вход по аккаунту


Developmental dyslexia in women Neuropathological findings in three patients.

код для вставкиСкачать
Developmental Dyslexia in Women:
Neuropathological Findngs in Three Patients
Peter Humphreys, MD, FRCP(C), Walter E. Kaufmann, MD, and Albert M. Galaburda, MD
Brains from male cases with dyslexia show symmetry of the planum temporale and predominantly left-sided cerebrocortical microdysgenesis. We now report on three women with dyslexia. In all brains, the planum temporale was
again symmetrical. Also, in two of the brains, multiple foci of cerebrocortical glial scarring were present. In both
women, many of the scars were myelinated, suggesting origination during late intrauterine or early postnatal life. In
one, scars were mainly left perisylvian and involved portions of the vascular border zone of the temporal cortex. Inthe
other, scars were more numerous and occurred in the border zone of the anterior, middle, and posterior cerebral
arteries symmetrically. All three cases showed to a variable extent brain warts, molecular layer ectopias, andfocal
architectonic dysplasia identical to those seen in the male cases. Two women had primary brain neoplasms, an
oligodendroglioma and a low-grade astroqtoma, respectively, and two women showed small angiomas. Reexamination of previously reported male cases disclosed one with myelinated glial scars. Two control brains with asymmetrical
plana temporale showed myelinated glial scars as well. The significance of the anatomical findings is discussed, and
possible etiological factors are considered with known effects of autoimmune diseases on the nervous system.
Humphreys P, Kaufmann WE, Galaburda AM. Developmental dyslexia in women:
neuropathologicalfindings in three patients. Ann Neurol 1990;28:727-738
Previous reports 11, 23 from our laboratory described
consistent neuroanatomical characteristics in four male
cases with well-documented histories of developmental dyslexia. Comparable findings were observed
in an additional unpublished male case. Two main categories of anatomical change have been noted as follows: All five brains from males with dyslexia had (1)
symmetrical plana temporale instead of the more common leftward asymmetry 133 and (2) multiple foci of
cerebrocortical microdysgenesis. The cerebrocortical
microdysgenesis consisted of focal microgyria, nodular
dysgenesis characterized by either nests of neurons
and glia in layer I or true brain warts, or both 143.
Between 30 and 140 individual foci per brain of these
various microanomalies have been observed in the dyslexia cases, as compared with only a few ectopias present in an occasional control brain (see 153). We have
argued that the two findings occurring together are
significantly linked to developmental dyslexia (for a
detailed discussion, see 163).
To expand on our previous reports of male cases, we
now report three women with developmental dyslexia
whose brains were analyzed in whole-brain serial histological sections.
Materials and Methods
The serial sections for cytoarchitectonic and myeloarchitectonic analysis were processed for Nissl substance and myelin
From the Dyslexia Research Laboratory, Charles A. Dana Research
Institute, and Department of Neurology, Beth Israel Hospital, and
Harvard Medical School, Boston, MA.
by methods previously reported for the male dyslexics El, 2).
Some sections were stained for glial reaction by the phosphotungstic-acid-hematoxylin (FTAH) and hematoxylineosin (H-E) methods, and for axons by the Bodian technique.
All light-microscopic abnormalities were cataloged as to
type and cytoarchitectonic and laminar distribution. Histological abnormalities were mapped on drawings made from
photographs of the lateral, dorsal, and ventral aspects of the
brain taken before embedding.
The areas of the temporal plana were calculated on computer-assisted reconstructions made from the serial sections.
This method corrects for varying distance between sections
and side differences in the vertical axis level of the planum
temporale. A directional asymmetry coefficient (6) for the
area of the planum temporale was computed as 6 -. (R - L)/
%(R + L), which corrects for individual variability in the size
of the measured area. A value of 6 between -0.05 and
+ 0.05 was designated to be symmetrical.
Patient Descriptions and Findings
Information was obtained from medical and educational
records, and in some cases also from family members.
Case 1 (ORT-10-84)
This right-handed social worker died by suicide
at age 36 years. She was the product of an unremarkable
pregnancy and delivery, informally with normal early developmental milestones, including language. An automobile accident at age 2 years caused significant head injury, with
Received Dec 12, 1989, and in revised form Jun 6, 1990. Accepted
for publication Jun 10, 1990.
Address correspondence to Dr Galaburda, Department of Neurology, Beth Israel Hospital, 330 Brookline Ave, Boston, MA 02215.
Copyright 0 1990 by the American Neurological Association 727
coma and altered consciousness lasting approximately 10
days. There were no obvious resulting motor, linguistic, or
cognitive deficits, but she was noted to be a very active child
with a short attention span.
Major difficulty in learning to read was evident during the
early years of school. In adulthood, reading ability varied
from the third- to 12th-grade level, depending on the patient’s ability to profit from contextual cues. A Wechsler
Adult Intelligence Scale at the ages of 33 and 35 years
showed the I Q in the normal to above-average range with no
significant discrepancy between verbal and performance
scores. Subtests showed the most striking deficits in areas of
attention control, word retrieval, and confrontation naming,
and in the perception and production of multisyllabic words.
The patient suffered from various atopic disorders including asthma and drug allergies. A workup for collagenvascular disease was negative.
During her second and third decades, the patient had periods of depression and social withdrawal accompanied by
visual and auditory hallucinations and other paroxysmal
symptoms. There were repeated suicide attempts prompting
several psychiatric admissions. A sleep-deprived electroencephalogram and cranial computed tomographic scan were
normal. The patient had no children and no family history of
PATHOLOGICAL FINDINGS. The general postmortem examination was unremarkable except for the presence of several uterine fibroids. The brain weighed 1,255 gm. No gross
abnormality or anomaly of the cerebral hemispheres, cerebellum, or brainstem was noted. Microscopic evaluation of
the cresyl violet-stained sections revealed several abnormalities.
(1) Molecular layer neuronal ectopias (facal cerebrocortical microdysgenesiJ). Four ectopic nests, three in the left hemisphere, were noted in the orbital frontal regions bilaterally
and in the left temporal pole (Figs lA, 2). For purposes of
this analysis, we considered molecular layer neuronal ectopias and brain warts to be variants of the same pathogenetic disturbance during neuronal migration [7), and thus
recorded them together.
(2)Focal cortical neuronal lass and gliosis. This was the most
common microabnormality in the cerebral cortex (Fig 3;
Table). The typical lesion consisted of a fusiform, triangular,
circular, or occasionally irregularly shaped zone devoid of
neurons but having a preserved network of capillaries and
increased glial cells. Most of the lesions had an anteroposterior (A-P) diameter of less than 700 IJ-m,and many were
located in the depths of sulci and occurred in clusters. The
lesions varied in laminar location and sometimes involved
more than one layer. Most of the lesions were richly gliotic
(see Fig 3A), and in myelin sections two-thirds of these
showed dense staining for myelin (see Fig 3B; Table). While
glial reaction when present typically involved the whole lesion, myelin staining involved lesions of the deeper layers, or
the deep portions of lesions involving more than one layer.
Occasional sections passing through the central portion of
some lesions showed a perforating cortical arteriole, which
appeared unremarkable. A general survey of cortical and
meningeal vessels revealed no intimal thickening, hyalinitation, or round cell infiltrates.
728 Annals of Neurology
Fig I. Examples of molecular layer neuronal ectopias in Case 1
(A),Case 2 IB), and Case 3 1 0 . 1n A and C, the ectopiasform
actual brain warts, whereas in B, the ectopic nearom, althougb
reaching the pia, do not substantially disturb the pial .ru&we
and would not be visible on inspection of the brain surface. A l ~ o
note in C a dilated blood vessel near the center of the brain wart,
a common finding. (Bar {A} = 250 pm; bar {B} = I S O mm;
bar {C} = 500 p.)
Vol 28 No 6 December 1990
By sampling every 20th section (700 pm), 122 such foci of
selective neuronal loss were found, 106 in the left hemisphere and 16 in the right, with a predilection for left orbital
frontal, opercular frontal, and lateral temporal cortices (see
Fig 2; Table).
(3) Vascular malformution. There was a nest of dilated,
distorted thin-walled vessels in the left frontal subcortical
white matter with an A-P extension of 16.8 mrn (Fig 4A).
(4) Oligodendroglioma. Two succeeding sections of the left
mesial temporal lobe (Fig 5A) showed a microscopic multilobed focus in the anterior hippocampus having the characteristic features of an oligodendroglioma.
The areas of the temporal plana in the two hemispheres
were symmetrical.
Case 2
This left-handed psychiatrist died at age 88 years
of complications of arteriosclerotic heart disease. Her family
recalled no difficulties with delivery, neonatal problems, or
delay in early milestones. By school age, there were significant and enduring problems with learning to read, which she
finally accomplished at age 12 years. At age 25 years, she was
diagnosed as having typical dyslexia by Samuel T. Orton, an
influential early medical worker in chis field. No formal psychological assessment of the patient was ever performed.
Samples of handwriting from later life revealed syntactic and
spelling errors. Despite these handicaps, the patient had a
distinguished professional career.
The patient also had biopsy-proved ulcerative colitis, degenerative joint disease, episodes of irregular cardiac rhythm,
the latter dating to at least age 21 years, and later, she had
arteriosclerotic heart disease.Other long-standing problems
included grade 0 chronic lymphocytic leukemia, obstructive
pulmonary disease, and indolent urinary tract infection.
The patient had three children, all rght handed, who did
well academically. One of her sons, a physician, reported lifelong difficulty with spelling. One of the patient’s brothers
also had ulcerative colitis.
Fig 2. Diagrams of the brain of Case 1 showing topographic
distribution of histologicalabnormalities. Open circles denote molecular kayer ectopias or brain warts; closed circles represent myelinated glial scars in the cortex. Note that the distribution of
lesions on the convexity corresponds in part t o the border zone of
the middle and posterior cerebral arteries. The right hemisphere
is nearly free of pathology. The planum temporale is symmetrical
in this case. See text.
PATHOLOGICAL FINDINGS. Postmortem examination was
restricted to the brain, which weighed 1,375 gm. The external appearance of the brain and meninges was unremarkable.
The planum temporale was symmetrical. Microscopic evaluation of the Nissl sections, however, revealed widespread abnormalities.
( I ) Molecular layer neumnal ectopias cfocal cmbmcortical mimdysgenais). Sixty dysgenetic lesions, both molecular neuronal ectopias and brain warts, many in the depths of sulci,
were equally distributed between the two cerebral hemispheres (see Fig 1B;Table). As demonstrated in Figure 6,
the ectopias were densely clustered in the orbital frontal
regions and in arc-shaped bands of the lateral frontal cortex
corresponding to the border zone between anterior and middle cerebral artery territories. The posterior perisylvian cortex was not involved.
(2) Focal cortacal neuranal loss and gliosis. In the survey of
every 700 pm, 699 lesions identical in appearance to those in
Case 1 (see Fig 7, for example), but far more numerous,
were observed equally between the hemispheres (see Fig 6).
Again, there was a strong tendency for the lesions to be
located in the depths of sulci. Unlike Case 1, however, this
Humphreys et al: Dyslexia in Women
patient’s lesions occurred more often in the m
layers. Fibrillary gliosis in the cell-free areas I
often in this patient, being demonstrated in on
lesions stained with PTAH. Myelination of the
common, occurring in 40% of the lesions idei
myelin sections. As in Case 1, myelination of k
to occur more often in the deeper cortical lay
occasions, unmyelinated lesions contained cenu
of macrophages, suggesting a recent onset, but t
seen in myelinated lesions.
The distribution of the foci of neuronal loss (s
distinctly different from that noted in Case 1, bc
instead to a band of anterior orbital frontal, la
parasagittal frontoparietal, and inferior occipitot
tex corresponding to the entire border zone of
middle, and posterior cerebral arteries. Myelin
myelinated lesions alike followed this pattern.
frontal regions, the bands of foci of neuronal 1
ronal ectopias overlapped, the former tending 1
anterior and medial to the latter. Posteriorly
spheres, there tended to be more scars than ectc
for the border zone region, the orbital frontal
frontal opercular zones were free of foci of n
while studded with neuronal ectopias. As in Ca
and meningeal vessels appeared entirely unrerr
(3) Vascular malformation. There was a 1.4-1
tical “cryptic” microangioma in the left lateral f
(see Fig 4B) within a zone showing micrody
focal neuronal loss.
Case 3 CORT-14-84]
This right-handed, left-eyed woma
pokalemic cardiac complications of bulimia at ,
She was the product of an unremarkable pregn
livery (repeat cesarean section). Early develop
stones were normal. By age 4 years, there wa:
delayed l a g w e acquisition and easy distractibi
in kindergarten was slow, necessitating a trar
before first grade and a remedial reading progi
tlnued throughout elementary school. IQ testii
(Stanford-Binet) showed above-average score!
expressive vocabulary in the superior range but
tibility, poor pencil and paper skills, and probler
and auditory memory. Review of test results b
pendent experts led to conflicting impressions;
the patient as primarily dyslexic, and a third cc
Fig 3. Example of a myelinated glialscar from Case 1 showing
glial staining (A) on a phosphotungstic-acid-hemato~lin section and myelin staining (B) on a Loyez sectron. Note that the
full extent of the glzalscar is not myelinated. (Bars = 500 F.)
Distribution of Histological and Architectonic Anomalies in the Female Patients with Dyslexia
Patient 1
Patient 2
Cerebral Hemisphere
Neuronal ectopias (total)
Cortical scars (total Nissl)
Cortical scars (total Loyez)
Myelinated scars
Percentage of myelinated scars
The total number of scars seen in Nissl and Loyez sections differ in accordance with different staining frequencies for the tv
Patient 1, note that there is a bias to the left hemisphere, particularly for myelinated scars. In Patient 2, the general involven
symmetrical,but myelinated scars are still relatively more predominant in the left. See text for discussion. Patient 3 did not shs
730 Annals of Neurology Vol 28 N o 6 December 1970
F ig 5 . (A) Lw-power microphotographof the lobuluted olzgodendroglioma (arrows) in a Nissl-stained section of Case 1 (bar =
1 .OO mm) arising near the pyramidal cell layer of the lt$ hippocampus. (B) Cellstain ofa section showing a microscopic lowgrade astrocytoma (arrows) in Case 3. Under higher magnification,some neuronal elements are seen mingling with glial cells
(bar = 300 pm).
Pzg 4. (A) Sample section showing a small arteriovenousanomaly in the white matter of Case 1 (bar = 500 pn), and (Bi a
microscopicvascular anomaly embedded in dysplastic cortex of
Care 2 (bar = I .OO mm). Vascular pathology such as (B) IS
commonly seen in areas of cortical dysgenesis.
reading disability to be secondary to attention deficit disorder. Because many individuals with this pattern are diagnosed as dyslexic, we included this patient in this report. By
18 years, there was modest decline in cognitive performance.
The Wechsler Adult Intelligence Scale showed a full scale
IQ of 103, a verbal score of 99, and performance score of
108. Reading level was at the 47th, spelling at the 5th, mathematics at the 16th, and vocabulary and comprehension at
the 10th percentile. Part of the decline in functioning may
have been related to nutritional consequences of her eating
disorder, initially manifested at age 15.
Except for bulimia, the patient was in good health all of
her life, with no evidence of allergic or immunological disorders.
There was a strong family history of learning disability.
The patient’s two younger brothers were diagnosed as dyslexic, the youngest also having attention deficit disorder;
both were right-handed and left-eye dominant. Her righthanded mother had a history of spelling difficulties, right-left
disorientation, and poor writing. The maternal grandfather
may have been dyslexic by history, but there is no confirmatory documentation. On the paternal side, there was a remote history of left-handedness and rheumatoid arthritis but
no history of dyslexia.
At postmortem examination,
there was pulmonary congestion and edema. Generalized
atrophy of the viscera was noted, which was attributed to
chronic malnutrition. There was grade 1 atherosclerosis of
the coronary and cerebral arteries, also of the aorta. The
heart was otherwise normal.
The brain weighed 1,300 gm; the plana temporale were
symmetrical, and the external appearance of the brain and
Humphreys et al: Dyslexia in Women
Fig 6. Diagram of the brain of Case 2 showing topographic
distribution of histological abnormalities. Open circles denote moIecular layer ectopias or brain warts; closed circles represent myelinated glial scars in the cortex. Note that the distribution of
lesions on the convexity corresponds in large part to the border
zone of the anterior, middle, and posterior cerebral arteries. Also
note that the foci of dysgeneszs are clusteredfrontally, while foci
ofglial scarring are more posterior. The planurn temporale is
symmetrical in this case. See text.
meninges was unremarkable. In comparison with the other
two cases, microscopic evaluation of the Nissl sections revealed relatively few abnormalities.
( I ) Molecular layer neuronal ectopias vocal cerebrocortical microdysgenesis).Only l l such lesions were identified, 9 molecular layer ectopias and 2 brain warts (see Fig 1C). Nine ectopias were present in the right hemisphere (Fig 8; Table),
concentrated in the orbital frontal and perisylvian regions,
with a few lesions parasagittally in the arterial border zone
region. In the left hemisphere, there was a brain wart in the
lateral frontal area and an ectopia in Heschl’s gyrus.
732 Annals of Neurology
Fig 7. Example of a myelinated glial scar (arrows)from Case 2
showing cellstaining (A)on a cresyl violet section and myelin
staining (B) on a Loyez section. Note that there is n o m l intracortical myelin in radialfbers, and that this delicate pattern is
disrupted by the dense deposit of myelin in the scar. (Bars =
300 pm.)
Fig 8. Diagram of the brain of Case 3 showing the right hmisphere, which contained most of the histological abnormalities.
Open circles represent molecular layer ectopias or brain warts.
This brain also showed a small low-grade astrocytoma in the
cortex. The planum temporale zs symmetrical in thzs case. See
Vol 28 N o 6 December 1990
Fi g 9. (L&j Topography of distribution of myelinated scars in
control case M U-19. Note that the distribution of myelinated
focal scars is similar to Case 2, except that the orbital and opercukzr regions are not involved. The planum temporale exhibited
the standdrdform of asymmetry in this case. that is, ldt larger
than right. (Right) Topography of distribution of myelinated
scars in control case M U-98. Note that the distribution of myelinatedfocal scars is similar to Case 2, except that the orbital
and opercular regions are not involved. The involvement is milder
than in case M U-19. Theplanum lemporale exhibited the standurdform of asymmetry in this case, that is. 1ej.t larger than
right. Also see text.
(2) Microscopic intracortical glioma. A 1-mm glial nodule
(see Fig 5B) was noted within the cortical plate of right
dorsal area 4 (Brodmann). The tumor involved layers I to 111
and contained predominantly large, pale nuclei, occasionally
ovoid in shape, without cytoplasm or fibrillary reaction on
PTAH stain. Interspersed among the pale nuclei were occasional neurons and typical oligodendroglial nuclei. H-E stains
outlined a plethora of branching capillaries. The tumor was
most likely a low-grade protoplasmic astrocytoma, but an
oligondendroglioma could not be ruled out by our methods.
There were no glial scars or vascular anomalies.
In summary (see Table), foci of neuronal loss (cortical
scars), myelinated or not, were distributed in a strikingly
asymmetric manner in Case 1, favoring the left hemisphere;
the more numerous lesions in Case 2 were symmetrically
placed, except for the tendency of myelinated scars to occur
more often in the left hemisphere than the right. Xone of
the brains had discernible microdysgenesis or scarring in the
basal ganglia, diencephalon, or brainstern. Cases 1 and 2 had
a few microscopic scars in the cerebellar cortex, but none
that stained positive for myelin.
Control Cases
The control group, which was processed and analyzed identically to the dyslexics, included the same 10 male brains
(ORT-15, MU-89, MU-91, CAN-263, SP-6, STD-VI, SP14, V-63, MU-19, and MU-98) reported by Kaufmann and
Galaburda [4}. We added three male brains from the Orton
collection (ORT-7, ORT-9, ORT-12) at this laboratory, and
three female brains from the normative series (MU-92, MU85, MU-39), of the Yakovlev Collection, for a total of 16.
The nine male and three female controls from the Yakovlev
Collection were described as normal, but we cannot exclude
the possibility that some may have had dyslexia. The four
male Orton cases were most likely nondyslexic.
Neuronal ectopias were few or absent in the control
brains, in contrast to the dyslexic brains, which usually have
had large numbers of ectopias.
Three of 16 control cases showed myelinated cortical scars
as follows: SP-14 had a single scar in the right hemisphere;
MU-19 had 51 (31 left, 20 right) (Fig 9 , left); MU-98 had 15
(7 left, 8 right) (see Fig 9, right). In the latter two cases, the
orbital surface was spared and the frontal zones were either
lightly affected or spared. As in the dyslexic cases, the myelinated scars were distributed in the arterial border zone
regions but only in the extrafrontal portions.
Planurn temporale asymmetry was similar to that seen in a
previously reported large normative series [3}, that is, 11 of
16 with left asymmetry, 3 of 16 with right asymmetry, and 2
of 16 with symmetrical plana.
Reexurnination of Mule Dyslexic Brains
We reexamined for myelhated cortical scars the three male
dyslexic brains still available to us. ORT-2-82 had numerous
scars predominantly in the left hemisphere (Fig lo), involv-
Humphreys et al: Dyslexia in Women
chance is very small. It thus remains likely that the
presence of planum temporale symmetry represents a
significant anatomical association of dyslexia.
The manner by which symmetrical plana may relate
to the reading disorder of developmental dyslexia is
not known. Symmetrical cortical areas, however, show
excessive development of the side that is ordinarily
small 181 and an altered pattern of interhemispheric
connectivity bilaterally 191. The large side results from
an increase in the number of neurons {lo]. One may
speculate that these changes in neuronal circuitries
could support changes in cognitive capacity. The presence of symmetry alone, however, does not appear to
be sufficient for dyslexia to be expressed, and early
acquired, even mild pathology affecting the developing
cerebral cortex may be necessary as well.
Case 2 resembled the male dyslexics in having multiple microdysgenetic foci, but she had no left hemisphere preponderance and no significant involvement
of the perisylvian cortices. In common with the males,
she had involvement of the orbital, opercular, and lateral frontal regions. Case 1 had very few neuronal ectopias but instead had myelinated cortical scars (a phenomenon also seen extensively in Case 2) and a small
oligodendroglioma. Case 3 had only a few neuronal
ectopias located largely in the right frontal opercular
zone; she also had a primary brain tumor. Small vascular anomalies were present in Cases 1 and 2. Because
we have discussed focal cerebrocortical microdysgenesis elsewhere [l, 2, 51, we will concentrate here on the
other abnormalities.
Fig 10. Topography of distribution of myelinated scars and molecular Layer ectopias in previously reported dyslexic case ORT-2.
Open circles denote molecular layer ectopias or brain warts; closed
circles represent myelinated glial scars in the cortex. Note that the
distribution of myelinated focal scars overlaps in part the vascular watershed, and that the orbital sudace contains some lesions
(see cases M U-I 9 and M U-98). The planum temporale, as in
all other brains of dyslexics studied by us, was symmetrical.
ing the orbital frontal cortices, the perisylvian and lateral
temporal regions, and portions of the arterial border zone;
many of the scars were myelinated. The other two cases had
no myelinated scars. In retrospect, we also noted that some
of the neuronal ectopias in the male cases were located in
portions of the arterial border zone regions bilaterally.
Like the five males, the three women with dyslexia had
symmetrical temporal plana. This is now a uniform
finding in all dyslexic brains we have studied. Because
fewer than one-third of ordinary brains have symmetrical temporal plana [3,Sl, the probability of encountering eight dyslexic brains with symmetrical plana by
The Myelinated Scar
Available evidence from human 111-141 (for review,
see [15]) and animal [16-19] studies suggests that glial
scars in cerebral gray matter myelinate only if damage
occurs before myelination is completed. In humans,
myelination of gray matter scars has been observed in
the corpus striatum [20) and in the cerebral cortex
113, 201. The histological findings in cortex and striatum are considered to be morphologically equivalent.
In the vast majority of case reports of human brains
with myelinated scars, the cerebral insult clearly occurred at around the time of birth 111-141. Those
cases without birth or neonatal injury sustained cerebral insults some time during the first year of life 1127.
Myelinated scars in human cortex or corpus striatum
have to our knowledge never been documented after
insults to the brain in later childhood or adult life.
Animal studies essentially confirm the observations
in humans. Myers [l5} reported myelinated scars in
the basal ganglia but not cortex of rhesus monkeys
after an intrauterine, perinatal hypoxic-ischemic insult.
Myelinated scars anywhere could not be produced by
hypoxic-ischemic insults in older monkeys. Johnston
1171 produced myelinated scars in the striatum of rat
734 Annals of Neurology Vol 28 No 6 December 1990
pups by unilateral carotid ligation and reduction of
ambient oxygen. Deckel and Robinson El81 performed neocortical transplants from embryonic day 17
brain tissue into adult rat cortex and striatum, and produced some hypermyelinated regions reminiscent of
myelinated scars. The common thread in all these
animal studies seems to be that myelination of gray
matter scars requires the presence of perinatal brain.
We could find only one report of injury to adult animal
nervous tissue leading to scar myelination in the posterior columns of cat spinal cord 100 days after dorsal
root section 1191.
It seems highly likely that the myelinated scars seen
in three of our patients with dyslexia date from before,
possibly during, but not after intracortical myelinogenesis was completed. According to Yakovlev and Lecours { 2 11, human intracortical myelination begins,
with the exception of primary visual and sensorimotor
cortex, at 40 weeks’ gestation. While a minor degree of
progressive myelination may continue in the tangential
fibers of the cerebral cortex well into adult life, the
process is largely finished by age 5 years. Therefore,
the event or events leading to multifocal cortical
neuronal loss with myelinated gliosis must have occurred either during the last trimester of pregnancy or
during the first 5 years of postnatal life but well before
reading instruction.
Not all scars of fetal origin become myelinated, as
can be seen in animal studies in which fetal striatal cells
were transplanted to cortex or striarum with resulting
myelin-free scars {IS]. In the present study, we observed that scars invoiving deeper cortical layers were
more likely to show myelin, probabiy reflecting the
proximity of a rich supply of oligodendrocytes in the
immediately subjacent white matter and in the more
heavily myelinated inner stripe of Baillarger 1211.
Hence, it would not be surprising to discover that
more superficial cortical scars ordinarily escape detectable myelination, irrespective of time of injury.
It is reasonable to speculate that the foci of microdysgenesis and of cortical scarring have the same etiology, with their differing appearance simply reflecting
damage at slightly different stages of brain development. McBride and Kemper 171 have postulated that
micropolygyria, brain warts, and neuronal ectopias develop as a consequence of injury during neuronal
migration. Alternative views propose that these abnormalities can occur with injury shortly after the completion of neuronal migration 122). Microdysgenesis
has been produced experimentally with damage during
late phases of neuronal migration [23), which in the
human goes on until the fifth fetal month {24]. We
believe that a pathological process acting during the
second trimester could lead to the present form of
microdysgenesis, while the focal myelinated scars
could be the outcome of the same process acting dur-
ing the third trimester or early postnatal period. An
insult spanning the two periods, or finding two areas of
cortex at different developmental stages, could prod. ce both microdysgenesis and myelinated scars in the
same brain. Also, an identically timed insult might produce predominantly microdysgenesis in a brain that is
less mature, and predominantly glial scars in one that is
more mature despite similar chronological ages, for
example, male and female brains, respectively 1251.
This hypothesis could help explain the differences in
frequency of lesion type in the brains from male and
female dyslexics we have analyzed, with the former
more commonly having ectopias and the latter more
commonly having scars.
There is a topographic relationship between microdysgenesis and myelinated scars that supports the
above hypotheses. In Case 1,the myelinated scars cluster in the orbital and opercular frontal, perisylvian, and
lateral temporal cortices, the same areas in which
neuronal ectopias were concentrated in the male dyslexics. In Case 2, the neuronal ectopias and cortical
scars (myelinated and unmyelinared) overlap extensively in the posterior frontal regions, while the remaining ectopias segregate to the more rostral frontal
regions with the remaining scars segregating more posteriorly. Late-maturing frontal regions, for example,
may take the brunt of the ectopias, while cortices earlier to mature may express the insult as scars.
Another finding potentially linking the two types of
lesion is the central perforating cortical arteriole in
both, suggesting that ectopias and scars occur in the
territory of small-caliber cortical vessels. In molecular
layer ectopias, we observed that the centrally placed
arteriole is seen most often in the brain wart. Larroche
and Razavi-Encha {26) have recently made the same
Etiology of the Myelinated Scars
The presence of many discrete microscopic foci of
selective cortical neuronal loss (or microdysgenesis),
apparently in the territory of perforating cortical arterioles, suggests to us the possibihty of a microangiopathic etiological process. While reported cases
{ll-141 of rnyelinated scars have been clearly related
to perinatal hypoxic-ischemic insults or febrile illnesses
in the first year of life, no such history is present in any
of our dyslexic cases. The cortical pathology in cases of
perinatal hypoxia-ischemia is never that of a multifocal,
microscopic process but occurs at a macroscopic level
with confluent necrosis, ulegyria, and porencephaly.
Likewise, head trauma cannot explain the pathology
observed in our dyslexics (e.g., in Case 1, at age 2
years); cortical contusion necrosis is more extensive,
wedge-shaped, and occurs on the crests of gyri 1271.
While there are numerous potential causes for a cerebral microangiopathic process in early life (e.g., heHumphreys et al: Dyslexia in Women
molytic-uremic syndrome, posttraumatic fat embolism,
drug sensitivity, and systemic lupus erythematosus),
none is known to cause a clinically silent insult to the
brain, with the single exception of systemic lupus
erythematosus (SLE) 1281. Other collagen vascular diseases and acute immunopathogenic disorders may
cause cerebral vasculitis in infancy (e.g., rheumatoid
arthritis and polyarteritis nodosa; see {29] for review),
but these usually involve larger arteries in addition to
cortical arterioles and are almost invariably associated
with overt systemic symptomatology.
In SLE, the central nervous systemic vascular involvement nearly always includes arterioles and capillaries. During the early phase, the SLE vascular lesion
consists of an arteriole or capillary plugged by hyaline
thrombus, with little or no perivascular reaction. This
may be followed by reestablishment of the lumen, or
by a small surrounding zone of neuronal necrosis and
gliosis. The latter lesion resembles in every respect
those seen in four of our dyslexic cases, as well as the
brain lesions seen in "antiphospholipid" syndromes
Large-vessel involvement in the brain can also occur
in SLE, invariably as a consequence of embolization
from the heart, and with immediate, obvious clinical
It is now well recognized that microvascular cerebral
pathology with surrounding focal gliosis can occur in
SLE without neurological symptomatology {28}. We
have confirmed this observation in a whole-brain serial-section study of the brain of a young woman dying
of juvenile SLE, clinically apparent by age 8 years (unpublished data). Multiple microscopic cortical scars,
identical to those seen in dyslexics, were present in the
cerebral arterial border zones bilaterally, as well as portions of the middle cerebral artery territories. Many of
the scars were myelinated, suggesting the silent presence of brain involvement long before systemic symptoms of SLE appeared.
SLE can be present in newborn infants {31], usually
in the form of a skin rash, hemolytic anemia, thrombocytopenia, and congenital heart block. Clinical involvement of the central nervous system has never
been reported, nor have any rigorous pathological
studies of the brain been performed. The presumed
etiology of neonatal SLE is the placental transfer of
maternal antibodies 132). What is intriguing from the
viewpoint of our dyslexic subjects is that the majority
of mothers have no clinical evidence of SLE while
pregnant, although some may develop overt signs of
the disease months or years later {33}.
Given the close resemblance between the type and
distribution of cortical scars seen in our dyslexic subjects and those seen in patients with SLE, as well as the
silent fetal involvement seen in patients with neonatal
SLE, the possibility is raised that the dyslexic brain may
have undergone in utero the formation of multiple
microscopic cortical scars and, by extrapolation, neuronal ectopias.
The following are several lines of evidence supporting an immunopathogenic mechanism for the myelinated cortical scars and neuronal ectopias seen in dyslexics: (1) There is a large increase in the incidence of
learning disorders, particularly dyslexia, in the male
offspring of women with SLE f333. (2) Six of the eight
dyslexic subjects whose brains we studied, including
two of the three female cases reported here, have had
personal or family histories, or both, of clinically expressed atopic or autoimmune disease (e.g., ulcerative
colitis). (3) Focal cortical neuronal ectopias have been
noted in two strains of immunologically disordered
mice (NZB and BXSB) but not in controls {343. The
NZB mouse is considered a model for human SLE and
has been shown to have learning difficulties {35}.
While these various concurrent phenomena may be
circumstantial, they do strengthen the concept of a
cause-and-effect linkage between immunological disorders, cortical neuronal ectopias, and learning disorders.
The presently reported findings in three dyslexic subjects, of multifocal microscopic myelinated scars that
are (1) overlapping in distribution with neuronal ectopias in two cases and (2) identical in appearance with
the cortical lesions of SLE, give more concrete support
for an immunopathogenic contribution to the phenomenon of dyslexia.
Considering the human and animal evidence, we
propose that the putative dyslexic individual begins
with a familial predisposition to dyslexia (often along
the paternal line {36]), which is expressed through a
propensity to develop symmetrical temporal plana.
Additionally, there may be a familial tendency, possibly through the maternal line, to have autoimmune
and allergic disorders. A susceptible brain is exposed
during early development to antibodies, possibly maternally derived, directed to cerebral cortical capillary
and arteriolar endothelium, and possibly to other targets. Immune complex deposition on the endothelial
cell surface would lead to the formation of hyalineplatelet thrombi that occlude the lumen and produce
ischemia of the adjacent cortex. In regions where cortical perfusion is most marginal (e.g., arterial border
zones), metabolic requirements are exceptionally high,
or both, there would be tissue destruction. If the antibody-derived vascular plugs occurred during neuronal
migration, ectopias and brain warts, or perhaps frank
microgyria, would result. If the antibody circulation
began in or persisted into the third trimester and beyond, multifocal myelinated (and unmyelinated) scars
would develop in regions of greatest risk. Ultimately,
the pattern of learning disability that might evolve
736 Annals of Neurology Vol 28 N o 6 December 1990
would depend on the severity and localization of the
damage, the neural plasticity inherent in the system,
which provides for reorganization of circuitries, and
available compensatory cognitive strategies.
By this hypothesis, the presence alone of symmetrical temporal plana ( e g , most of the 24% of the normal population with symmetrical plana) or immunologically derived developmental brain pathology (e.g., the
Yakovlev brains with myelinated scars but asymmetrical plana) would not result in overt symptoms of dyslexia. Other forms of cerebral malfunction, independent of immunological disorders, could be involved in
some cases as well, for example, early traumatic brain
injury as in our Patient 1. A combination of individually subclinical or subtle untoward factors affecting
motivation, attention, and emotional makeup could tip
the scale and allow for the clinical expression of hereditary risk factors for dyslexia, such as the symmetry of
the planum temporale.
Connection between Developmental Brain Pathology
and the Cognitive Disorder of Dyslexia
Although all the dyslexic brains we have examined
demonstrate consistent symmetry of the planum temporale, and microscopic abnormalities that can be
dated to the periods of late neuronal migration and
subsequent cortical maturation, we cannot establish a
causal connection between the pathoanatomical findings and the cognitive disorder. If there are functional
consequences from the presence of the early cortical
lesions, their predominant location in orbital frontal,
lateral frontal, opercular, lateral temporal, and intrasylvim cortices might be expected to result in (among
other problems) language and attentional disorders,
disorders of planning and comportment, hyperactivity,
and disorders of auditory memory. This statement,
however, is based on knowledge of the effects of lesions in these areas occurring mostly later in life, and
does not take into consideration the remarkable reorganization effected by a younger brain (see, for example 1373, with mostly unknown consequences on
functional localization and functional capacity. The degree of brain pathology witnessed in the dyslexic
brains we have studied thus far would not be expected
to produce major cognitive deficits if acquired after
adolescence. We must postulate, therefore, that the
timing of the damage is such as to launch major reorganization of cortical circuits involved in language
function, perhaps through the preservation of transient
neurons and connections that would otherwise be ontogenetically eliminated 1371. We are currently addressing this hypothesis in our laboratory by using
murine and rat models of microdysgenetic cortical foci
created during late neuronal migration, and results are
supportive of the working hypothesis 138-40).
Supported in part by National Institutes of Health (NIH) Grants
HD-20806 and HD-19819, a grant from the Carl W. Herzog Foundation, and a grant from the Research Division of the Orton Dyslexia Society. P. H. performed this work while in Boston on a
sabbatical year, supported in part by the University of Ottawa Department of Pediatrics Research and Development Fund.
We thank Antis Zalkalns and Deborah Baker for technical assistance, William Baker, Eupil Choi, Regina Cicci, Drake Duane,
Gloria McAnulty, Marwet Rawson, Roger Saunders, Peter Schilder, Archie Silver, and Ana Sotrel for access to or interpretation of
anatomical and behavioral materials, or both, and Mohamad Haleem
of the Yakovlev Collection (Supported by N I H Grant NS-42303)
for access to normative specimens.
1. Galaburda AM, Kemper TL. Cytoarchitectonic abnormalities in
developmental dyslexia: a case study. Ann Neurol 1979;694100
2. Galaburda AM, Sherman GF, Rosen GD, et al. Developmental
dyslexia: four consecutive patients with cortical anomalies. Ann
Neurol 1985;18:222-2 33
3. Geschwind N, Levitsky W. Human brain: left-right asymmetries
in temporal speech region. Science 1968;161:186-187
4. Kaufmann WE, Galaburda AM. Cerebmcortical microdysgenesis in neurologically normal subjects: a histopathologic
study. Neurology 1989;39:238-244
5. Galaburda AM. The pathogenesis of childhood dyslexia. In:
Plum F, ed. Language, Communication, and the Brain. Association for Research in Nervous and Mental Disease series, vol66.
New York: Raven Press, 1987:127-137
6. Jacob H. Die feinere Oberflachengestaltung der Hirnwind
ungen, die Hirnwarzenbildung und die Mikropolygyrie. 2
Gesamte N e w 1 Psychiatr 1940;170:64-84
7 McBride MC, Kemper TL. Pathogenesis of four-layered micmgyric cortex in man. Acta Neuroparhol (Bed) 1982;57:93-
8 Galaburda AM, Corsigha J, Rosen GD, Sherman GF. Planum
temporale asymmetry: reappraisal since Geschwind and
Levitsky. Neuropsychologia 1987;25:853-868
9 Rosen GD, Sherman GF, Galaburda AM. Interhemispheric
connections differ between symmetric and asymmetric brain regions. Neuroscience 1989;33:52 5-5 33
10 Galaburda AM, Aboitiz F, Rosen GD, Sherman GF. Histological asymmetry in the primary visual cortex of the rat: implications for mechanisms of cerebral asymmetry. Cortex 1986;22:
11. Norman RM. Etat marbrP of the corpus striatum following birth
injury. J Neurol Neurosurg Psychiatry 1947;10:12-25
12. Malamud N. Status mannoratus: a form of cerebral palsy following either birth injury or inflammation of the central nervous
system. J Pediatr 1950;37:610-619
13. Azzarelli B, Meade P, Muller J. Hypoxic lesions in areas of
primary myelination-a distinct pattern in cerebral palsy. Childs
Brain 1980;7:132-145
14. Larroche JC. Fetal encephalopathies of circulatory origin. Biol
Neonate 1986;5 0:6 1- 74
15. Myers RE. Atrophic cortical sclerosis associated with status marmoratus in a perinatally damaged monkey. Neurology 1969;19:
16. Friede RL. Developmental Neuropathology. New York: Springer-Verlag, 1975
17. Johnston MV. Neurotransmitter alterations in a model of perinatal hypoxic-ischemic brain injury. Ann Neurol 1983;13:5115 18
Hwnphreys et al: Dyslexia in Women
18. Deckel AW, Robinson RG. Status marmoratus in fetal cortical
transplants. Exp Neurol 1986;91:212-218
19. Bignami A, Ralston RG. Myelination of hbrillary astragld processes in long term wallerian degeneration. The possible relationship to “status marmoratus.” Brain Res 1968;11:710-713
20. Oppenheim H, Vogt C. Nature et localisation de la paralyse
pseudobulhaire congenitale et infantile. J Psycho1 Neurol (Lyz)
21. Yakovlev PI, Lecours AR. The myelogenetic cycles of regional
maturation of the brain. In: Minkowski A, ed. Regional Development of the Brain in Early Life. London: Blackwell, 1967:
22. Caviness VS, Misson J-P, Gadisseux J-F. Abnormal neuronal
patterns and disorders of neocortical development. In: Gdaburda A, ed. From Reading to Neurons. Cambridge, MA: Massachusetts Institute of Technology PresdBradford Books, 1989:
401-43 5
23. Dvorak K, Feit J. Migration of neuroblasts through partial necrosis of the cerebral cortex in newborn rats: contribution to the
problem of morphological development and developmental period of cerebral microgyria. Acta Neuropathol (Bed) 1977;38:
203-2 12
24. Sidman RL,Rakic P. Neuronal migration, with special reference
to developing human brains: a review. Brain Res 1973;62:1-35
2 j. Taylor DC. Differential rates of cerebral maturation between
sexes and between hemispheres. Evidence from epilepsy. Lancet
26. Larroche JC, Ratavi-Encha F. Cytoarchitectonic abnormalities.
In: Myrianthopoulos NC, ed. Handbook of Clinical Neurology,
vol G(50). Malformations. New York: Elsevier, 1987:245-266
27. Lindenberg R. Tissue reactions in the gray matter of the central
nervous system. In: Haymaker W, Adams RD, eds. Histology
and Histopathology of the Nervous System, vol 1. Springfield:
Charles C Thomas, 1982:999
28. Johnson RT, Richardson EP. The neurological manifestations of
systemic lupus erythematosus. A clinical-pathologic study of 24
cases and review of the literature. Medicine 1968;47:337-369
738 Annals of Neurology
29. Moore, PM. Neurological complications of vasculitis. Ann
Neurol 1983;14:15 5- 167
30. Briley DP, Cod1 BM, Goodnight SH. Neurological disease associated with antiphospholipid antibodies. Ann Neurol 1989;
31. Watson RM, Lane AT, Barnett NK, et al. Neonatal lupus
erythematosus. A clinical, serological and immunogenetic study
with review of the literature. Medicine 1984;63:362-378
32. Lockshin MD, Bonfa E, Elkon K, Druzin MI.. Neonatal lupus
risk to newborns of mothers with systemic lupus erythematosus.
Arthritis Rheum 1988;31:697-701
33. m t a RG. Systemic lupus erythematosus: learning disability in
the male offspring of female patients and relationship to laterality. Psychoneuroendocrinology 1988;13:385-396
34. Sherman GF, Galaburda AM, Geschwind N. Cortical anomalies
in brains of New Zealand mice: a neuropathologic model of
dyslexia? Proc Natl Acad Sci USA 1985;82:8072-8074
35. Spencer DG, Humphries K, Mathis D, Lal H. Behavioral impairments related to cognitive dysfunction in the autoimmune
New Zealand Black mouse. Behav Neurosci 1986;100:353358
36. Geschwind N, Galaburda AM. Cerebral lateralization. Biological Mechanisms, Associations, and Pathology. Cambridge, MA:
Massachusetts Institute of Technology Press, 1987
37. Goldman-Rakic PS, Rakic P. Experimental modification of gyral
patterns. In: Geschwind N, Galaburda AM, eds. Cerebral
Dominance: The Biological Foundations. Cambridge, MA: Harvard University Press, 1984:179-192
38. Sherman GF, Stone JS, Press DM, et al. Abnormal architecture
and connections disclosed by neurofilament staining in the cerebral cortex of autoimmune mice. Brain Res (in press)
39. Sherman GF, Morrison L, Rosen GD, et al. Brain abnormalities
in immune defective mice. Brain Res (in press)
40. Sherman GF, Stone JS, Rosen GD, Galaburda AM. Neocortical
VIP neurons are increased in the hemisphere containing focal
cerebrocortical rnicrodysgenesis in New Zealand Black Mice.
Brain Res (in press)
Vol 28 No 6 December 1790
Без категории
Размер файла
1 399 Кб
development, women, patients, neuropathological, dyslexia, three, findings
Пожаловаться на содержимое документа