- Email: [email protected]
Somatosensory evoked potentials from the human brain-stem: Origins of short latency potentials
Electroencephalography and clinical Neuropl~}'swlogv, 1984, 57:221-227
221
Elsevier Scientific Publishers Ireland, Ltd.
S O M A T O S E N S O R Y EVOKED POTENTIALS FROM THE H U M A N BRAIN-STEM: ORIGINS OF S H O R T LATENCY POTENTIALS i I. HASHIMOTO
Department of Neurosurgerv, Tokyo Metropolitan Hospital of Fuchu, Musashidai 2-0-2, Fuchu CiO, Tokyo 183 (.lapan) (Accepted for publication: August 4, 1983)
Following median nerve stimulation, short latency somatosensory evoked potentials (SEPs) have been recorded from the human scalp, remote from their presumed sources, with the use of farfield averaging techniques. The clinical usefulness of recording short latency SEPs for localizing lesions in the afferent somatosensory pathways from the peripheral nerves to the cortex have been reported (Desmedt et al. 1966; Desmedt 1971; Noel and Desmedt 1975; Anziska et al. 1978; Nakanishi et al. 1978; Anziska and Cracco 1980; Chiappa et al. 1980). The sources of human short latency SEP components remain somewhat uncertain in spite of many previous studies using various electrode montages in normals (Cracco and Cracco 1976; Kimura et al. 1978; Kritchevsky and Wiederholt 1978; Yamada et al, 1980) and in patients with lesions of the somatosensory pathways at various levels (Nakanishi et al. 1978; Green and Mcleod 1979: Anziska and Cracco 1980; Chiappa et al. 1980). Since these potentials are expected to become a useful means for non-invasive evaluation of patients, the precise information of the sources of each short latency SEP component in humans is of a sine qua non for interpretation of the potentials. Direct recordings from the spinal cord (Ertekin 1976; Sances et al. 1978; Shimoji et al. 1978), the thalamus (Albe-Fessard et al. 1962; Ervin and Mark 1964; Pagni 1967; Goto et al. 1968; Larson and Sances 1968; Haider et al. 1972; Fukushima et al. 1976; Sances et al. 1978; Celesia 1979) and the cortex (Jasper et al. 1960; Pagni i This research has been supported by a research grant from the Tokyo Metropolitan Government.
1967; Celesia 1979) have been described, but failed to provide conclusive evidence so far. There have been only few reports on direct recordings from the human brain-stem (Liberson et al. 1970; Strassburg et al. 1979). In this preliminary report, short latency SEPs recorded from the human brain-stem are described and are correlated with those from the scalp.
Methods
Recordings were carried out in a patient with an arterio-venous malformation (AVM) of the superior vermis who underwent ventricular catheterization through a frontal burr hole for ventriculography combined with vertebral angiography for precise delineation of the AVM. Informed consent was obtained from the patient and his family. The recording electrodes consisted of 4 platinum ring electrodes with an outer diameter of 1.5 ram, located along the ventricular tube at equal intervals of 1 cm as described previously (Hashimoto 1982). Insulated silver wire leads with a diameter of 0.1 mm were passed through the catheter. The catheter was introduced into the IIlrd and through the sylvian aqueduct to the IVth ventricle under radioscopic guidance. The wrist contralateral to stimulation served as the common reference for intracranial and scalp recordings. Square wave pulses of 0.1 msec duration were delivered at 4 / s e c over the median nerve at the wrist (cathode proximal). Stimulus intensity was adjusted to produce a small muscle twitch of the thumb. Electrode impedances of less than 3 k~-2 were maintained. The surface electrode was placed over the
0013-4649/84/$03.00 '~' 1984 Elsevier Scientific Publishers Ireland, Ltd.
222
central region contralateral to the stimulated median nerve. The input signals were amplified with a frequency response down 3 dB at 30 and 3 kHz. One thousand responses were averaged for 25 msec employing a Nicolet Ca-1000 averager. The averages were written out with an X-Y plotter. Recordings were repeated twice to test reproducibility. The negativity of the intracranial and scalp electrodes was up in all recordings. The components were labeled by their polarities and peak latencies as recommended by Donchin et al. (1977). After the recordings were taken from the 4 electrodes, the catheter was withdrawn 5 cm from the initial position for recordings from more rostral loci. The recording electrode position within the IVth and IIIrd ventricles was confirmed by a metrizamide ventriculography.
1. HASHIMOTO
negative and positive (P22) waves. The negative wave was clearly bilobed (N15 and N20) with a positive notch (P17) separating this slow wave, and followed 4 small positive waves (P9, P l l , P13, and P14) (bottom trace in Fig. 2). Intracranial poten-
A
B
C
D Result
P13/~ The recording sites were mapped on the lateral view of the X-ray photo of the ventricular system (Fig. 1). The depth (from A to H in Fig. ]) of recording was given for each trace (Fig. 2). The potentials recorded from the contralateral central region, referenced to the wrist, were dominated by
E
F-
G
H
l uV
N15yN20 pllp13 j
J
i
i
i
12.5 msec Fig. 1. The recording sites of short latency somatosensory evoked potentials were mapped on the lateral view of the X-ray photo of the ventricular system and were indicated alphabetically. The sites A and B are in the IVth ventricle (vicinity of the pons); C and D in the aqueduct of Sylvius (the midbrain); E, F, and G in the lllrd ventricle (the thalamus); and H in the lateral ventricle.
~..~z P22 J
i
L
0.5 uV
i
25
Fig. 2. Short latency somatosensory evoked potentials recorded at the locations indicated in Fig. 1. S denotes the surface recording at the central region contralateral to the stimulated median nerve. The common reference is the contralateral wrist. The voltage calibration of 1 ktV applies to all intracranial recordings, while that of 0.5 /tV refers to the surface recording. Negativity is up in all traces.
DEPTH RECORDINGS OF EARLY SEP rials were characterized by 2 or 3 fast positive components ( P l l , P13, and P14) with approximately 1 msec duration and a large slow negative and positive component. The latency of P l l remained appreciably consistent in all traces and the amplitude showed only a slight change in recordings from caudal to more rostral electrodes. The peak latency of the component corresponds in time to that of P l l over the scalp. By contrast, P13 increased its peak latency with a progressively more rostral recording site up to the posterior IIIrd ventricle (site E). Radiographically, this electrode (E) lies nearest to thalamus relay nucleus ventralis posterolateralis (VPL). Rostrally, P13 was seen as a sm~ll notch at the same latency with the surface P13. P13 also showed a gradual increase and appeared largest at electrodes (C and D) within the aqueduct of Sylvius. The P14 component was absent in the recording from the most caudal electrode (A), and its peak latency showed a progressive increase from caudal to rostral and finally coincided with that of the surface P14. The negativity of about 6 msec duration which follows the 3 fast positive wavelets was the largest component in the intracranial SEPs. Its peak latency systematically increased rostrally. The latency shift recorded between the electrodes A and H exceeded 3 msec. The amplitude of the component measured from the onset of P13 remained almost the same in the recordings from the brain-stem, but was attenuated between the electrodes D and E and again very slowly further rostrally. The positive wave following the large negative wave was much smaller in amplitude in all recordings and its peak latency varied according to the depth electrode locations. The shortest latency of this component was recorded at electrode A and the longest at electrode H, and the latency shift was systematic, similar to the preceding negative wave. Discussion
Short latency SEPs recorded within the IVth and IIIrd ventricles and the sylvian aqueduct reflect activities generated from somatosensory path-
223 ways in close proximity to the electrodes and also volume-conducted potentials remote from the electrode sites. There has been considerable evidence to show that SEPs are mediated solely by the dorsal column-lemniscal system (Halliday and Wakefield 1963; Larson et al. 1966: Namerow 1968; Noel and Desmedt 1975, 1980; Desmedt and Cheron 1980). The earliest response from the intracranial recordings is the P l l with an onset latency of 10.3 msec which coincides in time with the surface P11 component. This potential exhibits little or no latency shift along the electrode array and the amplitude (less than 0.3 /~V) is not significantly changed. Therefore, this component recorded from the brain-stem reflects the far-field activity from below the brain-stem. We infer that P l l represents activities in fibers ascending the cuneate fasciculus, as suggested by Desmedt and Cheron (1980, 1981a) on grounds that the N l l onset closely corresponds to spinal entry time and that a clear latency shift of about 1 msec can be recorded for N l l between the lower and upper neck. The observations by Anziska and Cracco (1980) that P2 ( P l l in our nomenclature) peak latency was normal in patients with lesions rostral to the medulla and also direct recordings from the spinal cord by Lesser et al. (1981) are consistent with this interpretation. In contrast to P l l which behaves as a volumeconducted far-field potential, the intracranially recorded P13 exhibits a systematic change in latency from the pons (electrode A) to the diencephalon (electrode E). Furthermore, the wave form and amplitude of the component are almost the same in two different electrodes (electrodes C and D) within the aqueduct of Sylvius where the medial lemniscus runs fairly parallel with it. This implies that the potential must represent a synchronized axonal volley traveling in fiber tract al the level of these recording electrodes. The sudden amplitude reduction between the midbrain and thalamus (electrodes D and E) may be due to the fact that the medial lemniscus shifts its course: laterally at this level. Since electrodes located beyond the thalamic relay nucleus only see the volume-conducted far-field potential of this component, the potential from the rostral electrodes shows no
224
appreciable latency shift and slow amplitude attenuation toward the scalp. Desmedt and Cheron (1980) suggested that the P13 component originates from the medial lemniscus on the basis of the potential distribution using non-cephalic reference recording, coupled with their anatomical measurements of the somatosensory system in human cadavers. The absence or abnormally delayed peak latency of this P13 component in patients with brain-stem lesions caudal to the diencephalon (Anziska and Cracco 1980) is consistent with the present observation. Arezzo et al. (1979) have provided further evidence in monkeys with use of the semimicroelectrodes that the potential analogous to human P13 corresponds to the synchronized volleys of the action potentials within the medial lemniscus as it courses from the brain-stem to the thalamus. The intracranial P14 component which follows P13 by less than 1 msec at the rostral brain-stem (site D) appears to travel toward the rostral recording sites while P13 remains of the same latency beyond the midbrain. As a result, P13 and P14 are separated by about 1.5 msec at the surface. Recordings from the h u m a n midbrain medial lemniscus during stereotactic surgery for intractable pain demonstrated a potential with an onset latency of 11-13 msec (Liberson et al. 1970) and a peak latency of 14-16 msec (Strassburg et al. 1979). Arezzo et al. (1979) and Allison and Hume (1981) observed in their comparative studies in the monkey and the cat that the potential which appears homologous to the human P14 reflects bursts of action potentials traveling along the medial lemniscus and the initial portion of the thalamocortical projections. H u m e and Cant (1978), Nakanishi et al. (1978), and Anziska and Cracco (1980) demonstrated a normal P14 component in their patients with thalamic lesions. Taken together, these data suggest that the earlier portion of scalp P14 primarily reflects activity within the medial lemniscus and its branches to the various brain-stem nuclei (see the later discussion on the anatomy of dorsal columnolemniscal system). Following these potentials, the intracranial electrodes record a negative component which is the largest potential within 25 msec after the median nerve
I. H A S H I M O T O
stimulation. This potential is characterized by its definitely longer duration (approximately 6 msec) compared with the preceding small wavelets of brief duration (roughly 1 msec) and its slower traveling time in a caudo-rostral direction. While the preceding 4 wavelets have reasonably been ascribed to the axonal volleys ascending the dorsal column-lemniscal system, the question arises as to whether this slow negative wave represents the axonal volleys or postsynaptic potentials. Spinal cord potentials recorded intrathecally (Ertekin 1976) and epidurally (Shimoji et al. 1978) in man consist of the initial positive (P1) component of brief duration (1.8 3.2 msec) followed by a negative (N1) component with a longer duration (5.2--8.0 msec). P1 is considered to represent the incoming volleys through the dorsal roots and N1, the synaptic excitation of dorsal horn interneurons (Beall et al. 1977). The slow negative potential at the brain-stem and diencephalon closely resembles in configuration the N1 from the spinal cord. Provided that this negativity represents postsynaptic events rather than axonal activities, the possible neural structures responsible for its generation are the nucleus cuneatus in the medulla and the VPL in the thalamus which are the two main relay nuclei in the dorsal column-lemniscal system. However, the systematic latency shift as well as the similar amplitude and consistent duration of this component along the electrode array within the brain-stem would argue against the interpretation that this component is produced solely by overlapping of the activities from these two fixed sources. This systematic latency shift of the slow negativity within the brain-stem would lead us to postulate the existence of nuclear structures within the pons and midbrain which may be activated progressively by incoming volleys from the ascending lemniscal system. Recently, the nucleus cuneatus was found to project to the various brain-stem nuclei such as the accessory olives, the external nucleus of the inferior colliculus, the superior colliculus, and the nucleus tuber (Hand and Van Winkle 1976), providing an anatomical basis for the interpretation of the traveling slow negative wave in the brain-stem. Thus the surface N15 component may correspond to a traveling slow negative wave in the pontine and midbrain nuclei.
DEPTH RECORDINGS OF EARLY SEP
Direct recordings from the human thalamus have been described previously (Ervin and Mark 1964; Pagni 1967; Goto et al. 1968; Larson and Sances 1968; Haider et al. 1972; Fukushima et al. 1976; Sances et al. 1978; Celesia 1979). The initial potential of the recordings from the VPL following the median nerve stimulation has a peak latency of 12-18 msec and a duration of 6-10 msec which is close to our slow negative component in spite of a great variability in the amplitude and peak latency in their recordings. These data are so variable that they seem difficult to reconcile. Goto et al. (1968) found, however, by moving the recording electrode step by step within the thalamus, that the potential is diphasic and the initial positive phase is followed by a slow negative potential within the region of the nucleus ventralis intermedius (Vim). They found also that, below the ventral border of this nucleus, the slow negative component disappears and only the initial positive component is present (see Fig. 5 in Goto et al. 1968). This suggests that different neural structures are involved in the generation of their 'thalamic' potential; the initial positive component may represent axonal volleys in the medial lemniscus and the slow negative component, postsynaptic activities in the thalamic nucleus (VPL or Vim). There is no detectable activity in the depth recordings within the range for the surface N20 and P22. Thus N20 and P22 can be ascribed to a more rostral structure beyond the thalamus, presumably the primary somatosensory cortex (Desmedt and Cheron 1980, 1981b; Allison and Hume 1981).
Summa~ Short latency somatosensory evoked potentials (SEPs) were recorded in man from an array of electrodes within the IVth (the pons) and IIIrd ventricles (the thalamus) and the aqueduct of Sylvius (the midbrain). The slow negative (N15 and N20) and positive (P22) waves were preceded by 4 small positive (P9, P l l , P13, and P14) waves of approximately 1 msec duration in scalp recordings. The sources of these waves have been differentiated on the basis of their timing and spatial
225
gradients of corresponding intracranial potentials. While P l l is identified as volume conducted from below the brain-stem, P13 and P14 reflect synchronized volleys of the medial lemniscus and its branches to the various pontine and mesencephalic nuclei. In contrast to these earlier positive wavelets arising from fiber tracts, N15 may represent postsynaptic activities within these pontine and midbrain nuclei. N20 and P22 are the initial responses of the juxtarolandic cortex.
R~sum~ Potentiels OvoquOs somatosensoriels du tronc cOr~bral chez l'homme: origine des potentiels gl courte latence
Les composantes ~ courte latence des potentiels evoques somatosensoriels (PES) ont 6t6 enregistr6es chez l'homme a partir d'61ectrodes mobiles 'a l'interieur du 46me ventricule (voisinnage du pont), de l'aqueduc de Sylvius (mesenc+phale), du 36me ventricule (thalamus) et du ventricule lateral. Les ondes lentes negatives (N15 et N20) et positives (P22) 6taient preced6es de 4 petites ondes positives (P9, P l l , P13 et P14) d'une duree approximative de 1 msec dans les enregistrements sur le scalp. Les origines de ces ondes ont 6t6 differenciees par leur chronologie et les gradients spatiaux des potentiels intracrfiniens correspondants. Alors que P11 est identifiee comme lice ~ une conduction en volume ~ partir du tronc cerebral, P13 et P14 refl6tent des volees synchrones du lemnisque median et de ses branchements vers les divers noyaux pontiques et mesenc6phaliques. Contrairement a ces petites ondes positives precoces, en provenance de fasceaux de fibres, la N15 pourrait representer les activites postsynaptiques des noyaux pontiques et mesencephaliques eux-memes. N20 et P22 sont les reponses initiales du cortex juxtarolandique. The author wishes to thank Prof. John E. Desmedt, Brain Research Unit. University of Brussels, Brussels and Prof. Truett Allison. Department of Neurology, Yale University School of Medicine, New Haven, CT, for reviewing the manuscript and for helpful discussion, and to acknowledge the valuable advice from Dr. Muneo Shimamura, Department of Neurophysiology, Tokyo Metropolitan Institute of Neurosciences, Tokyo, and
226 Dr. Buichi lshijima, Department of Neurosurgery, Tokyo Metropolitan Neurological Hospital, Tokyo, in the preparation of the manuscript.
References Albe-Fessard, D., Arfel, G., Guiot. G., Hardy, J., Vourch, G., Hertzog, E., A16onard, P. et Derome, P. D6rivations d'activitbs spontan6es et evoqu6es darts les structures c6r6brales profoundes de l'homme. Rev. neurol., 1962, 106: 89 105. Allison, T. and Hume, A.L. A comparative analysis of shortlatency somatosensory evoked potentials in mare monkey, cat, and rat. Exp. Neurol., 1981, 72:592 611. Anziska, B. and Cracco, R.Q. Short latency somatosensory evoked potentials. Studies in patients with focal neurological disease. Electroenceph. clin. Neurophysiol., 1980, 49: 227 239. Anziskm B., Cracco, R.Q., Cook, A.W. and Feld, E.W. Somatosensory far-field potentials. Studies of normals and patients with multiple sclerosis. Electroenceph. clin. Neurophysiol., 1978, 45: 602-610. Arezzo, J., Legatt, A.D. and Vaughan, H.G. Topography and intracranial sources of somatosensory evoked potentials in the monkey. I. Early components. Electroenceph. clin. Neurophysiol., 1979, 4 6 : 1 5 5 172. Beall, J.E., Applebaum, A.E., Foreman, R.D. and Willis, W.D. Spinal cord potentials evoked by cutaneous afferents in the monkey. J. Neurophysiol., 1977, 40: 199-211. Celesia, G.G. Somatosensory evoked potentials recorded directly from h u m a n thalamus and SMI cortical area. Arch. Neurol. (Chic.), 1979, 36: 399-405. Chiappa, K.H., Choi, S.K. and Young, R.R. Short latency somatosensory evoked potentials following median nerve stimulation in patients with neurological lesions. In: J.E. I)esmedt (Ed.), Clinical Uses of Cerebral, Brainstem and Spinal Somatosensory Evoked Potentials. Progr. clin. Neurophysiol., Vol. 7. Karger, Basel, 1980: 264-281. Cracco, R.Q. and Cracco, J.B. Somatosensory evoked potentials in man: far-field potentials. Electroenceph. clin. Neurophysiol., 1976, 4 1 : 4 6 0 - 4 6 6 . De Divitiis, E., Giaquinto, S. and Signorelli, C.D. Peripheral influences on VPM-VPL thalamic nuclei in the human: a study on evoked potentials. Confin. neurol. (Basel), 1971. 33: 174-185. DesmedL J.E. Somatosensory cerebral evoked potentials in man. In: A. Remond (Ed.), Handbook of Electroenceph. ctin. Neurophysiol., Vol. 9. Elsevier, Amsterdam, 1971: 55-81. Desmedt, J.E. and Cheron, G. Central somatosensory conduction in man: neural generators and interpeak latencies of the far-field components recorded from neck and right or left scalp and earlobes. Electroenceph. clin. Neurophysiol.. 1980, 50: 328-403. DesmedL J.E. and Cheron, G. Prevertebral (oesophageal) re-
I. H A S H I M O T O cording of subcortical somatosensory evoked potentials in man: the spinal P13 component and the dual nature of the spinal generators. Electroenceph. clin. Neurophysiol., 1981a, 52: 257-275. Desmedt, J.E. and Cheron, G. Non-cephalic reference recording of early somatosensory potentials to finger stimulation in adult or aging normal man: differentiation of widespread N18 and contralateral N20 from the prerolandic P22 and N30 components. Electroenceph. din. Neurophvsiol., 1981b, 52:553 570. Desmedt, J.E., Franken, L., Borensteim S., l)ebecker, J., Lambert, C. et Manil, J. Le diagnostic des ralentissements de la conduction aff6rente dans les affections des nerfs p6riph6riques: intbr6t de l'extraction du potentiel evoqu,5 cer+bral. Rev. neurol., 1966, 115:255 262. Donchin, E., Callaway, E., Cooper, R., Desmedt, J.E., Goff, W.R., Hillyard, S.A. and Sutton, S. PublicatiOn criteria for studies of evoked potentials. Report of a committee. In: J.E. Desmedt (Ed.), Attention, Voluntary Contraction and Event-Related Cerebral Potentials. Progr. olin. Neurophysiol.. Vol. 1. Karger, Basel, 1977:1 11. Ertekin, C. Studies on the h u m a n evoked elecUospinogram: tile origin of the segmental evoked potentials. Acta. neurol. scan&, 1976, 53: 3-20. Ervin, F.R. and Mark, V.M. Studies of the h u m a n thalamus. IV. Evoked responses. Ann. N.Y. Acad. Sci., 1964. 112: 81 92. Fukushima, T., Mayanagi, Y. and Bouchard, G. Thalamic evoked potentials to somatosensory stimulation in man. Electroenceph. clin. Neurophysiol., 1976, 40: 481-490. Goto, A., Kosaka, K., Kubota, K., Nakamura, R. and Narabayashi, H. Thalamic potentials from muscle afferents in the human. Arch. Neurol. (Chic.), 1968, 19:302 309. Green, J.G. and Mcleod, S. Short latency somatosensory evoked potentials in patients with neurological lesions. Arch. Neurol. (Chic.), 1979, 36: 846-851. Haider, M., Gangleberger, J.A. and Groll-Knapp, E. Computer analysis of subcortical and cortical evoked potentials and of slow potential phenomena in humans. Confin. neurol. (Basel), 1972, 3 4 : 2 2 4 229. Halliday, A.M. and Wakefield. G.S. Cerebral evoked potentials in patients with dissociated sensory loss. J. Neurol. Neurosurg. Psychiat., 1963, 26: 211-219. Hand, P.J. and Van Winkle, T. The efferent connections of the feline nucleus cuneatus. J. comp. Neurol., 1976, 171:83 110. Hashimoto, I. Auditory evoked potentials from the human midbrain: slow brain stem responses. Electroenceph. olin. Neurophysiot., 1982, 53: 652-657. Hume, A.M. and Cant, B.R. Conduction time in central somatosensory pathways in man. Electroenceph. clin. Neurophysiol., 1978, 45: 361-375. Jasper, H., Lende, R. and Rasmussen, T. Evoked potentials from the exposed somatosensory cortex in man. J. nerv. ment. Dis., 1960, 130:526 537. Kimura, J., Yamada, T. and Kawamura, H. Central lalencies of somatosensory cerebral evoked potentials. Arch. Neurol. (Chic.), 1978, 35:683 688.
DEPTH RECORDINGS OF EARLY SEP Kritchevsky, M. and WiederholL W.C. Short latency somatosensory evoked responses in man. Arch. Neurol. (Chic.), 1978, 35:706 711. Larson, S.J. and Sances, A. Averaged evoked potentials in stereotaxic surgery. J. Neurosurg., 1968, 28: 227-232. Larson, S.J., Sances, A. and Christenson, P.C. Evoked somatosensory potentials in man. Arch. Neurol. (Chic.), 1966, 15: 88-93. Lesser, R.P., Lueders, H., Hahn, J. and Klem, G. Early somatosensory potentials evoked by median nerve stimulation: intraoperative monitoring. Neurology (Minneap.)~ 1981, 31:1519 1523. Liberson, W.T., Voris, H.C. and Uematsu, S. Recording somatosensory evoked potentials during mesencephalotomy for intractable pain. Confin. neurol. (Basel), 1970, 32: 185-194. Nakanishi, T., Shimada, Y., Sakuta, M. and Toyokura, Y. The initial positive component of the scalp-recorded somatosensory evoked potential in normal subjects and in patients with neurological disorders. Electroenceph. clin. Neurophysiol., 1978, 45: 26-34. ! Namerow, N.W. Somatosensory evoked responses in multiple sclerosis patients with varying sensory loss. Neurology (Minneap.), 1968, 18: 1197-1204. Noi)l, P. and Desmedt, J.E. Somatosensory cerebral evoked potentials after vaseular lesions of the brain stem and diencephalon. Brain, 1975, 98: 113-128.
227 Noi~l, P. and Desmedt, J.E. Cerebral and far-field somatosensory evoked potentials in neurological disorders involving the cervical spinal cord, brainstem, thalamus and cortex. In: J.E. Desmedt (Ed.), Clinical Uses of Cerebral, Brainstern and Spinal Somatosensory Evoked Potentials. Progr. clin. Neurophysiol., Vol. 7. Karger, Basel, 1980: 205-230. PagnL C.A. Somato-sensory evoked potentials in thalamus and cortex of man. In: P. Gloor (Ed.), Hans Berger on the EEG of Man. Electroenceph. clin. Neurophysiol., Suppl. 28. Elsevier, Amsterdam, 1967: 147-155. Sances, Jr., A.~ Larson, S.J., Cusick, J.F., Myklebust, J., Ewing, C., Jodat, R., Ackmann, J.J. and Walsh, P. Early somatosensory evoked potentials. Electroenceph. clin. Neurophysiol., 1978, 45:505 514. Shimoji, K., Shimizu, H. and Maruyama, Y. Origin of the somatosensory evoked responses recorded from the cervical skin surface. J. Neurosurg., 1978, 48:980 984. Strassburg, H.M., Thoden, U. and Mundinger, F. Mesencephalic chronic electrodes in pain patients. An electrophysiological study. Appl. Neurophysiol., 1979, 42: 284-293. Yamada, T., Kimura, J. and Nitz, D.M. Short latency somatosensory evoked potentials following median nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1980, 48: 367-376.
221
Elsevier Scientific Publishers Ireland, Ltd.
S O M A T O S E N S O R Y EVOKED POTENTIALS FROM THE H U M A N BRAIN-STEM: ORIGINS OF S H O R T LATENCY POTENTIALS i I. HASHIMOTO
Department of Neurosurgerv, Tokyo Metropolitan Hospital of Fuchu, Musashidai 2-0-2, Fuchu CiO, Tokyo 183 (.lapan) (Accepted for publication: August 4, 1983)
Following median nerve stimulation, short latency somatosensory evoked potentials (SEPs) have been recorded from the human scalp, remote from their presumed sources, with the use of farfield averaging techniques. The clinical usefulness of recording short latency SEPs for localizing lesions in the afferent somatosensory pathways from the peripheral nerves to the cortex have been reported (Desmedt et al. 1966; Desmedt 1971; Noel and Desmedt 1975; Anziska et al. 1978; Nakanishi et al. 1978; Anziska and Cracco 1980; Chiappa et al. 1980). The sources of human short latency SEP components remain somewhat uncertain in spite of many previous studies using various electrode montages in normals (Cracco and Cracco 1976; Kimura et al. 1978; Kritchevsky and Wiederholt 1978; Yamada et al, 1980) and in patients with lesions of the somatosensory pathways at various levels (Nakanishi et al. 1978; Green and Mcleod 1979: Anziska and Cracco 1980; Chiappa et al. 1980). Since these potentials are expected to become a useful means for non-invasive evaluation of patients, the precise information of the sources of each short latency SEP component in humans is of a sine qua non for interpretation of the potentials. Direct recordings from the spinal cord (Ertekin 1976; Sances et al. 1978; Shimoji et al. 1978), the thalamus (Albe-Fessard et al. 1962; Ervin and Mark 1964; Pagni 1967; Goto et al. 1968; Larson and Sances 1968; Haider et al. 1972; Fukushima et al. 1976; Sances et al. 1978; Celesia 1979) and the cortex (Jasper et al. 1960; Pagni i This research has been supported by a research grant from the Tokyo Metropolitan Government.
1967; Celesia 1979) have been described, but failed to provide conclusive evidence so far. There have been only few reports on direct recordings from the human brain-stem (Liberson et al. 1970; Strassburg et al. 1979). In this preliminary report, short latency SEPs recorded from the human brain-stem are described and are correlated with those from the scalp.
Methods
Recordings were carried out in a patient with an arterio-venous malformation (AVM) of the superior vermis who underwent ventricular catheterization through a frontal burr hole for ventriculography combined with vertebral angiography for precise delineation of the AVM. Informed consent was obtained from the patient and his family. The recording electrodes consisted of 4 platinum ring electrodes with an outer diameter of 1.5 ram, located along the ventricular tube at equal intervals of 1 cm as described previously (Hashimoto 1982). Insulated silver wire leads with a diameter of 0.1 mm were passed through the catheter. The catheter was introduced into the IIlrd and through the sylvian aqueduct to the IVth ventricle under radioscopic guidance. The wrist contralateral to stimulation served as the common reference for intracranial and scalp recordings. Square wave pulses of 0.1 msec duration were delivered at 4 / s e c over the median nerve at the wrist (cathode proximal). Stimulus intensity was adjusted to produce a small muscle twitch of the thumb. Electrode impedances of less than 3 k~-2 were maintained. The surface electrode was placed over the
0013-4649/84/$03.00 '~' 1984 Elsevier Scientific Publishers Ireland, Ltd.
222
central region contralateral to the stimulated median nerve. The input signals were amplified with a frequency response down 3 dB at 30 and 3 kHz. One thousand responses were averaged for 25 msec employing a Nicolet Ca-1000 averager. The averages were written out with an X-Y plotter. Recordings were repeated twice to test reproducibility. The negativity of the intracranial and scalp electrodes was up in all recordings. The components were labeled by their polarities and peak latencies as recommended by Donchin et al. (1977). After the recordings were taken from the 4 electrodes, the catheter was withdrawn 5 cm from the initial position for recordings from more rostral loci. The recording electrode position within the IVth and IIIrd ventricles was confirmed by a metrizamide ventriculography.
1. HASHIMOTO
negative and positive (P22) waves. The negative wave was clearly bilobed (N15 and N20) with a positive notch (P17) separating this slow wave, and followed 4 small positive waves (P9, P l l , P13, and P14) (bottom trace in Fig. 2). Intracranial poten-
A
B
C
D Result
P13/~ The recording sites were mapped on the lateral view of the X-ray photo of the ventricular system (Fig. 1). The depth (from A to H in Fig. ]) of recording was given for each trace (Fig. 2). The potentials recorded from the contralateral central region, referenced to the wrist, were dominated by
E
F-
G
H
l uV
N15yN20 pllp13 j
J
i
i
i
12.5 msec Fig. 1. The recording sites of short latency somatosensory evoked potentials were mapped on the lateral view of the X-ray photo of the ventricular system and were indicated alphabetically. The sites A and B are in the IVth ventricle (vicinity of the pons); C and D in the aqueduct of Sylvius (the midbrain); E, F, and G in the lllrd ventricle (the thalamus); and H in the lateral ventricle.
~..~z P22 J
i
L
0.5 uV
i
25
Fig. 2. Short latency somatosensory evoked potentials recorded at the locations indicated in Fig. 1. S denotes the surface recording at the central region contralateral to the stimulated median nerve. The common reference is the contralateral wrist. The voltage calibration of 1 ktV applies to all intracranial recordings, while that of 0.5 /tV refers to the surface recording. Negativity is up in all traces.
DEPTH RECORDINGS OF EARLY SEP rials were characterized by 2 or 3 fast positive components ( P l l , P13, and P14) with approximately 1 msec duration and a large slow negative and positive component. The latency of P l l remained appreciably consistent in all traces and the amplitude showed only a slight change in recordings from caudal to more rostral electrodes. The peak latency of the component corresponds in time to that of P l l over the scalp. By contrast, P13 increased its peak latency with a progressively more rostral recording site up to the posterior IIIrd ventricle (site E). Radiographically, this electrode (E) lies nearest to thalamus relay nucleus ventralis posterolateralis (VPL). Rostrally, P13 was seen as a sm~ll notch at the same latency with the surface P13. P13 also showed a gradual increase and appeared largest at electrodes (C and D) within the aqueduct of Sylvius. The P14 component was absent in the recording from the most caudal electrode (A), and its peak latency showed a progressive increase from caudal to rostral and finally coincided with that of the surface P14. The negativity of about 6 msec duration which follows the 3 fast positive wavelets was the largest component in the intracranial SEPs. Its peak latency systematically increased rostrally. The latency shift recorded between the electrodes A and H exceeded 3 msec. The amplitude of the component measured from the onset of P13 remained almost the same in the recordings from the brain-stem, but was attenuated between the electrodes D and E and again very slowly further rostrally. The positive wave following the large negative wave was much smaller in amplitude in all recordings and its peak latency varied according to the depth electrode locations. The shortest latency of this component was recorded at electrode A and the longest at electrode H, and the latency shift was systematic, similar to the preceding negative wave. Discussion
Short latency SEPs recorded within the IVth and IIIrd ventricles and the sylvian aqueduct reflect activities generated from somatosensory path-
223 ways in close proximity to the electrodes and also volume-conducted potentials remote from the electrode sites. There has been considerable evidence to show that SEPs are mediated solely by the dorsal column-lemniscal system (Halliday and Wakefield 1963; Larson et al. 1966: Namerow 1968; Noel and Desmedt 1975, 1980; Desmedt and Cheron 1980). The earliest response from the intracranial recordings is the P l l with an onset latency of 10.3 msec which coincides in time with the surface P11 component. This potential exhibits little or no latency shift along the electrode array and the amplitude (less than 0.3 /~V) is not significantly changed. Therefore, this component recorded from the brain-stem reflects the far-field activity from below the brain-stem. We infer that P l l represents activities in fibers ascending the cuneate fasciculus, as suggested by Desmedt and Cheron (1980, 1981a) on grounds that the N l l onset closely corresponds to spinal entry time and that a clear latency shift of about 1 msec can be recorded for N l l between the lower and upper neck. The observations by Anziska and Cracco (1980) that P2 ( P l l in our nomenclature) peak latency was normal in patients with lesions rostral to the medulla and also direct recordings from the spinal cord by Lesser et al. (1981) are consistent with this interpretation. In contrast to P l l which behaves as a volumeconducted far-field potential, the intracranially recorded P13 exhibits a systematic change in latency from the pons (electrode A) to the diencephalon (electrode E). Furthermore, the wave form and amplitude of the component are almost the same in two different electrodes (electrodes C and D) within the aqueduct of Sylvius where the medial lemniscus runs fairly parallel with it. This implies that the potential must represent a synchronized axonal volley traveling in fiber tract al the level of these recording electrodes. The sudden amplitude reduction between the midbrain and thalamus (electrodes D and E) may be due to the fact that the medial lemniscus shifts its course: laterally at this level. Since electrodes located beyond the thalamic relay nucleus only see the volume-conducted far-field potential of this component, the potential from the rostral electrodes shows no
224
appreciable latency shift and slow amplitude attenuation toward the scalp. Desmedt and Cheron (1980) suggested that the P13 component originates from the medial lemniscus on the basis of the potential distribution using non-cephalic reference recording, coupled with their anatomical measurements of the somatosensory system in human cadavers. The absence or abnormally delayed peak latency of this P13 component in patients with brain-stem lesions caudal to the diencephalon (Anziska and Cracco 1980) is consistent with the present observation. Arezzo et al. (1979) have provided further evidence in monkeys with use of the semimicroelectrodes that the potential analogous to human P13 corresponds to the synchronized volleys of the action potentials within the medial lemniscus as it courses from the brain-stem to the thalamus. The intracranial P14 component which follows P13 by less than 1 msec at the rostral brain-stem (site D) appears to travel toward the rostral recording sites while P13 remains of the same latency beyond the midbrain. As a result, P13 and P14 are separated by about 1.5 msec at the surface. Recordings from the h u m a n midbrain medial lemniscus during stereotactic surgery for intractable pain demonstrated a potential with an onset latency of 11-13 msec (Liberson et al. 1970) and a peak latency of 14-16 msec (Strassburg et al. 1979). Arezzo et al. (1979) and Allison and Hume (1981) observed in their comparative studies in the monkey and the cat that the potential which appears homologous to the human P14 reflects bursts of action potentials traveling along the medial lemniscus and the initial portion of the thalamocortical projections. H u m e and Cant (1978), Nakanishi et al. (1978), and Anziska and Cracco (1980) demonstrated a normal P14 component in their patients with thalamic lesions. Taken together, these data suggest that the earlier portion of scalp P14 primarily reflects activity within the medial lemniscus and its branches to the various brain-stem nuclei (see the later discussion on the anatomy of dorsal columnolemniscal system). Following these potentials, the intracranial electrodes record a negative component which is the largest potential within 25 msec after the median nerve
I. H A S H I M O T O
stimulation. This potential is characterized by its definitely longer duration (approximately 6 msec) compared with the preceding small wavelets of brief duration (roughly 1 msec) and its slower traveling time in a caudo-rostral direction. While the preceding 4 wavelets have reasonably been ascribed to the axonal volleys ascending the dorsal column-lemniscal system, the question arises as to whether this slow negative wave represents the axonal volleys or postsynaptic potentials. Spinal cord potentials recorded intrathecally (Ertekin 1976) and epidurally (Shimoji et al. 1978) in man consist of the initial positive (P1) component of brief duration (1.8 3.2 msec) followed by a negative (N1) component with a longer duration (5.2--8.0 msec). P1 is considered to represent the incoming volleys through the dorsal roots and N1, the synaptic excitation of dorsal horn interneurons (Beall et al. 1977). The slow negative potential at the brain-stem and diencephalon closely resembles in configuration the N1 from the spinal cord. Provided that this negativity represents postsynaptic events rather than axonal activities, the possible neural structures responsible for its generation are the nucleus cuneatus in the medulla and the VPL in the thalamus which are the two main relay nuclei in the dorsal column-lemniscal system. However, the systematic latency shift as well as the similar amplitude and consistent duration of this component along the electrode array within the brain-stem would argue against the interpretation that this component is produced solely by overlapping of the activities from these two fixed sources. This systematic latency shift of the slow negativity within the brain-stem would lead us to postulate the existence of nuclear structures within the pons and midbrain which may be activated progressively by incoming volleys from the ascending lemniscal system. Recently, the nucleus cuneatus was found to project to the various brain-stem nuclei such as the accessory olives, the external nucleus of the inferior colliculus, the superior colliculus, and the nucleus tuber (Hand and Van Winkle 1976), providing an anatomical basis for the interpretation of the traveling slow negative wave in the brain-stem. Thus the surface N15 component may correspond to a traveling slow negative wave in the pontine and midbrain nuclei.
DEPTH RECORDINGS OF EARLY SEP
Direct recordings from the human thalamus have been described previously (Ervin and Mark 1964; Pagni 1967; Goto et al. 1968; Larson and Sances 1968; Haider et al. 1972; Fukushima et al. 1976; Sances et al. 1978; Celesia 1979). The initial potential of the recordings from the VPL following the median nerve stimulation has a peak latency of 12-18 msec and a duration of 6-10 msec which is close to our slow negative component in spite of a great variability in the amplitude and peak latency in their recordings. These data are so variable that they seem difficult to reconcile. Goto et al. (1968) found, however, by moving the recording electrode step by step within the thalamus, that the potential is diphasic and the initial positive phase is followed by a slow negative potential within the region of the nucleus ventralis intermedius (Vim). They found also that, below the ventral border of this nucleus, the slow negative component disappears and only the initial positive component is present (see Fig. 5 in Goto et al. 1968). This suggests that different neural structures are involved in the generation of their 'thalamic' potential; the initial positive component may represent axonal volleys in the medial lemniscus and the slow negative component, postsynaptic activities in the thalamic nucleus (VPL or Vim). There is no detectable activity in the depth recordings within the range for the surface N20 and P22. Thus N20 and P22 can be ascribed to a more rostral structure beyond the thalamus, presumably the primary somatosensory cortex (Desmedt and Cheron 1980, 1981b; Allison and Hume 1981).
Summa~ Short latency somatosensory evoked potentials (SEPs) were recorded in man from an array of electrodes within the IVth (the pons) and IIIrd ventricles (the thalamus) and the aqueduct of Sylvius (the midbrain). The slow negative (N15 and N20) and positive (P22) waves were preceded by 4 small positive (P9, P l l , P13, and P14) waves of approximately 1 msec duration in scalp recordings. The sources of these waves have been differentiated on the basis of their timing and spatial
225
gradients of corresponding intracranial potentials. While P l l is identified as volume conducted from below the brain-stem, P13 and P14 reflect synchronized volleys of the medial lemniscus and its branches to the various pontine and mesencephalic nuclei. In contrast to these earlier positive wavelets arising from fiber tracts, N15 may represent postsynaptic activities within these pontine and midbrain nuclei. N20 and P22 are the initial responses of the juxtarolandic cortex.
R~sum~ Potentiels OvoquOs somatosensoriels du tronc cOr~bral chez l'homme: origine des potentiels gl courte latence
Les composantes ~ courte latence des potentiels evoques somatosensoriels (PES) ont 6t6 enregistr6es chez l'homme a partir d'61ectrodes mobiles 'a l'interieur du 46me ventricule (voisinnage du pont), de l'aqueduc de Sylvius (mesenc+phale), du 36me ventricule (thalamus) et du ventricule lateral. Les ondes lentes negatives (N15 et N20) et positives (P22) 6taient preced6es de 4 petites ondes positives (P9, P l l , P13 et P14) d'une duree approximative de 1 msec dans les enregistrements sur le scalp. Les origines de ces ondes ont 6t6 differenciees par leur chronologie et les gradients spatiaux des potentiels intracrfiniens correspondants. Alors que P11 est identifiee comme lice ~ une conduction en volume ~ partir du tronc cerebral, P13 et P14 refl6tent des volees synchrones du lemnisque median et de ses branchements vers les divers noyaux pontiques et mesenc6phaliques. Contrairement a ces petites ondes positives precoces, en provenance de fasceaux de fibres, la N15 pourrait representer les activites postsynaptiques des noyaux pontiques et mesencephaliques eux-memes. N20 et P22 sont les reponses initiales du cortex juxtarolandique. The author wishes to thank Prof. John E. Desmedt, Brain Research Unit. University of Brussels, Brussels and Prof. Truett Allison. Department of Neurology, Yale University School of Medicine, New Haven, CT, for reviewing the manuscript and for helpful discussion, and to acknowledge the valuable advice from Dr. Muneo Shimamura, Department of Neurophysiology, Tokyo Metropolitan Institute of Neurosciences, Tokyo, and
226 Dr. Buichi lshijima, Department of Neurosurgery, Tokyo Metropolitan Neurological Hospital, Tokyo, in the preparation of the manuscript.
References Albe-Fessard, D., Arfel, G., Guiot. G., Hardy, J., Vourch, G., Hertzog, E., A16onard, P. et Derome, P. D6rivations d'activitbs spontan6es et evoqu6es darts les structures c6r6brales profoundes de l'homme. Rev. neurol., 1962, 106: 89 105. Allison, T. and Hume, A.L. A comparative analysis of shortlatency somatosensory evoked potentials in mare monkey, cat, and rat. Exp. Neurol., 1981, 72:592 611. Anziska, B. and Cracco, R.Q. Short latency somatosensory evoked potentials. Studies in patients with focal neurological disease. Electroenceph. clin. Neurophysiol., 1980, 49: 227 239. Anziskm B., Cracco, R.Q., Cook, A.W. and Feld, E.W. Somatosensory far-field potentials. Studies of normals and patients with multiple sclerosis. Electroenceph. clin. Neurophysiol., 1978, 45: 602-610. Arezzo, J., Legatt, A.D. and Vaughan, H.G. Topography and intracranial sources of somatosensory evoked potentials in the monkey. I. Early components. Electroenceph. clin. Neurophysiol., 1979, 4 6 : 1 5 5 172. Beall, J.E., Applebaum, A.E., Foreman, R.D. and Willis, W.D. Spinal cord potentials evoked by cutaneous afferents in the monkey. J. Neurophysiol., 1977, 40: 199-211. Celesia, G.G. Somatosensory evoked potentials recorded directly from h u m a n thalamus and SMI cortical area. Arch. Neurol. (Chic.), 1979, 36: 399-405. Chiappa, K.H., Choi, S.K. and Young, R.R. Short latency somatosensory evoked potentials following median nerve stimulation in patients with neurological lesions. In: J.E. I)esmedt (Ed.), Clinical Uses of Cerebral, Brainstem and Spinal Somatosensory Evoked Potentials. Progr. clin. Neurophysiol., Vol. 7. Karger, Basel, 1980: 264-281. Cracco, R.Q. and Cracco, J.B. Somatosensory evoked potentials in man: far-field potentials. Electroenceph. clin. Neurophysiol., 1976, 4 1 : 4 6 0 - 4 6 6 . De Divitiis, E., Giaquinto, S. and Signorelli, C.D. Peripheral influences on VPM-VPL thalamic nuclei in the human: a study on evoked potentials. Confin. neurol. (Basel), 1971. 33: 174-185. DesmedL J.E. Somatosensory cerebral evoked potentials in man. In: A. Remond (Ed.), Handbook of Electroenceph. ctin. Neurophysiol., Vol. 9. Elsevier, Amsterdam, 1971: 55-81. Desmedt, J.E. and Cheron, G. Central somatosensory conduction in man: neural generators and interpeak latencies of the far-field components recorded from neck and right or left scalp and earlobes. Electroenceph. clin. Neurophysiol.. 1980, 50: 328-403. DesmedL J.E. and Cheron, G. Prevertebral (oesophageal) re-
I. H A S H I M O T O cording of subcortical somatosensory evoked potentials in man: the spinal P13 component and the dual nature of the spinal generators. Electroenceph. clin. Neurophysiol., 1981a, 52: 257-275. Desmedt, J.E. and Cheron, G. Non-cephalic reference recording of early somatosensory potentials to finger stimulation in adult or aging normal man: differentiation of widespread N18 and contralateral N20 from the prerolandic P22 and N30 components. Electroenceph. din. Neurophvsiol., 1981b, 52:553 570. Desmedt, J.E., Franken, L., Borensteim S., l)ebecker, J., Lambert, C. et Manil, J. Le diagnostic des ralentissements de la conduction aff6rente dans les affections des nerfs p6riph6riques: intbr6t de l'extraction du potentiel evoqu,5 cer+bral. Rev. neurol., 1966, 115:255 262. Donchin, E., Callaway, E., Cooper, R., Desmedt, J.E., Goff, W.R., Hillyard, S.A. and Sutton, S. PublicatiOn criteria for studies of evoked potentials. Report of a committee. In: J.E. Desmedt (Ed.), Attention, Voluntary Contraction and Event-Related Cerebral Potentials. Progr. olin. Neurophysiol.. Vol. 1. Karger, Basel, 1977:1 11. Ertekin, C. Studies on the h u m a n evoked elecUospinogram: tile origin of the segmental evoked potentials. Acta. neurol. scan&, 1976, 53: 3-20. Ervin, F.R. and Mark, V.M. Studies of the h u m a n thalamus. IV. Evoked responses. Ann. N.Y. Acad. Sci., 1964. 112: 81 92. Fukushima, T., Mayanagi, Y. and Bouchard, G. Thalamic evoked potentials to somatosensory stimulation in man. Electroenceph. clin. Neurophysiol., 1976, 40: 481-490. Goto, A., Kosaka, K., Kubota, K., Nakamura, R. and Narabayashi, H. Thalamic potentials from muscle afferents in the human. Arch. Neurol. (Chic.), 1968, 19:302 309. Green, J.G. and Mcleod, S. Short latency somatosensory evoked potentials in patients with neurological lesions. Arch. Neurol. (Chic.), 1979, 36: 846-851. Haider, M., Gangleberger, J.A. and Groll-Knapp, E. Computer analysis of subcortical and cortical evoked potentials and of slow potential phenomena in humans. Confin. neurol. (Basel), 1972, 3 4 : 2 2 4 229. Halliday, A.M. and Wakefield. G.S. Cerebral evoked potentials in patients with dissociated sensory loss. J. Neurol. Neurosurg. Psychiat., 1963, 26: 211-219. Hand, P.J. and Van Winkle, T. The efferent connections of the feline nucleus cuneatus. J. comp. Neurol., 1976, 171:83 110. Hashimoto, I. Auditory evoked potentials from the human midbrain: slow brain stem responses. Electroenceph. olin. Neurophysiot., 1982, 53: 652-657. Hume, A.M. and Cant, B.R. Conduction time in central somatosensory pathways in man. Electroenceph. clin. Neurophysiol., 1978, 45: 361-375. Jasper, H., Lende, R. and Rasmussen, T. Evoked potentials from the exposed somatosensory cortex in man. J. nerv. ment. Dis., 1960, 130:526 537. Kimura, J., Yamada, T. and Kawamura, H. Central lalencies of somatosensory cerebral evoked potentials. Arch. Neurol. (Chic.), 1978, 35:683 688.
DEPTH RECORDINGS OF EARLY SEP Kritchevsky, M. and WiederholL W.C. Short latency somatosensory evoked responses in man. Arch. Neurol. (Chic.), 1978, 35:706 711. Larson, S.J. and Sances, A. Averaged evoked potentials in stereotaxic surgery. J. Neurosurg., 1968, 28: 227-232. Larson, S.J., Sances, A. and Christenson, P.C. Evoked somatosensory potentials in man. Arch. Neurol. (Chic.), 1966, 15: 88-93. Lesser, R.P., Lueders, H., Hahn, J. and Klem, G. Early somatosensory potentials evoked by median nerve stimulation: intraoperative monitoring. Neurology (Minneap.)~ 1981, 31:1519 1523. Liberson, W.T., Voris, H.C. and Uematsu, S. Recording somatosensory evoked potentials during mesencephalotomy for intractable pain. Confin. neurol. (Basel), 1970, 32: 185-194. Nakanishi, T., Shimada, Y., Sakuta, M. and Toyokura, Y. The initial positive component of the scalp-recorded somatosensory evoked potential in normal subjects and in patients with neurological disorders. Electroenceph. clin. Neurophysiol., 1978, 45: 26-34. ! Namerow, N.W. Somatosensory evoked responses in multiple sclerosis patients with varying sensory loss. Neurology (Minneap.), 1968, 18: 1197-1204. Noi)l, P. and Desmedt, J.E. Somatosensory cerebral evoked potentials after vaseular lesions of the brain stem and diencephalon. Brain, 1975, 98: 113-128.
227 Noi~l, P. and Desmedt, J.E. Cerebral and far-field somatosensory evoked potentials in neurological disorders involving the cervical spinal cord, brainstem, thalamus and cortex. In: J.E. Desmedt (Ed.), Clinical Uses of Cerebral, Brainstern and Spinal Somatosensory Evoked Potentials. Progr. clin. Neurophysiol., Vol. 7. Karger, Basel, 1980: 205-230. PagnL C.A. Somato-sensory evoked potentials in thalamus and cortex of man. In: P. Gloor (Ed.), Hans Berger on the EEG of Man. Electroenceph. clin. Neurophysiol., Suppl. 28. Elsevier, Amsterdam, 1967: 147-155. Sances, Jr., A.~ Larson, S.J., Cusick, J.F., Myklebust, J., Ewing, C., Jodat, R., Ackmann, J.J. and Walsh, P. Early somatosensory evoked potentials. Electroenceph. clin. Neurophysiol., 1978, 45:505 514. Shimoji, K., Shimizu, H. and Maruyama, Y. Origin of the somatosensory evoked responses recorded from the cervical skin surface. J. Neurosurg., 1978, 48:980 984. Strassburg, H.M., Thoden, U. and Mundinger, F. Mesencephalic chronic electrodes in pain patients. An electrophysiological study. Appl. Neurophysiol., 1979, 42: 284-293. Yamada, T., Kimura, J. and Nitz, D.M. Short latency somatosensory evoked potentials following median nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1980, 48: 367-376.