Temporal dispersion of cortically evoked single motor unit potentials in ALS

Journal of the Neurological Sciences 180 (2000) 71–75 www.elsevier.com / locate / jns

Temporal dispersion of cortically evoked single motor unit potentials in ALS Nobuyuki Hirota, Markus Weber, Andrew Eisen* The Neuromuscular Diseases Unit, Vancouver Hospital, 855 West 12 th Avenue and The University of British Columbia, Vancouver, Canada, V5 Z 1 M9

Abstract Peristimulus time histograms (PSTHs) can be used to investigate corticomotoneuronal dysfunction in amyotrophic lateral sclerosis (ALS). The most characteristic change is temporal dispersion of the primary peak. We recorded PSTHs in the extensor digitorum communis with voluntary motor units activation (standard PSTHs) or at rest (non-activated PSTHs). Standard PSTHs were recorded in 29 motor units of 12 healthy control subjects and 12 sporadic ALS patients. Double primary peaks were seen in three motor units of two healthy control subjects and 10 motor units from five ALS patients. The number of subpeaks was up to three in most of the normal motor units as well as in the earlier component of double primary peaks. The subpeaks were smaller and less discernible in the later component of double primary peaks. Non-activated PSTHs of ALS patients demonstrated similar decomposition of subpeaks in the motor units with significantly increased variability of latency. Similar findings in the standard PSTHs and non-activated PSTHs suggest that the abnormalities seen in ALS are independent of the membrane potential of the spinal motoneuron and therefore supraspinal in origin. The decomposed additional later component may indicate activation of slow conducting corticospinal tracts.  2000 Elsevier Science B.V. All rights reserved. Keywords: Amyotrophic lateral sclerosis; Corticospinal tract; Magnetic stimulation; Motor unit

1. Introduction Amyotrophic lateral sclerosis (ALS) is a fatal disease, which selectively involves upper and lower motoneurons. The pathogenic mechanisms remain unclear. In the past decade transcranial magnetic stimulation (TMS) has been used to investigate corticomotoneuronal dysfunction in ALS. Motor evoked potentials evoked by TMS may be absent, delayed or reduced in amplitude [1,2]. A modified technique, peristimulus time histograms (PSTHs) can be used to estimate the excitatory postsynaptic potentials (EPSPs) evoked at the spinal motoneuron [3]. TMS at sub-threshold intensity influences the firing pattern of a voluntarily activated single motor unit. The changes of the firing probability are seen as a primary peak in the PSTHs. In the extensor digitorum communis (EDC) of normal *Corresponding author. Tel.: 11-604-875-4405; fax: 11-604-8755867. E-mail address: [email protected] (A. Eisen).

subjects, a well synchronized primary peak occurs 16–27 ms after a TMS [4]. In ALS various abnormalities have been described. The most characteristic abnormalities are delayed and desynchronized primary peaks [5–8]. Abnormalities of the primary peak are not seen in Kennedy’s disease, which selectively involves spinal motoneurons, implying that the abnormalities seen in ALS are supraspinal in origin [4]. Also, despite abnormal responses to TMS, responses of the same spinal motoneuron to Ia afferent input are normal in ALS patients [9,10]. This too indicates that abnormalities of the primary peak in ALS are caused by corticomotoneuronal dysfunction. In the PSTHs the amplitude of a primary peak changes with the firing rate of a single motor unit [11–13]. This indicates that the primary peak depends on the trajectory of a spinal motoneuron membrane potential. In this study we have compared PSTHs, in normal subjects and ALS using two different methods. Standard (or activated PSTH) and non-activated PSTH. The latter is independent of the

0022-510X / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0022-510X( 00 )00417-2

72

N. Hirota et al. / Journal of the Neurological Sciences 180 (2000) 71 – 75

trajectory of the spinal motoneuron, which, in the absence of firing, remains stable.

2. Materials and methods

lated for each response [17]. The DIF shows subtle changes in the firing probability similar to the cumulative sum (CUSUM) analysis [18] of the PSTHs. The onset latency and duration of the rising phase of the DIF was used to determine the onset and the duration of the primary peaks.

2.1. Patients 2.4. Non-activated PSTHs All patients fulfilled the El Escorial criteria for definite ALS [14]. Subjects voluntarily consented to participate in the experiments, which were approved by University of British Columbia Human Ethics Committee.

2.2. Transcranial magnetic stimulation and recording of single motor unit potentials A Dantec Magpro II magnetic stimulator delivered stimuli to the contralateral motor cortex through a large, cup-shaped round coil (MMC 140) fixed over the head and supported on a stand. The scalp site at which the lowest intensity stimulus capable of inducing a visible muscle contraction of the EDC was located. Stimulus intensity was expressed on a linear scale as a percentage of the maximum output of the device. Subjects sat comfortably with their arms pronated on a pillow. A Dantec Counterpoint EMG machine was used for recording the motor unit potentials (MUPs) with a monopolar needle electrode (Dantec 13R1) inserted into the EDC. The electromyographic response obtained by the Dantec Counterpoint was AD converted to the PC-486 compatible computer by using specific software and data were stored for off-line analysis.

We examined 13 healthy control subjects (seven men), aged 39612 years (mean6S.D.) and 11 patients (nine men) with sporadic ALS, aged 61611 years. This method is essentially similar to the recording of standard PSTHs except that EDC muscle is relaxes. To differentiate this technique from standard PSTHs, we named this method ‘non-activated PSTHs’. The stimulus intensity was adjusted to evoke single MUPs. With the EDC muscle at rest, a series of 100 to 120 stimuli were randomly delivered with an interval of 3 to 5 s. Responses could be obtained in between 10 and 80% of delivered stimuli. The electromyographic activities during 50 ms including 5 ms prior to stimuli were stored with the 50 kHz (20 ms resolution) sampling rate. Single MUPs were verified offline by visual inspection of individual sweeps. The poststimulus latency of 10 or more identical single MUPs was measured. The onset was defined as the shortest latency in each motor unit. The duration was calculated as the difference between the shortest and longest latency in each motor unit. The number of clear subpeaks was counted in the histograms using a bin width of 0.2 ms.

2.3. Standard PSTHs

3. Results

We examined 12 healthy control subjects (eight men), aged 42611 years (mean6S.D.) and 10 patients (five men) with sporadic ALS, aged 6268 years. The method for recording standard PSTHs has been previously reported [15]. Stimulus intensity was set at 2.5% smaller than activated threshold. While the subjects maintained a steady recruitment of single motor unit, series of 100 to 120 magnetic stimuli were randomly delivered at intervals of 1 to 5 s. MUPs occurring between 50 ms prior to stimuli and 200 ms after stimuli, were stored with a 10 kHz (100 ms resolution) sampling rate. Single MUPs were verified off-line by visual inspection. We used the rate of an inhomogeneous Poisson process (RIPP) to identify subpeaks in the primary peak. This neuronal output measure, which estimates the peristimulus spike density, was originally introduced by Awiszus [16]. In this study RIPP density was calculated as the firing probability based on five consecutive single MUPs, which were derived from all stimuli and sorted by each peristimulus time. The displaced impulses function (DIF) was also calcu-

3.1. Standard PSTH We compared 29 motor units from 12 healthy control subjects and 29 motor units from 10 sporadic ALS patients. The data are shown in Table 1. The stimulus intensity was significantly higher in the ALS patients (P,0.001 in Mann–Whitney U-test). There was no significant difference of the onset latency of the primary peak, but the duration was significantly longer in the ALS patients (P50.017 in Mann–Whitney U-test). In healthy control subjects most of the motor units showed single primary peak (Fig. 1A). Double primary peaks were observed in two healthy control subjects and five ALS patients (Fig. 1B). The primary peak was absent in two motor units of one ALS patient, but a clear suppression phase was seen between 50 and 100 ms. Most of the motor units with prolonged duration consisted of double primary peaks (Fig. 2). The number of subpeaks in the primary peak is plotted in Fig. 3. In healthy control subjects the number of subpeaks was up to three (Fig. 1A) except in one motor unit, which showed a double primary peak. In

N. Hirota et al. / Journal of the Neurological Sciences 180 (2000) 71 – 75

73

Table 1 Results of standard PSTHs and non-activated PSTHs a Standard PSTHs

Number of subjects Age (years) Number of motor units Stimulus intensity (%) Onset latency (ms) Duration (ms) a *

Non-activated PSTHs

Control

ALS

Control

ALS

12 42611 29 4666 19.962.4 5.564.5

10 6268 29 5668* 20.263.0 11.167.3*

13 39612 30 54610 17.663.0 1.560.9

11 61611 24 61614* 19.363.0 5.566.3*

Mean values are expressed as mean6S.D. P,0.05 in comparison to controls with Mann–Whitney U-test.

ALS patients, the early component of double primary peaks occurred at around 20 ms and consisted, as in control subjects, of up to three subpeaks (Fig. 1B). The later component of double primary peaks occurred at around 30 ms and consisted of smaller, less clearly discernible subpeaks than the early component.

3.2. Non-activated PSTH We analyzed 30 motor units from 13 healthy subjects and 24 motor units from 11 ALS patients. The results are summarized in Table 1. In healthy control subjects the duration of the poststimulus latency was up to 3.20 ms. The normal upper limit of the duration was 3.30 ms, defined by the mean

Fig. 2. The duration of primary peak in each standard PSTHs is plotted on the time scale. Double primary peaks are expressed as filled circle. The duration is significantly prolonged in ALS patients (P50.0017 in Mann–Whitney U-test). Most of the primary peaks with prolonged duration consist of double primary peaks.

plus two standard deviations. Using a bin width of 0.2 ms, a single peak was seen in 15 motor units (50%), two subpeaks were seen in nine motor units (30%) and three subpeaks were seen in six motor units (20%). In ALS patients the onset latency was prolonged, but the difference between the healthy control subjects and ALS patients was not significant. The duration was significantly

Fig. 1. Example of standard PSTHs in a healthy control subject (A) and an ALS patient (B). PSTHs are shown with 0.2 ms bin width collected over a 50 ms period. Stimuli were applied at time 0 ms. The rate of an inhomogeneous Poisson process (RIPP density) is superimposed on the much smaller routine form of displaying the PSTH. The components of the PSTH and sub-peaks are much more readily appreciated. The displaced impulses function (DIF) is shown for each response. It enables recognition of more subtle changes in the firing probability. In A, three subpeaks are clearly discernible in a single primary peak starting at around 22 ms. In B, the early component of double primary peak occurred at around 20 ms and consisted of two subpeaks. The later component of double primary peak occurred at around 30 ms and consisted of smaller, less clearly discernible subpeaks than the early component, probably reflecting conduction through a slowly conducting pathway.

Fig. 3. Histograms showing the numbers of subpeaks within the primary peaks in standard PSTHs. An increased number of subpeaks occur in ALS patients.

74

N. Hirota et al. / Journal of the Neurological Sciences 180 (2000) 71 – 75

longer in the ALS patients (P50.007 in Mann–Whitney U-test). The duration longer than the normal upper limit was seen in 10 MUPs of six ALS patients. Similar to standard PSTHs the motor units of healthy control subjects showed up to three clear subpeaks. In ALS patients the motor units with normal duration also showed up to three subpeaks. The motor units with prolonged duration had discernible subpeaks at around 20 ms and some of them possibly had increased number of subpeaks. However, the separation between the subpeaks was unclear especially at latencies later than 25 ms after the TMS. Mean stimulus intensity was significantly higher for the motor units of ALS patients than for healthy control subjects (P50.001 in Mann–Whitney U-test).

4. Discussion Standard PSTHs can only be recorded in the clinically less-affected muscles with .grade 4 MRC (Medical Research Council scale) strength. In ALS patients the standard PSTHs showed temporal dispersion of the primary peak, frequently associated with a double primary peak. The early component of double primary peaks had a normal latency and preserved subpeaks. There was decomposition of subpeaks in the later component of double primary peaks. Both, the standard and non-activated PSTHs revealed similar changes. But, the non-activated method can also be used in weak, wasted muscles. In standard PSTHs the spinal motoneuron membrane trajectory varies with the firing rate of the motoneuron. In the non-activated PSTH the spinal motoneuron membrane potential remains stable [19], implying that the abnormalities of the PSTH in ALS are supraspinal in origin. There are two possible explanations for the additional later component. It could be the result of increased repetitive firing of corticomotoneurons evoked by single TMS. A single cortical stimulus evokes a descending volley consisting of a direct (D) wave followed by a series of indirect (I) waves [20]. TMS with the current flowing in an anterio–posterior direction in the coil over the upper limb motor area predominantly provokes I waves [21], but at higher stimulus intensity, as frequently may be necessary in the ALS patients, D waves are also evoked. These have been recorded over the spinal cord during spinal surgery in humans [20]. The total duration of D and I waves is ,10 ms [21–23], whereas the total duration of the primary peak in ALS may far exceed this [4,6,8]. Therefore, excess repetitive firing of the corticomotoneuron cannot be the only explanation for the desynchronized primary peak typical of ALS. An alternative or additional explanation for the desynchronized primary peak and delayed components is activation of slow conducting corticospinal tracts (Fig. 4). The conduction velocities of corticospinal tracts in the cat range from 7 to 70 m / s with two peaks at about 14 m / s

Fig. 4. Hypothetical changes in ALS following preferential involvement of fast conducting corticospinal tract. Under normal conditions the activation of spinal motoneurons by slow conducting corticospinal tracts is masked by the predominant influence of the fast conducting corticospinal tract. As fast conducting corticospinal fibers preferentially degenerate at the early stage of ALS, activation of relatively spared slow conducting corticospinal tracts are unmasked.

and 42 m / s [24], suggesting the existence of a slow and a fast conducting corticospinal tract. The fast conducting corticospinal fibers in humans originate from giant Betz cells. Their axons are of large diameter and make monosynaptic connections with spinal motoneurons [25]. However, the vast majority of corticospinal fibers are slow conducting. In humans there are about one million pyramidal tract axons and only 30,000 giant Betz cells [26]. Ninety per cent of axons are thinner than 4 mm and conduct slowly [26]. To what extent they are direct or indirect in terms of the spinal motoneuron connections are unknown. One example of indirect pathway is the propriospinal tract in the cat. This is a polysynaptic pathway via the propriospinal interneurons located at the upper cervical spinal cord. These propriospinal interneurons receive monosynaptic excitation from corticomotoneurons and make monosynaptic contacts with spinal motoneurons [27]. The existence of a similar pathway has been suggested in humans [28]. Therefore, the most likely explanation for the abnormalities seen in PSTHs of ALS patients is that the phylogenetically new, fast conducting fibers are preferentially involved at the initial stages of ALS [29]. This leaves slower conducting pathways, which are reflected in a desynchronized and / or prolonged primary peak. Under normal conditions the later EPSPs generated by slow conducting corticospinal tracts are probably masked by strong inhibitory postsynaptic potentials following the EPSPs evoked by the fast conducting corticospinal tract [27]. In addition threshold intensity as required for standard PSTHs precludes activation of the high threshold, slow conducting fibers. Higher stimulation recruits additional motor units making analysis of the PSTH impossible. The predominant involvement of the fast conducting corticospinal tract certainly supports the ‘corticomotoneuronal hypothesis’, which postulates that the

N. Hirota et al. / Journal of the Neurological Sciences 180 (2000) 71 – 75

disorder is primarily one of the corticomotoneuron or its presynaptic terminal and secondarily affects lower motor and possibly other neurons [30,31].

References [1] Ingram DA, Swash M. Central motor conduction is abnormal in motor neuron disease. J Neurol Neurosurg Psychiat 1987;50(2):159–66. [2] Eisen A, Shytbel W, Murphy K, Hoirch M. Cortical magnetic stimulation in amyotrophic lateral sclerosis. Muscle Nerve 1990;13(2):146–51. [3] Day BL, Rothwell JC, Thompson PD, Dick JP, Cowan JM, Berardelli A et al. Motor cortex stimulation in intact man. 2. Multiple descending volleys. Brain 1987;110(Pt 5):1191–209. [4] Weber M, Eisen A. Assessment of upper and lower motor neurons in Kennedy’s disease: implications for corticomotoneuronal PSTH studies. Muscle Nerve 1999;22(3):299–306. [5] Awiszus F, Feistner H. Abnormal EPSPs evoked by magnetic brain stimulation in hand muscle motoneurons of patients with amyotrophic lateral sclerosis. Electroencephalogr Clin Neurophysiol 1993;89(6):408–14. [6] Mills KR. Motor neuron disease. Studies of the corticospinal excitation of single motor neurons by magnetic brain stimulation. Brain 1995;118(Pt 4):971–82. [7] Kohara N, Kaji R, Kojima Y, Mills KR, Fujii H, Hamano T et al. Abnormal excitability of the corticospinal pathway in patients with amyotrophic lateral sclerosis: a single motor unit study using transcranial magnetic stimulation. Electroencephalogr Clin Neurophysiol 1996;101(1):32–41. [8] Kohara N, Kaji R, Kojima Y, Kimura J. An electrophysiological study of the corticospinal projections in amyotrophic lateral sclerosis. Clin Neurophysiol 1999;110(6):1123–32. [9] Awiszus F, Feistner H. Comparison of single motor unit responses to transcranial magnetic and peroneal nerve stimulation in the tibialis anterior muscle of patients with amyotrophic lateral sclerosis. Electroencephalogr Clin Neurophysiol 1995;97(2):90–5. [10] Nakajima M, Eisen A, McCarthy R, Olney RK, Aminoff MJ. Reduced corticomotoneuronal excitatory postsynaptic potentials (EPSPs) with normal Ia afferent EPSPs in amyotrophic lateral sclerosis. Neurology 1996;47(6):1555–61. [11] Brouwer B, Ashby P, Midroni G. Excitability of corticospinal neurons during tonic muscle contractions in man. Exp Brain Res 1989;74(3):649–52. [12] Bawa P, Lemon RN. Recruitment of motor units in response to transcranial magnetic stimulation in man. J Physiol (Lond) 1993;471:445–64. [13] Olivier E, Bawa P, Lemon RN. Excitability of human upper limb motoneurones during rhythmic discharge tested with transcranial magnetic stimulation. J Physiol (Lond) 1995;485(Pt 1):257–69. [14] Brooks BR. El Escorial World Federation of Neurology criteria for the diagnosis of amyotrophic lateral sclerosis. Subcommittee on Motor Neuron Diseases /Amyotrophic Lateral Sclerosis of the World Federation of Neurology Research Group on Neuromuscular Dis-

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24] [25]

[26] [27]

[28]

[29]

[30]

[31]

75

eases and the El Escorial ‘Clinical limits of amyotrophic lateral sclerosis’ workshop contributors. J Neurol Sci 1994;124(Suppl):96– 107. Eisen A, Entezari-Taher M, Stewart H. Cortical projections to spinal motoneurons: changes with aging and amyotrophic lateral sclerosis. Neurology 1996;46(5):1396–404. Awiszus F. The RIPP density estimate: an alternative method for the estimation of peri-stimulus spike density. J Neurosci Meth 1992;45(1–2):55–62. Awiszus F. Quantification and statistical verification of neuronal stimulus responses from noisy spike train data. Biol Cybern 1993;68(3):267–74. Ellaway PH. Cumulative sum technique and its application to the analysis of peristimulus time histograms. Electroencephalogr Clin Neurophysiol 1978;45(2):302–4. Jones KE, Bawa P. A comparison of human motoneuron data to simulated data using cat motoneuron models. J Physiol Paris 1999;93(1–2):43–59. Nakamura H, Kitagawa H, Kawaguchi Y, Tsuji H. Intracortical facilitation and inhibition after transcranial magnetic stimulation in conscious humans. J Physiol (Lond) 1997;498(Pt 3):817–23. Nakamura H, Kitagawa H, Kawaguchi Y, Tsuji H. Direct and indirect activation of human corticospinal neurons by transcranial magnetic and electrical stimulation. Neurosci Lett 1996;210(1):45– 8. Kaneko K, Kawai S, Fuchigami Y, Shiraishi G, Ito T. Effect of stimulus intensity and voluntary contraction on corticospinal potentials following transcranial magnetic stimulation. J Neurol Sci 1996;139(1):131–6. Di Lazzaro V, Restuccia D, Oliviero A, Profice P, Ferrara L, Insola A et al. Effects of voluntary contraction on descending volleys evoked by transcranial stimulation in conscious humans. J Physiol (Lond) 1998;508(Pt 2):625–33. Lance JW, Manning RL. Origin of the pyramidal tract in the cat. J Neurophysiol 1954;124:385–99. Phillips CG, Porter R. Origin, course and distribution of pyramidal axons in cats and primates. In: Corticospinal neurones, London: Academic Press, 1977, pp. 93–152. Lassek AM. The pyramidal tract: basic considerations of corticospinal neurons. Res Publs Ass Nerv Ment Dis 1948;27:106–28. Porter R, Lemon R. Corticospinal influences on the spinal cord machinery for movement. In: Cortical function and voluntary movement, London: Clarendon Press, 1993, pp. 122–209. Burke D, Gracies JM, Mazevet D, Meunier S, Pierrot-Deseilligny E. Non-monosynaptic transmission of the cortical command for voluntary movement in man. J Physiol (Lond) 1994;480(Pt 1):191– 202. Kaji R, Kimura J. Amyotrophic lateral sclerosis — a cost of human evolution. In: Kimura J, Kaji R, editors, Physiology of ALS and related diseases, Amsterdam: Elsevier, 1997, pp. 5–8. Eisen A, Kim S, Pant B. Amyotrophic lateral sclerosis (ALS): a phylogenetic disease of the corticomotoneuron? Muscle Nerve 1992;15(2):219–24. Eisen A. Amyotrophic lateral sclerosis is a multifactorial disease. Muscle Nerve 1995;18(7):741–52.