| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Charité, Universitätsmedizin Berlin,
Schumannstraße 20/21, Berlin 10117, Germany.
2 Department of Urology, Charité, Universitätsmedizin Berlin, Berlin
10117, Germany.
3 Institute of Pathology, Charité, Universitätsmedizin Berlin,
Berlin 10117, Germany.
4 Department of Radiation Medicine, Charité, Universitätsmedizin
Berlin, Berlin 10117, Germany.
Received October 11, 2004;
accepted after revision December 6, 2004.
Address correspondence to D. Beyersdorff
(dirk.beyersdorff{at}charite.de).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Twenty-four patients with prostate cancer underwent MRI at 1.5 T using the combined endorectal–body phased-array coil and at 3.0 T using the torso phased-array coil, among them 22 before undergoing radical prostatectomy. The prostate was imaged with T2-weighted sequences in axial and coronal orientations at both field strengths and, in addition, with an axial T1-weighted sequence at 1.5 T. Preoperative analysis of all MR images taken together was compared with the histologic findings to determine the accuracy of MRI for the local staging of prostate cancer. In a retroanalysis, the image quality, tumor delineation, and conspicuity of staging criteria were determined separately for both field strengths and compared. Statistical analysis was performed using Wilcoxon's and the McNemar tests.
RESULTS. In the preoperative analysis, MRI (at both 1.5 and 3.0 T) had an accuracy of 73% for the local staging of prostate cancer. The retroanalysis yielded significantly better results for 1.5-T MRI with the endorectal–body phased-array coil in terms of image quality (p < 0.001) and tumor delineation (p = 0.012) than for 3.0-T MRI with the torso phased-array coil. Analysis of the individual staging criteria for extracapsular disease did not reveal a superiority of either of the two field strengths in the depiction of any of the criteria.
CONCLUSION. Intraindividual comparison shows that image quality and delineation of prostate cancer at 1.5 T with the use of an endorectal coil in a pelvic phased-array is superior to the higher field strength of 3.0 T with a torso phased-array coil alone. As long as no endorectal coil is available for 3-T imaging, imaging at 1.5 T using the combined endorectal–body phased-array coil will continue to be the gold standard for prostate imaging.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The clinical role of 3.0-T MRI scanners is currently being investigated. In principle, this new generation of scanners has a nearly two-fold SNR compared with conventional scanners with a field strength of 1.5 T, but the SNR does not increase linearly with the field strength. Only the torso phased-array coil is currently available for 3.0-T MRI of the prostate, and no endorectal coil has as yet been approved for clinical use at this field strength. In view of this situation, two questions arose in the course of the study: Can prostate MRI at 3.0 T match imaging at 1.5 T with a combined coil in terms of image quality, tumor delineation, and evaluation of the staging criteria, and is it still advantageous to use an endorectal coil? Is imaging at 3 T without an endorectal coil superior to imaging with the endorectal–body phased-array coil at 1.5 T, and does this imply that the endorectal coil can be abandoned at higher field strengths?
Our study was performed prospectively with an additional retroanalysis. The purpose of the prospective portion was to determine the accuracy of local prostate cancer staging using all MR images acquired at 1.5 and 3.0 T and to correlate the findings with histology. In the second part of the study, imaging at 1.5 and 3.0 T was compared by separately analyzing the sets of images acquired at both field strengths for image quality, delineation of prostate cancer, and the depiction of staging criteria for extracapsular extension and infiltration of adjacent organs.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The prostate-specific antigen (PSA) levels ranged from 2 to 14 ng/mL (mean, 7.5 ng/mL). Twenty-two patients underwent radical prostatectomy within 1–34 days (mean, 10 days) of the MRI examination. In one patient with iliac lymph node metastases proven by laparoscopic pelvic lymphadenectomy, the operation was discontinued without performing prostatectomy. In another patient, the first prostate biopsy performed externally yielded the histologic borderline diagnosis of well-differentiated adenocarcinoma. This patient underwent repeated endorectal sonographically guided biopsy with removal of 10 cores that did not contain any tumor tissue. No operation was performed, and the patient was followed up clinically with regular control of PSA levels. Histology of the surgical specimens in the 22 of 24 study patients operated on yielded stage T2 cancer in 17 patients and T3 cancer in five.
MRI Protocol
The first MRI examination was performed on a 3.0-T scanner (Signa 3T, GE
Healthcare) using a torso phased-array coil (USA Instruments). The sequence
parameters used had been optimized for the coils used in prior prostate
examinations. After a localizer scan, at least a T2-weighted angulated axial
fast spin-echo sequence with a TR/TE of 4,500/102 and an echo-train length of
8 or 16 and an angulated coronal fast spin-echo sequence (3,800/78.3;
echo-train length, 8) with a field of view of 16 x 16 cm and a matrix of
256 x 256 were acquired. Four acquisitions were performed with each
sequence at a slice thickness of 4 mm and a gap of 1 mm.
After giving informed consent, all patients underwent a second MRI examination of the prostate on a 1.5-T MRI scanner (Magnetom Vision or Magnetom Symphony, Siemens Medical Solutions) using a combination of an endorectal coil (Medrad) and a body phased-array coil. The prostate was imaged with a T2-weighted turbo spin-echo sequence in angulated axial (3,500/96; echo-train length, 13; 3 acquisitions) and coronal (4,522/112; echo-train length, 13; 2 acquisitions) slice orientations with a field of view of 16 x 16 cm. The matrix was 256 x 256. In addition, an angulated axial T1-weighted spin-echo sequence (680/14; echo-train length, 3) was acquired. The field of view was 16 x 16 cm and the matrix was 256 x 256. The slice thickness was 3 mm and the interslice gap, 0.9 mm.
Peristalsis was suppressed before imaging was begun by the IV administration of 40 mL of butyl scopolamine (Buscopan, Boehringer) in 21 patients and 1 mg of glucagon in two patients. One patient with contraindications to both drugs was examined without suppression of peristalsis.
Histology
Twenty-two of the 24 patients included in the study were treated with
radical prostatectomy. The surgical specimen was analyzed in 15–29
blocks, depending on the size of the prostate. This procedure ensured a
reproducible localization of abnormal changes in the prostate. For correlation
with the MRI findings, the histologic specimens were analyzed in terms of
tumor localization; extracapsular tumor extension; and infiltration of seminal
vesicles, urinary bladder, and rectum. One patient with a negative repeat
biopsy and removal of 10 cores did not undergo prostatectomy and was
classified in the statistical analysis as having no tumor and no extracapsular
tumor extension. In one patient with proven lymphadenopathy, the planned
prostatectomy was cancelled. Hence, extracapsular penetration could not be
evaluated in this patient, who therefore was not included in the statistical
analysis of the criteria for extracapsular disease.
Prospective Analysis
All MR images obtained at both field strengths were analyzed together for
preoperative staging in terms of tumor localization, extracapsular extension,
and infiltration of adjacent organs. This analysis was done as part of
clinical routine. A simplified version of the TNM staging system was used: T1
= microscopic tumor, T2 = tumor confined to prostate, T3 = extracapsular tumor
extension or infiltration of seminal vesicles, and T4 = infiltration of
adjacent organs or pelvic floor muscles. The criteria for extracapsular tumor
extension were contiguity of the tumor with the capsule over an extended area
with smooth bulging of the capsule, asymmetry of the neurovascular bundle,
displacement of the rectoprostatic angle, and continuous infiltration of
periprostatic tissue by the tumor. Infiltration of the seminal vesicles was
assumed when an intravesicular loss of signal or continuous tumor extension
was seen on T2-weighted images. A lymph node was classified as metastatic when
its minimal axial diameter was 10 mm or more.
Retroanalysis
The retroanalysis was done to compare the MR images obtained at the two
field strengths. To this end, two sets of images per patient were compiled,
those obtained with the combined coil at 1.5 T and those obtained with the
torso phased-array coil at 3 T. The two sets were analyzed separately and in
random order. (The reviewers were not blinded to the field strength because
the rectal coil was visible on the MR images acquired at 1.5 T.) The images
were interpreted by two radiologists in consensus and without knowledge of the
clinical data, the MRI findings obtained with the other field strength, or the
results of histology.
Overall image quality was subjectively classified as excellent, 1; good, 2; satisfactory, 3; moderate, 4; or poor, 5. Criteria were motion artifacts, coil artifacts, correct placement of the rectal coil, and sharpness and overall subjective impression. The quality score assigned to the MRI examination could not be higher than 4 when the coil was twisted by more than 30°.
The presence of artifacts was assessed separately, assigning the following scores: no artifacts, 1; few artifacts, 2; moderate artifacts, 3; considerable artifacts, 4; and pronounced artifacts resulting in an image that could not be analyzed satisfactorily, 5. The delineation of the tumor and the prostate capsule and zonal anatomy were likewise assessed on a 5-point scale ranging from excellent (1) to poor (5).
Periprostatic infiltration and invasion of adjacent organs were evaluated using the following criteria: smooth bulging of the capsule, irregular bulging of the capsule, contiguity of the tumor with the capsule over an extended area, infiltration of periprostatic fatty tissue, displacement of the rectoprostatic angle, and asymmetry of the neurovascular bundle. The presence of each of these criteria was scored as definitely not present, 1; highly unlikely, 2; somewhat likely, 3; probably present, 4; and definitely present, 5. The same 5-point scale was used to assess the likelihood of infiltration of the seminal vesicles, infiltration of the urinary bladder, and infiltration of the rectum.
|
|
|
|
The MRI findings and histologic results were compared for extracapsular extension by means of cross tables grouping the variables as negative results (score of 1 or 2) or positive results (score of 4 or 5), with a score of 3 being defined as inconclusive. Next, the sensitivity and specificity of each of the two field strengths were calculated for each criterion for 23 patients (22 operated on and one patient without tumor visualization at the repeat biopsy), and the results obtained with both field strengths were compared. The sensitivity and specificity of both field strengths were compared with the McNemar test and assuming significance at a p value of less than 0.05. In addition, the kappa coefficient was calculated as measure of random-corrected agreement.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
In the retroanalysis, the images of all 24 patients were used for comparing the two field strengths in terms of image quality. The images obtained at 1.5 T were assigned significantly better scores than those obtained at 3.0 T for overall image quality (p < 0.001) and the criteria of delineation of the prostate capsule (p = 0.007), depiction of the zonal anatomy (p = 0.003), and delineation of the tumor (p = 0.012). Artifacts were present at both field strengths, but there were fewer artifacts at 1.5 T caused by the endorectal coil than there were motion artifacts and technical artifacts at 3 T (p = 0.002).
In the comparison of the images obtained with the two coil set-ups and at
the different field strengths, six criteria of extracapsular tumor
infiltration were analyzed. For the images obtained at 1.5 T, the sensitivity
was highest for contiguity with capsule, followed by smooth bulging, irregular
bulging, direct periprostatic infiltration, and asymmetry of the neurovascular
bundle, with displacement of the rectoprostatic angle having the lowest
sensitivity. In terms of specificity, the reverse order was found for these
criteria. At 3.0-T MRI, the two criteria of asymmetry of the neurovascular
bundle and displacement of the rectoprostatic angle had a higher sensitivity
and lower specificity than at 1.5 T. At 1.5 T, smooth bulging and irregular
bulging were found to have a higher sensitivity but lower specificity. The
criteria of contiguity with capsule and direct extracapsular extension had the
same sensitivity at both field strengths, whereas one of each had a higher
specificity at either field strength (Table
2). In the McNemar test, none of the staging criteria investigated
was found to have a statistically significant advantage at either of the two
field strengths. No correlation was found for displacement of the
rectoprostatic angle, direct periprostatic infiltration, and extensive
contiguity with the capsule (
<0.2), whereas a low to moderate
correlation was found for smooth bulging, irregular bulging, and asymmetry of
the neurovascular bundle (, 0.25–0.69).
|
None of the images obtained at either field strength showed signs of infiltration of the urinary bladder and rectum in any of the patients. Histologically, no infiltration of these two organs was seen in any of the patients. No suspicious lymph node enlargement was seen at 1.5 or 3.0 T. The metastatic lymph node found in one patient contained a metastasis measuring 4 mm in diameter.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The introduction of new coil systems such as the endorectal coil, the pelvic phased-array coil, and the combined endorectal–body phased-array coil improved the spatial resolution and SNR of prostate MRI. The studies performed so far have shown the endorectal coil to be superior to the conventional body coil in staging prostate cancer [7, 9]. Further studies found the combined endorectal–body phased-array coil yielded slightly better results than the body phased-array coil alone, but the difference was not significant [10, 11]. The development of the coil technique was paralleled by an increase in the routinely used field strength from an initial 0.3 T [2] to the present standard of 1.5 T [4, 7, 10, 19]. Recently, some centers have been using MRI scanners with a field strength of 3.0 T or more [20, 21]. The aim of our study was to determine in a direct comparison whether the currently established technique of prostate MRI at 1.5 T using the combined endorectal–body phased-array coil could be improved by imaging with the torso phased-array coil alone on a 3.0-T MRI scanner. Endorectal coils approved for clinical use at 3.0 T unfortunately were not available at the time of the study.
The evaluation of the criteria for extracapsular tumor extension and infiltration of adjacent organs depends on the examiner's experience and skills [22, 23]. The two radiologists who performed the prospective analysis and the retroanalysis in our study had at least 5 years of experience with endorectal MRI. To reduce a bias that might have resulted from individual image interpretation approaches, the images were analyzed by consensus. Moreover, the different staging criteria were scored separately to eliminate an effect resulting from evaluating them together [22, 24].
|
|
|
|
|
Detection of microscopic extracapsular tumor invasion would require a much higher spatial resolution of MRI than the in-plane resolution of 0.6 mm used here (probably four- to fivefold), whereas the SNR is still adequate. Intraindividual comparison in our study showed no improvement of the SNR at the higher field strength compared with the combined coil set-up at 1.5 T. Because a higher field strength is expected in principle to improve the SNR, the fact that this was not the case in our study is attributable to the different coils used. This observation is supported by studies performed at 1.5 T showing that addition of the endorectal coil improves the image quality compared with MRI using the body phased-array coil alone [10, 11].
There were fewer and less-pronounced artifacts caused by the endorectal coil at 1.5 T than there were at 3.0 T. The artifacts occurring at the higher field strength were mostly caused by motion or blurring, most likely resulting from the high echo-train length used. Comparison of the different coil systems at 1.5 T showed similar results with respect to the occurrence of artifacts [10, 11]. In fact, use of the endorectal coil blocks the position of the rectum by inflating the attached balloon with about 80–100 mL air. Thus, motion artifacts are less likely. Compression of the rectoprostatic angles or artifacts caused by the coil material itself was not found to be a problem.
All patients included in the study underwent prostate biopsy at least 2 weeks before the MRI examination. Some patients showed more or less pronounced signal increases on the T1-weighted images, indicating hemorrhage. Such hemorrhage-related signal changes are more conspicuous on T2-weighted images and cannot be distinguished from the signal changes produced by prostate cancer [26]. This is a problem that occurs at both field strengths to the same extent.
Several studies investigating T1-weighted imaging with the IV administration of gadopentetate dimeglumine did not show a definitive advantage for contrast-enhanced MRI over unenhanced imaging [7, 27, 28]. Therefore, we did not include contrast-enhanced prostate MRI in the present study.
The coil systems used in our study enabled evaluation of locoregional lymph nodes at both field strengths. As expected, no differences between the two coil systems and two field strengths were seen. The micrometastasis present in one patient was missed on MRI at both field strengths. The diagnostic dilemma posed by lymph node metastases that do not enlarge the lymph node may possibly be solved by the use of specific contrast media [29].
A limitation of our study is the small patient population who underwent prostate imaging at both field strengths. This results from the fact that the strain associated with a second MRI examination would not be tolerated by all patients.
The results of our study suggest that the improvement in the SNR resulting from the higher field strength of 3.0 T does not reach that achieved with the endorectal coil at 1.5 T. Although our study did not reveal any statistically significant differences in the visualization of staging criteria between the two field strengths investigated, 1.5-T MRI of the prostate with the endorectal coil will continue to be the gold standard for MRI of the prostate because of its superior overall image quality compared with MRI at 3.0 T using only the torso phased-array coil. However, initial preclinical results obtained with prototypes of endorectal coils at 3.0 T suggest that the combination of higher field strength and endorectal coil leads to more improvements in SNR than achieved with either technique alone [30].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. W. T. P. J. Heijmink, J. J. Futterer, T. Hambrock, S. Takahashi, T. W. J. Scheenen, H. J. Huisman, C. A. Hulsbergen-Van de Kaa, B. C. Knipscheer, L. A. L. M. Kiemeney, J. A. Witjes, et al. Prostate Cancer: Body-Array versus Endorectal Coil MR Imaging at 3 T--Comparison of Image Quality, Localization, and Staging Performance Radiology, July 1, 2007; 244(1): 184 - 195. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
