Multimodality Evoked Potentials
in
Motor Neuron Disease Janice Shanthi Subramaniam, BSc(Hon), Con Yiannikas, MD, FRACP
\s=b\ We
performed median and tibial nerve
somatosensory evoked potentials (SEPs), pattern-shift visual evoked potentials (PSVEPs), and brain-stem auditory evoked potentials (BAEPs) on 27 patients with motor neuron disease (MND). Median and tibial nerve SEPs were abnormal in 8(30%) of 27 and 3 (14%) of 21 patients tested, respectively. Central and peripheral abnormalities were recorded in the absence of spondylosis. As a group, patients with MND and no evidence of cervical spondylosis had normal conduction to Erb's point following median nerve stimulation, but conduction times beyond this point were prolonged. The PSVEPs and BAEPs were within normal limits in all patients, excluding abnormalities attributable to other disease, but the group P100 latency was significantly prolonged in the group with MND. The BAEPs were normal in the group with MND. This study provides neurophysiological evidence of sensory system involvement in MND. (Arch Neurol. 1990;47:989-994)
/Totor neuron disease (MND) is clas-
sically described as a motor sys¬ involving progressive de¬ generation of the upper (corticospinal and corticobulbar tracts) and/or lower motor neurons. Clinically, it is mani¬ fested by weakness and amyotrophy with or without spasticity and hyper¬ reflexia. Limb, trunk, or bulbar mus¬ culature is affected. Clinically, sensory involvement has been thought not to occur, and the demonstration of sig¬ tem disorder
nificant sensory abnormalities on neu-
Accepted for publication January 29, 1990. From the Department of Clinical Neurophysiology, Westmead Hospital, New South Wales, Australia.
Reprint requests to Department of Clinical Neurophysiology, Westmead Hospital, West-
mead, New South Wales 2145, Australia (Dr Yiannikas).
rological or electrophysiological exam¬
ination has been grounds to reformu¬ late the diagnosis. However, a growing body of pathological18 and clinical810 evidence suggests that sensory path¬ ways may be involved in MND. Evoked potentials (EPs) provide an objective, noninvasive test of function in selected sensory pathways and have been used by several laboratories to investigate sensory involvement in MND.11 " To date, however, reports are conflicting and are confounded by in¬ adequate age matching of control data, insufficient screening of patients for additional disease, and methodological problems that make it impossible to localize abnormalities to the central nervous system or the periphery. In this study, we report the results of EP studies (median and tibial nerve somatosensory EPs [SEPs], patternshift visual EPs [PSVEPs], and brainstem auditory EPs [BAEPs]) per¬ formed on patients with different forms and varying degrees of MND. The patient group was carefully screened for additional disease and objectively assessed on a disability scale. Control data were age matched. PATIENTS AND METHODS
Twenty-seven patients (19 male, 8 fe¬ male) were diagnosed as having MND by a
neurologist based on the clinical and electromyographic findings and in the absence of alternate causes on radiologie investiga¬ tion. Patients were classified as having pseudobulbar palsy (exclusively upper mo¬ tor neuron signs), primary muscular atro¬ phy (PMA; exclusively lower motor neuron signs), or amyotrophic lateral sclerosis (ALS; both lower motor neuron and pyra¬ midal signs, with or without bulbar signs). Disease severity was assessed on a disabil¬ ity scale24 based on bulbar function, limb function, and patient independence. Low scores indicate severe disability; a maxi¬ mum score
of 25 indicates normal function.
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
Myelograms were performed to exclude other causes of upper motor neuron disease, and sensory changes were assessed as part of the neurological examination. Nerve con¬ duction studies and electromyography were performed. Disease severity and disease type were correlated with central conduc¬ tion time to median nerve stimulation for all patients with MND without concurrent cervical spondylosis. We also assessed the correlation among abnormalities of periph¬ eral nerve conduction, clinical sensory ex¬ amination, and EP abnormalities. Somatosensory EPs were elicited by uni¬ lateral percutaneous stimulation (0.2-millisecond square-wave pulse) at 2 Hz. Me¬ dian nerve SEPs were elicited by stimula¬ tion of the median nerve at the wrist at an intensity just sufficient to produce a small visible movement of the thumb. Tibial nerve SEPs were elicited by stimulation of the tibial nerve at the ankle at an intensity just sufficient to produce a slight plantar flexion of the foot. We recorded 256 or 512 responses to median and tibial nerve stim¬ ulation using 52- and 100-millisecond poststimulus sweeps, respectively. Recording band passes for subcortical and cortical re¬ sponses were 16 Hz to 3 kHz and 1.6 Hz to 3 kHz (—3 dB points), respectively. In median nerve SEPs, active electrodes over Erb's point, C-2, and the contralateral scalp 2 cm posterior to C-3 and C-4 (C3 and C4, respectively) were referred to P;. As¬ sessment was based on the absolute laten¬ cies of the Erb's point potential (N9), the cervicomedullary potential (P/N13), and the onset and peak of the thalamocortical potential (N190„ and N19pk, respectively). In addition, the N9-P/N13, P/N13-N19, N9N19, and N19on-N19pl interpeak latencies were measured, the latter to assess thalamocortical conduction. In tibial nerve SEPs, active electrodes at the level of the first lumbar vertebra and at C, were refer¬ enced to the iliac crest and F2, respectively. Assessment was based on the absolute la¬ tency of the spinal (N20) and cortical (N/ P37) potentials and the N20-N/P37 inter¬
peak latency.
We elicited PSVEPs by an alternating (2-Hz) black-and-white checkerboard pat¬ tern, each check subtending an angle of 50'
at the retina. The stimulus was
monocularly using
rotating
presented
mirror pat¬ tern-shift stimulator. Full-field stimula¬ tion was routinely used (angle subtended by full field at the retina was 23.8° horizon¬ tally, 21.2° vertically). Half-field stimula¬ tion was used when necessary to clarify waveforms. We averaged 256 or 512 re¬ sponses
over a
a
250-millisecond poststimu-
lus period (band width, points). Five-channel PSVEPs were re¬ corded using a midline electrode 5 cm above the inion and electrodes spaced 5 and 10 cm laterally on each side, referenced to Fs. Re¬ sponses were assessed on the latency of the first major positive component, P100. We elicited BAEPs by square-wave click stimuli (0.1 millisecond, 9 Hz) delivered through headphones. Rarefaction clicks were routinely used, but occasionally an al¬ ternating (condensation/rarefaction) stim¬ ulus was presented to clarify the position of wave I. Monaural click stimuli (70 dB above click threshold for each ear) and a con¬ tralateral masking white noise (40 dB above noise threshold for each ear) were pre¬ sented stimultaneously. We averaged 1024 or 2048 responses over a 10-millisecond poststimulus sweep (band width, 53.2 Hz to 3 kHz, -3 dB points). Two-channel BAEPs were recorded with active electrodes on the earlobes referenced to the vertex and were assessed on the latency of waves I, III, and V and the I-III, III-V, and I-V interpeak la¬ tencies. The EPs were routinely recorded using silver-silver chloride cup electrodes (im¬ pedance .l). Group comparisons of PSVEPs in 24 patients with MND and 24 age-matched controls (Table 4) showed a significant prolongation of the P100 latency at a level of signifi¬ cance of .01 ( .001). =
=
=
=
COMMENT
The demonstration of subjective and objective abnormalities in a small pro¬ portion of patients with MND810 sug-
Fig 1.—Median nerve somatosensory evoked potentials from four patients with sporadic mo¬ tor neuron disease and no other significant disease. Trials were replicated and superim¬ posed to ensure reproducibility in all patients. Center top and center bottom show the aver¬ age of the replicated trials. Top, Normal N9 and P/N13 absolute latencies but bilaterally pro¬ longed P/N13-N19conductionare shown (left interpeak latency [IPL], 7.9 ms; right IPL, 8.2 ms; mean age-matched control IPL + 2.5 SD, 7.0 ms). EP indicates Erb's point. Center top,
Normal N9 and P/N13 absolute latencies but bilaterally prolonged P/N13-N19 conduction (left IPL, 9.8 ms; right IPL, 10.7 ms; mean agematched control IPL + 2.5 SD, 7.0 ms) and bi¬ laterally prolonged 19on-N 19pk are shown (left IPL, 6.9 ms; right IPL, 6.3 ms; mean agematched control IPL + 2.5 SD, 5.0 ms). Center bottom, Normal N9 absolute latencies but pro¬ longed N9-P/N13 conduction on the left are shown (left IPL, 5.1 ms; mean age-matched control IPL + 2.5 SD, 4.8 ms). N9-P/N13 is at the upper limit of normal on the right (IPL, 4.8 ms). Bottom, Bilaterally delayed N9 potentials (left N9, 12.4 ms; right N9, 12.1 ms; mean age-matched control N9 + 2.5 SD, 10.4 ms).
gests that degeneration
Left
EP-F,
|2µ
CVF,
|2 ßV
|2µ
|4MV
\4ßV IPL + 2 SD O
IPL + 2.5 SD
5 10 15 20 25 30 35 40 45
Latency,
O
5 10 15 20 25 30 35 40 45
Latency,
ms
Left
|2MV
CVF,
|2µ
CVF,
|2mV
1.5
EP-F,
O
10
|1.S/»V
CVF,
2mV IPL + 2.5 SD
30
20
Latency,
nation results and abnormal sensory nerve conduction study results not as¬ sociated with any disease other than
µ
CVF,
IPL + 2.5 SD
pathways in this disease. In this study, two patients presented with subjective and objective sensory defi¬ cits (abnormal clinical sensory exami¬
AD
ms
Right
EP-F,
may involve
There is also convincing pathologi¬ cal evidence of sensory fiber involve¬ ment at a peripheral and central level. Autopsy studies of patients with ALS have demonstrated spinal ganglia involvement8 and marked pallor of the posterior columns, particularly in the fasciculus gracilis.3·8 A reduction in the neuronal population of Clarke's nucleus1 and a decrease in choline acetyltransferase level in the dorsal horns and in the region of Clarke's nu¬ cleus in patients with ALS has also been noted.5 Morphometric studies of the spinal cord at the level of L-57 demonstrated a 54% decrease in the number of large afferent neurons in the dorsal root ganglia and a 27% re¬ duction in the number of large myelinated fibers in the dorsal root. Other studies have demonstrated consistent segmental demyelination and remyelination in peripheral sensory nerves4 and a significant reduction in the den¬ sity of myelinated fibers in the sural nerve.2 Thus, there is convincing pathological and clinical evidence of sensory fiber involvement at central and peripheral levels. Hudson6 found the incidence of pos¬ terior column involvement to be less than 10% in sporadic ALS but as high
|2µ
EP-F
CVF,
sensory
MND).
Right
40
50
O
10
ms
20
Latency,
Left
40
30
50
SF
ms
Right
EP-F,
CVF,
1
µ\
EP-F,
1
ßV
CVF,
1
µ
µ
IPL + 2.5 SD N19
CVF,
1
10
20
30
Latency,
mV
C3-F,
1
40
10
ms
20
30
Latency,
Left
µ
40
ms
Right
EP-F,
EP-F,
CVF,
CVF,
C.-F,
CVF,
|1µ O
5 10 15 20 25 30 35 40 45
Latency,
ms
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
O
5 10 15 20 25 30 35 40 45
Latency,
ms
KL
as 70% in familial disease. Recently, it has been suggested25 that the incidence of familial MND may be frequently underestimated because of difficulty in obtaining sufficient comprehensive in¬ formation and that it may account for at least 10% of cases. Given the prom¬ inence of sensory disease in familial MND, one would predict a fairly high incidence of sensory disease in the population with MND at large. Such disease may well underlie the seem¬ ingly sporadic occurrences of sensory abnormalities in patients with MND. Several EP studies have investi¬ gated sensory tract involvement in MND, but to date most reports are equivocal. The early data of Dustman et al17 suggesting median nerve SEP abnormalities in patients with ALS were incomplete and not statistically analyzed. Most reports since then have been confounded by methodological problems. A number of studies report¬ ing median nerve SEP abnormalities fail to exclude or at least identify pa¬ tients with cervical spondylosis."·'6·"'23 The fact that all the patients of Matheson et al19 with abnormalities of the cortical potential also had P/N13 ab¬ normalities suggests that cervical my-
elopathy secondary to spondylosis may well have been a complicating factor in this study. Other studies1219 fail to lo¬
calize SEP abnormalities to the pe¬ riphery or the central nervous system, while still other reports of normal13·14·20 and abnormal21 median nerve SEPs in MND do not specify adequate age matching of patient and control data. Abnormal responses to median nerve stimulation were recorded in 11 of our 27 patients with MND compared with an age-matched control group. In 8 cases with no myélographie evidence of spondylosis, the abnormalities con¬ sisted of delayed N9 potentials, pro¬ longed N9-P/N13, P/N13-N19, and N9-N19 conduction times, as well as dispersion of the thalamocortical po¬ tential. In a further three patients with myélographie evidence of cervical spondylosis, the findings included pro¬ longed N9-P/N13 and N9-N19 conduc¬ tion times and normal central conduc¬ tion times, consistent with typical ab¬ normalities in cervical myelopathy.26 The findings that N9-P/N13 conduc¬ tion can be prolonged in cervical spondylosis and pure MND suggests that abnormalities between the bra¬ chial plexus and the lower medulla cannot be used in differentiating these disorders. However, an increase in cen¬ tral conduction time in the presence of normal N9-P/N13 conduction argues against spondylitic myelopathy as the cause.
Spondylosis
severe
enough
to
Right
|2MV
|4µ
CVF, C.-F,
12 µ
5 10 15 20 25 30 35 40 45
Latency,
O
5
10 15 20 25 30 35 40 45
Latency,
ms
VF
ms
nerve somatosensory evoked potentials from a patient with sporadic amyotrophic lateral sclerosis and concurrent C4-5 canal stenosis and C4-7 spondylosis. P/N13 potentials were absent bilaterally. N9-N19 conduction was at the upper limit of normal on the left and pro¬ longed on the right (left interpeak latency [IPL], 11.3 ms; right IPL, 12.3 ms; mean age-matched control IPL + 2.5 SE, 11.3 ms). EP indicates Erb's point.
Fig 2.—Median
Table
2.—Group Comparisons
in Evoked Potentials* MND
Potential Median nerve SEP No. of arms
Age,
57.2 ± 15.4
y
Latencies, ms N9 N9-P/N13
56.8 ± 15.0
9.8 ± 1.2 3.7 ± 0.6
P/N13-N19
.030 .008
3.4 ± 0.7
10.7 ± 1.2
.0001
61.4 ± 9.62
.50
N19„-N19Dh BAEP No. of
40
ears
Age, y Latencies,
ms
I
l-V
1.52 ± 0.21
1.60 ± 0.23
2.17 ± 0.22
1.95 ± 0.22
2.24 ± 0.22 1.90 ± 0.19
4.12 ± 0.25
4.14 ± 0.25
.10
PSVEP No. of eyes
Age,
62.6 ± 11.1
59.5 ± 12.3
y
100 latency,
ms
98.5 ± 6.6
102.7 ± 5.9
.0013
MND indicates motor neuron disease; SEP, somatosensory evoked potential; BAEP, brain-stem auditory evoked potential; and PSVEP, pattern-shift visual evoked potential. Values are mean ± SD. For SEPs, group data were from 11 patients with MND with normal cervical myelograms and 11 age-matched control subjects; for BAEP, group data were from 20 patients with MND with normal auditory thresholds and 20 age-matched con¬ trol subjects; and for PSVEP, group data were from 24 patients with MND with normal visual acuity and 24 age-
matched control subjects.
a central conduction delay would also severely disrupt N9-PN13 conduction.26 Our data do not support the suggestion made by Oh et al20 and Cascino et al13 that the presence of any SEP abnormality is useful in differen¬ tiating cervical spondylitic myelopa¬ thy from MND. However, it is an interesting observation that none of our group with pure MND had grossly dispersed or absent P/N13 potentials,
produce
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
whereas patients in our spondylitic group did. Perhaps the finding of se¬ verely disrupted N9-P/N13 conduction or absent P/N13 potentials is grounds to look for alternate or concurrent dis¬ ease.
Comparison of a group of patients with pure MND (normal cervical mye¬ lograms) with an age-matched control group showed no difference in periph¬ eral conduction (normal N9 latency)
but a prolongation of N9-P/N13 and P/N13-N19 conduction in the group with MND, with central conduction being more significantly affected. Given that the N9-P/N13 conduction was within normal limits in the ma¬ jority of individual studies, the group abnormalities may suggest a mild in¬ volvement of the large-diameter pe¬ ripheral sensory fibers in some pa¬ tients with MND. Central conduction may be more severely and/or fre¬
Right
C,-F
µ
1
C.-F,
µ
/ 37 N/P37 + 2.5 SD
N/P37 + 2.5 SD
quently affected, implying a more sig¬
nificant involvement of sensory tracts. Studies to date of tibial nerve SEPs in patients with MND also have meth¬ odological limitations. Abnormalities in the latency of the cortical potential and central conduction time have been reported,16·21·27 but coexisting spondylo¬ sis was not excluded by full myelogram in these studies. Other studies1321 do not adequately age match their patient and control groups. The report of Matheson et al19 of central conduction defects in 13 of their 32 patients is complicated by the L-l potential not having been recorded. They define a central conduction defect as an abnor¬ mal cortical potential in the setting of normal -reflexes. However, the SEPs were elicited by stimulation at the an¬ kle, while the -reflexes were elicited by stimulation at the popliteal fossa, leaving a significant segment of pe¬ ripheral sensory nerve unassessed by the -reflex. In addition, the H-reflex elicited at the popliteal fossa assesses a single cord segment (S-l), while stimulation of the tibial nerve at the ankle involves several segments (L-5S-2). We report abnormal spinal and/ or cortical responses to tibial nerve stimulation (in the setting of normal full spinal myelograms and normal nerve conduction study results) in 3 of 21 patients studied. In 2 of these, the conduction deficit may have been pe¬ ripheral; in the third, there was an un¬ equivocal central conduction deficit. Brain-stem auditory EPs have been reported as normal in a total of 30 patients.15·22-24 Two studies,19·21 in which control data were not adequately age matched, have reported BAEP abnor¬ malities. Our study, using agematched controls, found no BAEP ab¬ normalities attributable to MND. In particular, those patients with promi¬ nent oropharyngeal symptoms had normal BAEPs. Group statistics showed no BAEP differences between the group with MND and the control group, suggesting that degeneration may not affect this system. To date, PSVEPs have been normal in most MND studies.14·21·28 Matheson et al19 reported minor prolongation of
|2µ
|2µ
O
O 10 20 30 40 50 60 70 80 90
10 20 30 40 50 60 70 80 90
Latency,
Latency,
ms
EB
ms
Right
Left
C,-F,
|2µ
L,-it
|2µ
h µ
c-F,
|1 µ
N/P37 IPL + 2.5 SD O 10 20 30 40 50 60 70 80 90
O 10 20 30 40 50 60 70 80 90
Latency,
Latency,
ms
MC
ms
nerve somatosensory evoked potentials from two patients with motor neu¬ disease and no other concurrent disease. Top, Bilaterally absent N20 potentials and bilater¬ ally delayed N/P37 responses are shown (left latency, 52.4 ms; right latency, 50.4 ms; mean age-matched control latency + 2.5 SE, 50.0 ms). Bottom, Normal N20 absolute latency but bilat¬ erally dispersed /P37 potentials are shown. /P37 cannot be confidently identified on the right.
Fig 3.—Posterior tibial ron
0
2
4
6
Severe Disability
Newrick and Langton-Hewer Disability Score
Fig 4.—Plot of central conduction time (P/N13-N19) against Newrick and Langton-Hewer24 as¬ sessment of bulbar function, limb function, and patient independence in motor neuron disease (re¬ gression line and equation for data set are shown).
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
PlOO in 4 of 32 patients in a non-agematched study. In view of the signifi¬ cant increase in P100 latency after the fifth decade of life14 and the tendency of patients with MND to be in this older age group, age matching of control data is essential. No PSVEP abnor¬ malities were noted in individual pa¬ tients in our study using age-matched controls, but the average P100 latency for the patient group was significantly delayed compared with the control group (Table 4). Laxer and Spence18 reported a similar finding in a group of five patients, but Cascino and colleagues13 found no group PSVEP abnormalities in their patient group in a
non-age-matched study.
Previous studies19·21 have anecdot-
ally reported no relationship between disease severity and the presence of EP
abnormalities. Our data also suggest this, but the distribution of disability among our patient population too uneven to allow any conclu¬
scores was
sions. There
was no
tween clinical and EP
correlation be¬
abnormalities,
and clinical sensory deficits without evidence of alternate disease were rare and much less frequent than EP evi¬ dence of sensory involvement. This is consistent with the observation of others21 that SEPs may be a more sen¬ sitive index of sensory system involve¬ ment in MND than clinical examina¬ tion. CONCLUSION
This study reports SEP abnormali¬ ties in patients with MND with no al¬ ternative disease to explain the find¬ ings. The most common finding was prolongation of the central conduction time to median nerve stimulation, but prolonged peripheral, N9-P/N13, and N19on-N19pk conduction times were also noted. Since patients with cervical spondylosis also demonstrate abnor¬ mal N9-P/N13 and N9-N19 conduc¬ tion, disrupted conduction between the brachial plexus and the lower medulla is not a useful criterion for differenti¬ ating between MND and cervical spondylitic myelopathy. Increased
central conduction time in the pres¬ ence of normal N9-P/N13 conduction argues against spondylitic myelopathy
as the cause of the abnormality. Pe¬ ripheral and central conduction abnor¬
malities were recorded in the tibial SEPs. Occasional clinical sen¬ sory deficits were noted, but the fre¬ quency of these abnormalities com¬ pared with the frequency of SEP ab¬ normalities suggests that SEPs may be a more sensitive index of sensory disease in central pathways in MND than sensory examination. A small proportion of patients also had abnor¬ mal sensory nerve conduction study results, suggesting that peripheral fi¬ bers may also be involved in this dis¬ ease. The P100 latency in the patient group was significantly delayed, al¬ though the P100 latency of all individ¬ ual patients was within normal limits, suggesting that conduction in the vi¬ sual pathway is affected less severely than conduction in the large-fiber sen¬ sory system. We found no individual or group BAEP abnormalities. nerve
References 1. Averback P, Crocker P. Regular involvement of Clarke's nucleus in sporadic amyotrophic lateral sclerosis. Arch Neurol. 1982;39:155-156. 2. Bradley WG, Good P, Rasool CG, Adelman LS. Morphometric and biochemical studies of peripheral nerves in amyotrophic lateral sclerosis. Ann Neurol. 1983;14:267-277. 3. Castaigne P, Cambier J, Escourolle R, Brunet P. Sclerose laterale amyotrophique et lesions degeneratives des cordons posterieurs. J Neurol Sci. 1971;13:125-135. 4. Dayan AD, Graveson GS, Illis LS, Robinson PK. Schwann cell damage in motor neurone dis-
Neurology. 1969;19:242-246. Gillberg PG, Aquilonius SM, Eckernas S, Lundquist G, Winblad B. Choline acetyltransferase and substance P-like immunoreactivity in the human spinal cord: changes in ALS. Brain Res. 1982;250:394-397. 6. Hudson AJ. Amyotrophic lateral sclerosis and its association with dementia, Parkinsonism and other neurological disorders: a review. Brain. 1981;104:217-247. 7. Kawamura Y, Dyck PF, Shimono M, Okazaki H, Tateishi J, Doi H. Morphometric comparison of the vulnerability of peripheral motor and sensory neurons in amyotrophic lateral sclerosis. J Neuropathol Exp Neurol. 1981;40:675-687. 8. Luthy F, Martin F. Observation de cas de sclerose laterale amyotrophique avec consideration peculiere sur la participation sensitive. ease.
5.
Schweiz Med Wochenschr. 1947;25:694-697. 9. Dyck PJ, Stevens JC, Mulder DW, Espinosa RE. Frequency of nerve fibre degeneration of peripheral motor and sensory neurons in amyotrophic lateral sclerosis. Neurology. 1975;25:781\x=req-\
785. 10. Mulder DW, Bushek W, Spring E, Karnes J, Dyck PJ. Motor neuron disease (ALS): evaluation of detection thresholds of cutaneous sensation.
Neurology. 1983;33:1625-1627. 11. Anziska BJ, Cracco RQ. Short-latency somatosensory evoked potentials to median nerve stimulation in patients with diffuse neurologic disease. Neurology. 1983;33:989-993. 12. Bosch EP, Yamada T, Kimura J. Somatosensory evoked potentials in motor neuron disease. Muscle Nerve. 1985;8:556-562. 13. Cascino GD, Ring SR, King PJL, Brown RH, Chiappa KH. Evoked potentials in motor system diseases. Neurology. 1988;38:231-238. 14. Chiappa KH. Evoked Potentials in Clinical Medicine. New York, NY: Raven Press; 1983. 15. Chiappa KH, Harrison JL, Brooks EB, Young RR. Brainstem auditory evoked responses in 200 patients with multiple sclerosis. Ann Neurol. 1980;7:135-143. 16. Cosi V, Poloni M, Mazzini L, Callieco R. Somatosensory evoked potentials in amyotrophic lateral sclerosis. J Neurol Neurosurg Neuropsy-
chiatry. 1984;47:857-861.
17. Dustman RE, Snyder EW, Callner DA, Beck EC. The evoked potential as a measure of cerebral dysfunction. In: Begleiter H, ed. Evoked Brain Potentials and Behaviour. New York, NY: Plenum Press; 1979:321-363. 18. Laxer KD, Spence B. Visual evoked responses in amyotrophic lateral sclerosis. Electroencephalogr Clin Neurophysiol. 1980;50:168P. 19. Matheson JK, Harrington HJ, Hallett M. Abnormalities of multimodality evoked potentials in amyotrophic lateral sclerosis. Arch Neu-
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
rol. 1986;43:338-340. 20. Oh SJ, Sunwoo IN, Kim HS, Fraught E. Cervical and cortical somatosensory evoked potentials differentiate cervical spondylitic myelopathy from amyotrophic lateral sclerosis. Neurology. 1985;35:147-148. 21. Radtke RA, Erwin A, Erwin CW. Abnormal sensory evoked potentials in amyotrophic lateral sclerosis. Neurology. 1986;36:796-801. 22. Tsuji S, Muracka S, Kuroina Y, Chen KM, Gajdusek C. Auditory brainstem evoked response of Parkinson-dementia complex and amyotrophic lateral sclerosis in Guam and Japan. Rinsho
Shinkeigaku. 1981;21:37-41.
23. Yamada T, Bosch PE, Kimura J. Somatosensory evoked potentials in motor neuron disease. Electroencephalogr Clin Neurophysiol.
1984;58:54P.
24. Newrick PG, Langton-Hewer R. Motor neudisease: can we do better? a study of 42 patients. Br Med J. 1984;289:539-542. 25. Selby G. Hereditary motor neurone disease. Clin Exp Neurol. 1987;24:145-151. 26. Yiannikas C, Shahani BT, Young RR. Short\x=req-\ latency somatosensory-evoked potentials from radial, median, ulnar, and peroneal nerve stimulation in the assessment of cervical spondylosis. Arch Neurol. 1986;43:1264-1271. 27. Matheson JK, Harrington H, Hallett M. Abnormalities of somatosensory, visual, and brain stem auditory evoked potentials in amyotrophic lateral sclerosis. Muscle Nerve. 1983;6:529. 28. Dashieff RM, Drake ME, Brendle A, Erwin CW. Abnormal somatosensory evoked potentials in amyotrophic lateral sclerosis. Electroencephalogr Clin Neurophysiol. 1985;60:306-311. rone
in
Motor Neuron Disease Janice Shanthi Subramaniam, BSc(Hon), Con Yiannikas, MD, FRACP
\s=b\ We
performed median and tibial nerve
somatosensory evoked potentials (SEPs), pattern-shift visual evoked potentials (PSVEPs), and brain-stem auditory evoked potentials (BAEPs) on 27 patients with motor neuron disease (MND). Median and tibial nerve SEPs were abnormal in 8(30%) of 27 and 3 (14%) of 21 patients tested, respectively. Central and peripheral abnormalities were recorded in the absence of spondylosis. As a group, patients with MND and no evidence of cervical spondylosis had normal conduction to Erb's point following median nerve stimulation, but conduction times beyond this point were prolonged. The PSVEPs and BAEPs were within normal limits in all patients, excluding abnormalities attributable to other disease, but the group P100 latency was significantly prolonged in the group with MND. The BAEPs were normal in the group with MND. This study provides neurophysiological evidence of sensory system involvement in MND. (Arch Neurol. 1990;47:989-994)
/Totor neuron disease (MND) is clas-
sically described as a motor sys¬ involving progressive de¬ generation of the upper (corticospinal and corticobulbar tracts) and/or lower motor neurons. Clinically, it is mani¬ fested by weakness and amyotrophy with or without spasticity and hyper¬ reflexia. Limb, trunk, or bulbar mus¬ culature is affected. Clinically, sensory involvement has been thought not to occur, and the demonstration of sig¬ tem disorder
nificant sensory abnormalities on neu-
Accepted for publication January 29, 1990. From the Department of Clinical Neurophysiology, Westmead Hospital, New South Wales, Australia.
Reprint requests to Department of Clinical Neurophysiology, Westmead Hospital, West-
mead, New South Wales 2145, Australia (Dr Yiannikas).
rological or electrophysiological exam¬
ination has been grounds to reformu¬ late the diagnosis. However, a growing body of pathological18 and clinical810 evidence suggests that sensory path¬ ways may be involved in MND. Evoked potentials (EPs) provide an objective, noninvasive test of function in selected sensory pathways and have been used by several laboratories to investigate sensory involvement in MND.11 " To date, however, reports are conflicting and are confounded by in¬ adequate age matching of control data, insufficient screening of patients for additional disease, and methodological problems that make it impossible to localize abnormalities to the central nervous system or the periphery. In this study, we report the results of EP studies (median and tibial nerve somatosensory EPs [SEPs], patternshift visual EPs [PSVEPs], and brainstem auditory EPs [BAEPs]) per¬ formed on patients with different forms and varying degrees of MND. The patient group was carefully screened for additional disease and objectively assessed on a disability scale. Control data were age matched. PATIENTS AND METHODS
Twenty-seven patients (19 male, 8 fe¬ male) were diagnosed as having MND by a
neurologist based on the clinical and electromyographic findings and in the absence of alternate causes on radiologie investiga¬ tion. Patients were classified as having pseudobulbar palsy (exclusively upper mo¬ tor neuron signs), primary muscular atro¬ phy (PMA; exclusively lower motor neuron signs), or amyotrophic lateral sclerosis (ALS; both lower motor neuron and pyra¬ midal signs, with or without bulbar signs). Disease severity was assessed on a disabil¬ ity scale24 based on bulbar function, limb function, and patient independence. Low scores indicate severe disability; a maxi¬ mum score
of 25 indicates normal function.
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
Myelograms were performed to exclude other causes of upper motor neuron disease, and sensory changes were assessed as part of the neurological examination. Nerve con¬ duction studies and electromyography were performed. Disease severity and disease type were correlated with central conduc¬ tion time to median nerve stimulation for all patients with MND without concurrent cervical spondylosis. We also assessed the correlation among abnormalities of periph¬ eral nerve conduction, clinical sensory ex¬ amination, and EP abnormalities. Somatosensory EPs were elicited by uni¬ lateral percutaneous stimulation (0.2-millisecond square-wave pulse) at 2 Hz. Me¬ dian nerve SEPs were elicited by stimula¬ tion of the median nerve at the wrist at an intensity just sufficient to produce a small visible movement of the thumb. Tibial nerve SEPs were elicited by stimulation of the tibial nerve at the ankle at an intensity just sufficient to produce a slight plantar flexion of the foot. We recorded 256 or 512 responses to median and tibial nerve stim¬ ulation using 52- and 100-millisecond poststimulus sweeps, respectively. Recording band passes for subcortical and cortical re¬ sponses were 16 Hz to 3 kHz and 1.6 Hz to 3 kHz (—3 dB points), respectively. In median nerve SEPs, active electrodes over Erb's point, C-2, and the contralateral scalp 2 cm posterior to C-3 and C-4 (C3 and C4, respectively) were referred to P;. As¬ sessment was based on the absolute laten¬ cies of the Erb's point potential (N9), the cervicomedullary potential (P/N13), and the onset and peak of the thalamocortical potential (N190„ and N19pk, respectively). In addition, the N9-P/N13, P/N13-N19, N9N19, and N19on-N19pl interpeak latencies were measured, the latter to assess thalamocortical conduction. In tibial nerve SEPs, active electrodes at the level of the first lumbar vertebra and at C, were refer¬ enced to the iliac crest and F2, respectively. Assessment was based on the absolute la¬ tency of the spinal (N20) and cortical (N/ P37) potentials and the N20-N/P37 inter¬
peak latency.
We elicited PSVEPs by an alternating (2-Hz) black-and-white checkerboard pat¬ tern, each check subtending an angle of 50'
at the retina. The stimulus was
monocularly using
rotating
presented
mirror pat¬ tern-shift stimulator. Full-field stimula¬ tion was routinely used (angle subtended by full field at the retina was 23.8° horizon¬ tally, 21.2° vertically). Half-field stimula¬ tion was used when necessary to clarify waveforms. We averaged 256 or 512 re¬ sponses
over a
a
250-millisecond poststimu-
lus period (band width, points). Five-channel PSVEPs were re¬ corded using a midline electrode 5 cm above the inion and electrodes spaced 5 and 10 cm laterally on each side, referenced to Fs. Re¬ sponses were assessed on the latency of the first major positive component, P100. We elicited BAEPs by square-wave click stimuli (0.1 millisecond, 9 Hz) delivered through headphones. Rarefaction clicks were routinely used, but occasionally an al¬ ternating (condensation/rarefaction) stim¬ ulus was presented to clarify the position of wave I. Monaural click stimuli (70 dB above click threshold for each ear) and a con¬ tralateral masking white noise (40 dB above noise threshold for each ear) were pre¬ sented stimultaneously. We averaged 1024 or 2048 responses over a 10-millisecond poststimulus sweep (band width, 53.2 Hz to 3 kHz, -3 dB points). Two-channel BAEPs were recorded with active electrodes on the earlobes referenced to the vertex and were assessed on the latency of waves I, III, and V and the I-III, III-V, and I-V interpeak la¬ tencies. The EPs were routinely recorded using silver-silver chloride cup electrodes (im¬ pedance .l). Group comparisons of PSVEPs in 24 patients with MND and 24 age-matched controls (Table 4) showed a significant prolongation of the P100 latency at a level of signifi¬ cance of .01 ( .001). =
=
=
=
COMMENT
The demonstration of subjective and objective abnormalities in a small pro¬ portion of patients with MND810 sug-
Fig 1.—Median nerve somatosensory evoked potentials from four patients with sporadic mo¬ tor neuron disease and no other significant disease. Trials were replicated and superim¬ posed to ensure reproducibility in all patients. Center top and center bottom show the aver¬ age of the replicated trials. Top, Normal N9 and P/N13 absolute latencies but bilaterally pro¬ longed P/N13-N19conductionare shown (left interpeak latency [IPL], 7.9 ms; right IPL, 8.2 ms; mean age-matched control IPL + 2.5 SD, 7.0 ms). EP indicates Erb's point. Center top,
Normal N9 and P/N13 absolute latencies but bilaterally prolonged P/N13-N19 conduction (left IPL, 9.8 ms; right IPL, 10.7 ms; mean agematched control IPL + 2.5 SD, 7.0 ms) and bi¬ laterally prolonged 19on-N 19pk are shown (left IPL, 6.9 ms; right IPL, 6.3 ms; mean agematched control IPL + 2.5 SD, 5.0 ms). Center bottom, Normal N9 absolute latencies but pro¬ longed N9-P/N13 conduction on the left are shown (left IPL, 5.1 ms; mean age-matched control IPL + 2.5 SD, 4.8 ms). N9-P/N13 is at the upper limit of normal on the right (IPL, 4.8 ms). Bottom, Bilaterally delayed N9 potentials (left N9, 12.4 ms; right N9, 12.1 ms; mean age-matched control N9 + 2.5 SD, 10.4 ms).
gests that degeneration
Left
EP-F,
|2µ
CVF,
|2 ßV
|2µ
|4MV
\4ßV IPL + 2 SD O
IPL + 2.5 SD
5 10 15 20 25 30 35 40 45
Latency,
O
5 10 15 20 25 30 35 40 45
Latency,
ms
Left
|2MV
CVF,
|2µ
CVF,
|2mV
1.5
EP-F,
O
10
|1.S/»V
CVF,
2mV IPL + 2.5 SD
30
20
Latency,
nation results and abnormal sensory nerve conduction study results not as¬ sociated with any disease other than
µ
CVF,
IPL + 2.5 SD
pathways in this disease. In this study, two patients presented with subjective and objective sensory defi¬ cits (abnormal clinical sensory exami¬
AD
ms
Right
EP-F,
may involve
There is also convincing pathologi¬ cal evidence of sensory fiber involve¬ ment at a peripheral and central level. Autopsy studies of patients with ALS have demonstrated spinal ganglia involvement8 and marked pallor of the posterior columns, particularly in the fasciculus gracilis.3·8 A reduction in the neuronal population of Clarke's nucleus1 and a decrease in choline acetyltransferase level in the dorsal horns and in the region of Clarke's nu¬ cleus in patients with ALS has also been noted.5 Morphometric studies of the spinal cord at the level of L-57 demonstrated a 54% decrease in the number of large afferent neurons in the dorsal root ganglia and a 27% re¬ duction in the number of large myelinated fibers in the dorsal root. Other studies have demonstrated consistent segmental demyelination and remyelination in peripheral sensory nerves4 and a significant reduction in the den¬ sity of myelinated fibers in the sural nerve.2 Thus, there is convincing pathological and clinical evidence of sensory fiber involvement at central and peripheral levels. Hudson6 found the incidence of pos¬ terior column involvement to be less than 10% in sporadic ALS but as high
|2µ
EP-F
CVF,
sensory
MND).
Right
40
50
O
10
ms
20
Latency,
Left
40
30
50
SF
ms
Right
EP-F,
CVF,
1
µ\
EP-F,
1
ßV
CVF,
1
µ
µ
IPL + 2.5 SD N19
CVF,
1
10
20
30
Latency,
mV
C3-F,
1
40
10
ms
20
30
Latency,
Left
µ
40
ms
Right
EP-F,
EP-F,
CVF,
CVF,
C.-F,
CVF,
|1µ O
5 10 15 20 25 30 35 40 45
Latency,
ms
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
O
5 10 15 20 25 30 35 40 45
Latency,
ms
KL
as 70% in familial disease. Recently, it has been suggested25 that the incidence of familial MND may be frequently underestimated because of difficulty in obtaining sufficient comprehensive in¬ formation and that it may account for at least 10% of cases. Given the prom¬ inence of sensory disease in familial MND, one would predict a fairly high incidence of sensory disease in the population with MND at large. Such disease may well underlie the seem¬ ingly sporadic occurrences of sensory abnormalities in patients with MND. Several EP studies have investi¬ gated sensory tract involvement in MND, but to date most reports are equivocal. The early data of Dustman et al17 suggesting median nerve SEP abnormalities in patients with ALS were incomplete and not statistically analyzed. Most reports since then have been confounded by methodological problems. A number of studies report¬ ing median nerve SEP abnormalities fail to exclude or at least identify pa¬ tients with cervical spondylosis."·'6·"'23 The fact that all the patients of Matheson et al19 with abnormalities of the cortical potential also had P/N13 ab¬ normalities suggests that cervical my-
elopathy secondary to spondylosis may well have been a complicating factor in this study. Other studies1219 fail to lo¬
calize SEP abnormalities to the pe¬ riphery or the central nervous system, while still other reports of normal13·14·20 and abnormal21 median nerve SEPs in MND do not specify adequate age matching of patient and control data. Abnormal responses to median nerve stimulation were recorded in 11 of our 27 patients with MND compared with an age-matched control group. In 8 cases with no myélographie evidence of spondylosis, the abnormalities con¬ sisted of delayed N9 potentials, pro¬ longed N9-P/N13, P/N13-N19, and N9-N19 conduction times, as well as dispersion of the thalamocortical po¬ tential. In a further three patients with myélographie evidence of cervical spondylosis, the findings included pro¬ longed N9-P/N13 and N9-N19 conduc¬ tion times and normal central conduc¬ tion times, consistent with typical ab¬ normalities in cervical myelopathy.26 The findings that N9-P/N13 conduc¬ tion can be prolonged in cervical spondylosis and pure MND suggests that abnormalities between the bra¬ chial plexus and the lower medulla cannot be used in differentiating these disorders. However, an increase in cen¬ tral conduction time in the presence of normal N9-P/N13 conduction argues against spondylitic myelopathy as the cause.
Spondylosis
severe
enough
to
Right
|2MV
|4µ
CVF, C.-F,
12 µ
5 10 15 20 25 30 35 40 45
Latency,
O
5
10 15 20 25 30 35 40 45
Latency,
ms
VF
ms
nerve somatosensory evoked potentials from a patient with sporadic amyotrophic lateral sclerosis and concurrent C4-5 canal stenosis and C4-7 spondylosis. P/N13 potentials were absent bilaterally. N9-N19 conduction was at the upper limit of normal on the left and pro¬ longed on the right (left interpeak latency [IPL], 11.3 ms; right IPL, 12.3 ms; mean age-matched control IPL + 2.5 SE, 11.3 ms). EP indicates Erb's point.
Fig 2.—Median
Table
2.—Group Comparisons
in Evoked Potentials* MND
Potential Median nerve SEP No. of arms
Age,
57.2 ± 15.4
y
Latencies, ms N9 N9-P/N13
56.8 ± 15.0
9.8 ± 1.2 3.7 ± 0.6
P/N13-N19
.030 .008
3.4 ± 0.7
10.7 ± 1.2
.0001
61.4 ± 9.62
.50
N19„-N19Dh BAEP No. of
40
ears
Age, y Latencies,
ms
I
l-V
1.52 ± 0.21
1.60 ± 0.23
2.17 ± 0.22
1.95 ± 0.22
2.24 ± 0.22 1.90 ± 0.19
4.12 ± 0.25
4.14 ± 0.25
.10
PSVEP No. of eyes
Age,
62.6 ± 11.1
59.5 ± 12.3
y
100 latency,
ms
98.5 ± 6.6
102.7 ± 5.9
.0013
MND indicates motor neuron disease; SEP, somatosensory evoked potential; BAEP, brain-stem auditory evoked potential; and PSVEP, pattern-shift visual evoked potential. Values are mean ± SD. For SEPs, group data were from 11 patients with MND with normal cervical myelograms and 11 age-matched control subjects; for BAEP, group data were from 20 patients with MND with normal auditory thresholds and 20 age-matched con¬ trol subjects; and for PSVEP, group data were from 24 patients with MND with normal visual acuity and 24 age-
matched control subjects.
a central conduction delay would also severely disrupt N9-PN13 conduction.26 Our data do not support the suggestion made by Oh et al20 and Cascino et al13 that the presence of any SEP abnormality is useful in differen¬ tiating cervical spondylitic myelopa¬ thy from MND. However, it is an interesting observation that none of our group with pure MND had grossly dispersed or absent P/N13 potentials,
produce
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
whereas patients in our spondylitic group did. Perhaps the finding of se¬ verely disrupted N9-P/N13 conduction or absent P/N13 potentials is grounds to look for alternate or concurrent dis¬ ease.
Comparison of a group of patients with pure MND (normal cervical mye¬ lograms) with an age-matched control group showed no difference in periph¬ eral conduction (normal N9 latency)
but a prolongation of N9-P/N13 and P/N13-N19 conduction in the group with MND, with central conduction being more significantly affected. Given that the N9-P/N13 conduction was within normal limits in the ma¬ jority of individual studies, the group abnormalities may suggest a mild in¬ volvement of the large-diameter pe¬ ripheral sensory fibers in some pa¬ tients with MND. Central conduction may be more severely and/or fre¬
Right
C,-F
µ
1
C.-F,
µ
/ 37 N/P37 + 2.5 SD
N/P37 + 2.5 SD
quently affected, implying a more sig¬
nificant involvement of sensory tracts. Studies to date of tibial nerve SEPs in patients with MND also have meth¬ odological limitations. Abnormalities in the latency of the cortical potential and central conduction time have been reported,16·21·27 but coexisting spondylo¬ sis was not excluded by full myelogram in these studies. Other studies1321 do not adequately age match their patient and control groups. The report of Matheson et al19 of central conduction defects in 13 of their 32 patients is complicated by the L-l potential not having been recorded. They define a central conduction defect as an abnor¬ mal cortical potential in the setting of normal -reflexes. However, the SEPs were elicited by stimulation at the an¬ kle, while the -reflexes were elicited by stimulation at the popliteal fossa, leaving a significant segment of pe¬ ripheral sensory nerve unassessed by the -reflex. In addition, the H-reflex elicited at the popliteal fossa assesses a single cord segment (S-l), while stimulation of the tibial nerve at the ankle involves several segments (L-5S-2). We report abnormal spinal and/ or cortical responses to tibial nerve stimulation (in the setting of normal full spinal myelograms and normal nerve conduction study results) in 3 of 21 patients studied. In 2 of these, the conduction deficit may have been pe¬ ripheral; in the third, there was an un¬ equivocal central conduction deficit. Brain-stem auditory EPs have been reported as normal in a total of 30 patients.15·22-24 Two studies,19·21 in which control data were not adequately age matched, have reported BAEP abnor¬ malities. Our study, using agematched controls, found no BAEP ab¬ normalities attributable to MND. In particular, those patients with promi¬ nent oropharyngeal symptoms had normal BAEPs. Group statistics showed no BAEP differences between the group with MND and the control group, suggesting that degeneration may not affect this system. To date, PSVEPs have been normal in most MND studies.14·21·28 Matheson et al19 reported minor prolongation of
|2µ
|2µ
O
O 10 20 30 40 50 60 70 80 90
10 20 30 40 50 60 70 80 90
Latency,
Latency,
ms
EB
ms
Right
Left
C,-F,
|2µ
L,-it
|2µ
h µ
c-F,
|1 µ
N/P37 IPL + 2.5 SD O 10 20 30 40 50 60 70 80 90
O 10 20 30 40 50 60 70 80 90
Latency,
Latency,
ms
MC
ms
nerve somatosensory evoked potentials from two patients with motor neu¬ disease and no other concurrent disease. Top, Bilaterally absent N20 potentials and bilater¬ ally delayed N/P37 responses are shown (left latency, 52.4 ms; right latency, 50.4 ms; mean age-matched control latency + 2.5 SE, 50.0 ms). Bottom, Normal N20 absolute latency but bilat¬ erally dispersed /P37 potentials are shown. /P37 cannot be confidently identified on the right.
Fig 3.—Posterior tibial ron
0
2
4
6
Severe Disability
Newrick and Langton-Hewer Disability Score
Fig 4.—Plot of central conduction time (P/N13-N19) against Newrick and Langton-Hewer24 as¬ sessment of bulbar function, limb function, and patient independence in motor neuron disease (re¬ gression line and equation for data set are shown).
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
PlOO in 4 of 32 patients in a non-agematched study. In view of the signifi¬ cant increase in P100 latency after the fifth decade of life14 and the tendency of patients with MND to be in this older age group, age matching of control data is essential. No PSVEP abnor¬ malities were noted in individual pa¬ tients in our study using age-matched controls, but the average P100 latency for the patient group was significantly delayed compared with the control group (Table 4). Laxer and Spence18 reported a similar finding in a group of five patients, but Cascino and colleagues13 found no group PSVEP abnormalities in their patient group in a
non-age-matched study.
Previous studies19·21 have anecdot-
ally reported no relationship between disease severity and the presence of EP
abnormalities. Our data also suggest this, but the distribution of disability among our patient population too uneven to allow any conclu¬
scores was
sions. There
was no
tween clinical and EP
correlation be¬
abnormalities,
and clinical sensory deficits without evidence of alternate disease were rare and much less frequent than EP evi¬ dence of sensory involvement. This is consistent with the observation of others21 that SEPs may be a more sen¬ sitive index of sensory system involve¬ ment in MND than clinical examina¬ tion. CONCLUSION
This study reports SEP abnormali¬ ties in patients with MND with no al¬ ternative disease to explain the find¬ ings. The most common finding was prolongation of the central conduction time to median nerve stimulation, but prolonged peripheral, N9-P/N13, and N19on-N19pk conduction times were also noted. Since patients with cervical spondylosis also demonstrate abnor¬ mal N9-P/N13 and N9-N19 conduc¬ tion, disrupted conduction between the brachial plexus and the lower medulla is not a useful criterion for differenti¬ ating between MND and cervical spondylitic myelopathy. Increased
central conduction time in the pres¬ ence of normal N9-P/N13 conduction argues against spondylitic myelopathy
as the cause of the abnormality. Pe¬ ripheral and central conduction abnor¬
malities were recorded in the tibial SEPs. Occasional clinical sen¬ sory deficits were noted, but the fre¬ quency of these abnormalities com¬ pared with the frequency of SEP ab¬ normalities suggests that SEPs may be a more sensitive index of sensory disease in central pathways in MND than sensory examination. A small proportion of patients also had abnor¬ mal sensory nerve conduction study results, suggesting that peripheral fi¬ bers may also be involved in this dis¬ ease. The P100 latency in the patient group was significantly delayed, al¬ though the P100 latency of all individ¬ ual patients was within normal limits, suggesting that conduction in the vi¬ sual pathway is affected less severely than conduction in the large-fiber sen¬ sory system. We found no individual or group BAEP abnormalities. nerve
References 1. Averback P, Crocker P. Regular involvement of Clarke's nucleus in sporadic amyotrophic lateral sclerosis. Arch Neurol. 1982;39:155-156. 2. Bradley WG, Good P, Rasool CG, Adelman LS. Morphometric and biochemical studies of peripheral nerves in amyotrophic lateral sclerosis. Ann Neurol. 1983;14:267-277. 3. Castaigne P, Cambier J, Escourolle R, Brunet P. Sclerose laterale amyotrophique et lesions degeneratives des cordons posterieurs. J Neurol Sci. 1971;13:125-135. 4. Dayan AD, Graveson GS, Illis LS, Robinson PK. Schwann cell damage in motor neurone dis-
Neurology. 1969;19:242-246. Gillberg PG, Aquilonius SM, Eckernas S, Lundquist G, Winblad B. Choline acetyltransferase and substance P-like immunoreactivity in the human spinal cord: changes in ALS. Brain Res. 1982;250:394-397. 6. Hudson AJ. Amyotrophic lateral sclerosis and its association with dementia, Parkinsonism and other neurological disorders: a review. Brain. 1981;104:217-247. 7. Kawamura Y, Dyck PF, Shimono M, Okazaki H, Tateishi J, Doi H. Morphometric comparison of the vulnerability of peripheral motor and sensory neurons in amyotrophic lateral sclerosis. J Neuropathol Exp Neurol. 1981;40:675-687. 8. Luthy F, Martin F. Observation de cas de sclerose laterale amyotrophique avec consideration peculiere sur la participation sensitive. ease.
5.
Schweiz Med Wochenschr. 1947;25:694-697. 9. Dyck PJ, Stevens JC, Mulder DW, Espinosa RE. Frequency of nerve fibre degeneration of peripheral motor and sensory neurons in amyotrophic lateral sclerosis. Neurology. 1975;25:781\x=req-\
785. 10. Mulder DW, Bushek W, Spring E, Karnes J, Dyck PJ. Motor neuron disease (ALS): evaluation of detection thresholds of cutaneous sensation.
Neurology. 1983;33:1625-1627. 11. Anziska BJ, Cracco RQ. Short-latency somatosensory evoked potentials to median nerve stimulation in patients with diffuse neurologic disease. Neurology. 1983;33:989-993. 12. Bosch EP, Yamada T, Kimura J. Somatosensory evoked potentials in motor neuron disease. Muscle Nerve. 1985;8:556-562. 13. Cascino GD, Ring SR, King PJL, Brown RH, Chiappa KH. Evoked potentials in motor system diseases. Neurology. 1988;38:231-238. 14. Chiappa KH. Evoked Potentials in Clinical Medicine. New York, NY: Raven Press; 1983. 15. Chiappa KH, Harrison JL, Brooks EB, Young RR. Brainstem auditory evoked responses in 200 patients with multiple sclerosis. Ann Neurol. 1980;7:135-143. 16. Cosi V, Poloni M, Mazzini L, Callieco R. Somatosensory evoked potentials in amyotrophic lateral sclerosis. J Neurol Neurosurg Neuropsy-
chiatry. 1984;47:857-861.
17. Dustman RE, Snyder EW, Callner DA, Beck EC. The evoked potential as a measure of cerebral dysfunction. In: Begleiter H, ed. Evoked Brain Potentials and Behaviour. New York, NY: Plenum Press; 1979:321-363. 18. Laxer KD, Spence B. Visual evoked responses in amyotrophic lateral sclerosis. Electroencephalogr Clin Neurophysiol. 1980;50:168P. 19. Matheson JK, Harrington HJ, Hallett M. Abnormalities of multimodality evoked potentials in amyotrophic lateral sclerosis. Arch Neu-
Downloaded From: http://archneur.jamanetwork.com/ by a Western University User on 06/07/2015
rol. 1986;43:338-340. 20. Oh SJ, Sunwoo IN, Kim HS, Fraught E. Cervical and cortical somatosensory evoked potentials differentiate cervical spondylitic myelopathy from amyotrophic lateral sclerosis. Neurology. 1985;35:147-148. 21. Radtke RA, Erwin A, Erwin CW. Abnormal sensory evoked potentials in amyotrophic lateral sclerosis. Neurology. 1986;36:796-801. 22. Tsuji S, Muracka S, Kuroina Y, Chen KM, Gajdusek C. Auditory brainstem evoked response of Parkinson-dementia complex and amyotrophic lateral sclerosis in Guam and Japan. Rinsho
Shinkeigaku. 1981;21:37-41.
23. Yamada T, Bosch PE, Kimura J. Somatosensory evoked potentials in motor neuron disease. Electroencephalogr Clin Neurophysiol.
1984;58:54P.
24. Newrick PG, Langton-Hewer R. Motor neudisease: can we do better? a study of 42 patients. Br Med J. 1984;289:539-542. 25. Selby G. Hereditary motor neurone disease. Clin Exp Neurol. 1987;24:145-151. 26. Yiannikas C, Shahani BT, Young RR. Short\x=req-\ latency somatosensory-evoked potentials from radial, median, ulnar, and peroneal nerve stimulation in the assessment of cervical spondylosis. Arch Neurol. 1986;43:1264-1271. 27. Matheson JK, Harrington H, Hallett M. Abnormalities of somatosensory, visual, and brain stem auditory evoked potentials in amyotrophic lateral sclerosis. Muscle Nerve. 1983;6:529. 28. Dashieff RM, Drake ME, Brendle A, Erwin CW. Abnormal somatosensory evoked potentials in amyotrophic lateral sclerosis. Electroencephalogr Clin Neurophysiol. 1985;60:306-311. rone
