Journal of

Neurology

J Neurol (1985) 232 : 162-166

© Springer-Verlag 1985

Epilepsia partialis continua:
active cortical spike discharges and high cerebral blood flow
in the motor cortex and enhanced transcortical long loop reflex
Y. Kuroiwa 1, H. Tohgi l, A. Takahashi 2, and H. Kanaya 2
Departments of 1Neurology and 2Neurosurgery, Iwate Medical University, 19-1 Uchimaru, Morioka 020, Japan

Summary. We report a patient suffering from persistent myoclonic jerks in the right forearm without any definite E E G
abnormality under routine recording conditions. By computer
summation, using the jerk-locked averaging technique, a
sharp spike was recognized as a precisely time-locked event in
relation to myoclonic twitches. A cranial CT scan revealed a
small cortical lesion, which was found very close to the
sensorimotor cortex of the right arm. Cerebral blood flow
study using the xenon inhalation method revealed a discrete
focus of high flow, which corresponded well with the CT
lesion. On electrical stimulation of the right median nerve, a
large somatosensory evoked potential and an enhanced transcortical long loop reflex were observed. Electrocorticogram
showed active focal spike discharges localized at the left
precentral gyrus. We postulate that an epileptogenic focus in
the motor cortex and an enhanced transcortical long loop
reflex appear to be important for the occurrence of epilepsia
partialis continua in this patient.

Key words: Electrocorticogram - Epilepsia partialis continua
- Long loop reflex - Myoclonus - Somatosensory evoked
potential

Introduction
Epilepsia partialis continua (EPC) refers to clonic muscular
twitching repeated at fairly regular short intervals in one part
of the body for a period of days, weeks or sometimes years [2,
8]. We now describe a case of EPC, in which active focal spike
discharges were recorded directly from the surface of the
motor cortex. Computer summation of scalp E E G in this
patient disclosed a time-locked spike preceding myoclonic
twitches. The essential cause of EPC seems to be abnormal
excitability of the motor cortex. Since some patients with
myoclonus show enhanced transcortical long loop reflexes [7,
9, 11, 12] and EPC differs from other types of myoclonus only
in its consistently unilateral manifestation [8], we studied
transcortical long loop reflexes and somatosensory evoked
potentials in our patient.

Case report
A 59-year-old right-handed man noted abrupt intermittent
brief jerks confined to the right forearm and weakness of the

Offprint requests to."Dr. Y. Kuroiwa (address see above)

right hand on 22 December 1982. The myoclonic twitches
occurred continuously at intervals varying from 3 to 4 s. The
myoclonic jerks were aggravated when he voluntarily moved
his right arm and disappeared only during sleep. On 1 January
1983, he developed a Jacksonian motor seizure when grasping
chopsticks with his right hand. Initially it affected the right
arm and spread to the right leg, as well as to the right hemifacial muscles. The seizure was preceded by severe head-heaviness and was associated with a brief period of cyanosis and
clouded consciousness, and he was admitted to the Department of Neurology, Iwate Medical University. He had been
diagnosed as having diabetes mellitus at the age of 40 years
and hypertension at age 45. Neurological examination revealed mild monoparesis and continuous jerks affecting the right
upper limb. Otherwise no abnormality was found. During the
hospital course, the myoclonic twitches became more frequent
and affected bifrontal muscles as well as the right arm. The
monoparesis also progressed. Various anticonvulsants, including diazepam, phenytoin, phenobarbital and clonazepam
were used, but no adequate effect was obtained on the persistent myoclonic jerks.
A blood cell count, routine blood chemistry values and
urine analysis were normal. Cerebrospinal fluid was normal
with negative V D R L results. Coronal and sagittal sections of
cranial CT scan showed a small low-density lesion, which was
found just below the skull, approximately at a coronal slice of
central sulcus (Fig. 1). Cerebral blood flow study using the
xenon inhalation method revealed a discrete focus of high
flow, corresponding well with the CT lesion.

Electrophysiological findings
Surface EMG. A pair of disc electrodes was placed 3 cm apart
over several muscles. The E M G showed continuous synchronous clonic jerks, affecting the right thenar muscle, extensor
carpi radialis, biceps, triceps, deltoid, and bifrontal muscles
(Fig. 2). The myoclonic twitches recorded from the surface
E M G became less frequent during drowsiness and disappeared after the patient fell asleep.
Scalp EEG and jerk-locked averaging study. The routine scalp
E E G recorded from surface electrodes placed according to a
10-20 international system showed no abnormality, except for
E M G artefacts related to clonic jerks. Then the scalp E E G
activity was studied by the jerk-locked averaging technique
[11, 12]. The E E G analysis was triggered when the amplitude
of a myoclonic discharge from the right frontal muscle reached

163

Fig.l. Cranial computed tomography. A low-density lesion is shown on coronal and sagittal sections

~lti~h~im1ii~qlii~1iiiiiiiiiii~iiiiiii~iiiiiiiiNi~iiii11iiiiiii~ii1~ii~i1~ii~uii1imi1iiiIiiiiii~i~iIiiiiii~i~1iii~1iii~ii~i1ii1ii~iiiiiiiiiii~1i~iiiiiiiiiiii1

Rt.obd .p.brev.
Rt.ext.carpi
rodialis

Rt. deltoid

Rt.trcapezi us

Rt. f r ontalis

Lt. frontatis

""

r-

,

I

.

,

'!

I sec
Fig.2. Surface EMG. Myoclonic twitches occur at regular intervals

a certain level. Pretriggered analysis was carried out by averaging 200 samples of scalp E E G , time-locked to the myoclonic
discharge from the right frontal muscle. While the patient was
awake, a positive spike was observed at onset latency of
20-24ms, preceding the triggering jerk (Fig. 3). The positive
spike was predominantly seen from the electrodes opposite

the triggered muscle contractions, but was not recognized in
the jerk-locked averaging study, while the patient was drowsy.

Somatosensory evoked potential and long loop reflex. Somatosensory evoked potentials were obtained with a square wave
electrical pulse of 0.5 ms duration, applied every second to the

AVERAGED

EMG ( R I G H T

FRONTALIS)

~,,~o~o~.~.~ .,/,%/,~,,~
_.,

T3.A1

C3-A1

T5.A1

P3"A1

C4-A2

T4.A2

P4-A2

T6.A2

....

01.A1

I 2o pV

-204.8
I

/
"~,,

~"~

,4t,~

0

msec.
~

~ '

÷ 204.8
|

,

~
~ ~

,

Fig.3. Jerk-locked averaging study. Positive spikes precede myoclonic twitches from the right frontalis
Right median

•
.

~

nerve

stimulation

,~ % ,
', ,

,. : , ~

•

~

~#~ ~

t-~"

,~¢,~ke

'~.f? ~ ~ '

SEP

f

C3.A1
C4. A2
~

, ,~,~~
,.v.~ :"~
' 5*,~L,,~ ,,~~ae~, Y~

(.
Surface
EMG
(A.P.B.)

right

Fig.4. Somatosensory evoked potential and averaged surface EMG
from abductor pollicis brevis on electrical stimulation of the right
median nerve at the wrist. A r r o w indicates a negative peak latency of
21.4 ms. Five stars indicate repetitive long loop reflexes

left

20

/JV

0
F

msc.

20l,.8
!

165

4
I eec ~

I~lO00 jJV

Fig.5. Electrocorticogram. Electrodes 1-3 were placed over the surface of the cerebral cortex. An electrode 4 was inserted into the precentral
gyrus. Arrows indicate central sulcus
right or left median nerve at the wrist. Two hundred samples
of scalp E E G from C3-A1 and C4-A2 were averaged. The
peak latency of these somatosensory evoked potentials was
normal. However, an evoked potential with stimulation of the
right median nerve had large peak-to-peak amplitude
(62.2gV, N20-P25). Then 200 samples of E E G from C3-A1
and surface E M G from the right thenar muscle were simultaneously averaged with electrical stimulation of the right
median nerve at the wrist. The stimulus strength was adjusted
to 5% above the threshold for contraction of the right thenar
muscle. The first negative peak and the first positive peak of
the cortical somatosensory evoked potential had a peak
latency of 21.4ms and 27.0ms, respectively. These cortical
potentials were followed by myographic discharges from the
thenar muscle, M1-M5 having a negative peak latency in each
of 41.6 ms, 98.8 ms, 116.0 ms, 128.8 ms, and 142.2 ms (Fig. 4).
These latency values were fairly constant throughout five
sessions of recording.

Electrocorticogram. During the operative procedure for brain
biopsy, the electrocorticogram showed active spike discharges
from the deep electrode inserted in the left precentral gyrus
and from the surface electrode placed over the same gyrus
(Fig.5). General anaesthesia was maintained with nitrous
oxide, fluothane and oxygen. Administration of fluothane had
been discontinued 30 min before electrocorticographic recording was initiated.

Discussion

Many authors [1, 10, 13, 14] claim that the cause of EPC is
solely to be found in the cortex. Most patients with subcortical

lesions have also shown pathological changes in the cortex
upon histological examination [3, 4, 6, 15]. Thomas et al. [14]
demonstrated an active focal spike abnormality, recorded
from the cortical surface, in all five patients with EPC for
whom electrocorticography was carried out. Therefore, the
essential lesion for EPC is presumed to be situated in the
cortex. However, the pathophysiological mechanism of continuous rhythmic clonic jerks in EPC is still obscure.
In our case, the CT scan showed a small low-density area
in the left cerebral cortex, closely adjacent to the sensorimotor
representation of the right upper limb, which exhibited continuous rhythmic jerks. The spike focus of the electrocorticogram at the left precentral gyrus and a high flow spot in cerebral blood flow study are presumed to correspond with the
localization of the above CT abnormality. In a jerk-locked
averaging study [11, 12], the latency from the onset of a positive spike to a myoclonic discharge was 20-24ms in our
patient, which is adequate for an impulse to be conducted
from the motor cortex to the right frontal muscle via the pons
and the right facial nerve. Therefore, we concluded that the
origin of EPC was in the left motor cortex.
The transcortical long loop reflex (TLLR) was obtained
easily at rest by electrical stimulation of the right median
nerve at the wrist. In normal subjects, T L L R is rarely obtained at rest [5]. The first response, M1, having a peak latency of
41.6 ms, is presumed to correspond with the preceding cortical
response, having a negative peak of 21.4 ms in somatosensory
evoked potential. The TLLR, M1, was followed by several
myographic discharges, M2-Ms. The time interval between
M1 and Ma (57.2ms) is sufficient for proprioceptive impulse
produced by contraction of the right thenar muscle (M1), to be
transmitted to the sensorimotor cortex, than eliciting the
second TLLR (M2), via a transcortical pathway. The time

166
intervals b e t w e e n M2 a n d M3, M3 a n d M4, M4 and M5 r a n g e d
f r o m 12.8 to 17.2 ms. T h e s e interval values are too short for an
impulse to b e t r a n s m i t t e d t h r o u g h a transcortical long circuit.
It is p o s t u l a t e d t h a t some k i n d of i n t r a c e r e b r a l short circuit
plays a role in g e n e r a t i n g repetitive discharges, M3-Ms. O u r
study suggests t h a t T L L R plays some role for an epileptogenic
focus in the m o t o r cortex to g e n e r a t e E P C .

References
1. Burr CW (1915) Continuous clonic spasm of the left arm (Epilepsia
continua) caused by a tumor of the brain. Am J Med Sci 149:169174
2. Celesia GG (1982) Prognosis in convulsive status epilepticus. In:
Delgado-Escueta AV, Wasterlain CG, Treiman DM, Porter RJ
(eds) Advances in neurology, vol 34. Status epilepticus. Raven
Press, New York pp 55-59
3. Chatrian GE, Shaw CM, Plum F (1964) Focal periodic slow
transients in epilepsia partialis continua. Clinical and pathological
correlations in two cases. Electroencephalogr Clin Neurophysiol
16 : 387-393
4. Chatrian GE, Shaw CM, Leffman H (1964) The significance of
periodic lateralized epileptiform discharges in EEG. An electrographic, clinical and pathological study. Electroencephalogr Clin
Neurophysiol 17 : 177-193
5. Chauvel P, Louvel J, Lamarche M (1978) Transcortical reflexes
and focal motor epilepsy. Electroencephalogr Clin Neurophysiol
45 : 309-318

6. Ehrler P (1945) Ober epilepsia partialis continua (Kojewnikoff)
mit anatomischen Befund. Monatsschr Psychiatr Neurol 110:
280-290
7. Hallett M, Chadwick D, Marsdcn CD (1979) Cortical reflex myoclonus. Neurology (Minneap) 29:1107-1125
8. Juul-Jensen P, Denny-Brown D (1966) Epilepsia partialis continua. A clinical, electroencephalographic, and neurophysiological
study of nine cases. Arch Neurol 15 : 563-578
9. Kelly Jr J J, Sharbrough FW, Daube JR (1981) A clinical and
electrophysiological evaluation of myoclonus. Neurology (NY)
31 : 581-589
10. Kugelberg E, Wid6n L (1954) Epilepsia partialis continua. Electroencephalogr Clin Neurophysiol 6 : 503-506
11. Shibasaki H, Kuroiwa Y (1975) Electroencephalographic correlates
of myoclonus. Electroencephalogr Clin Neurophysiol 39 : 455-463
12. Shibasaki H, Yamashita Y, Kuroiwa Y (1978) Electroencephalographic studies of myoclonus. Myoclonus-related cortical spikes
and high amplitude somatosensory evoked potentials. Brain
101 : 447-460
13. Spiller WG, Martin E (1909) Epilepsia partialis continua occurring in cerebral syphilis. JAMA 52:1921-1922
14. Thomas JE, Reagan TJ, Klass DW (1977) Epilepsia partialis continua. A review of 32 cases. Arch Neurol 34 : 266-275
15. T6bel F, Schaltenbrand G (1947) Beitrag zur Frage der Lokalisation und Entstehung des Kojewnikoffschen Syndrome (Epilepsia
partialis continua). Nervenarzt 18 : 501-505

Received February 6, 1985/Accepted March 14, 1985