Letters, Notice, Book review
this aboratory succeeded in preparing a
monoclonal antibody (BC-1) specific for
oncofetal FN, an isoform characterised by a
peculiar sequence (ED.B) in the central part
of its molecule. On the other side, diVerent
monoclonal antibodies were prepared, which
are either able to recognise all FN isoforms
(IST-4), or are specific for FNs that do not
contain oncofetal sequence (IST-6).1 A previous study, in which all these monoclonal
antibodies were employed showed that,
whereas FNs are widely distributed in the
endothelia of human gliomas, the occurrence
of the oncofetal isoform correlates with the
degree of malignancy. Similarly, the expression of the oncofetal FN in the vascular
endothelium of benign meningiomas was
found to be restricted,2 by contrast with the
findings reported here.
Preliminary findings by using the Tc-99mlabelled anti-oncofetal FN monoclonal antibody BC-1, that was administered intravenously, showed that this monoclonal
antibody does reach malignant meningiomas
in vivo, as well as malignant gliomas, allowing
scintigraphic evidence of the tumour and
providing a possible basis for vascular
targeting.4 In this regard, it can be speculated
that administration of tumoricidal radionuclides linked to monoclonal antibody
BC-1 might prove to be able to selectively
destroy endothelia of meningeal malignancies, while sparing the normal vessels
of the brain, and to favourably aVect the
course of the illness in terms of recurrence
rate.

413
Repetitive sentence writing after a left
anterior cerebral artery infarct
Patients with cerebral lesions out of motor
and sensory strips are often overlooked, especially those with frontal lobe symptoms,
which are subtle and elusive. We report on a
patient with a left anterior cerebral artery infarct, who had an unusual manifestation—
namely, repetitive sentence writing.
A 58 year old right handed woman had a
transient lapse of consciousness when riding
a bicycle on 3 June 1995. She had no vomiting, convulsion, or sphincter problems, and
walked home after having taken a rest. Brain
CT performed at a local hospital was normal.
Next day, she visited our hospital because of
right leg weakness, urinary urgency, persistent mutism, and slow mentality. On admission, the patient seemed alert and had normal
vital signs, but was mute and responded
slowly to oral commands. Neurological
examination disclosed a grasp reflex on both
hands, being stronger on the right, gegenhalten rigidity, right leg weakness, and clonic
motor perseveration of the right hand. With
the right hand, she would hit the bed, her own
body, cutlery, and nearby objects before her.
She knocked herself on the head if we asked
her: “Do you have a headache?”, and on the
leg if we asked her: “Do you have leg pain?”.
The clonic motor perseveration stopped
when she held objects, made a fist, changed
posture, or restrained the right hand with the

left one, but would recur several seconds
later. It never ceased for long unless she fell
asleep. On her second day in hospital, she was
still abulic, mute, and indiVerent. She used a
toothbrush, chopsticks, towels, and fastened
and undid buttons normally. On the fourth
day, the frequency and the amplitude of the
clonic motor perseveration decreased. She
laughed, displayed more facial expressions,
and pronounced an “e” sound. On the 11th
day, she began to communicate with her family by writing. Next day, she spoke a complete
sentence to complain about the noise made
by the neighbouring patient, which was the
only one sentence she had spoken since her
admission. Brain MRI on the 13th day
showed a left anterior cerebral artery infarct
(figure A, B).
Neuropsychological tests were done from
day 6 to 11. The patient had had a high
school education. She had no spontaneous
speech and could not repeat even simple
words, but her verbal comprehension was
intact and she could correctly name real
objects and photographs of them. She
performed line bisection and cancellation
tasks very well. On day 11, the patient wrote
about her illness on request (figure C),
although she was still mute. The handwriting
was legible and we easily transcribed it to a
printed form (figure D). Initially, she wrote
quite normally, “I am....My physical condition is not good and (I had) a syncope on the
road, and was sent to this hospital by a friend.

This study was partially supported by the National
Research Council (CNR), AIRC, and the Ministry
of University and Scientific Research (MURST).
We thank Mr Sergio Deseri for his technical help in
the preparation of the manuscript.
ANTONIO PAU
LUCA BRUZZONE
ALESSANDRA DORCARATTO
GIUSEPPE VIALE
Division of Neurosurgery, University of Genoa Medical
Scool, Genoa, Italy
GIULIANO MARIANI
Departement of Internal Medicine, Nuclear Medicine
Service, University of
Genoa Medical School, Genoa, Italy
PATRIZIA CASTELLANI
ANNALISA SIRI
LUCIANO ZARDI
Laboratory of Cell Biology, Istituto per la Ricerca sul
Cancro, Genoa, Italy
Correspondence to: Dr Antonio Pau, Clinica Neurochirurgica dell’Universita degli Studi Ospedale S,
Martino Largo Rosanna Benzi, 10 16132 Genova,
Italia. Telephone 039 10 3537610; fax 039 10
352104.

1 Carnemolla B, Balza E, Siri A, et al. A tumor
associated fibronectin isoform generated by
alternative splicing of messenger-RNA precursors. J Cell Biol 1989;108:1139–48.
2 Castellani P, Viale GL, Dorcaratto A, et al. The
fibronectin isoform containing the ED-B oncofetal domain: a marker of angiogenesis. Int J
Cancer 1994;59:612–8.
3 Chamberlain MC. Adjuvant combined modality therapy for malignant meningiomas. J Neurosurg 1996;84:733–6.
4 Mariani G, Lasku A, Villa G, et al. A pilot pharmacokinetic and immunoscintigraphic study
with the 99m Tc- labeled monoclonal antibody
BC-1 in patients with brain tumors. J Nuclear
Medicine 1996;37:238.
5 Sternberger LA, Shirley AJ. The unlabeledantibody method. J Histochem Cytochem 1979;
27:1424–9.

Cerebral MRI; (A) axial T2 weighted, TR 2500, TE 90 ms; (B) coronal T1 weighted, TR 600, TE 15
ms showing a left mesial frontal lesion; (C) patient’s original penmanship; (D) transcription of (C).

Letters, Notice, Book review

414

Now I am well” (figure D, the first and
second lines). In figure D, the seventh
character right of the comma on the second
line (we defined as L2R7) and L3R6 should
be L4R7, which is an orthographic paragraphia. Then, she began to rewrite sentences
from L3, although not exactly identical.
These repetitive sentences were divided into
two clauses by a comma. The first clause said,
“My physical condition is not very good”,
which she repeated 17 times (L3 to L19).
The second clause said, “( I ) was sent to this
hospital by a doctor of this hospital”. There
were several minor errors in the second
clauses, including character omissions (between L3R4 and L3R5, L10R6 and L10R7,
and L12R3 and L12R4) and redundant
characters (at L7R6 and L23R10 to R12).
She changed “friend” (L3R2 and L3R3) to
“doctor” (L8R4 and L8R5) and unnecessarily repeated the modifier “this hospital” twice
(L14R2 and L14R3 as well as L14R8 and
L14R9). She wrote five identical copies of a
long sentence (L14 to L18 of figure D). She
stopped writing when the space on the paper
was used up.
Our patient was discharged on the 14th
day. Her husband reported that on the 17th
day she watered flowers, but played with the
water at the same time. She rapped kitchen
utensils with a spatula when cooking, and
rapped bowls with a spoon when eating.
Three days later, she began to repeat phrases,
and followed events in the stock market on
television. However, she became angry with
her husband without reason on the 21st day,
which was unprecedented. On the 26th day,
her verbal output seemed normal. The clonic
motor perseveration subsided gradually and
eventually stopped.
Most of this patient’s symptoms had been
described elsewhere.1 The repetitive writing
had been reported only for letters, characters,
and phrases.2 To our knowledge, repetition of
a long meaningful sentence has never been
reported. Our patient seemed totally involved
when writing the sentences, and would not be
distracted. Furthermore, the clonic motor
perseveration and the repetitive sentence
writing showed similarities in that both were
repetitive and elicited by a stimulus. The
clonic motor perseveration was elicited by
nearby objects attracting her, and the repetitive sentence writing by an idea to describe
her illness.
Regarding our patient’s ability to write
while she was mute, we suggest two points.
Firstly, her inner language was intact. This
resembles “dynamic aphasia” with which
patients seem almost mute in conversational
speech, but show a dramatic preservation of
the ability to name objects, to read, and to
repeat sentences.3 Kleist thought that patients
with dynamic aphasia had intact verbal
propositional thought and intact “sentence
schema” but these were disconnected.4 Our
patient wrote sentences with correct grammar and word selection to describe a
meaningful idea, which supports this notion.
Secondly, the mechanism for writing was
intact. The neural substrates accounting for
writing were unaVected in our patient, as
shown by the proper sentence syntax and legible ideograms.
Dominant mesial frontal lesion may impair
the processes of modification and monitoring
of the motoric components of writing, resulting in disinhibition and loss of monitoring.
Moreover, Shallice5 proposed an idea of
“supervisory attentional system” which controls and modulates the lower level processes.

This system is activated when a person is
dealing with a non-habitual condition and is
thought to be a frontal lobe function. This
explains why our patient rewrote the sentences so many times, yet did not detect the
errors in the writing.
We thank Professor Atsushi Yamadori of Tohoku
University, Japan for his helpful comments in the
preparation of this manuscript.
MING-CHYI PAI
ING-JER HUANG
Division of Behavioral Neurology, Department of
Neurology, National Cheng Kung University Hospital,
Tainan, Taiwan
Correspondence to: Dr Ming-Chyi Pai, Division of
Behavioral Neurology, Department of Neurology,
National Cheng Kung University Hospital, 138,
Sheng Li Road, Tainan 704, Taiwan. Telephone 886
6 2353535 ext 3579; fax 886 6 2759036; email:
pair@mail.ncku.edu.tw
1 Bogousslavsky J, Regli F. Anterior cerebral
artery territory infarction in the Lausanne
Stoke Registry. Clinical and etiological patterns. Arch Neurol 1990;47:144–50.
2 Yamadori A. Perseveration in aphasia. Aphasia
Research 1987;7:25–9. (In Japanese.)
3 Luria AR, Tsvetkova LS. Towards the mechanism of “dynamic aphasia”. Acta Neurol Belg
1967;67:1045.
4 Kleist K. Gehirnpathologie. Leipzig: Barth,
1934. Cited by: McCarthy RA, Warrington
EK, eds. Cognitive neuropsychology. San Diego:
Academic Press, 1990:191.
5 Shallice T. From neuropsychology to mental structure. Cambridge: Cambridge University Press,
1988.

Peripheral neuropathies among patients
with HIV infection
Peripheral neuropathies often aVect patients
with HIV disease,1 and various peripheral
neuropathies have been described at different
stages of HIV infection, from seroconversion
to AIDS. Acute inflammatory demyelinating
polyneuropathies have been found in association with primary HIV infection; chronic
inflammatory demyelinating polyneuropathies, isolated mononeuritis, and multiplex
mononeuritis have more commonly been
described in those in advanced stages of HIV
disease, and distal symmetric polyneuropathy
has been found to be the most common neuropathy among persons with AIDS. In most
cases, the aetiopathogenesis of these neuropathies is still unclear, although various
factors have been implicated (infectious,
metabolic, immunological, inflammatory, nutritional, and toxic).2 To date, relatively few
studies have estimated the incidence of
peripheral neuropathies among HIV infected
patients belonging to diVerent population
groups. Data derived from the Multicenter
AIDS Cohort Study showed annual rates
slightly higher than 1.5/100 person-years for
sensory neuropathy.3 Another study found a
roughly 1% annual incidence of neuropathies
in symptomatic patients without AIDS.4 To
estimate the incidence of peripheral neuropathies during the diVerent stages of HIV
disease, we analysed data from a cohort study
of patients with known dates of seroconversion (a documented HIV seronegative test
followed by a confirmed positive test within
two years). The seroconversion date was estimated as the midpoint in time between the
last negative and the first positive HIV test.
After seroconversion, patients were followed
up to obtain clinical information and data on
laboratory indices about every six months.
The study design and methodology have been
described in detail elsewhere.5 The present
analysis included four clinical centres, which

performed a detailed neurological examination of the participants of the study. These
centres were contacted to provide complete
information on peripheral neuropathies by
completing a standardised data collection
form for each episode, providing detailed
information
on
diagnostic
criteria
(neurological signs and symptoms, instrumental diagnosis), AIDS prophylaxis, and
antiretroviral drug therapy. The study population consisted of 621 HIV seroconverters.
The median age of the participants was 27
(range 14–66) years, 29 for males and 24 for
females. Of these patients; 267 (42.9%) were
intravenous drug users, 217 (34.9%) were
homosexual men, and 127 (20.5%) were heterosexual contacts; 10 (1.6%) patients had
unknown or undetermined risk factors.During a median follow up time of 5.7 years, 19
(3.1%) patients developed symptoms suggestive of peripheral neuropathies. The estimated incidence of peripheral neuropathies
was 5.5/1000 person-years of follow up. The
median age at the time of diagnosis was 35.2
years (range 24.8–53.6 years). Thirteen
events (68.4%) occurred among males and
six (31.6%) among females. For the exposure
category, peripheral neuropathy was found
among seven (2.6%) intravenous drug users,
five (2.3%) homosexual men, and seven
(5.51%) heterosexual contacts. The incidence rate of peripheral neuropathies was
4.4/1000 person-years in intravenous drug
users, 4.3/1000 person-years in homosexual
men, and 11.4/1000 person-years in heterosexual contacts. Acute infection was observed
in 43 of 621 (7%) patients. One episode of
peripheral neuropathy occurred during acute
infection (the incidence rate of peripheral
neuropathies was 42.4/100 person-years), six
episodes during the asymptomatic phase
(0.2/100 person-years), eight in the preAIDS symptomatic phase (2.4/100 personyears), and four episodes in patients who had
already presented with diseases indicative of
AIDS (3.8/100 person-years). The CD4 cell
count at diagnosis of peripheral neuropathies
was 266.2 for the patients with acute
infection; the median was 324.6 (range
149.7–655.2) for those in the asymptomatic
phase, 216.5 (range 14.9–1440.0) among
those with pre-AIDS symptoms, and 38.0
(range 7.2–124.8) among persons with
AIDS. Of the 19 patients with peripheral
neuropathies, 15 (78.9%) had distal symmetric polyneuropathy, two (10.5%) had multiplex mononeuritis, and two (10.5%) acute
inflammatory demyelinating polyneuropathies. The median CD4 cell count at diagnosis of peripheral neuropathy was 31.2 (range
16.3–47.0) for persons with multiplex
mononeuritis, 184.1 (range 7.5–655.4) for
those with distal symmetric polyneuropathy,
and 304.3 (range 342.1–266.3) for those with
multiplex mononeuritis All 19 patients had
sensory neuropathy, and 14 (73.7%) also had
motor impairment. Of these 19 patients, 13
(68.4%) underwent electrophysiological investigations, showing abnormalities of nerve
conduction. Another patient presenting with
a mononucleosis-like syndrome and meningoradiculoneuritis had a raised number of
cells (184/µl) in the CSF. The CSF of the
other participants was normal. No participant reported the use of known neurotoxic
drugs such as DDC; 10 patients had received
AZT and two had received DDI (both
patients who were treated with DDI had discontinued treatment six months before the
onset of peripheral neuropathies; one had
multiplex mononeuritis and one distal sym-