still be worse for patients with large aneurysms
(RR ⫽ 1.34; 95% CI 0.94 to 1.90). We found one other
study on aneurysm size and outcome; in that study
no relation was found.9 Two factors may have introduced a bias toward patients with good outcome in
that study. First, the study was performed in a tertiary care center; secondly, aneurysm size was assessed on conventional angiograms, and thus not in
patients who are not candidates for operation because of a poor condition. Yet, in that study, patients
with large aneurysms had a longer period of unconsciousness after the hemorrhage and patients with
small aneurysms had a greater chance of complete
recovery. The data of this previous study are therefore in agreement with our data.
In our series, patients with large aneurysms were
less frequently operated. This implies that the poor
outcome in these patients might partly be explained
by an increase in frequency of rehemorrhage, but
this was not the case (see table 2). The larger proportion of patients with large aneurysms who have not
been operated can be explained by the worse clinical
condition from the outset, and is therefore a consequence and not so much a causal factor for the poor
outcome.

Apraxic agraphia due
to thalamic infarction

Acknowledgment
The authors thank Prof. J. van Gijn for critical review of this
manuscript.

References
1. Hop JW, Rinkel GJE, Algra A, van Gijn J. Case fatality rates
and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke 1997;28:660 – 664.
2. van Crevel H, Habbema JDF, Braakman R. Decision analysis
of the management of incidental intracranial aneurysms. Neurology 1986;36:1335–1339.
3. Rinkel GJE, Djibuti M, Algra A, van Gijn J. Prevalence and
risk of rupture of intracranial aneurysms. Stroke 1998;29:251–
256.
4. Raaymakers TWM, Rinkel GJE, Limburg M, Algra A. Mortality and morbidity from surgery of unruptured aneurysms: a
systematic review. Stroke 1998;29:1531–1535.
5. Vieco PT, Shuman WP, Alsofrom GF, Gross CE. Detection of
circle of Willis aneurysms in patients with acute subarachnoid
hemorrhage: a comparison of CT angiography and digital substraction angiography. Am J Radiol 1995;165:425– 430.
6. Drake CG. Report of World Federation on Neurological Surgeons committee on a universal subarachnoid hemorrhage
grading scale. J Neurosurg 1988;68:985–986.
7. Hijdra A, Brouwers PJAM, Vermeulen M, van Gijn J. Grading
amount of blood on computed tomograms after subarachnoid
hemorrhage. Stroke 1990;21:1156 –1161.
8. Hop JW, Rinkel GJE, Algra A, van Gijn J. Initial loss of consciousness and risk of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. Stroke 1999;30:2268 –2271.
9. Rosenørn J, Eskesen V. Patients with ruptured intracranial
saccular aneurysms: clinical features according to size. Br J
Neurosurg 1994;8:73–78.

Article abstract—The authors report a patient of pure apraxic agraphia
with normal praxis due to left thalamic infarction. 15O-gas-PET showed reduced oxygen metabolism in the left thalamus and the left dorsolateral premotor area, while MRI and 11C-fulumazenil-PET showed no remarkable
lesions in the frontal cortex. The patient’s word imaging remained normal.
The authors hypothesize that thalamic destruction causes pure apraxic
agraphia by exerting a remote effect on left dorsolateral premotor area and
blocking somewhere between graphemic area and motor programming. Key
words: Apraxic agraphia—Left thalamic infarction—Dorsolateral premotor—PET.
NEUROLOGY 2000;54:2336–2339

T. Ohno, MD; M. Bando, MD; H. Nagura, MD; K. Ishii, MD; and H. Yamanouchi, MD

Apraxic agraphia with normal praxis, or pure
apraxic agraphia, is a selective impairment of writing characterized by deteriorated orthographic production with otherwise normal praxis, sensory motor
functions, and preserved oral spelling and typing.1

From the Department of Neurology, Tokyo Metropolitan Geriatric Hospital
(Drs. Ohno, Bando, Nagura, and Yamanouchi); and the Positron Medical
center, Tokyo Metropolitan Institute of Gerontology (Dr. Ishii), Tokyo,
Japan.
Received September 27, 1999. Accepted in final form March 17, 2000.
Address correspondence and reprint requests to Dr. T. Ohno, Department of
Neurology, Graduate School of Medicine, The University of Tokyo, 7-3-1
Hongo, Bunkyo-ku, Tokyo 113-8655, Japan; e-mail: ttakae-tky@umin.ac.jp
2336

Copyright © 2000 by the American Academy of Neurology

Neuroanatomic investigations of the syndrome have
been few. Some reports have implicated lesions in
the left parietal lobe,1-3 and others the frontal lobe.3-5
Most of the reported cases have been accompanied by
aphasia, visuospacial defects, or apraxia.
Thalamic agraphia has been reported seldomly,
and its exact location and pathomechanism are unknown.6,7 One study favored the centromedian nucleus, another the anterior portion of the left
thalamus, and both suggested the possibility of secondary impairment in the frontal or parietal cortex
by a remote effect.
We report a patient presenting with pure apraxic
agraphia after an infarction restricted to the left dor-

somedial thalamic nucleus. Details of the pathomechanism and clinical pattern of his apraxic
agraphia are discussed.
Patient presentation. A 78-year-old right-handed man
was hospitalized after the sudden onset of a writing disorder. Since graduating from business school, he had worked
in an office for more than 50 years. He had cervical spondylosis, which disturbed his walking, but was otherwise
healthy. On December 28, 1998, the patient suddenly could
not write any characters or numbers and was referred to
our hospital. Physical examination results were normal.
The neurologic examination showed bilateral hyperreflexia
with severe spasticity of the lower extremities due to cervical spondylosis and pure agraphia in both hands. The patient was alert and his hands had no sensory-motor
deficits. His verbal IQ was 91, performance IQ 71, and he
had a total IQ score of 80 on the Wechsler Adult Intelligence Scale Revised. His score on Raven’s Colored Progressive Matrices was 18 of 36. He had no aphasia, alexia,
apraxia, neglect, amnesia or constructional disturbance.
His scores on Wisconsin card sorting test for 1st and 2nd
trial were: total correct 48 of 128, 84 of 128, respectively;
category completed 2, 5; perseverartive responses 35, 25.
MRI showed a left thalamic infarction in dorsomedial nucleus (figure 1). Intensity change and swelling were
present in the bilateral splenium but disappeared within a
few weeks. The lesion was not enhanced by gadolinium.
15
O-gas-PET showed significant oxygen hypometabolism
in the left dorsolateral premotor area as well as the left
thalamus (figure 2). His cerebral metabolic rate of oxygen
(CMRO2) was significantly low in the left frontal lobe as
compared to the control ⫾ 2SD (2.22 and 3.16 ⫾ 0.36). The
blood flow and oxygen metabolism were matched in the left
frontal cortex and there were no evidence of ischemia in
the left middle cerebral arterial territory. No focal defect
was found with 11C-flumazenil-PET (FMZ-PET).
Language evaluation. The Western Aphasia Battery
was performed 14 days after onset. The patient’s spontaneous speech, auditory comprehension, repetition, naming,
reading, praxis, and construction were perfect (10 points
all subjects). In contrast, he could not write kanji, kana,
the Roman alphabet, or Arabic numerals (0.9 points for
writing) with either hand. Scores on other naming tests
were also perfect. His scores of kanji, kana, the Roman
alphabet, and numbers were as follows: for reading 20 of

20, 40 of 40, 20 of 20, and 10 of 10; for copying 20 of 20, 40
of 40, 20 of 20, and 10 of 10; and for dictation 3 of 20, 4 of
40, 8 of 20, and 4 of 10. When the patient would recognize
his errors he would stops writing. The majority of errors
involved the partial omission or addition of characters.
Scrawling, no reaction, neographism, and complete substitution were not seen. He was able to spell kanji orally and
make sentences by pointing at kana in the Japanese syllabary. He could orally describe correctly the orientation of
each stroke of a kanji character.
To clarify the mechanism of the patient’s agraphia, we
devised a special task. Ten kanji of various levels of difficulty were selected, and he was asked to write them while
describing the orientation of each stroke of a kanji. His
writing, without the oral guide, was compared with the
results in A-B-B-A order (A: with the oral guide, B: without
it, the scores being 6 of 10, 2 of 10, 2 of 10 and 7 of 10) ( p ⬍
0.01). Interestingly, the patient found it easier to write
with the oral guide.

Discussion. The patient had pure apraxic agraphia
caused by tissue hypoactivity of the left frontal cortex
secondary to the remote effect of a left thalamic infarction. This frontal area, which had reduced oxygen metabolism in 15O-gas-PET, showed no lesion in
MRI or FMZ-PET; this supports the remote effect
hypothesis. Furthermore, the thalamo-cortical connection from the dorsal medial nucleus reaches to
the frontal cortex.8 Pure agraphia by thalamic destruction, however, has rarely been reported.6,7 It
may be that when the lesion is extended, thalamic
dementia appears and the agraphia cannot be detected, or it may be that when the lesion is small, it
can be overlooked.
The lesions of thalamic agraphia are reported to
be the centromedian nucleus, parafascicular nucleus,
dorsal medial nucleus, or anterior nuclei, all of which
belong to either the paramedian thalamic or premammillary artery territories or both.6,7 This is compatible with our patient’s lesion. The possibility of
the secondary effect of thalamic involvement in the
frontal or parietal cortex was discussed in these reports, but the patients had complications such as
alexia, anomia, or mild consciousness disturbance.
To our knowledge, apraxic agraphia due to thalamic
involvement has not been reported.

Figure 1. T1-weighted coronal (A), saggital (B), and axial (C) images of the brain showing a left thalamic infarction. The
lesion is localized in the dorsal medial nucleus.
June (2 of 2) 2000

NEUROLOGY 54

2337

Figure 2. 15O-gas-PET (oxygen metabolism; upper), FMZ-PET (distribution
volume of the benzodiazepin receptor;
lower middle), and T1-weighted axial
(lower). 15O-gas-PET shows significant
hypometabolism in the left dorsolateral
premotor as well as the left thalamus,
while FMZ-PET shows no significant
neuronal loss in the frontal lobe.

Apraxic agraphia with otherwise normal praxis
was first described in 1983,1 but its mechanism and
the anatomic substrate are not clear. This syndrome
is characterized by difficulty in forming written letters despite normal praxis and the preservation of
oral spelling or typing. We call it pure apraxic
agraphia, but most of the reported patients had such
complications as aphasia, visuospatial defects,
apraxia, possible agraphia due to callosal lesion, or
confusion caused by a severe mass effect. True pure
apraxic agraphia, therefore, has rarely been found.
Although previous neuroanatomic findings suggest
two main sites—the left frontal and left parietal
lobes—the definite location has not been determined.
Written spelling agraphia had been classified into
two subgroups.2 In one, the level of visual letter imaging is disturbed9; in the other, the graphemic area
is intact, but there is failure to reach the correct
graphemic motor pattern. The lesion for the former
group was proposed as being in the dominant superior parieto-occipital, and proposed for the latter as
being in the frontal lobe. Our patient’s agraphia can
be classified in the latter. Moreover, the hypoactivated cortical area secondary to thalamic infarc-

tion was in the left dorsolateral premotor area.
Although a transient abnormal lesion was found in
the corpus callosum, we hypothesize that it did not
have an effect on the patient’s agraphia. His
agraphia was for both hands, which negates callosal
agraphia, and his writing disorder remained even
after the lesion disappeared.
While trying to clarify the mechanism of the
apraxic agraphia, we noted a unique phenomenon:
the patient found it easier to write kanji by describing the visual image of the letter structure. This
shows that the graphemic area and the visual letter
imaging are preserved enough to rebuild the letters.
The patient usually wrote without using visual letter
imaging, but he could write by building and copying
a visual letter image. Therefore, there is damage
somewhere between the graphemic area and motor
programming. To clarify this phenomenon, we used
the model of cognitive spelling processes (figure 3).10
The area inside the shadow suggests the site of destruction. The direct route from the visual letter
imaging to handwriting, which is usually unused,
is the theoretical route that explains our patient’s
phenomenon.

Figure 3. A model of cognitive spelling
processes. Dotted lines show the new
proposed route. The shadowed part is
the suggested area of destruction. A direct route from the visual letter imaging
to motor programming is assumed.
2338 NEUROLOGY 54

June (2 of 2) 2000

References
1. Roeltgen D, Heilman KM. Apraxic agraphia in a patient with
normal praxis. Brain Lang 1983;18:35– 46.
2. Friedman RB, Alexander MP. Written spelling agraphia.
Brain Lang 1989;36:503–517.
3. Alexander MP, Fisher RS, Friedman R. Lesion localization in
apraxic agraphia. Arch Neurol 1992;49:246 –251.
4. Coslett HB, Gonzalez Rothi LJ, Valenstein E, Heilman KM.
Dissociation of writing and praxis: two cases in point. Brain
Lang 1986;28:357–369.
5. Hodges JR. Pure apraxic agraphia with recovery after drainage of a left frontal cyst. Cortex 1991;27:469 – 473.

Unilateral pure
thalamic asterixis:
Clinical,
electromyographic,
and topographic
patterns

6. Sugishita M, Ishijima B, Hori T. “Pure” agraphia after CMthalamotomy. Rinsho-shinkeigaku 1973;9:568 –574.
7. Araki S, Kawamura M, Isono O. Reading and writing deficit
in cases with localized infarction of the left anterior thalamus.
No-to-Shinkei 1990;42:65–72.
8. Talairach BJ, Tournoux P. Co-planar stereotaxic atlas of the
human brain. New York: Thieme Medical Publishers,1988:12.
9. Crary MA, Heilman KM. Letter imagery deficits in a case of
pure apraxic agraphia. Brain Lang 1988;34:147–156.
10. Roeltgen D. Agraphia. In: Heilman KM, Valenstein E, eds.
Clinical Neuropsychology, 3rd ed. New York: Oxford University Press, 1993:63– 89.

Article abstract—Eleven patients (nine with infarctions and two with primary hematomas) with isolated thalamic lesions and contralateral asterixis
were examined using a standard electromyographic and neuroimaging protocol. Asterixis was a short-duration phenomenon associated with a hemiataxia
hypesthesia syndrome in all patients. Electromechanical synchronization was
constant for the two silent period types. The anatomic data strongly suggest
that ventral lateral or lateral posterior thalamus are concerned in the pathophysiology of thalamic asterixis. Key words: Asterixis—Negative myoclonus—Thalamus—Electromyography—Stroke.
NEUROLOGY 2000;54:2339–2342

L. Tatu, MD; T. Moulin, MD; V. Martin, MD; G. Monnier, MD; and L. Rumbach, MD

Asterixis is defined as an abrupt, brief, arrhythmic
lapse of tonically contracting muscles.1 It is best seen
in the extended wrist but can also be observed in the
dorsiflexed foot or on the face. Electromyographically, asterixis is characterized by brief periods of
electrical silence in muscles activated to sustain posture, and it is now considered as a form of negative
myoclonus.2,3 Unilateral asterixis is a reliable sign of
structural neurologic disease and is associated with
various focal brain lesions in which the thalamus is
frequently involved.4,5
We report 11 cases of unilateral asterixis related
to an isolated contralateral thalamic vascular lesion.
Electromyographic patterns were studied for all patients. Anatomic and radiologic correlations were
performed to assess the precise lesion topography.
Methods. Between 1990 and 1998, every patient admitted for stroke to the Neurology Department of the Besanĉon University Hospital was systematically examined
for asterixis. The patients, eyes closed, were asked to hold
their arms outstretched with their hands dorsiflexed at the
wrist and their fingers extended and abducted. The patients were also asked to dorsiflex the ankle and to show

From the Fédération des Sciences Neurologiques, Centre Hospitalier Universitaire Jean Minjoz, Besançon, France.
Received October 19, 1999. Accepted in final form February 29, 2000.
Address correspondence and reprint requests to Dr. Laurent Tatu, Service
de Neurologie, CHU Jean Minjoz, 25030 Besançon, France; e-mail:
laurent.tatu@univ-fcomte.fr

their teeth in order to demonstrate a leg or face asterixis.
The duration of asterixis was noted.
Among 35 patients with unilateral asterixis and stroke,
28 had at least one thalamic involvement and 11 (nine
with infarction and two with primary intracerebral hematoma [PIH]) had an isolated thalamic lesion and underwent a
standard evaluation allowing inclusion in this study.
Electromyography (EMG) was performed in all patients
with asterixis. EMG of forearm or finger flexor and extensor muscles and ankle flexor and extensor muscles was
made with surface electrodes. An accelerometer was placed
on the dorsum of the hand or fingers and the dorsum of the
foot. For the face, the surface electrodes were placed over
the levator labii superioris, the orbicularis oris, and thedepressor anguli oris muscles. According to the analysis of
Ugawa et al.,6 EMG silences were classified as silent period type I and type II. In type I, the silent period followed
usual EMG background activities. In type II, the silent
period followed a somewhat large EMG discharge preceded
by a short EMG silence. The characteristics of this synchronization of mechanical effect and silent periods were
studied in all patients.
MRI was performed in all patients using the center-tocenter bicommissural plane as a reference. The site of the
thalamic lesion and, for thalamic infarction, the arterial
territory were systematically assessed.7 Standard EEG
was carried out in all patients but without simultaneous
EEG-EMG studies. Sensitive-evoked potentials were not
measured. All patients were questioned on possible druginduced asterixis. Laboratory studies were performed in
order to rule out metabolic disorders.
Copyright © 2000 by the American Academy of Neurology

2339