335

Journal of Neurology, Neurosurgery, and Psychiatry 1990;53:335-338

Pure agraphia of kanji due to thrombosis of the
Labbe vein
Takanori Yokota, Sumio Ishiai, Tetsuo Furukawa, Hiroshi Tsukagoshi

Abstract
A case is described of a 56 year old Japanese male with pure agraphia of kanji (the
Japanese morphograms) due to haemorrhagic infarction of the left temporal lobe
caused by the rare condition of cortical
vein thrombosis of Labbe. Writing kanji
was severely impaired without disturbed
consciousness, aphasia or apraxia. On the
other hand, writing kana (the Japanese
syllabograms), and reading kanji and
kana were almost intact. This suggests
that the process of writing kanji involves a
different pathway from that of reading
kanji in the left temporal lobe. Pure agraphia of kanji is considered to be similar to
lexical agraphia in Indo-European languages, in that the writing system with a
poor or irregular. phoneme-grapheme
transformation is impaired by the left
temporal lesion. This case indicates
the necessity for considering thrombosis
of the Labbe vein when a subcortical
haematoma is detected in a temporal lobe
on computed tomography of the brain.

Department of
Neurology, Tokyo
Medical and Dental
University, Tokyo,
Japan
T Yokota
S Ishiai
T Furukawa
H Tsukagoshi
Correspondence to:
Dr Takanori Yokota,
Department of Neurology,
Tokyo Metropolitan
Neurological Hospital, 2-6-1
Musashidai, Fuchu-shi
Tokyo 183, Japan.
Received 10 April 1989 and
in revised form 31 July 1989.
Accepted 21 August 1989

The Japanese language has two different writing systems, kana and kanji, and Japanese
sentences usually consist of combinations of
both. Generally, kanji are used for writing most
nouns and roots of verbs, adverbs and adjectives, while kana are mainly used for inflectional endings, conjunctions and postpositions.
Kana are the simple syllabograms with unique
phonetic readings. Kanji are the structurally
complex morphograms introduced from
China, which have several phonetic readings.
All kanji can be represented by kana. Japanese
can be written exclusively with 104 kana, which
are learnt in the first year of primary school. On
the other hand, the 2000 commonly used kanji
should be learnt by the end of the ninth grade.
It has been suggested that the processing of
kanji and kana involves a different intrahemispheric mechanism, as judged mainly through
studies on patients with alexia and agraphia.'"
We report a patient with thrombosis of the left
Labbe vein who developed pure agraphia of
kanji and compare it with agraphia in IndoEuropean languages.
Case report
A 56 year old right handed man, educated
through high school, was admitted to another
hospital for convulsion and disturbed consciousness following dysarthria and slight right

hemiparesis lasting for a few hours on 24
August 1984. He soon recovered consciousness, and his dysarthria and right hemiparesis
improved during the next four days. On 7
September 1984, he was admitted to our hospital having developed agraphia of kanji for the
first time. On admission, neurological examination disclosed slightly increased deep tendon
reflexes on the right side with an extensor
plantar reflex. Consciousness, orientation,
cranial nerve functions, including ocular
movements and visual fields, muscle strength,
sensation and gait were normal. Neuropsychological examinations disclosed slightly disturbed immediate ability to recall, name and
read kanji and kana but with severely disturbed
ability to write kanji. Brain computed tomography (CT) on 7 September 1984, disclosed 'a
high density area around the posterior portion
of the left temporal lobe and angular gyrus (fig
1A). In three weeks, neuropsychological
problems other than the writing of kanji
improved rapidly when the haematoma was
absorbed leaving a low density area in the
posterior portion of the left temporal lobe on
brain CT (fig IB). Cerebral angiography on 22
October, revealed a filling defect of contrast
medium in the portion in which the left Labbe
vein descends between the junction of the
transverse and sigmoid sinuses (fig 2).
He had a second attack of consciousness
disturbance and right hemiparesis, which
improved in a few days in March 1987, and he
was admitted to our hospital for a second
cerebral angiography in June 1987. The
neurological findings were normal. The
neuropsychological findings were as follows:
his Wechsler Adult Intelligence Scale was
average (Verbal IQ = 118, Performance IQ
= 114, Full Scale = 117). Neither hemineglect (bisection of a line, Albert's line-crossing test and copying the picture of a flower) nor
constructional apraxia were noted. Limb and
facial praxis to command, facial recognition,
left-right discrimination and finger naming
were intact. His spontaneous speech was fluent
and well articulated. On spontaneous writing
he used more kana instead of kanji than would
be expected from his intelligence level (fig 3).
His speech characteristics were normal on the
Western Aphasia Battery (WAB).' He made
no mistakes in the WAB tests of spontaneous
speech, auditory comprehension, repetition
and naming. In reading tasks the scores were
maximum except for the following two sections: in Japanese-translated WAB, the
"spelled word recognition" and "spelling"
sections were changed for kanji and kana.

336

Yokota, Ishiai, Furukawa, Tsukagoshi

Figure)I CT scan (1 4
days after the first attack)
(A), axial view. A high
density area with a vague
margin was detected
around the angular gyrus
(top) and left posterior
temporal area (bottom).
Two months after the
attack (B), a low density
area was recognised in the
left posterior temporal
area (arrow) without
abnormality in the angular
gyrus (top).

Tested kanji letters consisted of two elements,
which had their own meanings and phonetic
readings. "Spelled word recognition" section
was changed to the recognition of kanji by
hearing the structure of them, for example, the
names of the elements and their positional

..j :. .

*

,
/
a

:.

, -

Az,+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
...

Figure 2 Angiography of the left carotid artery, venous phase, lateral view. Afilling
defect of contrast medium (arrow) was recognised in the Labbe vein (L) falling into the
junction between transverse ( T) and sigmoid (S) sinuses.

relation between the elements. The "spelling"
section was changed to the test of explaining
the structure of kanji. In these two sections the
scores were low (50o/, each). In writing tasks
the scores for writing kanji was low (50-62 5 % )
except for copying kanji. On the other hand,
the scores for writing kana and numbers were
maximum. We performed special examinations
for reading and writing kanji and kana. He was
asked to read and write on dictation 218 kanji
and 104 kana, which he had learnt in the first
two years of primary school. He could not write
82 kanji out of 218 trials (37 6%). He tended to
fail kanji characters which were more complex
and had been learnt later. There was neither
partial omissions, simplification nor confusion
with other characters. On the other hand, he
made no mistakes in reading kanji or in reading
and writing kana letters. In contrast to the
prompt responses in writing kana, and reading
kana and kanji, he often took some time to write
kanji, even when he eventually proved able to
write them. When he could not write some
kanji, he reported that he knew but could not
recall them. In fact, he was able to write

337

Pure agraphia of kanji due to thrombosis of the Labbi vein

Discussion
Writing is a very complex function involving
linguistic, acoustic, praxic, kinesthetic, visual
sv
and
motor components. Chedru and Gesch-7.
,-o_
wind6 first indicated that writing was first
II
disturbed in general brain dysfunction. It has
1:
9
been suggested that agraphia was of no clinical
value in the localisation of brain damage.7
3
*1
However, recently, much evidence has
accumulated indicating that isolated writing
5
disturbances do occur in cases of localised brain
*
damage, which could not be explained by the
'/
1)
-1,
disturbance of any other higher functions.
('l
-t
Three specific areas for pure agraphia have
been reported: the left posterior frontal lobe,8 9
4i~
the left superior parietal lobe'"'2 and the left
i'~)
temporal lobe.""'6.
A?
d
Our case showed no disturbance of cons4
ciousness or intelligence, or aphasic or apraxic
errors in the chronic phase even after a
k^J l *tt thorough examination. The pure agraphia in
.
5.0
it
our case was therefore considered to be a focal
cerebral sign. The lesion was clearly demonstrated in the left temporal lobe on CT and
MRI, which was probably due to the thrombosis of the Labbe vein.
Mochizuki and Ohtomo'7 reported a case of
Figure 3 Spontaneous writing. He was requested to use kanji as much as po ssible. He
wrote the corresponding kana (circle) instead of kanji, which were learnt in* primary
infarction of the left occipital lobe and inferior
school. On the other hand, kana were written correctly. There were no parag raphSerrors.
temporal gyrus, who initially had the difficulty
t
The meanings of the sentences are; "On 17 November I went to Shinjuku and visited the
in kanji reading and writing and kana writing,
I
dropped
into
my
in
and
on
the
way
home
grave
a temple with my wife,
chihd's house,
and I arrived at Seibu-Shinjuku station, and Ifelt ill when I got on the trai zn,and I was but later kanji reading improved with little
brought to Tokyo Medical and Dental University Hospital by ambulance."
improvement in kanji writing. They proposed
that the left inferior temporal lobe might be
complex kanji immediately after Iglancing at indispensable for writing kanji. Recently Soma
them, but could not recall them spo ntaneously. et al" first described "pure agraphia of kanji"
When he wrote kanji and kana w,ith his left in three cases with haematoma, infarction and
trauma. All these cases had the lesion in the left
hand, the results did not change maLrkedly.
Magnetic resonance imaging (MR{I) revealed posterior temporal area probably extending to
the lesions in the posterior part of le ft temporal the angular gyrus on CT. Although they had
cortex (fig 4A) and the adjacent white matter amnesic aphasia, alexia and agraphia in the
(fig 4B) with dilatation of the inferior horn of acute phase, disturbances other than agraphia
the left lateral ventricle. There wer4 e no abnor- of kanji disappeared in a few months. The
malities in the left angular gyrus. Oni the second clinical course, symptoms and lesion of our case
cerebral angiography, the same findling, that is, was very similar to those of Soma's cases. Our
the filling defect of the left Labb4 e vein, was case, however, clearly demonstrated that
amnesic aphasia and alexia, associated with
seen as before.
'It

f-

0

,)
-I

0

le

Figure 4 MRI (3 months
after the second attack),
T2-weighted image
(TR = 2500 ms,
TE = 105 ms). (A)
Axial view. Low signal
area in the posterior

portion of left middle
temporal lobe (arrow).
(B) Frontal view. High
signal area in the white
matter of left temporal
lobe (arrow head).

12

Yokota, Ishiai, Furukawa, Tsukagoshi

338

agraphia of kanji in the acute phase, was due to
the lesion in the left angular gyrus, and that the
remaining symptom of pure agraphia of kanji
was caused by the left posterior temporal
lesion. The haematoma, seen around the posterior temporal area and angular gyrus in the
acute phase, was absorbed leaving the localised
lesion in the posterior temporal area with no
abnormality in the angular gyrus on CT and
MRI.
In reviewing reported Japanese cases with
alexia and agraphia, Iwata' proposed the following pathway for the writing of kanji: Wernicke's area(W)-posterior inferior temporal

acute phase, two cases2223 had sensory aphasia

and the other2" had motor aphasia. However,
there was no detailed description about agraphia in the reports. Cambria's case2' had carbon
monoxide poisoning, and Komiyama's22 case
was deficient in anti-thrombin III. In our
report and that of Kawase23 the cause of the
thrombosis could not be determined. All the
reported cases including ours were accompanied by lobar subcortical haemorrhages, which
were located around the left posterior temporal
lobe and angular gyrus. Therefore, in such
situations, it is necessary to consider Labbe
vein thrombosis.

area(T)-occipital lobe(O)-angular gyrus(A)anterior speech area of Broca and motor
association area of hand(M). This is the pathway involved in selecting the correct kanji We thank Drs Ryoko Okusa, Yoshinori Ito and Kyoko
graphemes according to the meaning of the Eguchi for their help.
words, and thus recalled visual engrams of
letters are sent to the angular gyrus which
mediates the transformation of them into the
1 Iwata M. Kanji versus Kana. Neuropsychological correlates
of the Japanese writing system. Trans Neurosci 1984;
motor engrams of letters. On the other hand,
7:290-3.
was
postulated 2 Oka H, Asano T, Hattori S, Mochio S, Sugishita M. Pure
the pathway for reading of kanji
alexia and agraphia caused by the left inferior temporal
to be O-T-W, for the writing of kana, W-A-M,
and fusiform gyri. Neurol Med (Tokyo) 1985;23:73-6.
and for reading kana, O-A-W. Our case sugges3 Kawahata N, Tagawa K, Hirata Y, Nagata K, Shishido F. A
case of alexia with agraphia due to the infarction in the left
ted that the process of writing kanji is different
inferior-posterior temporal lobe. Clin Neurol (Tokyo)
from that of reading kanji or kana, or writing
1987;27:420-7.
4 Kawamura M, Hirayama K, Hasegawa K, Takahashi N,
kana. According to Iwata's hypothesis, in our
Yamaura A. Alexia with agraphia of Kanji (Japanese
case, the writing pathway for the kanji from W
morphograms). J Neurol Neurosurg Psychiatry 1987;
50:1125-9.
to T was selectively disturbed by the lesion in
5 Kertesz A. The Western Aphasia Battery. (Translated into
the
white
reading
the temporal
matter, sparing
Japanese, Sugishita Mj et al, Tokyo: Igaku-Shoin, 1986)
New York: Grune and Stratton, 1982.
pathway for kanji from T to W.
6 Chedru F, Geschwind N. Writing disturbances in acute
In Indo-European languages two different
confusional states. Neuropsychologia 1972;10:343-53.
A: The agraphias. In: Vinken PJ and Bruyn GW
agraphias, lexical and phonological, have 7 Leischner
(eds): Handbook of Clinical Neurology, Amsterdam, 1969;
been discussed.'6 1820 Beavois and Derouesne'9
4:141-80.
W, Roberts L. Speech and brain mechanisms,
argued that writing a word can be performed 8 Penfield
Princeton: Princeton University Press, 1959.
either by using a direct lexical method of 9 Dubois J, Hecaen H, Marcie
P. L'agraphie "pure". Neuropsychologia 1969;7:271-86.
obtaining its spelling (the "lexical route") or by 10 Russel
R, Espir MLE. Traumatic aphasia. Oxford: Oxford
sounding-out the word into phonemes and
University Press, 1961. DB.
Agraphia selective for
M, Rosenfield
employing phoneme-grapheme transforma- 11 Kinsboume
written spelling: an experimental case study. Brain Lang
tions. Lexical agraphia consists of a specific
1974;l:215-26.
SH, Alexander MP. Pure agraphia and unilateral
difficulty in writing those words which use 12 Auerbachataxia
associated with a left superior parietal lobule
optic
any irregular or ambiguous phonemelesion. J Neurol Neurosurg Psychiatry 1981;44:430-2.
Ein Fall von reiner motorischer Agraphie. Arch
grapheme transformations, while the writing 13 Boettiger A. Nervenkr
1922;65:87-103.
Psychiatr
of regular words and nonsense syllables is 14 Rosati
P. Pure agraphia: a discrete form of
Bastiani
G,
Neurol
J
Psychiatry 1979;42:266-9.
Neurosurg
agraphia.
preserved. Kana are the simple syllabograms 15 Soma
A, Sugishita M, Maruyama S, Kitamura K, Tsubaki
with unique phonetic reading, and the words
T. "Pure agraphia of Kanji" associated with the left
posterio-inferior temporal lesions. Neurol Med (Tokyo)
using kana are orthographically regular. On the
1988;29: 172-8.
other hand, kanji are the structurally complex 16 Croisile
B: Agraphies pures, agraphies parietales, agraphia
lexicales? A propos de trois observations. These de
morphograms with several phonetic readings.
Medicine, Lyon, 1987.
Kanji cannot be read aloud without knowing 17 Mochizuki
H, Ohtomo R. Pure alexia in Japanese and
agraphia without alexia in Kanji. Arch Neurol 1988;
how they should be pronounced, even when
45:1157-9.
their meanings are understood. Therefore, 18 Shallice
T. Phonological agraphia and the lexical route in
writing. Brain 1981;104:413-29.
in
agraphia of kanji is similar to lexical agraphia
M-F, Derouesne J. Lexical or orthographic
Indo-European languages in that the writing 19 Beauvois
agraphia. Brain 1981;104:21-49.
20
Roelgen DP, Sevush S, Heilman KM. Phonological
system with a poor or irregular phonemeagraphia: Writing by the lexical-semantic route.
The
is
transformations
impaired.
grapheme
Neurology 1983;33:755-65.
21 Cambria S. Infarctus cerebral hemorragique par thrombose
case reported by Croisile'6 had the lesion in the
de la veine de Labbe. Rev Neurol 1980;136:321-6.
left posterior temporal area, which was similar 22 Komiyama
A, Hirayama K, Watanabe Y. Cerebral cortical
vein thrombosis presenting with lobar hemorrhage in a
to those in our case and those reported by
with
familial antithrombin III deficiency. Clin
patient
Soma.
Neurol (Tokyo) 1988;28:493-6.
So far as cerebral vein thrombosis is concer- 23 Kawase T, Tazawa T, Mizukami M.andCerebral venous
operative findthrombosis-Computed tomography
ned, only three cases with thrombosis of the
ings in 4 cases of simulated subcortical hemorrhage. CT
Study (Tokyo) 1983;5:19-27.
Labbe vein2"23 have been reported. In their