Impact of Electrode Positioning on MEP Variability and Reliability: Implications for Optimal Pulse Number in TMS Studies
DOI: Pending DOI
Synopsis
Motor evoked potentials (MEPs) obtained through transcranial magnetic stimulation (TMS) can present substantial variability depending on surface electrode positioning (7). This study investigated the impact of three surface electrode-montages (Pseudomopolar over innervation zone [Piz] proposed by Garcia et al. (2017); Pseudomopolar over muscle belly [Pmb] proposed by the International Federation of Clinical Neurophysiology (IFCN) and Modified Bipolar over muscle belly [MBmb] proposed by Munneck et al. (2018)) (Figure 1) on the variability, reliability of motor evoked potentials (MEPs), and the optimal number of transcranial magnetic stimulation (TMS) pulses. Eight healthy subjects underwent MEP acquisition from the biceps brachii simultaneously using the three surface electrode montages. Peak-to-peak value (MEPp-p), coefficient of variation (CV), cumulative mean, and confidence intervals were extracted from the MEP data for analysis. The Pmb montage showed higher MEPp-p, lower CV, and greater inter-subject consistency, representing the most robust configuration. The MBmb montage showed faster stabilization of the cumulative mean but higher variability, suggesting lower reliability. The Piz montage showed intermediate behaviour but greater sensitivity to inter-subject variability and external influences.
Background
Motor evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS) are widely used to assess corticospinal excitability and motor pathway integrity (1,8,14). However, their application is limited by high intra- and inter-subject variability, underscoring the need for methodological standardization (3–5). Among the contributing factors, surface electrode positioning plays a critical role in surface electromyographic (sEMG) signal acquisition and significantly influences MEP peak-to-peak (MEPp-p) values and variability (6).
MEP variability is influenced by both physiological and methodological factors, including fluctuations in cortical excitability, TMS coil positioning, and sEMG acquisition parameters (4,5,10). Thus, improving methodological rigour is essential for enhancing reproducibility. This study aimed to evaluate how different surface electrode positioning strategies affect MEPp-p, variability, and temporal stability, and to estimate the minimum number of pulses required for reliable measurements.
Methods
Eight healthy subjects participated in accordance with established safety guidelines for TMS application (3,11). MEPs were recorded from the biceps brachii using surface electromyography (sEMG) (EMG System do Brasil, EMG 410C; 4 channels; gain: 2000; sampling rate: 3.5 kHz; band-pass filter: 20–500 Hz; 12-bit A/D converter) in both upper limbs. Participants were seated comfortably, with muscles at rest and the upper limbs supported on a surface, positioned at approximately 90° of elbow flexion. Three surface electrode montages were evaluated simultaneously: Pseudomopolar over the innervation zone [Piz]; Pseudomopolar over the muscle belly [Pmb]; and Modified bipolar over the muscle belly [MBmb]. Skin preparation included trichotomy when necessary and cleaning with neutral soap to reduce impedance. The innervation zone (IZ) was initially estimated from an anatomical atlas and subsequently confirmed using an electronic acupuncture pen (3.7 V; 300 mA ± 50 mA; pulse frequency: 0.01–300 Hz). Disposable Ag/AgCl electrodes were used, and interelectrode distances were recorded. A reference electrode was placed over C7. TMS was delivered using a Magstim 200² with a figure-of-eight coil positioned at 45° to target the biceps brachii hotspot in relation to the sagittal plane. Stimulation intensity was set at 120% of resting motor threshold (IFCN-based). Approximately 30 single pulses were delivered at 5–10 s intervals. Neuronavigation was used to ensure consistent coil positioning. sEMG signals were processed using Signal Hunter (MATLAB), with semi-automatic latency detection. Data were analyzed in Python (https://colab.research.google.com/drive/1rXrZ7MaMfiPQjC3ZWwzjJs6tBTyoqoC3?usp=sharing), including mean, standard deviation, coefficient of variation (CV), cumulative mean, and 95% confidence intervals (CI). Temporal stability was assessed through cumulative mean convergence and probability of remaining within the CI.
Results
MEPp-p varied across subjects and electrode configurations but followed a consistent pattern. The Pmb showed the highest MEPp-p values (0.31–0.52 mV), followed by Piz (0.20–0.41 mV), while the bipolar configuration MBmb showed the lowest ones (0.14–0.34 mV) (Figure 2), consistent with previous findings on electrode positioning (6,7).
Variability analysis revealed that Pmb consistently exhibited the lowest CV values (30–50%), indicating greater signal stability. Piz showed intermediate variability (40–80%), whereas MBmb demonstrated the highest instability, with CV values exceeding 100% in some cases. These findings reinforce that methodological factors significantly influence MEP variability.
Cumulative mean analysis indicated that stabilization occurred at different time points depending on the configuration. MBmb stabilized earlier ~6 pulses, Piz at ~8 pulses, and Pmb required ~12 pulses. These findings align with prior evidence suggesting that the number of pulses required for reliable MEP estimation varies depending on signal characteristics.
Confidence interval analysis showed that Pmb had higher MEPp-p values and more consistent CI ranges across participants. Piz maintained intermediate, predictable behaviour, whereas MBmb exhibited marked intersubject variability.
The probability of cumulative mean values remaining within the 95% CI varied substantially across configurations. Pmb showed high consistency in some subjects (>90%) but reduced performance in others, indicating subject-dependent variability. MBmb exhibited highly unstable behaviour, whereas Piz showed intermediate performance.
Overall, the surface electrode montage significantly influenced MEPp-p, variability, and temporal stability, directly affecting the number of pulses required for reliable MEP estimation.
Discussion
These findings demonstrate that electrode positioning is a critical determinant of MEP reliability, supporting evidence that methodological factors substantially contribute to signal variability (4,5). The Pmb montage (proposed by the IFCN) provided the best balance between amplitude and variability, supporting its use in clinical and research applications (3,11,14).
Although the MBmb configuration proposed by Munneck et al. (2018) and the most widely used montage showed faster stabilization, their high variability limits their reliability, indicating that rapid convergence does not necessarily reflect stable signal acquisition. These findings highlight that MEPp-p variability is not solely neurophysiological but also strongly influenced by acquisition parameters, including electrode positioning, coil type, coil orientation, and other methodological factors (4,5,14). On the other hand, the Piz showed intermediate results; the present surface electrode montage exhibits rapid stabilization and high MEPp-p values, but shows greater consistency across subjects.
Importantly, the optimal number of TMS pulses depends on electrode configuration and signal characteristics, reinforcing the need for protocol-specific standardization (4,7,9).
These results contribute to improving reproducibility and support the development of standardized guidelines for MEP acquisition in neuromodulation research.
Disclosures
The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior -Brasil (CAPES) -Finance Code 001 and the Universidade Federal de Juiz de Fora (UFJF).
References
1. Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of human motor cortex. The Lancet. 1985;1:1106–1107.
2. Barbero M, Merletti R, Rainoldi A. Atlas of Muscle Innervation Zones. Springer; 2012.
3. Biabani M, et al. Minimal number of TMS trials for reliable assessment. Neuroscience Letters. 2018;674:94–100.
4. Chang WH, et al. Optimal number of stimuli for reliable MEPs. Clinical Neurophysiology. 2016;127:417–423.
5. Chipchase L, et al. Checklist for methodological quality in transcranial magnetic stimulation studies. Clinical Neurophysiology. 2012;123:1698–1704.
6. Farina D, Merletti R, Enoka RM. Neural strategies from surface electromyography. Journal of Applied Physiology. 2014;117:1215–1230.
7. Garcia MAC, Souza VH, Vargas CD. Can the recording of motor potentials evoked by transcranial magnetic stimulation be optimized? Frontiers in Human Neuroscience. 2017;11:e413. doi:10.3389/fnhum.2017.00413
8. Garcia MAC, Vieira TMM. Surface electromyography: why, when and how to use it. Revista Andaluza de Medicina del Deporte. 2011;4:17–28.
9. Garcia MAC, et al. Motor potential evoked by transcranial magnetic stimulation depends on the placement protocol of recording electrodes: A pilot study. Biomedical Physics & Engineering Express. 2020. doi:10.1088/2057-1976/ab950a
10. Garcia MAC, et al. Surface electrode placement determines MEP magnitude. Biomedical Signal Processing and Control. 2023;84.
11. Groppa S, Oliviero A, Eisen A, Quartarone A, Cohen LG, Mall V, et al. IFCN diagnostic transcranial magnetic stimulation guidelines. Clinical Neurophysiology. 2012;123:858–882.
12. Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Recommendations for surface electromyography sensors and sensor placement procedures. Journal of Electromyography and Kinesiology. 2000;10:361–374.
13. Osnabruegge M, et al. Reliability of motor evoked potentials. Frontiers in Human Neuroscience. 2023.
14. Rossi S, Hallett M, Rossini PM, Pascual-Leone A. Safety and application guidelines for transcranial magnetic stimulation. Clinical Neurophysiology. 2009;120:2008–2039.
15. Rossini PM, Burke D, Chen R, Cohen LG, Daskalakis Z, Di Iorio R, et al. Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: Basic principles and procedures for routine clinical and research application. Clinical Neurophysiology. 2015;126:1071–1107.
16. Yoo EK, et al. Number of pulses required for reliable corticospinal excitability assessment. Frontiers in Human Neuroscience. 2017;11.