702

J Neurol Neurosurg Psychiatry 2001;70:702–710

LETTERS TO
THE EDITOR
Posterior alexia after right
occipitotemporal infarction
Posterior or pure alexia is an uncommon
acquired reading disturbance in which the
loss of the ability to read is not associated
with other language deficits.1 It has been
reported almost without exception after left
(dominant) hemispheric lesions involving
visual temporo-occipital pathways and splenium of the corpus callosum, and is considered to be a strong lateralising feature. The
explanations of this deficit are on the basis of:
(a) a disconnection syndrome—that is, the
lesion prevents visual information from
reaching the language area, or (b) the
presence of a lateralised visual language
association area residing in the dominant
posterior inferior temporal lobe. We describe
a patient with an atypical interhemispheric
organisation who developed pure alexia and
features of apperceptive agnosia after a right
occipitotemporal infarction.
The patient is a 71 year old man, with a
college degree education. He writes, uses a
fork, spoon, or knife with the right hand, but
prefers to use his left hand or leg for every
other activity. He has no antecedents of sinistrality in first degree relatives. At the age of 68
he had a myocardial infarction complicated
with an interventricular communication and
heart failure. He underwent cardiac surgery
and recovered. Immediately after surgery his
examination showed a left visual hemianopia
and he complained of inability to read. However, no neurological assessment was required at that time and no further studies
were conducted.
Three years later he was referred to the
neurology clinic because of a syncopal
episode and further attention was then paid
to his cognitive complaints. He referred that
during the first year after his surgery letters
were just unmeaning lines for him and he had
Table 1

some confusion with numbers. Later, he
learnt gradually to discriminate numbers and
to recognise many letters but was still unable
to put them together into syllables or
meaningful words.
He was given a cognitive examination
(quantified data are summarised in table 1).
He was cooperative and fully oriented. On a
line bisection task he scanned from left to
right and upside down bisecting all lines. He
had no visuospatial disorientation according
to his performance in tests of the visual object
and space perception battery.2 Visuospatial
perception and visuoconstructive tasks disclosed features consistent with apperceptive
agnosia. He was able to recognise and copy
simple shapes and figures (circle, cube...) but
was impaired in recognising schematic drawings (for example, from the Boston naming
test or cookie theft scene) or when he tried to
copy more complex figures such as a house or
Rey’s figure. The main deficit pertained to
piecing together separate visual stimuli, especially line drawings, into a whole percept.
Recognition of real objects and pictures of
famous people was appropriate. Colour
matching was normal but he committed
errors when asked to name a colour on confrontation or to point to a certain colour.
Because he reported to have always had some
diYculty with colour discrimination it was
not certain the interpretation of this deficit
was achromatopsia.
Spontaneous speech, comprehension, and
repetition were normal. Confrontation naming with the Boston naming test was in the
lower range disclosing very mild word finding
problems (in addition to several failures
clearly due to perceptual diYculties). Handwriting was meaningful and syntactically correct both spontaneously and to dictation. He
was perfectly able to copy single letters and
words. When trying to read, the patient was
able to recognise many letters but he
committed frequent errors such as between
M and V, R and A, or H and R. He was
unable to read syllables or simple words. On
the contrary, he recognised with ease three
syllable words spelled aloud by the examiner.
He was able to read and to recognise
numbers. Calculation was normal, with
isolated errors. He had no problems distinguishing between right and left sides.

Cognitive assessment of the patient
Results (correct/total)

Interpretation

30/30

No neglect
Appropriate spatial analysis

Spatial attention/orientation:
Line cancellation task
VOSP†
Shape detection screening test
Dot counting
Number location
Position discrimination

20/20
10/10
8/10
17/20

Visuoconstructive abilities:
Copy of simple drawings
Copy of complex Rey figure

2/3*
21/36 (centile 1)*

Visual recognition
Simple shapes
Schematic drawings
Real objects
Faces (photographs)
Colour matching
Colour discrimination

5/5
10/15*
5/5
5/5
5/5
6/10

Language
Boston naming test
Writing sentences
Reading words
Recognition of spelling aloud words

22‡/30
3/3
0/10
6/6

Apperceptive agnosia *

No prosopagnosia
Achromatopsia?
No aphasia
Mild naming deficit
No agraphia
Posterior alexia

†Visual object and space perception battery; ‡Plus three items after phonemic clues (includes just the errors
due to naming impairment).

www.jnnp.com

Brain CT shows an infarction in the territory of
the right posterior cerebral artery involving
occipital visual cortex and periventricular white
matter extending to the forceps major of the
corpus callosum.
Brain CT showed an old infarction in the
territory of the right posterior cerebral artery
involving the occipital visual cortex and
periventricular white matter extending to the
forceps major of the corpus callosum (figure).
No lesions were found in the left hemisphere.
The development of posterior alexia and
apperceptive agnosia after a right occipitotemporal infarction is exceptional—only one
case has been reported, in a left handed person, by Gloning et al,3 and discloses a rather
atypical interhemispheric organisation in this
patient. Firstly, he had a right hemispheric
dominance for language, which is uncommon
even in left handed patients. The neuropsychological profile of his reading disorder
showed the typical features of a posterior
alexia due to dominant hemispheric damage.
His alexia could not be explained on the basis
of neglect, visuospatial disorientation, simultanagnosia, or a global visual agnosia. Neither
could it be due to his diYculties with
complex visual perception, as he was perfectly able to copy words that afterwards he
was unable to read. It is noteworthy that he
was not able to recover any capability to read
relying on his intact left temporo-occipital
cortex.
The second remarkable deficit was consistent with the descriptions of Lissauer (1889)
and of Bauer and Rubens (1985) of apperceptive agnosia.4 He had clear diYculties in
perceiving the overall configuration of complex line figures or schematic drawings, which
impaired both their copy and their recognition. This deficit could not be attributed to
primary visual deficit, deficit in visual scanning, visuospatial disorientation, or just to
constructional apraxia. Apperceptive agnosia
is classically related to right temporooccipital lesions whereas associative agnosia
occurs after similar left lesions. All previously
reported cases of posterior alexia, when
accompanied by agnosic features, have been
of the associative type consistent with left
damage. On the contrary, this patient encompassed a striking combination of agnosic
features—that is, a selective associative agnosia for words plus features of apperceptive
agnosia for other complex visual stimuli. This
combination of deficits suggests that reverse
hemispheric dominance for language was not
accompanied by reverse dominance for other
functions.
Finally, the clinicoanatomical relation in
this case mirrors the posterior alexia after left

J Neurol Neurosurg Psychiatry 2001;70:702–710
hemispheric lesions,5 suggesting a similar linguistic organisation when the right hemisphere is dominant. In this patient damage to
the posterior transcallosal fibres on their way
to the left dominant angular gyrus and/or to
the dominant occipitotemporal gyrus was
also the crucial anatomical substrate in
producing the syndrome of pure alexia.
ESTRELLA GÓMEZ-TORTOSA
ANTONIO DEL BARRIO
Department of Neurology, Fundación Jiménez Díaz,
Avda Reyes Católicos 2, Madrid 28040, Spain
Correspondence to: Dr E Gómez-Tortosa
egomezt@meditex.es
1 Benson DF. Alexia. In: JAM Fredericks, ed.Handbook of clinical neurology: clinical neuropsychology. Vol 1. Amsterdam: Elsevier, 1985:433–
55.
2 Warrington EK, James M. The visual object and
space perception battery. Bury St Edmunds:
Thames Valley Test Company, 1991.
3 Gloning I, Gloning K, Seittelberger F, et al. Ein
Fall von reiner Wortblindheit mit Obduktionsbefund.Wiener Zeitung Nervenheilkd 1955;12:
194–215.
4 Bauer RM, Rubens AB. Agnosia. In: Heilman
KM, Valenstein E, eds. Clinical neuropsychology.
New York: Oxford University Press, 1985;187–
241.
5 Binder JR, Mohr JP. The topography of callosal
reading pathways. A case control analysis.
Brain 1992;115:1807–26.

De novo partial duplication of 17p
associated with Charcot-Marie-Tooth
disease type 1A
Charcot-Marie-Tooth disease type 1A
(CMT1A) is the most frequent form of CMT
(also known as hereditary motor and sensory
neuropathy, HMSN). The inheritance is
autosomal dominant and is usually associated
with a duplication at chromosome 17p11.2.
This region contains the gene of the peripheral myelin protein 22 (PMP22) and an
increased concentration of PMP22 seems to
be responsible for neurological phenotype in
patients with 17p11.2 duplication.1 Therefore, a gene dosage eVect is the most likely
pathogenetic mechanism for CMT1A.
Partial trisomy 17p is an unusual chromosomal disorder rarely reported in Europe, the
clinical features of which have not been definitely established.2–6 Growth retardation,
craniofacial anomalies, and developmental
delay seem to constitute a characteristic phenotype associated with this condition.6
Moreover, the identification of patients with
17p trisomy with a uniform demyelinating
neuropathy similar to that seen in patients
with CMT1A, has provided further in vivo
evidence supporting the gene dosage
hypothesis.7–9 In this report, we describe an
additional case of a child with de novo partial
duplication of 17p associated with CMT1A.
A 7 year old boy was born at 41 weeks of
gestation by spontaneous vaginal delivery
after artificial amniorrexis. There was prenatal history of oligohydramnios. His parents
were healthy and non-consanguineous and
the family history for neurological or genetic
diseases was negative. At the time of birth, the
apgar score was 7 because he had meconium
Table 1

703

aspiration syndrome. Physical examination
showed low weight (2.755 kg, percentile 3),
normal length (49.5 cm, percentile 25) and
microcephaly (33 cm, percentile 3). Facial
dysmorphic characters included hypertelorism, downslanted palpebral fissures, migrognathia, plane philtrum, and prominent
ears. At the age of 3 months, there was a continued growth failure (weight 4.100 kg,
percentile 3; length 56.5 cm, percentile 3)
and a right microphthalmia and divergent
strabismus were noted. At 8 months of age,
there was a marked global developmental
delay. At that time, cerebral echography,
brain auditory evoked potentials, and MRI
were within normal limits. Subsequently, the
patient was sent to a centre for mentally
retarded children where he has been taking
intensive schooling including psychomotor
and speech therapies.
In June 1999, the patient was referred to
our hospital for electrophysiological evaluation to test the possibility of impairment of
the peripheral nervous system. At that time,
the patient had no weakness, gait disturbances, or sensory complaints but the diagnosis of partial duplicaton of 17p was the
reason for testing. Neurological examination
showed facial dysmorphic characters (as
mentioned above), remarkable hyperactivity,
pronounced expressive speech delay, generalised hyporeflexia, pes cavus, and incipient
tibioperoneal atrophy. Gower’s sign was
negative. Enlarged nerves were not found.
There was no sensory deficit, but detailed
testing could not be performed. Electrophysiological studies were done with a
MEDELEC Shappire 4 electromyographer.
The EMG showed a mixed pattern and polyphasic motor unit potentials. No fibrillation
potentials or positive sharp waves were found.
Motor nerve conduction velocities were
obtained on the patient and his mother. The
patient’s motor conduction velocities were
uniformly slowed in the limbs and F waves
were markedly delayed (table 1).
Sensory nerve conduction measurements
were not done because of the child’s anxiety .
These findings were consistent with the presence of a demyelinating neuropathy in
keeping with the diagnosis of CMT1. By
contrast, his mother’s median and ulnar
motor nerve conduction velocities were
normal (53.7 m/s and 70 m/s, respectively).
Chromosome analysis of the patient
showed the presence of extra chromosomal
material in 17p. Fluorescence in situ hybridisation confirmed the diagnosis of partial
duplication of the short arm of chromosome
17 and demonstrated that extra material was
located in the region beyond band 17p13.3.
The mother’s and father’s karyotypes were
normal. Therefore, these findings were compatible with a de novo interstitial duplication
of 17p. Duplication analysis was done using
microsatellite markers (D17S921 and
DS17S955) from within the region of chromosome 17p11.2, known to be duplicated in
CMT1A and containing the PMP22 gene.
Gene dosage experiments with these microsatellite markers demonstrated that the patient

was a carrier of a chromosome 17 with
17p11.2 duplication, which is often associated with the CMT1A phenotype. His
parents were heterozygous for these markers
but signals of equal intensity were found for
both alleles.
Most of the clinical features described here
are similar to those previously detailed in several children with de novo partial duplications of 17p.2–6 All of them had low birthweight, growth retardation, developmental
delay, and craniofacial dysmorphic characters
such as hypertelorism, downslanted palpebral
fissures, smooth philtrum, and abnormal
ears. In addition, our patient had divergent
strabismus, microphthalmia, pronounced hyperactivity, and remarkable speech delay. To
date, only a few cases of this rare chromosomal condition have been published and
clinical data are scarce. However, our finding
supports the hypothesis that partial trisomy
17p has a typical phenotype.5 6 Recently,
some authors have also shown abnormal
nerve conduction velocities in patients with
this disorder.7–9 Therefore, we carried out an
electrophysiological study to examine the
status of the peripheral nerves, detecting the
existence of a demyelinating neuropathy in
keeping with the diagnosis of CMT1. It is
important to emphasise that clinical manifestations such as weakness, gait disturbances,
and sensory deficits were absent in the child.
This fact strongly supports the need for electrophysiological evaluation in all patients
with trisomy 17p even when the anomalies
denoting the existence of peripheral nervous
system impairment are absent.
Lupski et al1 studied a patient with 17p
partial trisomy with decreased nerve conduction velocities whose molecular analysis
showed the CMT1A duplication. On the
basis of their findings, the authors hypothesised that the CMT1A phenotype resulted
from a gene dosage eVect. Later Chance et al 7
ratified the association of trisomy 17p and
CMT1A and provided further evidence supporting a gene dosage mechanism. Since
then, only a few cases have been published.8 9
One of these studies examined the presence
of PMP22 gene duplication in four patients
with partial trisomy.8 Two of these patients
showed slowing of conduction velocities and
exhibited PMP22 gene duplication. By contrast, in two patients with normal conduction
velocities the PMP22 gene was not duplicated. This investigation demonstrated that
the presence of demyelinating neuropathy is
directly associated with PMP22 gene duplication and corroborated the gene dosage
model.
In summary, we describe another case of a
patient with 17p trisomy whose electrophysiological evaluation showed the existence of a
demyelinating neuropathy type CMT1. Our
finding contributes support to the idea that
trisomy 17p is responsible for a typical
phenotype and that nerve conduction studies
should be included in the routine examinations in all patients with this chromosomal
disorder. In addition, our patient adds more
in vivo evidence supporting a gene dosage
eVect in CMT1A.

Results of electrophysiological studies
We are grateful to Dr Orera for the initial
cytogenetic evaluation and Dr O Combarros for his
kind contribution to the bibliographic review.

Motor conduction values
Nerve

Ampd (mV)

Ampp ( mV)

DML ( ms)

CV (m/s)

FL (ms)

Median
Peroneal
Tibial

3.4
3.6
2.9

2.1
2.8
2.5

8.8
10.5
8.3

14.1
16.0
14.1

ND
77.9
83.9

www.jnnp.com

J L FERNÁNDEZ-TORRE
Servicio de Neurofisiología Clínica, Hospital de
Cabueñes, Cabueñes s/n, 33394 Gijón, Spain
B OTERO
Servicio de Neuropediatría