Generator sources of median somatosensory evoked potentials

A Comprehensive Review and Practical Guide of the Applications of Evoked Potentials in Neuroprognostication After Cardiac Arrest

Cardiorespiratory arrest is a very common cause of morbidity and mortality nowadays, and many therapeutic strategies, such as induced coma or targeted temperature management, are used to reduce patient sequelae. However, these procedures can alter a patient's neurological status, making it difficult to obtain useful clinical information for the reliable estimation of neurological prognosis. Therefore, complementary investigations are conducted in the early stages after a cardiac arrest to clarify functional prognosis in comatose cardiac arrest survivors in the first few hours or days. Current practice relies on a multimodal approach, which shows its greatest potential in predicting poor functional prognosis, whereas the data and tools to identify patients with good functional prognosis remain relatively limited in comparison. Therefore, there is considerable interest in investigating alternative biological parameters and advanced imaging technique studies. Among these, somatosensory evoked potentials (SSEPs) remain one of the simplest and most reliable tools. In this article, we discuss the technical principles, advantages, limitations, and prognostic implications of SSEPs in detail. We will also review other types of evoked potentials that can provide useful information but are less commonly used in clinical practice (e.g., visual evoked potentials; short-, medium-, and long-latency auditory evoked potentials; and event-related evoked potentials, such as mismatch negativity or P300).

Misconceptions in IONM Part I: Interleaved Intraoperative Somatosensory Evoked Potential Stimulation

A misconception in the field of intraoperative neurophysiological monitoring (IONM) is that continuous, multi-nerve (four-limb), interleaved somatosensory evoked potential (SSEP) stimulation, while advantageous, is not universally utilized due to variety of misunderstandings regarding this approach to SSEP stimulation. This article addresses the rationale for this misconception. We find that continuous, multi-nerve, interleaved SSEP stimulation is superior to all other stimulation paradigms in most operative scenarios, allowing the fastest acquisition of SSEPs at low stimulation repetition rates, which generate the highest amplitude cortical responses.

Somatosensory evoked potentials in clinical practice: a review

The authors present a review of the current use of somatosensory evoked potentials (SSEPs) in neurological practice as a non-invasive neurophysiological technique. For this purpose we have reviewed articles published in English or Portuguese in the PubMed and LILACS databases. In this review, we address the role of SSEPs in neurological diseases that affect the central nervous system and the peripheral nervous system, especially in demyelinating diseases, for monitoring coma, trauma and the functioning of sensory pathways during surgical procedures. The latter, along with new areas of research, has become one of the most important applications of SSEPs.

Evaluation of different cortical source localization methods using simulated and experimental EEG data

Different cortical source localization methods have been developed to directly link the scalp potentials with the cortical activities. Up to now, these methods are the only possible solution to noninvasively investigate cortical activities with both high spatial and time resolutions. However, the application of these methods is hindered by the fact that they have not been rigorously evaluated nor compared. In this paper, the performances of several source localization methods (moving dipoles, minimum Lp norm, and low resolution tomography (LRT) with Lp norm, p equal to 1, 1.5, and 2) were evaluated by using simulated scalp EEG data, scalp somatosensory evoked potentials (SEPs), and upper limb motor-related potentials (MRPs) obtained on human subjects (all with 163 scalp electrodes). By using simulated EEG data, we first evaluated the source localization ability of the above methods quantitatively. Subsequently, the performance of the various methods was evaluated qualitatively by using experimental SEPs and MRPs. Our results show that the overall LRT Lp norm method with p equal to 1 has a better source localization ability than any of the other investigated methods and provides physiologically meaningful reconstruction results. Our evaluation results provide useful information for choosing cortical source localization approaches for future EEG/MEG studies.

Median nerve somatosensory evoked potentials in migraine

In visual evoked potential studies, habituation during stimulus repetition with the same stimulus at a constant intensity has been found to be abnormal in migraineurs between attacks. The purpose of this study was to investigate habituation of somatosensory evoked potentials (SEPs) and the effects of migraine on them. Eighty-five subjects were included in the study: 30 healthy volunteers (HVs) and 55 migraineurs [30 with migraine without aura (MO), 25 with migraine with aura (MA)]. During continuous stimulation at 3 Hz, four blocks of 100 responses were sequentially averaged of Erb's point (N9), cervical (N13), and cortical (N20) median nerve SEPs. Mean amplitude changes in the second, third and fourth blocks are expressed as percentages of the first block. There was habituation to N13 and N20 in the second, third and fourth blocks in HVs. In the migraine groups, there was no habituation; on the contrary, potentiation was found. This potentiation was statistically significant only in the second blocks for N13 (MO P=0.007, MA P=0.01 versus HVs). However, in both migraineur groups, the rate of N20 potentiations was statistically significant versus that in HVs for all blocks (all P < 0.05). It is concluded that whilst physiological habituation occurs in HVs for cervical and cortical SEPs, in migraine patients there is an interictal deficit of habituation of this sensory modality.

Generators of short latency human somatosensory-evoked potentials recorded over the spine and scalp

Somatosensory evoked potentials (SEPs) are most commonly obtained after stimulation of the median nerve and the posterior tibial nerve. SEPs reflect conduction of the afferent volley along the peripheral nerve, dorsal columns, and medial lemniscal pathways to the primary somatosensory cortex. Short-latency SEPs are recorded over the spine and scalp. After posterior tibial nerve stimulation, the following waveforms are recorded: N22, W3, the dorsal column volley, N29, P31, N34, and P37. After median nerve stimulation, the brachial plexus volley, dorsal column volley (N11), N13, P14, N18, N20, and P22 potentials are recorded. We discuss the current state of knowledge about the generators of these SEPs. Such information is crucial for proper interpretation of SEP abnormalities.

Sensory potentials evoked by magnetic stimulation of the cervical spine

Magnetic stimulation of cervical spinal roots was shown to elicit sensory potentials (MESP) which could easily be recorded at the fingers with ring electrodes. The latency of the MESP recorded at digit I was significantly shorter and the amplitude higher than of digits III and V. The latencies were largely independent of stimulus strength. In an attempt to localize the place of depolarization, the latencies of these potentials were compared with the N11 of the SEP (reflecting the arrival in the spinal cord) and with F-wave latencies and motor evoked potentials (MEP) to abductor pollicis brevis. The MESP latencies showed a very constant difference with the N11, being 0.6 ms faster. The mean difference between F latency and MEP was 1.2 ms. It is concluded that the origin of these MESPs is very near the spinal foramina, possibly in the sensory ganglia.

Sympathetic skin response in hemispheric lesions

We recorded the sympathetic skin response (SSR) from electrical nerve stimulation in 16 patients with cerebrovascular accident (CVA). Location and nature of the lesion were documented by computerized tomography (CT). Median (pre-rolandic and parietal) somatosensory evoked potentials (SEP) were also recorded. SSR was absent bilaterally in eight hemiplegics after stimulation of the plegic side and present bilaterally after stimulation of the normal side in the first weeks after CVA. Parietal and pre-rolandic SEPs were absent in the affected hemispheres. SSR was present bilaterally after stimulation of each side in the remaining CVA cases with reduced amplitude SEPs. Absence of the electrically evoked SSR in hemispheric lesions may be due to involvement of central afferent pathways or temporary suppression of suprasegmental excitatory influences.

Epilepsy in schizencephaly: abnormal cortical organization studied by somatosensory evoked potentials

Median nerve short-latency somatosensory evoked potentials (MN-SSEP) are recorded from the scalp to assess parietal lobe function and from the cortex to identify primary sensory and motor areas before epilepsy surgery. Nevertheless, the origins of many of the MN-SSEP waveforms and the reliability of this technique for localizing the central sulcus are not definitively known. We studied a child with a unilateral, closed, right parietal schizencephalic cleft and frequent simple partial seizures before the child underwent cortical resection. The sensory examination, neuroimaging, and electrical brain stimulation findings indicated a normal thalamus and an abnormal parietal lobe. Scalp-recorded MN-SSEPs showed intact widespread N18 potentials bilaterally, but absent right, although normal left parietal N20 and P27 waveforms. Cortically recorded MN-SSEPs could not localize the central sulcus owing to an absence of the expected negative potential over the right postcentral gyrus and the presence of waves with abnormal latencies over the precentral cortex. These findings suggest that: (a) the N18 potential probably originates at or below the level of the thalamus, (b) the N20 and P27 peaks are most likely generated by parietal cortex or white matter, and (c) cortically recorded MN-SSEPs can fail to localize the central sulcus before epilepsy surgery when congenital anomalies exist in the parietal lobe.

Polarity reversal of N20 and P23 somatosensory evoked potentials between scalp and depth recordings

From depth and scalp electrodes, we recorded MN-SSEPs of a 33-year-old man with right parietal dysfunction and refractory right temporal seizures. A depth lead with 8 electrodes was implanted deep in each parietal-temporal region. Stimulation and recording parameters followed American EEG Society guidelines. Scalp recordings had well-defined P9, P13-14, N18, N20, and P23 potentials with normal conduction times bilaterally. Depth recordings showed potentials of greater number, voltage, and coherence. P13-14 and N18 were recorded at all depth sites with latencies similar to those at the scalp. N18 had markedly greater voltage and duration near the thalamus, with multiple fast components on its ascending phase. In the deep parietal region there was a positivity corresponding to the scalp N20 and a negative potential equal in latency to scalp P23. These findings support an origin of P13-14 caudal to the thalamus, multiple thalamic and possibly rostral brain-stem generators for N18, and generation of N20 and P23 in sensory cortex of subjacent white matter.

Electrophysiologic studies on locked-in patients: heterogeneity of findings

Somatosensory evoked potentials, brain-stem auditory evoked potentials and electroencephalograms were obtained from 9 patients with the diagnosis of 'locked-in' syndrome. No pattern of evoked potential abnormality was specific to this syndrome, with findings ranging from bilaterally normal to unilaterally or bilaterally absent. The evoked potential studies complemented radiographic findings in defining the extent of the lesion and revealed that a portion of the pontine tegmentum was usually involved. Pathology from 2 patients corroborated the findings of the evoked potential studies. The value of evoked potential studies of patients with locked-in syndrome is to provide early objective evidence of brain-stem involvement independent of the clinical examination, EEG and radiographic studies.

Somatosensory evoked potentials of the dog: recording techniques and normal values

Median and tibial nerve somatosensory evoked potentials (SSEPs) of 5 sedated dogs were studied to determine their normal features and optimal stimulation and recording techniques. Cortical potentials were mapped from an extensive array of skull electrodes as each limb was independently stimulated with subdermal needles. The effects of bandpass and stimulus intensity and rate were also assessed. Three cortical components (P1, N1, P2) were evoked by median or tibial nerve stimulation and were localized along the coronal suture at lateral and medial electrodes, respectively. SSEP voltage varied much more than morphology, topography, or latency. The inion was a stable, indifferent reference site. Cortical SSEP frequency content was mostly below 250 Hz. Maximal SSEP voltage was achieved only at stimulus intensities 2-3 times motor threshold. Appropriate methods minimize technical difficulties and consistently yield legible SSEPs.

The use of somatosensory evoked potentials in infants and children

Somatosensory evoked potentials (SSEPs) are a useful, reliable means of assessing function of the somatosensory system. Complex maturational changes of the CNS such as synaptogenesis and myelination, as well as body growth, complicate interpretation of SSEPs. An understanding of these factors enhances clinical interpretation in infants and children.

Dipole-tracing of 'awareness' attenuating the cortical components of somatosensory evoked potentials

Using the dipole-tracing method, the source generators of N18, P22 and P40 of the somatosensory evoked potential (SEP) were estimated as the equivalent dipole. After voluntary action of the thumb flexion, no changes were observed in N18 or P40, but the amplitude of P22 was suppressed. The after-effects of intention accompanied by a voluntary action or the subject's awareness that electrical stimulation will be given after the voluntary action were treated as 'awareness'. By subtracting the pure SEP from SEP during 'awareness', it was found that the equivalent dipole of 'awareness' of P22 was located at the same region of pure P22, but the vector was of opposite orientation. 'Awareness' attenuated the perceptive potential of SEP like P22 generated in the cortex.

Topographical displays of somatosensory evoked potentials

The distribution of somatosensory evoked potentials (SEPs) after stimulation of the median nerve at the wrist was examined in 10 normal subjects using isopotential maps. The latencies of continuous negative and positive peaks were measured in each lead. The differences of the potentials at these latencies were measured in all the leads and the isopotential maps were constructed. The distribution of P0-NI was all similar. The latencies of P0 were almost the same in all the leads at about 13 msec. The distribution of NI-PI-NII was divided into three types--N16-P20-N28 localized in the frontal region, N17-P22-N30 localized in the central region and N19-P25-N33 distributed in the parieto-occipito-temporal regions. The distributions of NII-PII and PII-NIII were all similar, with high amplitudes in the central region. The latencies of PII and NIII were almost the same in all the leads at about 45 msec and 68 msec.

Clinical effectiveness of dermatomal evoked cerebrally recorded somatosensory responses

Among 129 patients with spine pain and radiculopathy in both the cervical and lumbar region, there were significant differences between the upper and lower extremity DSER groups, noting the higher number of normal studies by our criteria in the upper extremities as compared to the lower extremities. This difference may be due to the intertwining of the "nerve circuitry" of the input pathways. It appears that the dermatomal somatosensory evoked response is a study of low sensitivity and of high specificity.

Selectivity of attenuation (i.e., gating) of somatosensory potentials during voluntary movement in humans

Attenuation of somatosensory evoked potentials (SEPS) recorded from the scalp during voluntary movement occurs for specific combinations of the finger moved and the peripheral nerve stimulated. The cerebral potential component occurring at a latency of 27 msec (P27) evoked either by stimulation of median nerve at the wrist or by stimulation of 1st and 2nd digit nerves in the fingers were selectively attenuated during movement of 1st digit but were not altered during movement of 5th digit. By contrast, the cerebral P27 component evoked by stimulation of ulnar nerve at the wrist or by stimulation of 5th digital nerve were attenuated during movement of that digit but were not altered during movement of 1st digit. Gating of somatosensory activity is a selective phenomenon occurring when movement involves the areas being stimulated.

Scalp-recorded short and middle latency peroneal somatosensory evoked potentials in normals: comparison with peroneal and median nerve SEPs in patients with unilateral hemispheric lesions

Peroneal somatosensory evoked potentials (SEPs) were performed on 23 normal subjects and 9 selected patients with unilateral hemispheric lesions involving somatosensory pathways. Recording obtained from right and left peroneal nerve (PN) stimulations were compared in all subjects, using open and restricted frequency bandpass filters. Restricted filter (100-3000 Hz) and linked ear reference (A1-A2) enhanced the detection of short latency potentials (P1, P2, N1 with mean peak latency of 17.72, 21.07, 24.09) recorded from scalp electrodes over primary sensory cortex regions. Patients with lesions in the parietal cortex and adjacent subcortical areas demonstrated low amplitude and poorly formed short latency peroneal potentials, and absence of components beyond P3 peak with mean latency of 28.06 msec. In these patients, recordings to right and left median nerve (MN) stimulation showed absence or distorted components subsequent to N1 (N18) potential. These observations suggest that components subsequent to P3 potential in response to PN stimulation, and subsequent to N18 potential in response to MN stimulation, are generated in the parietal cortical regions.

Comparison of somatosensory evoked responses from root and cord recorded by skin and epidural electrodes using stimulation of the median nerve in cervical radiculopathy and radiculomyelopathy

Somatosensory evoked potentials (SEPs) were recorded by stimulating the median nerve at the wrist from the skin and epidural space of the 7th cervical spine in patients suffering from cervical radiculopathy or radiculomyelopathy. The patients were divided into four subgroups according to the severity of the disease. Skin and epidural SEPs were calculated and compared with each other and with control values. Usually only one negative potential N13 was identified in the skin recording, but two potentials N11 and N13 occurred in the epidural recording. Lower amplitudes were obtained from the skin than from the epidural space. In the skin SEPs the mean of the central latency of N13 was significantly prolonged in the severe radiculomyelopathy groups, while the mean of the amplitude N13 showed only a tendency to decrease. In contrast, in the epidural SEPs a significant decrease in the mean of the N11 and N13 amplitudes together with a significant prolongation in the mean of the central latency of N13 could be found. In the epidural recording the amplitude changes in particular increased with the severity of the disease, but the highest number of abnormalities (61%) could be seen in the central latency of N13.

Dissociation of frontal N100 from occipital P100 in pattern reversal visual evoked potentials

We studied the relationship between occipital P100 and frontal N100 in visual evoked potentials produced by pattern reversal in normal subjects and two groups of patients. Recording derivation was critical for interpretation since both Fz and Oz electrode sites are active. In 9 patients, but no normal subjects, P100 was absent. In these patients, use of a standard Oz-Fz montage resulted in the erroneous impression of a 'normal' P100 since a downward deflection was produced by the inverting effect of the amplifier on an intact N100 at Fz. When both P100 and N100 were present (at Oz and Fz respectively), their latencies were usually similar but not identical which contributed to apparent latency shifts or W-shaped wave forms in the Oz-Fz derivation. We conclude that use of a non-cephalic or relatively inactive scalp position (such as the mastoid) should be used as a reference site in addition to Fz to reduce interpretive errors.

What determines the latency and amplitude of stationary peaks in far-field recordings?

In 20 radial nerves from 10 healthy persons, a referential derivation from the tip of the first or second digit registered two biphasic stationary peaks, PI-NI and PII-NII, following stimulation of the nerve in the forearm. These two peaks occurred slightly before the arrival of the propagating impulse at the wrist and at the base of the digit, respectively. With stepwise reduction of stimulation from a maximal to a threshold intensity, the far-field potential decreased in amplitude linearly with the near-field potential recorded at the junction of the volume conductor. Abduction of the first digit or flexion of the second and third digits altered the waveform and, to a lesser degree, the latency of the far-field peaks. However, these changes appeared in an inconsistent manner, which we were unable to characterize. We conclude that in far-field recording a stationary peak results at the border of the adjoining volume conductors in proportion to the magnitude of the propagating axonal volley approaching the boundary.

Some comments on the clinical use of evoked potentials

In this survey we describe the uses of somatosensory, visual, and auditory evoked potentials (EPs), adding some critical comments on the values and pitfalls of these methods. It must be stressed that the application of EPs is most valuable when combined with a thorough neurological examination. There is general agreement that EP measurement is one of the best techniques for objective, noninvasive study of brain function in humans.

'Gating' of somatosensory evoked potentials begins before the onset of voluntary movement in man

The inflow of somatosensory information to the cerebral cortex is modified before and during active movement in animals. This phenomenon has been termed 'gating' and occurs at several levels of the sensory pathway. We studied somatosensory evoked potentials (SEPs) to stimulation of the median nerve at the wrist during voluntary movement of the ipsilateral thumb in man. Results indicate that SEPs are attenuated shortly after a command to move (approximately 100 ms before the onset of the electromyogram (EMG)), become maximally attenuated with maximum EMG and return to normal size when movement is finished.