nously, then 6 mg every 6 hours) have
been thought to aid in the survival of
patients with AHL.3,4 More recently,
plasma exchange was used successfully
in one patient with neurologic sequelae
of mycoplasmal pneumonia.1' Until now,
to our knowledge the use cyclophospha¬
mide has not been reported in the treat¬
ment of AHL. Its effect in this patient

remains uncertain. Effective treatment
probably consists of a combination of
early diagnosis based on clinical suspi¬
cion, neuroimaging, brain biopsy, life
support with reduction of elevated ICP
in the critical stage, and early, sus¬
tained treatment with appropriate

immunosuppressants.

References
1. Hurst EW. Acute hemorrhagic leukoencephalitis: a previously undefined entity. Med J Aust.

1941;2:1-6.

2. Beham PO, Moore MJ, Lamarche JB. Acute

necrotizing hemorrhagic encephalopathy. Postgrad Med. 1973;54:154-160.
3. Byers RK. Acute hemorrhagic leukoencepha-

litis: report of three cases and review of the literature. Pediatrics. 1975;56:727-735.
4. Foley JM, Kane CA. Clinical pathologic conference. Neurology. 1957;7:135-141.
5. Coxe WS, Luse SA. Acute hemorrhagic leukoencephalitis: a clinical and electron-microscopic
report of two patients treated with surgical decompression. J Neurosurg. 1963;20:584-596.
6. Gosztonyi G. Acute hemorrhagic leukoencephalitis (Hurst's disease). In: Vinken PJ, Brun GW,
eds. Handbook of Clinical Neurology. Amsterdam, the Netherlands: Elsevier Science Publishers; 1978;3(pt

2):587-604.

7. Lander H. Acute hemorrhagic leukoencephalitis. Australas Ann Med. 1958;7:55-68.
8. Kulich SA. Acute hemorrhagic leukoencephalitis: report of a case and review of the literature.
Boston Med Q. 1960;11:120-130.
9. Martins AN, Kempe LG, Harp GJ. Acute
hemorrhagic leukoencephalitis with a concurrent
primary herpes simplex infection. J Neurol Neurosurg Psychiatry. 1964;27:493-501.
10. Yesnick L. Central nervous system compli-

cations of primary atypical pneumonia. Arch Intern
Med. 1956;97:93-98.
11. Fisher RS, Clark AW, Wolinsky JS, Parhad
IM, Moses H, Mardiney MR. Postinfectious leukoencephalitis complicating Mycoplasma pneumoniae infection. Arch Neurol. 1983;40:109-113.
12. Rothstein TL, Shaw CM. Computerized tomography as a diagnostic aid in acute hemorrhagic
leukoencephalitis. Ann Neurol. 1983;13:331-333.
13. Watson RT, Ballinger WE, Quisling RG.
Acute hemorrhagic leukoencephalitis: diagnosis by
computed tomography. Ann Neurol. 1984;15:611\x=req-\
612.
14. Levine S, Wenk EJ. A hyperacute form of
allergic encephalomyelitis. Am J Pathol. 1965;
47:61-88.
15. Lamarche JB, Behan PO, Segarra JM, Feldman RG. Recurrent acute necrotizing hemorrhagic
encephalopathy. Acta Neuropathol(Berl). 1972;
22:79-87.
16. Kristianson K, Harkmark W, Cohen MM.
Acute hemorrhagic encephalitis. Neurology (Min-

neap). 1956;6:503-509.
17. Cotter FE, Bainbridge D, Newland AC.
Neurological deficit associated with Mycoplasma
pneumonia reversed by plasma exchange. BMJ.
1983;286:22.

Isolated Congenital Hemianopia Caused
by Prenatal Injury to the Optic Radiation
Nicola K. Ragge, MRCP FCOphth; A. James Barkovich, MD;
William F. Hoyt, MD; Scott R. Lambert, MD

\s=b\ Isolated congenital hemianopias are
typically caused by developmental abnormalities of the occipital lobe cortex. We describe two patients with an incidental partial
hemianopia associated with unilateral periventricular leukomalacia that was acquired
prenatally. Magnetic resonance imaging
scans suggest a late second or early third

trimester unilateral cerebral ischemic event.
(Arch Neurol. 1991 ;48:1088-1091)

Tsolated congenital occipital lesions

with accompanying hemianopia usu¬
ally involve gray and white matter. Ex¬
amples include porencephaly1"3; vascular
malformations, either isolated or in as¬
sociation with Sturge-Weber-type mal¬

formation without cutaneous signs4; colpocephaly'; and polymicrogyria of the
striate cortex.'' ' This report documents
two cases of congenital hemianopia
caused by prenatal injury to the peri¬
ventricular white matter.

REPORT OF CASES

Case l.—A

healthy 21-year-old white

woman was found

to have an inferior quadrantic field defect during a routine eye ex¬
amination. She had a history of herpetic keratitis in her right eye. There was no history of
prenatal or postnatal illness or neurologic
disorder. Her visual acuities were 20/50 in
the right eye and 20/20 in the left eye. There
were no abnormalities of color vision. Her
pupil reflexes were equal bilaterally. Both
Goldmann and computerized visual field test¬
ing showed a right inferior homonymous qua-

drantanopia with minor incongruity along
the horizontal meridian (Fig 1). There were
signs in both retinas of nerve fiber loss corre¬
sponding to that seen in homonymous hemioptic atrophy. Both discs were small in
diameter. The right disc showed band atro¬
phy, with loss of nerve fiber layer especially
prominent at the nasal disc edge, where the
pigment epithelial edge could be seen. The
superior and inferior arcuate bundles were
well preserved. On the left disc, a full com¬
plement of nasal nerve fiber layer could be
seen, but temporal pallor and relative loss of
the superior and inferior arcuate bundles

Fig 1. —Patient 1. Right inferior homonymous quadrantic defect demonstrated by Goldmann
perimetry.

Accepted for publication April 18,1991.
From the Departments of Neurology, Neurosurgery, and Ophthalmology (Drs Ragge and Hoyt)
and Radiology (Dr Barkovich), School of Medicine,
University of California, San Francisco, and the
Department of Ophthalmology, Emory University
School of Medicine, Atlanta, Ga (Dr Lambert).

Reprint requests to Neuro-ophthalmology Unit,
Department of Neurosurgery, 126 UC, University
of California Medical Center, San Francisco, CA
94143 (Dr Hoyt).

Downloaded From: http://archneur.jamanetwork.com/ by a University of California - San Diego User on 06/02/2015

Fig 2. —Patient 1. Left, Coronal ,-weighted image at the level of the ventricular trigone shows a nearly complete absence of
periventricular white matter ¡nferolaterally on the left. The depths of the cortical gyri nearly abut the ventricular surface. Center,
Heavily T2-weighted axial image shows the expanded trigone and paucity of periventricular white matter. The lateral margin of
the ventricle is poorly defined. Right, Spin-density image af the same level as center shows high signal intensity in the
periventricular white matter, representing damaged tissue. The presence of this area of high intensity indicates that the damage
occurred later than that in patient 2 (compare with Fig 4, right).
the ventricle and more clearly showed the
close approximation of the depths of the cor¬
tical gyri to the ventricular wall. T2-weighted
images showed normal intensity of the peri¬
ventricular white matter. No enhancement
was seen after administration of intravenous
contrast.

COMMENT

Fig 3. —Patient 2. Left superior homonymous quadrantic defect demonstrated by Goldmann
perimetry.

were

present.

Magnetic resonance images of the brain
(Fig 2) showed ex vacuo enlargement of the
trigone and occipital horn of the left lateral
ventricle. The close approximation ofthe cor¬
tical gyri, particularly the posterior insular
cortex, to the ventricular surface was charac¬

teristic of the pattern seen as a result of loss
of the intervening white matter. In this loca¬
tion, the white matter is partially composed
of the optic radiations. T2-weighted images
showed a thin strip of high signal intensity in
the immediate periventricular white matter
indicative of replacement of normal tissue by
gliotic white matter.
CASE 2.—A healthy 10-year-old black boy,
who was being followed up in the eye clinic
for a left esotropia and amblyopia, was found
to have a left homonymous hemianopia.
There was no significant obstetric or perina¬
tal history. His visual acuities were 20/25 in
the right eye and 20/200 in the left eye. The

pupils were 3 mm in diameter, equal, and
fully reactive. There was a mild left relative
afferent pupillary defect. The patient had a
left superior homonymous quadrantanopia
(Fig 3). The retinal nerve fiber layer demon¬
strated a pattern of nerve fiber loss indicat¬

ing homonymous hemioptic atrophy." The re¬
markable feature of both discs was that they
were large in diameter, with pronounced
physiological cups and a nasal entrance ofthe
vessels. The left disc was vertically oval and
kidney shaped with band atrophy. There was
an atrophie nerve fiber layer at the nasal
border of the left disc compared with a full
nasal nerve fiber layer adjacent to the right
disc.

Computed tomography ofthe head showed
enlargement of the right lateral ventricle,
most notably the trigone, as a result of loss of
periventricular white matter. Magnetic res¬
onance imaging with gadolinium-pentetic
acid (Fig 4) showed the same enlargement of

We have described two cases of uni¬
lateral white matter injury in the region
of the optic radiations causing isolated
partial hemianopias. In both cases the
ophthalmologic evidence of transsynaptic nerve fiber degeneration establishes
the occipital injury as a prenatal
event/11 The occurrence of such notice¬
able retinal nerve fiber layer and optic
disc changes is somewhat unexpected:
the predicted loss of nerve fibers by
transsynaptic degeneration from a le¬
sion that causes a quadrantanopia
would be about 25% from each optic
nerve. We have no explanation for the
fact that both patients' homonymous
field defects were quadrantic, while the

magnetic resonance imaging scans
showed diffuse, not focal, thinning of
the visual radiation in the involved oc¬
cipital lobes.
The affected regions of both brains
have magnetic resonance imaging char¬
acteristics that fit those reported for
end-stage periventricular leukomalacia.1"11 These are (1) ventriculomegaly,
sometimes with irregular outline of the
ventricular surface; (2) reduced
amounts of periventricular white mat¬
ter at the trigone and, in more severe
cases, the entire centrum semiovale; (3)

Downloaded From: http://archneur.jamanetwork.com/ by a University of California - San Diego User on 06/02/2015

Fig 4. —Patient 2. Left, Axial ,-weighted image after infusion of intravenous contrast shows enlarged frontal horn and trigone of
the right lateral ventricle. Note the paucity of white matter between the cortical gyri and the ventricular surface on the right side
compared with the left. No contrast enhancement is apparent. Center, Coronal ,-weighted image at the level of the ventricular
trigones shows that the absence of periventricular white matter is more marked interiorly at the level of the optic radiations.
Right, Axial spin-density image at the level of the ventricular trigones shows an absence of high signal intensity in the
periventricular white matter (compare with Fig 2, right). This implies an earlier gestational event.

deep and prominent cortical sulci with
subcortical white matter approaching
the ventricles with little interposed
white matter; and (4) prolonged T2 re¬
laxation time in the immediate peritrigonal white matter, sometimes extend¬
ing anteriorly along the ventricular
body.
Periventricular leukomalacia occurs

most commonly in premature infants

who also have other complications in¬
cluding hyaline membrane disease, per¬
sistent fetal circulation, and germinal
matrix hemorrhage.12'3 Both pathologi¬
cal and neuroimaging studies have
shown a characteristic progression that
begins with edema, followed by coagu¬
lation necrosis and eventually cavitation, usually in 3 to 6 weeks.1214 The
cysts usually collapse, the affected
brain exhibiting gliosis and reduction in
the quantity of white matter in the peri¬
ventricular region. The ependymal sep¬
aration between the cavities and the
ventricle may break down, with conse¬
quent incorporation of the cavities into
the ventricle.
Since periventricular leukomalacia
occurs most commonly in the peritrigonal region,15 through which the geniculocalcarine radiations course, it is not
surprising that patients with this condi¬
tion commonly have visual impair¬
ment.11'1' However, this is usually seen
in the context of a premature infant with
bilateral defects. Several authors have
described unilateral periventricular
leukomalacia18"21; however, all have con¬
centrated on motor dysfunction and
none have described visual impairment.

The pathogenesis of most cases of
periventricular leukomalacia is now
widely accepted as being the result of

ischemia in the immature brain.1122,21
The "watershed areas" of the prema¬
ture brain are located in the periventri¬
cular region.24 At about 36 weeks' gesta¬
tional age, the watershed areas begin to
move

peripherally, eventually reaching

the subcortical white matter and cere¬
bral cortex at term. It is a logical exten¬
sion of this concept to postulate that
unilateral periventricular leukomalacia
is most likely to be the result of in utero
hypoperfusion of a single hemisphere.
Gilles et al25 have shown that the glial
response to brain injury is minimal prior
to about 26 weeks' gestational age. Bar¬
kovich and Truwit11 have further shown
that asphyxie brain injury prior to 28
weeks' gestational age shows minimal
periventricular T2 prolongation. The
area of prolonged T2 relaxation time ad¬
jacent to the enlarged trigone in patient
2 indicates an injury between about 28
and 34 weeks' gestational age, whereas
the absence of periventricular pro¬
longed T2 relaxation in patient 1 indi¬
cates an earlier injury, most likely dur¬
ing the second half of the second
trimester.

This study was supported, in part, by grants
from the Wellcome Trust and the Royal Society of
Medicine, London, England. Dr Ragge is a Keeler
Scholar, Keeler Ltd, Windsor, England.

References
1. Remillard GM, Ethier R, Andermann F.
Temporal lobe epilepsy and perinatal occlusion of
the posterior cerebral artery: a syndrome analo-

gous to infantile hemiplegia and a demonstrable
etiology in some patients with temporal lobe epilepsy. Neurology. 1974;24:1001-1009.
2. Hoyt WF. Congenital occipital hemianopia.

Neuro-ophthalmol Jpn. 1985;2:252-259.

3. Novakovic P, Taylor DSI, Hoyt WF. Localising patterns of optic nerve hypoplasia\p=m-\retinato
occipital lobe. Br J Ophthalmol. 1988;72:176-182.
4. Andriola M, Stolfi J. Sturge-Weber syndrome: report of an atypical case. AJDC.

1972;123:507-510.

5. Garg BP. Colpocephaly: an error of morphogenesis? Arch Neurol. 1982;39:243-246.
6. Tychsen L, Hoyt WF. Relative afferent pupillary defect in congenital occipital hemianopia. Am J
Ophthalmol. 1985;100:345-346.
7. Taylor DC, Falconer MA, Bruton CJ, Corsellis JAN. Focal dysplasia of the cerebral cortex in
epilepsy. J Neurol Neurosurg Psychiatry.

1971;34:369-387.
8. Hoyt WF, Rios-Montenegro EN, Behrens
MM, Eckelhoff RJ. Homonymous hemioptic hypoplasia: fundoscopic features in standard and redfree illumination in three patients with congenital
hemiplegia. Br J Ophthalmol. 1972;56:537-545.
9. Miller NR, Newman SA. Transsynaptic degeneration. Arch Ophthalmol. 1981;99:1654.
10. Flodmark 0, Lupton B, Li D, et al. MR

imaging of periventricular leukomalacia in child-

hood. AJNR. 1989;10:111-118.
11. Barkovich AJ, Truwit CL. Brain damage
from perinatal asphyxia: correlation of MR findings
with gestational age. AJNR. 1990;11:1087-1096.
12. Volpe JJ. Hypoxic-ischemic encephalopathy:
neuropathology and pathogenesis. In: Neurology of
the Newborn. 2nd ed. Philadelphia, Pa: WB
Saunders Co; 1987:209-235.
13. Friede RL. Developmental Neuropathology. 2nd ed. New York, NY: Springer-Verlag
NY Inc; 1989:69-97.
14. De Reuck J, Chattha AS, Richardson EP.
Pathogenesis and evolution of periventricular leukomalacia in infancy. Arch Neurol. 1972;27:229\x=req-\
236.
15. Banker BQ, Larroche J-C. Periventricular
leukomalacia of infancy. Arch Neurol. 1962;7:386\x=req-\
410.

16. Lambert SR, Hoyt CS, Jan JE, Barkovich
AJ, Flodmark O. Visual recovery from hypoxic

cortical blindness during childhood: computed to-

Downloaded From: http://archneur.jamanetwork.com/ by a University of California - San Diego User on 06/02/2015

Neuropediatrics.

mography and magnetic resonance imaging predic-

by congenital hemiparesis.

17. Flodmark O, Jan JE, Wong PKH. CT of the
brains of children with cortical visual impairment.
Dev Med Child
1990;32:611-620.
18. Wiklund L-M, Uvebrant P, Flodmark 0.
Morphology of cerebral lesions in children with congenital hemiplegia: a study with computed tomog-

20. Claeys V, Deonna T, Chrzanowski R. Congenital hemiparesis: the spectrum of lesions. Helv

tors. Arch Ophthalmol. 1987;105:1371-1377.

raphy. Neuroradiology. 1990;32:179-186.

19. Molteni B, Oleari G, Fedrizzi E, Bracchi M.
Relation between CT patterns, clinical findings and
etiological factors in children born at term, affected

1987;18:75-80.

Pediatr Acta. 1983;38:393-455.
21. Kotlarek F, Rodewig R, Brull D. CT findings
in congenital hemiparesis in childhood and their
relation to etiology and prognosis. Neuropediatrics. 1981;12:101-109.
22. Pape KE, Wigglesworth JS. Haemorrhage,
Ischaemia, and the Perinatal Brain. Philadelphia,
Pa: JB Lippincott; 1979;69/70. Clinics in Develop-

mental Medicine.
23. Wiklund L-M, Uvebrant P, Flodmark O. CT
as an adjunct in etiological analysis of hemiplegic
cerebral palsy. Neuropediatrics. In press.
24. Takashima S, Tanaka K. Development of cerebrovascular architecture and its relationship to
periventricular leukomalacia. Arch Neurol.

1978;35:11-16.
25. Gilles FH, Averill DR Jr,Kerr C. Changes in

neonatally induced cerebral lesions with advancing
age. J Neuropathol Exp Neurol. 1977;36:666-684.

Visual Hallucinations and Racing Thoughts
on Eye Closure After Minor Surgery
C. Miller Fisher, MD

\s=b\ Two hours after minor surgery under local anesthesia, a 67-year-old man developed
vivid visual hallucinations that were present
only when he closed his eyes. After lasting 4
hours, the hallucinations were succeeded by
racing thoughts, lasting 2 hours, and then by
illusionary movements of his body. The reaction was attributed to the effect of lidocaine.
The case resembles one previously reported.
(Arch Neurol. 1991 ;48:1091 -1092)

rPhe occurrence of visual hallucinations on eye closure in a fully alert
patient is a rarely reported phenome¬
non. Recently, one case was described

in which the hallucinations were attrib¬
uted to atropine toxicity and lasted 11
days.1 The present case bears a close
resemblance; but, in addition to visual
hallucinations, the patient experienced
"racing thoughts" (his term) and halluci¬
nations of body movement on eye clo¬
sure. The symptoms began shortly after
a minor operation that was performed
under local anesthesia.
Abnormal mental events are of inter¬
est, not only as they aid in clinical inter¬
pretation, but also as they extend our
knowledge of the brain's behavioral

capability.

REPORT OF A CASE

A 67-year-old hypertensive physician re¬
ported a 15-minute spell of weakness of the
right side of the face and the right hand.
Results of neurological examination were
normal, as was a cranial computed tomo¬
graphic scan. Angiography showed stenosis
of both internal carotid arteries to less than
1-mm residual lumen, worse on the right
side. A right-sided carotid endarterectomy
was attempted, but patency could not be re¬
stored, and the internal carotid artery was

Accepted for publication April 4, 1991.
From the Neurology Service of the Massachusetts General Hospital, Boston.
Reprint requests to Neurology Service, Massachusetts General Hospital, Boston, MA 02144 (Dr
Fisher).

ligated. Immediate recovery was unevent¬
ful; but, 24 hours later, on removing the drain
from the wound, bleeding occurred into the
neck, forming a mass that displaced the tra¬
chea. The blood clot was promptly evacuated
using 15 mL of 0.01% lidocaine, with 0.005%
phenylephrine hydrochloride as the local an¬
esthetic. The only additional medications
were codeine, 60 mg, given orally 2 hours
before the operation; cefazolin sodium, 1 g;

and labetalol, 5 mg, administered twice dur¬
ing the procedure, which was uncomplicated
and lasted about 1 hour. Cerebral hypoxia
did not occur. The labetalol temporarily re¬
placed atenolol, which the patient had been
taking for about 6 years.
At 12:30 PM, two hours after surgery,
while drowsing, the patient began to experi¬
ence visual hallucinations; these continued
for about 4 hours. According to his daughter,
who sat at his bedside as he described the
events to her, he was lucid and appropriate at
all times. He was first seen by me a few hours
after the aberrant ideation had ceased; at this
time, his memory ofthe happenings was still
fresh.
The visual hallucinations consisted of four
main topics: books and other printed materi¬
al, street scenes, snowy landscapes, and ani¬
mals of many kinds. The images followed one
another in quick succession, "nonstoj) with no
rhyme or reason," each one lasting 2 or 3
seconds. When he opened his eyes, the vi¬
sions disappeared. Although he was wide
awake and apparently clear mentally, it was
for some reason difficult to keep his eyes
open. "It was as if they were being forced
shut." He was not frightened or alarmed, but
was restless and unduly excited; he attribut¬
ed this to a feeling of being out of control and
helpless. The images were colored and pro¬
jected into the center of space before him,
without lateralization. They were silent, and
figures showed no facial movements to sug¬
gest that they were speaking. The individual
hallucinations were entirely unrelated to one
another, with no recognizable narrative con¬
nection. It was broad daylight, and the room
was brightly lit.
The printed matter consisted of books and
pages; in addition, handwritten material was
present. The colored street scenes were not

recognized as being in anyr particular city;
many people about 1 m tall were rashing
across the street from side to side at

great

speed, dashing hither and thither, creating a
bewildering confusion. The streets were
lined with buildings on each side. The people

seemed to be further away than the sur¬
roundings, creating the appearance of minia¬
turization. There were no autos. The snow
scenes stretched for miles over low hills. As
the hospitalization was in the winter time,
when there was snow, this may have been a
determining factor; but, a snow scene could
quickly be followed by one without snow.
There were no trees. The animals were de¬
formed and repugnant; for example, there
were ugly, one-eyed Picasso-like monsters,
or, grotesque, nongeometrical shapes that
were everchanging. The animals were sta¬
tionary, as in a picture. There were no recog¬
nizable normal animals, such as dogs, cats,
horses, or rodents. The coloring was mostly
of water colors and pastels, except for white,
which was quite bright, and orange, which
was vivid.
As the visual hallucinations subsided, they
were followed by a most unpleasant episode
of "racing thoughts." The patient was not
aware of any visual or auditory images. Un¬
connected thoughts followed one another in
quick succession, without any sense. They
passed so swiftly, the patient was unable to
recognize them. It was his impression that
they were about people, rather than things
or places. They probably involved family
members, including a grandson. Each
thought might consist of three staccato bits,
only to be interrupted by an entirely differ¬
ent thought with its own few bits, with the
whole series being fragmentary and mean¬
ingless. Also, the thinking spree was present
only when his eyes were closed; this was
immediately dispelled on opening the eyes;
but, again, there was difficulty in keeping the
eyes open in the absence of sleepiness. The
episode of rushing thoughts was stranger
and more unpleasant than the visual halluci¬
nations, although the thoughts themselves
were not frightening or bizarre. The patient
agreed that the thoughts might have been

ordinary thinking, speeded up to a dizzying
degree. In light of the patient's statement

Downloaded From: http://archneur.jamanetwork.com/ by a University of California - San Diego User on 06/02/2015