Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van der Graaf, M.
Right arrow Articles by Heerschap, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van der Graaf, M.
Right arrow Articles by Heerschap, A.
(Radiology. 1999;213:919-925.)
© RSNA, 1999

Technical Developments

Human Prostate: Multisection Proton MR Spectroscopic Imaging with a Single Spin-Echo Sequence-Preliminary Experience1

Marinette van der Graaf, PhD, Hendrikus J. van den Boogert, BS, Gerrit J. Jager, MD, PhD, Jelle O. Barentsz, MD, PhD and Arend Heerschap, PhD

1 From the Department of Radiology, Medical Faculty, University Hospital Nijmegen, Geert Grooteplein Zuid 18, 6525 GA Nijmegen, the Netherlands. Received October 12, 1998; revision requested November 13; revision received March 19, 1999; accepted April 29. M.v.d.G. supported by Dutch Cancer Society grant KUN 95-1016. Address reprint requests to M.v.d.G. (e-mail: M.vanderGraaf@rdiag.azn.nl).


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The authors investigated the feasibility of a multisection proton magnetic resonance (MR) spectroscopic imaging technique for the acquisition of metabolic information in the human prostate. Multisection MR spectroscopic imaging was performed of a citrate phantom and of the prostates of eight adult volunteers. High-quality proton MR spectra and citrate metabolite maps of the prostate were obtained with this method.

Index terms: Magnetic resonance (MR), spectroscopy, 844.12145 • Prostate, MR, 844.12141, 844.12145 • Prostate, neoplasms, 844.30


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Prostate cancer is one of the most frequently diagnosed malignancies in Western men. Conventional noninvasive imaging techniques such as transrectal ultrasonography, computed tomography, and magnetic resonance (MR) imaging often cannot reliably differentiate prostate cancer from benign prostatic hyperplasia and normal tissue (1); therefore, other techniques are being evaluated for the identification of prostate cancer. One of these techniques is proton MR spectroscopy, with which it is possible to determine the content of various metabolites in the prostate such as citrate, choline, and creatine. Prostate cancer is characterized by a decreased level of citrate (2–12) and an increased level of (phospho)choline (7,9–11,13). Tumor tissue can be identified by means of a reduced signal ratio for citrate to choline or citrate to choline and creatine (10,11), especially in the peripheral zone of the prostate.

As cancer foci may occur at any location in the prostate, analysis of the entire prostate is necessary to make MR spectroscopy useful as a clinical tool. One of the approaches to achieve this is three-dimensional MR spectroscopic imaging (10,14) with volume preselection by means of point-resolved spatially localized spectroscopy (PRESS) (15,16) or stimulated-echo acquisition mode, or STEAM (17,18). Another possibility is multisection proton MR spectroscopic imaging, which was introduced almost simultaneously by two groups (19,20). They used this multisection method for studies of the human brain. The purpose of our study was to evaluate the feasibility of multisection MR spectroscopic imaging in studies of the human prostate.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Phantom
A 90-mmol/L citrate solution in a glass sphere with an outer diameter of 10 cm was used to test the multisection MR spectroscopic imaging sequence and to study the J-modulation of the citrate multiplet signal. Several salts were added to this phantom, as described previously (11,21), to obtain a citrate solution that corresponds to the average composition of human prostatic fluid (22).

In experiments to test the multisection MR spectroscopic imaging sequence, the phantom was positioned in a container filled with sunflower oil to mimic periprostatic fat. The container was positioned on an endorectal surface coil and surrounded by four bottles filled with water doped with 5 mmol/L NiSO4 and 85 mmol/L NaCl.

Volunteers
This study was approved by our institutional review board. After informed consent was given by eight healthy volunteers (age range, 20–45 years; mean age, 30 years), they underwent multisection MR spectroscopic imaging of the prostate in the supine position. The body phased-array coil was applied around the abdomen with slight compression to reduce respiratory motion. No antiperistaltic drug was used.

MR Spectroscopic Imaging and MR Imaging
All MR examinations were performed with a 1.5-T MR spectrometer (Magnetom Vision; Siemens Medical Systems, Erlangen, Germany). The body radio-frequency coil was used for excitation, and a disposable endorectal coil (MRInnervu; Medrad, Pittsburgh, Pa) was used in combination with the body phased-array coil for signal reception. A special interface with phased-array–compatible, low-noise preamplifier technology enabled integration of the unmodified disposable endorectal coil into the arrangement of the body phased-array coil. MR images were acquired on the basis of a combination of all four signals, but MR spectra were acquired with the signal of only the endorectal coil. To study the J-modulation of the citrate multiplet, MR spectroscopic imaging of the phantom was performed with a regular circular polarized head coil instead of with the coil set-up described previously.

Prior to MR spectroscopic imaging of the prostate, multisection T2-weighted MR imaging was performed in slightly oblique axial, sagittal, and coronal orientations. A turbo spin-echo (SE) sequence was used: repetition time msec/echo time msec of 3,000/132, 5-mm section thickness, 1-mm intersection gap, rectangular field of view of 140 x 280 mm, matrix of 210 x 512, and two signals acquired. In addition, background images of two or three 10-mm-thick sections with orientation and position identical to the MR spectroscopic imaging planes were acquired with the same turbo SE sequence. A noniterative shim procedure (23) was used to optimize magnetic field homogeneity.

Figure 1 shows the general scheme of the multisection MR spectroscopic imaging pulse sequence used in this study, which was based on that developed by Duyn et al (19) for applications in the human brain. Our pulse sequence consists of n times a single SE sequence for section selection and 32 x 32 phase-encoding steps in the directions within each section. Crusher gradients, which dephase all spins that are not properly inverted by the 180° pulse between them, are present around the 180° section-selection pulse in the section-selection direction, Gs, and in the second phase-encoding direction, Gp2. The latter crusher gradients do not appear in the general scheme in Figure 1. As many as six outer volume saturation (OVS) pulses may be applied to suppress signals that originate from the region around the volume of interest. Each OVS pulse is a 90° optimized 2.56-msec sinc pulse, which excites a region selected by an OVS band, followed by crusher gradients in all three directions. To obtain better suppression of water and lipid signals, we replaced the chemical shift-selective (CHESS) water suppression (24) used by Duyn et al (19) with a solvent-suppression technique used by Mescher et al (25) with frequency-selective 180° pulses surrounded by dephasing gradients in opposite directions. The 180° pulses consist of a summation of two 25.6-msec Gaussian pulses with bandwidths of 60 Hz and center frequencies at 0 Hz (water) and -220 Hz (lipid), respectively. Both Gaussian pulse shapes were multiplied by using a Hamming filter. The first two dephasing gradients (10 mT/m) were applied for 1.64 msec each and the last two for 1.5 msec each. The acquisition period for each section consisted of 400 msec during which the second half of the echo was recorded (512 data points).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Schematic depicts the pulse sequence used for multisection MR spectroscopic imaging with a single SE. The basic sequence consists of two section-selective 90° and 180° sinc-shaped pulses (5.12 msec) in the presence of gradients (Gs) followed by data acquisition (Acq). This sequence is repeated n times, which results in measurement of n sections. In the other two directions within each section, 32 x 32 phase-encoding gradients (Gp1 and Gp2) are present. OVS is applied at the start of each SE sequence to suppress signals from regions outside the volume of interest. In addition, two solvent-suppression frequency-selective 180° pulses (180°(sel)) surrounded by two crusher gradients in opposite directions are present before and after the section-selective 180° pulse for the suppression of water and fat signals in the whole plane (see Materials and Methods).

 
For multisectional MR spectroscopic imaging of the prostate, two or three 10-mm-thick parallel axial or oblique sections with a small angle (<15°) to the coronal direction were chosen in such a way that sections perpendicular to the rectal wall were obtained. The intersection gap was 1–3 mm, and field of view was 200–240 mm with nominal voxel sizes of 0.39–0.56 cm3. One image was acquired for each of the 32 x 32 phase-encoding steps after a preparation period of 6 seconds at the beginning of each study. Total measurement time (1,200–1,630/80–100) was 20–28 minutes. OVS bands with a thickness of 35–100 mm were positioned perpendicular to the sections through the prostate to suppress signals that originate from fatty tissues around the prostate.

Comparable studies were performed of the citrate phantom in the container with sunflower oil to mimic the prostate and surrounding fat. In addition, the effectiveness of the OVS bands and the water and fat suppression were tested. MR examinations were performed with and without OVS bands and with and without the solvent-suppression 180° pulses (25). When the solvent-suppression 180° pulses were left out, CHESS water suppression (24) was used.

Because the shape of the AB-type multiplet of citrate strongly depends on pulse sequence and timing (26), we also studied the behavior of this signal as a function of echo time (TE) with both a single SE sequence (90° - 1/2TE - 180° - 1/2TE - Acq) and a double SE PRESS sequence (90° - {tau} - 180° - 1/2TE - 180° - [1/2TE - {tau}] - Acq), where Acq is acquisition. In the single SE experiments, the signal of an axial 15-mm-thick plane was acquired. In the PRESS experiments, the signal of a voxel of 22 x 22 x 22 mm was acquired. CHESS water suppression was used without solvent-suppression pulses and gradients. The CHESS pulse was 25.6 msec long. Citrate spectra (5,000/30–180) were acquired with two preparation scans followed by 32 scans with 1,024 data points (acquisition time, 1,024 msec). The delay {tau} in the PRESS sequence was 7.5 msec. In addition, two scans without water suppression were acquired at each echo time for eddy-current correction (27).

Postprocessing
Postprocessing of the MR spectroscopic data was performed with the system software (LUISE; Siemens Medical Systems). Raw data obtained to study the behavior of the citrate multiplet were processed with use of a protocol consisting of eddy-current correction (27), zero filling to 4,096 data points, Fourier transformation, and minor first-order phase correction. The line widths in the spectra from the single voxel (PRESS studies) were slightly smaller than those in the spectra from the section through the phantom (SE studies). Therefore, to obtain citrate multiplet patterns with comparable line widths, an additional multiplication by using a Lorentz filter with its maximum at 0 msec and half-width at 256 msec was applied to the time-domain data obtained with the PRESS sequence.

Multisection MR spectroscopic imaging raw data obtained in vitro from the citrate phantom or in vivo from the prostate were processed per section. Prior to Fourier transformation in the two spatial directions, a Hamming filter with a filter ratio of 50% and an offset correction were applied. Postprocessing of the time-domain data consisted of zero filling from 512 to 2,048 data points, multiplication by a Gaussian filter with its maximum at 64 msec and half-width at 256 msec, Fourier transformation, and phase correction. Results of these multisection MR spectroscopic imaging studies were presented as spectral maps, metabolite maps, or both. Metabolite maps appear as smoothed images in which the pixel signal intensities are calculated by means of interpolation of the peak volumes derived from the individual spectra of the voxels.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Modulation of the Citrate Multiplet
Modulation of the citrate multiplet as a function of the echo time is different for an SE sequence than for a PRESS sequence (Fig 2). This behavior of the citrate multiplet is in agreement with the theoretic description of the effect of such single or double SE sequences on AB-type multiplets, as presented by Mulkern and Bowers (26). The intensities of the most intense central peaks obtained with a single SE sequence are rather constant at echo times of 70–120 msec. This indicates that an SE sequence allows use of echo times shorter than 120–140 msec, which are often used in a PRESS sequence, without loss of intensity in the central lines of the citrate signal.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Spectra depict modulation of the citrate multiplet as a function of the echo time (TE). Top: Double SE PRESS sequence: 90° - {tau} - 180° - 1/2TE - 180° - [1/2TE - {tau}] - Acq, where Acq is acquisition, with {tau} = 7.5 msec. Bottom: Single SE sequence: 90° - 1/2TE - 180° - 1/2TE - Acq.

 
Another phenomenon visible in Figure 2 is a shift of the modulation pattern of the two outer lines of the citrate multiplet. The shape of these lines obtained with an echo time of X msec in the single SE sequence is comparable to the shape obtained with an echo time of (X + 10) msec in the PRESS sequence.

In Vitro Multisection MR Spectroscopic Imaging of the Citrate Phantom
Figure 3 shows the results of multisection MR spectroscopic imaging of the citrate phantom in an arrangement to mimic the human prostate and surrounding fatty tissues (Fig 3a). Spectra obtained with the multisection MR spectroscopic imaging sequence with both OVS and solvent-suppression water and fat suppression (Fig 3c, part 1) are compared to those obtained with a sequence in which OVS is left out (Fig 3c, part 2) and/or the solvent-suppression water and fat suppression has been replaced by CHESS water suppression (Fig 3c, parts 3 and 4). Figure 3a shows the positions of the OVS bands that cover the container with sunflower oil as much as possible, and Figure 3b illustrates the four voxels (black rectangle) from which the spectra in Figure 3c originated.



View larger version (75K):
[in this window]
[in a new window]
  		Figure 3a. (a) T2-weighted MR image shows the phantom arrangement used to mimic the human prostate. A sphere with citrate solution in a container of sunflower oil is centered on an endorectal surface coil and surrounded by four bottles of doped water within the body phased-array coil. Four OVS bands suppress the lipid signals that originate from the sunflower oil. (b) Enlargement of the central part of the image in a with the MR spectroscopic voxels overlaid. The black rectangle indicates the four voxels from which the spectra in c were obtained. (c) Four proton MR spectra (0.5-6.0 ppm, magnitude mode) derived from the four voxels indicated in b were acquired with various combinations of water and fat suppression methods: 1, solvent-suppression water and fat suppression with OVS; 2, solvent-suppression water and fat suppression without OVS; 3, CHESS water suppression with OVS [H2O = water, Cit = citrate, Lip = lipids]; 4, CHESS water suppression without OVS. The spectra originated from the middle of three MR spectroscopic sections (1,712/100, 32 x 32 matrix, field of view of 240 mm) and were derived from the signal received by the endorectal coil without correction for the coil reception profile.

 


View larger version (113K):
[in this window]
[in a new window]
  		Figure 3b. (a) T2-weighted MR image shows the phantom arrangement used to mimic the human prostate. A sphere with citrate solution in a container of sunflower oil is centered on an endorectal surface coil and surrounded by four bottles of doped water within the body phased-array coil. Four OVS bands suppress the lipid signals that originate from the sunflower oil. (b) Enlargement of the central part of the image in a with the MR spectroscopic voxels overlaid. The black rectangle indicates the four voxels from which the spectra in c were obtained. (c) Four proton MR spectra (0.5-6.0 ppm, magnitude mode) derived from the four voxels indicated in b were acquired with various combinations of water and fat suppression methods: 1, solvent-suppression water and fat suppression with OVS; 2, solvent-suppression water and fat suppression without OVS; 3, CHESS water suppression with OVS [H2O = water, Cit = citrate, Lip = lipids]; 4, CHESS water suppression without OVS. The spectra originated from the middle of three MR spectroscopic sections (1,712/100, 32 x 32 matrix, field of view of 240 mm) and were derived from the signal received by the endorectal coil without correction for the coil reception profile.

 


View larger version (19K):
[in this window]
[in a new window]
  		Figure 3c. (a) T2-weighted MR image shows the phantom arrangement used to mimic the human prostate. A sphere with citrate solution in a container of sunflower oil is centered on an endorectal surface coil and surrounded by four bottles of doped water within the body phased-array coil. Four OVS bands suppress the lipid signals that originate from the sunflower oil. (b) Enlargement of the central part of the image in a with the MR spectroscopic voxels overlaid. The black rectangle indicates the four voxels from which the spectra in c were obtained. (c) Four proton MR spectra (0.5-6.0 ppm, magnitude mode) derived from the four voxels indicated in b were acquired with various combinations of water and fat suppression methods: 1, solvent-suppression water and fat suppression with OVS; 2, solvent-suppression water and fat suppression without OVS; 3, CHESS water suppression with OVS [H2O = water, Cit = citrate, Lip = lipids]; 4, CHESS water suppression without OVS. The spectra originated from the middle of three MR spectroscopic sections (1,712/100, 32 x 32 matrix, field of view of 240 mm) and were derived from the signal received by the endorectal coil without correction for the coil reception profile.

 
With use of solvent-suppression water and fat suppression (Fig 3c, parts 1 and 2), almost no signal from fat or water is present in the upper spectra, which originate from voxels situated in the citrate sphere. In spectral maps of the whole plane (not shown), no spectrum with a lipid signal larger than the citrate signal is observed anywhere inside the citrate sphere, and fewer than 10 voxels show a spectrum with the water signal higher than the citrate signal. The voxels situated inside the container with sunflower oil clearly show a residual fat signal, which is suppressed approximately 90% with use of OVS. In the spectra obtained with CHESS water suppression (Fig 3c, parts 3 and 4), much higher water and fat signals are present. Spectral maps of the whole plane show no spectra with the water signal lower than the citrate signal. When the lipid signals in spectra obtained with the solvent-suppression technique are compared to those obtained with CHESS water suppression, a signal reduction of 90% can also be observed. With use of a combination of OVS and solvent-suppression technique, the total reduction in the fat signal around 1.2–1.3 ppm is about 99%. Citrate signal intensities obtained with use of the four different methods are almost identical.

In Vivo Multisection MR Spectroscopic Imaging of the Human Prostate
The results of multisection MR spectroscopic imaging in the prostates in eight healthy volunteers show that high-quality spectra of large parts of the human prostate can be obtained with this method (Fig 4). Figure 4a and 4b show the orientations of three oblique MR spectroscopic imaging sections through and four OVS planes around the prostate in one of the volunteers. The MR spectroscopic imaging sections and OVS bands are positioned perpendicular and parallel to the rectal wall, respectively. Figure 4c shows three spectral maps with spectral regions from 1.2 to 3.6 ppm, obtained from sections 1–3. The signals in the spectra of the two outer planes (Fig 4c, parts 1 and 3) are generally somewhat lower and broader than those in the central plane (Fig 4c, part 2), which indicates the homogeneity of the magnetic field in these two outer planes is lower than that in the central plane. Also, the results obtained in the other volunteers show some prostate areas with less well-resolved spectra, which is due to the difficulty of obtaining optimal shim values for the whole prostate.



View larger version (125K):
[in this window]
[in a new window]
  		Figure 4a. (a) Sagittal T1-weighted image of the abdomen in a 25-year-old volunteer. 1-3 indicate the positions of three oblique (axial to coronal, 11°) 10-mm-thick sections through the prostate measured at MR spectroscopic imaging. The intersection distance is 1 mm, and 32 individual voxels are shown within the sections. Two of the four OVS bands are indicated. (b) Oblique T2-weighted turbo SE image obtained from section 2 in a. The MR image section thickness is 10 mm, which exactly covers the MR spectroscopic imaging section. The four OVS bands are indicated around the prostate. (c) Spectral maps obtained from sections 1-3 in a were acquired with the multisection MR spectroscopic imaging method (1,565/80, 32 x 32 matrix, field of view of 240 mm, section thickness of 10 mm). All spectra (absorption mode) show the region from 1.2 to 3.6 ppm and are identically scaled, without correction for the reception profile of the endorectal coil. The background images show lower intensity deviations as they are derived from the combined signals received by the endorectal and body phased-array coils. The spectrum from the voxel outlined by the white box in section 2 is displayed in e. (d) Citrate maps show the citrate distribution in sections 1-3. These metabolite maps were derived by integrating the spectra from 2.4 to 2.85 ppm. (e) Proton MR spectrum (1.2-3.6 ppm) from the voxel (0.56 cm3) indicated by the white box in c, part 2. Cho = choline, Cit = citrate, Cre = creatine, PA = polyamines.

 


View larger version (135K):
[in this window]
[in a new window]
  		Figure 4b. (a) Sagittal T1-weighted image of the abdomen in a 25-year-old volunteer. 1-3 indicate the positions of three oblique (axial to coronal, 11°) 10-mm-thick sections through the prostate measured at MR spectroscopic imaging. The intersection distance is 1 mm, and 32 individual voxels are shown within the sections. Two of the four OVS bands are indicated. (b) Oblique T2-weighted turbo SE image obtained from section 2 in a. The MR image section thickness is 10 mm, which exactly covers the MR spectroscopic imaging section. The four OVS bands are indicated around the prostate. (c) Spectral maps obtained from sections 1-3 in a were acquired with the multisection MR spectroscopic imaging method (1,565/80, 32 x 32 matrix, field of view of 240 mm, section thickness of 10 mm). All spectra (absorption mode) show the region from 1.2 to 3.6 ppm and are identically scaled, without correction for the reception profile of the endorectal coil. The background images show lower intensity deviations as they are derived from the combined signals received by the endorectal and body phased-array coils. The spectrum from the voxel outlined by the white box in section 2 is displayed in e. (d) Citrate maps show the citrate distribution in sections 1-3. These metabolite maps were derived by integrating the spectra from 2.4 to 2.85 ppm. (e) Proton MR spectrum (1.2-3.6 ppm) from the voxel (0.56 cm3) indicated by the white box in c, part 2. Cho = choline, Cit = citrate, Cre = creatine, PA = polyamines.

 


View larger version (112K):
[in this window]
[in a new window]
  		Figure 4c. (a) Sagittal T1-weighted image of the abdomen in a 25-year-old volunteer. 1-3 indicate the positions of three oblique (axial to coronal, 11°) 10-mm-thick sections through the prostate measured at MR spectroscopic imaging. The intersection distance is 1 mm, and 32 individual voxels are shown within the sections. Two of the four OVS bands are indicated. (b) Oblique T2-weighted turbo SE image obtained from section 2 in a. The MR image section thickness is 10 mm, which exactly covers the MR spectroscopic imaging section. The four OVS bands are indicated around the prostate. (c) Spectral maps obtained from sections 1-3 in a were acquired with the multisection MR spectroscopic imaging method (1,565/80, 32 x 32 matrix, field of view of 240 mm, section thickness of 10 mm). All spectra (absorption mode) show the region from 1.2 to 3.6 ppm and are identically scaled, without correction for the reception profile of the endorectal coil. The background images show lower intensity deviations as they are derived from the combined signals received by the endorectal and body phased-array coils. The spectrum from the voxel outlined by the white box in section 2 is displayed in e. (d) Citrate maps show the citrate distribution in sections 1-3. These metabolite maps were derived by integrating the spectra from 2.4 to 2.85 ppm. (e) Proton MR spectrum (1.2-3.6 ppm) from the voxel (0.56 cm3) indicated by the white box in c, part 2. Cho = choline, Cit = citrate, Cre = creatine, PA = polyamines.

 


View larger version (71K):
[in this window]
[in a new window]
  		Figure 4d. (a) Sagittal T1-weighted image of the abdomen in a 25-year-old volunteer. 1-3 indicate the positions of three oblique (axial to coronal, 11°) 10-mm-thick sections through the prostate measured at MR spectroscopic imaging. The intersection distance is 1 mm, and 32 individual voxels are shown within the sections. Two of the four OVS bands are indicated. (b) Oblique T2-weighted turbo SE image obtained from section 2 in a. The MR image section thickness is 10 mm, which exactly covers the MR spectroscopic imaging section. The four OVS bands are indicated around the prostate. (c) Spectral maps obtained from sections 1-3 in a were acquired with the multisection MR spectroscopic imaging method (1,565/80, 32 x 32 matrix, field of view of 240 mm, section thickness of 10 mm). All spectra (absorption mode) show the region from 1.2 to 3.6 ppm and are identically scaled, without correction for the reception profile of the endorectal coil. The background images show lower intensity deviations as they are derived from the combined signals received by the endorectal and body phased-array coils. The spectrum from the voxel outlined by the white box in section 2 is displayed in e. (d) Citrate maps show the citrate distribution in sections 1-3. These metabolite maps were derived by integrating the spectra from 2.4 to 2.85 ppm. (e) Proton MR spectrum (1.2-3.6 ppm) from the voxel (0.56 cm3) indicated by the white box in c, part 2. Cho = choline, Cit = citrate, Cre = creatine, PA = polyamines.

 


View larger version (11K):
[in this window]
[in a new window]
  		Figure 4e. (a) Sagittal T1-weighted image of the abdomen in a 25-year-old volunteer. 1-3 indicate the positions of three oblique (axial to coronal, 11°) 10-mm-thick sections through the prostate measured at MR spectroscopic imaging. The intersection distance is 1 mm, and 32 individual voxels are shown within the sections. Two of the four OVS bands are indicated. (b) Oblique T2-weighted turbo SE image obtained from section 2 in a. The MR image section thickness is 10 mm, which exactly covers the MR spectroscopic imaging section. The four OVS bands are indicated around the prostate. (c) Spectral maps obtained from sections 1-3 in a were acquired with the multisection MR spectroscopic imaging method (1,565/80, 32 x 32 matrix, field of view of 240 mm, section thickness of 10 mm). All spectra (absorption mode) show the region from 1.2 to 3.6 ppm and are identically scaled, without correction for the reception profile of the endorectal coil. The background images show lower intensity deviations as they are derived from the combined signals received by the endorectal and body phased-array coils. The spectrum from the voxel outlined by the white box in section 2 is displayed in e. (d) Citrate maps show the citrate distribution in sections 1-3. These metabolite maps were derived by integrating the spectra from 2.4 to 2.85 ppm. (e) Proton MR spectrum (1.2-3.6 ppm) from the voxel (0.56 cm3) indicated by the white box in c, part 2. Cho = choline, Cit = citrate, Cre = creatine, PA = polyamines.

 
Citrate metabolite maps were obtained by integrating the spectral region between 2.4 and 2.85 ppm (Fig 4d). These maps clearly depict the shape of the prostate in the corresponding plane and show a relatively higher citrate content in the peripheral zone than in the central gland, which is in agreement with earlier reports about citrate distribution in the healthy human prostate (10,28–30). However, part of the signal at the ventral side adjacent to the prostate, which was derived from the spectra in the upper row of voxels in Figure 4c, parts 1–3, does not originate from citrate but from periprostatic lipid signals and some baseline distortions.

Figure 4e shows an enlargement of the spectrum that originated from the voxel indicated by the white box in Figure 4c, part 2. Although the shape of the citrate multiplet indicates high field homogeneity, the resonances of choline (3.2 ppm) and creatine (3.0 ppm) are not resolved. This phenomenon is observed more often and may be caused by signal intensity around 3.1 ppm, which originates from polyamines (31). The polyamine spermine is present in relatively high concentrations of 10–20 mmol/L in healthy prostatic fluid (22,32).


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Results in this study show that MR spectroscopic imaging data sets from multiple sections through the human prostate can be obtained simultaneously with relatively high spatial resolution (0.4–0.6 cm3). The first reports of multisection MR spectroscopic imaging were studies of the human brain (19,20). To our knowledge, no other publications have appeared about application of this technique to another human organ. When application of multisection MR spectroscopic imaging to the brain is compared with application to the prostate, some differences appear. The most important difference is that a section through the brain (particularly an axial section) contains mainly signals of the region of interest, but a section through the prostate contains many signals that originate from surrounding tissues. A limited field of view can be chosen for study of the brain as no signal is present outside the skull, and the lipid signals that originate from the fatty tissue just below the skull can be suppressed quite easily by means of OVS (19). A relatively large field of view has to be chosen for study of the prostate to avoid aliasing problems, however, and it is more difficult to suppress all the signals from tissues outside the prostate with use of OVS. However, the limited reception profile of the endorectal coil positioned near the prostate reduces aliasing problems.

Because suppression of the signals that originate from tissues outside the prostate with use of OVS is not sufficient to get rid of all lipid signals in the region of interest, it is important to combine OVS with the frequency-selective solvent-suppression technique (25). This suppression method is used to minimize not only the fat signal but also the water signal. One of the greater advantages of solvent-suppression water suppression compared with CHESS water suppression is that optimization of the strength of the solvent-suppression 180° pulses is not necessary, whereas the CHESS pulse has to be optimized to obtain good water suppression. In the present study, we evaluated the effectiveness of the OVS and solvent-suppression techniques with a phantom. Lipid signal suppression by means of OVS is approximately 90%, which is in agreement with data reported by Duyn et al (19) for the skull area. In addition, the amount of suppression of the lipid signal obtained with use of the solvent-suppression technique with Gaussian frequency-selective pulses is also about 90% in the present study. The solvent-suppression technique may be further improved by increasing the strength of the crusher gradients and replacing the Gaussian 180° pulses with Shinnar–le Roux 180° pulses (33–35), as described by Star-Lack et al (36).

Because frequency-selective pulses are used in the solvent-suppression technique, it is important that no or only limited movement is present and that the magnetic field is homogeneous over the whole volume studied. These conditions can be fulfilled more easily in the human brain than in the prostate. First, the prostate may be subject to displacement as a result of respiratory motion and peristaltic movements. Second, it is difficult to obtain a good homogeneous magnetic field over the complete prostate owing to susceptibility differences between prostatic tissue; periprostatic lipid tissue; the bladder, which contains urine with various salts; and the balloon of the endorectal coil filled with air in the rectum. The results of this preliminary study show plane-to-plane differences in signal intensity and spectral resolution within individual prostates, which is probably caused by differences in magnetic field homogeneity. Since these differences must be reduced as much as possible in a clinical setting, future studies should be performed to test if the shimming can be improved (eg, by including higher order shims). In addition, the line shapes of the metabolite resonances may be corrected during postprocessing of the data.

As only a limited acquisition time per plane is possible in the multisection MR spectroscopic imaging method, free induction decays (or echoes) of metabolite signals with long T2 values may be truncated. The maximum T2 value of proton spins for the main metabolites observed in MR spectra of the prostate (ie, 227 msec for choline [30]) is shorter than that for metabolites in the brain observed at MR spectroscopic imaging (ie, 512 msec for N-acetylaspartate in white matter [37]). For that reason, the acquisition window per section used for the prostate can be shorter than that used for the brain.

The multisection MR spectroscopic imaging method may be used as an alternative to three-dimensional MR spectroscopic imaging with PRESS or stimulated-echo acquisition mode, or STEAM, preselection. On the one hand, use of three-dimensional MR spectroscopic imaging with a preselection method has the advantage that mainly signals of the preselected volume of interest are obtained, which allows use of a smaller field of view. In addition, only the magnetic field of the volume of interest has to be optimized by means of shimming. On the other hand, preselection of a volume of interest requires section selection in three dimensions, which results in section profile distortions and chemical shift artifacts in all three dimensions. With the multisection MR spectroscopic imaging method, section selection is applied in only one dimension by means of two section-selective pulses, which results in better section definition and no chemical shift artifacts in the two dimensions in the plane in which phase encoding is used. Furthermore, the signals from one plane cannot leak into the adjacent planes as a result of the data point spread function being less than ideal for localization (38). Another advantage of the multisection approach is that multiple sections are acquired within one repetition time, which increases the signal-to-noise ratio per unit time. With the three-dimensional approach, however, only one free induction decay is acquired per repetition time. Finally, the multisection sequence is based on an SE sequence that requires only two pulses, whereas the three-dimensional MR spectroscopic imaging method with volume preselection consists of three pulses. Use of an additional pulse may result in signal loss due to the nonideality of the pulse in some areas of the volume acquired and to differences in J-modulation of the citrate multiplet. Kurhanewicz et al (10) and Kaji et al (39) successfully used three-dimensional MR spectroscopic imaging with PRESS preselection and a matrix of 8 x 8 x 8 or 16 x 8 x 8 to identify the presence of prostate cancer in the peripheral zone of the prostate. The echo time of 130 msec that they used with PRESS preselection is in agreement with the relative optimum for citrate signal obtained with the PRESS sequence in our study (Fig 2). In a single SE multisection MR spectroscopic imaging study, shorter echo times can be chosen without loss of intensity of the citrate signal and with less attenuation of the other metabolite signals as a result of less T2 relaxation.

For identification of prostate tumors, the multisection MR spectroscopic imaging technique may be combined with a fast dynamic gadolinium-enhanced multisection MR imaging technique (40). One 10-mm-thick section studied at MR spectroscopic imaging can cover two 5-mm-thick MR images. A large part of the prostate can be covered with use of three 10-mm-thick MR spectroscopic imaging sections (particularly when the sections are chosen with a more or less coronal orientation); measurement time for three sections is less than 28 minutes with the present experimental conditions. In large prostates, use of more sections may be necessary, which would require longer measurement times. In these cases, the total measurement time can be reduced by using shorter acquisition times or a partial k-space sampling scheme that limits the total number of phase-encoding steps.

In conclusion, the results of this study show that it is possible to successfully apply multisection MR spectroscopic imaging to the human prostate. The shape of the entire prostate can be easily visualized with this technique, which is important in studies of prostate cancer. MR spectroscopic images can be compared directly to MR images acquired with the same oblique orientation.


   Acknowledgments
 
We thank Siemens Medical Systems, Erlangen, Germany, and Medrad, Pittsburgh, Pa, for their support in development of the special endorectal coil interface. We also thank the former for providing us with a basic multisection MR spectroscopic sequence and work-in-progress postprocessing software. We thank Oliver Heid, PhD, of Siemens Medical Systems for the modifications he made in his noniterative shim procedure to enable calculation of the shim currents on the basis of a voxel subset. Finally, we thank Wim M.M.J. Bovée, PhD, (Delft University, the Netherlands) for stimulating discussions and critical reading of the manuscript.


   Footnotes
 
Abbreviations: CHESS = chemical shift selective OVS = outer volume saturation PRESS = point-resolved spatially localized spectroscopy SE = spin echo

Author contributions: Guarantors of integrity of entire study, M.v.d.G., A.H.; study concepts, M.v.d.G., A.H.; study design, M.v.d.G., H.J.v.d.B., A.H.; definition of intellectual content, M.v.d.G., A.H.; literature research, M.v.d.G.; experimental studies, all authors; data acquisition, all authors; data analysis, M.v.d.G.; manuscript preparation, M.v.d.G.; manuscript editing and review, all authors.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Jager GJ, Barentsz JO, Ruijter ETG, de la Rosette JJMCH, Oosterhof GON. Primary staging of prostate cancer. Eur Radiol 1996; 6:134-139.[Medline]
  2. Halliday KR, Fenoglio-Preiser C, Sillerud LO. Differentiation of human tumors from nonmalignant tissue by natural-abundance 13C NMR spectroscopy. Magn Reson Med 1988; 7:384-411.[Medline]
  3. Costello LC, Franklin RB. Concepts of citrate production and secretion by prostate. I. Metabolic relationships. Prostate 1991; 18:25-46.[Medline]
  4. Fowler AH, Pappas AA, Holder JC, et al. Differentiation of human prostate cancer from benign hypertrophy by in vitro 1H NMR. Magn Reson Med 1992; 25:140-147.[Medline]
  5. Narayan P, Kurhanewicz J. Magnetic resonance spectroscopy in prostate disease: diagnostic possibilities and future developments. Prostate Suppl 1992; 4:43-50.[Medline]
  6. Schiebler ML, Miyamoto KK, White M, Maygarden SJ, Mohler JL. In vitro high resolution 1H-spectroscopy of the human prostate: benign prostatic hyperplasia, normal peripheral zone and adenocarcinoma. Magn Reson Med 1993; 29:285-291.[Medline]
  7. Cornel EB, Smits GA, Oosterhof GO, et al. Characterization of human prostate cancer, benign prostatic hyperplasia and normal prostate by in vitro 1H and 31P magnetic resonance spectroscopy. J Urol 1993; 150:2019-2024.[Medline]
  8. Kurhanewicz J, Dahiya R, Macdonald JM, Chang LH, James TL, Narayan P. Citrate alterations in primary and metastatic human prostatic adenocarcinomas: 1H magnetic resonance spectroscopy and biochemical study. Magn Reson Med 1993; 29:149-157.[Medline]
  9. Kurhanewicz J, Vigneron DB, Nelson SJ, et al. Citrate as an in vivo marker to discriminate prostate cancer from benign prostatic hyperplasia and normal prostate peripheral zone: detection via localized proton spectroscopy. Urology 1995; 45:459-466.[Medline]
  10. Kurhanewicz J, Vigneron DB, Hricak H, Narayan P, Carroll P, Nelson SJ. Three-dimensional H-1 MR spectroscopic imaging of the in situ human prostate with high (0.24–0.7-cm3) spatial resolution. Radiology 1996; 198:795-805.[Abstract/Free Full Text]
  11. Heerschap A, Jager GJ, van der Graaf M, et al. In vivo proton MR spectroscopy reveals altered metabolite content in malignant prostate tissue. Anticancer Res 1997; 17:1455-1460.[Medline]
  12. Liney GP, Turnbull LW, Lowry M, Turnbull LS, Knowles AJ, Horsman A. In vivo quantification of citrate concentration and water T-2 relaxation time of the pathologic prostate gland using H-1 MRS and MRI. Magn Reson Imaging 1997; 15:1177-1186.[Medline]
  13. Heerschap A, Jager G, de Koster A, et al. 1H MRS of prostate pathology (abstr) In: Proceedings of the 9th Annual Meeting of the Society of Magnetic Resonance in Medicine. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1993; 213.
  14. Kurhanewicz J, Vigneron DB, Hricak H, et al. Prostate cancer: metabolic response to cryosurgery as detected with 3D H-1 MR spectroscopic imaging. Radiology 1996; 200:489-496.[Abstract/Free Full Text]
  15. Ordidge RJ, Bendall MR, Gordon RE, Connelly A. Volume selection for in-vivo biological spectroscopy. In: Govil G, Khetrapal CL, Saran A, eds. Magnetic resonance in biology and medicine. New Delhi, India: Tata McGraw-Hill, 1985; 387-397.
  16. Bottomley PA. Spatial localization in NMR-spectroscopy in vivo. Ann NY Acad Sci 1987; 508:333-348.[Abstract]
  17. Granot J. Selected volume excitation using stimulated echoes (VEST): applications to spatially localised spectroscopy and imaging. J Magn Reson 1986; 70:488-492.
  18. Frahm J, Merboldt KD, Hanicke W. Localized proton spectroscopy using stimulated echoes. J Magn Reson 1987; 72:502-508.
  19. Duyn JH, Gillen J, Sobering G, van Zijl PC, Moonen CT. Multisection proton MR spectroscopic imaging of the brain. Radiology 1993; 188:277-282.[Abstract/Free Full Text]
  20. Spielman DM, Pauly JM, Macovski A, Glover GH, Enzmann DR. Lipid-suppressed single- and multisection proton spectroscopic imaging of the human brain. JMRI 1992; 2:253-262.
  21. van der Graaf M, Heerschap A. Effect of cation binding on the proton chemical shifts and the spin-spin coupling constant of citrate. J Magn Reson 1996; 112(series B):58-62.
  22. Kavanagh JP. Sodium, potassium, calcium, magnesium, zinc, citrate and chloride content of human prostatic and seminal fluid. J Reprod Fertil 1985; 75:35-41.[Abstract]
  23. Heid OH. Noniterative localized in vivo shimming in <15s (abstr) In: Proceedings of the Fourth Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1996; 363.
  24. Haase A, Frahm J, Hanicke W, Matthaei D. 1H NMR chemical shift selective (CHESS) imaging. Phys Med Biol 1985; 30:341-344.[Medline]
  25. Mescher M, Tannus A, Johnson MO, Garwood M. Solvent suppression using selective echo dephasing. J Magn Reson 1996; 123(series A):226-229.
  26. Mulkern R, Bowers J. Density matrix calculations of AB spectra from multipulse sequences: quantum mechanics meets in vivo spectroscopy. Concepts Magn Reson 1994; 6:1-23.
  27. Klose U. In vivo proton spectroscopy in presence of eddy currents. Magn Reson Med 1990; 14:26-30.[Medline]
  28. Lowry M, Manton DJ, Turnbull LW, Blackband SJ, Horsman A. Quantitative proton MRS assessment of citrate heterogeneity in normal prostate. MAGMA 1994; 2:483-485.
  29. Lowry M, Liney GP, Turnbull LW, Manton DJ, Blackband SJ, Horsman A. Quantification of citrate concentration in the prostate by proton magnetic resonance spectroscopy: zonal and age-related differences. Magn Reson Med 1996; 36:352-358.[Medline]
  30. Heerschap A, Jager GJ, van der Graaf M, Barentsz JO, Ruijs SHJ. Proton MR spectroscopy of the normal human prostate with an endorectal coil and a double spin-echo pulse sequence. Magn Reson Med 1997; 37:204-213.[Medline]
  31. van der Graaf M, Schipper RG, Oosterhof GON, Schalken JA, Verhofstad AAJ, Heerschap A. 1H MRS of prostatic tissue focused on the detection of spermine, a possible biomarker of malignant behavior in prostate cancer (abstr) In: Proceedings of the Sixth Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1998; 613.
  32. Smith RC, Litwin MS, Lu Y, Zetter BR. Identification of an endogenous inhibitor of prostatic carcinoma cell growth. Nat Med 1995; 1:1040-1045.[Medline]
  33. Shinnar M, Bolinger L, Leigh JS. The use of finite impulse response filters in pulse design. Magn Reson Med 1989; 12:81-87.[Medline]
  34. Le Roux P. Exact synthesis of radio frequency waveforms (abstr) In: Proceedings of the Seventh Annual Meeting of the Society for Magnetic Resonance in Medicine. Berkeley, Calif: Society for Magnetic Resonance in Medicine, 1988; 1049.
  35. Pauly J, Le Roux P, Nishimura D, Macovski A. Parameter relations for the Shinnar–Le Roux RF design algorithm. IEEE Trans Med Imaging 1991; 10:53-65.
  36. Star-Lack J, Nelson SJ, Kurhanewicz J, Huang LR, Vigneron DB. Improved water and lipid suppression for 3D PRESS CSI using RF band selective inversion with gradient dephasing (BASING). Magn Reson Med 1997; 38:311-321.[Medline]
  37. 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; 30:424-437.[Medline]
  38. Mareci TH, Brooker HR. Essential considerations for spectral localization using indirect gradient encoding. J Magn Reson 1991; 92:229-246.
  39. Kaji Y, Kurhanewicz J, Hricak H, et al. Localizing prostate cancer in the presence of postbiopsy changes on MR images: role of proton MR spectroscopic imaging. Radiology 1998; 206:785-790.[Abstract/Free Full Text]
  40. Jager GJ, Barentsz JO, Ruijter ET, Oosterhof GO, Ruijs SH. MR imaging in prostate cancer: current role, future applications (abstr). Radiology 1997; 205(P):577.[Free Full Text]



This article has been cited by other articles:


Home page
 RadiologyHome page
C. K. Kuhl, F. Traber, J. Gieseke, W. Drahanowsky, N. Morakkabati-Spitz, W. Willinek, M. von Falkenhausen, C. Manka, and H. H. Schild
Whole-Body High-Field-Strength (3.0-T) MR Imaging in Clinical Practice * Part II. Technical Considerations and Clinical Applications
Radiology, April 1, 2008; 247(1): 16 - 35.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
H. J. Huisman, J. J. Futterer, E. N. J. T. van Lin, A. Welmers, T. W. J. Scheenen, J. A. van Dalen, A. G. Visser, J. A. Witjes, and J. O. Barentsz
Prostate Cancer: Precision of Integrating Functional MR Imaging with Radiation Therapy Treatment by Using Fiducial Gold Markers
Radiology, July 1, 2005; 236(1): 311 - 317.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
J. R. Thornbury, D. K. Ornstein, P. L. Choyke, C. P. Langlotz, and J. C. Weinreb
Prostate Cancer: What Is the Future Role for Imaging?
Am. J. Roentgenol., January 1, 2001; 176(1): 17 - 22.
[Full Text]

Home page
 RadiologyHome page
F. Feldman
Musculoskeletal Radiology: Then and Now
Radiology, August 1, 2000; 216(2): 309 - 316.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van der Graaf, M.
Right arrow Articles by Heerschap, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van der Graaf, M.
Right arrow Articles by Heerschap, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE