DOI:10.2214/AJR.05.1918
AJR 2007; 188:409-414
© American Roentgen Ray Society
High-b Value Diffusion-Weighted MRI for Detecting Pancreatic Adenocarcinoma: Preliminary Results
Tomoaki Ichikawa1,
Sukru Mehmet Erturk2,
Utarou Motosugi1,
Hironobu Sou1,
Hiroshi Iino3,
Tsutomu Araki1 and
Hideki Fujii3
1 Department of Radiology, University of Yamanashi, Shimokato, Japan.
2 Department of Radiology, Sisli Etfal Hospital, No. 10/1 Dogancilar, Uskudar
Istanbul 81160, Turkey.
3 Department of First Surgery, University of Yamanashi. Shimokato, Japan.
Received October 29, 2005;
accepted after revision February 28, 2006.
Address correspondence to S. M. Erturk
(mehmeterturk{at}superonline.com).
Abstract
OBJECTIVE. The objective of our study was to evaluate the usefulness
of high-b value diffusion-weighted MRI (DWI) in the detection of pancreatic
adenocarcinoma.
SUBJECTS AND METHODS. Twenty-six patients with pancreatic
adenocarcinoma were included in the study. Twenty-three other patients who
were being followed up due to pancreatic diseases other than adenocarcinoma
were included as control subjects. All patients and subjects underwent DWI,
and the images were evaluated by three blinded radiologists.
RESULTS. Receiver operating characteristic (ROC) curve analysis
yielded Az values (i.e., area under the ROC curve) of
0.998, 0.998, and 0.995 for the three radiologists. The mean sensitivity and
specificity for the detection of pancreatic adenocarcinoma were 96.2% and
98.6%, respectively. The kappa values indicating interobserver agreement
between different pairs of radiologists were in the category of excellent.
CONCLUSION. High-b value DWI allows the detection of pancreatic
adenocarcinoma with a high sensitivity and specificity.
Keywords: diffusion-weighted MRI • intraductal papillary mucinous tumor • MRI • oncologic imaging • pancreas • pancreatic adenocarcinoma
Introduction
Pancreatic cancer is one of the most lethal human cancers and
continues to be a major unsolved health problem at the start of the 21st
century [1]. It is well known
that the 5-year survival rate of patients with pancreatic adenocarcinoma is
dismal, being less than 10%; at the initial diagnosis, fewer than 10-15% of
patients can undergo surgical resection, which is the only potential curative
treatment [2]. Although much
effort has been devoted to increase the sensitivity for detecting earlystage
pancreatic adenocarcinomas with conventional imaging techniques, such as
sonography, CT, or MRI, the sensitivity to detect pancreatic cancer is still
insufficient [1].
High-b value diffusion-weighted MRI (DWI) is different from morphologically
oriented imaging techniques in that it can sensitively depict
disease-associated changes of random translational molecular motion, known as
diffusion or brownian water motion
[3]. The authors of several
studies have reported that quantification of diffusion as apparent diffusion
coefficient (ADC) values is useful in tissue characterization
[4,
5]. However, mainly because of
the insufficient signal-to-noise ratio (SNR) of the images, the standard DWI
applications are not qualitative but are quantitative, being based on complex
ADC calculations. Therefore, this technique is still not available for use as
a detection tool for abdominal malignancies. However, recently, it became
possible to obtain high-b value DW images with an improved SNR
[6]. More recently, Ichikawa et
al. [7] reported that direct
visual assessment of high-b value DW images has a high sensitivity and
specificity in the detection of colorectal cancer.
The purpose of this study was to evaluate DWI as a diagnostic tool for
depicting pancreatic adenocarcinomas using direct visual assessment of the
obtained images.
Subjects and Methods
Patients
During a 12-month period between June 2003 and May 2004, 26 patients (12
women, 14 men; mean age, 62 ± 7.8 years) who had a diagnosis of
pancreatic adenocarcinoma based on various diagnostic techniques, such as
ERCP, CT, MRI, and imaging-guided fine-needle aspiration biopsy, were included
in the study. The lesions ranged from 16 to 49 mm in size (mean, 28 mm) and
were located in the head (n = 17), body (n = 6), or tail
(n =3) of the pancreas. Twenty-three other patients (13 women, 10
men; mean age, 69 ± 9.2 years) who were being followed up at our
institution because of pancreatic diseases other than pancreatic
adenocarcinoma were included in the study as control subjects. Twenty of the
control subjects had chronic pancreatitis, and three had intraductal papillary
mucinous tumors. All patients with pancreatic adenocarcinoma underwent
surgery, and the diagnoses were confirmed at pathologic evaluation. The
control subjects had clinical and CT follow-up examinations for at least 12
months after completion of the study; no evidence of pancreatic adenocarcinoma
was detected in any of the control subjects during the follow-up period. Our
institutional review board approved this study, and informed consent was
obtained from all study participants.
MR Protocol and Parameters
All MRI examinations were performed using a commercially available 1.5-T
superconducting MR unit (Signa LX, GE Healthcare). First, breath-hold
fat-saturated T1-weighted MR images were obtained with a combination of
gradient-echo sequences and chemical shift selective fat-suppression
techniques (TR range/TE range, 160-227/1.4-5.3; matrix, 256 x 256).
Second, respiratory-triggered fat-saturated T2-weighted fast spin-echo images
were acquired (TR range/TE, 2,500-6,432/80; echo-train length, 8-12; matrix,
192-256 x 256). Third, each patient underwent MR
cholangiopancreatography using a single-shot fast spin-echo sequence
(TR/firstecho TE, second-echo TE, infinite/80, 700; matrix size, 256 x
256) and a true-fast imaging with steadystate free precession (FISP) sequence
(TR/TE, 4.6/2.3; matrix size, 256 x 256). Fourth, breath-hold,
multiphasic contrast-enhanced MRI with fast gradient-echo sequences was
performed (170/2.3-4.2; flip angle, 90°; matrix, 256-512 x 512).
MR images were obtained after bolus injection of 20 mL of gadopentetate
dimeglumine (Magnevist, Schering). The patients underwent a multiphasic
contrast-enhanced MRI protocol including arterial phase, pancreatic
parenchymal phase, and portal venous phase imaging. The arterial phase images
were initiated 25 seconds after the beginning of the injection of contrast
material; pancreatic parenchymal phase images, 45-55 seconds after contrast
injection; and portal venous phase images, 70-90 seconds after contrast
injection.
Finally, all patients underwent DWI using the body coil of the MR unit. The
DWI technique was a modified version of the original DWI protocol
[6,
7]. The detailed parameters of
DWI were as follows: sequence, single-shot spin-echo echo-planar with
chemical-shift selective fat-suppression technique; scan direction, axial;
respiration, non-breath-hold method; b value, 1,000 s/mm2 (with
diffusion-weighted gradients applied in three orthogonal directions); TR
range/TE range, 8,000-10,000/73.2-73.4; inversion time, 70 milliseconds;
matrix, 128 x 64; slice thickness/gap, 4 mm/0 mm; field of view, 40 cm;
number of excitations, 6; and acquisition time, approximately 5 minutes. All
axial source images were provided with black-and-white reversal display to
facilitate lesion detection. Coronal maximum-intensity-projection (MIP) images
were also reconstructed from the axial source images and were evaluated using
the 3D rotational cine mode.
View larger version (156K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —Adenocarcinoma in head of pancreas with extensive necrosis in
58-year-old woman. Respiratory-triggered transverse T2-weighted fast spin-echo
MR image shows heterogeneously hyperintense mass (arrow) at head of
pancreas. Signal intensity of mass is high, similar to that of kidneys, which
might suggest cystic nature of mass.
|
|
View larger version (161K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —Adenocarcinoma in head of pancreas with extensive necrosis in
58-year-old woman. Breath-hold contrast-enhanced transverse T1-weighted
gradient-echo MR image obtained during pancreatic parenchymal phase clearly
shows extensive cystic area in mass (arrow).
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —Adenocarcinoma in head of pancreas with extensive necrosis in
58-year-old woman. Non-breath-hold transverse diffusion-weighted MR image with
inverted black-and-white image contrast clearly depicts mass (arrow)
showing strong signal intensity despite its cystic nature.
|
|
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —Adenocarcinoma in head of pancreas with extensive necrosis in
58-year-old woman. Fusion image—combination of T1- and
diffusion-weighted MR images—can facilitate identification of mass
(arrow) presenting at head of pancreas.
|
|
Image Analysis
On the basis of surgical-pathologic results, two radiologists serving as
study coordinators attempted to determine the location of the lesions and to
anatomically correlate the pathologically confirmed lesions with the imaging
findings as accurately as possible to allow detection of false-positive
interpretations. The study coordinators had access to all imaging, clinical,
and surgical data for the patients and control subjects; they also used fusion
images created from conventional MR images and DW images.
All high-b value DW images were then independently interpreted in random
order and blinded fashion by three abdominal radiologists (reviewers) other
than the study coordinators. The reviewers were aware that the study was being
performed to detect pancreatic adenocarcinoma. However, they were blinded to
all other information, such as patient identity, clinical history, the
findings of other imaging examinations, and results of histopathologic
evaluations. The reviewers interpreted only high-b value DW image series,
including axial source and MIP images alone without referring to any other MR
images. MIP images were evaluated with the rotational cine mode together with
the axial source images in different windows on diagnostic monitors. Each
reviewer graded the presence (or absence) of lesions on a 5-grade confidence
scale on the basis of the strength and appearance of dark signals on the
high-b value DW images as follows: 1, definitely absent (no signal); 2,
probably absent (nonlocalized, mild to moderate signal); 3, undetermined
(localized, mild to moderate signal); 4, probably present (localized, strong
signal with no definite margins); and 5, definitely present (localized, strong
signal with definite margins). If a lesion was considered to be present on a
high-b value DW image, the location of the lesion was recorded. Only lesions
recorded at the correct location determined by the study coordinators were
accepted as true-positive findings.
View larger version (104K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —Pseudocyst related to chronic pancreatitis in 60-year-old man.
Respiratory-triggered transverse T2-weighted fast spin-echo MR image reveals
well-defined mass (arrow) in tail of pancreas is showing extremely
high signal intensity, which may indicate cystic nature of mass.
|
|
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —Pseudocyst related to chronic pancreatitis in 60-year-old man.
Respiratory-triggered transverse diffusion-weighted MR image with inverted
black-and-white image contrast shows no significant signals in area
corresponding to area where mass is shown in A.
|
|
Statistical Analysis
Receiver operating characteristic (ROC) curves were used to represent the
performance of individual radiologists for tumor detection. The diagnostic
accuracy for each radiologist was determined by calculating the area under the
ROC curve (Az). Grade 4 and grade 5 were accepted as
positive for the presence of pancreatic adenocarcinoma, and sensitivity and
specificity were calculated with 95% CIs.
The interobserver agreement among reviewers for tumor detection was
calculated with linearweighted kappa statistics. A kappa statistic greater
than 0.75 was considered to indicate excellent agreement beyond chance;
0.4-0.75, fair to good agreement; and less than 0.4, poor agreement.
Results
ROC analysis yielded Az values of 0.998 (95% CI,
0.993-1), 0.998 (95% CI, 0.993-1), and 0.995 (95% CI, 0.984-1) for the three
reviewer radiologists (Figs.
1A,
1B,
1C,
2A,
2B,
2C,
2D,
3A,
3B,
4A,
4B,
4C). For each reviewer, the
sensitivity was 96.2% (25/26; 95% CI, 81.1-99.3%). All reviewers missed the
same pancreatic adenocarcinoma by grading it as "3" (Fig.
5A,
5B). The specificity was 100%
(23/23; 95% CI, 88.7-100%) for two radiologists and 98.6% (22/23; 95% CI,
79-99.2%) for the third one. The mean sensitivity and specificity of high-b
value DWI for the detection of pancreatic adenocarcinoma were 96.2% (75/78)
and 98.6% (68/69), respectively
View larger version (115K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —Intraductal papillary mucinous tumor, side branch type, in
71-year-old woman. Coronal MR cholangiopancreatography image shows dilated
main pancreatic duct (asterisks) and cystic mass (arrows) in
head of pancreas.
|
|
View larger version (150K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —Intraductal papillary mucinous tumor, side branch type, in
71-year-old woman. Axial T2-weighted MR image depicts same cystic mass
(arrowhead) as that shown in A. Spleen is marked with arrow to
serve as landmark for correlation with diffusion-weighted MR image.
|
|
View larger version (72K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C —Intraductal papillary mucinous tumor, side branch type, in
71-year-old woman. Respiratory-triggered transverse diffusion-weighted MR
image with inverted black-and-white image contrast shows no significant
signals at area corresponding to area where mass is shown in A and
B. Spleen is marked with arrow.
|
|
View larger version (93K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —Pancreatic adenocarcinoma in body of pancreas in 56-year-old man.
Using diffusion-weighted MR image corresponding to A, all three
reviewers missed lesion (arrows) by grading it as "3"
(undetermined; localized, mild to moderate signal).
|
|
All values indicating interobserver agreement were in the category of
excellent (
= 0.81-0.87).
Discussion
It is well known that diffusion is caused by random translational molecular
motion, also known as brownian water motion
[7]. DWI is the only imaging
method that can be used to evaluate the diffusion process in vivo. The speed
with which water molecules diffuse differs in extracellular and intracellular
components of tissues [8]. In
the intracellular component, diffusion is relatively slow because of the
presence of cellular membranes. Thus, ADCs, which are quantitative expressions
of the diffusion characteristics of tissues, are related to the proportion of
extracellular and intracellular components. ADC values tend to decrease with
increased tissue cellularity or cell density
[3]. On the other hand, the
cell density may be indicative of tumor aggressiveness; the results of several
clinical studies suggest an increased metastatic capacity of tumors with high
cellularity [9]. Moreover, in
addition to the cellular membranes, the intracellular cytoskeleton,
organelles, matrix fibers, and soluble macromolecules contribute to diffusion
restrictions in tumors [3];
therefore, DWI should be sensitive for the differentiation of histopathologic
tissue characteristics. In fact, several authors have already reported
decreased ADC values in various malignant lesions
[4,
5,
8]. However, no previous
studies, to our knowledge, have used direct visual assessment of DW images to
report the diagnostic performance of this technique for the detection of
pancreatic adenocarcinoma.
Although the technique used in the present study is principally based on
DWI, our concept was different from the standard use of this technique for the
detection of abdominal disease. The standard application of DWI is not
qualitative but quantitative and is based on ADC measurements
[4,
8]. The high-b value DWI
technique for this study uses an acquisition method with multiple excitations
and without breath-holding to improve SNR. The limitation in scanning time
imposed by breath-holding does not permit thin-slice DW images to be obtained
with adequate SNR and with multiple excitations that can be used as source
images for multiplanar reconstructions
[6,
7]. On the other hand, an
increase in motion artifacts might be assumed as a theoretic shortcoming
[4]; however, in practice,
motion artifacts are averaged during multiexcitations by the motion-probing
gradients applied for DWI and become inconspicuous on the reconstructed
images. Thus, images with a good SNR are achieved in exchange for absolute ADC
values that become impossible to calculate because of signal averaging. Most
recently, using the same technique, Ichikawa et al.
[7] reported a sensitivity of
91% and a specificity of 100% for the detection of colorectal cancer using
DWI. We obtained a similar high sensitivity (96.2%) and specificity (98.6%)
for the detection of pancreatic adenocarcinoma in the present study.
DWI can be easily performed as an adjunct to a conventional MRI study.
Furthermore, DW images can be fused with conventional MR images, like the
fusion images obtained with PET/CT scanners, to achieve better anatomic
resolution [10]. Although the
reviewers in our study did not have access to this kind of fusion images, the
study coordinators noted their usefulness (Figs.
1C and
2D). Regarding its high
sensitivity and specificity, DWI might have potential to become the imaging
method of choice for screening patients with symptoms suggesting pancreatic
adenocarcinoma. Moreover, this technique can also be used for screening people
who have a hereditary predisposition for pancreatic cancers, such as patients
with hereditary pancreatitis; multiple endocrine adenomatosis, type 1; and
Gardner's syndrome [11].
There are some limitations to our study. First, the study population was
relatively small, so our results need to be confirmed in larger clinical
studies. Second, the study included only cases of chronic pancreatitis and
intraductal papillary mucinous tumor as the negative cases and did not include
other benign conditions. Thus, the specificity reported in the present study
needs to be considered relative rather than absolute.
In conclusion, according to the results of our preliminary study, high-b
value DWI might be a useful tool for detecting pancreatic adenocarcinoma; it
shows a high sensitivity and specificity. Nevertheless, further studies in
larger clinical settings are needed to support our findings because of the
limitations mentioned.
References
- Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer.
Lancet 2004
27;363: 1049-1057[CrossRef][Medline]
- American Cancer Society. Cancer facts & figures
2002. Atlanta, GA: American Cancer Society,2002
- Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and
necrotic fraction in human melanoma xenografts by diffusion-weighted magnetic
resonance imaging. Magn Reson Med 2000;43
: 828-836[CrossRef][Medline]
- Dzik-Jurasz A, Domenig C, George M, et al. Diffusion MRI for
prediction of response of rectal cancer to chemoradiation.
Lancet 2002; 360:307
-308[CrossRef][Medline]
- Thoeny HC, De Keyzer F, Chen F, et al. Diffusion-weighted MR
imaging in monitoring the effect of a vascular targeting agent on
rhabdomyosarcoma in rats. Radiology 2005;234
: 756-764[Abstract/Free Full Text]
- Takahara T, Imai Y, Yamashita T, Yasuda S, Nasu S, Van Cauteren M.
Diffusion weighted whole body imaging with background body signal suppression
(DWIBS): technical improvement using free breathing, STIR and high resolution
3D display. Radiat Med 2004;22
: 275-282[Medline]
- Ichikawa T, Erturk SM, Motosugi U, et al. High-b-value
diffusion-weighted MRI in colorectal cancer. AJR2006; 187:181
-184[Abstract/Free Full Text]
- Roth Y, Tichler T, Kostenich G, et al. High-b-value
diffusion-weighted MR imaging for pretreatment prediction and early monitoring
of tumor response to therapy in mice. Radiology2004; 232:685
-692[Abstract/Free Full Text]
- Sugahara T, Korogi Y, Kochi M, et al. Usefulness of
diffusion-weighted MRI with echo-planar technique in the evaluation of
cellularity in gliomas. J Magn Reson Imaging1999; 9:53
-60[CrossRef][Medline]
- Hosten N, Lemke AJ, Wiedenmann B, Bohmig M, Rosewicz S. Combined
imaging techniques for pancreatic cancer. Lancet2000; 356:909
-910[CrossRef][Medline]
- Bae S, Kurtz RC. Strategy after diagnosis of pancreatic dysplasia.
Lancet 2000; 355:85
-86[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
W. Luboldt, R. Kufer, N. Blumstein, T. L. Toussaint, A. Kluge, M. D. Seemann, and H.-J. Luboldt
Prostate Carcinoma: Diffusion-weighted Imaging as Potential Alternative to Conventional MR and 11C-Choline PET/CT for Detection of Bone Metastases
Radiology,
December 1, 2008;
249(3):
1017 - 1025.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. M. Charles-Edwards, C. Messiou, V. A. Morgan, S. S. De Silva, N. A. McWhinney, M. Katesmark, A. D. Attygalle, and N. M. deSouza
Diffusion-weighted Imaging in Cervical Cancer with an Endovaginal Technique: Potential Value for Improving Tumor Detection in Stage Ia and Ib1 Disease
Radiology,
November 1, 2008;
249(2):
541 - 550.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. Inan, A. Arslan, G. Akansel, Y. Anik, and A. Demirci
Diffusion-Weighted Imaging in the Differential Diagnosis of Cystic Lesions of the Pancreas
Am. J. Roentgenol.,
October 1, 2008;
191(4):
1115 - 1121.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Kim, D. E. Morgan, H. Zeng, W. E. Grizzle, J. M. Warram, C. R. Stockard, D. Wang, and K. R. Zinn
Breast Tumor Xenografts: Diffusion-weighted MR Imaging to Assess Early Therapy with Novel Apoptosis-Inducing Anti-DR5 Antibody
Radiology,
September 1, 2008;
248(3):
844 - 851.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
