Changes in Cerebral
Glucose Metabolism
in Newborn Infants
With Cerebral
Infarction
Takashi Kusaka, MD*,
Sonoko Ijichi, MD†,
Yuka Yamamoto, MD‡, and
Yoshihiro Nishiyama, MD‡
Cerebral infarction in infants is not uncommon, and it
differs in many important ways from cerebral infarction
in older children and adults. Computed tomography,
ultrasound, and conventional and diffusion-weighted
magnetic resonance imaging are useful for diagnosing
cerebral infarction, but these imaging techniques cannot
be used to measure cerebral blood flow and metabolic
activity. Abnormality in those parameters seems to follow
a different pattern and time course than those in older
patients. In this study, the rapid changes in regional
cerebral blood flow and metabolic rate of glucose were
estimated by single-photon emission computed tomography and positron emission tomography during the acute
and subacute phases of neonatal infarction. Subacute
increases in blood flow and metabolic rate in the infarcted
area of a term infant with multiple apneic episodes within
2 days after birth were observed, as well as acute increases in both in the infarcted area of a term infant with
acute clonic seizures within 24 hours after birth. Follow-up studies at 4 months for the first infant and at 10
days for the second infant demonstrated that both the
blood flow and metabolic rate in the infarcted region
decreased. The results of this study should contribute to
an understanding of the relationship between blood flow
and metabolic rate changes after neonatal infarction as
well as to improvement of diagnosis of neurologic impairments in neonates. © 2005 by Elsevier Inc. All rights

From the *Maternal Perinatal Center and the Departments of
†
Pediatrics and ‡Radiology, Kagawa University, Kagawa, Japan.

46

PEDIATRIC NEUROLOGY

Vol. 32 No. 1

reserved.
Kusaka T, Ijichi S, Yamamoto Y, Nishiyama Y. Changes in
cerebral glucose metabolism in newborn infants with cerebral
infarction. Pediatr Neurol 2005;32:46-49.

Introduction
Cerebral infarction in infants is not uncommon, and it
differs in many important ways from cerebral infarction in
older children and adults. The incidence of neonatal
cerebral infarction has been estimated to be 1 in 4000 live
term-birth infants [1]. Several imaging techniques such as
computed tomography, ultrasound, and conventional and
diffusion-weighted magnetic resonance imaging have
been used to investigate cerebral infarction [2,3]. However, these imaging techniques cannot be used to estimate regional cerebral blood flow and metabolic activity.
Measurements of regional cerebral metabolic rate of glucose (rCMRglu) in infants have been performed in a
few studies using positron emission tomography (PET)
[4,5]. In this study, we measured acute-phase and steadyphase regional cerebral blood flow using Tc-99m l,
l-ethyl cysteinate dimer single-photon emission computed
tomography (99mTc-ECD SPECT) and rCMRglu using
18
F-fluorodeoxyglucose positron emission tomography
(FDG-PET) imaging in 2 patients with neonatal infarction.
This investigation is the first report on regional cerebral
blood flow and rCMRglu abnormalities during acute and
subacute phases in infants with cerebral infarction.
Methods
99m

Tc-ECD SPECT Procedure

99m
Tc-ECD (111 MBq) was injected intravenously while the patient
was lying down in a quiet and dim room. 99mTc-ECD SPECT images
were obtained using a Prism 3000 (Picker International, Cleveland, Ohio)
and high-resolution fan-beam collimators with a spatial resolution of
8 mm full width at half maximum. The projection data were obtained in
a 128 ⫻ 128 matrix (magnification, ⫻ 1.123) for 24 angles at 30 seconds
each over a range of 120°. Image reconstruction was performed using
ramp-filtered backprojection. Transverse sections of 6.8 mm in thickness
were reconstructed. Scanning was parallel to the orbitomeatal line. Each
infant was sedated with thiamylal sodium (5 mg/kg) before both
FDG-PET and SPECT transmission scanning. Written informed consent
was obtained from the parents of each infant, and the study was approved
by a local ethics committee.

Communications should be addressed to:
Dr. Kusaka; Department of Pediatrics; Faculty of Medicine; Kagawa
University; Mikicho 1750-1, Kitagun; Kagawa 761-0793; Japan.
Received July 29, 2003; accepted June 10, 2004.

© 2005 by Elsevier Inc. All rights reserved.
doi:10.1016/j.pediatrneurol.2004.06.012 ● 0887-8994/05/$—see front matter

FDG-PET Procedure
For each patient, feeding was discontinued for 3 hours before intravenous administration of FDG (120 MBq). The FDG was administered in
a quiet, dimly lit room. The patients remained in this environment for
40 minutes after FDG administration and then were positioned for a PET
scan using an ECAT EXACT HR⫹ scanner (CTI, Knoxville, Tennessee/
Siemens Medical Systems, Inc., Hoffman Estates, Illinois). Transaxial
data were acquired, and emission data were corrected by using calculated
attenuation correction. The data were displayed as transaxial, coronal,
and sagittal images.

Clinical Evaluations
Patient 1
This female child was delivered with normal body
weight after 41 weeks of gestation. A physical examination after birth did not reveal any abnormalities. Gestation
and delivery had been uneventful. At 36 hours after
delivery, however, apneic episodes occurred. Other than
these episodes, findings of a physical examination were
normal. Findings of transfontanel ultrasound and lumbar
puncture were normal, ruling out the possibility of intracranial hemorrhage or meningitis. Computed tomography
was performed 4 days after the onset of symptoms, and a
diagnosis of left middle artery infarction was made. The
first magnetic resonance imaging was performed 10 days
after delivery. T2-weighted images revealed a left middle
cerebral artery infarction. On the same day, SPECT
documented a slight increment of regional cerebral blood
flow in the infarcted area, and FDG-PET revealed an
increment of rCMRglu in the same area in which an
increase in regional cerebral blood flow was observed by
SPECT. A follow-up magnetic resonance imaging study
was performed 4 months later. The infarcted brain area
was clearly visible on T2-weighted images, and the subarachnoid space was apparently increased over the entire
left hemisphere, which appeared to be smaller than the
right hemisphere. SPECT disclosed a decrease in regional
cerebral blood flow, and FDG-PET revealed a decrease in
rCMRglu in the infarcted region. The patient manifested
normal growth and development at 6 months of age.
Patient 2
This female child was delivered with normal body weight
after 41 weeks of gestation. A physical examination after
birth did not reveal any abnormalities. Gestation and delivery
had been uneventful. There were no risk factors. After 24
hours, clonic seizures of the right arm spontaneously developed. Transfontanel ultrasound and cerebrospinal fluid studies did not reveal any abnormality. She was treated with
phenobarbital and midazolam for the seizures. The first
magnetic resonance imaging was performed 5 hours later.
Diffusion-weighted magnetic resonance imaging revealed an
abnormality in the territory of the left middle cerebral artery,
and T2-weighted images revealed a slight high intensity.

Based on these findings, a diagnosis of left middle artery
infarction was made. On the same day, SPECT documented
an increment of regional cerebral blood flow in the infarcted
area, and FDG-PET recorded an increment of rCMRglu in
almost the same area in which an increase in regional cerebral
blood flow was observed by SPECT. A follow-up scan was
performed 10 days later because conventional sequences
demonstrated infarction, and the infarcted brain area was
clearly visible on T2-weighted images. The diffusionweighted sequence revealed a decreased signal in the infarcted area. SPECT disclosed a decrease in regional cerebral
blood flow, and FDG-PET indicated a decrease in rCMRglu
in the same region (Fig 1). Furthermore, the right cortex was
more metabolically active in rCMRglu than the thalamus and
basal ganglia. The patient manifested normal growth at 2
months of age, and no seizure occurred.
Discussion
The main findings of this study of infants with cerebral
infarction are as follows: subacute increases in regional
cerebral blood flow and rCMRglu in a term infant with
normal results on an initial physical examination, and apneic
episodes within 2 days after birth; and acute increases in
regional cerebral blood flow and rCMRglu in the infarcted
region in a term infant with no known abnormalities at birth
and acute clonic seizures within 24 hours after birth. In the
acute phase of infarction in adults, increased cerebral blood
flow and rCMRglu are also observed for up to 14 days after
the attack (under the condition of luxury perfusion) [6].
Therefore the increments in both regional cerebral blood flow
and rCMRglu in the infarcted area in the cases reported here
may indicate a condition of luxury perfusion. This observation may indicate the occurrence of anaerobic glycolysis and
production of lactate by aerobic glycolysis as a postischemic
metabolic abnormality in the infarcted area.
In Patient 2, T2-weighted sequences did not reveal an
infarction, but diffusion-weighted magnetic resonance imaging revealed an abnormality in the territory of the left
middle cerebral artery. Diffusion-weighted magnetic resonance imaging is particularly valuable for prediction of
the timing of a cerebrovascular insult because it can reveal
parenchymal injury earlier than can conventional magnetic
resonance imaging, abnormality being revealed within a
few hours after the onset of stroke [2,3]. Therefore it is
likely that infarction occurred at the time of birth and that
regional cerebral blood flow and rCMRglu increased in
the infarcted area and then began to decrease from day 10
after birth. The timing of the increment and decrement of
regional cerebral blood flow and rCMRglu was earlier
than the timing in adults. This difference in timing may be
the result of the differences in etiology of infarction,
fibrinolytic state, energy source and metabolism, and this
difference in timing is related to the fact that glycolytic
enzymes in the developing brain are expressed earlier and
in greater quantity than they are in the adult brain [7]. A
further noteworthy point in Patient 2 is that the right cortex

Kusaka et al: Metabolism of Neonatal Infarction 47

Figure 1. Diffusion-weighted (spin-echo, echo-planner imaging; B factor ⫽ 1000; slice thickness 5 mm [contiguous]; TR ⫽ 6000 ms and TE ⫽ 110 ms on
day 2; TR ⫽ 8700 ms and TE ⫽ 110 ms on day 10) and T2-weighted magnetic resonance imaging (fast spin echo; slice thickness 5 mm [contiguous]; TR ⫽
4050 ms and TE ⫽ 102 ms on day 2; TR ⫽ 4100 ms and TE ⫽ 102 ms on day 10), 99mTc-ECD SPECT, and FDG-PET images obtained on day 2 (upper row)
and on day 10 after birth (bottom row) in Patient 2. On day 2, a diffusion-weighted magnetic resonance image clearly revealed an infarction in the territory
of the left middle cerebral artery. At the same time, regional cerebral blood flow measurement indicated spontaneous luxury perfusion and an increment of
rCMRglu in this area. In the diffusion-weighted magnetic resonance image obtained on day 10, the infarcted parenchyma is slightly hypotense compared with
the unaffected area, and a T2-weighted magnetic resonance image documented middle cerebral infarction clearly in the white matter. At the same time, the
regional cerebral blood flow and rCMRglu in the infarcted area had decreased.

was more metabolically active than the thalamus and basal
ganglia, whereas the opposite phenomenon has been
reported in infants with and without hypoxic-ischemic
encephalopathy at the same age [4,7]. Activation in the
contralateral side of the infarcted area during the subacute
period may be related to the plasticity of this neonate.
Seizures were observed in Patient 2 at the time of the
first magnetic resonance imaging scan. Seizures in neonates with cerebral infarction are often observed during
the first week after birth, and it is thought that seizures can
occur easily during the period of increases in regional
cerebral blood flow and rCMRglu, just after infarction.
In this study, only a relative assessment of the infarcted
area compared with other regions on the same scanning
was made, because only relative values of regional cerebral blood flow and rCMRglu could be used. If quantitative values of regional cerebral blood flow and rCMRglu
can be measured, it will be possible to compare values at
different time points and in different patients [7]. Furthermore, we must consider the effects of repeated sedation
because barbiturates reduce metabolism in the brain.
Studies using more cases and more time points must be

48

PEDIATRIC NEUROLOGY

Vol. 32 No. 1

performed to assess variations in regional cerebral blood
flow and rCMRglu, their prognostic significance, and the
degree of recovery in order to obtain better insights into
the natural course of infantile infarction and to develop
better treatment strategies.
In summary, simultaneous measurements of regional
cerebral blood flow and rCMRglu in two infants with
cerebral infarction are reported here. It is expected that the
estimation of changes in regional cerebral blood flow and
rCMRglu during the acute and subacute phases of cerebral
infarction in infants will be used in the future for determining prognosis and predicting results of treatments.
This study was supported by grants-in-aid for scientific research (C), no.
15591158, 15591159 and 14571051, and for the encouragement of young
scientists, no. 14770571 and 13770619, from the Ministry of Education,
Culture, Sports, Science and Technology of Japan.

References
[1] Estan J, Hope P. Unilateral neonatal cerebral infarction in full
term infants. Arch Dis Child Fetal Neonatal Ed 1997;76:F88-93.

[2] Mader I, Schoning M, Klose U, Kuker W. Neonatal cerebral
infarction diagnosed by diffusion-weighted MRI: pseudonormalization
occurs early. Stroke 2002;33:1142-5.
[3] Miller V. Neonatal cerebral infarction. Semin Pediatr Neurol
2000;7:278-88.
[4] Chugani HT, Phelps ME. Mazziotta JC. Positron emission
tomography study of human brain functional development. Ann Neurol
1987;22:487-97.

[5] Thorngren-Jerneck K, Ohlsson T, Sandell A, et al. Cerebral
glucose metabolism measured by positron emission tomography in term
newborn infants with hypoxic ischemic encephalopathy. Pediatr Res
2001;49:495-501.
[6] Heiss WD, Herholz K. Assessment of pathophysiology of stroke
by positron emission tomography. Eur J Nucl Med 1994;21:455-65.
[7] Powers WJ. Cerebral blood flow and metabolism in newborn
infants. What can they teach us about stroke? Stroke 1993;24:I33-4.

Kusaka et al: Metabolism of Neonatal Infarction 49