© 1987 S. Karger AG, Basel
0014-3022/87/0272-0065$2,75/0

Eur. Neurol. 27: 65-71 (1987)

Alexia without Agraphia or Hemianopia
Paolo Caffarra
Department of Neurology, Fidenza/Parma, Italy

KeyWords. Pure alexia • Reading • Nuclear magnetic resonance correlate • Cerebral
haematoma
Abstract. The case of a right-handed man with pure alexia without hemianopia due to
intraparenchymal haematoma is reported. Ct scan and nuclear magnetic resonance showed a
lesion in the lateral temporo-occipital area confirming the assumption that the anatomic
pathways responsible for reading run close to the lateral temporo-occipital or inferior tem­
poral gyrus.

Alexia without agraphia or pure word
blindness was first described by D6jerine [8]
in 1892 as a result of lesions disconnecting
the pathways between the occipital lobes and
the temporoparietal association areas. Sev­
eral other cases were reported later but it was
Geschwind [10] who stressed the importance
of disconnection to explain this syndrome.
Based on this assumption, any lesions
preventing bilateral visual stimuli from
reaching the left angular gyrus might pro­
duce alexia without agraphia. The clinical
features of the syndrome generally include
right-sided hemianopia but in principle no
visual defects should be expected as a conse­
quence of lesions sparing the optic tracts and
radiations, the primary visual cortex, and
part of the visual association areas.
Several cases of pure alexia without hem­
ianopia have been reported [1,2, 12-14, 16,
18, 20, 21]. However, Luhdorf and Paulson

[19] have questioned whether hemianopia
was actually absent in some of these cases.
This study reports the case of a patient
with pure alexia without hemianopia or co­
lour-naming defects due to a small intrapar­
enchymal haematoma verified by CT scan
and nuclear magnetic resonance (NMR).

Case Report
M.R., a 48-year-old right-handed man, was hospi­
talized in August 1984 because of progressive frontal
headache lasting 3 days and sudden reading difficul­
ties. The physical and neurological examinations
upon admission were normal. In particular, no visual
field defects were observed (fig. 1). The patient re­
ported only great difficulties in reading (‘I can see, 1
can write, but I can’t read whole words’) and in pass­
ing from one line to the next. His abilities in conver­
sational speech, language comprehension, sponta­
neous and dictation writing were intact (fig. 2). In
addition, the patient was able to read numbers and
letters accurately and use a letter-by-letter decoding
strategy to understand short words.

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

Introduction

Three weeks after the stroke the patient was mod­
erately dyslexic and with some effort was able to read
long words and non-words. His reading disability sub­
sided gradually leaving him with only a mild deficit
10 months later. (The patient complained o f ‘slower
reading’ than before the stroke.)
In order to investigate possible differences in
reading between the two visual fields, a tachistoscopic
examination was taken 3 weeks after the stroke. The
results arc reported in table II. Stimuli were presented
at different exposure times on different days and
appeared at random 3.5-4° to the left and right of the
fixation point. As shown in table II, the patient exhib­
ited great difficulties in reading trisyllables and bisyl­
lables in both visual fields and no significant differ­
ences were found between them.

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

A standard neuropsychological testing battery (ta­
ble I) taken 3 days after the stroke showed no evi­
dence of aphasia, apraxia, agnosia, visual extinction
or neglect. Score was 103 on verbal WAIS and 104 on
performance WAIS. Total IQ was 104. The Wechsler
Memory Quotient (WMQ) was 86, i.e. below his gen­
eral intelligence level, and a slight impairment was
apparent on verbal memory subtests. His delayed ver­
bal retrieval (after 30 min) was 60% as against 100%
for non-verbal memorandum. Furthermore, the pa­
tient showed no difficulties in colour identification
and discrimination (panel D-15 test) [9]. He could
name colours or recognize colours when given their
name. In the tachistoscopic colour-matching task he
perceived the colours in the right hemifield slightly
paler than in the opposite half.

Fig. 2. Spontaneous handwriting sample.

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

Pure Alexia
67

68

CaiTarra

Fig. 3. CT scan taken 3 days after onset of symptoms, showing a left occipital haematoma with mild
perifocal oedema.
Fig. 4. CT scan taken 17 days after onset of symptoms, showing a reduction in size of the haematoma
located in the left paraventricular portion of the white matter with sparing of the corpus callosum and angular
gyrus.

WA1S verbal
WAIS performance
WAIS total
WMQ
Logic memory (delayed recall)
Non-verbal memory (delayed recall)
Token test
Verbal fluency
Graphaesthesia (for single letters on
the back and palm of both hand)
Graphaesthesia (for bisyllabic abstract
and concrete words)
On the back
On the left hand
On the right hand
Object naming
1 Wechsler Memory Scale.

103
104
104
86'
60%'
100%'
34/36
24/36
100%
correct

10/10
10/10
9/10
20/20

Table il. Tachistoscopic tests
Presen­
tation
time, ms
Letter naming
Colour naming
Colour matching
Matching of
non-verbal pictures
Reading of numbers
Reading of trisyllables
Reading of bisyllables

Left

30
25/25
100
12/12
100
12/12
50 same 8/10
different 10/10
12/12
50
50
6/10
1/14
200

Right

25/25
12/12
12/12
10/10
8/10
12/12
4/10
0/14

The results of the tachistoscopic tests are ex­
pressed as correct responses. For each test several
attempts with different target stimuli at different ex­
posure times were made in order to obtain the best
approximate performance with the lowest presenta­
tion time.

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

Table I. Standard neuropsychological examina­
tion

Laboratory Findings and Anatomic Localization
Complete blood count, serum electrolytes and
blood chemistry were all in the normal range. The
ECG and EEG revealed no abnormalities. Skull Xrays were also normal.
A CT scan (fig. 3) taken 3 days after the stroke

showed a left occipital intraparenchymal haematoma.
Left carotid and vertebral angiography revealed a
small avascular mass in the left paricio-occipital area
and no vascular malformations. A repeat CT scan
taken 14 days after the first one (fig. 4) confirmed the
previous findings and showed a reduction of the hae-

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

Fig. 5. MRI. Transverse section at three different levels: orbital plane (a); upper thalamus (b); bodies of the
lateral ventricles (c). Cortical-subcortical lesion in the left lateral temporo-occipital junction without involve­
ment of the corpus callosum, forceps major, angular gyrus and calcarine cortex, also evidenced in the left
parasagittal view (d). (e, f) Coronal view (anteroposterior sequence): lacunar area involving the left lateral
temporo-occipital gyrus.

70

Caffarra

taken in some other cases reported in the lit­
erature and have questioned whether hemia­
nopsia was truly absent in those cases. In our
patient visual perimetry performed 5 days
after the stroke was normal and serial bed­
side visual field examinations taken during
the acute stage showed no abnormalities.
Though infrequently, normal colour percep­
tion or naming may actually be present in
about 30% of the cases [11] with pure word
blindness. As suggested by Damasio and Da­
masio [7], colour anomia probably appears
under the 3 following conditions: (a) in­
volvement of the left lingual gyrus; (b) dam­
age to the left hippocampal region, and, (c)
Discussion
right-sided hemianopia. Neurological exami­
According to the anatomical classifica­ nation demonstrated that our patient had no
tion proposed by Greenblatt [15], alexia lesions in those areas. Furthermore, it has
without agraphia can be (a) splenio-occipital been postulated that the absence of colour
with or without hemianopia; (b) lateral or anomia may be due to the fact that the left
medial occipital, and (c) subangular with or dorsal fibres of the calcarine regions [2, 13]
without hemianopia. Based on the CT scan or even the dorsal fibres of the splenium are
and MRI findings, our case resembles more intact, which might actually be the case in
closely the lateral occipital form with sparing our patient. In addition, the preserved abil­
of the splenium fibres. This hypothesis ity to name letters or bisyllabic words by graseems to be confirmed by the absence of phaesthesic stimulation in both hands and in
the back indicates that the anterior part of
‘neighbourhood’ parietal signs.
Following the classification introduced the corpus callosum is also intact.
Based on some clinical observations and
by Damasio and Damsio [7], this case could
be a variety of type III, found in patients a review of the literature, Geschwind [10]
with intact visual field, without colour suggested that some anatomic pathways
anomia, and with or without hemiachroma- from the non-dominant hemisphere might
topsia. In this particular type, the lesion is be involved in reading processes. A possible
located in the temporo-occipital area and mechanism would be for the information to
includes the paraventricular portion of the proceed from the right Brodmann area 17 to
white matter. The integrity of the visual field the ipsilateral visual association cortex and
is uncommon in pure word blindness and to from there to the right angular gyrus and
date only 7 cases have been reported in then across the corpus callosum to the left
which anatomical classification is possible angular gyrus. A second mechanism would
[2, 3, 5, 13, 14, 17, 21], However, Luhdorf include fibres running from areas 18 and 19
and Paulson [19] have pointed out that a on the right to areas 18 and 19 on the left and
complete visual field examination was not from there to the left angular gyrus. Finally,

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

matoma, which now appeared to be located in the left
paraventricular portion of the white matter, sparing
the splenium of the corpus callosum and the left angu­
lar gyrus.
Ten months later, during a routine neurological
follow-up examination, the patient was found to read
normally and was submitted to magnetic resonance
imaging (MR1) in order to better determine the size of
the lesion and its relationship with the surrounding
structures. The MR1 (fig. 5) revealed a cortical and
subcortical posthaemorrhagic lacuna in the posterior
part of the lateral temporo-occipital gyrus. The splcnium of the corpus callosum, forceps major, calcarine
cortex, angular and parahippocampal gyri, thalamus
and right hemisphere were intact.

71

Pure Alexia

Acknowledgments
Our special thanks are due to Prof. P. Bassi and
Prof. A. Mazzucchi for their precious suggestions and
valuable technical assistance.

References
1 Adler, A.: Disintegration and restoration of optic
recognition in visual agnosia. Archs Neurol. Psychiat. 57: 243-259 (1944).
2 Ajax, E.: Dyslexia without agraphia. Archs Neu­
rol. 77: 645-652 (1967).
3 Assal, G.; Hadj-Djilani, M.: Une nouvelle obser­
vation d’alexie pure sans hémianopsic. Cortex 12:
169-174 (1976).
4 Benson, D.F.: Alexia; in Handbook of clinical
neurology, vol. 45/1 (Elsevier, 1985).
5 Castro-Caldas, A.; Salgado, V.: Right hemifield
alexia without hemianopia. Archs Neurol. 41: 8487 (1984).
6 Cohen, D.; Salanga, V.; Hully, W.; Steinberg. M.;
Hardy, R.:Alcxia without agraphia. Neurology 26:
455-459 (1976).

7 Damasio, A.R.; Damasio. H.: The anatomic basis
of pure alexia. Neurology 33: 1573-1583 (1983).
8 Déjerine, J.: Contribution à l’étude anatomo-pa­
thologique et clinique des différentes variétés de
cécité verbale. C. r. Séanc. Soc. Biol. 4: 61-90
(1892).
9 Farnsworth, D.; The Farnsworth dichotomous
test for colour blindness panel D-I5 manual (Psy­
chol. Corporation, New York 1947).
10 Geschwind. N.: Disconnection syndrome in ani­
mals and man. Part 11. Brain 88: 585-644
(1965).
11 Gloning, I.; Gloning, K.; Hoff, H.: Neuropsycho­
logical symptoms and syndromes in lesions of the
occipital lobe and the adjacent areas, pp. 75-81
(Gauthier-Villar, Paris 1968).
12 Goldstein, M.; Joynt, R.; Goldblatt. D.: Word
blindness with intact central visual fields. Neurol­
ogy 123: 239-291 (1971).
13 Grccnblatt. S.H.: Alexia without agraphia or hem­
ianopia. Brain 96: 307-316 (1973).
14 Grccnblatt. S.H.: Subangular alexia without agra­
phia or hemianopia. Brain Lang. 3:229-245(1976).
15 Greenblatt. S.H.: Localization of lesions in alexia;
in Kertesz, Localization in neuropsychology (Aca­
demic Press, New York 1983).
16 Heilman, K.M.; Safran. A.; Geschwind, N.:
Closed head trauma and aphasia. J. Neurol. Neurosurg. Psychiat. 34: 265-269 (1971).
17 Henderson, V.W.; Friedman, R.B.; Teng, E.L.;
Weiner, J.M.; Left hemisphere pathways in read­
ing: inferences from pure alexia without hemiano­
pia. Neurology 35: 962-968 (1985).
18 Hoff. H.; Potzl. O.: Reine Wortblindheit beim
Hirntumor. Nervenarzt 73: 385-393 (1937).
19 Luhdorf, K.; Paulson, O.B.: Does alexia without
agraphia always include hemianopsia? Acta neur.
scand. 55: 323-329 (1977).
20 Peron, N.; Goutner, V.: Alexie pure sans hémia­
nopsic. Revue ncurol. 76: 81-82 (1944).
21 Vincent, F.M.; Sadowsky, C.H. Saunders, R.L.;
Reeves, A.G.: Alexia without agraphia, hemiano­
pia, or color-naming defects: a disconnection syn­
drome. Neurology 27: 689-691 (1977).
Received: July 15, 1986
Accepted: November 19, 1986
Paolo Caffarra. MD, Department of Neurology,
1-43036 Fidenza/Parma (Italy)

Downloaded by:
Kings's College London
137.73.144.138 - 10/24/2017 11:03:02 PM

Geschwind did not mention the non-homo­
topic interhemispheric connections which
would transfer visual information from area
18 on the right to the right-sided association
cortex and then to the left angular gyrus.
This last mechanism is indeed possible but
has not yet been demonstrated anatomically
[4], Pure alexia resulting from lesions of the
left occipital lobe is usually transient in na­
ture and Geschwind therefore concluded
that the first two mechanisms were probably
present and functionally independent of
each other.
As regards functional anatomy, our case
and that reported by Henderson et al. [17]
seem to confirm this assumption and suggest
that the visual connections responsible for
reading run close to the lateral temporo-occipital gyrus or the inferior temporal gyrus.