Mutsumasa Takahashi, M.D.


Despite recent improvement in the differential diagnosis of brain tumors and
grading of gliomas, it has been rather difficult to make a definitive diagnosis and
grading before therapeutic approaches, which may be different depending upon
the diagnosis and tumor grading. For improving overall prognosis of these
patients with brain tumors, it is important to improve the accuracy of tumor
grading and differential diagnosis. Perfusion sensitive MR imaging with GE-EPI
technique and diffusion weighted MR imaging with SE-EPI as well as proton
spectroscopy have recently been applied clinically in various types of brain
tumors and useful information has been obtained.


This imaging technique has been developed to observe susceptibility effects
after injection of Gd-DTPA, thus obtaining rCBV and rCBF as parameters.
Multiple images are obtained with EPI of 256 x 256 matrices, TE 40 ms, FOV
22~26 cm, slice thickness 5 mm and slice numbers of 5~15. This technique can
visualize early infarctions and the vascularity in the tumors to good advantage.
We used rCBV ratio which is rCBV in the lesion divided by rCBV in the normal
adjacent white matter.

When the degree of vascularity on histology is compared with rCBV ratio, there
is good correlation between the two. In brain tumors, even with no tumor vessels
on conventional angiography, there is often rCBV present according to the
degree of histologic vascularity. In malignant gliomas and metastases, tumor
vascularity can be shown to good advantage; however, no perfusion is shown in
low-grade gliomas or lymphomas. Therefore, when perfusion is performed in
conjunction with Gd enhanced T1-weighted images, various histologic types of
tumors such as malignant gliomas, malignant lymphomas, low-grade gliomas,

metastases and radiation necroses can be differentiated on the basis of
perfusion study.


This imaging technique has been developed to observe diffusion of water
molecules in the abnormal tissues. Although minimal, diffusion of motor mole-
cules produces signal reduction on MR images. This imaging has become
possible with application of strong diffusion gradients. The images reflect
Brownian motion of water molecules within the normal and abnormal tissues of
the brain. From these images apparent diffusion coefficient (ADC) can be

Diffusion weighted images are acquired by using the diffusion weighted multi-
slice spin echo type echo planar sequence with application of motion proving
gradient (MPG) before and after 180 degree pulse. The sequences are
repeated for different values of MPG, producing images with different diffusion

Clinical application of diffusion weighted MR sequences was started with early
detection of acute ischemia, but this technique can be used to demonstrate
restricted motion of water molecules in the abnormal tissues such as brain
tumors. The ADC of gliomas correlates well with tumor cellularity, since highly
cellular gliomas have smaller interstitial space. It has been shown that tumor
cellularity correlates well with the minimum ADC value of gliomas, but not with
signal intensity on T2-weighted images. The minimum ADC of the high grade
gliomas is significantly lower than that of low grade gliomas. It has been found
that diffusion weighted MR imaging is a useful technique for assessing the tumor
cellularity and grading of gliomas.


The principle of magnetic resonance spectroscopy (MRS) is based upon the
chemical shift in which the resonance frequency of nucleus is different depending
upon its chemical environment. Protons have been used for MR spectroscopy

because of their high natural abundance and high nuclear magnetic sensitivity.
MRS is the only non-invasive technique which can measure chemicals or
metabolites within the body. With proton spectroscopy, various metabolites such
as N-acetyl aspatate (NAA), choline (Cho), creatine (Cr), alanine, lactate and
lipids can be measured.

NAA shows the largest peak as a single peak and assigned at chemical shift of
2.0 ppm. NAA reflects normal neurons and their metabolism. The second largest
peak in the normal spectra is Cho which is seen as a single peak that lies to the
left of NAA and is assigned a chemical shift of 3.2 ppm. Cho reflects active
cellular turnover and membrane metabolism. Cr is slightly smaller than NAA in
peak and is localized between Cho and NAA. Cr indicates the amount of cellular
energetics. The Cho and Cr peaks are about half the height of the NAA peak.
Lactate may be produced in the brain and is assigned at 1.32 ppm. Lipids at
0.8, 1.2, 1.5 and 0.6 ppm are also observed in necrosis of malignant tumors as
well as after treatment of tumors. Myoinositol is assigned at 3.56 ppm and
ususally observed in the tumors such as glioblastoma. Alanine is assigned at
1.3-1.4 ppm and has a doublet configuration, being located close to that of

In astrocytoma Cho level is elevated and the NAA level is considerably
decreased. Cr remains unchanged or is slightly decreased. When necrosis is
present, lactate is observed. In abcess, Cho level is decreased and lactate is
increased, while acetate, alanine and cytosolic amino acids are elevated.
Following radiation therapy or chemotherapy of astrocytomas, low levels of NAA,
Cho, and Cr are observed probably due to the presence of necrosis. Proton
MRS of metastasis demonstrates elevated levels of Cho, lactate and lipids with
low or absent NAA. Myoinositol may be decreased. In meningioma, alanine is
characteristically demonstrated in most cases. The alanine/Cr ratio is usually
3~4 times higher than that of the normal brain. In lymphoma, there are elevation
of Cho and lipids and reduction of Cr and NAA. Spectroscopy of
medulloblastoma shows high levels of Cho, reduced Cr and NAA and elevated
lactate. Ependymomas and astrocytomas often show similar findings.

By analizing patterns of metabolites including NAA, Cho, Cr, Alanine, lactate and
lipids, it has been possible to suggest differentiation of astrocytomas,
meningiomas, abscess, lymphoma and other tumors in the brain. Proton MRS is

a non-invasive means of the diagnosis and differential diagnosis of the brain


It has been shown that perfusion sensitive MR imaging with GE-EPI technique
provides useful information on grading of gliomas and differential diagnosis.
Diffusion weighted MR imaging with SE-EPI is also useful for assessing tumor
cellularity and grading of gliomas. Proton spectroscopy gives important
information in differential diagnosis of brain tumors through analysis of
metabolities of the normal brain and tumors.

                          SUGGESTED READING

1.    Sugahara T, Korogi Y, Kochi M, Ikushima I, Hirai T, Okuda T,
      Shigematsu Y, Liang L, Ge Y, Ushio Y, Takahashi M
      Correlation of MR imaging-determined cerebral blood volume maps with
      histologic and angiographic determination of vascularity of gliomas.
      AJR 1998; 171:1479-1486

2.    Sugahara T, Korogi Y, Tomiguchi S, Shigematsu Y, Ikushima I, Kira T,
      Liang L, Ushio Y, Takahashi M
      Post-therapeutic intraaxial brain tumors: Early detection of tumor
      recurrence with perfusion-sensitive contrast-enhanced MR imaging.
      AJNR 1999 (submitted)

3.    Sugahara T, Korogi Y, Shigematsu Y, Hirai T, Ikushima I, Liang L,
      Ushio Y, Takahashi M
      Perfusion-sensitive MRI of cerebral lymphomas: A preliminary report.
      JCAT 1999; 23:232-237

4.   Sugahara T, Korogi Y, Kochi M, Ikushima I, Shigematsu Y, Hirai T,
     Okuda T, Liang L, Ge Y, Komohara Y, Ushio Y, Takahashi M
     Usefulness of diffusion- weighted MRI with echo-planar technique in the
     evaluation of cellularity in gliomas.
     JMRI 1999; 9:53-60

5.   Castillo M, Kwock L, Scatliff J, Mukherji SK
     Proton MR spectroscopy in neoplastic and non-neoplastic
     brain disorders.
     MRI Clin North Am 1998; 6:1-20

6.   Kwock L
     Localized MR spectroscopy: basic principles.
     Neuroimagig Clin North Am 1998; 8:713-732

7.   Preul MC, Caramanos Z, Collins DL, et al.
     Linear discriminant analysis based on proton MR spectroscopic imaging
     of human tumors improves pre-operative diagnosis.
     Nature Med 1996; 2:323-325

8.   Somorjai RL, Dolenko B, Nikulin AK, et al.
     Classification of 1H MR spectra of human brain neoplasms: The influence
     of preprocessing and computerized consensus diagnosis on
     classification accuracy.
     J. Magn Reson Imaging 1996; 6:437-444

9.   Somorjai RL, Nikulin AE, Pizzi N, et al.
     Computerized consensus diagnosis: A classification strategy for
     the robust analysis of MR spectra. I. Application of the 1H spectra
     of thyroid neoplams.
     MRM 1995; 33:257-263

To top