IMAGING BRAIN TUMORS IN NEWBORNS AND EARLY CHILDHOOD: UTILITY OF COMBINING MR TECHNIQUES

1
IMAGING BRAIN TUMORS IN NEWBORNS AND EARLY
CHILDHOOD UTILITY OF COMBINING MR TECHNIQUES

• M. MORTILLA, M. ANTONELLO, C. CESARINI,
• L. TASCIOTTI, C. FONDA
• UNIVERSITY CHILDRENS HOSPITAL A. MEYER
• FIRENZE, ITALY

2
INTRODUCTION
In the childhood CNS tumors are the leading
cause of cancer-related death. In the last 20
years the advances in neuroimaging, neurosurgery,
radiation therapy and chemotherapy have
considerably improved the long term survival of
children with brain tumors. Conventional MR
images show definite details of brain tumor
location, extension and morphological
characteristics. Therefore MR imaging is widely
used in the diagnosis and follow-up of pediatric
patients with brain tumors.
3
Since conventional MRI does not provide
information about tissue chemistry and the
interpretation of these images may lead to poor
estimation of the extent of active tumor, non
conventional techniques, such as diffusion images
and proton spectroscopy, may be used contributing
to more accurate diagnosis, prognostication and
treatment planning. Tissue diagnosis remains the
gold standard. The purpose of the study is to
find a role of Diffusion-weighted Imaging (DWI)
and proton Spectroscopy in characterizing
intracerebral masses and finding a correlation
between these techniques and histologic analysis
of tumors.
4
PATIENTS METHODS
62 patients affected with brain tumors aged 1
month-6 years, were studied with a 1.5T MR
scanner (Eclipse, Philips) operating at 27mT/m
gradient strenght and 40 mT/m/ms slew rate. A
quadrature head coil was used. T2w SE images,
FLAIR, T1w SE pre and post Gadolinium injection
and sometimes GE T2 were obtained. Proton
Spectroscopy studies included single voxel
studies (PRESS TE 40/135/270ms, STEAM TE 20ms)
and/or CSI (PRESS TE 135/270.
5
Spectroscopy acquisition were performed before
the injection of the contrast media. Eight
children were able to undergo to the MR without
need of sedation despite the long duration of the
exam (50-60 minutes). The other children were
sedated with different modalities regarding
weight, age and critical conditions patients up
to 18 months of age (if the weight was below
10Kg) were sedated with chlorale hydrate (50-100
mg/Kg), while from 18mo. to 6yrs. patients were
sedated using Sevoflurane through laringeal mask
or, rarely, with barbiturates i.v.
6
Echo planar Imaging of Diffusion were obtained
with sensitization along the slice select, read
out and phase encoding axes (b- value of
800/1000/1200) with Echo Planar Single Shot
sequences (TR 6450ms, TE 145.8ms, FA 90, 5mm/1mm
slice thickness/gap, 81x81 gt 128x 128
reconstruction matrix, 1 NEX, chemical saturation
at FA 180, FAT saturation) with a total
acquisition time of 32 seconds/15 slices.
7
In all patients a set of 3 images along the long
orthogonal axes of gradient sensitization were
obtained and the DWI TRACE images were
post-processed. Apparent Diffusion Coefficient
(ADC) images along the same orthogonal axes were
also obtained and synthetic ADC map was produced
(ADC TRACE. ADC values are expressed as a number
x 10-3mm2/sec.
8
Multivoxel (CSI) were acquired with PRESS
sequence, TR 1500ms, TE 135ms, FA 90, thickness
1-1.5 cm, FOV 15 to 20cm. Chemical shift imaging
matrix size 16x16, signal averages 1/2 with 1
slice per batch. Single voxel were acquired with
PRESS sequence, TR 1500/2000ms, TE 270/135/40ms
and with STEAM TE 20ms, FA 90 acquisition
volume from 20x20x20mm, signal averages 128 and
reference averages 8.
9
Most of the tumors underwent to surgery
excision and the specimens were analyzed by an
expertise in pathology in order to make diagnosis
following the WHO classification and to determine
cell counting expressed as mean value (m.v.) over
an are of 0.083mm2. Some tumors were biopsied
only, such as germinoma because they are highly
responsive to therapy. Biopsy or surgical
excision was not performed when the tumor arised
from non resectable regions (hypotalamus,
brainstem,..)
10
OUR CASES

• Craniopharingioma 5
• Low grade glioma (WHO I) 11
• Glioma WHO II 2
• High grade glioma 1
• Pylocitic astrocytoma 10
• Medulloblastoma (MB-PNET) 11
• Germinoma 1
• Subependimal gigantic astrocytoma 2
• Ganglioblastoma 1
• Ganglioglioma 1
• Dysplasia 3
• DNET 2
• Teratoma 1
• Metastasi 1
• Altri 10

11
Diffusion-weighted Imaging
It has been reported that the ADC characterizes
the biophysical characteristics of tissue
microstructure and microdynamics and that it
provides information (based on pathophysiologic
characteristics) that differs from that obtained
with contrast-enhancing imaging. It has been
suggested that the minimum ADC value of
high-grade gliomas is significantly higher than
that of low-grade gliomas and that low ADC values
were found in areas of increased cellularity.
12
Others have suggested that the ADC may assist in
the early detection of responses to anticancer
therapy, because an increase in ADC values has
been noted after treatment. DWI measures the
molecular mobility of extracellular
water,alterations in water mobility appear to
reflect treatment-induced changes in tissue
structure. Current understanding is that water
diffusion increases acutely in tumor responsive
to therapy. This precede changes in tumor volume.
13
1H-MRS (proton magnetic resonance spectroscopy)

• 1H-MRS (Magnetic Resonance Spectroscopy)
  provides a qualitative and quantitative
  evaluation of brain chemistry
• In a proton spectrum at 1.5T the metabolites are
  spread out between 63 and 64 MHz, and near 300
  MHz at 7T
• The resonant frequencies are expressed in part
  per million (ppm), and are read from right to
  left

14
1H-MRS

• Fat and water are eliminated, because their
  peaks are to high and in spectrum scaling the
  brain metabolites would be invisible
• The water suppression is obtained with CHESS
  (Chemical Shift Selective) or IR (Inversion
  Recovery) techniques
• The peaks are separated into the individual
  frequencies through a Fourier Transform
• The magnetic field felt by the Protons in
  different molecules depends on electron clouds
  related to their different molecular position -gt
  different chemical shift -gt spread of single
  peaks over the ppm or hertz scale

15
Sequences in 1H-MRS

• STEAM (Stimulated Echo Acquisition Mode) 90o
  refocusing pulse, short echo time, less
  signal-to-noise ratio
• PRESS (Point Resolved Spectroscopy) 180o
  refocusing pulse, short and long echo time
• With short echo time (TE 20-40 ms) metabolites of
  both short and long T2 are visualized
• With long echo time (TE 270 ms) only metabolites
  with long T2 are seen . Echo Time of 135 ms
  allows the separation of lactate doublet from
  lipids peaks with phase inversion.
• TE 65 ms increases sensitivity in lipid detection
  nulling lactate

16
voxel

• MR Spectra may be acquired with a single voxel
  localized in region of interest (normal or
  pathological) with variable volume (usually of
  2x2x2 cm or more). Small volume are characterized
  by less signal to noise ratio. Large volumes
  experience higher averaging and are not indicated
  for higher resolution data collection, but only
  for mean value in a defined area

17
multivoxel

• MR Spectra may be acquired within a brain slice
  in 2D acquisition or more slices in 3D
  acquisition) with variable matrix and variable
  volume (usually of 1x1x1 cm or more) Small volume
  of voxels experience lower averaging than single
  voxel with larger volume. Chemical Shift Imaging
  (CSI) may create the metabolite maps with direct
  visualization of peak concentration
• Magnetic field inhomogeneity, insufficient
  shimming and lipids contamination frequently
  alter the quality of multivoxel spectra.

18

• Many metabolites may be identified in the proton
  magnetic resonance at 1.5 tesla, NAA, Cho, Cr,
  Lactate, myo-Inositol, Lipids are currently
  evaluated
• In the following slides there is a list brain
  peaks

19
List of metabolites that can be individualized
by 1H-MRS

• N-acetyl methylgroups
• (NAA N-acetylaspartate and NAAG
  N-acetylaspartylglutamate)
• Methyl and Methylene protons of total creatine
  (Cr PCr)
• Trimethylammonium groups Choline containing
  (Cho)
• Myo-Inositol (mI)
• Glycine co-resonating with main mI peak (Gly)
• Glutamate Glutamine with a and b-/g- protons
  (Glu/Gln)
• Glucose
• Scyllo-inositol
• Lactate
• GABA
• Glutathione
• Taurine
• Homo-carnosine
• Phospho-ethanolamine
• Macromolecules
• Lipids

20
Resonance intensitiesexpressed in ppm at 1.5T

• Lactate/lipids 1.33
• NAA 2.02
• Glx 2.2-2.4
• Cr/PCr 3.02
• Cho 3.22
• mI 3.56

21
Control 5 yrs old CSI PRESS TE 135ms
22

• NAA(N-acetyl aspartate) free aminoacid, high CNS
  concentration (just less to glutamate)
• in adults in neural tissue, axons and dendrites
• in brain maturation also in oligodendrocytes type
  2 and in non neuronal cells (mast cells)
• used as neuronal marker

23

• Cho (choline - N(CH3)3 GPC,PC) cellular membrane
  turnover marker
• High in tumors, demyelinating processes,
  inflammation
• Cr ( creatine - Cr PCr k), reference internal
  due to its stability, marker of byproducts of
  energy chains gtATP
• Lac (Lactate) expresses the anaerobic metabolism

24

• mI (myo inositol) glial pool marker. Small
  amounts from glycine
• Glu or Glx (glutamate) neurotransmitter,
  intermediate in aminoacid catabolism
• Gln (glutamine) metabolism of glutamate glial
  marker

25
PRESS TE 135ms
STEAM TE 20ms
? Cho ? Lac ? NAA
4 years old girl pilocytic astrocytoma Lactate
in solid nodule High choline peak In P.A. the
Choline Levels are usually below 3.0, while in
MB-PNET are usually higher
b
a
c
26
Mean ADC value 1.68 10-3 mm2 /sec
DWI TRACE
ADC TRACE
215 m.v.
4y.o. girl pilocytic astrocytoma. Histologic
surgical specimen and cell density counting
over mixoid (185 cells) and more compact (215
cells) portion of the tumor.
185 m.v.
Nr. of cell 0.083mm2
27
PRESS TE 270ms
PRESS TE 135ms
Boy, 13 months BRAINSTEM GLIOMA Cho/Cr NAA/Cr Lac
MRS data indicates that it
could be a pilocytic astrocytoma
28
6 years old boy Brainstem glioma
1st MRI diagnosis (WHO I) 2ndafter chemo High
Cho 3rd6mo. after Stop Therapy Higher Cho,
low NAA, high lac
FLAIR C.E. FSE T1 ADC TRACE
PRESS TE 270ms
29
DWI TRACE ADC TRACE
DWI TRACE ADC TRACE
At diagnosis b 800 DWI ADC
After chemotherapy
30
6 years old girl glioma of the midbrain (WHO I)
Mild reduction of NAA/Cr and mI/Cr
PRESS TE 40ms
31
Boy, 4 years old Astrocytoma WHO II-III
Important reduction of NAA/Cr and moderate
increase of Cho/Cr
32
b
VOXEL 8
STEAM TE 20 ms
CSI PRESS TE 270 ms
6 years old boy medulloblastoma
CSI PRESS TE 270ms high choline peak (Cho/Cr
4.09), low NAA intensity signal. Small amount of
lactate. STEAM TE 20ms evident lipids peak.
a
33
a
DWI TRACE
ADC TRACE
Mean ADC value 1.2 10-3 mm2 /sec
Medulloblastoma.
750 m.v.
Nr. of cell 0.083mm2
34
PRESS TE 135ms
b
STEAM TE 20ms
6 years old girl. PNET-MB
..
STEAM 20 evident mI, lipids and Glx peaks PRESS
TE 135ms high choline peak (Cho/Cr 12.4) and low
NAA intensity signal.
a
35
STEAM TE 20ms
PRESS TE 135ms
11 months old girl pinealoblastoma.
PRESS 135 high choline peak (Cho/Cr 2.6), low
NAA signal intensity. Presence of lactate.
STEAM 20 evident lipids and mI peaks.
36
Mean ADC value 0.4 x 10-3 mm2 /sec
DWI TRACE
ADC TRACE
Pinealoblastoma
900 m.v.
Nr. of cell 0.083mm2
37
6 years old girl choroid plexus carcinoma.
All metabolites, included Creatine, but Choline
are reduced Cho/Cr 31.5
Mean ADC value 1.0 10-3 mm2 /sec
PRESS TE 135ms
reduced Cr
neoplasm do not produce NAA
38
5 years old girl germinoma.
PRESS TE 135ms
high lipids signal.
Mean ADC value 0.57 10-3 mm2 /sec
Nr. of cell 0.083mm2
320 m.v.
39
Cho/Cr 6.1 Lac/Cr 6.8 NAA/Cr
CSI PRESS TE 135ms
PRESS TE 270ms
Girl, 23 months metastasis from rabdomyosarcoma
40
Cortical Dysplasia TAYLOR TYPE
Increased mI/Cr ratio
STEAM TE 20ms
41
STEAM TE 20ms
PRESS TE 40ms
Boy, 20 months Tuberous sclerosis Increased
mI/Cr ratio
42

ADC vs.Cells number correlation
Germinoma
more highly cellular gt smaller intercellular
spaces gt lower ADC
JDGG
Pinealoblastoma
PNET
MB-PNET
Pylocitic Astrocytoma (solid portion)
43
BRAIN ABSCESS vs. TUMOR
a
b
DWI ADC
DWI is able to discriminate between abscesses and
tumors a) brain abscess b) supratentorial
ependymoma
44
Boy, 6 years old brain abscess
Most of the metabolites are reduced. Lipids and
lactate are present
45
Conclusions
Brain tumors in children are highly heterogeneous
for histology, prognosis and therapeutic
response. Diagnosis and therapy of those, most of
which are low grade, can be complicated because
of their frequent adjacent location to crucial
structures that limits biopsy. The utility of
combining data from biologically important
intracellular molecules, obtained with proton MR
spectroscopy and from water mobility, obtained
with diffusion imaging, is clearly addressed to
increase the diagnostic accuracy in determining
the clinical grade of pediatric brain tumors.
46
DWI enable to better differenciate between low-
grade and high-grade tumor high-grade gliomas
have lower ADC values than low-grade gliomas. ADC
maps are easily generated from routine fast
diffusion-weighted imaging by use of software
available on many MR systems. ADC may be a more
direct indicator of changes in the brain than are
other physical parameters. The degree of
diffusion is strongly affected by microscopic
biological structures such as the number, type
and spatial arrangement of cells. These
structures create barriers to the free diffusion
of water so changes in diffusion may more
directly reflect changes occurring within and
between cells.
47
It still debating if ADC measurement can be used
to determine the extent of tumor infiltration and
to differentiate infiltration from peritumoral
edema. It has been suggested that tumor
infiltration is characterized by lower ADC values
than edema. In our experience we found that quite
often this is true but we always prefer to be
cautious adding spectroscopy data when is
possible to perform CSI. We also utilize
high-b-value DWI that increase the anisotropy so
is more accurate in the assessment of
infiltration. Since infiltration occurs within
and along white matter tracts, diffusion tensor
imaging may yield more useful information.
48
Proton MR Spectroscopy enables the measurements
of multiple chemical metabolites in normal and
abnormal brain parenchyma. Cho/Cr or Cho/NAA in a
lesion correlate with higher cellular
proliferation rate and reflect the presence of a
more malignant and rapidly growing tumor. It is
necessary to correlate MRI data since pilocytic
astrocytoma has high Choline and lactate despite
it is considered benign usually Cho/Cr ratio is
less than 3. NAA is considered a neuronal marker,
which decreases with replacement of neurons by
tumor (or other non-neuronal tissue, including
necrosis). In choroid plexus carcinoma a very low
NAA/Cr is characteristic since the tumor does not
produce NAA.
49
MRS can be used to follow tumors over time, since
patients can serve as their own control after
obtaining a baseline scan. CSI may be useful in
monitoring the surgical scar elevation of Cho/Cr
ratio is index of a relapse. We have found this
index reliable mostly in anaplastic ependymoma.
It has been suggested that children with higher
total creatine levels are more responsive to
radiation or chemotherapy. In our experience
children with low-grade gliomas that have
significantly higher baseline Cho/Cr ratio have
more chance to have tumor that progress over 2
years than those that have stable tumors. Lactate
is no more considered an indicator of malignancy
since it is found in benign tumors such as
pilocytic astrocytoma.
50
Lipids may be detected in enhancing and non
enhancing tumor regions. For a reliable detection
of lipid peak it would be better to use
acquisition protocols with a TE 65ms that null
lactate peak. Lipids represent microscopic tumor
cell necrosis or membrane breakdown that may
precede necrosis. Lipids may be present in viable
tumor, presumably because of poor perfusion and
hypoxia and they undergo major intensity changes
during apoptosis. They are found also in
radiation necrosis. Because glial tumors are
grade according to their cellularity,
proliferative activity and degree of necrosis,
Cho mapping (increased cellularity and
proliferative activity) may added value to MRI in
childrens with brain tumors, especially when it
is combined with lipid mapping (necrosis and/or
apoptosis).
51
Selected References

1. Arnold DL, De Stefano N. Magnetic resonance
   spectroscopy in vivo applications in
   neurological disorders. Ital J Neurol Sci 1997
   18 321-329
2. Ashwal S, Holshouser B, Tong K, Serna T,
   Osterdock R, Gross M, Kido D. Proton spectroscopy
   detected myoinositol in children with traumatic
   brain injury. Pediatric Research 2004 56630-638
3. Astrakas LG, Zurakowski D, Tzika AA, Zarifi MK,
   Anthony DC, De Girolami U, Tarbell NJ, Black PM.
   Noninvasive magnetic resonance spectroscopic
   imaging biomarkers to predict the clinical grade
   of pediatric brain tumors. Clin Cancer Res. 2004
   Dec 1510(24)8220-8.
4. Barkovich AJ, Good WV, Koch TK, et al.
   Mitochondrial disorders analysis of their
   linical and imaging characteristics. AJNR 1993
   14 1119-1137
5. Bottomley PA. Spatial localization in NMR
   spectroscopy in vivo. Ann NY Acad Sci 1987 508
   333348
6. Bottomley PA. Spatial localization in NMR
   spectroscopy in vivo. Ann NY Acad Sci 1987 508
   333-348
7. Brooks WM, Friedman SD, Gasparovic C. Magnetic
   resonance spectroscopy in traumatic brain injury.
   J Head Trauma Rehabil 2001 16146164
8. Castillo M, Kwock L, Scatliff J, Mukherji SK.
   Proton MR spectroscopy in neoplastic and
   non-neoplastic brain disorders. Magn Reson
   Imaging Clin N Am. 1998 6(1)1-20
9. Cecil KM, Jones BV. Magnetic resonance
   spectroscopy of the pediatric brain. Top Magn
   Reson Imaging 200112435452
10. Cendes F, Andermann F, Preul MC, et al.
    Lateralization of temporal lobe epilepsy based on
    regional metabolic abnormalities in proton
    magnetic resonance spectroscopic imaging. Ann
    Neurol 1994 35 211-216
11. Frahm J, Hanefeld F. Localized proton magnetic
    resonance spectroscopy of cerebral metabolites.
    Neuropediatrics 1996 27 64-69
12. Grodd W, Krageloh-Mann I, Klose U, et al.
    Metabolic and destructive brain disorders n
    children findings with localized proton MR
    spectroscopy. Radiology 1991 81 173-181
13. Howe FA, Maxwell RJ, Saunders DE, et al. Proton
    spectroscopy in vivo. Magn Reson Q 1993 9
    31-59
14. Hunter JV, Wang ZJ. MR spectroscopy in pediatric
    neuroradiology. Magn Reson Imaging Clin N Am.
    2001 9(1)165-89
15. Hwang JH, Egnaczyk GF, Ballard E, Dunn RS,
    Holland SK, Ball WS Jr. Proton MR spectroscopic
    characteristics of pediatric pilocytic
    astrocytomas. AJNR Am J Neuroradiol. 1998
    19(3)535-40
16. Kadota T, Horinouchi T, Kuroda C. Development and
    aging of the cerebrum assessment with proton MR
    spectroscopy. AJNR Am J Neuroradiol
    200122128135
17. Kreis R, Ernst T, Ross BD. Development of the
    human brain in vivo quantification of metabolite
    and water content with proton magnetic resonance
    spectroscopy. Magn Reson Med 1993 30 424437
18. Kruse B, Hanefeld F, Christen HJ, et al.
    Alterations of brain metabolites in
    metachromatic leukodystrophy as detected by
    localized proton magnetic resonance spectroscopy
    in vivo. J Neurol 1993 241 68-74

52

1. Leonard JC, Younkin DP, Chance B, Subramanian VH,
   Leigh JS Jr, Alavi A, Kressel HY, Zimmerman R,
   Delivoria-Papadopoulos M. Nuclear magnetic
   resonance an overview of its spectroscopic and
   imaging applications in pediatric patients. J
   Pediatr. 1985 106(5)756-61.
2. Moore GJ. Proton magnetic resonance spectroscopy
   in pediatric neuroradiology. Pediatr Radiol. 1998
   Nov28(11)805-14.
3. Norfray JF, Tomita T, Byrd SE, Ross BD, Berger
   PA, Miller RS. Clinical impact of MR spectroscopy
   when MR imaging is indeterminate for pediatric
   brain tumors. AJR Am J Roentgenol 1999 173 119
   125
4. Norfray JF, Tomita T, Byrd SE, Ross BD, Berger
   PA, Miller RS. Clinical impact of MR spectroscopy
   when MR imaging is indeterminate for pediatric
   brain tumors. AJR Am J Roentgenol. 1999
   173(1)119-25
5. Novotny E, Ashwal S, Shevell M. Proton magnetic
   resonance spectroscopy an emerging technology in
   pediatric neurology research. Pediatr Res. 1998
   44(1)1-10
6. Penrice J, Cady EB, Lorek A, Wyelinska M, Amess
   P, Aldridge R, Stewart A, Wyatt J. Proton
   magnetic resonance spectroscopy of the brain in
   normal preterm and term infants and early changes
   after perinatal hypoxia-ischemia. Pediatr Res
   1996 40 614
7. Pietz J, Kreis R, Schmidt H, Meyding-Lamadé UK,
   Rupp A, Boesch C. Phenylketonuria findings at MR
   imaging and localized in vivo H-1 MR spectroscopy
   of the brain in patients with early treatment.
   Radiology 1996 413420
8. Pouwels PJ, Frahm J. Regional metabolite
   concentrations in human brain as determined by
   quantitative localized proton MRS. Magn Reson Med
   1998395360
9. Rand SD, Prost R, Li SJ. Proton MR spectroscopy
   of the brain. Neuroimaging Clin N Am. 1999
   9(2)379-9
10. Robertson NJ, Lewis RH, Cowan FM, Allsop JM,
    Counsell SJ, Edwards AD, Cox IJ. Early increases
    in brain myo-inositol measured by proton magnetic
    resonance spectroscopy in term infants with
    neonatal encephalopathy. Pediatr Res. 2001
    50(6)692-700.
11. Ross BD. Biochemical considerations in 1H
    spectroscopy. Glutamate and glutamine
    myo-inositol and related metabolites. NMR Biomed
    1991 4 5963
12. Ross BD. Biochemical considerations in 1H
    spectroscopy. Glutamate and glutamine
    myo-inositol and related metabolites. NMR 1991
    Biomed 4 59-63
13. Rutherford MA, Pennock JM, Counsell SJ, Mercuri
    E, Cowan FM, Dubowitz LMS, Edwards AD. Abnormal
    magnetic resonance signal in the internal capsule
    predicts poor neurodevelopmental outcome in
    infants with hypoxic-ischemic encephalopathy.
    Pediatrics 1998 102 323328
14. Saunders DE, Howe FA, van den Boogaart A,
    Griffiths JR, Brown MM. Discrimination of
    metabolite from lipid and macromolecule
    resonances in cerebral infarction in humans using
    short echo proton spectroscopy. J Magn Reson
    Imaging 1997 7 11161121
15. Thompson JE, Castillo M, Kwock L. MR spectroscopy
    in the evaluation of epilepsy. Magn Reson Imaging
    Clin N Am. 1998 6(1)21-9.
16. Tzika A, Astrakas L, Kieran M, Zurakowski D,
    Zarifi M, Poussaint T Choline to
    n-acetylaspartate ratio is predictive of
    pediatric brain tumor progression. Radiology
    2001 221(Suppl)488
17. Tzika AA, Ball WS Jr, Vigneron DB, et
    al.Childhood adrenoleukodystrophy assessment
    with proton MR spectroscopy. Radiology 1993 189
    467-480
18. Tzika AA, Zarifi MK, Goumnerova L, Astrakas LG,
    Zurakowski D, Young-Poussaint T, Anthony DC,
    Scott RM, Black PM. Neuroimaging in pediatric
    brain tumors Gd-DTPA-enhanced, hemodynamic, and
    diffusion MR imaging compared with MR
    spectroscopic imaging. AJNR Am J Neuroradiol.
    200223(2)322-33
19. Warren KE. NMR spectroscopy and pediatric brain
    tumors. Oncologist 2004 9(3)312-8.