Somatosensory evoked potentials versus magnetospinography and magnetoneurography: is there a need for a new diagnostic modality in brachial plexus, cervical roots and spinal cord neurophysiology?
1Department of Neurology, UMass Chan Medical School, Worcester, MA, USA
2Department of Psychology and Neuroscience, Duke University, Durham, NC, USA
Correspondence to Pegah Afra Department of Neurology, UMass Chan Medical School, Worcester, MA 01655, USA Tel: +1-508-334-2527 Fax: +1-774-442-3687 E-mail: pegah.afra@umassmed.edu
• Received: May 28, 2025 • Revised: June 28, 2025 • Accepted: July 3, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
The electrical signals that pass through neuronal structures generate electrical currents and magnetic fields that can be measured by different electrography and magnetography modalities. In this article, we briefly review the fundamentals of recording bioelectrical currents and biomagnetic fields. This is followed by an analysis of the neurophysiologic substrates of the brachial plexus, cervical roots, and spinal cord, comparing the electrography-based modality of somatosensory evoked potential with the magnetography-based modality of magnetospinography/magnetoneurography (MSG/MNG). We aim to illustrate that MSG/MNG has the potential to address the shortcomings that currently exist in electrography modalities for direct neurophysiologic assessment of the aforementioned neural structures.
In the past several decades, non-invasive structural and anatomical neuroimaging technologies have markedly advanced.1 Their excellent spatial resolution allows direct structural assessment of the spinal cord and, to some extent, its roots and plexi. The noninvasive functional and neurophysiologic assessment of these structures have been limited to electrography modalities with excellent temporal, but low spatial resolution. Somatosensory evoked potentials (SSEPs) can indirectly evaluate dorsal columns, plexi and roots2 and electromyography (EMG)/nerve conduction studies (NCS) can indirectly assess the anterior horns (via EMG) and more proximal segments of spinal nerves (via F responses).3 Despite these available technologies, there are no direct clinical neurophysiological modalities for detailed noninvasive neurophysiological assessments of the cervical spinal cord, dorsal roots, and plexi.3 Magnetospinography/magnetoneurography (MSG/MNG) offers the same excellent temporal resolution as legacy technologies, but with better spatial resolution.4 However, MSG/MNG has only been used as a research tool within a few academic institutions worldwide.5,6
In this article, we briefly review the fundamentals of electrical currents and magnetic fields at the cellular level, followed by principles of electric (SSEP) and magnetic (MSG/MNG) neurophysiologic recordings. This is followed by a more detailed analysis of the neurophysiologic substrates of the brachial plexus, cervical roots and spinal cord. Substrates above this level are beyond the reach of MSG/MNG, while structures below this level fall within an anatomical region that is difficult to assess using SSEPs; therefore, both fall outside the scope of this article. Overall, we aim to illustrate the shortcomings that exist in current SSEP modalities, which MSG/MNG has the potential to address.
NEUROMAGENTIC ACTION POTENTIAL PROPAGATION AT THE CELLULAR LEVEL: ELECTRICAL CURRENTS VERSUS MAGNETIC FIELDS
An understanding of electrical currents and magnetic fields is needed to interpret findings from electrography and magnetography modalities. Briefly, the propagating action potential is a biologically generated ionic current with intra-axonal and transmembrane components.7 The intra-axonal flow generates a magnetic field perpendicular to and clockwise around the intra-axonal flow (as determined by Fleming’s right-hand rule of electromagnetism). Transmembrane currents generate an electric field that consists of an outward current ahead of leading depolarization; followed by an inward current flow between the leading depolarization and trailing repolarization; and finally, an outward current flow after the trailing repolarization.7 This results in two back-to-back dipoles: an initial +/- (leading depolarization), followed by a -/+ (trailing repolarization) (Fig. 1A). These two unequal dipolar sources make a quadrupolar source (+ - - +) that can be depicted as a tripole1 as shown in Fig. 1B.
The sum of electrical currents generated by the nervous system can be recorded via different electrography modalities (electroencephalography, SSEP, NCS, EMG), while the magnetic field components can be recorded by magnetography modalities (magnetoencephalography, MSG/MNG, and magnetomyography).5,8-11 In the following sections, we briefly summarize electrography-SSEP, followed by magnetography-MSG/MNG recording principles and modalities that are applicable to plexi, roots, and spinal cord.
PRINCIPLES OF ACTION POTENTIAL PROPAGATION RECORDING: ELECTROGRAPHY VS. MAGNETOGRAPHY MODALITIES
Electrography principles
The fundamental aspect of electrography modalities is the principle of near- and far-field potentials, as determined by the location of a recording electrode in relation to a signal source. Fig. 2 compares the spatial profile of a dipole in an infinite volume conductor (resulting only in a near-field region) versus a finite long cylindrical volume conductor (resulting in both near- and far-field regions).12 When near-field potentials are recorded, the signal passes under the recording electrodes and decays in voltage by distance, indicating that the recorded electrical field is generated within the source (i.e., the nerve membrane itself).12,13 In the case of far-field potentials, signals are recorded far from the source and are indicative of an electrical field generated in a volume conductor, resulting in a non-declining voltage over the recording region and are therefore termed “non-moving”, “junctional”, or “stationary” potentials.12,13 Stegeman et al.12 arbitrarily defined the far-field region as the region in which the potential decreases <5% over distances that are equal to or greater than 10% of a volume conductor’s maximal dimension (i.e., the region in which the recorded signal is relatively constant or stable).
Stationary far-field potentials are the result of a disturbance in uniformity or homogeneity of nerve propagation. The single nerve fiber is a long cylindrical volume conductor. When an action potential is propagating with uniformity, no far-field components are generated. Uniformity implies that electrical conductivity of the cylinder is homogeneous in the direction of propagation and that the impulse travels with constant velocity. Any disturbance in uniformity will result in the generation of far-field components with non-moving or stationary peaks. The disturbance implies; 1) a change in electrical conductivity of the extracellular medium; 2) a change in the spatial extent or volume geometry of the volume conductor; and 3) a change in anatomical orientation or direction of propagation.14 Clinically, this translates into changes in the size or resistance of the volume conductor around the nerve, the high curvature of the nerve (i.e., a sharp bend in the nerve, or bending of the nerve with an acute angle), and the end of the nerve.15,16
Electrography modalities (SSEPs)
SSEPs are obtained in response to any peripheral nerve stimulation. Usually, the stimuli are electrical square waves of variable rates (1-30 seconds) and durations (10 usec-2 msec), most commonly 100-200 usec. The stimulus intensity is adjusted to obtain a twitch produced in the muscle innervated by the stimulated nerve. This usually requires about 10-20 mA for most individuals. The nerves most stimulated clinically are the median and ulnar in the upper extremity, although the radial, musculocutaneous and digital nerves have also been used.17-19 Neural activity is recorded along the conduction pathway with use of surface electrodes. A bandpass filter of 1-30 Hz (low cut off/high pass filter) and 3 kHz (high cut off/low pass filter) is applied. The average is set for a total sweep duration of 50 msec for upper extremities and inter-sample interval (dwell time) of 200 msec. Usually 1,000-2,000 sweep repetitions are done, although this can vary depending on the amount of noise or artifact.19
Magnetography principles
The intra-axonal flow of the propagating action potential generates a magnetic field, which is unaffected by volume conducting properties of the surrounding tissue.5 Therefore, the magnetic field bypasses near- and far-field problems, although the problem of signal decay with distance remains. When magnetic evoked fields are recorded, intracellular currents (including leading and trailing intra-axonal currents, and inward currents at the depolarization site) are being measured. As a result, there is a more direct correlation between the anatomical location and the signal path itself.4 Therefore, with the same temporal resolution of electrography modalities, this improved spatial resolution can allow more direct localization in the plexi and roots, allowing the signal to be measured as it enters the spinal cord and propagates towards the brainstem.4
Magnetography modalities (MNG/MSG)
After electrical stimulation, the magnetic signal is acquired with magnetosensors (as detailed in the paragraph below). Most magnetography studies have been conducted with an MSG/MNG machine developed by the Kazanawa Institute of Technology in collaboration with Tokyo Medical and Dental University, which has been through several iterations.20-25
The MNG/MSG machine has 44 sensor locations: 40 with vector type superconducting quantum interference device gradiometers (two planar gradiometers and one axial gradiometer) and four with axial second-order gradiometer, resulting in 124 channels.24,25 The orientation of planar-type gradiometers is perpendicular to each other for detection of magnetic field components tangential to the body’s surface, and the axial gradiometers for detection of magnetic field components are radial to the body’s surface.24 Sensors in the current MSG/MNG machine are in a protrusion with a surface curvature (radius 200 mm) that makes it suitable to test spinal cord and peripheral nerve activity. After signal acquirement, post-processing is done with offset removal and digital filtering and applying denoising techniques like dual signal subspace projection, and common mode subspace projection for artifact removal.26,27 Thereafter, unit gain constraint recursively applied null-steering spatial filtering (UGRENS) is used. With a UGRENS beamformer,28 the acquired magnetic data (i.e., the evoked currents) are reconstructed as vectors (with current density and direction). They are converted to pseudocolor maps that can be superimposed onto X-rays.28 Virtual electrodes are placed onto X-rays for current waveform display and then waveform analysis is performed.28
NEUROPHYSIOLOGICAL-NEUROANATOMICAL CORRELATION: ELECTROGRAPHY VS. MAGNETOGRAPHY MODALITIES
SSEP is the electrography modality used to assess the brachial plexus, nerve roots, and spinal cord non-invasively. The potentials are elicited with stimulation of peripheral sensory nerves in the upper extremity and can be reproducibly recorded, peripherally, and over the spine and scalp. Anatomically, they reflect the electrical conduction of action potentials from peripheral nerves and brachial plexus, through the spinal roots and dorsal columns, and ultimately along the medial lemniscal pathways to the somatosensory cortex.2,19
For this article, we review part of the neuroanatomical path of SSEPs’ electrical signal, and its near- and far-field potentials, from the brachial plexus to its postsynaptic target nucleus cuneatus in the lower brainstem (i.e., the upper SSEP plexus, root, and spinal cord components). This is followed by a brief discussion of MNG/MSG findings of the same anatomical regions. The discussion of thalamocortical potentials (N18, P20, N22) that are external to the recording field of MNG/MSG are outside of the scope of this article but can be found elsewhere. 13,29
Upper extremity SSEP vs. MNG/MSG
SSEPs can be elicited with stimulation of any sensory nerve or dermatome. Stimulation of median, ulnar and radial nerves at the level of the wrist or digits are used, with the median nerve at the wrist being the most common,18,19 although the musculocutaneous nerve has also been used.17 Short latency responses to median nerve stimulation arrive at the brachial plexus at 9 msec and reach nucleus cuneatus at 13-14 msec (Fig. 3, Table 1). A summary of SSEP followed by MNG/MSG neurophysiologic events at each site (brachial plexus, cervical spinal root, dorsal column volley, and cervical potentials) are provided below.
The brachial plexus
The propagating brachial plexus volley has both transmembrane and intra-axonal currents. The transmembrane currents (transmembrane outward current flow before the leading depolarization components, transmembrane inward current flow in between leading and trailing intra-axonal currents, and transmembrane outward current flow after the trailing intra-axonal current) contribute mainly to recorded electrical fields.6,30-32 The transmembrane inward current is considered to be the point of depolarization. The intra-axonal currents (with leading depolarization and a trailing repolarization components) are the main contributors to the recorded magnetic fields, although some authors report that the recorded magnetic signal also includes the inward transmembrane currents (i.e., the depolarization point) mainly due to orientation of some of the sensors.30 As such, recording of the depolarization point allows for accurate correlation of magnetic and electric signals.
In SSEP recordings, the near-field potential of the brachial plexus volley (Fig. 3B) can be recorded at approximately 9 msec as a triphasic potential (positive-negative-positive waveform) with its highest amplitude negative peak (N9). When this potential is recorded in the anterior axillary line, it is called the axillary potential, and when it is recorded at the Erb’s point, it is called the Erb’s potential.13,33 The Erb’s points is 2 cm above the midpoint of the clavicle, at the angle between the clavicle and the posterior boarder of the sternocleidomastoid muscle. The axillary potential is immediately followed by the Erb’s point potential,13 both around 9 msec. Although it is known that the N9 potential is generated by sensory fibers, its precise location of generation is unknown.19 The far-field potential is recorded at the scalp (P9) after the axillary potential and immediately before the Erb’s potential.13,34,35 Hashimoto’s hypothesis posits that volume currents circulating in the trunk, and therefore volume conductor geometry, are the source of far field P9 generation. At approximately 10 msec, the N10 potential is a stationary/far-field potential that can be recorded in the neck by electrodes placed along the anterior boarder of the sternocleidomastoid muscles and concurrently at the scalp19 with amplitude maxima over the anterior lower neck. This stationary far-field potential is generated by the antidromic motor volley of interosseus nerve.36,37
The brachial plexus has been recorded with MNG/MSG.29-31 The reconstructed neural activity in brachial plexus magnetic field analysis is in good temporal agreement with N9 Erb’s potential.31 It has better spatial resolution and can differentiate the conduction pathways after median versus ulnar nerve stimulations.30 Additionally, the volume currents that are the source of the far-field P9 potential have been elucidated, following Hashimoto’s theory.32,34 Based on magnetoneurographic analysis, the far-field P9 potential peak latency is the result of the change in volume conductor size between the upper and lower thorax.32
Cervical spinal roots
Due to their low spatial resolution, spinal roots cannot be assessed directly with SSEPs. Multi-nerve SSEPs have been attempted to deduce information indirectly about spinal roots, as well as trunks and cords.17,18Table 2 summarizes these deductions. Magnetic evoked fields have been recorded and reconstructed with UGRENS because they have a higher spatial resolution and can differentiate pathways of neural activity as they flow through different intervertebral foramina with three components. The first leading depolarization component of intra-axonal currents (ipsilateral to the stimulation side) can be seen in current distribution maps that flow into the intervertebral foramina. This is followed by the second trailing repolarization component of intra-axonal current which also flows along the course of the nerve through the intervertebral foramina. The third component is an inward current flowing perpendicular to the course of the nerves between leading and tailing components (i.e., the depolarization point) and propagates cranially on the convex side of the nerve and outside the spinal canal.38 In most research subjects, the equivalent currents flow through C6/7 and C7/T1 intervertebral foramina after median nerve stimulation, and C7/T1 and T1/T2 intervertebral foramina after ulnar nerve stimulation at the wrist or elbow.28,38,39 In some subjects, more proximal flow through C4/5 and C5/6 was seen with median nerve stimulation, and C6/7 with ulnar nerve stimulation.38 Future research is needed to discern whether these reflect anatomical variations versus inaccuracies in signal source estimation. Therefore, MSG/MNG as research tool has been able to differentiate median versus ulnar nerves anatomical pathways up to the root level in individual subjects.
The dorsal column volley (DCV)
In SSEP recordings (Fig. 3C), once the afferent/action potential volley enters the spinal cord, it ascends through white matter in the cuneate fasciculus: termed the dorsal column volley (DCV). 13,40 Its near-field activity (N11, N12) can be recorded with a series of electrodes placed over the dorsal aspect of the neck (C1S-C7S) and its peak latency can increase by 0.8-1 msec as it is conducted from the lower to the upper cervical spine.29,41 Once the near field N12 is recorded at the levels of the first cervical vertebrae (C1S), the N12 far-field potential is seen in the scalp prior to its termination in the cuneate nucleus.13
The DCV has been recorded with MSG/MNG as evoked fields after median nerve stimulation at the wrist and elbow, reconstructed with UGRENS, and correlated with the N11 peak latency in SSEPs.28,38,39 The leading depolarization component of intra-axonal currents (contralateral to the side of stimulation) ascends in the spinal canal in current distribution maps superimposed on X-ray. These ascending currents are better visualized with elbow stimulation compared to wrist stimulation (due to less signal dispersion). The descending currents cannot be detected presumably due to magnetic fields produced by descending volleys not persisting and being canceled out by the ascending volley that lasts longer. The N11 peak latency in SSEP (recorded with C5S-Erbc derivation) correlates with the timing of the inward volume current (at depolarization site) at C5 flowing perpendicular to the spinal cord.39
Although the correlation in timing was used as proof-of-concept, it is important to note that it is practically difficult to detect the ascending neural activity with SSEP, while this can be done much easier with MSG. More importantly, the conduction velocity in the dorsal column at the level of the cervical spinal cord has been calculated with MSG following thoracic cord stimulation at 64.3 m/s28 and median nerve stimulation at 53.9-75.9 m/s.39
The upper and lower cervical potentials
In SSEP recordings (Fig. 3D), the DCV is followed by the posterior neck N13 potential, which has a dual nature. In the lower cervical cord, the caudal N13 (cN13) potential is a fixed generator potential.13,40 It can be recorded optimally over the 5th and 6th cervical spinal vertebrae (C5S and C6S) as negativity and over the anterior neck region as positivity. The negative and positive components can be recorded by using a ring electrode around the neck at the level of C5S posteriorly and superior boarder of thyroid cartilage anteriorly. This stationary cervical potential therefore has a horizontal dipole and was referred to as N13/P13. It represents postsynaptic activity in the grey matter of the cervical spinal cord (i.e., in dorsal horn interneurons generating a horizontal dipole). In the upper cervical levels (C2S), the rostral N13 (rN13) potential has a vertical dipole and its generator is suggested to be presynaptic neurons of nucleus cuneatus. Paired median nerve stimuli can be used to facilitate the presynaptic upper cervical response and suppress the post-synaptic lower cervical response.42 On the other hand, the separation of the lower cN13 and upper rN13 is easier with ulnar nerve stimulation due to its lower entrance to the cervical spinal cord compared to the median nerve.40 The lemniscal afferent volley from the cuneate nucleus then proceeds to the thalamus.
The far-field potentials P13/14 complex are positive deflections in the scalp. The P13/14 complex has three subcomponents: P13, P14a, and P14b.43 There are considerable inter-subject variability and not all the subcomponents are present in all individuals.43 There can be anywhere from 1, 2, or 3 subcomponents present in any given individual. Although the precise generators of P13/14 complex have not been widely agreed upon,13,44 postsynaptic generators have been suggested, specifically, the medial lemniscus for P13 and P14a, and more rostrally for P14b.33,43
With MSG/MNG, spinal cord evoked fields have been recorded with a high signal-to-noise ratio after median nerve stimulation at the wrist and reconstructed with UGRENS. A nontraveling stable current source in the posterior-anterior direction was estimated and compared with the peak latency of the cervical N13/P13 potential in the C5S-anterior cervical derivation. The peak intensity latency of the reconstructed current correlated with cervical N13/P13 peak latency, that is generated in spinal cord grey matter and dorsal horn with a horizontal dipole.39
LOCALIZATION EFFORTS FOR SSEPs
There were many efforts by different investigators to add more precise localizing value to SSEP recordings of the brachial plexus, roots, and spinal cord. At the recording technique level, these had included recordings from multiple nerves in one extremity35 and unsurpassed neurophysiological techniques in teasing out the generators of different near- and far-field potentials by different cephalic, non-cephalic, and spine derivations.13,19,33,43,45 At the clinical level, these included the extensive lesional literature that serves as practical proof (or lack thereof) of accuracy of underlying generator localization. All of these efforts, albeit successful, turned out to be too cumbersome for day-to-day routine of neurophysiology practices.
THE BIRTH OF OBLIGATE POTENTIALS
In 1996, the American Clinical Neurophysiology Society published the first evoked potential guidelines followed by a second in 2006 (known as the 9D guidelines) as a standard for recording and clinical interpretation of median nerve SSEPs in the upper extremity and posterior tibial nerve SSEPs in the lower extremity.46 The 2006 guidelines resulted in a standardization and consensus with recording obligate potentials (peripheral Erb’s potential EP, stationary cervical potential N13, subcortically generated far-field potentials of P14, N18 and cortical N20 potential) and therefore markedly improved inter-lab variability and unified clinical interpretation language in practice. These streamlining efforts coincided with a decline in exemplary attempts to add to the localization value of SSEPs.
Based on these guidelines, the obligate potentials recorded by SSEPs became a clinical tool for broad regional localization, from central (cortical, subcortical) to peripheral regions. Clinically, they have been used to look for silent lesions, to monitor clinical status changes over time, and to quantify dysfunction linked to a neuroradiologic finding44 and in intraoperative monitoring.47,48 Therefore, the clinical gap of obtaining neurophysiological information from the brachial plexus, cervical roots, and spinal cord with localization accuracy has remained. Clinicians have been waiting for a new neurophysiologic modality to be developed that would allow noninvasive localization in these anatomical structures. We hope that with the above comparison we have highlighted that, despite its excellent temporal resolution, and due to its limited spatial resolution and effects of volume conduction, the electrography modality of SSEP is not able to localize lesions in the brachial plexus, roots and spinal cord with high accuracy.
With a similar temporal resolution, MSG/MNG offers an improved spatial resolution, mainly due to being unaffected by volume conduction (from bones, muscles and fat that ordinarily surrounds these structures). Therefore, MSG/MNG can allow for observation and analysis of magnetic fields generated at the brachial plexus, spinal root entry, segmental dorsal horn, and the spinal cord itself, which can carry future clinical implications. For example, it has the potential to differentiate between central spinal canal stenosis and foraminal stenosis, allowing for determination of the precise structural abnormality responsible for symptoms (even in the presence of multiple abnormalities in neuroimaging) and, in the future, has the potential to help guide more focused and minimally invasive surgeries. Another potential future benefit is that MSG/MNG can measure conduction velocity in both spinal cord and nerves/plexi/roots. This has the potential to allow determination of precise site of conduction block in variety of peripheral and central as well as obstructive, lesional and demyelinating disorders. However, it remains to be determined whether factors such as equipment availability and cost will hinder the widespread clinical adoption of MSG/MNG.
CONCLUSION
SSEPs are all-encompassing signals recorded along the anatomical path of a sensory stimulus from the point of peripheral stimulation ultimately to somatosensory cortex. However, some of the recorded potentials are near-field, while others are far-field. This inhomogeneity in signal origin is the principal source of inaccuracies in SSEPs. Other sources of inaccuracies include anatomical gaps, inaccuracies in anatomical localization of near-field potentials (for example, the triphasic wave form of EP does not address the complex anatomical intricacies of the brachial plexus), and inaccuracies in anatomico-physiological correlation of far-field stations with anatomical location. In the past several decades, there have been multiple attempts to account for these clinical gaps by stimulating multiple nerves in a limb (to increase localization accuracy of near-field potentials), recording a variety of non-obligate potentials (to fill the gap in anatomico-physiological signal localization), et cetera. Various other obstacles (otherwise not discussed in this article) include subject variability, merging of waveforms, waveform duplication in the case of far-field potentials, and nonobligate waveforms, requiring several kinds of unsurpassed neurophysiological approaches to tease out neurophysiological information, all of which rendered these exemplary attempts not practical for everyday clinical practice.
The magnetography modalities of MSG/MNG are needed to fill these gaps in our clinical neurophysiology techniques: for instance, by improving the accuracy of anatomical localization. These gaps stem from limitations such as volume conduction and the lack of point-to-point correspondence between evoked potential fields and their underlying anatomical sources. In this context, MSG/MNG is a powerful window into the neurophysiology of the brachial plexus, roots and spinal cord that has the potential for accurate localization of signal transmission in relation to lesions within the bony spinal column and the neural spinal cord itself. It may also allow the observation of a signal in its continuity from the site of stimulation to its arrival at the brainstem. Most importantly, MSG/MNG has the potential to open new avenues of spinal cord research which has historically been difficult to study noninvasively in human clinical neurophysiology.
Notes
Conflicts of Interest
Glenn D. R. Watson was an employee of SK Life Science Inc. when drafting this article.
Funding
None.
Fig. 1.
Electromagnetic action potential propagation. (A) Transmembrane inward current (yellow hue) and transmembrane outward currents (blue hue), generating a quadripolar source with two unequal dipoles (leading depolarization +/- followed by trailing repolarization -/+). The magnetic field (green arrows) is perpendicular to and clockwise around the intra-axonal current. Direction of propagation denoted by red arrow. Image modified from Wikswo and van Egeraat.7 (B) Spatial profile of two unequal dipoles in opposite directions (i.e., quadripolar source), partially cancelling out and resulting in a triphasic wave; images modified from Stegeman et al.12
Fig. 2.
Spatial profiles of dipolar sources. (A) Spatial profile of a dipolar source in an infinite volume conductor only producing a near-field region; (B) spatial profile of a dipolar source in a finite long cylindrical volume conductor produces both near field (red arrow) and far field (blue arrows) regions. The far field region (blue arrows) is defined as the region in which the potential decreases <5% over distances that are ≥10% of the volume conductors maximal dimension. Figure modified from Stegeman et al.12
Fig. 3.
Somatosensory evoked potential following median nerve stimulation. (A) Schematic drawing of ipsilateral side view of electrode positions: in green electrodes placed vertically along the anterior boarder of sternocleidomastoid (SCM) muscles with ipsilateral (to side of stimulation) electrodes 1-3 illustrated and contralateral (to side of stimulation) electrodes 4-6 not shown but corresponding to positions of electrodes 1-3. In blue electrodes over the 1st to 7th cervical vertebrae (C1S to C7S). In red ring of electrodes placed at the level of C5S posteriorly and the superior boarder of the thyroid cartilage anteriorly. (B) Median SSEP recorded from electrodes placed on ipsilateral Erb’s point (EPi), along the anterior boarder of ipsilateral (electrodes 1-3) and contralateral (electrodes 4-6) SCM muscle and cephalic electrode Fz. (C) Median SSEP recorded from electrodes placed on ipsilateral EPi, via ring electrode around the neck at the level of C5S posteriorly and the superior boarder of the thyroid cartilage anteriorly, and cephalic electrode Fz. (D) Median SSEP recorded from electrodes placed on ipsilateral EPi, electrodes placed over the cervical spinous processes 1-7 (C1S to C7S) and cephalic electrode Fz. Figure modified from Emerson and Pedley.13 BPV, brachial plexus volley; NC REF, noncephalic reference (contralateral elbow); DCV, dorsal column volley; CERV P13, cervical P13; CERV N13, cervical N13; NEG, negative; SSEP, somatosensory evoked potential.
Table 1.
Summary of SSEPs after median nerve stimulation that is within the view of MSG/MNG
msec
Generator
Potential
Peripheral derivation
Cervical derivation
Scalp derivation
9
BPV
N9 (NFP)
AXi-AXc
AXi-REF
EPi-EPc
EPi-REF
SCMi-REF
P9 (FFP)
CS-REF
Ceph-REF
Ceph-Ceph
10
AVIN
N10 (FFP)
SCM-REF
CS-REF
Ceph-REF
11-12
DCV
N11 (NFP)
C1S-REF
Ceph-Ceph
N12 (NFP)
C2S-REF
Ceph-REF
C3S-REF
C4S-REF
C5S-REF
C6S-REF
C7S-REF
13
Dorsal horn interneurons
cN13 (FFP)
C6S-REF
N13/P13 (FFP)
C5S-REF
Ring-REF
Presynaptic neurons of NC
rN13
C2S-REF
P13 (FFP)
Ceph-REF
14
Caudal to thalamus
P14a (FFP)
Ceph-REF
P14b (FFP)
Ceph-REF
SSEPs, somatosensory evoked potentials; MSG, magnetospinography; MNG, magnetoneurography; BPV, brachial plexus volley; NFP, near field potential; AXi, axially electrodes ipsilateral; AXc, axillary electrode contralateral; REF, non cephalic reference electrode (elbow, hand or any bony prominence); EPi, Erb’s point ipsilateral; EPc, Erb's point contralateral; SCMi, electrodes along anterior boarder of sternocleidomastoidian muscle ipsilateral; FFP, far field potential; CS, cervical spine; Ceph, cephalic electrodes (can be Fz, CPi, CPz); AVIN, antidromic volley of interosseus nerve; SCM, sternocleidomastoid muscle; DCV, dorsal column volley; cN13, caudal N13; Ring, ring electrode; NC, nucleus cuneatus; rN13, rostral N13.
Table 2.
Multi-nerve SSEPs to help determine localization in proximal brachial plexus and roots
Nerve
Nerve
Median
Radial
Ulnar
Musculocutaneous
Cords
Cords
Lat: C6-7, cutaneous
Posterior
Medial
Lateral
Med: C8-T1, M. Aff
Trunks
Trunks
All
Upper
Lower
Upper
Roots
Roots
C6-T1
C6-7
C8-T1
C5-6
SSEPs, somatosensory evoked potentials; Lat, lateral; Med, medial; M. Aff, muscle affarents.
References
1. Stroman PW, Wheeler-Kingshott C, Bacon M, Schwab JM, Bosma R, Brooks J, et al. The current state-of-the-art of spinal cord imaging: methods. Neuroimage 2014;84:1070-1081.
3. Kimura J. The F wave and the A wave in electrodiagnosis in diseases of nerve and muscles principle and practice. 4th ed. New York, NY: Oxford University Press, 2013:149-179.
4. Adachi Y, Kawabata S. Squid magnetoneurography: an old-fashioned yet new tool for noninvasive functional imaging of spinal cords and peripheral nerves. Front Med Technol 2024;6:1351905.
7. Wikswo JP Jr, van Egeraat JM. Cellular magnetic fields: fundamental and applied measurements on nerve axons, peripheral nerve bundles, and skeletal muscle. J Clin Neurophysiol 1991;8:170-188.
12. Stegeman DF, Dumitru D, King JC, Roeleveld K. Near- and far-fields: source characteristics and the conducting medium in neurophysiology. J Clin Neurophysiol 1997;14:429-442.
15. Desmedt JE, Nguyen TH, Carmeliet J. Unexpected latency shifts of the stationary P9 somatosensory evoked potential far field with changes in shoulder position. Electroencephalogr Clin Neurophysiol 1983;56:628-634.
16. Kimura J, Mitsudome A, Yamada T, Dickins QS. Stationary peaks from a moving source in far-field recording. Electroencephalogr Clin Neurophysiol 1984;58:351-361.
17. Yiannikas C, Shahani BT, Young RR. The investigation of traumatic lesions of the brachial plexus by electromyography and short latency somatosensory potentials evoked by stimulation of multiple peripheral nerves. J Neurol Neurosurg Psychiatry 1983;46:1014-1022.
20. Adachi Y, Kawai J, Uehara G, Ogata H, Miyamoto M, Kawabata S, et al. A SQUID biomagnetometer system for measurement of a human cervical spinal cordevoked field. Supercond Sci Technol 2005;18:S303.
21. Adachi Y, Kawai J, Uehara G, Miyamoto M, Tomizawa S, Kawabata S. A 75-ch SQUID biomagnetometer system for human cervical spinal cord evoked field. IEEE Trans Appl Supercond 2007;17:3867-3873.
22. Adachi Y, Kawai J, Miyamoto M, Ogata H, Tomori M, Kawabata S, et al. A SQUID system for measurement of spinal cord evoked field of supine subjects. IEEE Trans Appl Supercond 2009;19:861-866.
23. Adachi Y, Miyamoto M, Kawai J, Uehara G, Ogata H, Kawabata S, et al. Improvement of SQUID magnetometer system for extending application of spinal cord evoked magnetic field measurement. IEEE Trans Appl Supercond 2010;21:485-488.
24. Adachi Y, Kawabata S, Sasano T, Oyama YH, Uehara G, Sekihara K. Biomagnetic measurement system for supine subjects with expanded sensor array and real-time noise reduction. Annu Int Conf IEEE Eng Med Biol Soc 2015;2015:7071-7074.
25. Adachi Y, Kawabata S, Fujihira JI, Uehara G. Multi-channel SQUID magnetospinogram system with closed-cycle helium recondensing. IEEE Trans Appl Supercond 2016;27:1-4.
26. Sekihara K, Kawabata Y, Ushio S, Sumiya S, Kawabata S, Adachi Y, et al. Dual signal subspace projection (DSSP): a novel algorithm for removing large interference in biomagnetic measurements. J Neural Eng 2016;13:036007.
27. Watanabe T, Kawabata Y, Ukegawa D, Kawabata S, Adachi Y, Sekihara K. Removal of stimulus-induced artifacts in functional spinal cord imaging. Annu Int Conf IEEE Eng Med Biol Soc 2013;2013:3391-3394.
29. Lee EK, Seyal M. Generators of short latency human somatosensory-evoked potentials recorded over the spine and scalp. J Clin Neurophysiol 1998;15:227-234.
30. Watanabe T, Kawabata S, Hoshino Y, Ushio S, Sasaki T, Miyano Y, et al. Novel functional imaging technique for the brachial plexus based on magnetoneurography. Clin Neurophysiol 2019;130:2114-2123.
31. Ozaki I, Akaza M, Watanabe T, Miyano Y, Adachi Y, Kawabata S. Detailed magnetoelectric analysis of a nerve impulse propagation along the brachial plexus. Clin Neurophysiol 2023;145:129-138.
32. Paku M, Ando M, Itakura T, Sato S, Tani Y, Ishihara M, et al. Magnetoneurography to investigate the mechanisms underlying the P9 far-field potential. Clin Neurophysiol 2023;150:197-204.
35. Yamada T, Machida M, Oishi M, Kimura A, Kimura J, Rodnitzky RL. Stationary negative potentials near the source vs. positive far-field potentials at a distance. Electroencephalogr Clin Neurophysiol 1985;60:509-524.
36. Fujita Y, Yamada T, Inoue K, Sato A, Katayama M, Ofuji A, et al. Origin of the “N10” stationary-field potential after median nerve stimulation. J Clin Neurophysiol 1999;16:69-76.
37. Inoue K, Mimori Y, Nakamura S. N10 potential as an antidromic motor evoked potential in a median nerve short-latency somatosensory evoked potential study. J Clin Neurophysiol 2002;19:37-42.
38. Miyano Y, Kawabata S, Akaza M, Sekihara K, Hoshino Y, Sasaki T, et al. Visualization of electrical activity in the cervical spinal cord and nerve roots after ulnar nerve stimulation using magnetospinography. Clin Neurophysiol 2020;131:2460-2468.
39. Akaza M, Kawabata S, Ozaki I, Miyano Y, Watanabe T, Adachi Y, et al. Noninvasive measurement of sensory action currents in the cervical cord by magnetospinography. Clin Neurophysiol 2021;132:382-391.
40. Zanette G, Tinazzi M, Manganotti P, Bonato C, Polo A. Two distinct cervical N13 potentials are evoked by ulnar nerve stimulation. Electroencephalogr Clin Neurophysiol 1995;96:114-120.
42. Araki A, Yamada T, Ito T, Urushibara N, Kohira R, Hsu SP, et al. Dissociation between upper and lower neck N13 potentials following paired median nerve stimuli. Electroencephalogr Clin Neurophysiol 1997;104:68-73.
43. Sonoo M, Genba-Shimizu K, Mannen T, Shimizu T. Detailed analysis of the latencies of median nerve somatosensory evoked potential components, 2: analysis of subcomponents of the P13/14 and N20 potentials. Electroencephalogr Clin Neurophysiol 1997;104:296-311.
48. Toleikis JR. Intraoperative monitoring using somatosensory evoked potentials. A position statement by the American Society of Neurophysiological Monitoring. J Clin Monit Comput 2005;19:241-258.
Somatosensory evoked potentials versus magnetospinography and magnetoneurography: is there a need for a new diagnostic modality in brachial plexus, cervical roots and spinal cord neurophysiology?
Fig. 1. Electromagnetic action potential propagation. (A) Transmembrane inward current (yellow hue) and transmembrane outward currents (blue hue), generating a quadripolar source with two unequal dipoles (leading depolarization +/- followed by trailing repolarization -/+). The magnetic field (green arrows) is perpendicular to and clockwise around the intra-axonal current. Direction of propagation denoted by red arrow. Image modified from Wikswo and van Egeraat.7 (B) Spatial profile of two unequal dipoles in opposite directions (i.e., quadripolar source), partially cancelling out and resulting in a triphasic wave; images modified from Stegeman et al.12
Fig. 2. Spatial profiles of dipolar sources. (A) Spatial profile of a dipolar source in an infinite volume conductor only producing a near-field region; (B) spatial profile of a dipolar source in a finite long cylindrical volume conductor produces both near field (red arrow) and far field (blue arrows) regions. The far field region (blue arrows) is defined as the region in which the potential decreases <5% over distances that are ≥10% of the volume conductors maximal dimension. Figure modified from Stegeman et al.12
Fig. 3. Somatosensory evoked potential following median nerve stimulation. (A) Schematic drawing of ipsilateral side view of electrode positions: in green electrodes placed vertically along the anterior boarder of sternocleidomastoid (SCM) muscles with ipsilateral (to side of stimulation) electrodes 1-3 illustrated and contralateral (to side of stimulation) electrodes 4-6 not shown but corresponding to positions of electrodes 1-3. In blue electrodes over the 1st to 7th cervical vertebrae (C1S to C7S). In red ring of electrodes placed at the level of C5S posteriorly and the superior boarder of the thyroid cartilage anteriorly. (B) Median SSEP recorded from electrodes placed on ipsilateral Erb’s point (EPi), along the anterior boarder of ipsilateral (electrodes 1-3) and contralateral (electrodes 4-6) SCM muscle and cephalic electrode Fz. (C) Median SSEP recorded from electrodes placed on ipsilateral EPi, via ring electrode around the neck at the level of C5S posteriorly and the superior boarder of the thyroid cartilage anteriorly, and cephalic electrode Fz. (D) Median SSEP recorded from electrodes placed on ipsilateral EPi, electrodes placed over the cervical spinous processes 1-7 (C1S to C7S) and cephalic electrode Fz. Figure modified from Emerson and Pedley.13 BPV, brachial plexus volley; NC REF, noncephalic reference (contralateral elbow); DCV, dorsal column volley; CERV P13, cervical P13; CERV N13, cervical N13; NEG, negative; SSEP, somatosensory evoked potential.
Fig. 1.
Fig. 2.
Fig. 3.
Somatosensory evoked potentials versus magnetospinography and magnetoneurography: is there a need for a new diagnostic modality in brachial plexus, cervical roots and spinal cord neurophysiology?
msec
Generator
Potential
Peripheral derivation
Cervical derivation
Scalp derivation
9
BPV
N9 (NFP)
AXi-AXc
AXi-REF
EPi-EPc
EPi-REF
SCMi-REF
P9 (FFP)
CS-REF
Ceph-REF
Ceph-Ceph
10
AVIN
N10 (FFP)
SCM-REF
CS-REF
Ceph-REF
11-12
DCV
N11 (NFP)
C1S-REF
Ceph-Ceph
N12 (NFP)
C2S-REF
Ceph-REF
C3S-REF
C4S-REF
C5S-REF
C6S-REF
C7S-REF
13
Dorsal horn interneurons
cN13 (FFP)
C6S-REF
N13/P13 (FFP)
C5S-REF
Ring-REF
Presynaptic neurons of NC
rN13
C2S-REF
P13 (FFP)
Ceph-REF
14
Caudal to thalamus
P14a (FFP)
Ceph-REF
P14b (FFP)
Ceph-REF
Nerve
Nerve
Median
Radial
Ulnar
Musculocutaneous
Cords
Cords
Lat: C6-7, cutaneous
Posterior
Medial
Lateral
Med: C8-T1, M. Aff
Trunks
Trunks
All
Upper
Lower
Upper
Roots
Roots
C6-T1
C6-7
C8-T1
C5-6
Table 1. Summary of SSEPs after median nerve stimulation that is within the view of MSG/MNG
SSEPs, somatosensory evoked potentials; MSG, magnetospinography; MNG, magnetoneurography; BPV, brachial plexus volley; NFP, near field potential; AXi, axially electrodes ipsilateral; AXc, axillary electrode contralateral; REF, non cephalic reference electrode (elbow, hand or any bony prominence); EPi, Erb’s point ipsilateral; EPc, Erb's point contralateral; SCMi, electrodes along anterior boarder of sternocleidomastoidian muscle ipsilateral; FFP, far field potential; CS, cervical spine; Ceph, cephalic electrodes (can be Fz, CPi, CPz); AVIN, antidromic volley of interosseus nerve; SCM, sternocleidomastoid muscle; DCV, dorsal column volley; cN13, caudal N13; Ring, ring electrode; NC, nucleus cuneatus; rN13, rostral N13.
Table 2. Multi-nerve SSEPs to help determine localization in proximal brachial plexus and roots
SSEPs, somatosensory evoked potentials; Lat, lateral; Med, medial; M. Aff, muscle affarents.
