fig. 1C) prospectively doses TMS at a fixed E-field magnitude across participants, typically at an intensity presumed to be suprathreshold for neuronal firing such as 100V/m [[9]Salvador R. Wenger C. Miranda P.C. Investigating the cortical regions involved in MEP modulation in tDCS.Front Cell Neurosci. 2015; 9Google Scholar]. This approach uses structural MRI scans for modeling but does not measure rMT. Some researchers have already utilized this fixed E-field dosimetry approach, particularly in the application of single pulses of TMS [[10]Casali A.G. Gosseries O. Rosanova M. Boly M. Sarasso S. Casali K.R. et al.A theoretically based index of consciousness independent of sensory processing and behavior.Sci Transl Med. 2013; 5 (198ra05)Crossref PubMed Scopus (494) Google Scholar]. While this method nicely utilizes E-field modeling to account for individualized tissue compositions and conductivities, the optimal E-field threshold is currently unclear and may differ due to age, population, or other factors. Moreover, it is possible that the precise E-field magnitude required to activate neurons may differ between individuals, which is not taken into account in this approach due to lack of an rMT measurement. Thus, a drawback of Method 3 is that the individual functional neuronal activation threshold is not measured (no rMT). In contrast, rMT dosing measures the within-individual stimulation intensity required to activate cortical tissue, albeit within the motor system where it is easily observable. In Fig. 1C, we show the E-field magnitude difference from 100% rMT stimulation over the left motor cortex and individual stimulation intensities to produce the group average E-field in the prefrontal region of interest of 158.2V/m. Due to this selected E-field threshold, the average within-individual deviation between E-fields from 100% rMT motor stimulation and prefrontal stimulation at 158.2V/m was 20.7% (SD = 14.5%, range = 0.2–66.4%). Therefore, the E-field deviation between the motor and prefrontal cortices using a set E-field intensity for every participant remains relatively high. We propose that APEX MT (Method 4; Fig. 1D) could be used to optimally combine the benefits of rMT measurements and E-field modeling simulations to prospectively dose prefrontal TMS. This approach takes five steps:Step 1: Acquire a structural MRI scan for E-field modeling.Step 2: Determine the motor hotspot and acquire the motor threshold.Step 3: Perform E-field modeling at the motor hotspot at the 100% rMT stimulation intensity. This modeling determines the individualized cortical motor threshold E-field that produces neuronal firing.Step 4: Calculate a prefrontal E-field model at the stimulation target using a set stimulation intensity (e.g. 50% machine output TMS). The chosen prefrontal stimulation intensity can be any magnitude, since it is used to determine the input-output curve in the prefrontal cortex in Step 5 and E-field model inputs and outputs scale linearly.Step 5: Use Equation 2 to determine the required stimulation intensity to produce an equivalent cortical E-field at the prefrontal target as 100% rMT stimulation in the motor cortex:IndividualizedPrefrontalDose=PrefrontalMachineOutputPrefrontalE–field∗MotorThresholdE–fieldUsing APEX MT ensures that there is 0% deviation between motor and prefrontal E-fields for each individual (Fig. 1D; Supplementary Section 2). For example, a participant who has a motor threshold E-field of 120V/m (Step 2, Step 3) and a prefrontal E-field of 90V/m from 50% machine output (Step 4) would have a prefrontal dose of: 50%MachineOutput90V/m∗120V/m = 66.7% Machine Output (Step 5). Thus, APEX MT combines the ability of rMT to determine a personalized E-field threshold for neuronal firing within each person's primary motor cortex, while also accounting for individualized tissue composition differences between the motor and prefrontal cortices via E-field modeling that is not taken into account with rMT alone. This combined approach avoids the potential pitfalls of using a fixed E-field intensity (Method 3), which could be prone to under- or over-dosing depending on the uniform E-field threshold applied across every person. In contrast, APEX MT individualizes the E-field threshold based on the motor threshold E-field, and applies this personalized E-field threshold forward in the prefrontal cortex. While APEX MT requires the cost of MRI scanning and time required to perform offline E-field modeling, current and forthcoming real-time integration of E-field modeling into neuronavigation software may help to move this method into research or clinical applications in the near future. Moreover, since APEX MT ensures that there is no deviation between the motor and prefrontal E-fields, this approach is potentially safer than traditional 120% rMT dosing [[1]Caulfield K.A. Li X. George M.S. A reexamination of motor and prefrontal TMS dosing in tobacco use disorder: time for personalized electric field TMS dosing?.Clin Neurophysiol. 2021; Crossref Scopus (1) Google Scholar](See Supplementary Section 2 for safety discussion). In summary, E-field modeling can play a valuable role in prefrontal TMS dosing, and we propose that APEX MT's approach of combining motor threshold measurements and E-field modeling may be an optimized prefrontal dosing technique that reduces under- or over-dosing. This work was supported by an U.S. NIH / NIDA grant (number: 1R21DA036752-01A1 ) to Dr. Xingbao Li, the National Center of Neuromodulation for Rehabilitation (NC NM4R; P2CHD086844 ), and the Center for Biomedical Excellence (COBRE) in Stroke Recovery ( P20GM109040 ). We confirm that there are no known conflicts of interest associated with this publication and there was no financial support for this work that could have influenced its outcome. The following is the Supplementary data to this article: Download .docx (.04 MB) Help with docx files Multimedia component 1