24
Journal of the Neurological Sciences, 108 (1992) 24-31 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022 510X/92/$05.00
JNS 03708
Recovery functions of somatosensory evoked potentials in parkinsonian patients Kenji N a k a s h i m a , T a t s u o N i t t a a n d K a z u r o T a k a h a s h i Dicision of Neurology, blstitute of Neurological Sciences, Tottori University Faculty of Medicine, 86 Nishimachi, Yonago (Japan) (Received 30 October, 1990) (Revised, received 17 September, 1991) (Accepted 25 September. 1991)
Key words: Somatosensory evoked potential; Recovery function; Paired stimulation; Sensory disturbance; Parkinson's disease
Summary The aim of this study was to analyze the recovery functions of median nerve somatosensory cvoked potentials (SEPs) and to clarify changes in the somatosensory system of patients with Parkinson's disease (PD). The central and frontal SEPs and the nerve potential of the median nerve were examined. The latencies and amplitudes of SEPs produced by a single shock in patients with PD were normal. The recovery functions of central SEPs showed a low degree of suppression in patients with PD, although the recovery curve of the nerve potential in PD patients was almost normal. The change in recovery curves of SEPs in patients with sensory complaints were more noticeable than that in patients without sensory disturbance. A low degree of suppression of central SEPs might play a role in the sensory complaints of PD patients.
Introduction Sensory disturbances are sometimes observed in patients with Parkinson's disease (PD). Normal somatosensory evoked potentials (SEPs) have been reported in PD patients (Chiappa 1983; Keller 1984; Grosswasser et al. 1985; J6rg et al. 1987). Recently, the recovery functions of SEPs have been reported (Allison 1962; Shagass et al. 1964; Namerow 1970; Lueders et al. 1983; Greenwood et al. 1987; J6rg et al. 1987; Reisin et al. 1988; Kanda et al. 1989). The purpose of this study is to clarify changes in the somatosensory system of patients with PD through the analysis of the recovery functions of SEPs. We evaluated the recovery functions of the nerve and Erb's potentials and of the frontal and central components of SEPs.
Subjects and methods Thirteen normal subjects (aged 27-75 years; mean 53.8) and 18 patients with PD (aged 41-68 years; mean
Correspondence to: K. Nakashima, Division of Neurology, Institute of Neurological Sciences, Tottori University Faculty of Medicine, 86 Nishimachi, Yonago, 683, Japan. Tel: 0859-34-8032; Fax: 0859-348083.
57.0; Hoehn and Yahr (1967) grade II, 3 patients; grade llI, 15 patients) were examined. All patients were receiving medication; levodopa-carbidopa, trihexyphenydil, amantadine, etc. All patients had almost normal intellect, evaluated through the usual clinical interviews. They were fully evaluated through the conventional neurological examination. All patients with PD were examined using head computed tomography and cervical spine radiographs. The patients who showed obvious abnormality in these examinations were excluded from the present study. Clinical details of the 18 patients with PD examined in the present study are shown in Table 1. Six of the 18 patients suffered from dysesthesia or pain sensation. The peripheral nerve conduction study of the median nerve in patients with sensory complaints was almost normal and not different from those in patients without sensory complaints (Table 1). Before the experiments, the consent of subjects was obtained after informing them fully about the aims and procedures of this study. In all the subjects, the recovery functions of central SEPs were evaluated. The recovery functions of frontal SEPs were evaluated in 8 of the 13 control subjects and 10 of the 18 patients with PD. The recovery function of the nerve potential of the peripheral nerve was also examined in 6 normal subjects and 9 PD patients. The subjects were seated in chairs and asked to close their eyes. The stimuli were 0.1-ms square wave
25 TABLE 1 SUBJECTS Duration: duration o f the illness (years). Scale: H o e h n and Yahr scale; nd: no*. done. Age
Sex
Duration
Scale
59 54 64 41 62
F M F F F
7 2 4 5 I0
111 111 111 111 111
63 60 62 62 56 58 59 62 56 63 07 24 62
F F M F F M M F F M F F M
3 2 1 3 l0 8 10 6 8 6 12 5 2
Ill 11 II !! Ill iii Ill Ill ill III Ill Ill !11
Sensory disturbance
dysesthesia in the legs dysesthesia in the right arm and leg dysesthesia in the extremities pain in the extremities dysesthesia in the u p p e r and lower extremities dysesthesia in the left arm -
electrical pulses delivered to the unilateral median nerve at the wrist. The intensity was 10% above the motor threshold for thumb twitch. Single and double shocks were given alternately at 2 Hz. For double stimulation, the same intensity was used for conditioning (preceding) and test (following) stimulation through the same stimulating electrodes. The intervals between conditioning and test shocks were 10, 20, 30, 40, 50, 60, 70 and 80 ms. A recording electrode for central SEPs was placed over the contralateral scalp, at a point 7 cm lateral and 2 cm posterior to the vertex, and for frontal SEPs at Fz (according to a 10-20 system) with the linked-ear reference electrodes. For the Erb potential, an active electrode was placed over the ipsilateral Erb's point with a reference electrode at the contralateral Erb's point. For the nerve potential of the median nerve, an active electrode was placed over the median nerve at the elbow, just medial to the brachial artery and with a reference electrode at a point 2 cm medial to the active electrode. Filter settings of 5-300 Hz for SEPs and 50-3000 Hz for Erb's and nerve potentials were used. Signals were fed into a NEC San-ei Signal Processor 7T18A. A summation was taken of 128-256 responses with an analysis time of 200 ms and repeated. A pre-stimulus trigger was used for 20 ms prior to the conditioning or single stimulus. The SEP components were labeled according to their positive (P) or negative (N) polarity and their peak latencies. We evaluated the latencies and the amplitudes (peak to peak) of frontal and central SEPs produced by the single shocks. Then the recovery curves
Median nerve conduction motor ( m / s )
sensory ( m / s )
50.6 50.1 55.8 57.0
53.4 55.6 64.9 57.2
51.0 52.6 56.8 nd nd nd nd nd 55.5 nd 57.4 54.9 61.4 nd
64.3 60.0 58.5 nd nd nd nd nd 60.6 nd 69.1 57.8 65.2 nd
of SEP amplitudes were calculated. In order to determine the real responses produced by the test shock (R2), the values of responses produced by the single shock (R1) were subtracted from the values produced by the combined conditioning and test shock (R1 + R2) at each stimulus interval (Fig. 1). Recovery functions were expressed as a percentage of the values produced by the single shocks. Statistical comparisons were made using a medical statistical package, "STAX", designed by the medical computer research association, University of Tokyo. In the study of recovery functions in normal subjects, the Wilcoxon signed rank test was used at each time interval between conditioning and test shocks. The Wilcoxon rank-sum test was used for statistical analysis of the data from the recovery functions of SEPs in normal subjects and patients with PD. Two-way analysis of variance was used for statistical analysis of the differences in recovery curves of SEPs between normal subjects and patients with PD. Since the test in this statistical package was designed for less than eight levels within each group, recovery functions with less than eight interval levels were evaluated in a comparison between normal subjects and patients with PD.
Results
1. Latencies and amplitudes of SEPs The latencies and amplitudes of central and frontal SEPs elicited by single stimulation showed no differ-
26 A,
72-year-old female: normal subject
% 150 4) "O Q.
E t~
/
_
•
¢:
oo
i B,
100
__
| __ ...~...--.
'N,T
, __
-~...
_
_
4) 01 4) 4)
Q.
62-year-old female: PD patient
5O o
. . . .
1 2 50 60 70 80 Conditioning-test interval (ms)
Fig. 2. Recovery curve of central PI3-NI8 amplitude. ** Wilcoxon rank-sum test, P < 0.01; * P < 0.05. Filled circles: normal subjects; empty circles: PD. Mean ( 4-1 S.E.M.) values are plotted.
Fig. I. Representative cases. RI: response produced by single shock. Ri + 2: response produced by the double shocks. R2: response obtained by subtraction of the RI response from the RI +2 response. Interstimulus interval between conditioning (~) and test ( | ) shocks: 60 ms.
ences between patients with PD and normal subjects (Table 2).
2. Recovery functions of central SEP amplitudes In normal subjects, P13-N18 showed a statistically
significant suppression at 10-20 ms intervals (Wilcoxon signed rank test: P < 0.05) and recovery at the 30-ms interval. Suppression was seen again between 40 and 80 ms (Wilcoxon signed rank test at 40: P < 0.01; at 60, 70 and 80 ms: P < 0.05). A low degree of suppression or a slight facilitation of amplitude in patients with PD was seen between 10 and 70 ms (Fig. 2; Wilcoxon rank-sum test at 70 ms: P 0 . 0 5 ; N16-P20, Fi,ii2= 0.139, P > 0.05; P20-N28, Fm.~l2 = 0.555, P > 0.05; N28P44 FL~12=0.675, P > 0 . 0 5 ) , except in two instances when slight facilitation of P14-N16 occurred at the time intervals of 20 and 80 ms (Wilcoxon rank-sum test P < 0.01).
4. Recocery function of the Erb's potential; N9 The recovery function of the Erb's potential in normal subjects did not show an obvious' change (Wilcoxon signed rank test at 10, 20, 30, 40, 50, 60, 70 and 80 ms:
®
25C
Q.
E ell 0 0
a.
I
200]
150
~\\
II!
100
50
G
,0
35 4'0
6.0 ,.0 s'0
Conditioning-test interval (ms)
Fig. 7. Recoveryfunctions of P24-N31 in patients with and without sensory complaints. * Wilcoxon rank-sum test, P < 0.05. Filled circles: PD without sensory complaints; empty circles: PD with sensory complaints. Mean ( + i S.E.M.)values are plotted.
29 % 200
.m
E
150
c 0 o
"6 0 ~B ¢D
10(~
5¢
d
10 20 3'0 4'0 5() 6'0 7'0 8'0 Coationing-test lateral (ms)
Fig. 8. Recovery functions of N31-P44 in patients with and without sensory complaints. Filled circles: PD without sensory complaints; empty circles: PD with sensory complaints. Mean ( _ I S.E.M.) values are plotted.
ms: P < 0.05, at 20 and 60 ms: P > 0.05). The recovery function of the nerve potential in patients with PD was not different from that in normal subjects (Fig. 9; analysis of variance between 10 and 70 ms: F,.9, = 0.886, P > 0.05; Wilcoxon rank-sum test at 10, 20, 30, 40, 50, 60, 70 and 80 ms: P > 0.05).
Discussion Some studies of SEPs in patients with PD have shown no abnormalities (Koller 1984; Grosswasser et
O---O
control (n:6) PD (n:9)
qO
: 120 100
O o 80
o.
' 10
' 20
310
' 40
' 50
60
;0
J 80
Conditioning-test interval (ms) Fig. 9. Recovery function of the nerve potential of the median nerve. No difference between normal subjects and PD patients is observed. Mean ( + 1 S.E.M.) values are plotted.
al. 1985; J6rg et al. 1987). Recently, Rossini et al (1989) showed an abnormally depressed frontal component, "N30" (our N28), in patients with PD, although almost normal responses of central components were observed. The frontal "N30" component may originate from the supplementary motor area (SMA). As the SMA may be strongly connected to the basal ganglia, this may explain why the abnormal N30 was observed in patients with PD. In the present study, the amplitude of P20-N28 in patients with PD was slightly lower than that in normal subjects, although this difference was not statistically significant. One of the reasons for the difference may be the band-pass filter. We used a rather narrow band-pass filter (5-300 Hz), while Rossini et al. (1989) used a wide band-pass filter (13000 Hz). In the recordings of SEPs, the ideal band-pass filter has been discussed (Desmedt et al. 1974; Lueders et al. 1981; Tsuji et ai. 1984). Tsuji et al. (1984) recommended the restricted band-pass filter of 30-250 Hz for definition of the early cortical potentials to posterior tibial nerve stimulation. In our preliminary study, the effect of a band-pass filter on the recovery functions of SEPs was examined. If a wide band-pass filter was used, the recovery curves fluctuated more than those obtained through the band-pass filter of 5-300 Hz. Stable records of SEPs should be used for subtraction through the adequate restricted band-pass filter. Our results were almost compatible with the previously reported U-shaped recovery curves (Lueders et al. 1983). A U-shaped recovery curve of SEPs may he caused by the initial facilitation in the first 10-20 ms produced by the short excitatory postsynaptic potentials and followed by the prolonged inhibitory postsynaptic potentials in the thalamic nuclei; VPL (Eccles 1966). The recovery curves of central SEPs indicated suppression, temporary recovery (or facilitation), suppression and recovery. The recovery curves of frontal SEPs showed no suppression. The different patterns of recovery curves might be caused by the different sources of SEP components (Greenwood 1987). Possible mechanisms for the suppressive effects shown by recovery curves of SEPs may include local recurrent, reticulofugal and corticofugal inhibitory effects (Greenwood 1987). Animal studies suggested that the responses of SEPs to repetitive stimulation were different from those of the peripheral nerve (Iragui-Madoz et al. 1977; Wiederholt et al. 1977; Meyer-Hardting et al. 1983). Using depth recording techniques in the cat, IraguiMadoz et al. (1977) studied the responses to repetitive stimulation. The amplitude of potentials in the cervical dorsal column and nucleus cuneatus did not change with frequencies of stimulation, although that in the ventral posterior thalamus and thalamic' radiation declined. As the intercorrelations between the peripheral nerve and cerebral response recovery functions in man
30 showed no coefficient, the influence of peripheral factors on the cerebral recovery function might be negligible (Shagass et al. 1964). Reisin et al. (1988) also found that the peripheral response differed in the manner of recovery from the N20 scalp response. In the present study, the recovery curves of the nerve and Erb's potentials in patients with PD were the same as those in normal subjects, though those of the central SEPs showed obvious differences between PD patients and normal subjects. Furthermore, the Erb's potential showed neither obvious facilitation nor suppression in control subjects. We supposed, therefore, that the central nervous system played an important role in the recovery functions of central SEPs. Since the recovery functions of frontal SEPs did not show obvious suppression, the subcortical components (P13 and N16) and the SMA component (N28) were not suppressed by preceding stimulation. In patients with Huntington's disease (HD), low amplitudes of SEPs were observed (Takahashi et al. 1972; Noth et al. 1984; Ehle et al. 1984; Kanda et al. 1989). The reduction of SEP amplitudes might reflect a reduced cortical activation by the thalamic efferent in HD (Ehle et al. 1984; Kanda et al. 1989). Kanda et al. (1989) studied the recovery functions of SEPs in patients with HD. The facilitation of amplitude in the recovery functions of N20-P24 (our NI8-P24) at the 10-ms interval was seen in patients with HD. They speculated that the affected striatum in HD influenced the thalamus, resulting in the reduction of SEP amplitudes, and that the conditioning stimuli removed some functional block in the thalamus. The disturbance of the basal ganglia might contribute to the low degree of suppression appearing in the recovery curves of SEP amplitudes. In 30-40% of PD patients, sensory disturbances have been observed (Snider et al. 1976; Koller 1984; Snider et al. 1987) They complained of burning sensations, pain or paresthesia-like sensations. The pathophysiological details of sensory disturbances in patients with PD are not yet fully understood. Sensory disturbances in PD patients may be caused by the disturbance of the peripheral or central nervous system (Snider et al. 1976). In the present study, the recovery curves of the peripheral nerve and Erb's potentials in patients with PD were the same as those in normal subjects. Clinical neurological examination, peripheral nerve conduction tests and cervical spine radiographs showed no abnormality leading to sensory disturbance in PD patients. Furthermore, the head computed tomography did not show any obvious abnormality. Only central SEPs showed differences in recovery curves between normal subjects and PD patients. Therefore, the changes in the central nervous system, rather than the peripheral factor, may contribute to the sensory disturbance in PD patients. The basal ganglia appears
to modulate sensory input (Koller 1984; Snider et al. 1976; Schneider et al. 1987). Snider et al. (1976) suggested that the locus ceruleus and/or the substantia nigra played roles in sensory release in patients with PD. The locus ceruleus, the substantia nigra and the striatum appear to have inhibitory effects on the sensory system (Krauthamer et al. 1967). The patients with sensory disturbance showed more obvious lack of suppression and/or bigger facilitation than patients without sensory complaints. The abnorma!ity of the recovery function of SEPs related to the sensory disturbance in PD. A reduced inhibitory function may contribute to sensory disturbance in PD patients. The basal ganglia receives input from the sensory afferents (Schneider et al. 1981; Crutcher et al. 1984). The sensory system is inhibited by the basal ganglia (Demetrescu et al. 1962; Krauthamer et al. 1967). Somatosensory stimulation may provoke not only SEP responses, but also the basal ganglial inhibition of the sensory system. This basal ganglial inhibition may contribute to central SEP amplitude recovery curves. The recovery functions of central SEP amplitudes in patients with PD showed a lower degree of suppression. Temporarily, we suspected that the basal ganglial inhibition of the sensory system was low in patients with PD. However, recovery curves of frontal SEPs in patients with PD showed little difference from those in normal subjects. The basal ganglial inhibitory effects on the frontal components of SEPs are slight. Another explanation for the lack of suppression a n d / o r the facilitation of the recovery curves of SEPs may be a change in the reticular formation. The reticular formation receives input from somatosensory stimulation, activates the cerebral cortex and might play a role, at least in part, in the recovery curves of SEPs. The reticular formation also receives input signals from the basal ganglia. In PD, it might be disturbed due to basal ganglial dysfunction. The abnormality of the recovery curves of SEPs might be produced by the changes in reticular formation function in PD. In conclusion, we have described the low degree of suppression in the recovery curves of central SEP amplitudes. The low degree of suppression in the sensory system may be related to the abnormal function of the basal ganglia. This study demonstrates an underlying physiological change of the sensory systems in patients with PD.
Acknowledgements This work was supported in part by funds from the Research Committee of CNS Degenerative Disease, the Ministry of Health and Welfare of Japan. This work was presented at the first International Congress of Movement Disorders in Washington, D.C., on April 25th, 1990. We are indebted to Dr. E. Rin for his arrangements in measuring the data, to Miss R. Maeda for her technical assistance and to Miss S. Ito for her preparation of figures.
31
References Allison, T. (1962) Recovery functions of somatosensory evoked responses in man, Electroencephalogr. Clin. Neuropbysiol., 14: 331-343. Chiappa, H.K. (1983) Parkinson's Disease, Evoked Potentials in Clinical Medicine, Raven Press, New York, p 298. Crutcher, M.D. and M.R. DcLong (1984) Single cell studies of the primate potamen. II. Relations to direction of movement and pattern of muscular activity. Exp. Brain Res., 53: 244-258. Demetrescu, M. and M. Demetrescu (1962) The inhibitory action of the caudate nucleus in cortical primary receiving areas in the cat, Electroencephalogr. Clin. Neuropbysiol. 14: 37-52. Desmedt, J.E., E. Brunko, J. Debecker and J. Carmeliet (1974) The system bandpass required to avoid distortion of early components when averaging somatosensor~' evoked potentials. Electroencephalogr. Clin. Neuropbysiol., 37: 407-410. Eccles, J.C. (1966) Properties and functional organization of cells in the ventrobasal complex of the thalamus. In: D.P. Purpura and M.D. Yahr (Eds), The Thalamus, Columbia University Press, New York, pp. 129-138. Ehle, A.L., R.M. Stewart, N.A. Lellelid and N.A. Leventhal (1984) Evoked potentials in Huntington's ~iscase. A comparative and longitudinal study. Arch. Neurol., 41: 379-382. Emnri, T., T. Yamada, Y. Seki, A. Yasuhara, K. Ando, Y. Honda, A.A. Leis and Vachatimanont (1991) Recovery functions of fast frequency potentials in the initial negative wave of median SEP. Electroencephalogr. Clin. Neurophysiol., 78:116-123. Greenwood, P.M. and W.R. Goff (1987) Modification of median nerve somatic evoked potentials by prior median nerve, peroneal nerve, and auditory stimulation. Electroencephalogr. Clin. Neurophysiol., 68: 295-302. Grosswasser, Z., A. Kriss, S. Dell, P. Toshniwal and A.M. Halliday (1985) Pattern VEPs, SEPs and BAEPs in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 61: SI66. Hoehn, M.M. and M.D. Yahr (1967) Parkinsonism: onset, progression, and mortality. Neurology, 17: 427-442. Iragui-Madoz, V.J. and W.C. Wiederholt (1977) Far-field somatosensory evoked potentials in the cat: correlation with depth record. ing. Ann. Neurol., 1: 569-574. Jfirg, J. and H. Gerhard (1987) Somatosensory, motor and special visual evoked potentials to single and double stimulation in "Parkinson's disease" an early diagnostic test? J. Neural. Transm., 25 (Suppl): 81-88. Kanda, F., K. Jinnai, K. Takahashi, H. Abe, M. Yasuda, K. Tada and T. Fujita (1989) Somatosensory evoked potentials in Huntington's disease - Studies with paired stimulation. Electromyogr. Clin. Neurophysiol., 29: 287-291. Koller, W.C. (1984) Sensory symptoms in Parkinson's disease. Neurology, 34: 957-959.
Krauthamer, G., P. Feltz and D. Albe-Fessard (1967) Neurons of the medial diencephalon. I!. Excitation of central origin. J. Neumphysiol., 30: 81-97. Lueders, H., R.P. Lesser, J. Hahn, D.S. Dinner and G. Klein (19831 Cortical somatosensory evoked potentials in response tc hand stimulation. J. Neurosurg., 58: 885-894. Lueders, H., J. Andrish, A. Gurd, G. Weiker and G. Klem (1981) Origin of far-field subcortical potentials evoked by stimulation of the posterior tibial nerve. Electroencephalogr. Clin. Neurophysiol., 52: 336-344. Meyer-Hardting, E. and W.C. Wiederholl (1977) Recovery function of short-latency components of the human somatosensory evoked potential. Arch. NeuroL, 40: 290-293. Namerow, N.S. (1970) Somatosensory recovery functions in multiple sclerosis patients. Neurology, 20: 813-817. Noth, J., L. Engel, H.-H. Friedemann and H.W. Lange (1984) Evoked potentials in patients with Huntington's disease and their offspring. I. Somatosensory evoked potentials. Electroencephalogr. Clin. Neurophysiol., 59: 134-141. Reisin, R.C., D.S. Goodin, M.J. Aminoff and M.M. Mantle (1988) Recovery of peripheral and central responses to median nerve stimulation. Electroencephalogr. Clio. Neurophysiol., 69: 585588. Rossini, P.M., F. Babiloni, G. Bernardi, L. Cecchi, P.B. Johnson, A. Malentacca, P. Stanzione and A. Urbano (1989) Abnormalities of short-latency somatosensory evoked potentials in parkinsonian patients. Electroencephalogr. Clin. Neurophysiol., 74: 277-289. Schneider, J.S. and T.I. Lidsky (1981) Processing of somatosensory information in striatum of behaving cats. J. Neurophysiol., 45: 841-851. Schneider, J.S., S.G. Diamond and C.K. Markham (1987) Parkinson's disease: sensory and motor problems in arms and hands. Neurology, 37: 951-956. Shagass, C. and M. Schwartz (1964) Recovery functions of somatosensory peripheral nerve and cerebral evoked responses in man. Electroencephalogr. Clin. Neurophysioi., 17: 126-135. Solder, S.R., S. Fahn, W.P. Isgreen and L.J. Cote 0976) Primary sensory symptoms in parkinsonism. Neurology, 26: 423-429. Snider, S. and R. Sandyk (1987) Sensory dysfunction. In: W.C. Koller (Ed), Handbook of Parkinson's disease, Marcel Dekker, New York, pp. 171-180. Takahashi, K., E. Okada and Y. Fujitani (1972) Somatosensory and visual evoked potentials in Huntington's chorea. Rinsho shinkeigaku, 12: 381-385. Tsuji, S., H. Liiders, R.P. Lesser, D.S. Dinner, and G. Klein (1984) Subcortical and cortical somatosensory potentials evoked by posterior tibial nerve stimulation: normative values. Electroencephalogr. Clin. Neurophysiol., 59: 214-228. Wiederholt, W.C. and V.J. Iragui-Madoz (1977) Far-field somatosensory potentials in the rat. Electroencphalogr. Clin. Neuropbysiol., 42: 456-465.
Journal of the Neurological Sciences, 108 (1992) 24-31 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022 510X/92/$05.00
JNS 03708
Recovery functions of somatosensory evoked potentials in parkinsonian patients Kenji N a k a s h i m a , T a t s u o N i t t a a n d K a z u r o T a k a h a s h i Dicision of Neurology, blstitute of Neurological Sciences, Tottori University Faculty of Medicine, 86 Nishimachi, Yonago (Japan) (Received 30 October, 1990) (Revised, received 17 September, 1991) (Accepted 25 September. 1991)
Key words: Somatosensory evoked potential; Recovery function; Paired stimulation; Sensory disturbance; Parkinson's disease
Summary The aim of this study was to analyze the recovery functions of median nerve somatosensory cvoked potentials (SEPs) and to clarify changes in the somatosensory system of patients with Parkinson's disease (PD). The central and frontal SEPs and the nerve potential of the median nerve were examined. The latencies and amplitudes of SEPs produced by a single shock in patients with PD were normal. The recovery functions of central SEPs showed a low degree of suppression in patients with PD, although the recovery curve of the nerve potential in PD patients was almost normal. The change in recovery curves of SEPs in patients with sensory complaints were more noticeable than that in patients without sensory disturbance. A low degree of suppression of central SEPs might play a role in the sensory complaints of PD patients.
Introduction Sensory disturbances are sometimes observed in patients with Parkinson's disease (PD). Normal somatosensory evoked potentials (SEPs) have been reported in PD patients (Chiappa 1983; Keller 1984; Grosswasser et al. 1985; J6rg et al. 1987). Recently, the recovery functions of SEPs have been reported (Allison 1962; Shagass et al. 1964; Namerow 1970; Lueders et al. 1983; Greenwood et al. 1987; J6rg et al. 1987; Reisin et al. 1988; Kanda et al. 1989). The purpose of this study is to clarify changes in the somatosensory system of patients with PD through the analysis of the recovery functions of SEPs. We evaluated the recovery functions of the nerve and Erb's potentials and of the frontal and central components of SEPs.
Subjects and methods Thirteen normal subjects (aged 27-75 years; mean 53.8) and 18 patients with PD (aged 41-68 years; mean
Correspondence to: K. Nakashima, Division of Neurology, Institute of Neurological Sciences, Tottori University Faculty of Medicine, 86 Nishimachi, Yonago, 683, Japan. Tel: 0859-34-8032; Fax: 0859-348083.
57.0; Hoehn and Yahr (1967) grade II, 3 patients; grade llI, 15 patients) were examined. All patients were receiving medication; levodopa-carbidopa, trihexyphenydil, amantadine, etc. All patients had almost normal intellect, evaluated through the usual clinical interviews. They were fully evaluated through the conventional neurological examination. All patients with PD were examined using head computed tomography and cervical spine radiographs. The patients who showed obvious abnormality in these examinations were excluded from the present study. Clinical details of the 18 patients with PD examined in the present study are shown in Table 1. Six of the 18 patients suffered from dysesthesia or pain sensation. The peripheral nerve conduction study of the median nerve in patients with sensory complaints was almost normal and not different from those in patients without sensory complaints (Table 1). Before the experiments, the consent of subjects was obtained after informing them fully about the aims and procedures of this study. In all the subjects, the recovery functions of central SEPs were evaluated. The recovery functions of frontal SEPs were evaluated in 8 of the 13 control subjects and 10 of the 18 patients with PD. The recovery function of the nerve potential of the peripheral nerve was also examined in 6 normal subjects and 9 PD patients. The subjects were seated in chairs and asked to close their eyes. The stimuli were 0.1-ms square wave
25 TABLE 1 SUBJECTS Duration: duration o f the illness (years). Scale: H o e h n and Yahr scale; nd: no*. done. Age
Sex
Duration
Scale
59 54 64 41 62
F M F F F
7 2 4 5 I0
111 111 111 111 111
63 60 62 62 56 58 59 62 56 63 07 24 62
F F M F F M M F F M F F M
3 2 1 3 l0 8 10 6 8 6 12 5 2
Ill 11 II !! Ill iii Ill Ill ill III Ill Ill !11
Sensory disturbance
dysesthesia in the legs dysesthesia in the right arm and leg dysesthesia in the extremities pain in the extremities dysesthesia in the u p p e r and lower extremities dysesthesia in the left arm -
electrical pulses delivered to the unilateral median nerve at the wrist. The intensity was 10% above the motor threshold for thumb twitch. Single and double shocks were given alternately at 2 Hz. For double stimulation, the same intensity was used for conditioning (preceding) and test (following) stimulation through the same stimulating electrodes. The intervals between conditioning and test shocks were 10, 20, 30, 40, 50, 60, 70 and 80 ms. A recording electrode for central SEPs was placed over the contralateral scalp, at a point 7 cm lateral and 2 cm posterior to the vertex, and for frontal SEPs at Fz (according to a 10-20 system) with the linked-ear reference electrodes. For the Erb potential, an active electrode was placed over the ipsilateral Erb's point with a reference electrode at the contralateral Erb's point. For the nerve potential of the median nerve, an active electrode was placed over the median nerve at the elbow, just medial to the brachial artery and with a reference electrode at a point 2 cm medial to the active electrode. Filter settings of 5-300 Hz for SEPs and 50-3000 Hz for Erb's and nerve potentials were used. Signals were fed into a NEC San-ei Signal Processor 7T18A. A summation was taken of 128-256 responses with an analysis time of 200 ms and repeated. A pre-stimulus trigger was used for 20 ms prior to the conditioning or single stimulus. The SEP components were labeled according to their positive (P) or negative (N) polarity and their peak latencies. We evaluated the latencies and the amplitudes (peak to peak) of frontal and central SEPs produced by the single shocks. Then the recovery curves
Median nerve conduction motor ( m / s )
sensory ( m / s )
50.6 50.1 55.8 57.0
53.4 55.6 64.9 57.2
51.0 52.6 56.8 nd nd nd nd nd 55.5 nd 57.4 54.9 61.4 nd
64.3 60.0 58.5 nd nd nd nd nd 60.6 nd 69.1 57.8 65.2 nd
of SEP amplitudes were calculated. In order to determine the real responses produced by the test shock (R2), the values of responses produced by the single shock (R1) were subtracted from the values produced by the combined conditioning and test shock (R1 + R2) at each stimulus interval (Fig. 1). Recovery functions were expressed as a percentage of the values produced by the single shocks. Statistical comparisons were made using a medical statistical package, "STAX", designed by the medical computer research association, University of Tokyo. In the study of recovery functions in normal subjects, the Wilcoxon signed rank test was used at each time interval between conditioning and test shocks. The Wilcoxon rank-sum test was used for statistical analysis of the data from the recovery functions of SEPs in normal subjects and patients with PD. Two-way analysis of variance was used for statistical analysis of the differences in recovery curves of SEPs between normal subjects and patients with PD. Since the test in this statistical package was designed for less than eight levels within each group, recovery functions with less than eight interval levels were evaluated in a comparison between normal subjects and patients with PD.
Results
1. Latencies and amplitudes of SEPs The latencies and amplitudes of central and frontal SEPs elicited by single stimulation showed no differ-
26 A,
72-year-old female: normal subject
% 150 4) "O Q.
E t~
/
_
•
¢:
oo
i B,
100
__
| __ ...~...--.
'N,T
, __
-~...
_
_
4) 01 4) 4)
Q.
62-year-old female: PD patient
5O o
. . . .
1 2 50 60 70 80 Conditioning-test interval (ms)
Fig. 2. Recovery curve of central PI3-NI8 amplitude. ** Wilcoxon rank-sum test, P < 0.01; * P < 0.05. Filled circles: normal subjects; empty circles: PD. Mean ( 4-1 S.E.M.) values are plotted.
Fig. I. Representative cases. RI: response produced by single shock. Ri + 2: response produced by the double shocks. R2: response obtained by subtraction of the RI response from the RI +2 response. Interstimulus interval between conditioning (~) and test ( | ) shocks: 60 ms.
ences between patients with PD and normal subjects (Table 2).
2. Recovery functions of central SEP amplitudes In normal subjects, P13-N18 showed a statistically
significant suppression at 10-20 ms intervals (Wilcoxon signed rank test: P < 0.05) and recovery at the 30-ms interval. Suppression was seen again between 40 and 80 ms (Wilcoxon signed rank test at 40: P < 0.01; at 60, 70 and 80 ms: P < 0.05). A low degree of suppression or a slight facilitation of amplitude in patients with PD was seen between 10 and 70 ms (Fig. 2; Wilcoxon rank-sum test at 70 ms: P 0 . 0 5 ; N16-P20, Fi,ii2= 0.139, P > 0.05; P20-N28, Fm.~l2 = 0.555, P > 0.05; N28P44 FL~12=0.675, P > 0 . 0 5 ) , except in two instances when slight facilitation of P14-N16 occurred at the time intervals of 20 and 80 ms (Wilcoxon rank-sum test P < 0.01).
4. Recocery function of the Erb's potential; N9 The recovery function of the Erb's potential in normal subjects did not show an obvious' change (Wilcoxon signed rank test at 10, 20, 30, 40, 50, 60, 70 and 80 ms:
®
25C
Q.
E ell 0 0
a.
I
200]
150
~\\
II!
100
50
G
,0
35 4'0
6.0 ,.0 s'0
Conditioning-test interval (ms)
Fig. 7. Recoveryfunctions of P24-N31 in patients with and without sensory complaints. * Wilcoxon rank-sum test, P < 0.05. Filled circles: PD without sensory complaints; empty circles: PD with sensory complaints. Mean ( + i S.E.M.)values are plotted.
29 % 200
.m
E
150
c 0 o
"6 0 ~B ¢D
10(~
5¢
d
10 20 3'0 4'0 5() 6'0 7'0 8'0 Coationing-test lateral (ms)
Fig. 8. Recovery functions of N31-P44 in patients with and without sensory complaints. Filled circles: PD without sensory complaints; empty circles: PD with sensory complaints. Mean ( _ I S.E.M.) values are plotted.
ms: P < 0.05, at 20 and 60 ms: P > 0.05). The recovery function of the nerve potential in patients with PD was not different from that in normal subjects (Fig. 9; analysis of variance between 10 and 70 ms: F,.9, = 0.886, P > 0.05; Wilcoxon rank-sum test at 10, 20, 30, 40, 50, 60, 70 and 80 ms: P > 0.05).
Discussion Some studies of SEPs in patients with PD have shown no abnormalities (Koller 1984; Grosswasser et
O---O
control (n:6) PD (n:9)
qO
: 120 100
O o 80
o.
' 10
' 20
310
' 40
' 50
60
;0
J 80
Conditioning-test interval (ms) Fig. 9. Recovery function of the nerve potential of the median nerve. No difference between normal subjects and PD patients is observed. Mean ( + 1 S.E.M.) values are plotted.
al. 1985; J6rg et al. 1987). Recently, Rossini et al (1989) showed an abnormally depressed frontal component, "N30" (our N28), in patients with PD, although almost normal responses of central components were observed. The frontal "N30" component may originate from the supplementary motor area (SMA). As the SMA may be strongly connected to the basal ganglia, this may explain why the abnormal N30 was observed in patients with PD. In the present study, the amplitude of P20-N28 in patients with PD was slightly lower than that in normal subjects, although this difference was not statistically significant. One of the reasons for the difference may be the band-pass filter. We used a rather narrow band-pass filter (5-300 Hz), while Rossini et al. (1989) used a wide band-pass filter (13000 Hz). In the recordings of SEPs, the ideal band-pass filter has been discussed (Desmedt et al. 1974; Lueders et al. 1981; Tsuji et ai. 1984). Tsuji et al. (1984) recommended the restricted band-pass filter of 30-250 Hz for definition of the early cortical potentials to posterior tibial nerve stimulation. In our preliminary study, the effect of a band-pass filter on the recovery functions of SEPs was examined. If a wide band-pass filter was used, the recovery curves fluctuated more than those obtained through the band-pass filter of 5-300 Hz. Stable records of SEPs should be used for subtraction through the adequate restricted band-pass filter. Our results were almost compatible with the previously reported U-shaped recovery curves (Lueders et al. 1983). A U-shaped recovery curve of SEPs may he caused by the initial facilitation in the first 10-20 ms produced by the short excitatory postsynaptic potentials and followed by the prolonged inhibitory postsynaptic potentials in the thalamic nuclei; VPL (Eccles 1966). The recovery curves of central SEPs indicated suppression, temporary recovery (or facilitation), suppression and recovery. The recovery curves of frontal SEPs showed no suppression. The different patterns of recovery curves might be caused by the different sources of SEP components (Greenwood 1987). Possible mechanisms for the suppressive effects shown by recovery curves of SEPs may include local recurrent, reticulofugal and corticofugal inhibitory effects (Greenwood 1987). Animal studies suggested that the responses of SEPs to repetitive stimulation were different from those of the peripheral nerve (Iragui-Madoz et al. 1977; Wiederholt et al. 1977; Meyer-Hardting et al. 1983). Using depth recording techniques in the cat, IraguiMadoz et al. (1977) studied the responses to repetitive stimulation. The amplitude of potentials in the cervical dorsal column and nucleus cuneatus did not change with frequencies of stimulation, although that in the ventral posterior thalamus and thalamic' radiation declined. As the intercorrelations between the peripheral nerve and cerebral response recovery functions in man
30 showed no coefficient, the influence of peripheral factors on the cerebral recovery function might be negligible (Shagass et al. 1964). Reisin et al. (1988) also found that the peripheral response differed in the manner of recovery from the N20 scalp response. In the present study, the recovery curves of the nerve and Erb's potentials in patients with PD were the same as those in normal subjects, though those of the central SEPs showed obvious differences between PD patients and normal subjects. Furthermore, the Erb's potential showed neither obvious facilitation nor suppression in control subjects. We supposed, therefore, that the central nervous system played an important role in the recovery functions of central SEPs. Since the recovery functions of frontal SEPs did not show obvious suppression, the subcortical components (P13 and N16) and the SMA component (N28) were not suppressed by preceding stimulation. In patients with Huntington's disease (HD), low amplitudes of SEPs were observed (Takahashi et al. 1972; Noth et al. 1984; Ehle et al. 1984; Kanda et al. 1989). The reduction of SEP amplitudes might reflect a reduced cortical activation by the thalamic efferent in HD (Ehle et al. 1984; Kanda et al. 1989). Kanda et al. (1989) studied the recovery functions of SEPs in patients with HD. The facilitation of amplitude in the recovery functions of N20-P24 (our NI8-P24) at the 10-ms interval was seen in patients with HD. They speculated that the affected striatum in HD influenced the thalamus, resulting in the reduction of SEP amplitudes, and that the conditioning stimuli removed some functional block in the thalamus. The disturbance of the basal ganglia might contribute to the low degree of suppression appearing in the recovery curves of SEP amplitudes. In 30-40% of PD patients, sensory disturbances have been observed (Snider et al. 1976; Koller 1984; Snider et al. 1987) They complained of burning sensations, pain or paresthesia-like sensations. The pathophysiological details of sensory disturbances in patients with PD are not yet fully understood. Sensory disturbances in PD patients may be caused by the disturbance of the peripheral or central nervous system (Snider et al. 1976). In the present study, the recovery curves of the peripheral nerve and Erb's potentials in patients with PD were the same as those in normal subjects. Clinical neurological examination, peripheral nerve conduction tests and cervical spine radiographs showed no abnormality leading to sensory disturbance in PD patients. Furthermore, the head computed tomography did not show any obvious abnormality. Only central SEPs showed differences in recovery curves between normal subjects and PD patients. Therefore, the changes in the central nervous system, rather than the peripheral factor, may contribute to the sensory disturbance in PD patients. The basal ganglia appears
to modulate sensory input (Koller 1984; Snider et al. 1976; Schneider et al. 1987). Snider et al. (1976) suggested that the locus ceruleus and/or the substantia nigra played roles in sensory release in patients with PD. The locus ceruleus, the substantia nigra and the striatum appear to have inhibitory effects on the sensory system (Krauthamer et al. 1967). The patients with sensory disturbance showed more obvious lack of suppression and/or bigger facilitation than patients without sensory complaints. The abnorma!ity of the recovery function of SEPs related to the sensory disturbance in PD. A reduced inhibitory function may contribute to sensory disturbance in PD patients. The basal ganglia receives input from the sensory afferents (Schneider et al. 1981; Crutcher et al. 1984). The sensory system is inhibited by the basal ganglia (Demetrescu et al. 1962; Krauthamer et al. 1967). Somatosensory stimulation may provoke not only SEP responses, but also the basal ganglial inhibition of the sensory system. This basal ganglial inhibition may contribute to central SEP amplitude recovery curves. The recovery functions of central SEP amplitudes in patients with PD showed a lower degree of suppression. Temporarily, we suspected that the basal ganglial inhibition of the sensory system was low in patients with PD. However, recovery curves of frontal SEPs in patients with PD showed little difference from those in normal subjects. The basal ganglial inhibitory effects on the frontal components of SEPs are slight. Another explanation for the lack of suppression a n d / o r the facilitation of the recovery curves of SEPs may be a change in the reticular formation. The reticular formation receives input from somatosensory stimulation, activates the cerebral cortex and might play a role, at least in part, in the recovery curves of SEPs. The reticular formation also receives input signals from the basal ganglia. In PD, it might be disturbed due to basal ganglial dysfunction. The abnormality of the recovery curves of SEPs might be produced by the changes in reticular formation function in PD. In conclusion, we have described the low degree of suppression in the recovery curves of central SEP amplitudes. The low degree of suppression in the sensory system may be related to the abnormal function of the basal ganglia. This study demonstrates an underlying physiological change of the sensory systems in patients with PD.
Acknowledgements This work was supported in part by funds from the Research Committee of CNS Degenerative Disease, the Ministry of Health and Welfare of Japan. This work was presented at the first International Congress of Movement Disorders in Washington, D.C., on April 25th, 1990. We are indebted to Dr. E. Rin for his arrangements in measuring the data, to Miss R. Maeda for her technical assistance and to Miss S. Ito for her preparation of figures.
31
References Allison, T. (1962) Recovery functions of somatosensory evoked responses in man, Electroencephalogr. Clin. Neuropbysiol., 14: 331-343. Chiappa, H.K. (1983) Parkinson's Disease, Evoked Potentials in Clinical Medicine, Raven Press, New York, p 298. Crutcher, M.D. and M.R. DcLong (1984) Single cell studies of the primate potamen. II. Relations to direction of movement and pattern of muscular activity. Exp. Brain Res., 53: 244-258. Demetrescu, M. and M. Demetrescu (1962) The inhibitory action of the caudate nucleus in cortical primary receiving areas in the cat, Electroencephalogr. Clin. Neuropbysiol. 14: 37-52. Desmedt, J.E., E. Brunko, J. Debecker and J. Carmeliet (1974) The system bandpass required to avoid distortion of early components when averaging somatosensor~' evoked potentials. Electroencephalogr. Clin. Neuropbysiol., 37: 407-410. Eccles, J.C. (1966) Properties and functional organization of cells in the ventrobasal complex of the thalamus. In: D.P. Purpura and M.D. Yahr (Eds), The Thalamus, Columbia University Press, New York, pp. 129-138. Ehle, A.L., R.M. Stewart, N.A. Lellelid and N.A. Leventhal (1984) Evoked potentials in Huntington's ~iscase. A comparative and longitudinal study. Arch. Neurol., 41: 379-382. Emnri, T., T. Yamada, Y. Seki, A. Yasuhara, K. Ando, Y. Honda, A.A. Leis and Vachatimanont (1991) Recovery functions of fast frequency potentials in the initial negative wave of median SEP. Electroencephalogr. Clin. Neurophysiol., 78:116-123. Greenwood, P.M. and W.R. Goff (1987) Modification of median nerve somatic evoked potentials by prior median nerve, peroneal nerve, and auditory stimulation. Electroencephalogr. Clin. Neurophysiol., 68: 295-302. Grosswasser, Z., A. Kriss, S. Dell, P. Toshniwal and A.M. Halliday (1985) Pattern VEPs, SEPs and BAEPs in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 61: SI66. Hoehn, M.M. and M.D. Yahr (1967) Parkinsonism: onset, progression, and mortality. Neurology, 17: 427-442. Iragui-Madoz, V.J. and W.C. Wiederholt (1977) Far-field somatosensory evoked potentials in the cat: correlation with depth record. ing. Ann. Neurol., 1: 569-574. Jfirg, J. and H. Gerhard (1987) Somatosensory, motor and special visual evoked potentials to single and double stimulation in "Parkinson's disease" an early diagnostic test? J. Neural. Transm., 25 (Suppl): 81-88. Kanda, F., K. Jinnai, K. Takahashi, H. Abe, M. Yasuda, K. Tada and T. Fujita (1989) Somatosensory evoked potentials in Huntington's disease - Studies with paired stimulation. Electromyogr. Clin. Neurophysiol., 29: 287-291. Koller, W.C. (1984) Sensory symptoms in Parkinson's disease. Neurology, 34: 957-959.
Krauthamer, G., P. Feltz and D. Albe-Fessard (1967) Neurons of the medial diencephalon. I!. Excitation of central origin. J. Neumphysiol., 30: 81-97. Lueders, H., R.P. Lesser, J. Hahn, D.S. Dinner and G. Klein (19831 Cortical somatosensory evoked potentials in response tc hand stimulation. J. Neurosurg., 58: 885-894. Lueders, H., J. Andrish, A. Gurd, G. Weiker and G. Klem (1981) Origin of far-field subcortical potentials evoked by stimulation of the posterior tibial nerve. Electroencephalogr. Clin. Neurophysiol., 52: 336-344. Meyer-Hardting, E. and W.C. Wiederholl (1977) Recovery function of short-latency components of the human somatosensory evoked potential. Arch. NeuroL, 40: 290-293. Namerow, N.S. (1970) Somatosensory recovery functions in multiple sclerosis patients. Neurology, 20: 813-817. Noth, J., L. Engel, H.-H. Friedemann and H.W. Lange (1984) Evoked potentials in patients with Huntington's disease and their offspring. I. Somatosensory evoked potentials. Electroencephalogr. Clin. Neurophysiol., 59: 134-141. Reisin, R.C., D.S. Goodin, M.J. Aminoff and M.M. Mantle (1988) Recovery of peripheral and central responses to median nerve stimulation. Electroencephalogr. Clio. Neurophysiol., 69: 585588. Rossini, P.M., F. Babiloni, G. Bernardi, L. Cecchi, P.B. Johnson, A. Malentacca, P. Stanzione and A. Urbano (1989) Abnormalities of short-latency somatosensory evoked potentials in parkinsonian patients. Electroencephalogr. Clin. Neurophysiol., 74: 277-289. Schneider, J.S. and T.I. Lidsky (1981) Processing of somatosensory information in striatum of behaving cats. J. Neurophysiol., 45: 841-851. Schneider, J.S., S.G. Diamond and C.K. Markham (1987) Parkinson's disease: sensory and motor problems in arms and hands. Neurology, 37: 951-956. Shagass, C. and M. Schwartz (1964) Recovery functions of somatosensory peripheral nerve and cerebral evoked responses in man. Electroencephalogr. Clin. Neurophysioi., 17: 126-135. Solder, S.R., S. Fahn, W.P. Isgreen and L.J. Cote 0976) Primary sensory symptoms in parkinsonism. Neurology, 26: 423-429. Snider, S. and R. Sandyk (1987) Sensory dysfunction. In: W.C. Koller (Ed), Handbook of Parkinson's disease, Marcel Dekker, New York, pp. 171-180. Takahashi, K., E. Okada and Y. Fujitani (1972) Somatosensory and visual evoked potentials in Huntington's chorea. Rinsho shinkeigaku, 12: 381-385. Tsuji, S., H. Liiders, R.P. Lesser, D.S. Dinner, and G. Klein (1984) Subcortical and cortical somatosensory potentials evoked by posterior tibial nerve stimulation: normative values. Electroencephalogr. Clin. Neurophysiol., 59: 214-228. Wiederholt, W.C. and V.J. Iragui-Madoz (1977) Far-field somatosensory potentials in the rat. Electroencphalogr. Clin. Neuropbysiol., 42: 456-465.
