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


     

Published online before print April 22, 2004, 10.1148/radiol.2313030797
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2313030797v1
231/3/767    most recent
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 Sugita, R.
Right arrow Articles by Takahashi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sugita, R.
Right arrow Articles by Takahashi, S.
(Radiology 2004;231:767-774.)
© RSNA, 2004

Gastrointestinal Imaging

Periampullary Tumors: High-Spatial-Resolution MR Imaging and Histopathologic Findings in Ampullary Region Specimens1

Reiji Sugita, MD, Akemi Furuta, MD, Kei Ito, MD, Naotaka Fujita, MD, Ryou Ichinohasama, MD and Shoki Takahashi, MD

1 From the Department of Radiology, NTT East Tohoku Hospital, 2–29-1, Yamatomachi Wakabayashi-ku, Sendai, Japan (R.S., A.F.); Department of Gastroenterology, Sendai City Medical Center, Japan (K.I., N.F.); Department of Pathology, Tohoku University School of Dentistry, Sendai, Japan (R.I.); and Department of Radiology, Tohoku University School of Medicine, Sendai, Japan (S.T.). Received May 12, 2003; revision requested July 22; revision received October 20; accepted December 9. Address correspondence to R.S. (e-mail: reiji.s@thk.mhc.east.ntt.co.jp).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To prospectively determine the magnetic resonance (MR) signal intensity characteristics of structures of the ampullary region and to assess the potential use of MR imaging in evaluation of the extent of periampullary tumors in resected specimens.

MATERIALS AND METHODS: Twenty-five specimens from the ampullary region obtained in four autopsy cases without periampullary tumors and in 21 patients with periampullary tumors were examined with a 1.5-T MR system and a circular surface coil with 5-inch (12.7-cm) diameter. High-spatial-resolution MR images were obtained with field of view of 100 x 100 mm, matrix of 256 x 256 or 512 x 256, and section thickness of 2 mm. MR imaging findings were compared with histopathologic findings. Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of high-spatial-resolution MR imaging for assessment of tumor invasion into surrounding tissues were evaluated by two radiologists.

RESULTS: T1- and T2-weighted MR images clearly depicted normal structures in the ampullary region that included Oddi muscle, duodenal wall, common bile duct, and pancreas; these findings corresponded well with histologic findings. In 20 (95%) of 21 tumors, high-spatial-resolution MR imaging depicted location and extension of periampullary tumors precisely. Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of high-spatial-resolution MR imaging for assessment of tumor invasion into surrounding tissue were 88%, 100%, 96%, 100%, and 94%, respectively.

CONCLUSION: In this study, MR imaging correctly depicted location, extension, and origin of tumor. High-spatial-resolution MR imaging has potential for presurgical staging of tumors in this region.

© RSNA, 2004

Index terms: Bile ducts, neoplasms, 76.32, 767.32 • Duodenum, neoplasms, 73.32 • Pancreas, neoplasms, 77.32 • Specimens, MR, 73.121411, 76.121411, 77.121411


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Periampullary tumors arise within 2 cm of the ampulla in the duodenum and include pancreatic head cancer, lower common bile duct cancer, ampullary cancer, and periampullary duodenal cancer (1–6). Although patients with periampullary tumors have similar symptoms at presentation, they exhibit different clinical outcomes according to the origin of the tumors (1,2). Therefore, visualization of normal structures, differentiation among periampullary tumors, and evaluation of the extent of periampullary tumors are important in treatment planning (1–6). As for ampullary cancer, if tumor infiltration is limited to Oddi muscle, such cases are generally considered as early cancer (7–9). Thus, visualization of Oddi muscle is important in the staging of ampullary cancer. It is also essential in the evaluation of tumor extension into the ampullary region in regard to other periampullary tumors.

Although computed tomography (CT), conventional magnetic resonance (MR) imaging, and transabdominal ultrasonography (US) play important roles in the evaluation of biliary disease, the fine structures of the ampullary region are not sufficiently visualized. As a result, evaluation of local extension of periampullary tumors by these means is limited (10–17). Currently, MR cholangiopancreatography has been accepted as another method for the evaluation of the ampullary region (18–20). With this method, the ampullary lesion cannot be visualized directly. Endoscopic US and intraductal US have been used for the evaluation of the ampullary region because of their high spatial resolution (7–9,21,22). Particularly, it has been reported that intraductal US is the only modality with which the Oddi muscle can be visualized (7,8). However, both endoscopic and intraductal US are highly operator dependent and somewhat invasive modalities.

High-spatial-resolution MR imaging has been widely used in vitro as a research tool for assessment of the signal intensity characteristics of excised organs or tumors (23–27). To our knowledge, excised specimens of the ampullary region have not been examined with high-spatial-resolution MR imaging. The purpose of this study was to prospectively determine the MR signal intensity characteristics of the structures of the ampullary region and to assess the potential use of MR imaging in the evaluation of the extent of periampullary tumors in resected specimens.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Study Group
Between September 2001 and April 2003, we examined 25 specimens that were obtained from 21 patients (13 men and eight women; average age, 65 years; range, 47–74 years) who had undergone pancreatoduodenectomy because they were suspected of having pancreatobiliary diseases and four autopsy cases (three men and one woman; average age, 64 years; range, 61–69 years) without pancreatobiliary diseases according to the autopsy record and high-spatial-resolution MR imaging findings. In 21 patients, surgical specimens were obtained at pancreatoduodenectomy. The final histologic diagnoses in these patients were 12 pancreatic head cancers, three ampullary cancers, two periampullary duodenal cancers, and four lower common bile duct cancers. In all specimens, part of the ampullary region was imaged in this study. The study protocol was approved by the institutional review board, and informed consent concerning the use of the resected specimens for this study was obtained from all patients. Informed consent was obtained from relatives for use of specimens from the four autopsy cases.

Imaging Technique
High-spatial-resolution MR imaging was performed by using a 1.5-T MR system (Signa Horizon EchoSpeed; GE Medical Systems, Milwaukee, Wis) with a 23 mT/m maximum gradient capability and a circular surface coil with a diameter of 5 inches (12.7 cm). Conventional single-section sagittal, coronal, and transverse scout images of the specimens of the ampullary region were initially obtained.

Two specimens of periampullary tumor were imaged in a fresh state within 2 hours after resection and were reexamined after fixation in formalin. All other specimens were imaged after fixation. Four normal autopsy specimens were stored in 10% formalin for 10–30 days before MR imaging. Specimens of periampullary tumors were stored in 10% formalin for the time needed for complete fixation of the specimens. The pathologist (R.I.) did not manipulate the specimens before MR imaging.

In all specimens, T1-weighted spin-echo MR images were obtained with repetition time msec/echo time msec of 500/20, a bandwidth of 31.25 kHz, and six signals acquired. T2-weighted fast spin-echo MR images were obtained with 4,000/85 (effective), echo train length of 12, bandwidth of 31.25, and six signals acquired. All of these images were obtained with a field of view of 100 x 100 mm. A matrix of 256 x 256 (frequency encoding by phase encoding) was used in the first nine consecutive cases, whereas a matrix of 512 x 256 (frequency encoding by phase encoding) was used in the next 16 consecutive cases. A section thickness of 2 mm and an intersection gap of 0.5 mm were used. Acquisition times for T1- and T2-weighted MR images were approximately 12 and 8 minutes, respectively. The voxel size for a matrix of 256 x 256 was 0.4 x 0.4 x 2.0 mm, which equals 0.32 mm3, and the voxel size for a matrix of 512 x 256 was 0.2 x 0.4 x 2.0 mm, which equals 0.16 mm3. The maximum spatial resolution for a matrix of 256 x 256 was 400 µm/pixel, and that for a matrix of 512 x 256 was 200 µm/pixel. The orientation of both the T1- and T2-weighted MR images was perpendicular to the longitudinal axis of the resected descending portion of the duodenum, that is, roughly perpendicular to the distal common bile duct.

Image Analysis
After MR imaging, all 25 specimens from the resected ampullary region were sectioned perpendicular to the long axis of the descending portion of the duodenum to correspond to the orientation of the MR images. The specimen was then embedded in paraffin and cut with a microtome into 2.5-µm-thick sections. These sections were then stained with hematoxylin-eosin.

The four normal autopsy specimens were evaluated in terms of signal intensity, continuity, and uniformity of the normal structures, including the duodenal wall, the bile duct wall, the pancreas, and the ampullary region. Two radiologists (R.S., 16 years of experience; A.F., 17 years of experience) and an experienced pathologist (R.I., 23 years of experience) jointly compared MR images with specific histologic sections, and comparisons were made with visual analysis.

For the 21 periampullary cancers, the location and extension of tumors and involvement of lymph nodes were documented. The site of origin was determined on the basis of the largest site of tumor, according to guidelines in a previous publication (2). The MR imaging criteria used to determine the location and extent of tumor were as follows: discrete masses present in the ampullary and periampullary regions, focal areas of abnormal signal intensity in ampullary and periampullary regions, marked thickening of the complex structure of the common bile duct, Oddi muscle in the ampullary region, and thickening or polypoid mass in the lower bile duct wall (17,23–25). In the determination of tumor involvement, MR imaging findings of normal specimens were considered the reference standard.

MR imaging findings were compared with the histologic interpretations of the pathologist, who was unaware of the findings at MR imaging. All the hard-copy MR images were evaluated with separate interpretations made independently by two experienced radiologists. Although one (R.S.) of the two ex vivo readers was involved in the primary in vivo clinical reading, the reader did not have any clinical information for the ex vivo reading. The other ex vivo reader (A.F.) was involved only in the ex vivo reading but not in the primary in vivo clinical reading. An ex vivo reader interpreted MR images of the specimens. On the other hand, an in vivo reader interpreted presurgical clinical MR images obtained in the patients in whom the specimens were collected.

The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of high-spatial-resolution MR imaging for assessment of tumor invasion into the surrounding tissue were evaluated for the group, which included four normal autopsy specimens and 21 diseased specimens. In the case of disagreement as to the evaluation results, final consensus was achieved with interobserver discussion after the findings of the two independent interpreters (radiologists) were recorded. The degree of tumor extension into the surrounding structures was determined on the basis of the International Union Against Cancer classification (28) (Fig 1). For three ampullary tumors, invasion into the duodenal wall or pancreas was assessed. For the other three kinds of periampullary tumors (12 pancreatic head cancers, four lower bile duct cancers, two periampullary duodenal cancers), evaluation focused on whether the tumor invaded the ampullary region. We did not consider the spreading of the tumor into the extraampullary region (eg, invasion into large vessels and lower bile ducts in pancreatic head cancer or invasion into the pancreas in bile duct cancer).



View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a. Schemata of (a) normal duodenal ampullary region and (b-e) International Union Against Cancer classification of periampullary cancers with tumor classification. (a) Common bile duct and pancreatic duct terminals penetrate the duodenal wall and form a common channel surrounded by Oddi muscle. (1) m = mucosa, (2) sm submucosa, (3) pm = muscularis propria, (4) ss = serosa or subserosa, (5) Panc = pancreatic head, (6) PD = pancreatic duct, (7) BD = bile duct, and D = duodenal cavity. Structures are those depicted at MR imaging and are listed in the Table. (b) Ampullary cancer. T1 indicates tumor limited to Oddi muscle; T2, tumor with invasion of duodenal wall; and T3, tumor with invasion of pancreas. (c) Extrahepatic bile duct cancer. T1 indicates tumor confined to bile duct; T2, tumor with invasion beyond wall of bile duct; T3, tumor with invasion of pancreas; and T4, tumor with invasion of duodenum or adjacent organs. (d) Pancreatic cancer. T1 indicates tumor (<=2 cm in greatest dimension) limited to pancreas; T2, tumor (>2 cm in greatest dimension) limited to pancreas; and T3, tumor with extension beyond pancreas. (e) Duodenal cancer. T1 indicates tumor with invasion of lamina propria or submucosa; T2, tumor with invasion of muscularis propria; T3, tumor with invasion through muscularis propria into subserosa; and T4, tumor with direct invasion of other organs or structures.

 


View larger version (73K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b. Schemata of (a) normal duodenal ampullary region and (b-e) International Union Against Cancer classification of periampullary cancers with tumor classification. (a) Common bile duct and pancreatic duct terminals penetrate the duodenal wall and form a common channel surrounded by Oddi muscle. (1) m = mucosa, (2) sm submucosa, (3) pm = muscularis propria, (4) ss = serosa or subserosa, (5) Panc = pancreatic head, (6) PD = pancreatic duct, (7) BD = bile duct, and D = duodenal cavity. Structures are those depicted at MR imaging and are listed in the Table. (b) Ampullary cancer. T1 indicates tumor limited to Oddi muscle; T2, tumor with invasion of duodenal wall; and T3, tumor with invasion of pancreas. (c) Extrahepatic bile duct cancer. T1 indicates tumor confined to bile duct; T2, tumor with invasion beyond wall of bile duct; T3, tumor with invasion of pancreas; and T4, tumor with invasion of duodenum or adjacent organs. (d) Pancreatic cancer. T1 indicates tumor (<=2 cm in greatest dimension) limited to pancreas; T2, tumor (>2 cm in greatest dimension) limited to pancreas; and T3, tumor with extension beyond pancreas. (e) Duodenal cancer. T1 indicates tumor with invasion of lamina propria or submucosa; T2, tumor with invasion of muscularis propria; T3, tumor with invasion through muscularis propria into subserosa; and T4, tumor with direct invasion of other organs or structures.

 


View larger version (48K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c. Schemata of (a) normal duodenal ampullary region and (b-e) International Union Against Cancer classification of periampullary cancers with tumor classification. (a) Common bile duct and pancreatic duct terminals penetrate the duodenal wall and form a common channel surrounded by Oddi muscle. (1) m = mucosa, (2) sm submucosa, (3) pm = muscularis propria, (4) ss = serosa or subserosa, (5) Panc = pancreatic head, (6) PD = pancreatic duct, (7) BD = bile duct, and D = duodenal cavity. Structures are those depicted at MR imaging and are listed in the Table. (b) Ampullary cancer. T1 indicates tumor limited to Oddi muscle; T2, tumor with invasion of duodenal wall; and T3, tumor with invasion of pancreas. (c) Extrahepatic bile duct cancer. T1 indicates tumor confined to bile duct; T2, tumor with invasion beyond wall of bile duct; T3, tumor with invasion of pancreas; and T4, tumor with invasion of duodenum or adjacent organs. (d) Pancreatic cancer. T1 indicates tumor (<=2 cm in greatest dimension) limited to pancreas; T2, tumor (>2 cm in greatest dimension) limited to pancreas; and T3, tumor with extension beyond pancreas. (e) Duodenal cancer. T1 indicates tumor with invasion of lamina propria or submucosa; T2, tumor with invasion of muscularis propria; T3, tumor with invasion through muscularis propria into subserosa; and T4, tumor with direct invasion of other organs or structures.

 


View larger version (67K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1d. Schemata of (a) normal duodenal ampullary region and (b-e) International Union Against Cancer classification of periampullary cancers with tumor classification. (a) Common bile duct and pancreatic duct terminals penetrate the duodenal wall and form a common channel surrounded by Oddi muscle. (1) m = mucosa, (2) sm submucosa, (3) pm = muscularis propria, (4) ss = serosa or subserosa, (5) Panc = pancreatic head, (6) PD = pancreatic duct, (7) BD = bile duct, and D = duodenal cavity. Structures are those depicted at MR imaging and are listed in the Table. (b) Ampullary cancer. T1 indicates tumor limited to Oddi muscle; T2, tumor with invasion of duodenal wall; and T3, tumor with invasion of pancreas. (c) Extrahepatic bile duct cancer. T1 indicates tumor confined to bile duct; T2, tumor with invasion beyond wall of bile duct; T3, tumor with invasion of pancreas; and T4, tumor with invasion of duodenum or adjacent organs. (d) Pancreatic cancer. T1 indicates tumor (<=2 cm in greatest dimension) limited to pancreas; T2, tumor (>2 cm in greatest dimension) limited to pancreas; and T3, tumor with extension beyond pancreas. (e) Duodenal cancer. T1 indicates tumor with invasion of lamina propria or submucosa; T2, tumor with invasion of muscularis propria; T3, tumor with invasion through muscularis propria into subserosa; and T4, tumor with direct invasion of other organs or structures.

 


View larger version (63K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1e. Schemata of (a) normal duodenal ampullary region and (b-e) International Union Against Cancer classification of periampullary cancers with tumor classification. (a) Common bile duct and pancreatic duct terminals penetrate the duodenal wall and form a common channel surrounded by Oddi muscle. (1) m = mucosa, (2) sm submucosa, (3) pm = muscularis propria, (4) ss = serosa or subserosa, (5) Panc = pancreatic head, (6) PD = pancreatic duct, (7) BD = bile duct, and D = duodenal cavity. Structures are those depicted at MR imaging and are listed in the Table. (b) Ampullary cancer. T1 indicates tumor limited to Oddi muscle; T2, tumor with invasion of duodenal wall; and T3, tumor with invasion of pancreas. (c) Extrahepatic bile duct cancer. T1 indicates tumor confined to bile duct; T2, tumor with invasion beyond wall of bile duct; T3, tumor with invasion of pancreas; and T4, tumor with invasion of duodenum or adjacent organs. (d) Pancreatic cancer. T1 indicates tumor (<=2 cm in greatest dimension) limited to pancreas; T2, tumor (>2 cm in greatest dimension) limited to pancreas; and T3, tumor with extension beyond pancreas. (e) Duodenal cancer. T1 indicates tumor with invasion of lamina propria or submucosa; T2, tumor with invasion of muscularis propria; T3, tumor with invasion through muscularis propria into subserosa; and T4, tumor with direct invasion of other organs or structures.

 
Finally, the MR signal intensity, shape, and size of all the parts of the two fresh specimens obtained from surgical cases in the 21 patients were compared with those of the two fixed specimens.

Statistical Analysis
To assess whether or not there were differences in regard to age and sex between four autopsy cases and 21 patients, analysis with Student t and {chi}2 tests, respectively, was performed. To assess whether or not there were differences in regard to age and sex among patient groups with four kinds of periampullary tumors, analysis of variance and the {chi}2 test, respectively, were performed. In these analyses, we considered a difference with P < .05 as statistically significant. To assess interobserver variability in image interpretation between the two observers, the {kappa} statistic was used. The level of agreement for the {kappa} value was defined as follows: 0–0.20, slight; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; greater than 0.81, almost perfect (29). Then we assessed the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of high-spatial-resolution MR imaging for assessment of tumor invasion into the surrounding tissue. Statistical data were calculated by using commercially available software (SAS; SAS Institute, Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
We collected specimens in four autopsy cases without pancreatobiliary diseases and 21 patients with periampullary tumor. There were no significant differences in either age or sex between them or among patient groups with four kinds of periampullary tumors.

Normal Specimens
Signal intensity characteristics of the ampullary region.—Findings at histologic analysis of the ampullary region confirmed that it consisted of not only Oddi muscle but also the surrounding connective tissue.

The ampullary region protruded into the duodenal lumen and showed signal intensity different from that of the surrounding duodenal wall: It was hypointense in reference to the signal intensity of the duodenal mucosa and hyperintense in reference to that of the duodenal submucosa on T1-weighted MR images. It was hypointense in reference to the signal intensity of either the mucosa or the submucosa of the duodenal wall on T2-weighted MR images (Fig 2).



View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Images of normal ampullary region show pancreatic lobulus (P) and duodenal wall with four layers: mucosa (m), submucosa (sm), muscularis propria (pm), and subserosa or serosa (ss). (a) Oblique high-spatial-resolution T1-weighted spin-echo MR image (500/20). Intermediate- to low-signal-intensity band (arrowheads) indicates complex of Oddi muscle and connective tissue. (b) Oblique high-spatial-resolution T2-weighted fast spin-echo MR image (4,000/85; echo train length, 12). Low-signal-intensity band (arrowheads) indicates complex of Oddi muscle and connective tissue. (c) Photomicrograph shows that ampullary region is composed of complex (arrowheads) of Oddi muscle and connective tissue. Cystic lesion in ampullary wall shows glandular proliferation. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Images of normal ampullary region show pancreatic lobulus (P) and duodenal wall with four layers: mucosa (m), submucosa (sm), muscularis propria (pm), and subserosa or serosa (ss). (a) Oblique high-spatial-resolution T1-weighted spin-echo MR image (500/20). Intermediate- to low-signal-intensity band (arrowheads) indicates complex of Oddi muscle and connective tissue. (b) Oblique high-spatial-resolution T2-weighted fast spin-echo MR image (4,000/85; echo train length, 12). Low-signal-intensity band (arrowheads) indicates complex of Oddi muscle and connective tissue. (c) Photomicrograph shows that ampullary region is composed of complex (arrowheads) of Oddi muscle and connective tissue. Cystic lesion in ampullary wall shows glandular proliferation. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c. Images of normal ampullary region show pancreatic lobulus (P) and duodenal wall with four layers: mucosa (m), submucosa (sm), muscularis propria (pm), and subserosa or serosa (ss). (a) Oblique high-spatial-resolution T1-weighted spin-echo MR image (500/20). Intermediate- to low-signal-intensity band (arrowheads) indicates complex of Oddi muscle and connective tissue. (b) Oblique high-spatial-resolution T2-weighted fast spin-echo MR image (4,000/85; echo train length, 12). Low-signal-intensity band (arrowheads) indicates complex of Oddi muscle and connective tissue. (c) Photomicrograph shows that ampullary region is composed of complex (arrowheads) of Oddi muscle and connective tissue. Cystic lesion in ampullary wall shows glandular proliferation. (Hematoxylin-eosin stain; original magnification, x1.3.)

 
Signal intensity characteristics of other structures.—The signal intensity of the normal structures in the ampullary region is summarized in Figure 2 and the Table. MR images depicted the normal duodenal wall as consisting of four layers that corresponded well to the histologic layers: mucosa, submucosa, muscularis propria, and subserosa or serosa. The signal intensity of the bile duct wall was low on both T1- and T2-weighted MR images. Although the depiction of the duodenal wall on MR images was clear, depiction of the bile duct wall was not clear as consisting of four layers (mucosa, fibromuscularis, subserosa, and serosa).


View this table:
[in this window]
[in a new window]

  		SI of Normal Structures of Ampullary Region at High-Spatial-Resolution MR Imaging

 
MR images showed the pancreas as mixed tissues of lobulus and fat deposition. On T1-weighted MR images, the pancreas showed a variable degree of signal intensity from intermediate to markedly high, depending on the degree of fat deposition. The higher the proportion of fat, the higher the signal intensity of the pancreas. The pancreas also showed high signal intensity in proportion to the amount of fat content on T2-weighted MR images. The main pancreatic duct showed low signal intensity on both T1- and T2-weighted MR images.

Tumor Specimens
Detection and evaluation of extent of periampullary tumors at high-spatial-resolution MR imaging.—Twenty-one periampullary tumors that were examined included three ampullary cancers (one carcinoma confined to the Oddi muscle and two that invaded the periampullary duodenal wall), 12 pancreatic cancers (two carcinomas that invaded the periampullary duodenal wall and 10 carcinomas limited to the pancreas), four bile duct cancers (two carcinomas with periampullary invasion and two carcinomas without periampullary invasion) (Fig 3), and two duodenal cancers with invasion of the pancreas (Fig 4).



View larger version (102K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. Distal common bile duct cancer with invasion into ampullary region. (a) Oblique high-spatial-resolution T2-weighted MR image (4,000/85) shows tumor invasion (arrowheads) of pancreas (P). (b) Photomicrograph of same structure. (Hematoxylin-eosin stain; original magnification, x2.0.)

 


View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. Distal common bile duct cancer with invasion into ampullary region. (a) Oblique high-spatial-resolution T2-weighted MR image (4,000/85) shows tumor invasion (arrowheads) of pancreas (P). (b) Photomicrograph of same structure. (Hematoxylin-eosin stain; original magnification, x2.0.)

 


View larger version (75K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a. Periampullary duodenal cancer without invasion into ampullary region. (a) Oblique high-spatial-resolution T2-weighted MR image (4,000/85) shows tumor (arrowheads). Arrow shows marker for location of tumor. (b). Photomicrograph of tumor (arrowheads). (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b. Periampullary duodenal cancer without invasion into ampullary region. (a) Oblique high-spatial-resolution T2-weighted MR image (4,000/85) shows tumor (arrowheads). Arrow shows marker for location of tumor. (b). Photomicrograph of tumor (arrowheads). (Hematoxylin-eosin stain; original magnification, x1.3.)

 
For ampullary carcinoma that invaded the duodenal wall and pancreas, invasion of the muscularis propria of the duodenal wall and pancreatic parenchyma was demonstrated as a mass that extended into the pancreas with irregular disruption of the duodenal layer structure on T2-weighted MR images (Fig 5). In cases of pancreatic cancer, invasion of the periampullary duodenal wall was clearly distinguished from the normal duodenal wall (Fig 6).



View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. Ampullary tumor with invasion of pancreas. (a) Single-section in vivo MR cholangiopancreatographic image ({infty}/880 [effective]) shows filling defect (arrow) at distal end of common bile duct and main pancreatic duct. (b) In vivo coronal single-shot T2-weighted fast spin-echo MR image ({infty}/94 [effective]) shows abrupt obstruction of the proximally dilated bile duct (arrow). Tumor cannot be discriminated from pancreas or duodenal wall, all of which appear as a hypointense area. (c, d) Oblique high-spatial-resolution T1- (500/20) and T2-weighted (4,000/85) MR images show tumor (white arrowheads) with invasion of duodenal muscularis propria and pancreas. Tumor interrupts duodenal muscularis propria (black arrowhead) and infiltrates pancreas (arrows). (e) Photomicrograph reveals tumor with extension. Arrowheads and arrows indicate same structures as in c and d. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. Ampullary tumor with invasion of pancreas. (a) Single-section in vivo MR cholangiopancreatographic image ({infty}/880 [effective]) shows filling defect (arrow) at distal end of common bile duct and main pancreatic duct. (b) In vivo coronal single-shot T2-weighted fast spin-echo MR image ({infty}/94 [effective]) shows abrupt obstruction of the proximally dilated bile duct (arrow). Tumor cannot be discriminated from pancreas or duodenal wall, all of which appear as a hypointense area. (c, d) Oblique high-spatial-resolution T1- (500/20) and T2-weighted (4,000/85) MR images show tumor (white arrowheads) with invasion of duodenal muscularis propria and pancreas. Tumor interrupts duodenal muscularis propria (black arrowhead) and infiltrates pancreas (arrows). (e) Photomicrograph reveals tumor with extension. Arrowheads and arrows indicate same structures as in c and d. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (109K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c. Ampullary tumor with invasion of pancreas. (a) Single-section in vivo MR cholangiopancreatographic image ({infty}/880 [effective]) shows filling defect (arrow) at distal end of common bile duct and main pancreatic duct. (b) In vivo coronal single-shot T2-weighted fast spin-echo MR image ({infty}/94 [effective]) shows abrupt obstruction of the proximally dilated bile duct (arrow). Tumor cannot be discriminated from pancreas or duodenal wall, all of which appear as a hypointense area. (c, d) Oblique high-spatial-resolution T1- (500/20) and T2-weighted (4,000/85) MR images show tumor (white arrowheads) with invasion of duodenal muscularis propria and pancreas. Tumor interrupts duodenal muscularis propria (black arrowhead) and infiltrates pancreas (arrows). (e) Photomicrograph reveals tumor with extension. Arrowheads and arrows indicate same structures as in c and d. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d. Ampullary tumor with invasion of pancreas. (a) Single-section in vivo MR cholangiopancreatographic image ({infty}/880 [effective]) shows filling defect (arrow) at distal end of common bile duct and main pancreatic duct. (b) In vivo coronal single-shot T2-weighted fast spin-echo MR image ({infty}/94 [effective]) shows abrupt obstruction of the proximally dilated bile duct (arrow). Tumor cannot be discriminated from pancreas or duodenal wall, all of which appear as a hypointense area. (c, d) Oblique high-spatial-resolution T1- (500/20) and T2-weighted (4,000/85) MR images show tumor (white arrowheads) with invasion of duodenal muscularis propria and pancreas. Tumor interrupts duodenal muscularis propria (black arrowhead) and infiltrates pancreas (arrows). (e) Photomicrograph reveals tumor with extension. Arrowheads and arrows indicate same structures as in c and d. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5e. Ampullary tumor with invasion of pancreas. (a) Single-section in vivo MR cholangiopancreatographic image ({infty}/880 [effective]) shows filling defect (arrow) at distal end of common bile duct and main pancreatic duct. (b) In vivo coronal single-shot T2-weighted fast spin-echo MR image ({infty}/94 [effective]) shows abrupt obstruction of the proximally dilated bile duct (arrow). Tumor cannot be discriminated from pancreas or duodenal wall, all of which appear as a hypointense area. (c, d) Oblique high-spatial-resolution T1- (500/20) and T2-weighted (4,000/85) MR images show tumor (white arrowheads) with invasion of duodenal muscularis propria and pancreas. Tumor interrupts duodenal muscularis propria (black arrowhead) and infiltrates pancreas (arrows). (e) Photomicrograph reveals tumor with extension. Arrowheads and arrows indicate same structures as in c and d. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a. Pancreatic cancer with invasion into ampullary region. (a, b) Oblique high-spatial-resolution T2-weighted MR images (4,000/85). Pancreatic tumor (P) extends across muscularis propria (arrowheads and broken line) and infiltrates ampullary region (arrow) and surrounding duodenal wall. (b) Tumor (arrows) largely involves pancreas and compresses duodenal wall outward. (c) Photomicrograph shows a slight difference in position and angle between MR images and cross section of specimen. Arrow, arrowheads, and broken line indicate same structures as in a. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b. Pancreatic cancer with invasion into ampullary region. (a, b) Oblique high-spatial-resolution T2-weighted MR images (4,000/85). Pancreatic tumor (P) extends across muscularis propria (arrowheads and broken line) and infiltrates ampullary region (arrow) and surrounding duodenal wall. (b) Tumor (arrows) largely involves pancreas and compresses duodenal wall outward. (c) Photomicrograph shows a slight difference in position and angle between MR images and cross section of specimen. Arrow, arrowheads, and broken line indicate same structures as in a. (Hematoxylin-eosin stain; original magnification, x1.3.)

 


View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6c. Pancreatic cancer with invasion into ampullary region. (a, b) Oblique high-spatial-resolution T2-weighted MR images (4,000/85). Pancreatic tumor (P) extends across muscularis propria (arrowheads and broken line) and infiltrates ampullary region (arrow) and surrounding duodenal wall. (b) Tumor (arrows) largely involves pancreas and compresses duodenal wall outward. (c) Photomicrograph shows a slight difference in position and angle between MR images and cross section of specimen. Arrow, arrowheads, and broken line indicate same structures as in a. (Hematoxylin-eosin stain; original magnification, x1.3.)

 
In 20 (95%) of 21 tumors, high-spatial-resolution MR imaging depicted the location and extension of the tumor precisely. T2-weighted MR images were particularly useful in visualization of tumor extension into the surrounding tissues. The origins of all 20 detected tumors were correctly depicted. In the remaining bile duct tumor, nonpolypoid extension was observed within the bile duct epithelium.

MR imaging also depicted two periampullary lymph nodes (10 x 7 and 8 x 6 mm) in two patients with pancreatic cancer, and they were confirmed to be metastatic at histopathologic examination. These lymph nodes were hyperintense relative to the submucosa of the duodenum on T1-weighted MR images and hypointense on T2-weighted MR images. In two (10%) of 21 lesions, the radiologists did not agree in regard to their interpretations, and consensus was achieved with discussion. A {kappa} value of 0.79 between the two observers was obtained, and thus, interobserver agreement for the determination of location and extension of periampullary carcinomas was substantial.

Accuracy of high-spatial-resolution MR imaging.—In three (12%) of 25 lesions, the radiologists did not agree with respect to their interpretations, and consensus was achieved with discussion. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of high-spatial-resolution MR imaging in the specimens for the evaluation of invasion of periampullary carcinoma were 88% (seven of eight), 100% (17 of 17), 96% (24 of 25), 100% (seven of seven), and 94% (17 of 18), respectively. A false-negative determination was made in regard to one bile duct cancer.

Comparison of fresh and fixed specimens.—The morphologic features and signal intensity of the fresh and fixed specimens were compared in two cases. After fixation, the shape of all or part of the specimens remained unchanged; however, there was overall tissue shrinkage. In addition, signal intensity of the whole tissue decreased only slightly on both T1- and T2-weighted MR images. MR image interpretation was not impaired because of this change.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
Periampullary tumors are the third most frequent gastrointestinal tumors following gastric and colorectal cancers (4). These tumors exhibit similar symptoms but different clinical outcomes, according to their origin (1–6,28). That is, the prognoses for ampullary carcinoma and periampullary duodenal carcinoma are better than those for distal common bile duct carcinoma and pancreatic head carcinoma (1–6). Furthermore, presurgical staging of a periampullary tumor is prescribed on the basis of infiltration into the surrounding tissue (1–6). Therefore, differentiation among periampullary tumors and determination of their extent are important in the planning of treatment of these tumors.

For each of the periampullary cancers of four origins in the present series, pancreatoduodenectomy was performed because tumor origin or tumor extension could not be accurately determined. If such presurgical determination could have been possible, surgical procedures should have been different depending on the tumor origin and tumor extent (2). For example, when ampullary cancer is located within the Oddi muscle, minimally invasive surgery (transduodenal papillectomy) might be selected (30). In addition, it is reported that the presence or absence of infiltration into the pancreas is a prognostic factor of ampullary cancer (31). Therefore, presurgical determination of the extension of the ampullary tumor is important for both selection of therapeutic strategies and prediction of prognosis. Similarly, in duodenal cancer, local resection of the duodenum can be sufficient if the tumor is restricted (32,33). However, the staging of bile duct or pancreatic head cancer has no influence on the strategy of surgical treatment, but it seriously affects prognosis (2,33,34).

Although distant metastases generally preclude primary surgical treatment, we did not evaluate them because of the restricted field of view of high-spatial-resolution MR imaging in this study. Instead, we attempted to determine local extension of periampullary tumors by using high-spatial-resolution MR imaging, which is important for determination of both therapeutic strategies and prognosis.

Several diagnostic imaging procedures are available for presurgical staging of periampullary cancers (12,13,16). MR imaging also provides a good imaging method for the periampullary tumor, with high lesion detectability owing to technical progress (10,15). In addition, MR cholangiopancreatography offers additional information about the state of the pancreaticobiliary duct (11,14,17). With most of these procedures, visualization of the fine structures of the ampullary region cannot be achieved, and this lack of visualization results in limited accuracy in presurgical staging of periampullary carcinoma. Endoscopic and intraductal US have high diagnostic ability because of their high spatial resolution (7–9,21,22). Particularly, intraductal US is the only modality with which the Oddi muscle can be visualized (7,8). In contrast, both endoscopic and intraductal US are highly operator dependent and invasive modalities.

Our data demonstrated that ex vivo high-spatial-resolution MR imaging successfully depicted fine structures in the ampullary region, including the duodenal wall with four layers, the bile duct wall, pancreatic lobules, and main pancreatic duct wall. The thickness of the Oddi muscle ranges approximately from 200 to 1,000 µm (8), which is outside the scope of the spatial resolution (200–400 µm/pixel) of MR imaging in the present study. That is why the Oddi muscle itself was not depicted with MR imaging in this study, but the complex of Oddi muscle and surrounding connective tissue could be depicted. Our results showed that findings on MR images closely paralleled the structures identified on photomicrographs of the specimens. In this study, we were able to correctly determine the location, extension, and origin of the tumor in 20 (95%) of 21 specimens, including 21 tumors. In pancreatic and duodenal tumors, MR imaging depicted the tumor extension to the neighboring structure. In ampullary tumors, MR imaging demonstrated the area between the tumor and surrounding tissue.

In the present study, false-negative diagnosis occurred in one patient with nonpolypoid bile duct cancer; diagnosis of such disease in a patient with a tumor at this location with extension would be difficult with any modality. In ex vivo specimens, however, the bile duct was collapsed without bile remaining inside. On the contrary, in vivo MR imaging could help delineate the bile duct, thus improving the determination of location and extension of nonpolypoid bile duct tumors.

Our study had several limitations. The first limitation included the differences between ex vivo and in vivo MR imaging. The most important one was that the examination time was inordinately long. Without doubt, this flaw should have been accompanied by motion artifact, including peristasis, breathing, cardiovascular pulsations, and gross patient movements in the case of in vivo examination. In addition, the pulse sequences used in this study usually would not be performed in vivo. Ex vivo examination of the signal intensity of the structures and tumors in the periampullary region on T1- and T2-weighted MR images with this field of view (100 x 100 mm) seems to be important as a foundation. Development of faster MR imaging techniques with a reduced or targeted field of view will allow clinical use of high-spatial-resolution MR imaging of periampullary tumors.

The second limitation was that all the tumor specimens were obtained after radical surgery, and the MR image readers were not blinded to this fact, which might have caused selection bias.

The third limitation was that the patients had far more pancreatic head cancers (three to six times as many as other types of periampullary tumors), which possibly made fair comparison difficult among the MR imaging features of the four types of cancers.

Finally, signal intensity of fixed specimens was slightly different from that of fresh specimens. Some authors report that there are differences in signal intensity between fixed and fresh specimens (24,25,35). However, we believe that the influence of differences in signal intensity between fixed and fresh specimens is negligible in the interpretation of MR imaging findings in the ampullary region.

In conclusion, the present study findings demonstrated that ex vivo high-spatial-resolution MR imaging clearly depicted the fine structures in the ampullary region and suggest that MR imaging has the potential for accurate evaluation of the extent of tumor invasion in the ampullary region.


   ACKNOWLEDGMENTS
 
We thank Ichiro Tsuji, MD, professor of the Department of Public Health and Forensic Medicine, Tohoku University Graduate School of Medicine, Sendai, Japan, for statistical assistance and Taiki Chiba, RT, for technical assistance with this manuscript.


   FOOTNOTES
 
Author contributions: Guarantor of integrity of entire study, R.S.; study concepts and design, R.S.; literature research, R.S., S.T.; clinical data, R.S., K.I.; experimental studies, R.S., K.I.; data acquisition, R.S., K.I.; data analysis/interpretation, R.S., A.F., R.I.; statistical analysis, R.S.; manuscript preparation, R.S.; manuscript definition of intellectual content and editing, R.S., S.T., N.F.; manuscript revision/review and final version approval, all authors


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 

  1. Chan C, Herrera MF, Garza L, et al. Clinical behavior and prognostic factors of periampullary adenocarcinoma. Ann Surg 1995; 222:632-637.[Medline]
  2. Sarmiento JM, Nagorney DM, Sarr MG, Farnell MB. Periampullary cancers: are there differences? Surg Clin North Am 2001; 81:543-555.[CrossRef][Medline]
  3. Jones BA, Langer B, Taylor BR, Girotti M. Periampullary tumors: which ones should be resected? Am J Surg 1985; 149:46-52.[CrossRef][Medline]
  4. Coppola R, Riccioni ME, Ciletti S, et al. Periampullary tumors: analysis of 319 consecutive cases submitted to preoperative endoscopic biliary drainage. Surg Endosc 2001; 15:1135-1139.[CrossRef][Medline]
  5. Tomazic A, Pegan V. Preoperative staging of periampullar cancer with US, CT, EUS, and CA 19–9. Hepatogastroenterology 2000; 47:1135-1137.[Medline]
  6. Yeo CJ, Sohn TA, Cameron JL, Hruban RH, Lillemoe KD, Pitt HA. Periampullary adenocarcinoma: analysis of 5-year survivors. Ann Surg 1998; 227:821-831.[CrossRef][Medline]
  7. Itoh A, Goto H, Naitoh Y, Hirooka Y, Furukawa T, Hayakawa T. Intraductal ultrasonography in diagnosing tumor extension of cancer of the papilla of Vater. Gastrointest Endosc 1997; 45:251-260.[CrossRef][Medline]
  8. Itoh A, Tsukamoto Y, Naitoh Y, et al. Intraductal ultrasonography for the examination of duodenal papillary region. J Ultrasound Med 1994; 13:679-684.[Abstract]
  9. Palazzo L. Staging of ampullary carcinoma by endoscopic ultrasonography. Endoscopy 1998; 30(suppl 1):A128-A131.
  10. Semelka RC, Kelekis NL, John G, Ascher SM, Burdeny D, Siegelman ES. Ampullary carcinoma: demonstration by current MR techniques. J Magn Reson Imaging 1997; 7:153-156.[Medline]
  11. Lopez Hanninen E, Amthauer H, Hosten N, et al. Prospective evaluation of pancreatic tumors: accuracy of MR imaging with MR cholangiopancreatography and MR angiography. Radiology 2002; 224:34-41.[Abstract/Free Full Text]
  12. Freeny PC. Computed tomography in the diagnosis and staging of cholangiocarcinoma and pancreatic carcinoma. Ann Oncol 1999; 10(suppl 4):12-17.
  13. Campbell WL, Ferris JV, Holbert BL, Thaete FL, Baron RL. Biliary tract carcinoma complicating primary sclerosing cholangitis: evaluation with CT, cholangiography, US, and MR imaging. Radiology 1998; 207:41-50.[Abstract/Free Full Text]
  14. Soto JA, Alvarez O, Lopera JE, Munera F, Restrepo FC, Correa G. Biliary obstruction: findings at MR cholangiography and cross-sectional MR imaging. RadioGraphics 2000; 20:353-366.[Abstract/Free Full Text]
  15. Irie H, Honda H, Shinozaki K, et al. MR imaging of ampullary carcinomas. J Comput Assist Tomogr 2002; 26:711-717.[CrossRef][Medline]
  16. Lepanto L, Arzoumanian Y, Gianfelice D, et al. Helical CT with CT angiography in assessing periampullary neoplasms: identification of vascular invasion. Radiology 2002; 222:347-352.[Abstract/Free Full Text]
  17. Kim JH, Kim MJ, Chung JJ, Lee WJ, Yoo HS, Lee JT. Differential diagnosis of periampullary carcinomas at MR imaging. RadioGraphics 2002; 22:1335-1352.[Abstract/Free Full Text]
  18. Van Hoe L, Gryspeerdt S, Vanbeckevoort D, et al. Normal Vaterian sphincter complex: evaluation of morphology and contractility with dynamic single-shot MR cholangiopancreatography. AJR Am J Roentgenol 1998; 170:1497-1500.[Abstract/Free Full Text]
  19. Koike S, Ito K, Honjo K, Takano K, Yasui M, Matsunaga N. Oddi sphincter and common channel: evaluation with pharmacodynamic MR cholangiopancreatography using fatty meal and secretin stimulation. Radiat Med 2000; 18:115-122.[Medline]
  20. Kim JH, Kim MJ, Park SI, et al. Using kinematic MR cholangiopancreatography to evaluate biliary dilatation. AJR Am J Roentgenol 2002; 178:909-914.[Abstract/Free Full Text]
  21. Shoup M, Hodul P, Aranha GV, et al. Defining a role for endoscopic ultrasound in staging periampullary tumors. Am J Surg 2000; 179:453-456.[CrossRef][Medline]
  22. Mukai H, Yasuda K, Nakajima M. Tumors of the papilla and distal common bile duct: diagnosis and staging by endoscopic ultrasonography. Gastrointest Endosc Clin N Am 1995; 5:763-772.[Medline]
  23. Yamada I, Saito N, Takeshita K, et al. Early gastric carcinoma: evaluation with high-spatial-resolution MR imaging in vitro. Radiology 2001; 220:115-121.[Abstract/Free Full Text]
  24. Auh YH, Lim TH, Lee DH, et al. In vitro MR imaging of the resected stomach with a 4.7-T superconducting magnet. Radiology 1994; 191:129-134.[Abstract/Free Full Text]
  25. Imai Y, Kressel HY, Saul SH, et al. Colorectal tumors: an in vitro study of high-resolution MR imaging. Radiology 1990; 177:695-701.[Abstract/Free Full Text]
  26. Holland AE, Hendrick RE, Jin H, Russ PD, Barentsz JO, Holland R. Correlation of high-resolution breast MR imaging with histopathology: validation of a technique. J Magn Reson Imaging 2000; 11:601-606.[CrossRef][Medline]
  27. Goodwin DW, Zhu H, Dunn JF. In vitro MR imaging of hyaline cartilage: correlation with scanning electron microscopy. AJR Am J Roentgenol 2000; 174:405-409.[Abstract/Free Full Text]
  28. Sobin LH, Wittekind CH, eds. International Union Against Cancer (UICC): TNM classification of malignant tumors 6th ed. New York, NY: Wiley-Liss, 2002.
  29. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-174.[CrossRef][Medline]
  30. Bohra AK, McKie L, Diamond T. Transduodenal excision of ampullary tumours. Ulster Med J 2002; 71:121-127.[Medline]
  31. Beger HG, Treitschke F, Gansauge F, Harada N, Hiki N, Mattfeldt T. Tumor of the ampulla of Vater: experience with local or radical resection in 171 consecutively treated patients. Arch Surg 1999; 134:526-532.[Abstract/Free Full Text]
  32. Bakaeen FG, Murr MM, Sarr MG, et al. What prognostic factors are important in duodenal adenocarcinoma? Arch Surg 2000; 135:635-642.[Abstract/Free Full Text]
  33. Ahn YJ, Kim SW, Park YC, Jang JY, Yoon YS, Park YH. Duodenal-preserving resection of the head of the pancreas and pancreatic head resection with second-portion duodenectomy for benign lesions, low-grade malignancies, and early carcinoma involving the periampullary region. Arch Surg 2003; 138:162-168.[Abstract/Free Full Text]
  34. Fong Y, Blumgart LH, Lin E, Fortner JG, Brennan MF. Outcome of treatment for distal bile duct cancer. Br J Surg 1996; 83:1712-1715.[Medline]
  35. Thickman DI, Kundel HL, Wolf G. Nuclear magnetic resonance characteristics of fresh and fixed tissue: the effect of elapsed time. Radiology 1983; 148:183-185.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 RadioGraphicsHome page
S. Kim, N. K. Lee, J. W. Lee, C. W. Kim, S. H. Lee, G. H. Kim, and D. H. Kang
CT Evaluation of the Bulging Papilla with Endoscopic Correlation
RadioGraphics, July 1, 2007; 27(4): 1023 - 1038.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2313030797v1
231/3/767    most recent
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