147
ALS and other motor neuron disorders 2001 2, 147–152 © 2001 ALS and other motor neuron disorders. All rights reserved. ISSN 1466-0822
Original Research
Electrophysiologic evaluation of upper motor neuron involvement in amyotrophic lateral sclerosis
Amyotroph Lateral Scler Downloaded from informahealthcare.com by Tel Aviv University on 03/22/14 For personal use only.
Vivian E Drory1,2, Ivan Kovach1 and Galina B Groozman1 Department of Neurology, ALS Clinic EMG Laboratory, 1 Sourasky Medical Center 2 Sackler Faculty of Medicine Tel Aviv, Israel Correspondence: VE Drory, MD Department of Neurology Tel-Aviv Sourasky Medical Center 6 Weizmann Street 64239 Tel-Aviv Israel Tel. and Fax: (00) 972 3 6973689 E-mail:
[email protected] Received 30 April 2001 Revised 19 September 2001 Accepted 23 September 2001
BACKGROUND: The demonstration of upper motor neuron (UMN) dysfunction in patients with amyotrophic lateral sclerosis (ALS) with predominantly lower motor neuron (LMN) signs is clinically sometimes difcult. METHODS: We analyzed the possible role of F and H waves in the diagnosis of UMN dysfunction in 36 patients with different clinical forms of ALS and 20 controls. In both lower limbs we evaluated F-wave persistence and the F/M ratio of tibial nerves, Hmax/Mmax ratio, vibratory and recurrent (paired) inhibition of the Hreex. RESULTS: The persistence of the Fresponse was decreased signicantly in
the LMN group only. The F/M amplitude ratio, vibrated-H/rest-H amplitude ratio and conditioned H/rest H amplitude ratio were signicantly increased in patients with predominantly UMN, as well as in those with predominantly LMN signs. H/M amplitude ratios did not differ between the various patient groups. CONCLUSION: These ndings show that ALS patients, including those without clinical evidence of UMN involvement, have a marked disinhibition of anterior horn motor neurons. The simple tests described could support an UMN abnormality when clinical signs are lacking, and help to establish a diagnosis sooner and more accurately. (ALS 2001; 2: 147–152)
Keywords: amyotrophic lateral sclerosis – motor neuron disease – upper motor neuron – electrophysiology – F-response – H-reex
Introduction The hallmark of amyotrophic lateral sclerosis (ALS) is progressive degeneration of the motor system, involving both lower motor neurons (LMN), and upper motor neurons (UMN). During the early course of the disease, patients may lack overt clinical signs of an UMN lesion. In the later stages, severe LMN involvement may mask previously existent UMN signs. Early and accurate diagnosis of the disease is mandatory for the proper management of ALS patients. The World Federation of Neurology (WFN) agreed several diagnostic degrees of certainty of the disease, ranging from suspected to denite ALS, based on the clinical presentation, electrophysiologic and imaging studies.1 Patients with UMN signs alone are classied as “possible ALS”, while those with LMN signs in more than two regions but without UMN signs are considered as “suspected ALS”, and the combination of both is required for a “probable ALS” diagnosis. Patients with a pure UMN syndrome can be evaluated for a subclinical LMN lesion by electromyography (EMG). However, for patients with a clinically pure LMN syn-
drome there is no available diagnostic procedure for demonstrating a latent UMN decit. Magnetic resonance spectroscopy of the motor cortex,2,3 proton emission tomography,4 and, more recently, diffusion tensor magnetic resonance imaging (MRI),5 have been suggested for this purpose, as ALS patients frequently show abnormal results, but these techniques usually give abnormal results in patients with clinically obvious UMN signs and do not reveal subclinical corticospinal tract involvement in patients with predominant LMN involvement. Central motor conduction time measurements following transcranial magnetic stimulation demonstrate abnormalities in most “denite ALS” patients,6 but the sensitivity of the test reached only 50–71% in patients without overt clinical UMN involvement.7,8 Various other electrophysiologic tests for the analysis of UMN lesions have been described, some of them rather simple, others very sophisticated. Among the tests suggested, F-waves, Hreexes and their variations have been used.9–15 These tests are easy to perform, are not time-consuming and can be included in the routine electrophysiologic examination of suspected ALS patients. The F-wave is a response of part of the motor neuron
148
VE Drory, I Kovach, GB Groozman
Amyotroph Lateral Scler Downloaded from informahealthcare.com by Tel Aviv University on 03/22/14 For personal use only.
Original Research pool following antidromic supramaximal stimulation of a peripheral nerve. Various variables of this response, such as minimal F-wave latency, F/M amplitude ratio and persistence of F-waves with serial stimulation, have been described as indicators of UMN dysfunction.9-11 The H-reex represents a monosynaptic response from a muscle following electrical stimulation of Ia spindle afferents in a mixed nerve. It is most reliably recorded in adults from the soleus muscle. The H/M ratio depends on the number of excitable motor neurons in the spinal cord pool, which are activated by the Ia pathway. It tends to be increased in patients with UMN lesions.12–14 Variations of the technique have been recommended, using different facilitatory or inhibitory conditioning stimuli, in order to assess the state of excitability of spinal motor neurons in pyramidal lesions.15 In normal subjects, vibratory stimulation of the soleus muscle or its tendon, occurring at the same time as electrical stimulation during the recording of the H-reex, results in inhibition of impulses from spindle afferents and reduction of the amplitude of the H-response.16,17 It has been shown that this effect reects presynaptic inhibition of Ia afferents and is evoked by other Ia bers coming from the same muscle.18 In pathologic excitatory states the amplitude reduction of the H-reex during vibratory stimulation is less evident, probably due to reduced GABA-ergic activation.12,13 Paired stimulation of the tibial nerve can be used to activate Renshaw cells, causing partial inhibition of the Hreex and a decrease in its amplitude.19 In spastic states, in which the modulation of spinal motor neuron activity by upper structures is lacking, Renshaw-cell-induced (recurrent) inhibition is decreased and the amplitude of the test H-response is less diminished.20,21 The purpose of the present study was to investigate the usefulness of F-waves and H-reex studies in the evaluation of UMN dysfunction in patients with ALS. We hoped this might establish a reliable marker for UMN involvement.
Patients and methods We examined 36 consecutive patients with ALS attending our outpatient clinic (19 men, 17 women, ages 44–79, mean 60 years) and 20 normal volunteers (11 men, 9 women, ages 42–68, mean 53 years). Among the ALS patients, 16 were classied as “denite”, 11 as “probable” and nine as “possible” ALS by the new WFN criteria.1 The mean duration of disease in these patients was 19 ± 7 months. Normal volunteers were recruited among the department staff and patients’ relatives. Patients and controls gave their informed consent to the examination and the study protocol was approved by the local institutional ethics committee. Patients and controls were excluded from the study if they had any other disease which could affect the peripheral nervous system, such as diabetes, alcoholism, nutritional deciencies, etc. Patients on antispasticity drugs
were excluded from the study. Controls were excluded if their neurological examination revealed any pathological signs. All subjects were examined clinically by observation for atrophy and fasciculation, manual muscle testing, examination of tendon reexes and plantar responses. Muscle tone was assessed by the Ashworth score.22 Patients were also assessed for motor and sensory distal latencies, motor nerve conduction velocities, compound muscle action potential (CMAP) and sensory nerve action potential (SNAP) amplitude in at least one upper and one lower limb nerve. EMG was performed according to the clinical presentation in at least two muscles innervated by different roots and peripheral nerves in each limb. As one of the objectives of the study was to examine Hreexes, which are more easily and consistently elicited in the lower than in the upper limbs, especially in the presence of muscle wasting, only lower limb data are included in the present study. A lower limb was dened as showing a predominantly UMN syndrome if there was no marked atrophy, tendon reexes were present or brisk, the plantar response was indifferent or extensor, the Ashworth score was at least 1 and no spontaneous activity or high-amplitude wide motor units were seen on EMG. Limbs were dened as having a predominantly LMN syndrome if there was muscle atrophy, if fasciculation was noted, if at least one tendon reex was absent, if the plantar response was exor or unobtainable, the Ashworth score was 0 and the EMG showed denervation in at least two muscles (brillations and/or positive sharp waves at a minimum of two sites of the muscle out of at least ten sites sampled). Limbs were considered as normal if no signicant neurological abnormality was found on clinical examination or EMG (e.g. in patients with bulbar involvement or decit in the arms or opposite leg only). Right and left legs could be assigned to different groups in the same patient, due to the asymmetry often seen in ALS. Twenty-one limbs were included in the UMN group, 29 limbs in the LMN group, and 22 limbs were considered normal (N). The 40 limbs of the healthy subjects formed the control group (C). All electrophysiologic tests were performed on a Nicolet Viking IV electrodiagnostic system. Patients were examined supine in a quiet room and as relaxed as possible. Limb skin temperature was maintained at at least 30°C. For F-wave recording, a bipolar surface electrode was placed over the abductor hallucis muscle on the medial aspect of the foot. Twenty supramaximal square wave pulses of 0.1 ms duration, at a frequency of 0.5 Hz (cathode proximal), were applied to the tibial nerve in the medial malleolar region. Filter settings were 2 Hz–5 kHz, amplier gain was 0.1 mV for F and 1 mV for M responses. F-waves were considered present if their peak-to-peak amplitude was at least 20 µV and the waves were of variable shape and latency. Maximal peak-to-peak amplitude and persistence of the waves (i.e. number of measurable responses divided by the number of stimuli) were determined and the maximal F-amplitude/M-amplitude ratio (F/M ratio) was calculated.
149
F and H waves in amyotrophic lateral sclerosis
For H-reex recordings, patients lay with a exed hip joint and the knee of the examined leg exed to form a 120° angle, and supported by a triangle-shaped cushion. Patients were instructed to relax the leg, in order to minimize voluntary muscle activity, which would increase the variability of the response.23 The H-reex was obtained with recording electrodes placed over the soleus muscle and stimulation of the tibial nerve in the popliteal fossa about 15 cm proximally. The stimulation was started at low voltage with no Mor H-response and gradually raised by 10 V steps until a maximal H-response was elicited, then raised further until a maximal M-response was obtained. Stimulus duration was kept constant at 1 ms; its frequency was variable, but never exceeded 0.1 Hz. Filter settings were 2 Hz–5 kHz, amplier gain was set at 0.5 mV for H- and 2 mV for M responses. Surface silver/silver chloride disc electrodes were used, with a diameter of 10 mm, placed 2 cm apart on the lower part of the soleus muscle belly. H- and Mresponse amplitudes were measured from baseline to the negative peak. The H-response was considered present if its amplitude reached at least 20 mV. We measured the amplitude of the largest response and calculated the maximal Hamplitude/M-amplitude ratio (H/M ratio). Vibratory stimulation was performed in the same recording setting as for the standard H-reex examination, adding a continuous vibratory stimulus, produced by an eccentric DC 51motor taped over the triceps surae tendon, which elicited vibratory waves with a frequency of 100 Hz and an amplitude of 1 mm. Measurements were similar to the standard H-response recording. The vibrated-maximal H-amplitude/maximal H-amplitude ratio (vibH/H ratio) was calculated. For recurrent inhibition studies we used the technique described by Katz and Pierrot-Deseilligny,21 which is based on paired stimulation of the tibial nerve in the popliteal fossa. In this recording setting the rst (conditioning) stimulus is of increasing intensity, while the second (test) stimulus is constant and supramaximal, with an interstimulus interval of 10 ms. The rst stimulus produces an H-reex with low stimulus intensities; with higher intensities an M-response also. The second stimulus gives rise to a second H, H’. At low stimulus intensities, H’ is equal in amplitude to H, but at higher intensities, as H increases, H’ decreases progressively. This decrease of H’ amplitude indicates recurrent inhibition of motor neurons, elicited by the conditioning stimulus. The maximal H’ amplitude was measured and an H’/H amplitude ratio calculated. The elicitability of F- and H-responses in the different limb groups was compared by means of x 2 tests. Comparisons between the F persistence, F/M ratio, H/M ratio, vibH/H ratio and H’/H ratio in the different limb groups (UMN, LMN, N and C) were performed using univariate analysis of variance and Bonferroni post-hoc analysis with a level of signicance of
