- Email: [email protected]
Conduction characteristics of somatosensory evoked potentials to peroneal, tibial and sural nerve stimulation in man
Electroencephalography and clinical Neurophysiology, 1987, 68:287-294 Elsevier Scientific Publishers Ireland, Ltd.
287
EEG 03255
Conduction characteristics of somatosensory evoked potentials to peroneal, tibial and sural nerve stimulation in man Luciana Pelosi
1, Joan B.
Cracco and Roger Q. Cracco
Department of Neurology, State University of New York, Health Science Center at Brooklyn, Brooklyn, N Y 11203 (U.S.A.)
(Accepted for publication: 15 October, 1986)
Summary Lumbar spine and scalp short latency somatosensory evoked potentials (SSEPs) to stimulation of the posterior tibial, peroneal and sural nerves at the ankle (PTN-A, PN-A, SN-A) and common peroneal nerve at the knee (CPN-K) were obtained in 8 normal subjects. Peripheral nerve conduction velocities and lumbar spine to cerebral cortex propagation velocities were determined and compared. These values were similar with stimulation of the 3 nerves at the ankle but were significantly greater with CPN-K stimulation. CPN-K and PTN-A SSEPs were recorded from the L3, T12, T6 and C7 spines and the scalp in 6 normal subjects. Conduction velocities were determined over peripheral nerve-cauda equina (stimulus-L3), caudal spinal cord (T12-T6) and rostral spinal cord (T6-C7). Propagation velocities were determined from each spinal level to the cerebral cortex. With both CPN-K and PTN-A stimulation lhe speed of conduction over peripheral nerve and spinal cord was non-linear. It was greater over peripheral nerve-cauda equina and rostral spinal cord than over caudal cord segments. The CPN-K response was conducted significantly faster than the PTN-A response over peripheral nerve-cauda equina and rostral spinal cord but these values were similar over caudal cord. Spine to cerebral cortex propagation velocities were significantly greater from all spine levels with CPN-K stimulation. These data show that the conduction characteristics of SSEPs over peripheral nerve, spinal cord and from spine to cerebral cortex are dependent on the peripheral nerve stimulated. Key words: Somatosensory evoked potentials; Conduction characteristics
Short latency s o m a t o s e n s o r y potentials (SSEPs) to s t i m u l a t i o n of nerves in the lower extremity which arise in the c a u d a e q u i n a a n d in spinal cord afferent p a t h w a y s have been described in m a n using b o t h n o n - i n v a s i v e (surface) a n d invasive (intrathecal, epidural, s p i n o u s process or ligament) recording teclmiques (Shimoji et al. 1972, 1977; R. Cracco 1973; J. Cracco et al. 1975, 1979; Ertekin 1976a; Jones a n d Small 1978; Terao et al. 1979; Lueders et al. 1981; Jones et al. 1982; Kakigi et al.
1 Present address: Dept. of Clin. Neurophysiol., II School of Medicine, Naples, and Dept. of Neurol., Fondazione Clinica del Lavoro, Campoli, Italy. Correspondence to: Joan B. Cracco, MD, Dept. of Neurol., SUNY Health Science Center at Brooklyn, 450 Clarkson Avenue, Brooklyn, NY 11203 (U.S.A.).
1982; M a c o n a n d Poletti 1982; Y a m a d a et al. 1982; D e s m e d t a n d C h e r o n 1983; Maccabee et al. 1983; Seyal et al. 1983). SSEPs to posterior tibial nerve s t i m u l a t i o n at the ankle were recorded in most of these investigations (Jones a n d Small 1978; T e r a o et al. 1979; Lueders et al. 1981; Jones et al. 1982; Kakigi et al. 1982; Y a m a d a et al. 1982; D e s m e d t a n d C h e r o n 1983; Seyal et al. 1983); however, responses to posterior tibial or c o m m o n p e r o n e a l nerve s t i m u l a t i o n at the knee were obtained in a few (R. Cracco 1973; J. Cracco et al. 1975, 1979; Ertekin 1976b; Jones et al. 1982; M a c o n a n d Poletti 1982). Recordings were performed at different spinal levels in each of these studies a n d c o n d u c t i o n distances were n o t always m e a s u r e d in the same way. Because of these variations in s t i m u l a t i n g a n d recording techniques, different values for c o n d u c t i o n characteristics of
0168-5597/87/$03.50 © 1987 Elsevier Scientific Publishers Ireland, Ltd.
288
spine SSEPs were obtained in these studies and the data are difficult to compare. Therefore, in this study, SSEPs to stimulation of 4 nerves in the lower extremity were recorded from the spine and scalp in normal subjects and conduction characteristics along peripheral nerve and spine and from spine to scalp were compared using standardized recording methods.
Methods
Eleven slender normal adult subjects (6 males, 5 females, mean age 24.9 years) were studied. Recordings were performed with subjects unsedated and relaxed or drowsy in the prone position. Six millimeter tin disks attached to the skin with collodion and filled with conductive jelly served as both stimulating and recording electrodes. Recording electrode impedance was maintained under 3000 O. Stimulating electrodes were placed 3 cm apart (cathode proximal) along the nerve. Constant voltage stimuli (0.2 msec duration) were delivered to the left common peroneal nerve at the knee in the popliteal fossa (CPN-K) and to the left posterior tibial, sural and peroneal nerves at the ankle (PTN-A, SN-A, PN-A). Nerves were stimulated one at a time. With mixed nerve (CPN-K, PTN-A and PN-A) stimulation, the stimulus intensity was adjusted to produce vigorous but non-painful foot movement. With sensory nerve (SN-A) stimulation an intensity 2-2.5 times sensory threshold was used which resulted in non-painful paresthesia on the lateral aspect of the foot. With respect to sensory threshold, the stimulus intensity was the same for all nerves studied and varied between 2 and 2.5 times threshold. In 6 subjects SSEPs to C P N - K and PTN-A stimulation were recorded over 4 spine levels and from the scalp. Spine bipolar recordings were obtained by attaching the grid I electrodes over the L3, T12, T6 and C7 spinous processes and thegrid II electrodes 4 cm rostral to these locations. Scalp SSEPs were recorded from Cz'-Fpz' (2 cm behind Cz and FPz, International 10-20 system) and Cz'NC leads (the non-cephalic reference electrode (NC) was placed on the shoulder contralateral to
L. PELOSI ET AL.
the stimulated leg). Spine and scalp SSEPs were obtained separately using stimulus rates of 4.76.7/sec for spinal and 1.7/sec for scalp recordings and analysis times of the 35-40 msec for spine and 50-60 msec for scalp SSEPs. Computer sampling time was 116 and 192 /~sec/point for spine and scalp recordings, respectively. For spine recordings 2 - 4 trials of 2000-4000 averages were superimposed and for scalp recordings 2 trials of 500-1000 averages were superimposed. Recordings were also obtained over the tibial nerve in the popliteal fossa during PTN-A stimulation in 4 of these subjects; this recording electrode was placed 4 cm above the popliteal crease and the reference electrode was placed on the medial surface of the knee. Filter bandpass for all recordings was 5-1500 Hz and relative negativity in grid I produced an upward deflection. The onset latencies of the first negative potential recorded over peripheral nerve and spine (L3, T12, T6 and C7) were measured. This indicates the approximate time when impulses in the fastest fibers pass under the recording electrode (Gilliatt et al. 1965). For scalp recordings the peak latency of the prominent positive potential which precedes the first well defined negative potential served as the latency indicator. This was P27 with CPN-K and P37 with PTN-A stimulation. The peak of these components is thought to reflect activity arising in somesthetic cortex (Tsumoto et al. 1972; Rossini et al. 1981; Vas et al. 1981; Cruse et al. 1982; Desmedt and Bourguet 1985). Distances along the course of the nerves from the stimulating cathode at the knee or ankle to the L3 recording electrode were measured and stimulus to L3 conduction velocities (CVs) were calculated. In 4 subjects in whom potentials to PTN-A stimulation were recorded over the popliteal fossa, the onset latency of the response recorded at the knee was subtracted from that recorded at L3 to determine the CV in the knee to L3 segment. Distances along the skin surface between the L3, T12, T6 and C7 spines and from these spines to the scalp recording electrode (Cz') were measured with the subject prone. CVs for CPN-K and PTNA SSEPs were determined in the following segments: (1) peripheral nerve and cauda equina (stimulating cathode to L3); (2) cauda equina to
CONDUCTION CHARACTERISTICS OF SPINE AND SCALP SSEPs cervical spinal cord (L3 to C7); (3) spinal cord entry zone to cervical spinal cord (T12 to C7); (4) caudal spinal cord (T12 to T6); (5) rostral spinal cord (T6 to C7). Spine to cortex (L3, T12, T6, C7 to C z ' ) propagation velocities (PVs) were also calculated. The term propagation velocity was used for the spine to scalp segments since the term c o n d u c t i o n velocity implies that responses originating in the same fibers are recorded at 2 different sites 2. The mean, standard deviation and range were determined for each parameter. These values for C P N - K and PTN-A SSEPs were compared using the Student t test (Table I). In 8 subjects CPN-K, PTN-A, SN-A and PN-A were stimulated at rates of 1.7-4.7/sec and simultaneous recordings were obtained from a bipolar lead over the lumbar spine (L3 to 4 cm rostral) and from .3 scalp (Cz, Cz', Pz)-NC leads. Using analysis times of 60-100 msec, 2 or 3 trials of 1500-2000 averaged responses were superimposed. Using the onset latency of the negative potential at L3 and the peak latency of the first prominent positive scalp potential, CVs were determined over peripheral nerve and cauda equina (stimulating cathode to L3) and PVs were calculated from cauda equina to cortex (L3 to Cz). Mean, standard deviation and range for these two measurements were calculated for each nerve stimulated (Table II). These values were compared using the Student t test. Amplitudes were also measured at L3 from the peak of the first negative to the peak of the following positive potential and at the scalp from the peak of P27 (CPN-K) or P37 (PTN-A, SN-A and PN-A) to the peak of N35 or N45.
Results
Comparison of spine and scalp SSEPs to CPN-K and PTN-A stimulation With C P N - K and P T N - A stimulation small triphasic potentials often with poorly defined initial positive phases were recorded over the lumbar, 2 F o r the s a m e r e a s o n a n a r g u m e n t c o u l d be m a d e that the t e r m c o n d u c t i o n velocity is also i n a p p r o p r i a t e for spine to spine s e g m e n t s - - see Discussion.
289 PTN-A Stimulation
m$~. I0 C7-4cm. rostral
CPN-K Stimulation
60
5
55
~
T6 -4cm. rostrol
I2o.~.~ Ti2-4cm- rostral
L3-4crn. rostral
~
mm~. I0
40
5
Popliteol Fosso mN¢. 2
32
Fig. 1. Spine and scalp recordings of SSEPs to stimulation of PTN-A and CPN-K in 2 subjects. The response recorded at the popliteal fossa with PTN-A stimulation is shown in the bottom trace. Note differences in analysis time. The onset latency of the negative potentials recorded over peripheral nerve and spine and the peak latency of the initial scalp positive potential are indicated.
thoracic and cervical spines. These potentials progressively increased in latency rostrally (Fig. 1). With stimulation of either nerve potentials were usually larger and broader at T12 than at more rostral or caudal sites and they were small and at times difficult to record at T6 and C7. Typically, they were better defined with C P N - K stimulation. Scalp SSEPs to both C P N - K and PTN-A stimulation consisted of an initial positive component followed by a larger negative potential (Fig. 1). These components were labeled P27-N35 for C P N - K and P37-N45 for P T N - A (AEEGS Guidelines for Clinical Evoked Potential Studies 1984). From stimulus to L3 the CV was significantly slower with PTN-A than with C P N - K stimulation (Table I and Fig. 2). In the 4 subjects in whom PTN-A CVs were also determined over the knee to L3 segment, this value of 64.4 m / s e c was 5.2 m / s e c slower than the comparable C P N - K value. CVs of PTN-A SSEPs over the L3-C7, T12-C7
290
L. PELOSI ET AL. IOO
CPN-K PTN-A
L 3- 4cm. rostrol
flail! £ii:,
-:..:;.:
\
40
"k.:Ei <:/::
20
!:iiii!;! ~.iiiiii:i:
0
~ k
~
{:(:%
80
6o
C~ -Shoulder
SUBJECA T CPN-K SSEPs
SUBJECAT
.:...~
PTN-A SSEPs
<<:i~!
Stimulus- L3
T6-C7
T,2-T6
SUBJECAT
Fig. 2. Comparisonof mean valuesof CVs of SSEPsto CPN-K and PTN-A stimulation.
a n d T6-C7 segments were also significantly slower t h a n those of C P N - K SSEPs (Table I). W i t h b o t h C P N - K a n d F F N - A s t i m u l a t i o n the speed of c o n d u c t i o n over the spine was n o n - l i n e a r (Fig. 2). It was greater over peripheral n e r v e - c a u d a e q u i n a than over caudal cord a n d significantly increased over rostral spinal cord (for b o t h C P N - K a n d P T N - A T 1 2 - T 6 / T 6 - C 7 c o m p a r i s o n P was 0.001). Over rostral spinal cord, C P N - K SSEPs were c o n d u c t e d 1.5 times faster than P T N - A SSEPs. In contrast, there was n o significant difference in the speed of c o n d u c t i o n over the T12-T6 segment (Table I a n d Fig. 2). Based o n observations o n h u m a n cadavers the actual spinal cord length from lower thoracic to
. ~A /
~
PN-A SSEPs
SUBJECB T SN-A SSEPs
I
/
i
5
55
5
I
I05 msec
Fig. 3. Spine and scalp SSEPs to CPN-K, PTN-A, PN-A and SN-A stimulation in 2 subjects (subjects A and B). Note differences in calibration. The onset latency of the spine potential and the peak latency of the scalp positive potential are indicated.
TABLE, I Mean, standard deviation and range of peripheral and spinal conduction velocity and spine to scalp propagation velocity of SSEPs to CPN-K and PTN-A stimulation in 6 subjects. Stimulus-L3 T12-T6
T6-C7
CPN-K Mean S.D. Range
L3-C7
T12-C7
L3-Cz'
69.65 _+4.7 63.8-73.5
48.91 +4.9 41.9-56.4
100.5 70.52 70.16 39.65 +9.1 _+5.2 _+3.67 +1.67 91.4-116.2 64.4-78.3 66.4-76.47 36.9-41.7
PTN-A Mean S.D. Range
60.06 +3.5 54.2-65.2
50.52 _+4.7 45-58.05
P=
< 0.001
0.57
T12-Cz'
T6-Cz'
C7-Cz'
37.77 34.61 _+1.58 _+1.3 35.13-39.03 32.8-36.3
24.27 +1.9 21.23-26.14
64.88 58.91 _+6.7 +4.86 56.9-74.1 51.9-64.9
55.17 34.39 32.73 29.28 +5.2 _+2.3 _+1.98 _+2.03 46.57-60.7 30.4-37.03 29.24-34.4 26.4-31.9
21.87 _+2.63 18.58-24.48
< 0.001
< 0.001
0.002
< 0.001
< 0.001
< 0.001
< 0.001
CONDUCTION CHARACTERISTICS OF SPINE AND SCALP SSEPs
cervical areas is almost 13% shorter than that measured over the skin surface (Desmedt and Cheron 1983). Correcting for this factor the mean values for PTN-A CVs were reduced from 55.17 to 47.99 m / s e c for T12 to C7 segment, from 50.52 to 43.95 m / s e c for T12 to T6 and from 64.88 to 56.44 m / s e c for T6 to C7. Mean values for CPN-K CVs were reduced from 70.16 to 61.03 m / s e c for T12 to C7 segment, from 48.91 to 42.55 m / s e c for T12 to T6 and from 100.5 to 87.43 m / s e c for T6 to C7. Spine to scalp PVs were slower with PTN-A than with C P N - K stimulation (Table I). They were about 5 m / s e c slower from L3, T12 and T6 to scalp and about 2 m / s e c slower from C7 to scalp.
Comparison of L3 spine and scalp SSEPs to CPN-K, PTN-A, PN-A and SN-A stimulation SSEPS were recorded over L3 in all 8 subjects with CPN-K and PTN-A stimulation and in 7 subjects with PN-A and SN-A stimulation. These responses consisted of initially positive triphasic potentials. They were consistently lower in amplitude with SN-A and PN-A stimulation than with C P N - K and PTN-A stimulation (Fig. 3). With stimulation of each of the 4 nerves, scalp SSEPs consisted of an initial positive potential followed by a larger negative potential. Scalp SSEPs were consistently greater in amplitude with stimulation of PTN-A, SN-A and PN-A than with CPN-K stimulation (Fig. 3). The mean peak latency of the initial positive component was greater with SN-A (42.8 msec), PN-A (40.6 msec) and PTN-A (39.3
291
msec) stimulation than with CPN-K stimulation (27.7 msec). This value for PTN-A (39.3 msec) was similar to that reported by Jones and Small (1978) (39.7 msec) and by Riffel et al. (1984) (38.8 msec) but was greater than that noted by Chiappa (1983) (36.3 msec). Our value for CPN-A (27.6 msec) was similar to that found by Chiappa (27.3 msec). Our peak latencies for PN-A (40.6 msec) and SN-A (42.8 msec) were similar to those reported by Eisen and Elleker (1980) (39,9 and 42.1 msec, respectively) but the values for SN-A were greater than those published by Chiappa (38.7 msec). Table II shows the mean, standard deviation and range of the peripheral CVs and L3-Cz PVs for the 4 nerves stimulated. The peripheral CV was greatest with CPN-K stimulation (CPNK/PTN-A P=0.03; CPN-K/PN-A P=0.01; C P N - K / S N - A P = 0 . 0 0 1 ) . However, peripheral CVs were similar for all 3 nerves stimulated at the ankle and differences were not statistically significant. The PV from L3 to Cz was also greatest with CPN-K (CPN-K/PTN-A P = 0.004; CPNK / P N - A P = 0.003; C P N - K / S N - A P = < 0.001). In contrast, there were only slight differences in these mean values which were not statistically significant among the 3 nerves stimulated at the ankle. These PV values with CPN-K stimulation are somewhat greater than those previously reported from this laboratory (Schiff et al. 1984). This can be attributed to differences in methods used for measuring distances from spine to scalp. In the earlier study spine to scalp straight-line distances were used rather than measurements along the spine and scalp. In comparing these
TABLE II Peripheral nerve conduction velocity and spine to scalp propagation velocity of SSEPs to CPN-K, PTN-A, PN-A and SN-A stimulation in 7 subjects. CPN-K
PTN-A
PN-A
SN-A
Stimulus-L3 Mean S.D. Range
69.26 + 6.60 58.70-78.10
62.17 + 5.04 54.50-66.60
60.90 _+3.69 55.08-64.50
57.48 + 3.66 52.57-60.27
L3-Cz Mean S.D. Range
38.34 + 3.17 35.80-43.40
32.45 + 3.35 26.60-37.60
31.29 + 3.93 27.30-37.60
29.25 + 3.06 24.20-33.14
292 methods we found distances using a straight-line measurement to be about 10% less.
Discussion
Evoked potentials to stimulation of peripheral nerves have been recorded from individual spinal cord afferent tracts such as the dorsal and dorsolateral columns using small intradural electrodes in animals (Grundfest and Campbell 1942; Lloyd and McIntyre 1950). In contrast, large extradural or surface (skin) electrodes record spinal potentials arising in multiple afferent pathways (R. Cracco 1973; Happel et al. 1975; Sarnowski et al. 1975; R. Cracco and Evans 1978; Ducati and Schieppati 1980; Feldman et al. 1980; Jones et al. 1982). In human and animal extradural recordings the response recorded over mid and rostral spinal cord segments consists of several negative peaks (Happel et al. 1975; Sarnowski et al. 1975; R. Cracco and Evans 1978; Jones et al. 1982; Pelosi et al. 1985). The peak and interpeak latencies of these components progressively increase at more rostral recording locations (R. Cracco and Evans 1978). This is consistent with multiple afferent volleys which arise in fiber groups having different CVs. Studies in the cat, which compared CV over peripheral nerve and spinal cord to stimulation of lower extremity muscle and cutaneous afferent fibers, have shown that the cutaneous nerve evoked response has a CV of about 80 m / s e c over peripheral nerve-cauda equina. This volley slows to about 60-70 m / s e c over lower lumbar cord segments because of fiber branching and is then conducted rostrally at a similar velocity. In contrast, the CV of the response to stimulation of muscle nerves is greater than 100 m / s e c over peripheral nervecauda equina and over the dorsolateral fasciculus of the thoracic cord. However, the CV is considerably slower than that of the cutaneous nerve evoked response over lumbar cord segments because of synaptic delay and a greater degree of fiber branching (Lloyd and McIntyre 1950). The result of these complex relationships is that the volley to muscle nerve stimulation arrives at caudal cord segments before the cutaneous volley, is over-
L. PELOS1ET AL. taken by the cutaneous volley over caudal spinal cord and then again overtakes the cutaneous volley over rostral cord segments (Grundfest and Campbell 1942; Lloyd and McIntyre 1950). Hence, at least in the cat, the onset latency of the spinal cord compound action potential is probably not dependent on the same fibers at all spinal levels. Our data demonstrated that the peripheral and spinal CVs of the response to both CPN-K and PTN-A stimulation were non-linear. CVs were greater over peripheral nerve-cauda equina and rostral spinal cord than over caudal cord. CVs to PTN-A stimulation were greater over rostral spinal cord than over peripheral nerve-cauda equina when standard surface measurements were used. However, when correction was made for the overestimation of actual spinal cord length which results when using surface measurements (Desmedt and Cheron 1983), the adjusted CV was greater over peripheral nerve-cauda equina. With CPN-K stimulation the CV over rostral spinal cord was greater than it was over peripheral nerve-cauda equina using either standard or adjusted measurements. The slowing in speed of conduction over caudal cord can be explained on the basis of either fiber branching or synaptic delay or a combination of the two (Lloyd and McIntyre 1950). The increase in speed of conduction over rostral cord segments suggests that large diameter second order afferent fibers contribute to this response since the speed of conduction in first order fibers which ascend the length of the spinal cord would not be expected to increase in rostral cord segments (Lloyd and McIntyre 1950). The CV over caudal cord segments was similar with CPN-K and PTN-A stimulation. However, the CV over peripheral nerve-cauda equina and rostral cord and hence the degree of non-linearity in CV up the spine was much greater with CPN-K than with PTN-A stimulation. This indicates that the spinal response to CPN-K stimulation is mediated by more rapidly conducting pathways. This is not unexpected since rapidly conducting peripheral afferent fibers which supply muscle spindles and tendon organs would be expected to be more numerous at the knee than at the ankle. Additionally, it has been shown in primates that the ratio of afferent fibers arising from deep vs. cutaneous
CONDUCTION CHARACTERISTICS OF SPINE AND SCALP SSEPs
receptors is greater in proximal hind limb than in distal hind limb (Werner and Whitsel 1967; Whitsel et al. 1969). These afferents from deep receptors include the large group I fibers which synapse in caudal cord nuclei such as Clark's column which give rise to large rapidly conducting second order fibers. Second order cutaneous fibers such as those which ascend in Morin's tract (Morin 1955) could also contribute to the increase in CV over rostral spinal cord segments with both CPN-K and PTN-A stimulation. Peripheral CVs and spine to scalp (L3-Cz) PVs of SSEPs to PTN-A, PN-A and SN-A stimulation were all similar but significantly slower than those obtained with CPN-K stimulation. Lumbar, thoracic and cervical spine to scalp PVs with PTN-A stimulation were less than corresponding values obtained with CPN-K stimulation. These differences were least in the cervical spine to scalp segment. These data are consistent with the interpretation that the CPN-K response is mediated by larger faster conducting peripheral and spinal afferent fibers than those which mediate the response evoked by stimulation of nerves at the ankle. It also suggests that the central somatosensory pathways rostral to the cervical cord which mediate the response to stimulation of nerves at both ankle and knee are similar in diameter and conduction characteristics. Although this conclusion seems reasonable, it must be interpreted with caution since the fastest conducting fibers which contribute to the response and determine response latency may not be identical at different spinal levels and the fibers which determine the latency of the spinal potentials may not be the same as those which mediate the scalp response. The authors thank Mrs. Gracy George for her secretarial assistance.
References American Electroencephalographic Society (AEEGS). Guidelines for clinical evoked potential studies. J. clin. Neurophysiol., 1984, 1: 3-53. Chiappa, K.H. Evoked Potentials in Clinical Medicine. Raven Press, New York, 1983: 233-234. Cracco, J.B., Cracco, R.Q. and Graziani, L.J. The spinal evoked response in infants and children. Neurology (Minneap.), 1975, 25: 131-36.
293
Cracco, J.B., Cracco, R.Q. and Stolove, R. Spinal evoked potential in man: a maturational study. Electroenceph. clin. Neurophysiol., 1979, 46: 58-64. Cracco, R.Q. Spinal evoked response: peripheral nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1973, 35: 379-386. Cracco, R.Q. and Evans, B. Spinal evoked potential in the cat. Effects of asphyxia, strychnine, cord section and compression. Electroenceph. clin. Neurophysiol., 1978, 44: 187-201. Cruse, R., Klem, G., Lesser, R.P. and Lueders, H. Paradoxical lateralization of cortical potentials evoked by stimulation of posterior tibial nerve. Arch. Neurol. (Chic.), 1982, 39: 222-225. Desmedt, J.E. and Bourguet, M. Color imaging of parietal and frontal somatosensory potential fields evoked by stimulation of median or posterior tibial nerve in man. Electroenceph. clin. Neurophysiol., 1985, 62: 1-17. Desmedt, J.E. and Cheron, G. Spinal and far-field components of human somatosensory evoked potentials to posterior tibial nerve stimulation analysed with oesophageal derivations and non-cephalic reference recording. Electroenceph. clin. Neurophysiol., 1983, 56: 635-651. Ducati, A and Schieppati, M. Spinal pathways mediating somatosensory evoked potentials from cutaneous and muscle nerves in the cat. Acta neurochir. (Wien), 1980, 52: 99-104. Eisen, A. and Elleker, G. Sensory nerve stimulation and evoked cerebral potentials. Neurology (NY), 1980, 30: 1097-1105. Ertekin, C. Studies in the human evoked electrospinogram. I. The origin of the segmental evoked potentials. Acta neurol. scand., 1976a, 53: 2-20. Ertekin, C. Studies in the human evoked electrospinogram. II. The conduction velocity along the dorsal funiculus. Acta neurol, scand., 1976b, 53: 21-38. Feldman, M.H., Cracco, R.Q., Farmer, P. and Mount, F. Spinal evoked potential in the monkey. Ann. Neurol., 1980, 7: 238-244. Gilliatt, R.Q., Melville, I.P., Velate, A.S. and Willison, R.G. A study of normal nerve action potential using an averaging technique (barrier grid storage tube). J. Neurol. Neurosurg. Psychiat., 1965, 28: 191-200. Grundfest, H. and Campbell, B. Origin, conduction and termination of impulses in dorsal spinocerebellar tracts of cats. J. Neurophysiol., 1942, 5: 275-294. Happel, L.T., Le Blanc, H.J. and Kline, D.G. Spinal cord potentials evoked by peripheral nerve stimulation. Electroenceph, clin. Neurophysiol., 1975, 38: 349-354. Jones, S.J. and Small, D.G. Spinal and subcortical evoked potentials following stimulation of the posterior tibial nerve in man. Electroenceph. clin. Neurophysiol., 1978, 44: 299-306. Jones, S.J., Edgar, M.A. and Ransford, A.O. Sensory nerve conduction in the human spinal cord: epidural recordings made during scoliosis surgery. J. Neurol. Neurosurg. Psychiat., 1982, 45: 446-451. Kakigi, R. Shibasaki, H., Hashizume, A. and Kuroiwa, Y. Short-latency somatosensory evoked spinal and scalp-recorded potentials following posterior tibial nerve stimula-
294 tion in man. Electroenceph. clin. Neurophysiol., 1982, 53: 602-611. Lloyd, D.P.C. and Mclntyre, A.K. Dorsal column conduction of group I muscle afferent impulses and their relay through Clarke's column. J. Neurophysiol., 1950, 13: 39-54. Lueders, H., Andrish, J., Gurd, A., Weiker, G. and Klem, G. Origin of far-field subcortical potentials evoked by stimulation of the posterior tibial nerve. Electroenceph. clin. Neurophysiol., 1981, 52: 336-344. Maccabee, P.J., Levine, D.B., Pinkhasov, E.I., Cracco, R.Q. and Tsairis, P. Evoked potentials recorded from scalp and spinous processes during spinal column surgery. Electroenceph. clin. Neurophysiol., 1983, 56: 569-582. Macon, J.B. and Poletti, C.E. Conducted somatosensory potentials during spinal surgery. I. Control conduction velocity measurements. J. Neurosurg., 1982, 57: 349-353. Morin, F. A new spinal pathway for cutaneous impulses. Amer. J. Physiol., 1955, 183: 245-252. Pelosi, L., Caruso, G., Russo, G., Milano, C., Paolino, G., Lotti, G. and Greco, P. Spinal evoked potentials during surgery for scoliosis. Electroenceph. clin. Neurophysiol., 1985, 61: $28. Riffel, B., Stohr, M. and Korner, S. Spinal and cortical evoked potentials following stimulation of the posterior tibial nerve in the diagnosis and localization of spinal cord diseases. Electroenceph. clin. Neurophysiol., 1984, 58: 400-407. Rossini, P.M., Cracco, R.Q., Cracco, J.B. and House, W.J. Short-latency somatosensory evoked potentials to peroneal nerve stimulation: scalp topography and the effect of different frequency filters. Electroenceph. clin. Neurophysiol., 1981, 52: 540-552. Sarnowski, R.J., Cracco, R.Q., Vogel, H.B. and Mount, F. Spinal evoked response in the cat. J. Neurosurg., 1975, 43: 329-336. Schiff, J.A., Cracco, R.Q., Rossini, P.M. and Cracco, J.B. Spine
L. PELOSI ET AL. and scalp somatosensory evoked potentials in normal subjects and patients with spinal cord disease: evaluation of afferent transmission. Electroenceph. clin. Neurophysiol., 1984, 59: 374-387. Seyal, M., Emerson, R.G. and Pedley, T.A. Spinal and early scalp-recorded components of the somatosensory evoked potential following stimulation of the posterior tibial nerve. Electroenceph. clin. Neurophysiol., 1983, 55: 320-330. Shimoji, K., Kano, T., Higashi, H., Morioka, T. and Henschel, E.O. Evoked spinal electrograms recorded from epidural space in man. J. appl. Physiol., 1972, 33: 468-471. Shimoji, K., Matsuki, M. and Shimizu, H. Wave form characteristics in man. J. Neurosurg., 1977, 46: 304-310. Terao, A., Matsuda, E. and Fukunaga, H. Conduction velocity measurement in human spinal cord and tibial nerve using skin electrodes recording of the spinal evoked potential. Kawasaki med. J., 1979, 5: 71-78. Tsumoto, T., Hirose, N., Nonaka, S. and Takahashi, M. Analysis of somatosensory evoked potentials to lateral popliteal nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1972, 33: 379-388. Vas, G., Cracco, J.B. and Cracco, R.Q. Scalp recorded short latency cortical and subcortical somatosensory evoked potentials to peroneal nerve stimulation. Electroenceph. clin. Neurophysiol., 1981, 52: 1-8. Werner, G. and Whitsel, B.L. The topology of dermatomal projection in the medial lemniscal system. J. Physiol. (Lond.), 1967, 192: 123-144. Whitsel, B.L., Petrucelli, L.M. and Sapiro, G. Modality representation in the lumbar and cervical fasciculus gracilis of squirrel monkeys. Brain Res., 1969, 15: 67-78. Yamada, T., Machida, M. and Kimura, J. Far-field somatosensory evoked potentials after stimulation of the tibial nerve. Neurology (NY), 1982, 32: 1151-1158.
287
EEG 03255
Conduction characteristics of somatosensory evoked potentials to peroneal, tibial and sural nerve stimulation in man Luciana Pelosi
1, Joan B.
Cracco and Roger Q. Cracco
Department of Neurology, State University of New York, Health Science Center at Brooklyn, Brooklyn, N Y 11203 (U.S.A.)
(Accepted for publication: 15 October, 1986)
Summary Lumbar spine and scalp short latency somatosensory evoked potentials (SSEPs) to stimulation of the posterior tibial, peroneal and sural nerves at the ankle (PTN-A, PN-A, SN-A) and common peroneal nerve at the knee (CPN-K) were obtained in 8 normal subjects. Peripheral nerve conduction velocities and lumbar spine to cerebral cortex propagation velocities were determined and compared. These values were similar with stimulation of the 3 nerves at the ankle but were significantly greater with CPN-K stimulation. CPN-K and PTN-A SSEPs were recorded from the L3, T12, T6 and C7 spines and the scalp in 6 normal subjects. Conduction velocities were determined over peripheral nerve-cauda equina (stimulus-L3), caudal spinal cord (T12-T6) and rostral spinal cord (T6-C7). Propagation velocities were determined from each spinal level to the cerebral cortex. With both CPN-K and PTN-A stimulation lhe speed of conduction over peripheral nerve and spinal cord was non-linear. It was greater over peripheral nerve-cauda equina and rostral spinal cord than over caudal cord segments. The CPN-K response was conducted significantly faster than the PTN-A response over peripheral nerve-cauda equina and rostral spinal cord but these values were similar over caudal cord. Spine to cerebral cortex propagation velocities were significantly greater from all spine levels with CPN-K stimulation. These data show that the conduction characteristics of SSEPs over peripheral nerve, spinal cord and from spine to cerebral cortex are dependent on the peripheral nerve stimulated. Key words: Somatosensory evoked potentials; Conduction characteristics
Short latency s o m a t o s e n s o r y potentials (SSEPs) to s t i m u l a t i o n of nerves in the lower extremity which arise in the c a u d a e q u i n a a n d in spinal cord afferent p a t h w a y s have been described in m a n using b o t h n o n - i n v a s i v e (surface) a n d invasive (intrathecal, epidural, s p i n o u s process or ligament) recording teclmiques (Shimoji et al. 1972, 1977; R. Cracco 1973; J. Cracco et al. 1975, 1979; Ertekin 1976a; Jones a n d Small 1978; Terao et al. 1979; Lueders et al. 1981; Jones et al. 1982; Kakigi et al.
1 Present address: Dept. of Clin. Neurophysiol., II School of Medicine, Naples, and Dept. of Neurol., Fondazione Clinica del Lavoro, Campoli, Italy. Correspondence to: Joan B. Cracco, MD, Dept. of Neurol., SUNY Health Science Center at Brooklyn, 450 Clarkson Avenue, Brooklyn, NY 11203 (U.S.A.).
1982; M a c o n a n d Poletti 1982; Y a m a d a et al. 1982; D e s m e d t a n d C h e r o n 1983; Maccabee et al. 1983; Seyal et al. 1983). SSEPs to posterior tibial nerve s t i m u l a t i o n at the ankle were recorded in most of these investigations (Jones a n d Small 1978; T e r a o et al. 1979; Lueders et al. 1981; Jones et al. 1982; Kakigi et al. 1982; Y a m a d a et al. 1982; D e s m e d t a n d C h e r o n 1983; Seyal et al. 1983); however, responses to posterior tibial or c o m m o n p e r o n e a l nerve s t i m u l a t i o n at the knee were obtained in a few (R. Cracco 1973; J. Cracco et al. 1975, 1979; Ertekin 1976b; Jones et al. 1982; M a c o n a n d Poletti 1982). Recordings were performed at different spinal levels in each of these studies a n d c o n d u c t i o n distances were n o t always m e a s u r e d in the same way. Because of these variations in s t i m u l a t i n g a n d recording techniques, different values for c o n d u c t i o n characteristics of
0168-5597/87/$03.50 © 1987 Elsevier Scientific Publishers Ireland, Ltd.
288
spine SSEPs were obtained in these studies and the data are difficult to compare. Therefore, in this study, SSEPs to stimulation of 4 nerves in the lower extremity were recorded from the spine and scalp in normal subjects and conduction characteristics along peripheral nerve and spine and from spine to scalp were compared using standardized recording methods.
Methods
Eleven slender normal adult subjects (6 males, 5 females, mean age 24.9 years) were studied. Recordings were performed with subjects unsedated and relaxed or drowsy in the prone position. Six millimeter tin disks attached to the skin with collodion and filled with conductive jelly served as both stimulating and recording electrodes. Recording electrode impedance was maintained under 3000 O. Stimulating electrodes were placed 3 cm apart (cathode proximal) along the nerve. Constant voltage stimuli (0.2 msec duration) were delivered to the left common peroneal nerve at the knee in the popliteal fossa (CPN-K) and to the left posterior tibial, sural and peroneal nerves at the ankle (PTN-A, SN-A, PN-A). Nerves were stimulated one at a time. With mixed nerve (CPN-K, PTN-A and PN-A) stimulation, the stimulus intensity was adjusted to produce vigorous but non-painful foot movement. With sensory nerve (SN-A) stimulation an intensity 2-2.5 times sensory threshold was used which resulted in non-painful paresthesia on the lateral aspect of the foot. With respect to sensory threshold, the stimulus intensity was the same for all nerves studied and varied between 2 and 2.5 times threshold. In 6 subjects SSEPs to C P N - K and PTN-A stimulation were recorded over 4 spine levels and from the scalp. Spine bipolar recordings were obtained by attaching the grid I electrodes over the L3, T12, T6 and C7 spinous processes and thegrid II electrodes 4 cm rostral to these locations. Scalp SSEPs were recorded from Cz'-Fpz' (2 cm behind Cz and FPz, International 10-20 system) and Cz'NC leads (the non-cephalic reference electrode (NC) was placed on the shoulder contralateral to
L. PELOSI ET AL.
the stimulated leg). Spine and scalp SSEPs were obtained separately using stimulus rates of 4.76.7/sec for spinal and 1.7/sec for scalp recordings and analysis times of the 35-40 msec for spine and 50-60 msec for scalp SSEPs. Computer sampling time was 116 and 192 /~sec/point for spine and scalp recordings, respectively. For spine recordings 2 - 4 trials of 2000-4000 averages were superimposed and for scalp recordings 2 trials of 500-1000 averages were superimposed. Recordings were also obtained over the tibial nerve in the popliteal fossa during PTN-A stimulation in 4 of these subjects; this recording electrode was placed 4 cm above the popliteal crease and the reference electrode was placed on the medial surface of the knee. Filter bandpass for all recordings was 5-1500 Hz and relative negativity in grid I produced an upward deflection. The onset latencies of the first negative potential recorded over peripheral nerve and spine (L3, T12, T6 and C7) were measured. This indicates the approximate time when impulses in the fastest fibers pass under the recording electrode (Gilliatt et al. 1965). For scalp recordings the peak latency of the prominent positive potential which precedes the first well defined negative potential served as the latency indicator. This was P27 with CPN-K and P37 with PTN-A stimulation. The peak of these components is thought to reflect activity arising in somesthetic cortex (Tsumoto et al. 1972; Rossini et al. 1981; Vas et al. 1981; Cruse et al. 1982; Desmedt and Bourguet 1985). Distances along the course of the nerves from the stimulating cathode at the knee or ankle to the L3 recording electrode were measured and stimulus to L3 conduction velocities (CVs) were calculated. In 4 subjects in whom potentials to PTN-A stimulation were recorded over the popliteal fossa, the onset latency of the response recorded at the knee was subtracted from that recorded at L3 to determine the CV in the knee to L3 segment. Distances along the skin surface between the L3, T12, T6 and C7 spines and from these spines to the scalp recording electrode (Cz') were measured with the subject prone. CVs for CPN-K and PTNA SSEPs were determined in the following segments: (1) peripheral nerve and cauda equina (stimulating cathode to L3); (2) cauda equina to
CONDUCTION CHARACTERISTICS OF SPINE AND SCALP SSEPs cervical spinal cord (L3 to C7); (3) spinal cord entry zone to cervical spinal cord (T12 to C7); (4) caudal spinal cord (T12 to T6); (5) rostral spinal cord (T6 to C7). Spine to cortex (L3, T12, T6, C7 to C z ' ) propagation velocities (PVs) were also calculated. The term propagation velocity was used for the spine to scalp segments since the term c o n d u c t i o n velocity implies that responses originating in the same fibers are recorded at 2 different sites 2. The mean, standard deviation and range were determined for each parameter. These values for C P N - K and PTN-A SSEPs were compared using the Student t test (Table I). In 8 subjects CPN-K, PTN-A, SN-A and PN-A were stimulated at rates of 1.7-4.7/sec and simultaneous recordings were obtained from a bipolar lead over the lumbar spine (L3 to 4 cm rostral) and from .3 scalp (Cz, Cz', Pz)-NC leads. Using analysis times of 60-100 msec, 2 or 3 trials of 1500-2000 averaged responses were superimposed. Using the onset latency of the negative potential at L3 and the peak latency of the first prominent positive scalp potential, CVs were determined over peripheral nerve and cauda equina (stimulating cathode to L3) and PVs were calculated from cauda equina to cortex (L3 to Cz). Mean, standard deviation and range for these two measurements were calculated for each nerve stimulated (Table II). These values were compared using the Student t test. Amplitudes were also measured at L3 from the peak of the first negative to the peak of the following positive potential and at the scalp from the peak of P27 (CPN-K) or P37 (PTN-A, SN-A and PN-A) to the peak of N35 or N45.
Results
Comparison of spine and scalp SSEPs to CPN-K and PTN-A stimulation With C P N - K and P T N - A stimulation small triphasic potentials often with poorly defined initial positive phases were recorded over the lumbar, 2 F o r the s a m e r e a s o n a n a r g u m e n t c o u l d be m a d e that the t e r m c o n d u c t i o n velocity is also i n a p p r o p r i a t e for spine to spine s e g m e n t s - - see Discussion.
289 PTN-A Stimulation
m$~. I0 C7-4cm. rostral
CPN-K Stimulation
60
5
55
~
T6 -4cm. rostrol
I2o.~.~ Ti2-4cm- rostral
L3-4crn. rostral
~
mm~. I0
40
5
Popliteol Fosso mN¢. 2
32
Fig. 1. Spine and scalp recordings of SSEPs to stimulation of PTN-A and CPN-K in 2 subjects. The response recorded at the popliteal fossa with PTN-A stimulation is shown in the bottom trace. Note differences in analysis time. The onset latency of the negative potentials recorded over peripheral nerve and spine and the peak latency of the initial scalp positive potential are indicated.
thoracic and cervical spines. These potentials progressively increased in latency rostrally (Fig. 1). With stimulation of either nerve potentials were usually larger and broader at T12 than at more rostral or caudal sites and they were small and at times difficult to record at T6 and C7. Typically, they were better defined with C P N - K stimulation. Scalp SSEPs to both C P N - K and PTN-A stimulation consisted of an initial positive component followed by a larger negative potential (Fig. 1). These components were labeled P27-N35 for C P N - K and P37-N45 for P T N - A (AEEGS Guidelines for Clinical Evoked Potential Studies 1984). From stimulus to L3 the CV was significantly slower with PTN-A than with C P N - K stimulation (Table I and Fig. 2). In the 4 subjects in whom PTN-A CVs were also determined over the knee to L3 segment, this value of 64.4 m / s e c was 5.2 m / s e c slower than the comparable C P N - K value. CVs of PTN-A SSEPs over the L3-C7, T12-C7
290
L. PELOSI ET AL. IOO
CPN-K PTN-A
L 3- 4cm. rostrol
flail! £ii:,
-:..:;.:
\
40
"k.:Ei <:/::
20
!:iiii!;! ~.iiiiii:i:
0
~ k
~
{:(:%
80
6o
C~ -Shoulder
SUBJECA T CPN-K SSEPs
SUBJECAT
.:...~
PTN-A SSEPs
<<:i~!
Stimulus- L3
T6-C7
T,2-T6
SUBJECAT
Fig. 2. Comparisonof mean valuesof CVs of SSEPsto CPN-K and PTN-A stimulation.
a n d T6-C7 segments were also significantly slower t h a n those of C P N - K SSEPs (Table I). W i t h b o t h C P N - K a n d F F N - A s t i m u l a t i o n the speed of c o n d u c t i o n over the spine was n o n - l i n e a r (Fig. 2). It was greater over peripheral n e r v e - c a u d a e q u i n a than over caudal cord a n d significantly increased over rostral spinal cord (for b o t h C P N - K a n d P T N - A T 1 2 - T 6 / T 6 - C 7 c o m p a r i s o n P was 0.001). Over rostral spinal cord, C P N - K SSEPs were c o n d u c t e d 1.5 times faster than P T N - A SSEPs. In contrast, there was n o significant difference in the speed of c o n d u c t i o n over the T12-T6 segment (Table I a n d Fig. 2). Based o n observations o n h u m a n cadavers the actual spinal cord length from lower thoracic to
. ~A /
~
PN-A SSEPs
SUBJECB T SN-A SSEPs
I
/
i
5
55
5
I
I05 msec
Fig. 3. Spine and scalp SSEPs to CPN-K, PTN-A, PN-A and SN-A stimulation in 2 subjects (subjects A and B). Note differences in calibration. The onset latency of the spine potential and the peak latency of the scalp positive potential are indicated.
TABLE, I Mean, standard deviation and range of peripheral and spinal conduction velocity and spine to scalp propagation velocity of SSEPs to CPN-K and PTN-A stimulation in 6 subjects. Stimulus-L3 T12-T6
T6-C7
CPN-K Mean S.D. Range
L3-C7
T12-C7
L3-Cz'
69.65 _+4.7 63.8-73.5
48.91 +4.9 41.9-56.4
100.5 70.52 70.16 39.65 +9.1 _+5.2 _+3.67 +1.67 91.4-116.2 64.4-78.3 66.4-76.47 36.9-41.7
PTN-A Mean S.D. Range
60.06 +3.5 54.2-65.2
50.52 _+4.7 45-58.05
P=
< 0.001
0.57
T12-Cz'
T6-Cz'
C7-Cz'
37.77 34.61 _+1.58 _+1.3 35.13-39.03 32.8-36.3
24.27 +1.9 21.23-26.14
64.88 58.91 _+6.7 +4.86 56.9-74.1 51.9-64.9
55.17 34.39 32.73 29.28 +5.2 _+2.3 _+1.98 _+2.03 46.57-60.7 30.4-37.03 29.24-34.4 26.4-31.9
21.87 _+2.63 18.58-24.48
< 0.001
< 0.001
0.002
< 0.001
< 0.001
< 0.001
< 0.001
CONDUCTION CHARACTERISTICS OF SPINE AND SCALP SSEPs
cervical areas is almost 13% shorter than that measured over the skin surface (Desmedt and Cheron 1983). Correcting for this factor the mean values for PTN-A CVs were reduced from 55.17 to 47.99 m / s e c for T12 to C7 segment, from 50.52 to 43.95 m / s e c for T12 to T6 and from 64.88 to 56.44 m / s e c for T6 to C7. Mean values for CPN-K CVs were reduced from 70.16 to 61.03 m / s e c for T12 to C7 segment, from 48.91 to 42.55 m / s e c for T12 to T6 and from 100.5 to 87.43 m / s e c for T6 to C7. Spine to scalp PVs were slower with PTN-A than with C P N - K stimulation (Table I). They were about 5 m / s e c slower from L3, T12 and T6 to scalp and about 2 m / s e c slower from C7 to scalp.
Comparison of L3 spine and scalp SSEPs to CPN-K, PTN-A, PN-A and SN-A stimulation SSEPS were recorded over L3 in all 8 subjects with CPN-K and PTN-A stimulation and in 7 subjects with PN-A and SN-A stimulation. These responses consisted of initially positive triphasic potentials. They were consistently lower in amplitude with SN-A and PN-A stimulation than with C P N - K and PTN-A stimulation (Fig. 3). With stimulation of each of the 4 nerves, scalp SSEPs consisted of an initial positive potential followed by a larger negative potential. Scalp SSEPs were consistently greater in amplitude with stimulation of PTN-A, SN-A and PN-A than with CPN-K stimulation (Fig. 3). The mean peak latency of the initial positive component was greater with SN-A (42.8 msec), PN-A (40.6 msec) and PTN-A (39.3
291
msec) stimulation than with CPN-K stimulation (27.7 msec). This value for PTN-A (39.3 msec) was similar to that reported by Jones and Small (1978) (39.7 msec) and by Riffel et al. (1984) (38.8 msec) but was greater than that noted by Chiappa (1983) (36.3 msec). Our value for CPN-A (27.6 msec) was similar to that found by Chiappa (27.3 msec). Our peak latencies for PN-A (40.6 msec) and SN-A (42.8 msec) were similar to those reported by Eisen and Elleker (1980) (39,9 and 42.1 msec, respectively) but the values for SN-A were greater than those published by Chiappa (38.7 msec). Table II shows the mean, standard deviation and range of the peripheral CVs and L3-Cz PVs for the 4 nerves stimulated. The peripheral CV was greatest with CPN-K stimulation (CPNK/PTN-A P=0.03; CPN-K/PN-A P=0.01; C P N - K / S N - A P = 0 . 0 0 1 ) . However, peripheral CVs were similar for all 3 nerves stimulated at the ankle and differences were not statistically significant. The PV from L3 to Cz was also greatest with CPN-K (CPN-K/PTN-A P = 0.004; CPNK / P N - A P = 0.003; C P N - K / S N - A P = < 0.001). In contrast, there were only slight differences in these mean values which were not statistically significant among the 3 nerves stimulated at the ankle. These PV values with CPN-K stimulation are somewhat greater than those previously reported from this laboratory (Schiff et al. 1984). This can be attributed to differences in methods used for measuring distances from spine to scalp. In the earlier study spine to scalp straight-line distances were used rather than measurements along the spine and scalp. In comparing these
TABLE II Peripheral nerve conduction velocity and spine to scalp propagation velocity of SSEPs to CPN-K, PTN-A, PN-A and SN-A stimulation in 7 subjects. CPN-K
PTN-A
PN-A
SN-A
Stimulus-L3 Mean S.D. Range
69.26 + 6.60 58.70-78.10
62.17 + 5.04 54.50-66.60
60.90 _+3.69 55.08-64.50
57.48 + 3.66 52.57-60.27
L3-Cz Mean S.D. Range
38.34 + 3.17 35.80-43.40
32.45 + 3.35 26.60-37.60
31.29 + 3.93 27.30-37.60
29.25 + 3.06 24.20-33.14
292 methods we found distances using a straight-line measurement to be about 10% less.
Discussion
Evoked potentials to stimulation of peripheral nerves have been recorded from individual spinal cord afferent tracts such as the dorsal and dorsolateral columns using small intradural electrodes in animals (Grundfest and Campbell 1942; Lloyd and McIntyre 1950). In contrast, large extradural or surface (skin) electrodes record spinal potentials arising in multiple afferent pathways (R. Cracco 1973; Happel et al. 1975; Sarnowski et al. 1975; R. Cracco and Evans 1978; Ducati and Schieppati 1980; Feldman et al. 1980; Jones et al. 1982). In human and animal extradural recordings the response recorded over mid and rostral spinal cord segments consists of several negative peaks (Happel et al. 1975; Sarnowski et al. 1975; R. Cracco and Evans 1978; Jones et al. 1982; Pelosi et al. 1985). The peak and interpeak latencies of these components progressively increase at more rostral recording locations (R. Cracco and Evans 1978). This is consistent with multiple afferent volleys which arise in fiber groups having different CVs. Studies in the cat, which compared CV over peripheral nerve and spinal cord to stimulation of lower extremity muscle and cutaneous afferent fibers, have shown that the cutaneous nerve evoked response has a CV of about 80 m / s e c over peripheral nerve-cauda equina. This volley slows to about 60-70 m / s e c over lower lumbar cord segments because of fiber branching and is then conducted rostrally at a similar velocity. In contrast, the CV of the response to stimulation of muscle nerves is greater than 100 m / s e c over peripheral nervecauda equina and over the dorsolateral fasciculus of the thoracic cord. However, the CV is considerably slower than that of the cutaneous nerve evoked response over lumbar cord segments because of synaptic delay and a greater degree of fiber branching (Lloyd and McIntyre 1950). The result of these complex relationships is that the volley to muscle nerve stimulation arrives at caudal cord segments before the cutaneous volley, is over-
L. PELOS1ET AL. taken by the cutaneous volley over caudal spinal cord and then again overtakes the cutaneous volley over rostral cord segments (Grundfest and Campbell 1942; Lloyd and McIntyre 1950). Hence, at least in the cat, the onset latency of the spinal cord compound action potential is probably not dependent on the same fibers at all spinal levels. Our data demonstrated that the peripheral and spinal CVs of the response to both CPN-K and PTN-A stimulation were non-linear. CVs were greater over peripheral nerve-cauda equina and rostral spinal cord than over caudal cord. CVs to PTN-A stimulation were greater over rostral spinal cord than over peripheral nerve-cauda equina when standard surface measurements were used. However, when correction was made for the overestimation of actual spinal cord length which results when using surface measurements (Desmedt and Cheron 1983), the adjusted CV was greater over peripheral nerve-cauda equina. With CPN-K stimulation the CV over rostral spinal cord was greater than it was over peripheral nerve-cauda equina using either standard or adjusted measurements. The slowing in speed of conduction over caudal cord can be explained on the basis of either fiber branching or synaptic delay or a combination of the two (Lloyd and McIntyre 1950). The increase in speed of conduction over rostral cord segments suggests that large diameter second order afferent fibers contribute to this response since the speed of conduction in first order fibers which ascend the length of the spinal cord would not be expected to increase in rostral cord segments (Lloyd and McIntyre 1950). The CV over caudal cord segments was similar with CPN-K and PTN-A stimulation. However, the CV over peripheral nerve-cauda equina and rostral cord and hence the degree of non-linearity in CV up the spine was much greater with CPN-K than with PTN-A stimulation. This indicates that the spinal response to CPN-K stimulation is mediated by more rapidly conducting pathways. This is not unexpected since rapidly conducting peripheral afferent fibers which supply muscle spindles and tendon organs would be expected to be more numerous at the knee than at the ankle. Additionally, it has been shown in primates that the ratio of afferent fibers arising from deep vs. cutaneous
CONDUCTION CHARACTERISTICS OF SPINE AND SCALP SSEPs
receptors is greater in proximal hind limb than in distal hind limb (Werner and Whitsel 1967; Whitsel et al. 1969). These afferents from deep receptors include the large group I fibers which synapse in caudal cord nuclei such as Clark's column which give rise to large rapidly conducting second order fibers. Second order cutaneous fibers such as those which ascend in Morin's tract (Morin 1955) could also contribute to the increase in CV over rostral spinal cord segments with both CPN-K and PTN-A stimulation. Peripheral CVs and spine to scalp (L3-Cz) PVs of SSEPs to PTN-A, PN-A and SN-A stimulation were all similar but significantly slower than those obtained with CPN-K stimulation. Lumbar, thoracic and cervical spine to scalp PVs with PTN-A stimulation were less than corresponding values obtained with CPN-K stimulation. These differences were least in the cervical spine to scalp segment. These data are consistent with the interpretation that the CPN-K response is mediated by larger faster conducting peripheral and spinal afferent fibers than those which mediate the response evoked by stimulation of nerves at the ankle. It also suggests that the central somatosensory pathways rostral to the cervical cord which mediate the response to stimulation of nerves at both ankle and knee are similar in diameter and conduction characteristics. Although this conclusion seems reasonable, it must be interpreted with caution since the fastest conducting fibers which contribute to the response and determine response latency may not be identical at different spinal levels and the fibers which determine the latency of the spinal potentials may not be the same as those which mediate the scalp response. The authors thank Mrs. Gracy George for her secretarial assistance.
References American Electroencephalographic Society (AEEGS). Guidelines for clinical evoked potential studies. J. clin. Neurophysiol., 1984, 1: 3-53. Chiappa, K.H. Evoked Potentials in Clinical Medicine. Raven Press, New York, 1983: 233-234. Cracco, J.B., Cracco, R.Q. and Graziani, L.J. The spinal evoked response in infants and children. Neurology (Minneap.), 1975, 25: 131-36.
293
Cracco, J.B., Cracco, R.Q. and Stolove, R. Spinal evoked potential in man: a maturational study. Electroenceph. clin. Neurophysiol., 1979, 46: 58-64. Cracco, R.Q. Spinal evoked response: peripheral nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1973, 35: 379-386. Cracco, R.Q. and Evans, B. Spinal evoked potential in the cat. Effects of asphyxia, strychnine, cord section and compression. Electroenceph. clin. Neurophysiol., 1978, 44: 187-201. Cruse, R., Klem, G., Lesser, R.P. and Lueders, H. Paradoxical lateralization of cortical potentials evoked by stimulation of posterior tibial nerve. Arch. Neurol. (Chic.), 1982, 39: 222-225. Desmedt, J.E. and Bourguet, M. Color imaging of parietal and frontal somatosensory potential fields evoked by stimulation of median or posterior tibial nerve in man. Electroenceph. clin. Neurophysiol., 1985, 62: 1-17. Desmedt, J.E. and Cheron, G. Spinal and far-field components of human somatosensory evoked potentials to posterior tibial nerve stimulation analysed with oesophageal derivations and non-cephalic reference recording. Electroenceph. clin. Neurophysiol., 1983, 56: 635-651. Ducati, A and Schieppati, M. Spinal pathways mediating somatosensory evoked potentials from cutaneous and muscle nerves in the cat. Acta neurochir. (Wien), 1980, 52: 99-104. Eisen, A. and Elleker, G. Sensory nerve stimulation and evoked cerebral potentials. Neurology (NY), 1980, 30: 1097-1105. Ertekin, C. Studies in the human evoked electrospinogram. I. The origin of the segmental evoked potentials. Acta neurol. scand., 1976a, 53: 2-20. Ertekin, C. Studies in the human evoked electrospinogram. II. The conduction velocity along the dorsal funiculus. Acta neurol, scand., 1976b, 53: 21-38. Feldman, M.H., Cracco, R.Q., Farmer, P. and Mount, F. Spinal evoked potential in the monkey. Ann. Neurol., 1980, 7: 238-244. Gilliatt, R.Q., Melville, I.P., Velate, A.S. and Willison, R.G. A study of normal nerve action potential using an averaging technique (barrier grid storage tube). J. Neurol. Neurosurg. Psychiat., 1965, 28: 191-200. Grundfest, H. and Campbell, B. Origin, conduction and termination of impulses in dorsal spinocerebellar tracts of cats. J. Neurophysiol., 1942, 5: 275-294. Happel, L.T., Le Blanc, H.J. and Kline, D.G. Spinal cord potentials evoked by peripheral nerve stimulation. Electroenceph, clin. Neurophysiol., 1975, 38: 349-354. Jones, S.J. and Small, D.G. Spinal and subcortical evoked potentials following stimulation of the posterior tibial nerve in man. Electroenceph. clin. Neurophysiol., 1978, 44: 299-306. Jones, S.J., Edgar, M.A. and Ransford, A.O. Sensory nerve conduction in the human spinal cord: epidural recordings made during scoliosis surgery. J. Neurol. Neurosurg. Psychiat., 1982, 45: 446-451. Kakigi, R. Shibasaki, H., Hashizume, A. and Kuroiwa, Y. Short-latency somatosensory evoked spinal and scalp-recorded potentials following posterior tibial nerve stimula-
294 tion in man. Electroenceph. clin. Neurophysiol., 1982, 53: 602-611. Lloyd, D.P.C. and Mclntyre, A.K. Dorsal column conduction of group I muscle afferent impulses and their relay through Clarke's column. J. Neurophysiol., 1950, 13: 39-54. Lueders, H., Andrish, J., Gurd, A., Weiker, G. and Klem, G. Origin of far-field subcortical potentials evoked by stimulation of the posterior tibial nerve. Electroenceph. clin. Neurophysiol., 1981, 52: 336-344. Maccabee, P.J., Levine, D.B., Pinkhasov, E.I., Cracco, R.Q. and Tsairis, P. Evoked potentials recorded from scalp and spinous processes during spinal column surgery. Electroenceph. clin. Neurophysiol., 1983, 56: 569-582. Macon, J.B. and Poletti, C.E. Conducted somatosensory potentials during spinal surgery. I. Control conduction velocity measurements. J. Neurosurg., 1982, 57: 349-353. Morin, F. A new spinal pathway for cutaneous impulses. Amer. J. Physiol., 1955, 183: 245-252. Pelosi, L., Caruso, G., Russo, G., Milano, C., Paolino, G., Lotti, G. and Greco, P. Spinal evoked potentials during surgery for scoliosis. Electroenceph. clin. Neurophysiol., 1985, 61: $28. Riffel, B., Stohr, M. and Korner, S. Spinal and cortical evoked potentials following stimulation of the posterior tibial nerve in the diagnosis and localization of spinal cord diseases. Electroenceph. clin. Neurophysiol., 1984, 58: 400-407. Rossini, P.M., Cracco, R.Q., Cracco, J.B. and House, W.J. Short-latency somatosensory evoked potentials to peroneal nerve stimulation: scalp topography and the effect of different frequency filters. Electroenceph. clin. Neurophysiol., 1981, 52: 540-552. Sarnowski, R.J., Cracco, R.Q., Vogel, H.B. and Mount, F. Spinal evoked response in the cat. J. Neurosurg., 1975, 43: 329-336. Schiff, J.A., Cracco, R.Q., Rossini, P.M. and Cracco, J.B. Spine
L. PELOSI ET AL. and scalp somatosensory evoked potentials in normal subjects and patients with spinal cord disease: evaluation of afferent transmission. Electroenceph. clin. Neurophysiol., 1984, 59: 374-387. Seyal, M., Emerson, R.G. and Pedley, T.A. Spinal and early scalp-recorded components of the somatosensory evoked potential following stimulation of the posterior tibial nerve. Electroenceph. clin. Neurophysiol., 1983, 55: 320-330. Shimoji, K., Kano, T., Higashi, H., Morioka, T. and Henschel, E.O. Evoked spinal electrograms recorded from epidural space in man. J. appl. Physiol., 1972, 33: 468-471. Shimoji, K., Matsuki, M. and Shimizu, H. Wave form characteristics in man. J. Neurosurg., 1977, 46: 304-310. Terao, A., Matsuda, E. and Fukunaga, H. Conduction velocity measurement in human spinal cord and tibial nerve using skin electrodes recording of the spinal evoked potential. Kawasaki med. J., 1979, 5: 71-78. Tsumoto, T., Hirose, N., Nonaka, S. and Takahashi, M. Analysis of somatosensory evoked potentials to lateral popliteal nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1972, 33: 379-388. Vas, G., Cracco, J.B. and Cracco, R.Q. Scalp recorded short latency cortical and subcortical somatosensory evoked potentials to peroneal nerve stimulation. Electroenceph. clin. Neurophysiol., 1981, 52: 1-8. Werner, G. and Whitsel, B.L. The topology of dermatomal projection in the medial lemniscal system. J. Physiol. (Lond.), 1967, 192: 123-144. Whitsel, B.L., Petrucelli, L.M. and Sapiro, G. Modality representation in the lumbar and cervical fasciculus gracilis of squirrel monkeys. Brain Res., 1969, 15: 67-78. Yamada, T., Machida, M. and Kimura, J. Far-field somatosensory evoked potentials after stimulation of the tibial nerve. Neurology (NY), 1982, 32: 1151-1158.