DOI:10.2214/AJR.05.0873
AJR 2006; 187:W582-W588
© American Roentgen Ray Society
Comparison of Lymphotropic Nanoparticle-Enhanced MRI Sequences in Patients with Various Primary Cancers
Mansi Saksena1,
Mukesh Harisinghani1,
Peter Hahn1,
John Kim1,
Anuradha Saokar1,
Benjamin King1 and
Ralph Weissleder1
1 All authors: Division of Abdominal Imaging, Massachusetts General Hospital and
Harvard Medical School, White 270, 55 Fruit Street, Boston, MA 02114.
Received May 23, 2005;
accepted after revision November 17, 2005.
Address correspondence to M. Harisinghani
(mharisinghani{at}partners.org).
WEB
This is a Web exclusive article.
Abstract
OBJECTIVE. This study was performed to empirically evaluate
T2-weighted fast spin-echo, moderately T2*-weighted
gradient-refocused echo (GRE), and heavily T2*-weighted GRE
sequences to determine which sequence is the most effective for nodal
characterization on lymphotropic nanoparticle-enhanced MRI (LNMRI).
MATERIALS AND METHODS. The study included 65 patients who had proven
primary cancer and were scheduled for either surgical lymph node dissection or
imaging-guided lymph node biopsy. All patients underwent LNMRI using
T2-weighted fast spin-echo, moderately T2*-weighted GRE, and
heavily T2*-weighted GRE sequences. Unequivocal correlation of
histopathology and MRI could be made in 140 nodes and only these were included
in the analysis. Two blinded reviewers performed qualitative analysis of the
nodes. Alternative free-response receiver operating characteristic (ROC)
curves with a continuous rating scale were plotted for each sequence for both
reviewers and the diagnostic accuracy of fast spin-echo T2-weighted and GRE
T2*-weighted images were compared by calculating the area under the
curve (Az). A two-tailed Student's t test was
performed to test the significance (p < 0.05) of the differences
between the ROC curves derived from the three sequences.
RESULTS. Irrespective of reviewer experience,
T2*-weighted sequences showed better nodal characterization when
compared with T2-weighted sequences. For both reviewers, there was a
statistically significant difference between the Az for
T2- and the two T2*-weighted sequences (p < 0.05).
Neither reviewer showed a statistically significant difference between the two
T2*-weighted sequences.
CONCLUSION. GRE T2*-weighted sequences are superior for
nodal characterization on LNMRI to fast spin-echo T2-weighted sequences.
Imaging protocols for LNMRI should include fast spin-echo T2-weighted imaging
for anatomic localization, but characterization of nodes should be based on
their appearance on contrast-enhanced T2*-weighted images. The
T2*-weighted images acquired with dual TE values, one of which is
intermediate and the other longer, improve nodal characterization.
Keywords: cancer • lymph nodes • MR contrast agents • MRI
Introduction
Accurate pretreatment staging of an oncologic patient helps in both
selecting appropriate therapy and assessing prognosis. Cross-sectional imaging
is routinely used, not only to evaluate the primary tumor (T stage) and detect
regional or distant metastatic disease (M stage), but also for nodal
evaluation (N stage). Over the past decade, CT and MRI have been the mainstays
of nodal staging. However, the size criteria used to detect nodal metastatic
disease with these techniques are both nonspecific and insensitive
[1-4].
The inaccuracy of size criteria for nodal characterization has led to the
development of newer techniques that evaluate nodal function.
Lymphotrophic nanoparticle-enhanced MRI (LNMRI) using ferumoxtran-10
(Combidex, Advanced Magnetics; Sinerem, Guerbet) is a particularly promising
technique for nodal evaluation in the setting of malignancy.
This technique is highly accurate for nodal staging in patients with
various primary cancers
[5-15].
It evaluates nodal macrophage function and does not rely on nodal size to
detect metastatic disease. It is postulated that macrophages present in normal
benign nodes take up the contrast agent causing a drop in nodal signal
intensity on contrast-enhanced imaging. Benign nodes show homogeneous
ferumoxtran-10 uptake and signal intensity drop on contrast-enhanced
T2-weighted fast spin-echo and T2*-weighted gradient-refocused echo
(GRE T2*) images, whereas a node replaced by malignant cells shows
no change in signal intensity after ferumoxtran-10 administration
[16]. Therefore, imaging with
ferumoxtran-10 typically involves 2D axial T1-weighted gradient-echo, 2D axial
T2-weighted fast spin-echo, and 2D axial T2*-weighted
gradient-refocused echo sequences, the ideal imaging parameters for which have
been described [17].
To accurately characterize nodes, the optimal sequence for LNMRI should
have a good contrast-to-noise ratio (CNR). T2-weighted images have a good
signal-to-noise ratio (SNR) but are not very sensitive to the change in
intranodal susceptibility caused by ferumoxtran-10. The T2*
sequence is a heavily T2-weighted gradient-echo sequence with additional
sensitivity for the susceptibility changes induced by the intranodal
ferumoxtran-10 in normal nodes
[18]. T2*-weighted
images have good CNR ratio but lower SNR. With increasing TE, T2*
weighting becomes progressively stronger, with better CNR but lower SNR. The
T1-weighted images are used for anatomic localization of nodes and for the
detection of a fatty hilum, whereas nodal characterization depends on its
signal intensity on the contrast-enhanced T2- and T2*-weighted
images.
Multiple clinical studies have evaluated the technique using the above
mentioned sequences
[5-15].
Harisinghani et al. [9]
evaluated 80 patients with prostate cancer and reported a sensitivity of 100%
with a specificity of 95.7% in characterizing pelvic lymph nodes
[9]. Anzai et al.
[7], reporting on the overall
phase III multicenter trial in evaluating various primary cancers, reported a
sensitivity, specificity, and accuracy of 83%, 77%, and 80%, respectively,
with paired unenhanced and contrast-enhanced MRI. Tabatabaei et al.
[15] reported a sensitivity of
100% and a specificity of 97% on seven patients with squamous cell carcinoma
of the pelvis. All studies, however, evaluated primary efficacy parameters of
sensitivity, specificity, and accuracy for nodal characterization on LNMRI
without attention being paid to the accuracy of T2 and T2*
sequences individually. This study empirically evaluates T2-weighted fast
spin-echo, moderately T2*-weighted GRE, and heavily
T2*-weighted GRE sequences to determine which is the most effective
sequence for nodal characterization on LNMRI.
Materials and Methods
Patients
The study included 65 patients (51 men, 14 women; mean age, 60.5 years;
range, 28-80 years) with proven primary cancer (bladder, 15; breast, 10;
renal, 1; penile, 4; prostate, 27; rectal, 1; testicular, 5; ureteral, 2) who
were scheduled for either surgical lymph node dissection or imaging-guided
lymph node biopsy. Patient inclusion criteria were age older than 18 years, a
confirmed primary cancer, and suspected nodal metastases on staging CT scans.
The study was approved by the institutional review board. Patients provided
written informed consent for the examinations performed.
Ferumoxtran-10 Administration
Ferumoxtran-10 was provided as a lyophilized powder consisting of
ultrasmall superparamagnetic iron oxide nanoparticles covered with
low-molecular-weight dextran. The contrast material was reconstituted using 10
mL of normal saline after which a weight-adjusted dose (2.6 mg of iron per kg
of body weight) was withdrawn and diluted with 50 mL of saline and infused
through a 5-µm filter at a rate of 4 mL/min. For a 70-kg patient the
injected volume of reconstituted ferumoxtran-10 was 9.1 mL (0.13 mL per kg of
body weight). Total volume of diluted contrast material injected was 59.1 mL.
None of the patients suffered any severe adverse reactions to the agent,
although seven patients reported back pain during the administration of
contrast material. This resolved when administration was temporarily withheld
with no recurrence on resuming administration.
MRI
MRI was performed at 1.5 T (System 9X, GE Healthcare) with region-specific
phased-array coils. Identical sequences were performed before and 24-36 hours
after ferumoxtran-10 administration. The pulse sequences performed included
T2-weighted fast spin-echo (TR/TE, 4,000-4,500/80; flip angle, 90°; field
of view, 24-28 cm; slice thickness, 3 mm; matrix, 256 x 256; number of
excitations, 3; echo-train length, 8-12; number of slices, 25-32; average
acquisition time, 4.2 minutes), a dual TE T2*-weighted
gradient-echo (2,100/12.2, 21; flip angle, 70°; field of view, 26-28 cm;
slice thickness, 3 mm; matrix, 128 x 256; number of excitations, 2;
average acquisition time, 9 minutes), and a T1-weighted gradient-echo sequence
obtained in axial and coronal planes (175/1.8; flip angle, 80°; field of
view, 22-30 cm; slice thickness, 4 mm; matrix, 128 x 256; number of
excitations, 1; average acquisition time, 22 seconds). Care was taken to
ensure that all relevant locoregional nodal drainage areas were included in
the imaging. All sequences were performed in the standard axial plane, because
this allowed a consistent inclusion of all local and regional lymph nodes; for
T1-weighted gradient-echo sequences, additional coronal and oblique images
were also obtained to localize the nodes in relationship to vascular anatomic
landmarks. The above listed imaging sequences and parameters were optimized to
reduce motion artifacts; maximize SNR; and provide diagnostically useful
images of the pelvis, abdomen, and chest within clinically acceptable time
limits. In particular, when scanning the abdomen and lower chest, respiratory
triggering was used for the T2-weighted fast spin-echo sequences, respiratory
gating was used for the T2*-weighted gradient-echo sequences, and
breath-holding was used for the T1-weighted gradient-echo sequences. To
further reduce respiratory artifacts from anterior abdominal wall motion,
anterior saturation bands were placed on the subcutaneous fat.
Pathologic Correlation
Histopathologic correlation was obtained by using surgical resection in 20
(31%) patients, imaging-guided biopsy in 41 (63%) patients, and sentinel node
biopsy in four (6%) patients.
The average time interval between imaging and pathologic analysis was 8.3
days (range, 0-34 days). Surgically resected nodes were placed on a grid
identifying their location and orientation and sent for histopathologic
analysis. All nodal tissue was routinely processed and embedded in paraffin.
Surgically resected nodes were sliced into multiple parallel slices of 2-3 mm
and stained with H and E. The slides were reviewed by a pathologist who had no
knowledge of the MRI findings. The histopathologic results for each lymph node
were catalogued for subsequent comparison with MRI findings. The nodes that
underwent imaging-guided biopsy were correlated at biopsy by two radiologists.
Unequivocal correlation of histopathology and MRI could be made in 140 nodes
and only these were included in the analysis.
Interpretation of MR Images
Two radiologists who were blinded to the histopathologic diagnosis and not
involved in patient enrollment or the primary patient evaluation performed
qualitative analysis of the nodes. Reviewer 1 had recently completed a
fellowship in abdominal imaging that included occasional discussion of
ferumoxtran-10-enhanced MRI. Reviewer 2 had 6 years of experience interpreting
ferumoxtran-10-enhanced MR images of lymph nodes. All images were viewed and
analyzed on an Advantage Windows workstation (version 4.0, GE Healthcare).
Nodes that had undergone pathologic evaluation were marked with arrows and
provided to the reviewers. The reviewers did not evaluate the entire study.
The images were reviewed in three sessions, with an 8-week interval between
each session. To eliminate learning bias, the cases were presented randomly in
each session.
In the first session, the reviewers evaluated combined unenhanced and
contrast-enhanced fast spin-echo T2-weighted images followed 8 weeks later by
the unenhanced and contrast-enhanced GRE T2*-weighted images
obtained with a TE of 12.2 msec. In the third session, 8 weeks after the
second session, the reviewers evaluated the combined unenhanced and
contrast-enhanced GRE T2*-weighted images obtained with a higher TE
of 21 msec. The reviewers used established diagnostic guidelines for
subjective nodal characterization
[7]
(Fig. 1). At each session, the
reviewers assigned a confidence rating to each node based on a 6-point scale
as follows: 1, definitely benign; 2, most likely benign; 3, probably benign;
4, probably malignant; 5, most likely malignant; and 6, definitely
malignant.
Statistical Analysis
Statistical analysis was performed using Med-Calc for Windows, version
7.4.1.0 (MedCalc Software). Alternative free-response receiver operating
characteristic (ROC) curves with a continuous rating scale were plotted for
each sequence for both reviewers on a per-nodal basis. The diagnostic accuracy
of fast spin-echo T2 and GRE T2* images was compared by calculating
the area under the curve (Az). A two-tailed Student's
t test was performed to test the significance of the differences
among the ROC curves derived from the three sequences (p <
0.05).
Because the reviewers used a 6-point scale for nodal characterization,
primary efficacy parameters of sensitivity, specificity, and accuracy were
also calculated on a node-to-node basis for T2- and T2*-weighted
sequences (scale points 1-3 were designated as benign and 4-6 were designated
as malignant). Interobserver agreement was assessed using a weighted kappa
statistic, with a kappa value equal to or less than 0.2 indicating slight
agreement; 0.21-0.40 indicating fair agreement; 0.41-0.60 indicating moderate
agreement; 0.61-0.80 indicating substantial agreement; and 0.81-1.00
indicating almost perfect agreement.
Results
On imaging, 140 nodes were seen that were unequivocally correlated with
histopathology. The various primary tumors were bladder (38 nodes), breast (19
nodes), renal (1 node), penile (10 nodes), prostate (58 nodes), rectal (2
nodes), testicular (8 nodes) and ureteral (4 nodes). Twenty-five patients
contributed one node, 16 patients contributed two nodes and 24 patients
contributed three or more nodes. Of the 140 nodes, 89 (63.5%) were evaluated
by surgical dissection, 47 (33.5%) by imaging-guided biopsy, and four (2.8%)
by sentinel node biopsy. Of the evaluated lymph nodes (mean size, 10.5 mm), 46
(32.8%) were benign and 94 (67.2%) were malignant on histopathologic analysis;
seven nodes showed fibrotic changes on pathologic analysis.
ROC analysis evaluating the performances of the two reviewers was
performed. Table 1 shows the
Az values for each reviewer. The Az
value for reviewer 1 was highest for T2*-weighted sequences
obtained with a higher TE of 21 (95% confidence interval [CI], 0.85
[0.78-0.90]). The Az value for reviewer 2 was highest for
T2*-weighted sequences obtained with a lower TE of 12.6 (95% CI,
0.94 [0.89-0.97]). For both reviewers there was a statistically significant
difference between the Az for T2- and the two
T2*-weighted sequences (p < 0.05). Neither reviewer
showed a statistically significant difference between the two
T2*-weighted sequences. These results indicate that irrespective of
reviewer experience, T2*-weighted sequences show better nodal
characterization than do T2-weighted sequences (Figs.
2A,
2B,
2C,
2D,
2E,
2F,
3A,
3B,
3C,
3D,
3E, and
3F).
View this table:
[in this window]
[in a new window]
|
TABLE 1: Area Under the Curve for Receiver Operating Characteristic Curves
Obtained for Each Sequence for the Two Reviewers
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —85-year-old woman with bladder cancer. Unenhanced
T2*-weighted MRI image shows enlarged left external iliac lymph
node (arrow) with more prominent loss in signal intensity than on
T2-weighted images. This is characteristic of benign lymph node and diagnosis
was confirmed by CT-guided biopsy.
|
|
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —85-year-old woman with bladder cancer. Unenhanced heavily
T2*-weighted MRI image shows enlarged left external iliac lymph
node (arrow) with more prominent loss in signal intensity than on
T2-weighted images. This is characteristic of benign lymph node and diagnosis
was confirmed by CT-guided biopsy.
|
|
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2E —85-year-old woman with bladder cancer. Contrast-enhanced
T2*-weighted MRI image shows enlarged left external iliac lymph
node (arrow) with more prominent loss in signal intensity than on
T2-weighted images. This is characteristic of benign lymph node and diagnosis
was confirmed by CT-guided biopsy.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2F —85-year-old woman with bladder cancer. Contrast-enhanced
heavily T2*-weighted MRI image shows enlarged left external iliac
lymph node (arrow) with more prominent loss in signal intensity than
on T2-weighted images. This is characteristic of benign lymph node and
diagnosis was confirmed by CT-guided biopsy.
|
|
View larger version (90K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —61-year-old man with prostate cancer. Unenhanced heavily
T2*-weighted image shows hyperintense 7-mm left external iliac
lymph node (arrow). Node is benign by size criteria. This node was
sampled at surgery and found to be completely replaced by metastatic prostate
cancer.
|
|
View larger version (80K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D —61-year-old man with prostate cancer. Contrast-enhanced
T2-weighted image of 7-mm left external iliac lymph node (arrow)
shows no drop in signal intensity. Such an appearance on contrast-enhanced
lymphotropic nanoparticle-enhanced MRI is diagnostic of metastatic node.
|
|
View larger version (90K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3E —61-year-old man with prostate cancer. Contrast-enhanced
T2*-weighted image of 7-mm left external iliac lymph node
(arrow) shows no drop in signal intensity. Such an appearance on
contrast-enhanced lymphotropic nanoparticle-enhanced MRI is diagnostic of
metastatic node.
|
|
View larger version (87K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3F —61-year-old man with prostate cancer. Contrast-enhanced
heavily T2*-weighted image of 7-mm left external iliac lymph node
(arrow) shows no drop in signal intensity. Such an appearance on
contrast-enhanced lymphotropic nanoparticle-enhanced MRI is diagnostic of
metastatic node. This node was sampled at surgery and found to be completely
replaced by metastatic prostate cancer.
|
|
Table 2 summarizes the
sensitivity and specificity values. The images with a TE of 21 msec showed
higher specificity but low sensitivity, whereas the images with a TE of 12.2
msec showed higher sensitivity but decreased specificity.
The weighted kappa value was 0.754. This is interpreted as good agreement
within the reviewers (weighted kappa value = 0.61-0.80; strength of agreement
between reviewers, good.)
Discussion
Ferumoxtran-10 is an ultrasmall superparamagnetic iron oxide nanoparticle
characterized by a large magnetic moment and a high dipolar relaxivity when
compared with conventional paramagnetic agents, which gives it a strong
T2* effect [19].
Consequently, the GRE T2*-weighted sequence, which is a heavily
T2-weighted sequence, is very sensitive to the susceptibility changes induced
by the intranodal ferumoxtran-10 and allows for detection of small quantities
of intranodal ferumoxtran-10. When administered IV, these nanoparticles are
internalized by nodal macrophages, resulting in a decrease in signal intensity
of healthy lymph nodes on contrast-enhanced T2*-weighted images
[9] (Figs.
3A,
3B,
3C,
3D,
3E, and
3F). Therefore, intuitively,
one would expect better nodal characterization on T2*-weighted
images.
Although an overview of optimized imaging parameters for LNMRI was recently
published [17], no published
studies compare the performance of various sequences for nodal
characterization. This article compares T2 fast spin-echo and GRE
T2* sequences to determine which sequence provides better lymph
node characterization on LNMRI. In addition, a comparison of GRE
T2* sequences obtained with two different TE values evaluates the
optimal TE for nodal characterization.
Our results show that T2*-weighted images have a higher accuracy
in contrast-enhanced nodal characterization than do T2-weighted images
(Table 1). Independent of
reviewer experience, there is a statistically significant difference in nodal
characterization of the two sequences with T2*-weighted sequences,
demonstrating improved detection of metastatic nodal disease. This may be
attributable to the fact that the T2*-weighted images are more
sensitive to the susceptibility changes caused by intranodal ferumoxtran-10,
allowing for the detection of small amounts of ferumoxtran-10 taken up by the
healthy regions of partially infiltrated nodes. This makes smaller metastatic
deposits easier to identify. However, T2* images did have a number
of false-positives, particularly on the sequence with the shorter TE. The
T2* sequence with the longer TE had fewer false-positives and,
hence, a higher specificity. All of the nodes that were seen as false-positive
on T2*-weighted images showed intranodal fibrosis on
histopathologic examination (Figs.
4A and
4B). We postulate that
fibrosis replaces healthy nodal cells, including intranodal macrophages,
preventing ferumoxtran-10 uptake and mimicking nodal infiltration by
metastatic tumor.
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —59-year-old man with penile cancer. Unenhanced heavily
T2*-weighted (A) and contrast-enhanced heavily
T2*-weighted (B) images obtained with TE of 21 msec show two
left inguinal lymph nodes (arrows, A). Contrast-enhanced image
(B) shows a drop in signal intensity (arrows, B)
compared with unenhanced image (A). This was interpreted as malignant
nodal infiltration. Patient underwent inguinal lymphadenectomy, which revealed
nodal fibrosis without any evidence of malignancy.
|
|
View larger version (120K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —59-year-old man with penile cancer. Unenhanced heavily
T2*-weighted (A) and contrast-enhanced heavily
T2*-weighted (B) images obtained with TE of 21 msec show two
left inguinal lymph nodes (arrows, A). Contrast-enhanced image
(B) shows a drop in signal intensity (arrows, B)
compared with unenhanced image (A). This was interpreted as malignant
nodal infiltration. Patient underwent inguinal lymphadenectomy, which revealed
nodal fibrosis without any evidence of malignancy.
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —58-year-old man patient with prostate cancer. Reprinted with
permission from [17].
Contrast-enhanced T2* image obtained with shorter TE shows 4-mm
left external iliac node with central heterogeneity (arrow), a
finding that can be interpreted as representing malignant infiltration.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —58-year-old man patient with prostate cancer. Reprinted with
permission from [17].
Contrast-enhanced heavily T2*-weighted image obtained at same time
as A but with longer TE shows more homogeneous drop in signal intensity
(arrow), which is consistent with benign node. This node was benign
on pathologic analysis.
|
|
Although T2*-weighted sequences are sensitive to the
susceptibility effects of ferumoxtran-10, the parameters used to acquire
images influence the signal drop induced by intranodal accumulation of
ferumoxtran-10 [17]. The TE
must be long enough to adequately decrease the signal in a benign lymph node
after it takes up the contrast. Our purpose in comparing the two TE values of
12.2 and 21 msec was to identify the optimal TE for ferumoxtran-10-enhanced
T2* imaging. In our study, the experienced reviewer had a higher
Az value on the more heavily T2*-weighted
sequence, whereas the other reviewer had a higher Az value
on T2*-weighted sequences obtained with a lower TE. Neither
reviewer, however, had a statistically significant difference between the
Az values for the two sequences. On comparing primary
efficacy parameters, we saw that the shorter TE was more sensitive, whereas
the longer TE was more specific. This may be because the longer TE detected
even small concentrations of ferumoxtran-10 within nodes, hence identifying
more true-negative nodes, but also had more false-negatives because metastases
were missed in small nodes. This could be a result of blooming artifact caused
by ferumoxtran-10 in the healthy nodal regions obscuring small metastases.
However, the images with a TE of 12.2 msec had higher false-positives,
indicating that in some benign nodes, the intranodal ferumoxtran-10 was not
seen (Figs. 5A and
5B). Consequently, we
recommend using dual-echo T2* sequences with nodal evaluation being
based on assessment of images obtained with both TE values. This has the added
advantage of allowing numeric estimation of the T2* value, which
has the potential for semiautomated nodal cancer staging by imaging
[20]. However, because of the
drop in SNR on the higher TE value, it may not be possible to make anatomic
distinctions, regardless of contrast. This may be the limiting factor to how
much the TE can be increased.
The main limitation of our study was that the reviewers did not undertake
nodal detection. Nodes that were unequivocally correlated with histology were
marked out and shown to the reviewers. As mentioned before, T2*
sequences have poor SNR. This may pose a problem in nodal detection,
especially in anatomic locations where there may be significant artifact, such
as close to bowel. Therefore, although GRE T2* sequences have the
highest accuracy in characterizing nodes on LNMRI, fast spin-echo T2-weighted
images may have an added value in detection and anatomic localization of
nodes, which this study did not assess. Additionally, although learning bias
was minimized by a time interval of 8 weeks between interpretation of various
sequences, reviewers were presented with all images of one sequence first,
followed by images from the other sequence, which may have introduced some
bias at interpretation.
In conclusion, our results show that the GRE T2*-weighted
sequences are superior for nodal characterization on LNMRI to fast spin-echo
T2-weighted sequences. Imaging protocols for LNMRI should include fast
spin-echo T2-weighted imaging for anatomic localization, but characterization
of nodes should be based on their appearance on contrast-enhanced
T2*-weighted images. The T2*-weighted images acquired
with dual TE values, one of which is intermediate and the other longer,
improve nodal characterization.
References
- Jager GJ, Barentsz JO, Oosterhof GO, Witjes JA, Ruijs SJ. Pelvic
adenopathy in prostatic and urinary bladder carcinoma: MR imaging with a
three-dimensional TI-weighted magnetization-prepared-rapid gradient-echo
sequence. AJR 1996;167
: 1503-1507[Abstract/Free Full Text]
- Bipat S, Glas AS, van der Velden J, Zwinderman AH, Bossuyt PM,
Stoker J. Computed tomography and magnetic resonance imaging in staging of
uterine cervical carcinoma: a systematic review. Gynecol
Oncol 2003; 91:59
-66[CrossRef][Medline]
- Tiguert R, Gheiler EL, Tefilli MV, et al. Lymph node size does not
correlate with the presence of prostate cancer metastasis.
Urology 1999; 53:367
-371[CrossRef][Medline]
- Borley N, Fabrin K, Sriprasad S, et al. Laparoscopic pelvic lymph
node dissection allows significantly more accurate staging in
"high-risk" prostate cancer compared to MRI or CT.
Scand J Urol Nephrol 2003;37
: 382-386[CrossRef][Medline]
- Anzai Y, Blackwell KE, Hirschowitz SL, et al. Initial clinical
experience with dextran-coated superparamagnetic iron oxide for detection of
lymph node metastases in patients with head and neck cancer.
Radiology 1994;192
: 709-715[Abstract/Free Full Text]
- Anzai Y, McLachlan S, Morris M, Saxton R, Lufkin RB. Dextran-coated
superparamagnetic iron oxide, an MR contrast agent for assessing lymph nodes
in the head and neck. Am J Neuroradiol1994; 15:87
-94[Abstract]
- Anzai Y, Piccoli CW, Outwater EK, et al. Evaluation of neck and
body metastases to nodes with ferumoxtran 10-enhanced MR imaging: phase III
safety and efficacy study. Radiology2003; 228:777
-788[Abstract/Free Full Text]
- Bellin MF, Roy C, Kinkel K, et al. Lymph node metastases: safety
and effectiveness of MR imaging with ultrasmall superparamagnetic iron oxide
par ticles—initial clinical experience.
Radiology 1998;207
: 799-808[Abstract/Free Full Text]
- Harisinghani MG, Barentsz J, Hahn PF, et al. Noninvasive detection
of clinically occult lymph-node metastases in prostate cancer. N
Engl J Med 2003; 348:2491
-2499[Abstract/Free Full Text]
- Harisinghani MG, Barentsz JO, Hahn PF, et al. MR lymphangiography
for detection of minimal nodal disease in patients with prostate cancer.
Acad Radiol 2002;9
[suppl 2]:S312
-S313
- Mack MG, Balzer JO, Straub R, Eichler K, Vogl TJ. Superparamagnetic
iron oxide-enhanced MR imaging of head and neck lymph nodes.
Radiology 2002;222
: 239-244[Abstract/Free Full Text]
- Michel SC, Keller TM, Frohlich JM, et al. Preoperative breast
cancer staging: MR imaging of the axilla with ultrasmall superparamagnetic
iron oxide enhancement. Radiology 2002;225
: 527-536[Abstract/Free Full Text]
- Nguyen BC, Stanford W, Thompson BH, et al. Multicenter clinical
trial of ultrasmall superparamagnetic iron oxide in the evaluation of
mediastinal lymph nodes in patients with primary lung carcinoma. J
Magn Reson Imaging 1999; 10:468
-473[CrossRef][Medline]
- Rockall AG, Sohaib SA, Harisinghani MG, et al. Diagnostic
performance of nanoparticle-enhanced magnetic resonance imaging in the
diagnosis of lymph node metastases in patients with endometrial and cervical
cancer. J Clin Oncol 2005;23
: 2813-2821[Abstract/Free Full Text]
- Tabatabaei S, Harisinghani M, McDougal WS. Regional lymph node
staging using lymphotropic nanoparticle enhanced magnetic resonance imaging
with ferumoxtran-10 in patients with penile cancer. J
Urol 2005; 174:923
-927; discussion 927[CrossRef][Medline]
- Weissleder R, Elizondo G, Wittenberg J, Lee AS, Josephson L, Brady
TJ. Ultrasmall superparamagnetic iron oxide: an intravenous contrast agent for
assessing lymph nodes with MR imaging. Radiology1990; 175:494
-498[Abstract/Free Full Text]
- Harisinghani MG, Dixon WT, Saksena MA, et al. MR lymphangiography:
imaging strategies to optimize the imaging of lymph nodes with ferumoxtran-10.
RadioGraphics 2004;24
: 867-878[Abstract/Free Full Text]
- Stets C, Brandt S, Wallis F, Buchmann J, Gilbert FJ,
Heywang-Kobrunner SH. Axillary lymph node metastases: a statistical analysis
of various parameters in MRI with USPIO. J Magn Reson
Imaging 2002; 16:60
-68[CrossRef][Medline]
- Koenig SH, Kellar KE. Theory of 1/T1 and 1/T2 NMRD profiles of
solutions of magnetic nanoparticles. Magn Reson Med1995; 34:227
-233[Medline]
- Harisinghani MG, Weissleder R. Sensitive, noninvasive detection of
lymph node metastases. PLoS Med 2004;1
: e66[CrossRef][Medline]
- Harisinghani MG, Saksena MA, Hahn PF, King B, Kim J, Torabi MT,
Weissleder R. Ferumoxtran-10-enhanced MR lymphangiography: does
contrast-enhanced imaging alone suffice for accurate lymph node
characterization? AJR 2006;186
: 144-148[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
