Neuromagnetic investigation of somatotopy of human hand somatosensory cortex

Summary In order to investigate functional topography of human hand somatosensory cortex we recorded somatosensory evoked fields (SEFs) on MEG during the first 40 ms after stimulation of median nerve, ulnar nerve, and the 5 digits. We applied dipole modeling to determine the three-dimensional cortial representations of different peripheral receptive fields. Median nerve and ulnar nerve SEFs exhibited the previously described N20 and P30 components with a magnetic field pattern emerging from the head superior and re-entering the head inferior for the N20 component; the magnetic field pattern of the P30 component was of reversed orientation. Reversals of field direction were oriented along the anterior-posterior axis. SEFs during digit stimulation showed analogous N22 and P32 components and similar magnetic field patterns. Reversals of field direction showed a shift from lateral inferior to medial superior for thumb to little finger. Dipole modeling yielded good fits at these peak latencies accounting for an average of 83% of the data variance. The cortical digit representations were arranged in an orderly somatotopic way from lateral inferior to medial superior in the sequence thumb, index finger, middle finger, ring finger, and little finger. Median nerve cortical representation was lateral inferior to that of ulnar nerve. Isofield maps and dipole locations for these components are consistent with neuronal activity in the posterior bank of central fissure corresponding to area 3b. We conclude that SEFs recorded on MEG in conjunction with source localization techniques are useful to investigate functional topography of human hand somatosensory cortex non-invasively.     

Cerebral potentials and electromyographic responses evoked by stretch of wrist muscles in man

Summary The cerebral evoked potential produced by rapid extension of the wrist was recorded from scalp electrodes in normal subjects while they exerted a small background flexor torque (0.24 Nm) against an electric motor. The initial part of the response consisted of a negative deflection (N1) with an average latency of 24.7 ms. This was followed by a biphasic P1/P2 (32 ms) response and a large later negative wave (N2) (76 ms). Passive wrist extension also evoked reflex EMG responses in the forearm flexor muscles which could be resolved into a short latency (25 ms) and long-latency (52 ms) component. The cerebral responses persisted almost unchanged during complete ischaemic anaesthesia of the hand produced by a pressure cuff around the wrist, and were reduced if the stretch was given during voluntary wrist flexion. The primary component (N1-P1/ P2) of the cerebral response probably represents the arrival at the cortex of the muscle afferent volley. However, the significance of the secondary component (P1/P2-N2) is unknown. Under certain conditions, its size was related to the size of the long latency stretch reflex evoked by stretch of the flexor muscles. Thus, increasing the velocity of stretch or decreasing the repetition rate (from 1.0 to 0.15 Hz) at which stretches were applied, increased the size of both the muscle reflex and the cerebral response. The secondary component also could be changed by voluntary reaction to wrist stretch. Changes in the size of the secondary component of the evoked response may represent the earliest cortical sign of interaction between sensory input and motor output.     

Genetic influence demonstrated for MEG‐recorded somatosensory evoked responses

We tested for a genetic influence on magnetoencephalogram (MEG)‐recorded somatosensory evoked fields (SEFs) in 20 monozygotic (MZ) and 14 dizygotic (DZ) twin pairs. Previous electroencephalogram (EEG) studies that demonstrated a genetic contribution to evoked responses generally focused on characteristics of representative brain potentials. Here we demonstrate significantly smaller amplitude differences within MZ compared to DZ twin pairs for the complete SEF time series (across left and right hand SEFs: 0.37 vs. 0.60 pT² and 0.28 vs. 0.39 pT² for primary [SI] and secondary [SII] sensory cortex activation) and higher MZ than DZ wave shape correlations (.71 vs. .44 and .52 vs. .35 for SI and SII activation). Our findings indicate a genetic influence on MEG‐recorded evoked brain activity and also confirm our recent conclusion ( van 't Ent, van Soelen, Stam, De Geus, & Boomsma, 2009 ) that higher MZ resemblance for EEG amplitudes is not trivially reflecting greater MZ concordance in intervening biological tissues.     

Activation of the human posterior parietal cortex by median nerve stimulation

We recorded somatosensory evoked magnetic fields from ten healthy, right-handed subjects with a 122-channel whole-scalp SQUID magnetometer. The stimuli, exceeding the motor threshold, were delivered alternately to the left and right median nerves at the wrists, with interstimulus intervals of 1, 3, and 5 s. The first responses, peaking around 20 and 35 ms, were explained by activation of the contralateral primary somatosensory cortex (SI) hand area. All subjects showed additional deflections which peaked after 85 ms; the source locations agreed with the sites of the secondary somatosensory cortices (SII) in both hemispheres. The SII responses were typically stronger in the left than the right hemisphere. All subjects had an additional source, not previously reported in human evoked response data, in the contralateral parietal cortex. This source was posterior and medial to the SI hand area, and evidently in the wall of the postcentral sulcus. It was most active at 70–110 ms.     

Somatosensory evoked magnetic fields following passive movement compared with tactile stimulation of the index finger

Cortical processing of passive finger movement was assessed magnetoencephalographically in 12 healthy volunteers and compared with somatosensory evoked magnetic fields (SEF) following tactile stimulation. A new device comprising a clamp-like digit holder facilitated bilateral guidance of the briskly elevated index finger. Both passive movement and tactile stimulation induced activation of the contralateral primary somatosensory (SI) cortex, indicated by six SEF deflections with inter-individually rather consistent peak latencies of 20–230 ms following proprioceptive and 20–300 ms following tactile stimulation. SEF responses to the two stimulus modalities clearly differed with regard to peak latencies, amplitudes and orientations of equivalent current dipoles (ECDs). The strength and orientation of proprioception-related ECDs suggested sequential activation of SI generators, with possible involvement of areas 3a and/or 2 at around 20 ms, area 4 at approximate peak latencies of 65 and 100 ms and area 3b between 150 to 230 ms. Passive movement elicited additional activation of cortical regions outside SI, including the bilateral perisylvian regions and the contralateral cingulate gyrus at latencies of 40–470 and 150–500 ms respectively. The study provides new results with respect to the spatiotemporal analysis of proprioception-related cortical processing and may contribute to a better understanding of the modality-specific organization of the human somatosensory cortex. Electronic Publication     

Topographical organization of the cortical afferent connections to the cortex of the anterior ectosylvian sulcus in the cat

Summary The cortical afferents to the cortex of the anterior ectosylvian sulcus (SEsA) were studied in the cat, using the retrograde axonal transport of horseradish peroxidase technique. Following injections of the enzyme in the cortex of both banks, fundus and both ends (postero-dorsal and anteroventral) of the anterior ectosylvian sulcus, retrograde labeling was found in: the primary, secondary, and tertiary somatosensory areas (SI, SII and SIII); the motor and premotor cortices; the primary, secondary, anterior and suprasylvian fringe auditory areas; the lateral suprasylvian (LS) area, area 20 and posterior suprasylvian visual area; the insular cortex and cortex of posterior half of the sulcus sylvius; in area 36 of the perirhinal cortex; and in the medial bank of the presylvian sulcus in the prefrontal cortex. Moreover, these connections are topographically organized. Considering the topographical distribution of the cortical afferents, three sectors may be distinguished in the cortex of the SEsA. 1) The cortex of the rostral two-thirds of the dorsal bank. This sector receives cortical projections from areas SI, SII and SIII, and from the motor cortex. It also receives projections from the anterolateral subdivision of LS, and area 36. 2) The cortex of the posterior third of the dorsal bank and of the posterodorsal end. It receives cortical afferents principally from the primary, secondary and anterior auditory areas, from SI, SII and fourth somatosensory area, from the anterolateral subdivision of LS, vestibular cortex and area 36. 3) The cortex of the ventral bank and fundus. This sulcal sector receives abundant connections from visual areas (LS, 20, posterior suprasylvian, 21 and 19), principally from the lateral posterior and dorsal subdivisions of LS. It also receives abundant connections from the granular insular cortex, caudal part of the cortex of the sylvian sulcus and suprasylvian fringe. Less abundant cortical afferents were found to arise in area 36, second auditory area and prefrontal cortex. The abundant sensory input of different modalities which appears to converge in the cortex of the anterior ectosylvian sulcus, and the consistent projection from this cortex to the deep layers of the superior colliculus, make this cortical region well suited to play a role in the control of the orientation movements of the eyes and head toward different sensory stimuli.Supported by FISSS grants 521/81 and 1250/84     

Assessing functioning of the prefrontal cortical subregions with auditory evoked potentials in sleep-wake cycle

Our previously observations showed that the amplitude of cortical evoked potentials to irrelevant auditory stimulus (probe) recorded from several different cerebral areas was differentially modulated by brain states. At present study, we simultaneously recorded auditory evoked potentials (AEPs) from the dorsolateral prefrontal cortex (DLPFC) and the ventromedial prefrontal cortex (VMPFC) in the freely moving rhesus monkey to investigate state-dependent changes of the AEPs in the two subregions of prefrontal cortex. AEPs obtained during passive wakefulness (PW), active wakefulness (AW), slow wave sleep (SWS) and rapid-eye-movement sleep (REM) were compared. Results showed that AEPs from two subregions of prefrontal cortex were modulated by brain states. Moreover, a significantly greater increase of the peak-to-peak amplitude (PPA) of N1-P1 complexes appears in the DLPFC during PW compared to that during AW. During REM, the PPA of N1-P1 complexes presents a contrary change in the two subregions with significant difference: a significant increase in the DLPFC and a slight decrease in the VMPFC compared to that during AW. These results indicate that the modulation of brain states on AEPs from two subregions of the prefrontal cortex investigated is also not uniform, which suggests that different subregions of the prefrontal cortex have differential functional contributions during sleep-wake cycle.     

Centrifugal regulation of a task-relevant somatosensory signal triggering voluntary movement without a preceding warning signal

A warning signal followed by an imperative signal generates anticipatory and preparatory activities, which regulate sensory evoked neuronal activities through a top-down centrifugal mechanism. The present study investigated the centrifugal regulation of neuronal responses evoked by a task-relevant somatosensory signal, which triggers a voluntary movement without a warning signal. Eleven healthy adults participated in this study. Electrical stimulation was delivered to the right median nerve at a random interstimulus interval (1.75–2.25 s). The participants were instructed to extend the second digit of the right hand as fast as possible when the electrical stimulus was presented (ipsilateral reaction condition), or extend that of the left hand (contralateral reaction condition). They also executed repetitively extension of the right second digit at a rate of about 0.5 Hz, irrespective of electrical stimulation (movement condition), to count silently the number of stimuli (counting condition). In the control condition, they had no task to perform. The amplitude of short-latency somatosensory evoked potentials, the central P25, frontal N30, and parietal P30, was significantly reduced in both movement and ipsilateral reaction conditions compared to the control condition. The amplitude of long-latency P80 was significantly enhanced only in the ipsilateral reaction condition compared to the control, movement, contralateral reaction, and counting conditions. The long-latency N140 was significantly enhanced in both movement and ipsilateral reaction conditions compared to the control condition. In conclusion, short- and long-latency neuronal activities evoked by task-relevant somatosensory signals were regulated differently through a centrifugal mechanism even when the signal triggered a voluntary movement without a warning signal. The facilitation of activities at a latency of around 80 ms is associated with gain enhancement of the task-relevant signals from the body part involved in the action, whereas that at a latency of around 140 ms is associated with unspecific gain regulation generally induced by voluntary movement. These may be dissociated from the simple effect of directing attention to the stimulation.     

Interstimulus Interval Has No Effect on a Mid-Latency Scalp Potential Generated by Innocuous-Related Activity in the Primary Somatosensory Cortex

The sural nerve-evoked somatosensory evoked potential (SEP) scalp topography was separated into stable periods, where a stable period refers to consecutive time points with the same topographic pattern. The stimulus intensity-amplitude function, conduction velocity measurements, and a dipole source localization analysis of one of these stable periods, SP1 (60-90 ms post-stimulus), strongly suggests that it is generated by the response of neurons in the primary somatosensory cortex (SI) to inputs arising from the innocuous A peripheral afferents. Interstimulus intervals (ISI) ranging between 2.5 s and 10.0 s had no effect on SP1 amplitude. This contrasts with an earlier report from this laboratory demonstrating that subjective magnitude ratings and the amplitude of another stable period that appears at about 160-180 ms post-stimulus and that is also generated in SI, increase with decreasing ISI. Thus, ISI appears to affect perception and the late but not the mid-latency responses in SI.     

A neuromagnetic study of movement-related somatosensory gating in the human brain

Neuromagnetic fields from the left cerebral hemisphere of five healthy, right-handed subjects were investigated under three different experimental conditions: (1) electrical stimulation of the right index finger (task S); (2) voluntary movement of the same finger (M); (3) M+S condition, consisting of voluntary movements of the right index finger triggering the electrical stimulus at the very beginning of the electromyogram. The three conditions were administered in random order every 5–8 s. In addition, the task somatosensory evoked fields (task SEFs) gathered during condition (1) were compared with control SEFs recorded at the beginning of the experiment during rest. In all subjects the overlay of somatosensory stimulation on movement provoked a decrement in brain responsiveness (gating) as determined by the amplitude of gated SEFs. The latter was found as the difference between the neuromagnetic fields during M+S condition (overlaying of movement and sensory stimulation) minus neuromagnetic fields under M condition (M only). The gating effect was found to begin approximately 30 ms after movement onset, and to last for the whole period of the ongoing movement. The theoretical locus of gating was estimated by dipole localisation of the difference between task SEFs and gated SEFS using a moving dipole model. The site of the early gating effect (<40 ms) was found to be more anteriorly located than the later (>40 ms) gating effect. The task SEFs were found to be larger (significant after 30 ms) than the control SEFs elicited under the basal condition. The results are discussed with respect to timing, mechanism (centrifugal and centripetal), locus and selectivity of gating. In addition, the results are discussed with regard to clinical application (measuring attentional deficits in patients with impairments of higher mental functions and measuring gating deficits in patients with disturbed sensorimotor integration).     

Contribution of somatosensory cortex to responses in the rat cerebellar granule cell layer following peripheral tactile stimulation

The spatial coincidence of somatosensory cerebral cortex (SI) and trigeminal projections to the cerebellar hemisphere has been previously demonstrated. In this paper we describe the temporal relationship between tactilely-evoked responses in SI and in the granule cell layer of the cerebellar hemisphere, in anesthetized rats. We simultaneously recorded field potentials in areas of common receptive fields of SI and of the cerebellar folium crus IIa after peripheral tactile stimulation of the corresponding facial area. Response of the cerebellar granule cell layer to a brief tactile stimulation consisted of two components at different latencies. We found a strong correlation between the latency of the SI response and that of the second (long-latency) cerebellar component following facial stimulation. No such relationship was found between the latency of the SI response and that of the first (short-latency) cerebellar component, originating from a direct trigeminocerebellar pathway. In addition, lidocaine pressure injection in SI, cortical ablation, and decerebration all significantly affected the second cerebellar peak but not the first. Further, when tactile stimuli were presented 75 ms apart, the response in SI failed, as did the second cerebellar peak, while the shortlatency cerebellar response still occurred. We found a wide spatial distribution of the upper lip response beyond the upper lip area in crus IIa for the long-latency component of the cerebellar response. Our results demonstrate that SI is the primary contributor to the cerebellar long-latency response to peripheral tactile stimulation. These results are discussed in the context of Purkinje cell responses to tactile input.     

Modification of neuromagnetic responses of the human auditory cortex by masking sounds

Summary We have studied the effects of masking sounds on auditory evoked magnetic fields (AEFs) of healthy humans. The AEFs were elicited by 25-ms tones presented randomly to the left or to the right ear, and the responses were recorded over the right auditory cortex. Without masking, the 100-ms deflection (N100m) was of somewhat higher amplitude and of shorter latency for contrathan ipsilateral stimuli. Continuous speech, music, or intermittent noise, delivered to the left ear, dampened N100m to stimulation of both ears without correlated changes in sensation. Intermittent noise had a weaker effect on N100m than speech or music. Continuous noise fed to the left ear dampened both the sensation of and the responses to the left-ear stimuli, with no significant effect on the responses to the right-ear stimuli. The results suggest that the masking effects of continuous noise, seen at the auditory cortex, derive mainly from the periphery whereas the effects of sounds with intensity and frequency modulations take place at more central auditory pathways.     

Somatosensory amplification and its relationship to somatosensory, auditory, and visual evoked and event-related potentials (P300)

Somatosensory amplification refers to the tendency to experience benign and ambiguous somatic sensation as intense, noxious, and disturbing. The construct is helpful in assessing the perceptual style of a variety of somatizing conditions, but there is no human study clarifying the effects of neurological function on somatosensory amplification. The present study examines the relationship between somatosensory amplification and different types of evoked potentials. In 33 healthy volunteers (mean age 24 years, 18 men), latencies and amplitudes were recorded using the following parameters: short-latency somatosensory, brainstem-auditory, and visual evoked potentials (SSEP, BAEP, and VEP, respectively) and auditory event-related potentials (ERP). All subjects completed questionnaires for the Somatosensory Amplification Scale (SSAS), 20-item Toronto Alexithymia Scale (TAS-20), and Profile of Mood State (POMS). The SSAS scores were significantly associated with the P200 latency (p=0.020) and P300 amplitude of ERP (p=0.041), controlling for the significant effect of the TAS and POMS depression and tension-anxiety scales. The SSEP, BAEP, and VEP latencies or amplitudes were not statistically significant (all p>0.05). When the subjects were divided into high and low SSAS groups based on the median of the SSAS scores, the P300 amplitude of ERP significantly discriminated the two groups (p=0.023) by multiple logistic regression analysis. Although the findings should be viewed as preliminary because of the small sample size, somatosensory amplification appears to reflect some aspects of long-latency cognitive processing rather than short-latency interceptive sensitivity from the viewpoint of encephalography.     

New depth short-latency somatosensory evoked potential (SEP) component recorded in human SI area

To analyse short and long-latency (SEPs) recorded by chronically stereotactically electrodes implanted in SI area of two epileptic patients. Two drug-resistant epileptic patients (2 females, 38 and 15 years, respectively) suffering from left temporal and right frontal epilepsy respectively, were investigated by an electrode-chronically implanted in SI area. Short and long latency somatosensory evoked potentials were recorded by depth electrodes 10 days after implantation. This is the first study to describe a depth N36 response by an intracerebral recording electrode in the SI area, probably generated by a radially oriented generator, located in area 1. Furthermore, we confirmed a role of SI in the genesis of N60 component. Finally, our present data suggest that the SI area is still active at 120 ms after the stimulus, since in one patient (no. 2) we identified a N120 potential, reaching its maximal amplitude at the same depth as the N20 response.     

Centrifugal regulation of task-relevant somatosensory signals to trigger a voluntary movement

Many previous papers have reported the modulation of somatosensory evoked potentials (SEPs) during voluntary movement, but the locus and mechanism underlying the movement-induced centrifugal modulation of the SEPs elicited by a task-relevant somatosensory stimulus still remain unclear. We investigated the centrifugal modulation of the SEPs elicited by a task-relevant somatosensory stimulus which triggers a voluntary movement in a forewarned reaction time task. A pair of warning (S1: auditory) and imperative stimuli (S2: somatosensory) was presented with a 1 s interstimulus interval. Subjects were instructed to respond by moving the hand ipsilateral or contralateral to the somatosensory stimulation which elicits the SEPs. In four experiments, the locus and selectivity of the SEPs’ modulation, the contribution of cutaneous afferents and the effect of contraction magnitude were examined, respectively. A control condition where subjects had no task to perform was compared to several task conditions. The amplitude of the frontal N30, parietal P30, and central P25 was decreased and that of the long latency P80 and N140 was increased when the somatosensory stimuli triggered a voluntary movement of the stimulated finger compared to the control condition. The N60 decreased with the movement of any finger. These results were considered to be caused by the centrifugal influence of neuronal activity which occurs before a somatosensory imperative stimulus. The present findings did not support the hypothesis that the inhibition of afferent inputs by descending motor commands can occur at subcortical levels. A higher contraction magnitude produced a further attenuation of the amplitude of the frontal N30, while it decreased the enhancement of the P80. Moreover, the modulation of neuronal responses seems to result mainly from the modulation of cutaneous afferents, especially from the moved body parts. In conclusion, the short- and long-latency somatosensory neuronal activities evoked by task-relevant ascending afferents from the moved body parts are regulated differently by motor-related neuronal activities before those afferent inputs. The latter activities may be associated with sensory gain regulation related to directing attention to body parts involved in the action.     

Stimulus duration influences the dipole location shift within the auditory evoked field component N100m

The equivalent source of the neuromagnetic auditory evoked field (AEF) component N100m shifts systematically within its latency range. In the current study, possible effects of stimulus duration on this shift were analysed. 15 subjects were stimulated monaurally with tones of different duration (50, 100, 200 ms) and AEFs were recorded successively over both hemispheres. Dipoles were calculated in 5-ms-steps from 15 ms before to 15 ms after the N100m peak maximum. A dipole location shift within the N100m latency from posterior to anterior and from superior to inferior was observed. The shift in anterior-posterior direction was found to be larger in the right compared to the left hemisphere. Stimulus duration significantly affected the degree of dipole shift in this direction. It was found to be shorter the shorter the stimulus.     

Modulation of somatosensory evoked responses in the primary somatosensory cortex produced by intracortical microstimulation of the motor cortex in the monkey

Summary Previous studies have shown that the amplitude of somatosensory evoked potentials is diminished prior to, and during, voluntary limb movement. The present study investigated the role of the motor cortex in mediating this movement-related modulation in three chronically prepared, awake monkeys by applying low intensity intracortical microstimulation (ICMS) to different sites within the area 4 representation of the arm. Air puff stimuli were applied to the contralateral arm or adjacent trunk at various delays following the ICMS. Somatosensory evoked potentials were recorded from the primary somatosensory cortex, areas 1 and 3b, with an intracortical microelectrode. The principal finding of this study was that very weak ICMS, itself producing at most a slight, localized, muscle twitch, produced a profound decrease in the magnitude of the short latency component of the somatosensory evoked potentials in the awake money. Higher intensities of ICMS (suprathreshold for eliciting electromyographic (EMG) activity in the target muscle, i.e. that muscle activated by area 4 stimulation) were more likely to decrease the evoked response and produced an even greater decrease. The modulation appeared to be, in part, central in origin since (i) it preceded the onset of EMG activity in 23% of experiments, (ii) direct stimulation of the muscle activated by ICMS, which mimicked the feedback associated with the small ICMS-induced twitch, was often ineffective and (iii) the modulation was observed in the absence of EMG activity. Peripheral feedback, however, may also make a contribution. The results also indicate that the efferent signals from the motor cortex can diminish responses in the somatosensory cortex evoked by cutaneous stimuli, in a manner related to the somatotopic order. The effects are organized so that the modulation is directed towards those neurones serving skin areas overlying, or distal to, the motor output.     

Reduction of somatosensory evoked fields in the primary somatosensory cortex in a one-back task

In the present study, responses of the somatosensory cortex to sensory input of ten human volunteers were investigated during a one-back task with different conditions of attention. During an condition of attention subjects were requested to detect a predefined sequence of tactile stimuli applied to two different fingers of the dominant hand while a series of visual stimuli was presented simultaneously with an asynchronous stimulus-onset to the tactile stimuli. During an condition of distraction subjects received the identical series of visual and tactile stimuli like in the condition of attention but were now requested to detect a predefined stimulus sequence within the visual stimulus domain. In both conditions, somatosensory evoked magnetic fields (SEFs) to the tactile stimuli were recorded by means of a 31-channel magnetoencephalograph (MEG) from subjects‘ contralateral primary somatosensory cortex. The mean global field power, the dipole strength, the maximum current density, and the first component of the singular value decomposition (SVD) of magnetic fields were used to compare early components of the SEF in the conditions of attention versus distraction. Surprisingly, results revealed significant decreases of measures of all four parameters during the condition of attention as compared to the condition of distraction indicating that early responses of the primary somatosensory cortex became significantly reduced in the condition of attention. We hypothesize that changes in the centre-periphery-relationship of receptive fields in the primary somatosensory cortex may account for this unexpected result.     

Reproducibility and stability of neuromagnetic source imaging in primary somatosensory cortex

The present study investigated the test-retest reliability of magnetoencephalography (MEG) source localization of somatosensory evoked fields (SEFs) over an extended time period. Five healthy subjects were stimulated pneumatically at the first and fifth digit in two sessions spaced several months apart. At each location 400 stimuli were presented. The validation of the results was performed by overlay of the dipole localizations into the individual anatomic structure of the subjects' cortex by the use of magnetic resonance images (MRIs). The source localizations of the SEF component were found to be highly reproducible. The mean standard deviation of the dipole locations of the first digit was 1.55 mm in the x-, 1.55 mm in the y- and 3.49 mm in the z-direction. The mean standard deviation of the fifth digit was 3.69 mm in the x-, 4.27 mm in the y- and 6.60 mm in the z-direction. These results support the use of MEG recordings combined with MRI as an adequate method to define the organization of the human primary somatosensory cortex and provide a useful approach to the rapid detection of neuroplasticity.     

Primary somatosensory evoked magnetic fields elicited by sacral surface electrical stimulation

To explore the brain response to sacral surface therapeutic electrical stimulation (SSTES) for the treatment of refractory urinary incontinence and frequent micturition, evoked magnetic fields were measured in six healthy males. Electrical stimuli were applied between bilateral surface electrodes over the second through fourth posterior sacral foramens with intensity just below the pain threshold. Somatosensory evoked magnetic fields (SEFs) for the bilateral median (MN) and posterior tibial nerves (PTN) were also measured for the comparison. Sources of the early SEF peaks were superimposed on individual magnetic resonance images. The first peak latency for sacral stimuli, M30, occurred at 30.2 ± 0.8 ms (mean ± standard deviation, N = 6), with shorter latency than those for PTN stimulus (39.3 ± 1.4 ms, N = 12) and longer latency than those for MN stimulus (21.0 ± 0.9 ms, N = 12). The second peak latency for sacral stimuli, M50, occurred at 47.2 ± 2.9 ms (N = 6). Both M30 and M50 peaks showed a single dipole pattern over the vertex in the isofield maps. The equivalent current dipoles of M30 and M50 were both estimated near the medial end of the central sulcus with approximately posterior current direction. These results suggest that the sacral M30 and M50 are responses from the primary somatosensory cortex. The relatively long time lag between the onset and peak of M30 suggests that SSTES directly affects both the cauda equina and cutaneous nerve of the sacral surface.