Pain, 47(1991) 329-336
© 1991 Elsevier Science Publishers B.V. All rights reserved 0304-3959/91/$03.50

329

PAIN 01904

Central pain and thalamic hyperactivity: a single photon emission
computerized tomographic study
Pierre Cesaro a, Michael W. Mann a, Jean Luc Moretti b, Gilles Defer a, Brigitte Rouald~s ~,
Jean Paul Nguyen a and Jean Denis Degos a
a D~partement de Neurosciences Mddicales, C.H.U. Henri Mondor, 94000 Creteil (France) and b Sercice de M~decine Nucldaire, CHU A L,icenne,
93000 Bobigny (France)

(Received 5 July 1990, revision received 9 April 1991, accepted 30 April 1991)

Summary
Five patients with central post-stroke pain (CPSP) accepted to be studied according to the following
paradigm: a single photon emission computerized tomography (SPECT) using [~23I]N-isopropyl-iodoamphetamine
(IMP) was made in each patient 20 min following i.v. injection of IMP; during this time, the patients were
stimulated in order to reproduce their spontaneous pain. Of the five patients, two had CPSP with hyperpathia
following a stroke (with a lesion on CT scan involving the thalamo-cortical pathway in one and involving the
thalamus in the other); two had CPSP following a stroke in the middle cerebral artery area, without hyperpathia;
and the last patient suffered pain from algodystrophia following a fracture of the wrist. In the two cases with
hyperpathia, SPECT demonstrated a contralateral relative hyperactivity in a central region corresponding to the
thalamic area. This was not observed in the three other patients. In the two patients with hyperpathia, a second
SPECT scan with stimulation of the contralateral pain-free arm did not demonstrate any hyperactivity in the
thalamic area. These results suggest that a thalamic neuronal hyperactivity may characterize some hyperpathic
syndromes and, in accordance with our previous results obtained in the rat, that the loss of inhibition on medial
thalamic neurons may be a main feature of hyperpathia following certain cerebral stroke syndromes.
Key words: Thalamic pain; Central pain; Thalamus; SPECT; Brain activation

Introduction
The origin of pain in the thalamic syndrome [11] or
central post-stroke pain (CPSP) [18] remains an unsolved problem. There are two main pathophysiological
hypotheses of this peculiar type of pain: irritation and
disinhibition. Irritation of sensory fibers terminating in
the thalamic ventral posterior nucleus (VP) has been
held responsible for pain according to Roussy [27].
According to Garcin [15] this theory is supported by
the fact that pain appears only in some patients, due to
the anatomical characteristics of the lesion that vary
from patient to patient. The irritative phenomenon
Correspondence to: Dr. Pierre Cesaro, D6partement de Neurosciences M6dicales, C.H.U. Henri Mondor, 51 Avenue du Mar6chal
De Lattre De Tassigny, 94000 Creteil, France.

could be caused by the lesion itself, the glial scar or
ephaptic activation of axons [22]. As far as disinhibition
is concerned, the hypothesis assumes that the transmission of nociceptive stimuli within the fast neospinothalamic and the slow paleospinothalamic system is inhibited at different levels by quite different mechanisms.
Some authors suggest a modulation of the paleospinothalamic system by the lemniscal system [14,24,26].
Lhermitte et al. [20] argued that the loss of the 'filtering function' of the thalamus for ascending input is
responsible for thalamic pain. Finally, at the cortical
level Head and Holmes [17] claim that 'under normal
conditions the activity of the thalamic centre ... is
dominated by that of the cortex' so that cortical lesions
or lesions of the corticothalamic fibers might provoke
'exaggerated responses in cases where the thalamic
centre has been freed from control.'

331)
Pathological thalamic activity in CPSP could bc
demonstrated by two methods: electrophysiology and
functional imaging. Tasker [32] states that the electrical
stimulation of the paleospinothalamic tract (which projects to the medial thalamic nuclei) at midbrain level
and of the medial thalamus itself causes burning or
painful responses in patients with deafferentation pain,
whereas this tract is silent during stimulation in subjects without pain. More recently, Lenz et al. [19]
recorded abnormal bursts in the VP of patients suffering pain from spinal deafferentation.
The purpose of the present paper is to show thalamic hyperactivity in two patients with CPSP by single
photon emission computerized tomography (SPECT)
using [123I]N-isopropyl-iodoamphetamine (IMP) and to
discuss the possible pathophysiological significance of
this finding.

Methods

Functional SPECT imaging
IMP I 123 was prepared by an exchange reaction
using a commercially available kit which has been
described elsewhere [12]. SPECT was performed with a
rotating gamma camera (Gammatome 1 CGR). Blindfolded and with ears plugged, patients were intravenously injected with 5 - 7 mCi of IMP I 123 (10 mg of
IMP) 20 rain before the tomographic acquisition (64
views along 64 angles accumulated during 40 rain).
Tomographic images were obtained by a filtered back
projection method, a reorientation of the axial slices
parallel to the cantomeatal line and compensation of
attenuation by a Chang method [10].
Data analysis was performed according to our previous studies comparing control cases and patients [12].
Briefly speaking, the region of interest (ROI) was
visually selected and delineated by a square or a rectangle of at least 25 pixels, leading to a semiquantitative
analysis giving left-right differential percentage of activity (DPA) between symmetrically located ROI. DPA
was considered significant when it was higher than
10% according to all our results with the same methodology. In this study analysis was done on frontal and
sagittal slices because of better visualisation of the
thalamus on these slices compared to classical
transversal slices.
In each patient a SPECT scan was performed in
order to compare left-right cerebral difference in IMP
activity after sensory stimulation of the painful side of
the body. In two patients (cases 1 and 2) a pair of
SPECT scans were done after separate sensory stimulation of each side of the body. In these cases a
complete washout of IMP was ensured by a time interval of at least 8 days between the two SPECT scans.

Stereotaxic localization of the lesions
Stereotaxic localization was performed as Iollows.
First, the position of the lesion in a parasagittal plane
and the position of anterior-posterior commissure
( A C - P C ) line in a sagittal plane were located using
sagittal and parasagittal reconstructions [6,25]. The distance which separates the lesion from A C - P C line (z)
and AC and PC (y) can be deduced. Second, lateral
coordinates (x: distance between the lesion and the
midlinc) are measured on axial CT films and corrected
by the magnifying factor of CT films. Thc 3 coordinates
(x, y, z) are used to select maps from a stereotactic
atlas [31]. Digitalization of the stereotactic maps permits easy correction of some individual variations such
as the length of the third ventricle or the A C - P C
distance.
Patients
Five patients underwent SPECT, four with CPSP
due to a stroke and one with pain due to algodystrophia from a forearm fracture. Case reports are
detailed below.

Results

Case ]
A 69-year-old man was admitted to the Henri Mondor Hospital on June, 1982, following a stroke. He
presented a paresis of the left hand with hypoesthesia
to all somesthetic modalities of the left upper limb and
the left side of the face. Plain CT scan showed a small
right parietal subcortical hypodensity that was enhanced after injection of contrast material. A parietal
subcortical ischemic lacunar infarction was diagnosed
(Fig. 1). Two weeks later he complained of prolonged
burning pain in his left hand after a repetitive stimulation by light touch and pressure. One month later he
began to complain of spontaneous burning pain and,
on clinical examination, he showed a clear hyperpathia:
light touch or pressure on the left forefinger was followed by a burning sensation in the arm and left side
of the face. A first SPECq" with IMP was performed at
this time, before any antidepressant drug was given.
After stimulation of the left hand 15 min before and
after i.v. injection of IMP, SPECT showed a relative
right hyperactivity within the thalamic area (left-right
difference was t6% and 26% on two consecutive slices
at the thalamic level). One week later a second SPECT
was performed with the same stimulation of the painless right hand; no asymmetry in thalamic activity was
observed (Fig. 2). The patient was then successfully
treated with 150 mg amitriptyline daily. Another
SPECT was performed 3 months later with the same
stimulation of the left hand that had previously provoked a relative thalamic hyperactivity. No asymmetry

331
slow abnormal movements of the right hand appeared.
SPECT with IMP was performed at this time.
The right hand was immersed into warm water
(41 ° C) for 15 min before and after i.v. injection of
IMP, and SPECT showed a relative hyperactivity of the
left thalamic area (the left-right difference was 39%
and 20% on two consecutive slices at the thalamic
level). A second SPECT was performed 3 weeks later
with the same thermal stimulation of the painless left
hand; no significant thalamic hyperactivity was found
(Fig. 4). Six months later the hyperpathy to warm water
slowly disappeared. At this time, thermal and pinprick
sensations were normal, and the amplitude of the
abnormal movements was diminished. SPECT was repeated with warm water stimulation of the right hand
and no thalamic hyperactivity was observed.
The stereotaxic localization of the lesions of patients
1 and 2 is shown in Fig. 5.

Fig. 1. Case 1. Contrast enhanced CT scan showing a right parietal
lesion adjacent to the cortical ribbon (positivecontrast enhancement).

was observed. One year later the treatment had to be
stopped because of side effects and pain reappeared;
all sensory events, emotions, games with his children,
crossword playing were followed by the same extensive
and prolonged pain of the left side. SPECT again
demonstrated a relative hyperactivity in the right thalamic area.
Case 2
A 58-year-old man was admitted to the Henri Mondor Hospital on March, 1984, following a stroke. He
presented with paresis and numbness of the right hand,
a hypoesthesia for thermal and pinprick stimulation of
the right side of the body and slight aphasia. Plain CT
scan showed a left thalamo-capsular hyperdensity, and
2 months later a small lucency, suggesting a sequela of
a hematoma (Fig. 3). Two months after the stroke he
complained of prolonged burning pain in his right hand
when this hand was dipped into warm water. The
hyperpathia was exclusively provoked by warm water
and by no other sensory stimulus. At the same time,

Case 3
A 59-year-old right-handed woman with arterial hypertension rapidly developed in April, 1979, a right
hemiparesia, aphasia and vertical diplopia, preceded
by a perioral sensation of cold and followed by a
transient somnolence. An acute stroke was diagnosed
(no CT scan available). Seven months later she developed a burning pain of the right thigh spreading to the
heel; during the following year, it progressed to the
right hand, and then to the entire right upper limb and,
finally in 1983, to the right side of the face. The pain
was constant at a low level of increasing intensity over
the day, sometimes of a paroxysmal character. Neurological examination in 1984 disclosed a right-sided
hemiparesia with pyramidal signs, a right-sided hemihypesthesia for tactile, arthrokinetic, nociceptive and
thermal stimuli, more pronounced in the lower limb
than on the face. No hyperpathia was detected. The
patient reported an intermittent sensation of having a
third upper limb which was moving. Vertical diplopia
due to a partial left oculomotor paralysis persisted as
well as a slight degree of aphasia. There was no
astereognosia and no visual field defect. Higher brain
functions were otherwise normal. Psychiatric examination revealed a moderate depression.
A high resolution contrast enhanced CT scan in
1984 showed a low absorption area at the left sylvian
fissure, a small round lucency within the left thalamus
and a small lucency at the border of the right thalamus
and the internal capsule, consistent with a stroke in the
territory of the left superficial middle cerebral artery
(MCA), and 2 lacunar strokes involving the thalamus.
Treatment with 100 mg amitriptyline i.m., then with 75
mg given orally, clearly improved the right-sided burning pain and the depression for 8-9 months; then the
initial symptoms recurred. A therapeutic trial with the
antidepressant drugs mianserine and desipramine was

332

Fig. 2. Case 1. SPECT imaging. The four upper slices (1 horizontal, 1 coronal, and 2 parasagittal, all at thalamic level) were obtained without any
stimulation. The left-right difference of activity was within the control range (10%). The lower four slices were obtained 1 week before. These
SPECT slices were obtained following 15 rain sensory tactile, smooth stimulation of the hyperpathic finger of the left hand. Each slice
demonstrates a left-right difference, ranging from 16% (coronal slice) lo 26% (parasagittal slices) in favor of the right thalamic area.

333

Fig. 4. Case 2. SPECT imaging. Two consecutive coronal slices at
thalamic level. The left slices were performed without stimulation of
the right, painful hand. The right slices were obtained in a previous
session 8 days before following 15 min stimulation of the right hand
with warm water and left-right difference was increased by 39% and
20% in favor of the left side.

Fig. 3. Case 2. Plain CT scan showing a small left thalamic hematoma.

light pressure on and passive movement of the fingers,
which gave rise to increased pain but not to hyperpathia. No left-right difference in activity was measured
in the thalamic areas.

considered unsuccessful at follow-up in 1986 with regard to pain and depression.
In January 1985 a SPECT was performed (the patient being treated with amitriptyline), preceded by

Case 4
•In January 1980 a 68-year-old woman was admitted
to the Henri Mondor Hospital because of severe pain
of the left side of the body. She had suffered for many
years from ophthalmic migraine (i.e., recurrent

R
V

ML

LESION

m

LESION

VPL
AC-PC

A

AC-PC

B

Fig. 5. Anatomical representation of frontal planes according to the stereotaxic atlas of Talairach. The A C - P C line is found according to CT
scan slices by reconstructing the plane on a computer [23]. The digitalisation of the planes of the atlas allows them to be adapted easily to the
individual A C - P C line of each patient. A: case 2. The lesion was found to involve the nucleus lateralis posterior thalami (LP), the nucleus
ventralis posterior thalami (VPL) and the capsula interna (CI). The midline (ML), the lateral ventricle (V) and the A C - P C line (line drawn
between the anterior and posterior commissure of the third ventricle) are also represented. The section lies 5 m m rostral to the A C - P C line. B:
case 1. The lesion is located at the junction of the parietal gyruses (P1, P2) and the paracentral circumvolution (Prc = precuneus). O n this
coronal section, the following structures can be recognised: R = rolandic sulcus, G H = hippocampal gyrus, Fus = gyrus fusiformis, T2 = temporal
gyrus, T3 = temporal gyrus. The section lies 14 m m caudal to the posterior commissure.

334
headache preceded by a visual bilateral scotoma) with
no predilection for either side. In November 1984 she
presented a left hemiparesia and hemihypesthesia and
a left hemianopsia due to a stroke in the territory of
the right superficial MCA.
Four months after the stroke she developed a burning pain of the right side of the face, the tongue
(provoking a dysphagia), the left upper limb and less
frequently of the lower limb. A headache always preceded the burning type of pain, which extended during
15-20 rain over the left side of the body, lasted for
hours and disappeared simultaneously with the
headache. A burning pain could also be provoked
sometimes by light touch and by a movement of the left
thumb which would be followed by an ascending pain
of the medial surface of the left upper limb. This
burning pain was very intense, provoking suicidal ideas.
Sometimes a right-sided headache would occur alone,
without burning pain. No visual symptoms were associated with the headache, nor were left-sided migraine
attacks. No improvement was noted despite 60 rag/day
desipramine given orally, and she was hospitalized.
Neurological examination revealed a moderate leftsided hemiparesia with pyramidal signs and left visual
neglect. Diminished position sense of the left toe, a left
hemihypesthesia for light touch and pin prick and
extinction of left-sided sensory stimuli were found.
Thermal sensation was normal. Higher brain functions
were normal. No psychic abnormalities were noted. A
CT scan showed a large hypodense sequela of a stroke
in the territory of the right superficial MCA, sparing
the thalamus.
After a washout period of 3 days without antidepressant drugs, a SPECT was performed preceded by
light pressure and passive movement of the fingers of
the left hand which gave rise to increased pain but not
to a clear-cut hyperpathia. No left-right difference in
activity was measured within the thalamic area.
Case 5
A 46-year-old healthy subject consulted 10 months
after a fracture of the right forearm, complicated by an
algodystrophia with arm and hand edema and hyperesthesia. X-rays demonstrated demineralization of the
bones of the hand, and bone scintigraphy was abnormal. In 1984 a SPECT was performed, preceded by
gentle pressure and passive movement of the hyperesthetic fingers and wrist, which gave rise to increased
pain but not to hyperpathy. No left-right difference of
activity was measured within the thalamic area.

Discussion
Current although incomplete data suggest that
SPECT using IMP might be an appropriate method for
functional cerebral imaging. SPECT results are not

quantitative and our method has not yet been well
assessed for studying cerebral activation. Furthermore,
this method does not allow symmetric brain activation
to be measured following the stimulation we performed. Nevertheless IMP distribution could be related to nervous tissue metabolism [1], and focal cerebral activation was demonstrated in humans by
S P E C T - I M P following natural [16] or epileptic [28]
brain activation. Long cerebral retention time of IMP
allows delayed scanning 4-5 h after an intravenous
injection. These data suggest that altered activity of
nervous tissue can be visualized by SPECT using IMP.
All of our patients, except for the case with algodystrophia, presented some characteristics but not the full
picture of hyperpathia. In various combinations they
showed severe, widespread, burning pain that could be
elicited by different, normally non-painful stimulations
(light touch, emotions: case 1; warm water: case 2;
migraine, light movement: case 4). A sensory deficit
was noted in all patients at the time when abnormal
pain could be elicited. The lesion was localized on CT
scan within the left thalamus in one patient (case 2), in
the left parietal subcortical area in one patient (case 1),
and in the territory of the left (case 3) and right (case
4) superficial MCA in the remaining two cases. CT
scan of case 3 also displayed a deep left and a deep
right lacunar lucency, the latter one without any known
corresponding clinical history. Clearly, a suprathalamic
lesion can give rise to a CPSP. An extensive review of
such lesions of the parietal lobes has been made [31]
and the anatomically verified case of Biemond [7], who
reported a small, subcortical, ischemic lesion opposite
to the parietal operculum at the origin of central pain,
is particularly interesting. Our first case, although having a parietal lesion, closely resembles the classical
form of thalamic pain with hyperpathia. Thus the question is raised whether CPSP resulting from a suprathalamic or from a thalamic lesion share a common path()genesis.
Early and delayed SPECT scans with IMP in our
first two cases revealed a relative (as compared with
the opposite side) hyperactivity of the thalamic region
contralateral to the side of the body where the hyperpathia was provoked by appropriate stimulation,
whereas no thalamic hyperactivity was found in several
control SPECT scans with stimulation of the painless
side of the body nor with stimulation of the initially
hyperpathic side at a time when the patient was pain
free. The correlation between the SPECT results and
the clinical symptoms is striking in these patients. One
can suggest that the 'hyperactivity' results in fact from
a lessened activity in the unstimulated side or from
other mechanisms than neuronal hyperactivity. Positron
emission tomography (PET) studies are needed to answer the question and are currently in progress in a
larger group of patients.

335

No such relative thalamic hyperactivity was found in
the other three patients, either because of insufficient
sensitivity of our imaging technique or because of
different pathogenesis of pain, which is likely in the
patient with atgodystrophia, but more hypothetical in
the other two subjects with preceding infarction in the
MCA area. Central pain due to cortico-subcortical
softening in the MCA area might clinically be somewhat different from thalamic pain due to thalamic or
subcortical lesions, inasmuch the pain tends to be more
continuous and without hyperpathia. Admittedly we do
not know whether this clinical difference corresponds
to a different pathogenesis and to different SPECT
results.
What might be the significance of this thalamic
hyperactivity? First, it is compatible with either of the
two main pathophysiological theories: irritation and
disinhibition. Second, our results suggest, yet they do
not demonstrate, that thalamic pain is related to an
abnormally increased activity of some thalamic structures. Spatial resolution of SPECT is not presently
sufficient to differentiate thalamic structures, but there
is a considerable body of evidence that the medial
thalamic nuclei (i.e., mainly the intralaminar nuclei:
centralis lateralis, parafascicular, centrum medianum,
but also submedius) produce most of the observed
hyperactivity. (1) Both, the medial and the lateral thalamic nuclei are involved in pain processing. The lateral thalamic nuclei are particularly prone to destruction in case of ischemic 'pure sensory stroke' [11,13]
and, therefore, it seems unlikely that they themselves
could be the site of a hyperactivity. (2) The ill-defined
temporo-spatial character of CPSP might be a function
of the diffuse projections of the non-specific medial
nuclei [4]. (3) In patients with deafferentation pain,
medial thalamic and paleospinothalamic stimulation at
midbrain level elicit pain similar to spontaneous pain,
whereas the same areas are silent when stimulated in
patients without pain [32]. However, it must be noted
that CPSP may be associated with large thalamic lesions including the major part of medial thalamus [18],
but this latter observation is documehted only by CT
results and not by anatomical data.
How can medial thalamic nuclei acquire abnormal
functions? The destruction of the ventral posterior
thalamic nucleus (VP) is unlikely to act directly on the
medial nuclei, since there are no direct connections
between these two structures [29]. The nucleus reticularis thalami (RT) could be a relevant interface; the
vast majority of the thalamic inhibitory neurons are
localized within the RT [29]. The dominant orientation
of the dendritic mass of RT cells is perpendicular to
the course of thalamo-cortical and cortico-thalamic
radiations, thus allowing multiple and somewhat nonspecific input to RT. The RT projects exclusively to the
thalamus, mainly to the medial nuclei. Our own prelim-

inary experimental data suggest that extensive destruction of the lateral thalamus including RT is followed by
abnormal tonic hyperactivity in the medial thalamus
[5,9]. Furthermore, this medial thalamic hyperactivity is
increased by electrical stimulation of the VP nucleus,
whereas the same stimulation is inhibitory in normal
conditions [5,9].
Our hypothesis is that lesions of RT or altered input
to RT, due to cortical subcortical or thalamic lesions,
disturb the inhibitory function of RT upon the medial
thalamic nuclei, thus giving rise to medial thalamic
hyperactivity. The time of degeneration of RT neurons
(which in experimental conditions degenerate following
decortication) may explain the delayed onset of pain.

Acknowledgement
This study was supported by Research Grant No.
1154 R1 from the Universit6 Paris, Val de Marne.

References
1 Ackermann, R.H., Of cerebral blood flow, stroke and SPECT,
Stroke, 15 (1984) 1-4.
2 Ajurriaguerra de, J., La Douleur darts les Affections du Syst~me
Nerveux Central, Paris, Doin, 1937.
3 Alajouanine, Th., Thurel, R. and Brunelle, A., Les douleurs
alternes dans les l~sions bulbo-protuberantielles. Contribution
l'6tude de la physio-pathologie des douleurs centrales, Rev. Neurol., 63 (1935) 837-838.
4 Albe-Fessard, D. and Besson, J.M. (1973) Convergent thalamic
and cortical projections: the non-specific system. In: A. Iggo
(Ed.), Handbook of Sensory Physiology, Vol. II. Somatosensory
System, Springer Verlag, Berlin, pp. 490-548.
5 Amsallem, B., Cesaro, P., Joulin, Y., Pollin, B. and NguyenLegros, J., Possible role of the nucleus reticularis thalami in the
control between specific and non specific thalamic nuclei activity,
J. Physiol. (Lond.), 406 (1988) 171.
6 Archer, C.R., llinsky, I.A., Golfader, P.R. and Smith, Jr., K.R.,
Aphasia in thalamic stroke: CT stereotactic localization, J. Cornput. Assist. Tomogr., 5 (1981) 427-432.
7 Biemond, A., The conduction of pain above the level of the
thalamus opticus, Arch. Neurol. Psych., 75 (1956) 231-244.
8 Cassinari, V. and Pagni, C.A., Central Pain, Harvard University
Press, Cambridge, MA, 1969, 191 pp.
9 Cesaro, P., Amsallem, B., Pollin, B., Nguyen-Legros, J. and
Moretti, J.L., Organisation des noyaux m~dians et intralaminaires
du thalamus: hypotheses sur leur r61e dans la survenue de certaines douleurs centrales, Rev. Neurol., 142 (1986) 297-302.
10 Chang L.T., A method for attenuation correction in radionuclide
computed tomography, IEEE Trans. Nucl. Sci., 25 (1978) 638-643.
11 D6jerine, J. and Roussy, G., Le syndrome thalamique, Rev.
Neurol., 12 (1906) 521-532.
12 Defer, G., Moretti, J.-L., Cesaro, P., Sergent, A., Raynaud, C.
and Degos, J.-D., Early and delayed SPECT using N-isopropyl
p-iodoamphetamine iodine 123 in cerebral ischemia, Arch. Neurol., 44 (1987) 715-718.
13 Fisher, C.M., Pure sensory stroke and allied conditions, Stroke,
13 (1982) 434-447.

336
14 Frazier, C.H., Lewy, F.H. and Rowe S.N., The origin and mechanism of paroxysmal neuralgic pain and the surgical treatment of
central pain, Brain, 60 (1937) 44-51.
15 Garcin, R., La douleur dans les affections organiques du syst~me
nerveux central, Rev. Neurol.. 68 (19371 105-153
16 Goldenberg, G., Podreddu, J., Holl, K. and Dali Bianca, P., IMP
SPECT for visualization of changes in brain activity caused by
acoustic memory tasks, in: Amersham Buchler GmbH (Ed.), Eur.
Nucl. Med. Cong. Book of Abstracts, Braunschweig. F.R.G.,
1985, p. 111.
17 Head, H. and Holmes, G., Sensory disturbances from cerebral
lesions, Brain, 34 (1911) 102-254.
18 Leijon, G., Boivie, J. and Johansson, I., Central post-stroke pain:
neurological symptoms and pain characteristics, Pain, 36 (1989)
13-27.
19 Lenz, F.A., Kwan, H.C., Martin, R., Tasker, R. and Dostrovsky,
J.O., Characteristics of spontaneous neuronal activity at different
locations in ventrocaudal thalamus of patients with central pain
following spinal cord transection, Pain, Suppl. 5 (1990) 955.
2(1 Lhermitte, J., de Ajuriaguerra, J. and Hecaen, H., Le Syndrome
Thalamique. Etude Clinique, Vol. 1, IVe Congr~s Neurologique
International, Masson, Paris, 1949, pp. 71-82.
21 Manfredi, M. and Cruccu, G., Thalamic pain revisited. In: Carlo
Loeb (Ed.), Studies in Cerebrovascular Disease, Masson, Paris,
1981, pp. 73-95.
22 Melzack, R. and Wall, P.D., Pain mechanisms: a new theory,
Science, 150 (1965) 971-979.
23 Moretti, J.L., Askienazy, S., Raynaud, C. et al., N-isopropyl I
123-p-iodoamphetamine: an agent for brain imaging with single
photon emission computerized tomography. In: P.L. Magistretti

(Ed.), Functional Radionuclide Imaging ol Ihc 13rain, Raven
Press, New York. 1983, pp. 231 24(~.
24 Noordenbos, W., Pain: problems pertaining to lhe transmission ol
nerve impulses which give rise 1o pain. Elsevier. Amsterdam,
1959.

25 Nguyen, J.P., Szikla, G. and Missir, O., Fiabilit¢} d'une mt}lhode
simplifide de transposition des images TDM. Corrdlations au
rcp~rage stdr~:otaxique clans 3(1 cas, Neurochinlrgie, 28 (1982)
271 274.
26 Riddoch, G., The clinical features of central pain, Lancel. 234
(19381 1093-1098, 1150-1156, 1205-1209.
27 Roussy, G., La Couche Optique, Thesis, Steinheil, Paris, 1907.
28 Sanabria, E., Chauvel, P., Askienazy, S., Vignal, J.P., Trottier, S.,
Chodkiewicz, J.P. and Bancaud, J., Single photon emission computed tomography (SPECT) using 123 l-isopropyldodoamphetamine (lAMP) in partial epilepsy. In: M. Baldy-Moulinier,
D.I-|. lngvar and B.S. Meldrum (Eds), Current Problems in
Epilepsy. I. Cerebral Blood Flow, Metabolism and Epilepsy. John
Libbey, London, 1984, pp. 82-89.
29 Scheibel, M.E. and Scheibel, A.B., The organization of the nucleus reticularis thalami: a Golgi study, Brain Res., I (1966)
43-62.
30 Talairach, J., Tournoux, P. and Bancaud, J., La chirurgie paridtale
de la douleur, Acta Neurochir. (Vienna), 8 (19601 153-2511.
31 Talairach, J., Szikla, G., Tournoux, P., Prossalentis, A., BordasFerrer, Covello, L, lacob, M. and Mempel, E., Atlas d'Anatomie
Stc}rdotaxique du T~lencdphale, Masson, Paris, 1967.
32 Tasker, R.R., Effets sensitifs et moteurs de la stimulation thalamique chez l'homme. Applications cliniques, Rev. Neurol., 142
(1986) 316-332.