Original Paper
Received: June 3, 2002
Accepted: October 10, 2002

Eur Neurol 2003;49:131–136
DOI: 10.1159/000069075

Positron Emission Tomographic Study
of Post-Ischaemic-Hypoxic Amnesia
J. De Reuck a I. Vanwalleghem a D. Hemelsoet a M. De Weweire a
K. Strijckmans b I. Lemahieu c
a Department of Neurology, Ghent University Hospital, b Laboratory of Analytical Chemistry,

Institute of Nuclear Sciences, and c Laboratory for Electronics and Information Systems, Ghent University,
Ghent, Belgium

Key Words
Post-ischaemic-hypoxic amnesia W Positron emission
tomography W Acetazolamide vasoreactivity W Frontal
lobe dysfunction

Abstract
Background: Despite extensive research, it still remains
controversial as to what the precise location of the critical lesions underlying amnesia actually is. The amnesic
syndrome is believed to be heterogeneous and due to
several distinct functional deficits. Patients and Methods: Two patients, a 45-year-old woman and a 56-yearold man, with sudden cardiopulmonary arrest and successful resuscitation, were left with a clear amnesic syndrome as main neurological sequela. During their revalidation period, they underwent a positron emission tomographic (PET) examination, utilizing the 13NH3 bolus
technique at rest and after intravenous acetazolamide
administration. Results: Both PET studies showed more
or less similar features with a decrease in regional cerebral blood flow (rCBF) in the frontal, temporal and parietal lobes. In addition, the rCBF was increased in both
thalami of the 45-year-old woman and in the striata of the
56-year-old man. Acetazolamide vasoreactivity was
most lost in the frontal lobes. Conclusions: In the present
PET study, we demonstrated that destruction of the

ABC

© 2003 S. Karger AG, Basel
0014–3022/03/0493–0131$19.50/0

Fax + 41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com

Accessible online at:
www.karger.com/ene

inhibitory pathways to the thalamus and basal ganglia
by ischaemic-hypoxic frontal lesions could be one of the
mechanisms leading to amnesia.
Copyright © 2003 S. Karger AG, Basel

Introduction

Organic amnesia is a syndrome in which preservation
of intelligence and immediate memory are typically accompanied by various degrees of antegrade and retrograde amnesia. In general, amnesia may be caused by
encoding, storage and retrieval deficits [1]. Despite extensive research it still remains controversial as to what the
precise location of the critical lesions underlying amnesia
actually is. Evidence is gradually accumulating that several distinct functional deficits underlie the syndrome [2].
If they exist, it is possible to isolate the responsible
structures by using positron emission tomography (PET)
or functional magnetic resonance imaging (fMRI). There
are two main ways in which these procedures can be used:
recording activity during the resting state in patients with
known structural lesions and comparison with matched
control subjects [3–5] or using memory challenge techniques in which the pattern of activation is compared
between normal people and amnesics [6–8].

Jacques De Reuck, MD, PhD
Department of Neurology
Ghent University Hospital, De Pintelaan 185
B–9000 Gent (Belgium)
Tel. +32 9 240 45 39, Fax +32 9 240 49 71, E-Mail jacques.dereuck@yucom.be

MRI and PET studies in posthypoxic amnesia have
revealed hypometabolism in the medial temporal lobes
and hippocampi [9–12], but also in the limbic-diencephalic regions including the thalami [11–13], the caudate
nucleus and the cerebellum [11].
The present report describes unusual PET findings in 2
patients with post-ischaemic-hypoxic amnesia.

Patients and Methods
Two patients, a 45-year-old woman and a 56-year-old man, with
sudden cardiopulmonary arrest and successful resuscitation, were
left with a clear amnesic syndrome as main neurological sequela.
During their revalidation period, 3 and 9 months after the acute
event, respectively, they underwent a PET examination utilizing the
13NH bolus technique at rest and after intravenous acetazolamide
3
administration. Although 13NH3 is an old PET tracer that reflects
mainly capillary perfusion [14], it is also an indirect indicator of local
cerebral metabolism [15] and it is suitable for assessing acetazolamide vasoreactivity in hypoperfused areas [16].
Both patients underwent a bedside cognitive evaluation on admission, using a brief mental status questionnaire (Jacobs test),
adapted specifically to diagnose diffuse organic syndromes [17] and
translated into Dutch. This test was validated by our local neuropsychological department [unpubl. data]. Subjects with scores of less
than 20/30 were considered to have a significantly impaired cognitive capacity.
A full neuropsychological examination was performed in the first
patient and attempted in the second at the time of the PET scan. The
used mental status tests included the copying and subsequent reproduction from memory of the Rey-Osterrieth complex figure, Hooper
visual organization test, Corsi block test, serial recitation test for
numbers, Rey auditory verbal learning test, token test, assessing verbally mediated right-left discrimination and the Boston naming test.
Permission from the Hospital Ethics Committee was granted,
and oral informed consent for the PET examination was obtained
from the near relatives of the patients. A Neuro-Ecat IV scanner
(EG&G/Ortec, Oak Ridge, Tenn., USA) with double rings of detectors and lateral and axial resolutions of 8.1 and 14.0 mm, respectively, was used. The distance between detector ring centres was 32 mm,
and measurements were made in high-resolution mode using septa
and shadow shields. After performing phantom scans with an external 68Ge/68Ga ring source, the patient’s head was positioned to accumulate data from transaxial planes centred 3 and 35 mm above the
orbitomeatal line. A transmission scan was then performed using the
same external source to correct subsequent emission scans for photon
attenuation.
For measuring regional cerebral blood flow (rCBF), an intravenous bolus of 0.7 GBq 13NH3 was administered to the patients, and
after 3 min, cephalic radioactivity was scanned for 300 s. As the halflife of 13NH3 is 10 min, a time interval of 25 min was respected
before performing the consecutive PET examination. Acetazolamide
at a dose of 22 mg/kg body weight was then administered intravenously, and after another 10 min, a similar bolus of 0.7 GBq 13NH3
was given with scan time starting after 3 min for 300 s. The high-dose
acetazolamide was well tolerated by both patients.

132

Eur Neurol 2003;49:131–136

Three slice reconstructions were obtained 3, 19 and 35 mm above
the orbitomeatal line. Both patients had technically reliable measurements of the PET slices, including one through the basal frontal and
temporal cortices and cerebellum, a second through the basal ganglia
and thalami and a third at the level of the lateral ventricles. All PET
planes could be superimposed on corresponding CT slices obtained
around the time of the PET examination.
For each patient, ellipsoid regions of interest (ROIs) with axial
axes of 30 mm and lateral axes of 15 mm (twice the camera resolution) were drawn over the cortical rim of the frontal, temporal, parietal and occipital lobes of both cerebral hemispheres. For the striata,
thalami and cerebellar hemispheres circular ROIs, adapted to the
size of the structures, were used. The cerebellum, both at rest and
after acetazolamide administration, was used as reference region
(100%).
The vasoreactivity was determined semiquantitatively in all
ROIs taking normalized resting rCBF as 100%. The degree of acetazolamide reactivity was the percent difference between the number
of counts in a particular ROI after administration of the drug against
that at rest, using the same dose of 13NH3. In our previous study, it
was shown in a control group that the mean rCBF at rest increases on
average with 36.6% (B5.7) in all regions after acetazolamide administration [16].
Kruskal-Wallis one-way analysis of variance and two-group
Mann-Whitney U tests were used for statistical analysis. Statistical
significance was set at p values less than 0.05.

Results
Case 1
A 45-year-old woman with a prior history of hypothyroidism and
chronic alcohol abuse was transferred to the University Hospital following a cardiopulmonary arrest at home, probably due to myocardial infarction and ventricular fibrillation. After a successful reanimation and defibrillation by the resuscitation team, she was admitted in coma to the intensive-care unit. She needed artificial ventilation and was also treated with antibiotics for bronchopneumonia.
She initially improved, but 24 days later she developed again an
acute respiratory distress syndrome and had an episode of generalized tonic-clonic seizures, which was treated with phenytoin. She
became afterwards alert and cooperative, and obeyed all commands.
Initially she had an action myoclonus and a mixed positional and
intentional tremor that disappeared after a few weeks. Except for a
positive suck reflex, no other neurological deficits could be detected.
She was transferred to the neurological department. The most
remarkable finding was that, although she had fluent and normal
speech, the verbal communication was inaccurate. She was disoriented in time and space and could not tell correctly about herself
and her family. She confabulated and was euphoric. On a Jacobs test
for cognitive function the score was 19/30. A CT scan of the brain
revealed global cortical atrophy and moderate ventricular dilation
without focal lesions.
A neuropsychological examination, 2.5 months after the acute
event, revealed severe visuospatial dysfunction. There was an important attention deficit and copying was difficult but possible with perseverance. Encoding was difficult and explicit episodic memory was
poor. Language understanding was good without aphasia. The examination concluded to a dysfunction of the frontal-limbic-reticular
pathways more pronounced on the right side.

De Reuck/Vanwalleghem/Hemelsoet/
De Weweire/Strijckmans/Lemahieu

1

Fig. 1. 13NH3 PET slices of the brain at rest
and after acetazolamide administration in
case 1. The tracer uptake is high in both thalami and low in the anterior regions. After
acetazolamide administration, rCBF is increased in all regions but less in the frontal
lobes.
Fig. 2. Semiquantitative data of the tracer
distribution in the frontal (F), temporal (T),
parietal (P) and occipital (O) cortex and in
the striatum (St), thalamus (Th) and cerebellum (C) at rest and after acetazolamide administration in case 1. rCBF is low in the
frontal, temporal and parietal lobes and high
in the thalami. Acetazolamide vasoreactivity
is low in the frontal compared to the temporal and parietal lobes.

2

The PET examination was performed 3 months after the cardiac
arrest. Visual analysis disclosed a high tracer activity in both thalami
compared to the decreased activity in the frontotemporoparietal
regions. After acetazolamide administration the 13NH3 activity increased in most regions but less in the frontal lobes. There were no
significant asymmetries between both cerebral hemispheres (fig. 1).
The semiquantitative data confirmed the decreased rCBF in the

frontal, temporal and parietal regions, compared to the cerebellum
and the increased rCBF in the thalami (fig. 2). Acetazolamide vasoreactivity was less pronounced in the frontal lobes (18.4%) compared
to the temporal (36.5%) and the parietal lobes (51.5%).
The mental status of the patient remained unchanged, and she
was transferred to a chronic psychiatric department 1 month later.

PET in Post-Ischaemic-Hypoxic Amnesia

Eur Neurol 2003;49:131–136

133

3

Fig. 3. 13NH3 PET slices of the brain at rest
and after acetazolamide administration in
case 2. The tracer uptake is high in both
striata and low in the anterior regions. After
acetazolamide administration, rCBF is increased in all regions but less in the frontal
lobes.
Fig. 4. Semiquantitative data of the tracer
distribution in the frontal (F), temporal (T),
parietal (P) and occipital (O) cortex and in
the striatum (St), thalamus (Th) and cerebellum (C) at rest and after acetazolamide administration in case 2. rCBF is low in the
frontal, temporal and parietal lobes and high
in the striatum. Acetazolamide vasoreactivity is low in the frontal compared to the temporal and parietal lobes.

Case 2
A 56-year-old man with an unremarkable prior history was
admitted to a peripheral hospital after a cardiopulmonary arrest due
to myocardial infarction and ventricular fibrillation. The resuscitation had been arduous. The patient remained in deep coma for 48 h
and needed artificial ventilation. His consciousness recovered slowly
over a period of 7 weeks. He was discharged 1 week later. Due to

134

Eur Neurol 2003;49:131–136

4

severe cognitive problems he was admitted to our neurological
department 7 months after the acute event. The patient was confused
and disoriented in time and space but could recognize his family.
When alone, he was shouting and swearing. He could repeat numbers
and words. He was able to make small calculations but on reading
inverted the order of words. On a Jacobs test, he scored 7/30. He was
incontinent and had a hesitating gait with small steps. There was a

De Reuck/Vanwalleghem/Hemelsoet/
De Weweire/Strijckmans/Lemahieu

moderate rigidity of arms and legs. Tendon reflexes were brisk without Babinski signs. Suck and grasp reflexes were present.
A full neuropsychological examination was not possible due to
the extreme distraction of the patient.
CT scan and MRI of the brain showed severe cortical atrophy
mainly of the anterior regions with enlarged ventricles. No focal
lesions could be observed.
The clinical status remained unchanged. The PET examination,
performed 9 months after the acute event, disclosed on visual analysis a high tracer activity in both striata compared to the decreased
13NH activity in the frontotemporoparietal regions. After acetazo3
lamide administration, the 13NH3 activity increased in most regions
but less in the frontal lobes. There were no significant asymmetries
between both cerebral hemispheres (fig. 3). The semiquantitative
data confirmed the decreased rCBF in the frontal, temporal and parietal lobes, compared to the cerebellum, and the increased rCBF in the
striata (fig. 4). Acetazolamide vasoreactivity was less pronounced in
the frontal lobes (28.3%), compared to the temporal (49.7%) and the
parietal lobes (47.7%).
The patient was discharged home 3 months after his admission
without any improvement.

Discussion

In patients with sequelae of ischaemic-hypoxic encephalopathy after cardiopulmonary arrest and successful resuscitation the global decrease in blood flow and oxygen
metabolism on PET scan correlates with the degree of
brain damage and the clinical status of the patients [18].
The remarkable PET findings in our 2 patients with
post-ischaemic-hypoxic amnesia are the preserved or possibly increased rCBF in the thalami in the first patient and
in the striata in the second compared to the low rCBF in
the cortical areas, supplied by the anterior circulation. As

the acetazolamide vasoreactivity is most disturbed in the
frontal lobes, the memory deficits of both patients should
be attributed mainly to structural damage of those regions.
Cardiopulmonary arrest can damage the brain by a
lack of cerebral blood supply (ischaemic hypoxia) or by a
lack of cerebral oxygen supply (anoxic hypoxia), although
both mechanisms are likely to be involved to some degree.
The main frontal involvement and the relative preservation of rCBF in the deep structures in our cases should
suggest mainly an ischaemic-hypoxic cause rather than an
anoxic one [19].
The frontal lobes are involved in encoding and retrieval. The right frontal lobe is activated during encoding
and the left one during retrieval [20, 21]. Frontal lobe
function is mediated through connections with temporal,
parietal and associated visual-cortical areas [22] but also
by frontal-subcortical circuits. The latter have in common
a loop beginning in the frontal cortex, projecting to the
caudate nucleus and putamen, in turn to the globus pallidus and substantia nigra, then to the thalamus and finally
back to the frontal cortex [22]. In particular the dorsolateral frontal circuit is thought to be involved in attention
span and working memory such as motor planning, deciding which stimuli to attend to and shifting cognitive sets
[22–26]. The orbitofrontal circuit is heavily involved in
emotional life and personality structure [27]. In both of
our patients these frontosubcortical circuits must have
been disturbed leading to loss of the inhibitory GABAergic processes [28] of the corticothalamic projections in the
first patient and of the corticocaudate projections in the
second.

References
1 Verstichel P: Syndrome amnésique de Korsakoff. Presse Méd 2000;29:1670–1676.
2 Duyckaerts C, Suarez S, Hauw JJ: Les mémoires: données clinico-pathologiques. Rev
Neurol (Paris) 1998;154(suppl 2):8–17.
3 Heilman KM, Bowers D, Watson RT, Day A,
Valenstein E, Hammond E, Duara R: Frontal
hypometabolism in a patient with abnormal
orienting and retrosplenial amnesia. Neuropsychologia 1990;28:167–169.
4 Clarke S, Assal G, Bogousslavksy J, Regli F,
Towsend DW, Leenders KL, Blecic S: Pure
amnesia after unilateral left polar thalamic infarct: Topographic and sequential neuropsychological and metabolic (PET) correlations. J
Neurol Neurosurg Psychiatry 1994;57:27–34.

PET in Post-Ischaemic-Hypoxic Amnesia

5 Markowitsch HJ, Weber-Luxenburger G,
Ewald K, Kessler J, Heiss DW: Patients with
heart attacks are not valid models for medial
temporal lobe amnesia: A neuropsychological
and FDG-PET study with consequences for
memory research. Eur J Neurol 1997;4:178–
184.
6 Shalice T, Fletscher P, Frith CD, Grasby P,
Frackowiak RSJ, Dolan RJ: Brain regions associated with acquisition and retrieval of verbal
episodic memory. Nature 1994;368:633–635.
7 Grady CL, McIntosh AR, Horwitz B, Maisog
JM, Ungerleider LG, Mentis MJ, Pietrini P,
Shapiro MB, Haxby JV: Age-related reductions
of human recognition memory due to impaired
coding. Science 1995;269:218–221.

8 Maddock RJ, Garrett AS, Buonocore MH: Remembering familiar people: The posterior cingulate cortex and autobiographic memory retrieval. Neuroscience 2001;140:667–676.
9 Zola-Morgan S, Squire LR, Amaral DG: Human amnesia and the medial temporal region:
Enduring memory impairment following a bilateral lesion limited to field CA1 of the hippocampus. J Neurosci 1986;6:2950–2967.
10 Press GA, Amaral DG, Squire LR: Hippocampal abnormalities in amnesic patients revealed
by high resolution magnetic resonance imaging. Nature 1989;341:54–57.
11 Kuwert T, Homberg V, Steinmetz H, Unverhau S, Langen KJ, Herzog H, Feinendegen LE:
Posthypoxic amnesia: Regional cerebral glucose consumption measured by positron emission tomography. J Neurol Sci 1993;118:10–
16.

Eur Neurol 2003;49:131–136

135

12 Reed LJ, Marsden P, Lasserson D, Sheldon N,
Lewis P, Stanhope N, Guinan E, Kopelman
MD: FDG-PET analysis and findings in amnesia resulting from hypoxia. Memory 1999;7:
599–612.
13 Fazio F, Perani D, Gilardi MC, Colombo F,
Cappa SF, Vallar G, Bettinardi V, Paulesu E,
Alberoni M, Bressi S, Fracheschi M, Lenzi GL:
Metabolic impairment in human amnesia: A
PET study of memory networks. J Cereb Blood
Flow Metab 1992;12:353–358.
14 Phelps ME, Huang SC, Hoffmann EJ, Selin C,
Kuhl DE: Cerebral extraction of N-13 ammonia: Its dependence on cerebral blood flow and
capillary permeability-surface area product.
Stroke 1981;12:607–619.
15 Kuhl DE, Phelps ME, Howell AP, Metter EJ,
Selin C, Winter J: Effects of stroke on local
cerebral metabolism and perfusion: Mapping
by emission computed tomography of 18FDG
and 13NH3. Ann Neurol 1980;8:47–60.
16 De Reuck J, Decoo D, Hasenbroekx MC, Lamont B, Santens P, Goethals P, Strijckmans K,
Lemahieu I: Acetazolamide vasoreactivity in
vascular dementia: A positron emission tomographic study. Eur Neurol 1999;41:31–36.

136

17 Jacobs JW, Bernhard MR, Delgado A, Strain
JJ: Screening for organic mental syndromes in
the medically ill. Ann Intern Med 1977;86:40–
46.
18 De Reuck J, Decoo D, Vienne J, Strijckmans
K, Lemahieu I: Significance of white matter
lucencies in posthypoxic-ischemic encephalopathy: Comparison of clinical status and of computed and positron emission tomographic findings. Eur Neurol 1992;32:334–339.
19 Adams HJ, Graham DI: An Introduction to
Neuropathology. Edinburgh, Churchill Livingstone, 1988, pp 70–73.
20 Nijberg L, Cabeza R, Tulving E: PET studies of
encoding and retrieval: The HERA model. Psychonom Bull Rev 1996;2:134–147.
21 Wheeler MA, Stuss DT, Tulving E: Toward a
theory of episodic memory: The frontal lobes
and autonoetic consciousness. Psychol Bull
1997;121:331–354.

Eur Neurol 2003;49:131–136

22 Goldmann-Rakic P: Regional and cellular fractionation of working memory. Proc Natl Acad
Sci USA 1996;93:13473–13480.
23 Cumming JL: Frontal-subcortical circuits and
human behavior. Arch Neurol 1993;50:873–
880.
24 Taylor JG, Taylor NR: Analysis of recurrent
cortico-basal ganglia-thalamic loops for working memory. Biol Cybern 2000;82:415–432.
25 Menon V, Anagnoson RT, Glower GH, Pfefferbaum A: Basal ganglia involvement in memory-guided sequencing. Neuroreport 2000;11:
3641–3645.
26 Kalivas PW, Jackson D, Romanidies A, Wijndham L, Duffy P: Involvement of pallido-thalamic circuitry in working memory. Neuroscience 2001;104:129–136.
27 Lhermitte F, Pillon B, Serdaru M: Human
autonomy and the frontal lobes. I. Imitation
and utilization behavior: A neuropsychological
study of 75 patients. Ann Neurol 1986;19:326–
334.
28 Steriade M: Impact of network activities on
neuronal properties in corticothalamic systems. J Neurophysiol 2001;86:1–39.

De Reuck/Vanwalleghem/Hemelsoet/
De Weweire/Strijckmans/Lemahieu

Copyright: S. Karger AG, Basel 2003. Reproduced with the permission of S. Karger AG, Basel. Further
reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright
holder.