such as computed tomography and
magnetic resonance imaging do not
lead to the correct diagnosis, strong di¬
agnostic consideration should be given
to a vascular dissection, which must be
demonstrated by arteriography.
We believe that facial numbness
and dysesthesia represent previously
unemphasized features of internal
carotid artery dissection. Spontane¬
ous vascular dissections should be
considered in the differential diagno¬
sis of patients presenting with unilat¬
eral head and neck pain, oculosympa-

thetic paresis, dysgeusia, and facial
numbness.

carotid aneurysm: Raeder's paratrigeminal syndrome (group 2) case report. J Neurosurg

References

Spontaneous dissection of the cervical internal

1. Fisher CM: The headache and pain of spontaneous carotid dissection. Headache 1982;22:60\x=req-\
65.
2. Mokri B, Sundt TM, Houser OW: Spontaneous internal carotid dissection, hemicrania, and

Horner's syndrome. Arch Neurol 1979;36:677\x=req-\

680.
3. West TET, Davies RJ, Kelly RE: Horner's
syndrome and headache due to carotid artery
disease. Br Med J 1976;3:818-820.
4. Davis RH, Daroff RB, Hoyt WF: Hemicra-

1968;29:94-96.
5. Mokri B, Sundt TM, Houser OW, et al:

carotid artery. Ann Neurol 1986;19:126-138.
6. Thompson HS, Mensher JH: Adrenergic
mydriasis in Horner's syndrome: Hydroxyamphetamine test for diagnosis of postganglionic
defects. Am J Ophthalmol 1971;72:472-480.
7. Carpenter MB: Human Neuroanatomy. Baltimore, Williams & Wilkins, 1976.
8. Anderson CA, Collins GJ, Rich NM, et al:
Spontaneous dissection of the internal carotid
artery associated with fibromuscular dysplasia.
Am Surg 1980;46:263-266.

nia, oculosympathetic paresis and subcranial

Loss of Deep Dyslexic Reading Ability
From a Second Left-Hemispheric Lesion
David P. Roeltgen, MD

\s=b\ It has been hypothesized that the
residual reading ability in people with

deep dyslexia (an acquired dyslexia in
which the subjects make semantic paralexias, eg, child read as boy) utilizes right
hemispheric structures. A patient who
was deep dyslexic following an initial left
hemispheric stroke was studied. Following a new left hemispheric stroke, he lost
his residual reading ability. In this patient,
deep dyslexic reading abilities were
dependent on left hemispheric structures.

(Arch Neurol 1987;44:346-348)

Qubjects with deep dyslexia produce

semantic paralexias when reading
aloud. Other characteristics, includ¬
ing visual errors, function word sub¬
stitutions, derivational errors, dem¬
onstration of an inability to derive
phonology from print and reading
high imagery words better than low
imagery words, nouns better than
adjectives, adjectives better than
verbs, and nouns better than function
words, also occur as part of the syn¬
drome.1 Although there is agreement
Accepted for publication Oct 13, 1986.
From the Department of Neurology, University of Missouri-Columbia and the Harry S Truman Memorial Veterans Hospital, Columbia,

Mo.
Presented at the Eighth European Conference
of the International Neuropsychological Society,
Copenhagen, June 1985.
Reprint requests to Department of Neurology,
M741 Health Science Center, University of Missouri-Columbia, Columbia, MO 65212 (Dr Roelt-

gen).

general clinical presentation of
patients with deep dyslexia, there is
considerable controversy regarding
potential anatomic substrates under¬
lying their residual reading ability.
We addressed this issue by assessing
the residual reading ability in a
patient with deep dyslexia after he
on the

had a second cerebral lesion. Conse¬
quently, we were able to draw some
conclusions regarding the anatomic
substrate for his reading abilities fol¬
lowing his first stroke.
REPORT OF A CASE

A 54-year-old right-handed man had a
left cerebral infarction in 1981. He had a
transient right-sided paresis and mild sta¬
ble aphasia. He was studied 1lk years
following his stroke and except for a mild
aphasia, results of an examination were
normal. His spontaneous speech was mild¬
ly nonfluent with good articulation. His
comprehension for conversation and com¬
mands was good, but deteriorated with
increased complexity of syntax. Naming to
confrontation was good. Based on history
and results of the Western Aphasia Bat¬
tery (WAB)2 (Table 1), he could best be
described as a recovered Broca's aphasie.
He was considered deep dyslexic because
he made semantic errors when reading
aloud, especially when reading para¬

graphs.

When reading single words (n 420) he
made 58 errors: 3% semantic ("clothing"
for "fabric"), 14% derivational ("confi¬
dent" for "confidence"), 43% visual
("melt" for "man"), and 4% function word
substitutions ("should" for "such"). When
reading words in paragraphs (n 387) he
made 55 errors: 11% semantic, 9% deriva=

=

tional, 20% visual, and 35% function word

substitutions. Analysis of his ability to
read single words is presented in Table 2.
The patient had adequate comprehen¬
sion of single words and moderately good
comprehension of paragraphs, with a raw
score of 65 on the paragraph comprehen¬
sion subtest of the Peabody Individual
Achievement Test.3 On phonological read¬
ing tasks, he had difficulty reading nonwords (14 [23%] of 60 correct) but repeated
them well (20 of 20) (Battery of Linguistic
Analysis of Writing and Reading)4; (Bat¬
tery of Adult Reading Function [L. J.
Rothi, H. B. Coslett, K. M. Heilman, unpub¬
lished data, 1983]). He also made errors on
a visually presented rhyming task when
asked to say whether two words rhymed
(16 of 20 correct; chance equals ten of 20)
and on a picture-pointing task for nonwords that were homophonic for real
words (pseudohomophones) (Battery of
Adult Reading Function; 36 of 40 correct;
chance equals 13 of 40) (Table 3).
In September 1984, the patient sustained
a new stroke, with a new right hemiparesis
and marked worsening of his language
performance. The results of the rest of the
examination were unchanged. On exami¬
nation, his fluency and comprehension
were more impaired than previously, and
his repetition was limited to approximate¬
ly two word phrases. Naming ability
declined such that he was able to name
only rare objects with phonemic cues. Clin¬
ical evaluation and results from the WAB2
(Table 1) suggested that he could be best
classified as having a mixed aphasia.
His language performance was studied
at two, four, and six weeks following his
second stroke, and results from clinical
and formal examinations were constant.
He had fairly consistent scores on the
WAB2 (Table 1).

Downloaded From: http://archneur.jamanetwork.com/ by a Michigan State University User on 06/10/2015

He had little residual reading ability
after the second stroke. Whereas previous¬
ly he had read 100% of a list of highly
imageable nouns and 90% of function
words, he now read only 5% and 0%,

respectively, despite correctly repeating

95% of the nouns and 75% of the function
words. He correctly read only rare single
letters. Error analysis revealed that there
were no semantic paralexias and perhaps
one function word substitution. Some
errors had slight visual similarity to the
stimulus word ("dog" for "egg") but most
errors were unrelated real words ("stamp"
for "fruit") or neologisms ("staems" for
whatever). He could not read any nonsense
words, but repeated 50%. He had no pho¬
nological reading ability; he could not per¬
form the previously presented visual
rhyming task or the pseudohomophone
pointing task. He comprehended some sin¬
gle words, including occasional words that
he could not read aloud. Single word com¬
prehension on the Reading Comprehension
Battery for Aphasia5 was slightly greater
than chance.

Computed tomography (CT) performed

within 48 hours after the new stroke dem¬
onstrated well-defined areas of low densi¬
ty, one small one in the right frontal region
and one large one in the left posterior
temporoparietal region. The right frontal
lesion involved part of the inferior frontal
gyrus, the lower half of the precentrai
gyrus, and part of the mid and anterior
insula cortex. The left-sided lesion
involved the cortex surrounding the poste¬
rior half of the superior temporal sulcus as
well as the angular gyrus, and extended
subcortically to the lateral ventricle. In
addition, a less well-defined low-density
region was present in the left frontoparietal region. Two weeks later, a repeated CT
scan demonstrated enhancement of the left
frontal lesion. This lesion involved the
lower portions of the precentrai and postcentral gyri, the insula cortex, the posteri¬
or half of the superior temporal gyrus, and
part of the body of the caudate nucleus,
independent neurologists and a radiologist
interpreted the clinical and radiologie evi¬
dence as indicating two previous old ische¬
mie infarctions and a recent left frontoparietal ischemie infarction (Figure).
COMMENT

There is considerable controversy
regarding possible anatomic sub¬
strates subserving the residual read¬
ing ability in patients with deep dys¬
lexia, and much of this controversy
centers on the production of semantic

paralexias.612

Some authors suggest that the com¬
pensatory reading strategies used by

deep dyslexic patients, especially
those strategies responsible for the
production of semantic paralexias,
utilize, at least in part, reading mech¬
anisms located in the right hemi¬
sphere (right hemisphere hypothe¬
sis).7·8·11·12 Major support for this con¬
clusion is the qualitative similarity
between the reading ability of

Table 1.—Results of the Western Aphasia Battery*

Fluency

_Information
Initial
Second assessment
Second week
Sixth week

Comprehension

Repetition

Naming

assessment_00_6 _a2_7^_9.1
0.0
0.0

5.0

6.25

5.0

.f

'Adapted from Kertesz.2

.

4.5

1.5

.

tSimilar result on one subtest (Auditory Word Comprehension).

Table 2.—Analysis of Single-Word Reading Ability:
A Battery of Linguistic Analysis for Writing and Reading*

HighLow-Imagery
Orthographic
-, .Imageable Function Imagery ,_Nounsf Wordst Nounst Nounsj Adjectives^ Verbst Regular§ lrregular§
Correct, %
100_90_97_80_83_75_82_71

'Adapted from Roeltgen et al.4
tMatched for length, frequency, and regularity (n 40 nouns and function words).
^Matched for length and frequency (n 30).
§Matched for length, frequency, and word class (n 45).
=

=

=

patients with deep dyslexia and the
responses of some subjects, normal
and commisurotomized, to ortho¬
graphic imput presented only to the
left visual field.12·13 These groups all
have been shown to produce semantic

paralexias.
Alternatively, other authors sug¬
gest that the right hemisphere has
little or no reading ability, and that
the oral reading ability of deep dys¬
lexic patients depends on residual left
hemispheric mechanisms.914
Following his initial symptomatic
stroke (presumably the stroke that
caused the left parietal lesion), this
patient had a mild stable aphasia. He
also had a dyslexia characterized by
semantic, derivational, and visual
paralexias, as well as function word

Table 3.—Phonetic Reading

Test
Nonword oral reading from

Score
Correct/
Given
8/20

Battery of Linguistic
Analysis of Writing and
Reading (one and two

syllables)4

Nonword oral reading from

Battery of Adult Reading
Function (one, two, and
three syllables with com¬
plex phonemes, eg, blends)5

Rhyming task (real words)

16/20

(chance,
10/20)

Picture-pointing task (three
pictures) when stimulus
was a nonword homo-

36/40

(chance,
13/40)

phonic for a real word

substitutions. He also read nouns of

high imagery better than matched
nouns of low imagery, and read nouns

better than matched function words
and nouns better than matched adjec¬
tives, which he read better than
matched verbs (Table 2). In addition,
he showed difficulty in deriving pho¬
nology from print. He read only 14 of
60 nonwords and had difficulty on two

phonology recognition tasks, rhyming
and pseudohomophone picture-point¬
ing. These findings are consistent
with the diagnosis of deep dyslexia.
Following his most recent stroke,
the patient lost much of his speech
ability and almost all of his reading
ability, including reading of real
words, deriving phonology from print
and comprehension.
Computed tomography following
the last stroke indicated that he had
most likely sustained a recent left
frontal cerebral infarction. Although
no old CT scans were available for
comparison, the clinical and radiolog-

ic evidence indicate that the left fron¬
tal lesion was recent and the other
lesions were old. The size and location
of the right frontal lesion, close to or
involving the motor area, make it
unlikely that an acute ischemie stroke
in that location would be undetectable
clinically. Similarly, the new right
hemiparesis is consistent with a new
left hemispheric lesion. The nonchanging, well-defined lesions, poste¬
riorly on the left and on the right, as
found on the initial CT scan, are
inconsistent with the diagnosis of
acute ischemie infarction. In contrast,
the left frontal lesion that was initial¬
ly ill-defined and later enhanced is
typical for an acute ischemie infarc¬
tion.
These clinical and radiologie data
are consistent with the conclusion
that in this patient, the residual read¬
ing ability following the first stroke
was dependent on left frontal mecha¬
nisms.

Downloaded From: http://archneur.jamanetwork.com/ by a Michigan State University User on 06/10/2015

from the present patient, as well as
those from the patient described by
Heilman et al suggest that residual
mechanisms necessary to analyze
reading may be present in the left or
right hemisphere, but that in some
patients both mechanisms are dis¬
rupted by new left hemispheric
lesions.
This study was supported by Teacher Investi¬

gator Development award K07 NS00807 to the

University of Missouri-Columbia.

The author thanks Lisa Carr for helping test
the subject, and Evelyn Hampton for manuscript

preparation.

References
1. Coltheart M: Deep dyslexia: A review of the
syndrome, in Coltheart M, Patterson KE, Marshall JC (eds): Deep Dyslexia. London, Routledge
& Kegan Paul, 1980, pp 22-47.
2. Kertesz A: Western Aphasia Battery.
London, University of Western Ontario,

Second computed tomographic study after latest stroke. Stippled area is enhancing lesion,
consistent with recent ischemie infarction. Shaded areas are well-defined lucencies, consistent
with old ischemie infarctions. Brain slices were plotted using technique of Damasio and
Damasio." The lateral view was extrapolated from transverse sections and plotted on lateral
view of left hemisphere (modified after Gado et al'8).

The right hemisphere hypothesis
for reading includes at least two com¬
ponents, analysis and response.10
Analysis is thought to include those
mechanisms necessary for perception
and description of the stimulus word
so as to generate a semantic code. This
information is then thought to be
transferred to the left hemisphere for
response. The response may include
oral reading and/or lexical decision.
Prior to his last stroke, our patient
was able to perform analysis (compre¬
hension of written words and slightly
better than chance ability to point to
pictures when pseudohomophones
were presented) and production (oral
reading). Following the last stroke,

despite retaining some speech ability
(comprehension, single-word repeti¬

tion, and one-to-three-word sponta¬
neous utterances), he lost ability to
perform both analysis and production
for reading. Such findings are most

consistent with the conclusion that, in
this patient, reading analysis and pro¬
duction, including lexical access for
comprehension and production of
semantic paralexias, were dependent
on left hemispheric mechanisms that
were present after his earliest aphasie
stroke.
There are at least two ways that
this patient's residual reading abili¬
ties may have been dependent on left
hemispheric mechanisms. First, those
mechanisms may have been present in

the left hemisphere and subsequently
disrupted by the second left hemi¬
spheric stroke. Alternatively, the sec¬
ond left hemispheric stroke may have
disrupted access to language mecha¬
nisms that remained intact in the
right hemisphere. The second expla¬
nation is unlikely because the occipi¬
tal lobes, the temporo-occipital re¬
gion, and the splenium of the corpus

callosum, structures usually thought
to be necessary for the perception of

and interhemispheric transfer of visu¬
ally perceived written language,15
were intact after the second left hemi¬
spheric stroke. Consequently, percep¬
tion and interhemispheric transfer of
visual information, including reading,
should have remained intact after the
second left hemispheric stroke.
Although these data are consistent
with the conclusion that this patient's
deep dyslexic reading abilities, both
analysis and production, depended on
intact left hemispheric functions,
such a conclusion may not be true for
all patients with deep dyslexia. Heilman et al16 described a patient with a
global aphasia following a left hemi¬
spheric stroke. Their patient had pre¬
served reading comprehension that
was lost after a new right hemispheric
stroke. Their results indicate that,
unlike the present patient, some
patients may have mechanisms
present in the right hemisphere capa¬
ble of analysis of reading. The results

1980.
3. Dunn LN, Markwardt FC: Peabody Individual Achievement Test. Circle Pines, Minn, American Guidance Service, 1970.
4. Roeltgen DP, Cordell C, Sevush S: A battery
of linguistic analysis for writing and reading. Int
Neuropsychol Soc Bull, October 1983, p 31.
5. La Pointe LL, Horner J: Reading Comprehension Battery for Aphasia. Tigard, Ore, CC
Publications, 1979.
6. Saffran EM, Bogyo LD, Schwartz MF, et al:
Does deep dyslexia reflect right-hemisphere
reading? in Coltheart M, Patterson K, Marshall
JC (eds): Deep Dyslexia. London, Routledge &
Kegan Paul, 1980, pp 381-406.
7. Coltheart M: Deep dyslexia: A right hemisphere hypothesis, in Coltheart M, Patterson KE,
Marshall JC (eds): Deep Dyslexia. London, Routledge & Kegan Paul, 1980, pp 22-47.
8. Landis T, Regard M, Graves R, et al: Semantic paralexia: A release of right hemispheric
function from left hemispheric control? Neuro-

psychologica 1983;21:359-364.
9. Marshall JC, Patterson KE: Semantic paralexia and the wrong hemisphere: A note on
Landis, Regard, Graves, and Goodglass. Neuropsychologia 1983;21:425-427.
10. Patterson K, Besner D: Is the right hemisphere literate? Cog Neuropsychol 1984;1:315\x=req-\

341.
11. Rabinowicz B, Moscovitch M: Right hemisphere literacy: A critique of some recent

approaches. Cog Neuropsychol 1984;1:343-350.
12. Zaidel E, Schweiger A: On wrong hypotheses about the right hemisphere: Commentary on
Patterson K, Besner D, "Is the right hemisphere
literate?" Cog Neuropsychol 1984;1:351-364.
13. Regard M, Landis T: Experimentally
induced semantic paralexias in normals: A property of the right hemisphere. Cortex 1984;20:263\x=req-\
270.

14. Besner D: Deep dyslexia and the right-$
hemisphere hypothesis: Evidence from the USA
and the USSR. Can J Psychol 1983;37:565-571.
15. Friedman RB, Albert ML: Alexia, in Heilman KM, Valnestein E (eds): Clinical Neuropsychology. New York, Oxford University Press Inc,

1985, 49-73.

16. Heilman KM, Rothi L, Campanella D, et al:
Wernicke's and global aphasia without alexia.
Arch Neurol 1979;36:129-133.
17. Damasio H, Damasio A: 'Paradoxic' ear
extinction in dichotic listening: Possible anatomic significance. Neurology 1979;29:644-653.
18. Gado M, Hanaway J, Frank R: Functional
anatomy of the cerebral cortex by computed
tomography. J Comput Assist Tomogr 1979;3:1\x=req-\
19.

Downloaded From: http://archneur.jamanetwork.com/ by a Michigan State University User on 06/10/2015