0148-396X/79/0501-0001$02.00/0
NEUROSURGERY
Copyright © 1979 by the Congress of Neurological Surgeons

Vol. 5, No. 1, Part 1, 1979
Printed in U.S.A.

Clinical and scientific communications

Anatomical Study of a Posterior Cerebral Lesion Producing
Dyslexia

J. M. Van Buren, M.D.

Department of Neurological Surgery, University of Miami School of Medicine, Miami, Florida

After an “occipital lobectomy” that resulted in a severe dyslexia and a moderate dysgraphia-dyscalculia, anatomical study
showed damage to the posterior extremity of the angular gyrus and degeneration in the posteroinferior pulvinar. This is
in contrast to an earlier case that had degeneration in the anterosuperior pulvinar associated with a small anterior
temporoparietal infarct and a well-documented receptive-expressive aphasia. However, the role of the pulvinar in speech
function remains uncertain. The surgeon should be aware of the short distance between the angular gyrus and both the
midline and the occipital pole because a lesion here during an “occipital lobectomy” produces a distressing and durable

speech impairment.

Key words: Angular gyrus, Dyscalculia, Dysgraphia, Dyslexia, Occipital lobectomy, Pulvinar, Speech

INTRODUCTION

Déjerine recognized four major areas in which lesions cause
clinically distinguishable speech defects: (a) Broca’s region (in
which Déjerine included the posterior part of the 3rd frontal
gyrus and the surrounding area); (b) Wernicke’s region, corre-
sponding to the posterior part of the Ist and 2nd temporal
convolutions and considered as a center for the auditory images
of words; (c) the angular gyrus, considered as the center for
visual images of words; and (d) the subcortical region, in which
lesions cause verbal deafness and pure alexia (5, 6).

Déjerine’s opinions regarding the localization of specific
speech functions were countered by the globalistic doctrines of
Marie (14), but (with Foix (15)) Marie did acknowledge the
predominance of alexia in lesions of the angular gyrus. The
swing toward specific localization of speech disturbances was
continued by Henschen (11, 12) and Kleist (13), reaching an
extreme position with Neilsen’s monograph (18).

Current textbooks show a remarkable diversity in the cortical
regions designated as areas subserving language function.
Many texts restrict the temporoparietal speech representation
to the cortex immediately about the posterior extremity of the
sylvian fissure of the dominant hemisphere, perhaps because
of evidence presented by Penfield and Roberts (22). As a result
of the restricted cortical exposures used for anterior temporal
lobectomy, they found the majority of the aphasic responses to
cortical stimulation in this region. However, aphasic responses
can also be elicited well back to the parieto-occipital junction
(34), conforming to Déjerine’s original outline (Fig. 1).

In the present case of dyslexia with a mild dysgraphia-
dyscalculia, it was possible to define the cortical damage and
thalamic degeneration with some precision. There has been no
previous detailed anatomical study of a lesion lying this far
posteriorly in the dominant hemisphere that resulted in the
disturbance of language.

CASE REPORT
Summary

This 38-year-old woman had an acute hemorrhage about a
metastatic melanoma in the left occipital region. This was
decompressed with amputation of the occipital pole. For 2
weeks she had cortical blindness attributed to transient
compression of the posterior cerebral arteries by acute trans-
tentorial herniation. Later, she had a right hemianopsia with a
trace of visual sparing in the right periphery.

For the remaining 9 months of her life, the patient remained
oriented and spoke without difficulty. She could give a reason-
able description of a picture, named objects without difficulty,
differentiated right from left, and showed no apraxia. She
continued, however, to have difficulty reading. Reading aloud
was laborious and words were often substituted without rec-
ognition of the error. There was little or no comprehension of
what she read either aloud or silently. Writing and calculation
were accomplished with considerable effort, and the results
were much below the level expected from her educational
background.

Report

This right-handed woman was 38 years old at the time of
her death on September 9, 1957. In April 1954, a primary
melanoma was discovered in the left axilla and was resected
with a wide axillary lymph node dissection. In May 1955, there
were multiple cutaneous metastases and metastases to the
lungs. The cutaneous metastases subsequently disappeared
spontaneously. In February 1956, the patient had a convulsion,
nausea, and headache.

On December 18, 1956, examination showed an incomplete
right homonymous hemianopsia with retention of vision of
gross objects in the upper portion of the visual field on the
right. By December 25, 1956, this hemianopsia had become

2 VAN BUREN

complete and there was bilateral early papilledema. There was
a fine nystagmus on extreme lateral gaze to either side. On
December 27, 1956, a left occipital craniotomy was carried out
because of sudden semicoma. The metastatic melanoma and
surrounding blood clot were removed from the occipital lobe.

During the postoperative period the patient showed confu-
sion and right-left disorientation. Although she denied any
difficulty with vision, she was obviously blind to light and
movement. The pupils reacted to light and the extraocular
movements were full. By January 5, 1957, she began to have
some correct visual responses but often confabulated about a
variety of impossible activities while she was actually confined
to bed. By January 9, 1957, vision was 10/30 in both eyes
(chart held 10 feet away over the foot of the bed) but there was
a dense right homonymous hemianopsia.

By March I1, 1957, she was affable but slow to respond. She
was unable to read usefully, substituting letters (e.g., “forever”
was read as “foreven”). She gave a sparse but reasonable
description of a picture, did simple sums orally, named simple
objects, and differentiated right from left.

On June 3, 1957, speech examination was as follows:

Verbal commands. “Close your eyes and open your mouth”
and “Pat your left ear with your hand and open your mouth
three times” (done correctly).

Written commands. “Stick out your tongue and close your
eyes” (read slowly, then reread, but she did it correctly). “Pat
your head three times” (done correctly).

Writing. Wrote name correctly. Correctly wrote “It is a rainy
day” to dictation. When asked to describe what she saw out
the window, she wrote, “I see lots of green trees and the golf
course. There are some houses in the background.” She re-
marked that it was hard work writing. “I never was a writer.”
(She was a college graduate.)

Drawing. A daisy and a clock face were drawn without loss
of symmetry.

Oral arithmetic. The following were done correctly: 12 + 13;
17 + 18; 11 x 12; 27 — 18; 47 — 13; and 3X 7— 1X24 10.

Written arithmetic.

347 When she was asked to check this she changed the
+123 answer to 472.

365
674 67 1
—238 x13 13 3142.0
—136 201 13 (gave up attempt)
67 12

871

Reading. She remarked that “things don’t look right” and
that she could not read for pleasure. When asked to read aloud
she did so very slowly, one word at a time and without
inflection. She replaced words at times, such as “bulk” for
“bull,” and did not recognize the error. Her comprehension of
the paragraph was very scanty, but what was given was correct.

In March, June, and August 1957 she was examined by J.
M. Olhoeft, M.A., speech pathologist and audiologist. The tests
administered were the Halstead-Wepman aphasia screening
st, the Eisenson test, and proverbs and nouns. Her responses
stabilized by the summer of 1957 and examination on August
15, 1957, was reported as follows:

Despite her obvious discomfort, Mrs. S
well and appeared alert.

Testing revealed no demonstrable agnosias, apraxias or left-right
disorientation. Simple arithmetic reasoning, thought problems, con-
tinued to present difficulties but paper and pencil executions of
simple arithmetic problems were completed correctly. She stated
that she no longer has difficulty in finding words and almost no

cooperated

Neurosurgery, Vol. 5, No. 1, Part 1

Fic. |. The major speech areas of the dominant hemisphere as
outlined by Déjerine. The angular gyrus (Pe, pli corbe) is outlined
medially by the interparietal sulcus (unidentified in the present illus-
tration). (4, Wernicke’s area; B, Broca’s area.) Only the posteromedial
margin of the area associated by Déjerine with the “visual images of
words” was affected in the present case. (From J. Déjerine: Semiologie
des affections du systéme nerveux. Paris, Masson et Cie, 1914.)

aphasic errors were observed. Verbal abstractions fluctuated and
appeared to be of the same quality as those given in June. Written
abstractions were not tested.

Silent and oral reading remained difficult for the patient. She is
a slow word by word reader who tends to mis-see or misread pictures
and words. (A button was called a ball. Trees was read “tries” ).

Tangent screen examination on June 6 and July 31, 1957,
with 1- and 10-mm white objects at a l-m distance showed a
right hemianopsia with mild enlargement of the left blind spot.
On perimeter examination (330 mm) the patient reported that
she saw the test object in poorly defined areas 25 to 45 mm to
the right of the fixation point with both eyes. The responses
were not consistent enough to plot. The visual acuity was
normal.

In the spring of 1957 the patient received a course of thiotepa
(Lederle Laboratories, Pearl River, New York), and then she
received methotrexate (Lederle Laboratories), 4 mg/kg, for
three doses at 3-week intervals in the summer of 1957. She died
on September 9, 1957, of widespread metastatic malignant
melanoma and pneumonia.

MATERIAL AND METHODS

The brain was suspended in 10% neutral formalin. After 2
weeks, the pia-arachnoid was removed from the left hemisphere
to clarify the gross cortical morphology. The thalamus and
brain stem were dissected away and embedded in paraffin.
Sections transverse to the longitudinal axis of the hemispheres
disclosed a 3- x 4- x 5-cm metastatic melanoma in the medial
right temporal region. Sections of the left hemisphere at |-cm
intervals were prepared in cresyl violet and myelin (Heiden-
hain) pairs from the lateral geniculate body to the occipital
pole. In the brain stem, similar sections (25 p thick and oriented
perpendicular to the longitudinal axis of the brain stem) were
retained at 250-y intervals. These ran from the thalamus down
to the upper third of the pons.

The reconstructions were made by placing the slides in the
negative carrier of a 5- X 7-inch enlarger and preparing II- x
14-inch negative prints on Kodak photomechanical “T” paper.
Enlargements were used for the cerebral sections (about 3x)
and for the brain stem (12x). At the latter magnification, the

BS

July 1979

larger neurons appeared clearly in the cresyl violet prepara-
tions. The reconstruction was made from cresyl violet myelin
pairs of sections at about 500-u intervals. The differences
between numbers of the reconstructions indicate the interval
in microns.

The nomenclature used in the figures was derived from
several sources. The cortical nomenclature was taken from
Economo and Koskinas (7), the tract terminology from Riley
(25), and the nomenclature of the brain stem nuclei from
Olszewski and Baxter (20). Regretfully, we found it necessary
to abandon the macaque thalamic nomenclature (37), which
does not conform to the realities of the human thalamus in the
ventral and lateral region. In its place we have used the system
proposed by the Vogts (35, 36) with slight modification (30
32). This system has been presented recently with cytoarchitec-
tural diagrams (33), the approximate Anglo-American equiv-
alents are given in Table 1.

EXAMINATION OF THE BRAIN

In the right medial temporal region, a metastatic melanoma
invaded the hippocampal gyrus and extended anteriorly to the
uncus and posteriorly to the transverse level of the colliculi.
The ventricular system was displaced to the left, and the right
cerebral peduncle was indented on its anteroinferior aspect.
The inferior aspect of both hemispheres showed prominent
tentorial notching that extended to all of the cerebral cortex
immediately adjoining the tentorial opening on both sides.

No other metastases were found in the brain. No lesion other

ANATOMY OF DYSLEXIA 3

than the region of cortical infarction and surgical removal was
seen in the left hemisphere.

In the left occipital lobe, the surgical removal extended 2.5
cm anterior to the pole and destroyed the posterior calcarine

Fic. 2. The surgical lesion destroyed the cortex within the inter-
parietal sulcus (jp) and the medial portion of the adjacent angular
gyrus (Ang). R, fissura Rolandi; £-S, fissura Sylvii; 4 (t2), sulcus tem-
poralis primus (secundus); G sm, gyrus supramarginalis

TABLE |
Abbreviations

A Nucleus amygdalae Pl Pallidum laterale
B Nucleus basalis Pm Pallidum mediale
Bg Brachium conjunctivum P pd Nucleus peripeduncularis
BCoi Brachium colliculi inferioris Pret Regio pretectalis
Ce Nucleus centralis (Nucleus centrum medianum) prG Nucleus praegeniculatus
Cm Corpus mammillare Pu Nucleus pulvinaris
Cma Commissura anterior Put Putamen
Cm p Commissura posterior Py Nucleus paraventricularis hypothalami
Coe Nucleus locus coeruleus R Nucleus reticularis.
Coi Colliculus inferior Ru Nucleus ruber
Cos Colliculus superior Smth Stria medullaris thalami
Dbg Decussatio brachii conjunctivi Sn Substantia nigra
EW Nucleus Edinger Westphal So Nucleus supraopticus hypothalami
F Fornix Spt Nucleus supratrochlearis
Fa Nucleus fasciculosus (Nucleus ventralis anterior, medial Sth Nucleus subthalamicus

part) ™ Tractus Meynerti
F Is Fasciculus longitudinalis superior Tmth Tractus mammillothalamicus
Gl Corpus geniculatum laterale TO Tractus opticus
Gm Corpus geniculatum mediale Ve Nucleus ventrocaudalis (Nucleus ventralis posterior, pos-
iLa Nucleus intralamellaris (the intralaminar nuclei) terior part)
Is Nucleus interstitialis Vee Nucleus ventrocaudalis externus (Nucleus ventralis pos-
IV Nervus trochlearis terolateralis, posterior part)
Li Nucleus limitans Vei Nucleus ventrocaudalis internus (Nucleus ventralis poster-
Lm Lemniscus medialis omedialis, posterior part)
Lpo Nucleus lateropolaris (Nucleus ventralis anterior, lateral Ve pe Nucleus ventrocaudalis parvocellularis (Nucleus ventralis

two-thirds) posterior medialis parvocellularis)
M Nucleus medialis (Nucleus medialis dorsalis) Vim Nucleus ventrointermedius (Nucleus ventralis posterior,
N Nucleus anterior part)
N Il Nucleus nervi oculomotorii Voe Nucleus ventrooralis externus (Nucleus ventralis lateralis,
NIV Nucleus nervi trochlearis ventrolateral part)
N V me Nucleus nervi trigemini mesencephalicus Voi Nucleus ventrooralis internus (Nucleus ventralis lateralis,
N tub Nucleus tuberalis ventromedial part)
PE Nucleus parafascicularis Zi Zona incerta

Pi Pineal

4 VAN BUREN

cortex. The surgical defect extended 2.5 cm lateral to the
midline at the occipital pole and forward from the occipital
pole to destroy the most posterior and medial portion of the
angular gyrus (Fig. 2). The inferior and basal portions of the
parietal cortex were normal. On the medial aspect of the
occipital lobe, the resection included the posterior 2.0 to 2.5
cm of the calcarine cortex, the upper portion of the lingual
gyrus, and the lower portion of the cuneus (Fig. 3).

In the left parieto-occipital region (Fig. 4), the cavity occu-
pied by the hemorrhage and metastatic lesion removed at
operation undercut the angular gyrus some 0.5 cm anterior to
its cortical extent on the lateral surface. The posterior portion

Neurosurgery, Vol. 5, No. 1, Part 1

of the remaining calcarine cortex was surrounded by the op-
erative cavity. A narrow extension of the cavity came to the
cortical surface in the lingual gyrus.

Examination of the histological sections (Fig. 5) disclosed
small areas of infarction in the superior lip of the calcarine
fissure and within the occipitotemporal sulcus. Segments of
myelin loss and gliosis appeared in the stratum sagittal exter-
num and internum that could be followed forward into the
temporal lobe. The tapetum remained well myelinated, and no
damage was found in the splenium of the corpus collosum
(Fig. 5c). Microscopic examination of the right occipital lobe
and calcarine cortex did not show any damage.

Fic. 3. Transverse sections of the left occipital lobe starting with remnants of the pole and then progressing anteriorly (a through d). The surgical
cavity undercuts slightly the remaining angular gyrus.

SSS

July 1979

Fic. 4. The medial aspect of the left occipital lobe to show the
levels from which the sections shown in Figure 5 were taken. Note the
shrinkage of the superior lip of the calcarine fissure, which was the site
of infarction. The corpus callosum including the splenium was intact.

ANATOMICAL RECONSTRUCTION STUDIES

Thalamus (Fig. 6)

Pulvinar. There were two distinct areas of cell loss and gliosis.
One was inferolateral, limited by the external medullary lamina
laterally and the posterior margin of the nucleus ventrocaudalis
anteroinferiorly. The degeneration extended downward to in-
clude the posterolateral inferior extremity of the pulvinar.

The second region of degeneration in the pulvinar was seen
on the ventricular aspect of the nucleus behind the habenula
and extended forward (#220) to the scattered groups of cells of
intralaminar type that separate the nucleus medialis and the
pulvinar. It retained this position in more inferior sections,
coming into contact with the posterior aspect of the nucleus
limitans. Due to the mode of trimming the block, the full
superior and posterior excursion of the pulvinar cell loss lesions
could not be determined.

Lateral geniculate body. Total cell loss and gliosis affected
the apex of the geniculate pyramid and became constricted to
a central region of degeneration (#8320) in the lower segment
of the nucleus. At this level, cells were retained in the anterior
as well as the medial and lateral margins of the nucleus. The

ANATOMY OF DYSLEXIA 5

remainder of the thalamus that was available for study was
negative.

Brain stem (Fig. 7)

In the caudal mesencephalon, there was a linear lesion of
small diameter that extended some 4 mm (#2820 to 6820) but
at no point exceeded 2 mm in its greatest diameter. This was
first noted in the inferolateral portion of the inferior colliculus;
then it gradually shifted to a position in the dorsal tegmentum
immediately lateral to the locus coeruleus and the nucleus
nervi trigemini mesencephalicus. The appearance of the lesion
with striking gliosis and loss of myelinated fibers suggested a
microinfarct rather than secondary degeneration.

This lesion seemed to be related to the patient’s nystagmus.
On March 11, 1957, she had horizontal nystagmus on lateral
gaze bilaterally, enhanced by gaze to the left. On June 3, 1957,
she had a horizontal central gaze nystagmus with the quick
phase to the right on fundiscopic examination. The amplitude
increased on lateral gaze bilaterally with a higher frequency on
gaze to the left.

DISCUSSION

The patient's right hemianopsia correlated with the extensive
lesion of the left calcarine cortex. The small remaining frag-
ment in the anterior portion apparently permitted her to va-
guely see large moving objects in the far periphery of the right
visual field.

Cell loss in the lateral geniculate body spared only a thin
rind on the anteromedial and anterolateral aspects of the
nucleus. The cell loss corresponds to the macular representation
found by Polyak and Hayashi (23) in the chimpanzee plus
additional anterior extension correlated with the lesion that
spared only the anterior calcarine cortex.

The hemianopsia was considered to contribute little to the
language defect. In our experience, even lesions of the anterior
radiation producing a macula-splitting right hemianopsia (Ref.
29, Case UN) do not diminish the patient’s ability to read or
enjoyment of reading. Another case of right hemianopsia with-
out reading difficulty was reported by Foix and Hillemand (9).
In agreement with Deéjerine (4, 5), these authors pointed out
that an additional lesion of the splenium was needed to produce
alexia.

Déjerine provided the first well-described case of verbal
“blindness” and agraphia with a reasonable anatomical corre-
lation (3). Immediately after his stroke, the patient had a
“probable” right hemianopsia. After recovery from paraphasia,
alexia remained complete although the patient could recognize
figures. Writing ability was limited to tracing his name; he
could write neither spontaneously nor to dictation. In Déjer-
ine’s drawing of the brain, he indicated a softening (“the size
of a 5-franc piece”) that occupied the inferior three-quarters of
the left angular gyrus. This lesion penetrated to the ventricle at
the trigone, transecting the radiations of Gratiolet.

Hécaen et al. made a detailed study of language and cogni-
tive defects after occipital lobectomy in seven patients (10).
Three of these cases were followed for 9 to 33 months after
operation. Immediately after the lesion there was total alexia
that included numbers and musical notation, but no agraphia.
Words posed more of a problem than letters. This improved in
about a month so that the patients could read familiar texts in
a broken fashion. After 4 to 5 months, reading became more
rapid but still involved hesitation, a tendency to spell words,
and occasional paralexias. The patients were often troubled if
the texts were exposed only briefly or they were asked to read
in silence. They tended to follow the text with a finger, com-

6 VAN BUREN

Neurosurgery, Vol. 5, No. 1, Part 1

Fic. 5. Sections of the left occipital lobe taken at the levels indicated in Figure 4. As best seen in Section c, the central (lateral) portion (between
the arrows) of the external sagittal stratum (ess) showed marked demyelination. The demyelination of the internal sagittal stratum (iss) was of lesser
degree and the vertical extent of this was less (between the arrowheads) than that in the external sagittal stratum. The tapetum (/), lying yet more
medially, showed no degeneration. Section f was taken through the posterior temporal region at the level of the lateral geniculate body. The myelin
loss in the sagittal strata remains more posterior but the external sagittal stratum at this level is thinner and is seen only as a light band on the lateral
margin of the internal sagittal stratum. (The dark cortical markings on the superomedial and inferolateral corners of Section a are staining artifacts.)

plaining that the lines ran together. A distinctive feature,
despite a greater or lesser degree of recuperation, was persistent
lack of inclination to read. The reading problems seemed to be
independent of other speech difficulties that might be minimal
or transitory. Postoperatively, none of the patients could un-
dertake the same social activity that they had enjoyed prior to

operation. Patients were tested with a variety of perceptive
tests. The most important defect was inability to abstract and
understand the total significance of a situation.

The “occipital lobectomies” encroached more or less upon
the parietotemporal cortex. Reference in some cases to opening
the occipital horn suggested that considerable undercutting of

July 1979
A 57-177

ANATOMY OF DYSLEXIA 7

Fic. 6. Sections through the thalamomesencephalic region oriented perpendicular to the longitudinal axis of the brain stem. Areas of cell loss are
indicated by horizontal hatching in the pulvinar (Pu) and the lateral geniculate (G /). See text for description. The other abbreviations are given in
Table |. The difference between two section numbers indicates the interval between the sections in microns.

the parietal cortex may have occurred. Unfortunately, there
was no detailed anatomical account to compare with the pains-
taking clinical records.

Understanding of the relationship between natural lesions of
the thalamus and language function has been hampered by a
lack of anatomical studies. Earlier cases suggested that damage
to the pulvinar itself might cause less severe dysphasia than
damage to the cortical regions to which it projects.

Only 2 of 24 thalamic tumors reported by McKissock and
Paine were accompanied by aphasia (16). Ciemins reported an
instance in which a hemorrhage seemed to destroy all of the
thalamic nuclei on the left side, but the patient had only a
decrease in spontaneous speech, difficulty in carrying out
complex commands, difficulty in writing, and a defect in
describing material already read (2). In other cases (1, 2, 8, 17,
26), there was only mild dysphasia despite large thalamic
lesions. In their case, Reynolds et al. showed a posterior
thalamic lesion by computed axial tomography, but clinical
details were limited to a report of “severe aphasia” (24).

The limited clinical data suggest that the stable “thalamic
speech deficit” may appear most commonly as anomia with
neologisms and paraphasic logorrhea but relatively intact rep-
etition and comprehension of spoken speech. Testing is fre-
quently difficult because of a fluctuating level of response.
None of the earlier reports provided more than a gross indi-
cation of the size and position of the lesion.

However, specific areas in the pulvinar may be related to
specific speech function. In one case of predominantly receptive
aphasia with good clinical documentation, cell loss in the
anterosuperior portion of the pulvinar was associated with a
lesion at the posterior extremity of the sylvian fissure severing
fibers from the supramarginal and anterior angular gyri (27).
Taken in conjunction with the stimulation evidence of Oje-
mann et al. (19), this seemed to support the importance of the
anterosuperior pulvinar to the receptive aspects of speech.

In the present case, the region of the pulvinar projecting to
the posterior angular gyrus (hence related to language function,
primarily reading) lay in the inferior, posterior, and lateral
portion. Another case studied in serial section (MU-64-64) also
suggested that the posteroinferior and inferolateral portion of
the pulvinar might be related to cortex needed for reading
skills (32). This patient could read but had little understanding
of what he read; however, these symptoms were overshadowed
by memory loss and other symptoms related to the bilateral
cerebral lesions. The cell loss in the pulvinar of the present case
was quite similar to that of MU-64-64 with the exception that
degeneration also extended into the medial portion of the
pulvinar immediately behind the nucleus medialis.

In reconstructions of 28 human hemispheres with degener-
ation in the pulvinar after focal cerebral lesions (31), only 1
case (W-69-47) showed cell loss immediately behind the nu-
cleus medialis in the medial portion of the pulvinar. This was
associated with an extensive lesion of the occipitoparietal re-
gion that spread forward on the medial aspect of the hemi-
sphere into the retrosplenial portion of the limbic (cingulate)
gyrus, including the anterior paracalcarine cortex. In contrast,
Case A-64-96-R, with a cortical infarct destroying the superior
medial face of the hemisphere in the precuneus and the superior
portion of the cuneus but sparing the anterior paracalcarine
cortex, had no degeneration in this region of the pulvinar. In
the present case, the lesion undercut the anterior paracalcarine
cortex, confirming a relationship between this cortical region
and the segment of the pulvinar lying immediately behind the
nucleus medialis. Although it is part of the visual association
area, this region and the associated anteromedial region of the
pulvinar seem unrelated to the primary cortex subserving
language.

From a surgical aspect, the present case emphasized the short
distance between the angular gyrus and the occipital pole. In
fixed postmortem brains, the distance between the tip of the

8 VAN BUREN

A 57-I77

Neurosurgery, Vol. 5, No. 1, Part 1

Fic. 7. Sections showing the small area of infarction (black) inferior and lateral to the left inferior colliculus (Co /). See text for description. The
other abbreviations are given in Table 1. The difference between two section numbers indicates the interval between the sections in microns.

occipital pole and the posterior margin of the angular gyrus
averaged only 3.5 cm (34). At operation, the narrow gyri may
identify the occipital lobe, but with the pia-arachnoid intact
the outlines of the angular gyrus may be vague. The most
reliable means of identification of cortex indispensable for
speech is electrical stimulation of the cortex during craniotomy
under local anesthesia (21, 28). Short of this, recognition of the
anatomical relationships should permit the surgeon to use a
more medial or posterior entry into the occipital region and
thereby spare language function.

Received for publication, January 20, 1979; accepted, February 18,
1979.

Reprint requests: J. M. Van Buren, M.D., Professor of Neurological
Surgery, Department of Neurological Surgery, University of Miami,
School of Medicine, Post Office Box 016960, Miami, Florida 33101.

REFERENCES

1. Bugiani, O., Conforto, C., and Sacco, G. Aphasia in thalamic
hemorrhage. Lancet, |: 1052, 1969 (letter).

2. Ciemins, V. A. Localized thalamic hemorrhage: A cause of aphasia.
Neurology (Minneap.), 20: 776-782, 1970.

3. Déjerine, J. Sur un cas de cécité verbale avec agraphie, suivi

d’ autopsie. C. R. Soc. Biol. (Paris), 43: 197-201, 1891.

4. Déjerine, J. Contribution 4 l'étude anatomo-pathologique et cli-
nique des différentes variétés de cécité verbale. C. R. Soc. Biol.
(Paris), 44: 61-90, 1892.

5. Déjerine, J. Anatomie des Centres Nerveux. Vol. 2. Paris, J. Rueff,
1901.

6. Déjerine, J. Semiologie des Affections du Systéme Nerveux. Patis,
Masson et Cie, 1914.

7. Economo, C. V. and Koskinas, G. N. Die Cytoarchietktonik der
Hirnrinde des erwachsenen Menschen. Wien, Springer, 1925.

8. Fisher, C. M. The pathologic and clinical aspects of thalamic
hemorrhage. Trans. Am. Neurol. Assoc., 84: 56-59, 1959.

9. Foix, C. and Hillemand, P. Role vraisemblable du splénium dans
la pathogénie de l'alexie pure par lesion de la cérébrale postérieure.
Bull. Mem. Soc. Med. Hop. Paris, 49: 393-395, 1925.

10. Hécaen, H., Ajuriaguerra, J., and David, M. Les déficits fonction-
nels aprés lobectomie occipitale. Monatsschr. Psychiatr. Neurol.,
123: 239-291, 1952.

11. Henschen, S. E. Ueber Aphasie, Amusie and Akalkulie. Part V of
Klinishe und anatomische Beitrage zur Pathologie des Gehirns.
Stockholm, Nordiska Bokhandeln, 1920.

12. Henschen, S. E. Ueber sensorische Aphasie. Part VI of Klinishe
und anatomische Beitrage zur Pathologie des Gehirns. Stockholm,
Nordiska Bokhandeln, 1920.

13. Kleist, K. Sensory Aphasia and Amusia: The Myeloarchitectonic
Basis. Oxford, Pergamon Press, 1962.

ee

July 1979

14. Marie, P. Travaux et Mémoires. Vol. |. Paris, Masson et Cie, 1926.

15. Marie, P. and Foix, C. Les aphasies de guérre. Rev. Neurol. (Paris),
1: 53-87, 1917.

16. McKissock, W. and Paine, K. W. E. Primary tumors of the
thalamus. Brain, 81: 4t-63, 1958.

17. Mohr, J. P., Watters, W. C., and Duncan, G. W. Thalamic hem-
orrhage and aphasia. Brain Lang., 2: 3-17, 1975.

18. Nielsen, J. M. Agnosia, Apraxia, Aphasia: Their Value in Cerebral
Localization. 2nd Ed. New York, Paul B. Hoeber, Inc., 1946.

19. Ojemann, G. A., Fedio, P., and Van Buren, J. M. Anomia from
pulvinar and subcortical parietal stimulation. Brain, 91: 99-116,
1968.

20. Olszewski, J. and Baxter, D. Cytoarchitecture of the Human Brain
Stem. Philadelphia, J. B. Lippincott, 1954.

21. Penfield, W. and Jasper, H. Epilepsy and the Functional Anatomy
of the Human Brain. Boston, Little, Brown & Co., 1954.

22. Penfield, W. and Roberts, L. Speech and Brain-Mechanisms.
Princeton, New Jersey, Princeton University Press, 1959.

23. Polyak, S. and Hayashi, R. The cerebral representation of the
retina in the chimpanzee. Brain, 59: 51-60, 1936.

24. Reynolds, A. F., Jr., Harris, A. B., Ojemann, G. A., and Turner, P.
T. Aphasia and left thalamic hemorrhage. J. Neurosurg., 48: 570-
574, 1978.

25. Riley, H. A. An Atlas of the Basal Ganglia, Brain Stem and Spinal
Cord. New York, Hafner Publishing Co., 1960, p. 672.

26. Samarel, A., Wright, T. L., Sergay, S., and Tyler, H. R. Thalamic
hemorrhage with speech disorder. Trans. Am. Neurol. Assoc., 101:
283-285, 1976.

27. Van Buren, J. M. The question of thalamic participation in speech
mechanisms. Brain Lang., 2: 31-44, 1975.

28. Van Buren, M. M., Ajmone-Marsan, C., Mutsuga, N., and Sa-
dowsky, D. Surgery of temporal lobe epilepsy. Adv. Neurol., 8:
155-196, 1975.

29. Van Buren, J. M. and Baldwin, M. The architecture of the optic
radiation in the temporal lobe of man. Brain, 81: 15-40, 1958.

30. Van Buren, J. M. and Borke, R. C. A re-evaluation of the “nucleus
ventralis lateralis” and its cerebellar connections: A study in man
and chimpanzee. Int. J. Neurol., 8: 155-177, 1971.

31. Van Buren, J. M. and Borke, R. C. Variations and Connections of
the Human Thalamus. Vol. |. The Nuclei and Cerebral Connections
of the Human Thalamus. Berlin, Springer-Verlag, 1972.

32. Van Buren, J. M. and Borke, R. C. The mesial temporal substratum
of memory: Anatomical studies in three individuals. Brain, 95:
599-632, 1972.

33. Van Buren, J. M., Borke, R. C., and Modesti, L. M. Sensory and
nonsensory portions of the nucleus “ventralis posterior” thalami of
chimpanzee and man. J. Neurosurg., 45: 37-48, 1976.

34. Van Buren, J. M., Fedio, P., and Frederick, G. C. Mechanism and
localization of speech in the parietotemporal cortex. Neurosurgery,
2: 233-239, 1978.

35. Vogt. C. La myeloarchitecture du thalamus du cerocopithéque. J.
Psychol. Neurol., 12: 285-324, 1909.

36. Vogt, C. and Vogt, O. Thalamusstudien I-III. J. Psychol. Neurol.
(Leipz), 50: 32-154, 1941.

37. Walker, A. E. The Primate Thalamus. Chicago, University of
Chicago Press, 1938.

COMMENTS

The case reported by Dr. Van Buren re-emphasizes the
importance of the posterior part of the language area of the
dominant hemisphere, particularly in the region of the angular
gyrus, to a disorder in which difficulty in reading is the most
prominent finding. In a case such as his, with a relatively large
lesion, it is hard to know whether the defect is due to cortical
damage or to destruction of fibers connecting with the intact
visual cortex of the other side or perhaps with the thalamus.
But isolated dyslexia can occur in the absence of a visual field
defect, as was seen in a patient of mine who has intact visual
fields, a CT scan-documented stroke involving the posterior
region of the parieto-occipital junction in the back end of the
posterior language area, and a persisting moderate dyslexia
without oral or written language disturbance. A somewhat

ANATOMY OF DYSLEXIA 9

similar case, although with a less persisting deficit, was de-
scribed by Greenblatt (2).

The difficulty with the models derived from these individual
cases is that, if one maps the location of the cortex concerned
with various aspects of language in individual patients, one
finds a substantial degree of individual variability. I have
reported this for oral language function, as measured by elec-
trical stimulation mapping during naming (3), and have seen
an even greater individual variability in the location of sites
where electrical stimulation alters reading. Indeed, these pres-
ently unpublished findings suggest that reading is a process
that involves rather wide brain areas, not only the cortex
around the posterior language area but also extending forward
into the superior temporal gyrus on occasion and in portions
of the frontal lobe superiorly and anteriorly to the classical
anterior language area of Broca. Of course, these findings are
all in a select patient population, only those coming to crani-
otomy under local anesthesia for medically intractable epilepsy,
but alexia without agraphia has also been recently described
with strokes in the frontal lobe of the dominant hemisphere
(1). The neurosurgeon should recognize that if he approaches
close to the language cortex from any direction he is running
a risk of a potentially quite disabling reading disturbance. I re-
emphasize Dr. Van Buren’s suggestion that the only way to
avoid this problem reliably is to individually map these lan-
guage functions in the cortex of the awake patient.

What role the thalamic deficits might play in reading is still
undetermined. Dr. Van Buren summarizes the data indicating
the relationship of the dominant thalamus to spoken language
and points out the considerable specificity of lesions within the
pulvinar that are associated with oral language deficits and, in
this case, a reading deficit. This high degree of specificity in
the pulvinar probably explains why the induced pulvinar le-
sions that at one time were popular for treatment of spasticity
did not seem to be associated with much of a language disturb-
ance because they were placed relatively low in the pulvinar,
apparently well away from the areas that may be related to
spoken language. Reading was less thoroughly tested in those
patients, however.

George Ojemann, M.D.
Seattle, Washington

Benson, D. F. The third alexia. Arch. Neurol., 34: 327-331, 1977.
2. Greenblatt, S. Subangular alexia without agraphia or hemianopsia.
Brain Lang., 3: 209-228, 1976.
3. Ojemann, G. Individual variability in cortical localization of lan-
guage. J. Neurosurg., 50: 164-169, 1979.

This case report by Van Buren is unique because of its
anatomical elegance. In this respect, there are two distinct and
seemingly unrelated topics that deserve comment:

1. The thalamic lesions. In the late 1960s and early 70s there
was a flurry of activity among investigators who were trying to
understand the relationship between known thalamocortical
connections and clinical disorders of language. A lot of appar-
ently valid data has been accumulated, and the work still goes
on, but a viable working model of the relationship still eludes
us. Detailed studies such as the present one can only serve to
advance the cause.

2. The cortical and connecting white matter lesions. What-
ever meager reading ability remained to the patient could have
been mediated only by visual information which was arriving
in the right occipital lobe and thence transmitted to the left
angular gyrus region. Because the splenium was intact, the
information could potentially get across the midline. But it
could not reach the largely intact left angular gyrus because
the connecting pathways in the left occipital lobe had been

10 VAN BUREN

removed. (For further details and diagrams of the pathways,
see Greenblatt (5)). Particularly important in the present case
is Van Buren’s notation of the fact that the angular gyrus had
been significantly undercut during the lobectomy. Undercut-
ting of the left angular gyrus can lead to alexia even without a
formal lobectomy (4). The pathways that are thought to be
interrupted include the extreme sagittal stratum (vertical occip-
ital fasciculus) (3, 5). In the present case the more anterior limb
of this pathway was found to be degenerated.

The practical surgical implications of Van Buren’s case are
all derived from the anatomy of the cortical and white matter
lesions.

1. The warning to the surgeon about dominant occipital
lobectomies invading the angular gyrus has been sounded
before by Hecaen et al. and Rasmussen (6), among others, and
the surgical location of the angular gyrus has been emphasized
(5), but caveats such as this are justifiably repeatable.

2. The importance of avoiding undercutting of the angular
gyrus, as well as direct damage to it, also bears repeating.

3. The urgent nature of the surgery in this particular case
shows the importance of having some useful surgical guidelines
to the location of language-related structures for those occa-
sions when it is impractical to do cortical stimulation. Even the
stimulation technique may lead one astray if specific reading
tests are not included.

4. With regard to “surgical locations,” we must be careful in
our statements about surgical distances. When Van Buren
speaks of a lobectomy carried 2.5 cm anterior from the occipital
pole, he presumably means that it was carried in a straight
posterior-to-anterior direction, at right angles to a true coronal
section. But surgical distances also may be measured (before
resection) along surfaces that are actually curvilinear, and then
the distances may seem to be much larger. For example, the

Neurosurgery, Vol. 5, No. 1, Part 1

approximate center of the angular gyrus is 9 cm forward from
the inion and 4% cm lateral to the midline when measured
along the surface of the exposed skull (5).

Many reports of cases such as this are strong on the clinical
side but weak anatomically. Here, on the other hand, we have
the reverse situation. It would be of interest to know more
about the patient’s ability to perform verbal repetition, her
reading of single words, her spelling, and her color-naming
abilities. In addition, the report would have been considerably
strengthened if more extensive formal psychological and read-
ing achievement tests had been done. By and large, these
omissions are probably attributable to the fact that the patient
was seen in the middle 1950s, a decade before the beginning of
the current surge of interest in the corticocortical disconnection
syndromes (1, 2). The publication of a case such as this in a
neurosurgical journal seems to indicate that renewed interest
in this fascinating field is beginning to develop among neuro-
surgeons as well as neurologists.

Samuel H. Greenblatt, M.D.
Toledo, Ohio

1. Gazzaniga, M. S., Bogen, J. E.. and Sperry, R. W. Some functional
effects of sectioning of the cerebral commissures in man. Proc.
Natl. Acad. Sci. U.S.A., 48: 1765-1769, 1962.

2. Geschwind, N. Disconnexion syndromes in animals and man.
Brain, 88: 237-294, 585-644, 1965.

3. Greenblatt, S. H. Alexia without agraphia or hemianopsia: Ana-
tomical analysis of an autopsied case. Brain, 96: 307.316, 1973.

4. Greenblatt, S. H. Subangular alexia without agraphia or hemi-

anopsia. Brain Lang., 3: 229-245, 1976.
. Greenblatt, S. H. Neurosurgery and the anatomy of reading: A
practical review. Neurosurgery, |: 6-15, 1977.

6. Rasmussen, T. Surgery for epilepsy arising in regions other than

the temporal and frontal lobes. Adv. Neurol., 8: 207-226, 1975.

w