Tonic "Seizures" in a
Patient With
Brainstem
Demyelination: MRI
Study of Brain and
Spinal Cord
Mark H. Libenson, MD,
Carl E. Stafstrom, MD, PhD, and
N. Paul Rosman, MD

Libenson MH, Stafstrom CE, Rosman NP. Tonic "seizures" in a patient with brainstem demyelination: MRI
study of brain and spinal cord. Pediatr Neurol 1994;11:
258-262.

Introduction
Tonic seizures are an unusual and poorly understood
manifestation of demyelinating disease, first reported in 4
patients with multiple sclerosis [1]. There has been much
speculation as to the clinicopathologic correlates of this
phenomenon, with only a few reported patients with computed tomography- or magnetic resonance-documented lesions which were believed to underlie the tonic seizures
[2-5]. We describe a patient in whom tonic extension of
the left limbs caused by a right-sided brainstem lesion was
the first manifestation of demyelinating disease.
Case Report

Tonic seizures are a poorly understood manifestation
of demyelinating disease, first reported in 4 patients
with multiple sclerosis. We describe a patient with
tonic extension of the left limbs caused by a right-sided
bralnstem lesion as the first manifestation of demyelinating disease. A 19-year-old man was referred with
a 4-month history of spontaneous attacks of mild paresthesias of the left arm and leg, followed by 15-45 s of
rigid extension of the left limbs, occurring up to 25
times per day. Two months after onset, an MRI scan
revealed areas of T 2 abnormality in the lateral right
cerebral peduncle and deep frontal white matter. The
EEG was normal, including during hyperventilation
which induced a typical episode. All attacks were successfully suppressed by carbamazepine, phenytoin,
and valproate monotherapy. Serologic testing for toxoplasmosis, cytomegalovirus, Epstein-Barr virus, Lyme
disease, and HIV was negative. Cerebrospinal fluid
oligocional bands were absent but cerebrospinal fluid
immunogiobulin G was mildly elevated (4.2 mg/dl).
Over the next 30 months, serial MRIs revealed a normal spinal cord and persistence of the midbrain lesion,
with resolution of some of the white matter lesions but
reappearance of others. At 46 months, the midbrain
lesion resolved on MRI, and the spasms no longer occurred spontaneously, nor could they be elicited by
hyperventilation. While two previous reports have
shown internal capsule lesions to underlie the tonic
spasms in demyelinating disease, this is the first report
in which a bralnstem lesion has been causative.

A 19-year-old black man was referred to the Floating Hospital for
Children with a 4-month history of spasms of the left arm and leg stiffness. The episodes occurred up to 25 times per day with mild paresthesias in the left limbs, followed seconds later by rigid extension of these
limbs and flexion of the left hand lasting 15-45 s, followed by several
minutes of left-sided weakness. These attacks often occurred spontaneously, though occasional attacks appeared to be triggered by exercise.
Consciousness was always preserved. Past medical history was remarkable for an episode of uveitis at 6 years of age with resultant right eye
blindness, and oral ulcers 2 months before the onset of the tonic attacks.
Two months after onset of the spasms, the patient underwent cranial
MRI which revealed several discrete areas of enhanced signal intensity
on T2-weighted images: lateral aspect of the right cerebral peduncle
extending toward the red nucleus, deep left frontal white matter, and
deep right parietal white matter (Fig 1). An electroencephalogram was
normal, including during hyperventilation which induced a typical episode.
Carbamazepine (100 mg three times daily) was administered after
which all episodes ceased, but development of a rash necessitated its
replacement with clonazepam, 2 mg twice daily. The episodes returned
and phenobarbital 120 mg daily was added, but without improvement.
General physical examination was normal except for changes secondary to uveitis in the right eye and a corneal opacity, both present since 6
years of age. Neurologic examination was normal, except for visual
acuity (20/400 O.D. and 20/20 O.S.) and deep tendon reflexes that were
slightly brisker on the left.
The following laboratory tests were normal: hemogram; serum electrolytes, calcium, and magnesium; thyroid function tests; urine toxic
screen; erythrocyte sedimentation rate; serum immunoglobulin G; antinuclear antibodies; complement studies; cryoglobulins; HLA-B27 (negative); angiotensin-converting enzyme; and antibody titers against toxoplasmosis, cytomegalovirus, Epstein-Barr virus, Lyme disease, and
HIV. Serum glutamic oxaloacetic transaminase was 91 mU/ml (normal:
<35 mU/ml), serum glutamic pyruvic transaminase 104 mU/ml (normal:
<28 mU/ml), and lactic dehydrogenase 208 mU/ml (normal: < 170 mU/
ml). Cerebrospinal fluid (CSF) examination, including myelin basic

From the Departments of Pediatrics and Neurology; Division of
Pediatric Neurology; The Floating Hospital for Children; New England
Medical Center; Boston, Massachusetts.

Communications should be addressed to:
Dr. Libenson; Box 393; The Floating Hospital for Children; 750
Washington Street, Boston, MA 02111.
Received June 10, 1994; accepted August 16, 1994.

258

PEDIATRIC NEUROLOGY

Vol. 11 No, 3

© 1994 by Elsevier Science Inc. • 0887-8994/94/$7.00

J
.....

A

j

B

Figure 1. Te-weighted MRI scans obtained 2 months after the onset of symptoms (1.0 T; TR: 3,000 ms, TE : 90 ms). (A ) Area of increased signal
intensity in the lateral right cerebral peduncle (arrow). (B) Area of increased signal intensity in the deep left frontal white matter (arrow).

protein and oligoclonal bands, was normal. CSF immunoglobulinG was
4.2 mg/dl (normal: 0.7-3.5 mg/dl). Visual evoked potentials confirmed
right eye blindness but were otherwise normal.
Repeat MRI scan 7 months after presentation indicated persistence of
the increased signal in the right cerebral peduncle on T2-weighted images, but disappearance of the lesions in the left frontal and right posterior white matter (Fig 2). MRI of the cervical spinal cord was normal.
The patient was switched to phenytoin monotherapy, 300 rag daily, and
remained symptom-free for 1 year, at which time he discontinued this
medication himself.
Thirty months after symptom onset, the MRI indicated persistence of
the right mesencephalic lesion and appearance of a new area of increased
T2 signal adjacent to the atrium of the right lateral ventricle. Four months
later, he had recurrence of the same tonic spasms and treatment with
valproic acid was initiated which completely suppressed his symptoms.
Forty-six months after symptom onset, the MRI revealed nearly complete
resolution of the mesencephalic lesion, with several new lesions in the
deep white matter (Fig 3). The spasms were no longer occurring, even
after induced hyperventilation.

Discussion
Our patient experienced tonic episodes of the left arm
and leg with a lesion involving the most lateral aspect of
the right cerebral peduncle present on 2 successive cerebral MRI scans, spontaneously resolving after 3V2 years.
Earlier left frontal and right parietal white matter lesions
had also resolved, but new white matter lesions had appeared. Thus, during the 5 years of follow-up multiple
white matter lesions have appeared and disappeared. MRI
of the cervical spinal cord was normal.

Given the age of onset, the appearance and disappearance of multiple white matter lesions on MRI, and the
mildly elevated CSF g a m m a globulin, it seems clear that
these findings reflect the formation and resolution of the
demyelinative plaques of multiple sclerosis. Moreover,
tonic spasms are characteristic of that disorder [1,2,6-9].
Infectious causes were excluded by negative titers, and
vasculitides by negative laboratory studies. The presence
of uveitis and mouth ulcers in our patient raises the question of Beh~et disease, a disorder that can present with
uveitis, oral and genital ulcers, arthritis, or central nervous
system manifestations (e.g,, meningoencephalitis, headache, corticospinal tract involvement, or focal brainstem
signs) [10,11]. In addition, MRI studies have demonstrated reversible cerebral and brainstem lesions in Beh~et
disease [12]. Several factors, however, weigh against
Beh~et disease which is extremely rare in children [13]:
the patients are typically quite ill, with uveitis rarely the
presenting symptom; and normal erythrocyte sedimentation rate, leukocyte count, and CSF protein, and the absence of CSF pleocytosis make the diagnosis unlikely
[11,14]. Furthermore, our patient does not fulfill the established criteria for Beh~et disease [15].
The frequency of tonic seizures in multiple sclerosis has
ranged from 4% to 17% in different studies [6,16]. In
about 10% of patients, the tonic episodes are the first
manifestation of the disease [2]. The attacks are stereo-

Libenson et al: Tonic "Seizures" and Brainstem Demyelination 259

~~

'2-

-~

~

"

~

%

i i~ii~i:
I

L
B

A

Figure 2. Repeat T2-weighted MRI scans 7 months after onset of symptoms (1.0 T; TR: 3,000 ms, TE: 90 ms). (A) The lesion in the right
cerebral peduncle persists (arrow). (B) The lesion previously evident in
the deep left frontal white matter is no longer visible. (C) No abnormal
signal is present in the spinal cord (1.0 T; TR: 3.0 ms, TE: 90 ms).

C
typed, usually beginning with paresthesias in 1 limb, followed by spread over that half of the body, followed seconds later by a tonic posturing of that limb. The leg is
involved less often than the ann, though both may be
affected, as in our patient. In some patients, the tonic
spasms occur contralateral to the side of the paresthesias,
suggesting a spinal cord localization [17]. The right mesencephalic lesion can explain both the sensory and motor
components of our patient's spasms, and location of corticobulbar fibers medially in the cerebral peduncle [18]
can explain the sparing of his face.

260

PEDIATRIC NEUROLOGY

Vol. 11 No. 3

Such tonic episodes typically last from 15 to 60 s, with
no change in consciousness or "post-ictal" phenomena,
and there is no seizure activity on EEG. Various factors
can trigger the attacks. The episodes are usually wellcontrolled with antiepileptic drugs, especially carbamazepine and phenytoin [2,7-9,16,19,20], as in our patient, although hyperventilation could still elicit attacks.
There has been some discussion about whether tonic
seizures in multiple sclerosis represent true "epileptic"
activity. Matthews noted the similarity of the tonic postures seen in these episodes to decorticate posturing,

~, ]IV, V

A

B

Figure 3. Repeat MRI scans 46 months after symptom onset demonstrating that the mesencephalic lesion has nearly completely resolved (A )
(1.5 T; TR: 3,000 ms, TE: 90 ms), but that new lesions are present in the deep cerebral white matter (B) (1.5 T; TR: 3,000 ms, TE: 30 ms).

which he presumed was a "negative" phenomenon [1] or
a brainstem "release" phenomenon manifesting the absence of normal functioning of an area of the brainstem.
Matthews subsequently revised this notion, documenting
several patients who did not assume this stereotyped posture. He then suggested that the phenomena could be more
readily explained by abnormal transmission within demyelinative plaques [8].
The possibility that such episodes could be caused by
abnormally discharging brainstem neurons has been studied experimentally in an animal model of demyelination
that has demonstrated several types of aberrant conduction
in demyelinated axons [21,22]. This could account for
some of the clinical phenomena observed in such cases. In
demyelinated areas, impulses can propagate from axon to
axon (so-called "cross-talk" or ephaptic transmission)
rendering the region hyperexcitable and capable of generating volleys of abnormal (ectopic) impulses spontaneously [23,24]. Such injured areas become especially hyperexcitable if exposed to a low calcium environment; this
could provide a physiologic explanation for tonic spells
induced by hyperventilation (which lowers the concentration of ionized calcium by increasing blood pH), as occurred in our patient.
There are 3 reports of patients with tonic seizures in
demyelinating disease in which cranial imaging has demonstrated cerebral abnormalities. Berger et al. found 4 of
5 CT scans normal, while 1 scan showed " m a n y small
periventricular and subcortical enhancing lesions" [2].
Watson and Chiu [3] reported a patient with multiple scle-

rosis and left-sided tonic spasms associated with an enhancing lesion of the posterior limb of the right internal
capsule, but the study was limited by poor technical resolution of the scan and the difficulty of identifying brainstem lesions by CT. Maimone et al. reported a young
woman with tonic spasms involving the left hemibody and
a lesion in the right internal capsule on MRI [4]. Muramoto et al. described a patient with left-sided tonic
spasms but a right-sided cervical cord lesion and questionable involvement of the medulla on MRI; the right-sided
spinal cord lesion does not, however, easily explain the
left-sided motor phenomena [5].
In our patient, the specific plaque location shown by
MRI (cerebral peduncle), the stereotyped nature of the
attacks, and their provocation by hyperventilation strongly
suggest that this lesion was responsible for his symptoms,
and is the first example of an MRI-documented brainstem
lesion causing tonic seizures.
References

[1] MatthewsWB. Tonic seizures in disseminated sclerosis. Brain
1958;81:193-206.
[2] Berger JR, SheremataWA, MelamedE. Paroxysmaldystoniaas
the initial manifestationof multiple sclerosis. Arch Neurol 1984;41:74750.
[3] WatsonCP, Chin M. Painful tonic seizuresin multiple sclerosis:
Localization of a lesion. Can J Neurol Sci 1979;6:359-61.
[4] Maimone D, RederT, FinocchiaroF, RecuperoE. Internalcapsule plaque and tonic spasms in multiple sclerosis. Arch Neurol 1991;
48:427-9.
15] Muramoto T, Kira J, YoshimuraT, Goto I. Acutedisseminated

Libenson et al: Tonic "Seizures" and BralnstemDemyelination 261

encephalomyelitis affecting medulla and upper cervical cord-Comparison of the clinical and MRI findings. Rinsho Shinkeigaku 1989;
29:1013-6.
[61 Shibasaki H, Kuroiwa Y. Painful tonic seizures in multiple sclerosis. Arch Neurol 1974;30:47-51.
[7] Ostermm PO, Westerberg C-E. Paroxysmal attacks in multiple
sclerosis. Brain 1975;98:189-202.
[8] Matthews WB. Paroxysmal symptoms in multiple sclerosis. J
Neurol Neurosurg Psychiatry 1975;38:617-23.
[9] Twomey JA, Espir MLE. Paroxysmal symptoms as the first
manifestation of multiple sclerosis. J Neurol Neurosurg Psychiatry 1980;
43:296-304.
[10] Choi SJ. Behcet's syndrome. In: Schurnacber HR, ed. Primer
on the rheumatic diseases. 9th ed. Atlanta: Arthritis Foundation,
1988;192-3.
[11] Sehotland DL, Wolf SM, White HH, Dubin H. Neurologic
aspects of Behfet's disease: Case report and review of the literature. Am
J Med 1963;34:544-53.
[12] Mhmkaml K, Shiraishi H, Tanaka Y, et al. CNS changes in
neuro-Behcet's disease: CT, MR, and SPECT findings. Comput Med
Imaging Graph 1992;16:401-6.
[13] Rakover Y, Adar H, Tal I, Lang Y, Kedar A. Behcet disease:
Long-term follow-up of three children and review of the literature. Pediatrics 1989;83:986-92.
[14] O'Duffy JD, Goldstein NP. Neurologic involvement in seven
patients with Behcet's disease. Am J Med 1976;61:170-8.

262 PEDIATRIC NEUROLOGY

Vol. 11 No. 3

[15] International Study Group of Behfet's disease. Criteria for
diagnosis of Beh~et's disease. Lancet 1990;335:1078-80.
[16] Espir MLE, Millac P. Treatment of paroxysmal disorders in
multiple sclerosis with carbamazepine (Tegretol). J Neurol Neurosurg
Psychiatry 1970;33:528-31.
[17] Ekbom KA, Westerberg C-E, Osterman PO. Focal sensorymotor seizures of spinal origin. Lancet 1968;1:67.
[18] Carpenter MB. Core text of neuroanatomy. Baltimore:
Williams & Wilkins, 1985;79, 197.
[19] Toyokura Y, Sakuta M, Nakanishi T. Painful tonic seizures in
multiple sclerosis (abstract). Neurology 1976;26(Part 2):18-9.
[20] Tan CT, Low ET. Paroxysmal tonic seizures in demyeimating
disease. Singapore Med J 1986;27:256-8.
[21] Rasminsky M. Hyperexcitability of pathologically myelinated
axons and positive symptoms in multiple sclerosis. Adv Neurol 1981;
31:289-97.
[22] Ruminsky M. Ephaptic transmission between single nerve fibers in the spinal nerve roots of dystrophic mice. J Physiol 1980;305:
151-69.
[23] Waxman SG. Clinical course and electrophysiology of multiple sclerosis. Adv Neurol 1988;47:157-84.
[24] Smith KJ, McDonald WI. Spontaneous and evoked electrical
discharges from a central demyelinating lesion. J Neurol Sci 1982;55:
39-47.