Neuroimaging Findings in a
Patient Recovering from Global
Alexia to Spelling Dyslexia
S. Lanzinger

R. Oettli

Department of Neurology
Kantonsspital St. Galien
St. Galien, Switzerland

Department of Nuclear Medicine
Kantonsspital St. Galien

B. Weder
Department of Neurology
Kantonsspital St. Galien

C. Fretz
Department of Radiology
Kantonsspital St. Galien

ABSTRACT
The authors report findings in a 67-year-old right-handed man
who had an ischemic infarct in the territory of the left posterior
cerebral artery. The clinical manifestation consisted mainly of
total alexia without agraphia. The patient gradually recovered,
subsequently showing the syndrome of spelling dyslexia. Cerebral MR-images revealed a circumscript infarction of medial and
basal parts of left temporal lobe. In the acute stage [99 mTc]HMPAO SPECT was characterized by a diminished uptake in the
definitely infarcted area and hyperfixation in the region of the
left forceps major. Because high retention of HM-PAO indicates
potentially salvageable tissue after an ischemic event, the depicted area might be correlated with the recovery of function.
Thus, the authors' neuroimaging data give further support to
the assumption that the left forceps major is a critical area for
global alexia, whereas spelling dyslexia is due to involvement of
the left medio-basal temporal lobe.
Key words: global alexia, spelling dyslexia, recovery of function,
[

99

mTc]HM-PAO SPECT, HM-PAO hyperfixation, magnetic reso-

nance imaging.
Lanzinger S, Weder B, Oettli R, Fretz C. Neuroimaging findings
in a patient recovering from global alexia to spelling dyslexia.

J Neuroimaging 1999;9:48-51

Global alexia without agraphia is a rare manifestation of
occipital ischemic infarction, first described by the French
neurologist Dejerine in 1893. 1 Since that time, further
cases have been added analysing the underlying conditions in order to elucidate the pathogenesis of the syndrome.2 We report a new case and neuroimaging data.
Patient Report

A 67-year-old right-handed man had an ischemic infarct in
the area supplied by the left posterior cerebral artery after
Received Apr 2, 1998, and in revised form Oct 1, 1998.
Accepted for publication Oct 7, 1998.
Address correspondence to Dr Weder, Department of Neurology, Kantonsspital St. Gallen, CH-9007 St. Gallen,
Switzerland.
48

transient premonitory symptoms such as incapacity to
read aloud, vertigo, nausea, and vomiting. Hours later he
realized that he had lost the capability to read.
After admission the patient presented with global alexia without agraphia. Although he was able to write subtly about a complex issue, he was unable to read what he
had written. The written text was correct regarding grammar and content, but there were a few phonetic errors.
Apart from slight anomia, speech was fluent and comprehension intact. Additional symptoms were a superior quadrantanopsia to the right and minimal ataxia of the trunk.
In general, the patient complained of amnestic difficulties.
Magnetic resonance imaging (MRI) was performed
on day 2 of the affection and additionally, a singlephoton emission computed tomography (SPECT), using
99
[ mTc]HM-PAO (hexamethyl prophylene amine oxime),
on day 8. Cerebral Tl- and T2-weighted MR-images
showed a circumscript cerebral infarction of the left medio-basal part of the temporal lobe (Fig lA, C) extending
to the occipital horn of the lateral ventricle. No definite
infarction was disclosed in the adjacent occipital white
matter, there was only minor ischemic signal alteration in
the acute phase (Fig lB, D). The [99 mTc]HM-PAO
SPECT study showed a mixed pattern: in the medio-basal
part of the left temporal lobe, where definite infarction
had occurred, HM-PAO-uptake was diminished (Fig lE);
and, in a circumscript zone of the left occipital white matter involving the forceps major, a marked HM-PAO hyperfixation was found, compatible with hyperperfusion after early reperfusion (Fig 1F). 3 •4 These areas correspond
to subterritories of the lateral branch of the posterior cerebral artery (see discussion), whereas the calcarine area
and the parieto-occiptal cortex, supplied by the medial
branch, were not affected in any of the two examination
modalities. 5
The patient gradually recovered over the next 3 months
showing two major reading strategies, i.e., a laborious letter-by-letter method 2 and later the so-called "whole-word
procedure" 6 : reading his hand-writing he first spelled the
letters and then put them together to form a word, sometimes misreading the beginning or the end of the word.
Later he tried to remember a word by associating the
occasion on which he wrote it. Finally, he was able to
recognize words faster then before, as he remembered
more and more the symbols representing a whole word.
Initially this was possible for short nouns only, later also
for longer words of any type.
Conditions did not change in the course of three years.
The compensatory strategies became more routine allowing more or less fluent, albeit slow reading. When fatigued
the patient sometimes confused letters such as "b," "p,"
and "d." Neuropsychological testing, at this stage, gave
evidence for intact visual object and visual space percep-

Copyright© 1999 by the American Society of Neuroimaging

Fig 2.

Fol low-up examination showing T2-w eighted MRimages (A, B) at levels comparable to Fig 1, 3 years later. Because
the examination was not coregistered with the f irst image plan is
slightly antef lected in compa rison to Figure 1. Encephalomalacic
area in the left medio-basal temporal lobe (A) and gliosis in the
occipital w hite matter (B)

Fig 1.

Synopsis of correlated T2- (A, B) and T1-weighted (C, D)
MRis, and [99mTc)HM-PAO brain SPECT slices (E, F) from days 2
and 8, respectively. In the left column definite infarction is demonstrated in the medio-basal part of the left temporal lobe by
MR-images (A, C), concordant with low HM-PAO uptake by SPECT
(E). The right column (B, D, F) shows the findings at the level of t he
posterolateral white matter. Prominent findings are hyperfixat ion
of HM-PAO by SPECT involving the forceps major, whereas in t he
corresponding MR-images subtle ischemic changes are seen in t he
occipital white matter. Image plan is parallel to the line between
anterior and posterior commissure. (For methodical questions see
reference 8). View is from below, so that the right hemisphere
appears on the left.

tion (VOSP). Visual-constructive abilities (Rey-Osterrieth
Complex Figure Test) were normal as well as figural learning and memory as evinced by the Rey Visual Design
Learning Test (RVDLT). There was no apraxia. Aachen
Aphasia Test and Basel-Minnesota Test showed normal
findings except for a few phonetic errors when reading
aloud ("" 5% ). This errors concerned preferentially abstract words and nonwords. Hand-writing was perfectly
preserved. Processing of numerals was intact. Learning
and memory for written (CERAD, Consortium to Establish a Registry for Alzheimer's Disease) and listened word
lists (CVLT, California Verbal Learning Test) were significantly impaired. In the Chapman-Cook Reading Test,
reading aloud was slowed and time consumed for comprehension of a text was significantly prolonged. Reading his

own written text was considerably slowed over reading a
printout of the same text (75 words/min vs. 106 words/
min). The ability to read and write shorthand was definitely lost. Summarizing, the observed behavioral pattern
suggested severe difficulties both for learning and recalling of verbal material. Obviously, difficulties of reading
increased in parallel to formal reductions of characters
used, from printed material to hand-written texts and, finally, to texts in shorthand.
Follow-up MRI did not reveal new aspects (Figs 2A,
B). Postischemic encephalomalacia had occurred in the
left medio-basal temporal lobe, whereas only a slight
gliomatous alteration was detected in the occipital white
matter sparing the forceps major.

Fig 3.

Parametric map of [ 9 9 mTc]HM-PAO brain SPECT, indicating a cluster of deviating pixels wit h a z-score of >+2 .5, after
integration into the appropriate anatomical MR-image. The brain
studies were spatially standardized according to t he ref erence
brain of CBA. The symbol (* ) indicates the lower part of the splenium corporis callosi . Note partial confirmation of the qualitat ive
assessment (see Fig 1F) by the quantitative analysis, showing significantly enhanced HM-PAO up-take in the left post ero-lateral
white matter including forceps major. According t o Tala irach and
Tournoux, 14 t he center of deviat ing pixels has t he coordinates x =
- 24, y = - 66, z = 8.

Significant Case Report: Lanzinger et al: Alexia and Neuroimaging

49

Image Analysis

For quantitative analysis and localization, SPECT data
were merged into a dedicated Computerized Brain Atlas
(CBA) system. 7 The system makes it possible to reorientate and standardize image data and, thus, to merge functional data with anatomical details of MR-images. 8 With
regard to SPECT data z-scores were calculated on a pixel
basis for each slice in comparison to a population of 24
normal volunteers (age range 25 to 63 years). By thresholding parametric images were obtained, indicating deviating pixels: by this analysis the zone of HM-PAO hyperfixation could be located outside the definitely infarcted
area. According to the z-statistics performed, HM-PAO
hyperfixation was confined to the adjacent white matter of
the temporo-occipital junction. It corresponded to a large
cluster of pixels, with a z-score of more than +2.5, which is
demonstrated in Figure 3 after integration into the corresponding MRI slice. 8
Discussion

Pure alexia without agraphia was first reported by Dejerine in 1893,1 defined as a syndrome characterized by the
inability to read aloud or to comprehend words in the
absence of aphasia. The traditional account attributes the
syndrome to a disconnection between circumscript areas
of the left and right hemisphere? Thus, lexical and semantic analysis of the right hemisphere is not accessible to the
language processors on the left. 6 While a few similarly
affected cases have been described by others, it is clear
that most patients with left posterior hemisphere injury do
not present with a uniform manifestation of global alexia.6·9-12 Binder and Mohr 2 differentiated between global
alexics and spelling dyslexics. The latter describes exactly
the stage of recovery in our patient. Characteristically he
used the residual capacities to overcome his deficit: first,
he predominantly applied the letter-by-letter method, and
later on the so-called "whole-word procedure" as described above. 6 Correlating reading deficits with lesion topography typical lesion sites within the left posterior cerebral artery territory were described: (1) normal readers
had lesions in the medial and ventral occipital lobe, sparing dorsal white matter and the medial-ventral temporal
lobe; (2) global and permanent alexia occurred only when
the splenium of the corpus callosum, forceps major or the
white matter above the occipital horn of the lateral ventricle were injured. (3) patients with "spelling dyslexia"
had lesions of the medial and ventral temporal lobe?
While our patient presented with global alexia, a mixed
perfusion pattern was present in the [99 mTc]HM-PAO
SPECT study, affecting the vascular territory of the lateral
branch of the posterior cerebral artery. 5 According to
Salomon and Huang, 13 this territory is subdivided into
subterritories supplied by an inferior temporal and an occipitotemporal artery. The two distinct perfusion patterns
50

Journal of Neuroimaging Vol 9 No 1 January 1999

indicated quite different prognosis for the affected brain
tissue: at the site of low HM-PAO uptake, at the left medio-basal temporal lobe, ischemic infarction was evident
even by MR-imaging in the acute stage. In contrast, the
zone of HM-PAO hyperfixation, in the white matter of the
occipita-temporal junction, recovered almost completely
as verified by follow-up MR-imaging. We suggest that recovery is due to early reperfusion as observed in recent
SPECT studies. However, since SPECT was performed at
a later stage of subacute infarction, reparative mechanisms, i.e., granulomatous tissue, might have accounted
additionally for enhanced HM-PAO uptake in the ischemic area? In conclusion, HM-PAO hyperfixation during
the acute stage of stroke might represent an index of endogenous neuroprotection by reduced oxidative stress
and, thus, indicate a favorable prognosis_3.4 Our observation is in accordance with the basis of the theory that the
splenium of the corpus callosum and the left forceps major
are critically involved in global alexia, and that a lesion of
the left medial and ventral temporal lobe is associated
with spelling dyslexia?
In conclusion, the mixed pattern of HM-PAO retention
in the SPECT study provided the opportunity to understand the partial recovery from global alexia. It showed a
differential affection of the susceptible areas, in which
HM-PAO hyperfixation delineated the potentially reversible part of the syndrome. With regard to Figure 3, we
suggest a temporary deafferentiation of the left occipitoparietal association cortices.
This work was supported by a grant from Ciba-GeigyJubilaumsstiftung. The authors appreciate the careful neuropsychological testing performed by Erika Forster from the Neurology Department of Kantonsspital St. Gallen.
References
1. Dejerine J, Vialet N. Contribution a !'etude de la localisation anatomique de Ia cecite verbale pure. Comptes rendus
hebdomadaires des seances et memoires de Ia societe de
biologie 1893;5:790-793
2. Binder JR, Mohr JP. The topography of callosal reading
pathways. Brain 1992;115:1807-1826
3. Alexandrov A V, Ehrlich LE, Bladin CF, Black SE. Clinical significance of increased uptake of HMP AO on brain
SPECT scans in acute stroke. J Neuroimaging 1996;6:150155
4. Baird AE, Donnan GA, Austin MC, McKay WJ. Early
reperfusion in the "spectacular shrinking deficit" demonstrated by single photon emission computed tomography.
Neurology 1995;45:1335-1339
5. Damasio H. A computed tomographic guide to the identification of cerebral vascular territories. Arch Neurol 1983;
40:138-142
6. Coslett HB, Saffran EM, Greenbaum S, Schwartz H.
Reading in pure alexia. Brain 1993;116:21-37
7. Greitz T, Bohm C, Holte S, Eriksson L. A computerized
brain atlas: construction, anatomical content, and some
applications. J Comput Assist Tomogr 1991;15:26-38
8. Weder B, Oettli R, Maguire RP, Vonesch T. Partial epi-

ler,tic seizure with versive movements examined by
9
[ rnTc]HM-PAO brain single photon emission computed
tomography: an early postictal case study analyzed by
computerized brain atlas 'methods. Epilepsia 1996;37:68-75
9. De Renzi E, Zambolin A, Crisi G. The pattern of neuropsychological impairment associated with left posterior
cerebral artery infarcts. Brain 1987;110:1099-1116
10. Binder JR, Lazar RM, Tatemichi TK, et al. Left hemiparalexia. Neurology 1992;42:562-569
11. Caplan LR, Hedley-Whyte T. Cuing and memory dysfunc-

tion in alexia without agraphia: a case report. Brain 1974;
97:251-262
12. Coslett HB, Saffran EM. Preserved object recognition and
reading comprehension in optic aphasia. Brain 1989;112:
1091-1110
13. Salamon G, Huang YP. Radiologic anatomy of the brain.
Berlin: Springer Verlag; 1976
14. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the
human brain. 3-dimensional proportional system: an approach to cerebral imaging. Stuttgart: Thieme Verlag; 1988

STATEMENT OF OWNERSHIP, MANAGEMENT AND CIRCULATION (Act of August 12, 1970: Section 3685, Title 39 United
States Code) Date of Filing-October 1, 1998. Title of Publication-Journal of Neuroimaging; Frequency of Issue--4 issues annually;
Annual Subscription Price-$160.00; Location of Known Office of Publication-Lippincott Williams & Wilkins, 12107 Insurance Way,
Hagerstown, MD 21740; Location of the Headquarters or General Business Offices of the Publisher-227 East Washington Square,
Philadelphia, PA 19106; Publisher-Lippincott Williams & Wilkins, 227 East Washington Square, Philadelphia, PA 19106; EditorLeon D. Prockop, MD, University of South Florida College of Medicine, Department of Neurology, 12901 Bruce B. Downs Boulevard,
Tampa, FL 33612; Managing Editor-Marie Chesny, Lippincott Williams & Wilkins, 227 East Washington Square, Philadelphia, PA
19106. Owner-American Society ofNeuroimaging, 5841 Cedar Lake Road, Suite 108, Minneapolis, MD 55416. Known Bondholders,
Mortgagees, and other security holders owning or holding 1 percent or more of total amount of bonds, mortgages, or other securitiesnone. A. Total no. copies printed (net press run), average 2100, actual, 1600; B. Paid circulation 1. Sales through dealers and carriers,
street vendors and counter sales, average none, actual none; 2. Mail subscriptions, average 862, actual 853; C. Total paid circulation,
average 862, actual 853; D. Free distribution by mail, carrier or other means. Samples, complimentary, and other free copies, average
129, actual 126; E. Total distribution (sum of C and D), average 991, actual 979; F. Copies not distributed 1. Office use, leftover,
unaccounted, spoiled after printing, average 1109, actual621; 2. Returns from news agents, none; G. Total (sum of E and F-should
equal net press runs shown in A), average 2100, actual1600. I certify that the statements made by me above are correct and complete.
Beverly Dietrich, Circulation Director.
Significant Case Report: Lanzinger et al: Alexia and Neuroimaging

51