Movement Disorders Vol. 9, No. 4, 1994, pp. 42-30 0 1994 Movement Disorder Society

Abnormalities of Somatosensory Evoked Potentials in Spasmodic Torticollis

L. Mazzini, M. Zaccala, and C. Balzarini

“Clinica del Lavoro” Foundation (IRCCS), Medical Centre of Rehabilitation, Veruno, Italy

Summary: This study examined the N20 and N30 waves of somatosensory evoked potentials (SEPs) from median nerve stimulation at three different stimulation rates (1, 3, and 6 Hz) in patients with idiopathic spasmodic tor- ticollis (ST). The data were compared with those collected from a group of patients affected by Parkinson’s disease (PD) and normal age-matched sub- jects. N30 amplitude was significantly decreased in both groups of patients with respect to the controls. The decrease was larger in patients with ST. The N20 wave remained stable. The latencies of both waves were unchanged. When the stimulus rate was increased, the N30 amplitude decreased signifi- cantly, with a similar trend observed in both patients and controls. The isolated abnormalities of the N30 wave in both ST and PD support the hypothesis of a common physiopathogenetic mechanism that involves the basal ganglia or their connections with the supplementary motor area. Key Words: Spasmodic tor- ticollis-Somatosensory evoked potentials-N30-Stimulus frequency-Basal ganglia-Supplementary motor area.

Abnormalities of somatosensory evoked poten- tials (SEPs) in basal ganglia disorders have been described by several investigators. A decrease in the amplitude of N30 has been reported in Parkin- son’s disease (PD) (1) and Huntington’s disease (2- 4), whereas an increase in the amplitude of N30 has been found in segmental dystonia of the arm and in generalized dystonia (5). An increasing number of observations (6,7) suggest that the N20 and the N30 are generated by different neural structures, namely the parietal cortex and the supplementary motor area (SMA), respectively (6,8). To explore the hy- pothesis that the basal ganglia or their projections may be involved in the pathophysiology of spas- modic torticollis (ST), we studied SEPs from me- dian nerve stimulation recorded over the parietal and frontal scalp in patients with ST. The data were compared with those found in a group of patients affected by PD and normal age-matched subjects.

Address correspondence and reprint requests to Dr. L. Mazzini at “Ciinica del Lavoro” Foundation (IRCCS), Medical Centre of Rehabilitation, 1-28010 Veruno, Italy.

METHODS

Nineteen patients (14 men and five women) with idiopathic ST (eight with the head rotation to the left and 11 to the right) were studied. The mean age of the group was 45 years (range 22-69). The dura- tion of the illness ranged from 1 to 12 years. The diagnosis was established on the basis of the clinical features and the absence of focal lesions or cerebral atrophy on computed tomography (CT) scan. The disease was classified as mild (grade 2-3) or severe (grade 4 5 ) according to the dystonia disability rat- ing scale described by Jankovic and Oman (9). We also studied a group of 15 patients (five men and 10 women) with PD (without tremor). The mean age of the group was 64 years (range 43-81). The duration of the disease ranged from 3 to 15 years. The sever- ity of the disease was assessed by means of the Webster rating scale. Nineteen normal volunteers (13 men and six women) served as the control group; none had a history of neurological illness. The mean age of the group was 53 years (range 33-78). The experimental procedure was approved

426

ABNORMALITIES OF SEPs IN SPASMODIC TORTICOLLIS 427

by the local ethical committee, and informed con- sent was obtained from all subjects. Dystonic pa- tients and controls were not receiving any medica- tion, whereas all the parkinsonian patients were on stable treatment with L-dopa. None had the long- term syndrome (10).

The study was performed in a quiet room with the subject lying comfortably on a bed. In this position, the dystonic patients were able to relax their neck muscles. SEPs were elicited by surface stimulation of the median nerve delivered at both wrists sepa- rately. Stimuli were square pulses (0.1 ms dura- tion), and the intensity was set at 1.5 times the mo- tor threshold to produce the thumb twitch. Stimulus rates of 1, 3, and 6 Hz were used. SEPs were re- corded bilaterally with Ag/AgCl disk electrodes placed over C3’ and C4’, Fz, F3, and F4 (according to the 10-20 International System). All electrodes were referred to the earlobe contralateral to the stimulated arm. Electrodes were also applied to Erb’s point to record the afferent volley. Electrode impedance was kept below 5 k a . Tracings were fil- tered (bandpass 10-3,000 Hz) and averaged (500 ar- tifact-free responses), analysis time was 50 ms. Two runs of 500 trials were stored separately and super- imposed.

Latencies were measured from the onset of the stimulus to the peak of the negative deflection. Ow- ing to the different influence of the stimulus rate on P14 and P27 waves (11,12), the amplitudes of the N20 at C3’ and C4’ were measured peak to peak both from the previous P14 and to the following P27. Amplitude of the N30 peak at Fz, F3, and F4 was measured from the previous P22. The ampli- tude of the N9 was measured from peak to peak. Student’s t test was used to compare the amplitudes of the waves in patients and controls. Statistical analysis of the differences between the amplitudes

of cortical SEPs at different stimulus repetition rates was performed using an analysis of variance (ANOVA). Spearman’s rank correlation coeffi- cients were calculated to examine the relationship between SEP components and clinical variables.

RESULTS A clear P14-N2O-P27 sequence at parietal scalp

and a P22-N30 sequence at frontal scalp were re- corded in all normal subjects. No significant differ- ence in the absolute and interpeak latencies was found between patients (ST and PD) and normal subjects. The peak latencies were not affected sig- nificantly by the stimulus rate in the patients or in the controls.

The parietal P141N20 and N201P27 complexes were identified in all patients with ST; all were of normal shape and amplitude. The frontal P22/N30 was significantly lower than that recorded in the control group. Figure 1 shows examples of SEP waveforms recorded from Erb’s point, C3’, and Fz at the stimulation rate of 3 Hz in a normal subject and in a dystonic patient. Typically, the frontal N30 is very small, whereas the amplitude and latency of N9 and N20 remain unchanged. The patient was a 47-year-old woman with a year-long history of typ- ical ST who presented with moderate-severity rota- tion of the head to the right. The mean values of the amplitudes of parietal P14/N20 and N20P27 in the ST group did not significantly differ from those of the controls, even when the direction of rotation or the severity of the disease were taken into account (Fig. 2, upper graph). The mean amplitude of the frontal N30 was significantly decreased in dystonics as compared with normal subjects (Fig. 2, lower graph). No correlation was discernible between SEPs and clinical variables: the amplitude of P22/ N30 did not correlate significantly with disease du-

Spasmodic T o r t i c o l l i s Normal Subject

FIG. 1. SEP to right median nerve stimulation in a 36-year-old normal subject (left panel) and in a 47-year- old patient with ST (right panel). The parietal and frontal compo- nents show a typical configuration. Although the amplitudes of N20 are similar in the two subjects, frontal N30 is much lower in the ST pa- tient.

N30 N30

F Z & 1’’’ Erb 15”” .

lOmsec

Movement Disorders, Vol. 9 , No. 4, 1994

428 L. MAZZINI ET AL.

6 - PV

PV . m 5 - U 3 0 ' z 3

E

4

D 3 4 - = a '

0. -

g 3 :

a 2

1 2

0 17

PI 4lN20

- N --t ST *

. , . , . , . , . , . , . ,

FIG. 2. Mean amplitude of parietal P14hJ20 and N20/P27 and frontal P22/N30 evoked at the stimulation rate of 3 Hz. Each bar represents the grand mean and the standard error. Because no interside difference of the amplitudes was found in the normal subjects, the reported mean values include the two sides. The left bars of each graph refer to the mean amplitude of the dif- ferent waves in controls (N) and ST patients. The middle bars correspond to the mean amplitudes of the waves on the side ipsilateral or contralateral to the side of rotation of the head. The right bars show the mean amplitude in the mild and severe pa- tients.

ration (r = 0.29) or the severity of the dystonia (r = 0.08). No significant correlation was found between the amplitude of N30 and the side of the head rota- tion, even when N30 was recorded at F3 or F4 (p = 0.49). A marked decrease in amplitude of the N30 wave was observed in all patients except three with

P221N30 PYBl T

0 1 2 3 4 5 6 7 k

FIG. 3. Effects of the stimulus repetition rate on the amplitude of P14/N20, N20/P27, and P22/N30. Each graph compares the mean values and standard errors of the amplitudes of the waves in controls and patients (ST and PD) at the three stimulus rates ( I , 3 , and 6 Hz).

Movement Disorders, Vol. 9, No. 4, 1994

ABNORMALITIES OF SEPs IN SPASMODIC TORTICOLLIS 429

N30 waves in the control group and patients. The P14/N20 amplitude was not affected by the stimulus frequency in the control group (p = 0.87) or the patients (p = 0.64). N20/P27 decreased as a func- tion of the frequency of stimulation in both patients with ST and in the control subjects, but not signif- icantly (p = 0.06); this was probably related to the sensitivity of P27 to the stimulus rate, as previously reported by other investigators (1 1,12). N30 signif- icantly decreased as the stimulus rate increased in patients with ST (p < 0.007), in patients with PD (p < 0.002), and in the control subjects (p < 0.002). However, there was a clear difference between the patients (ST and PD) and the control subjects, and between the patients with ST and PD at every stim- ulus frequency. The smaller amplitude of N30 in patients with ST as compared with those with PD was statistically significant at every stimulus fre- quency.

DISCUSSION Our data show a significant diminution of the am-

plitude of the N30 wave of median SEP in ST and PD. This change occurred despite the fact that the N20 did not differ from normal subjects. This was true at all stimulus frequencies. To our knowledge there have been no previous studies of N30 in ST. Narayan et al. (13), who studied N20 in ST, re- ported that it was in the normal range.

A “gating” mechanism could be hypothesized for the reduction of N30 in the present study. The grounds for such a suggestion are that active or pas- sive movements are known to reduce the amplitude of cortical SEPs, particularly of N30 (14,15), even if sustained tonic muscular activity does not gate sen- sory input (16). However, given that we found no statistically significant correlation between the re- duction of N30 and (a) the disease severity of either ST or PD, (b) the direction of the head rotation in ST, or (c) the rigidity in PD, that explanation does not seem to apply to our patients. Moreover, during SEP recording, the patients were fully relaxed, with the head in a neutral position.

Previous studies have failed to find any correla- tion between SEP changes and clinical variables in other movement disorders, such as Huntington’s (4) or Steel-Richardson’s disease (17). Yamada’s (3) re- port of abnormalities of SEPs in asymptomatic rel- atives of patients with Huntington’s disease adds further evidence that there is no correlation be- tween neurophysiological findings and disease se- verity. We also can rule out the suggestion that cor-

tical atrophy might have played a role because re- sults of all of the CT scan examinations were normal. Reilly et al. (5) recently reported an in- crease in the amplitude of N30 in 10 patients with segmental dystonia of the arm and generalized dys- tonia. This result could be explained by hypothesiz- ing a different phy siopathological mechanism that underlies the two forms of dystonia. In fact, some neurophysiological differences between the two forms of dystonia have been described by Panizza et al. (18), who found that the patients with ST show a “facilitation” of the late recovery of H-reflex curve similar to that found in parkinsonian patients, but this result was not evident in other forms of dystonia, such as writer’s cramp or blepharospasm. Moreover, in a clinicopathological study of symp- tomatic dystonia, Rothwell (19) reported that a le- sion was present in the ipsilateral head of the cau- date nucleus in all patients with torticollis, whereas a thalamic lesion was present in four of six patients with hand dystonia. Furthermore, different types of dystonia may have different neurochemical bases. For example, in parkinsonian patients the L-dopa- induced dystonia was seen as an off-period or peak- dose phenomenon. Each type shows a distinctive pattern of localization of dystonic spasms: peak- dose dystonia shows a marked predilection for the orofacial region or neck, whereas off-period dysto- nia is more frequent in the limbs (20).

The stimulus frequency affects the amplitude of N30 but not that of N20 in normal subjects, as pre- viously reported (1 1,12,21-23). The mechanism of this modulation is unknown. However, it has been suggested that it may be the result of a selective attenuation of the inhibitory postsynaptic potential at higher stimulus frequencies (12). The rate of stim- ulation affects the amplitude of the waves in the same way in normal subjects and in patients with ST or PD. The different response of N20 and N30 to the stimulation rate indicates that these waves are generated by separate neural structures. N30 seems to reflect the activity of multiple generators, with the main contribution from the SMA. In fact, the maximal peak amplitude of the N30 has been re- corded from the scalp overlying the SMA @,IS). Furthermore, unilateral SMA compression due to a meningioma of the falx was found to eliminate this wave on the side of the compression (6). The SMA receives many afferents from the basal ganglia, which could explain the changes in the N30 amplitude in basal ganglia diseases such as PD (l), Huntington’s disease ( 2 4 , and Steel-Richardson’s disease (17).

Muvement Disorders, Vol. 9 , No. 4, 1994

430 L . MAZZINI ET AL.

The predominant N30 wave abnormalities in ST, although it is not a conclusive data, point to a pos- sible functional impairment of the thalamic-basal ganglia-frontal cortex circuit in these patients (25,26). According to our data, N30 amplitude was smaller in ST than in PD, whereas N20 amplitude was of equal size. This finding may reflect a greater change in SMA in ST than in PD. From this view- point one may note that SMA has been shown to control the postural muscles (27,28). It would seem reasonable to hypothesize that two different popu- lations of neurons in SMA may be affected in PD and ST or that different afferents to the SMA are involved.

Acknowledgment: We thank Professor Marco Schiep- pati and Dr. Antonio Nardone for their constructive crit- icism of the paper and Gilian Jarvis for revising the En- glish text.

1.

2.

3.

4.

5 .

6.

7.

8.

9.

10

I! .

REFERENCES Rossini PM, Babiloni F, Bernardi G, et al. Abnormalities of short-latency somatosensory evoked potentials in parkinso- nian patients. Electroencephalogr Clin Neurophysiol 1989;

Abbruzzese G, Dall’Agata D, Morena M, Reni L, Favale E. Abnormalities of parietal and prerolandic somatosensory evoked potentials in Huntington’s disease. Electroencepha- logr Clin Neurophysiol 1990;77:340-346. Yamada T, Rodnitzky RL, Kameyama S, Matsuoka H, Kimura J. Alteration of SEP topography in Huntington’s patients and their relatives at risk. Electroencephalogr CIin Neurophysiol 1991 ;80:251-261. Topper R, Schwarz M, Podoll K, Domges F, Noth J. Ab- sence of frontal somatosensory evoked potentials in Hun- tington’s disease. Brain 1993;116:87-101. Reilly JA, Hallett M, Cohen LG, Tarkka IM, Dang N. The N30 component of somatosensory evoked potentials in pa- tients with dystonia. Electroencephalogr Clin Neurophysiol

Mauguitre F, Desmedt JE, Coujon J. Astereognosis and dissociated loss of frontal or parietal components of somato- sensory evoked potentials in hemispheric lesions: detailed correlations with clinical signs and computerized tomo- graphic scanning. Brain 1983;106:271-311. Hashimoto S , Segawa Y, Kawamura J, et al. Volume con- duction of the parietal N20 potential to the prerolandic fron- tal area. Brain 1990;113:1501-1509. Desmedt JE, Nguyen TH, Bourguet M. Bit-mapped color imaging of human evoked potentials with reference to the N20, P22, P27 and N30 somatosensory responses. Electro- encephalogr Clin Neurophysiol 1987;68: 1-19. Jancovic J , Orman J. Botulinum-A-toxin for cranial-cervical dystonia: a double blind, placebo-controlled study. Neurol- ogy 1987;37:616-623. Barbeau A. Long-term side-effects of levodopa. Lancet 1971 ;1:395. Delberghe X, Mavroudakis D, Zegers D, Brunko E. The effect of stimulus frequency on post- and pre-central short- latency somatosensory evoked potentials (SEPs). Eleciroen- cephalogr Clin Neurophysiol 1990;77:8&92.

74~277-289.

1992 ;84243-247.

12. Huttunen J , Homberg V. Influence of stimulus repetition rate on cortical somatosensory potentials evoked by median nerve stimulation: implications for generation mechanisms. J Neurol Sci 1991;105:37-43.

13. Narayan TM, Ludwig C, Sat0 S. A study of multimodality evoked responses in idiopathic spasmodic torticollis. Elec- troencephalogr Clin Neurophysiol 1986;63:239-241.

14. Cohen LG, Starr A. Localization, timing and specificity of gating of somatosensory evoked potentials during active movement in man. Brain 1987;110:451467.

IS. Cheron G, Borenstein S. Gating of the early components of

16.

17.

18.

19.

20.

21.

22.

23.

24.

2s.

26.

27,

28.

the frontal and parietal somatosensory evoked potentials in different sensory-motor interference modalities. Electroen- cephalogr Clin Neurophysiol 1991 ;80:522-530. Dimitrov B, Hallett M, Sanes JN. Differential influence of posture and intentional movement on human somatosensory evoked potentials evoked by different stimuli. Brain Res 1989;496:211-218. Abbruzzese G , Tabaton M, Morena M, Dall’Agata D, Favale E. Motor and sensory evoked potentials in progres- sive supranuclear palsy. Mov Disord 1991;6:49-54. Panizza M, Lelli S, Nilsson J , Hallett M. H-reflex recovery curve and reciprocal inhibition of H-reflex in different kinds of dystonia. Neurology 1990;40:824-828. Rothwell JC, Obeso JA. The anatomical and physiological basis of torsion dystonia. In: Marsden D, Fahn S, ed. Move- ment disorders. 2nd ed. Stoneham, MA: Butterworth, 1987:

Poewe WH, Lees AJ, Stem GM. Dystonia in Parkinson’s disease: clinical and pharmacological features. Ann Neurol

Tromberg C , Desmedt JE, Ozaki TH, Nguyen TH, Chalklin V. Mapping SEP to finger stimulation at intervals of 450 to 4000 ms and the issue of habituation when assessing early cognitive components. Electroencephalogr Clin Neurophys- iol 1989;74:347-358. Abruzzese G, Dall’Agata D, Morena M, Reni L, Trivelli G, Favale E. Selective effects of repetition rate on frontal and parietal somatosensory evoked potentials (SEPs). In: Rossini PM, Mauguiere F, eds. New trends and advanced techniques in clinical neurophysiology. EEG suppl41. New York: Elsevier Science, 1990: 145-148. Garcia Larrea L, Bastuji H, Mauguitre F. Unmasking of cortical SEP components by changes in stimulus rate: a to-

3 13-33 1 .

1988;23:73-78.

pographic study.- Electroencephalogr Clin Neurophysiol 1992 $4:7 1-83. Huttunen J , Homberg V, Lange HW. Pre-and postcentral somatosensory evoked potentials in Huntington’s disease: effects of stimulus repetition rate. J Neurol Sci 1993316: 119-124. Lidsky TI, Manetto C, Schneider JS. A consideration of sensory factors involved in motor functions of the basal gan- glia. Brain Res Rev 1985;9: 133-146. Schwarz M, Block F, Topper R, Sontag KH, Noth J. Ab- normalities of somatosensory evoked potentials in the quin- olinic acid model of Huntington’s disease: evidence that basal ganglia modulate sensory cortical input. Ann Neurol 1992;32:358-364. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12:366375. Massion J, Viallet F, Massarino R, Khalil R. La rtgion de I’aire motrice suppltmentaire est impliqute dans la coordi- nation entre posture et mouvement chez l’homme. C R Acad Sci [IZd 1989;308:417423.

Movement Disorders, Vol. 9 , No. 4, 1994