Indian Journal of Radiology Indian Journal of Radiology  

   Login   | Users online: 1062

Home Bookmark this page Print this page Email this page Small font sizeDefault font size Increase font size     


GASTROINTESTINAL Table of Contents   
Year : 2006  |  Volume : 16  |  Issue : 4  |  Page : 515-521
Pancreatic tumors: Prospective evaluation using MR imaging with MR cholangiography and MR angiography

Gb Pant Hospital , Maulana Azad Medical College , New Delhi (Delhi University), India

Click here for correspondence address and email

Date of Submission12-Jun-2006
Date of Acceptance08-Oct-2006


Objectives: To assess the efficacy of MR imaging, with MR Cholangiography and MR Angiography in pre operative evaluation of pancreatic tumors. Materials and Methods: 30 patients suspected of having pancreatic tumors underwent MR imaging with a 1.5T unit. Images were prospectively analyzed results correlated with surgery and biopsy. Results: Of the total 30 patients, 24 were histologically proven to be malignant of which 21 underwent surgery. MR characterization - benign from malignant lesions resulted in 27 correct diagnosis (accuracy 90%). 2 patients with histologically confirmed chronic pancreatitis were wrongly considered malignant. Among patients with histologically proven malignancy (n=24), MR imaging yielded correct diagnosis in 23 of 24 patients a sensitivity of 96 %. Overall accuracy for determining vascular involvement in patients who underwent surgery (n=21) was 86% (18 out of 21) a sensitivity of 75% (6 of 8). Sensitivity for detection of lymphadenopathy and liver metastasis was only 50% and 60% respectively. Among patients with final diagnosis of malignancy (n=24) who underwent surgery (n=21), overall accuracy for determining tumor nonresectabilty was 81 % (17 out of 21). Conclusion: A comprehensive combination of MR imaging sequences with MRCP and MR Angiography offers excellent potential to depict, characterize and preoperatively determine the resectability of pancreatic masses.

Keywords: MR Cholangiopancreatography, MR Angiography, Pancreatic tumors

How to cite this article:
Chhibber S, Sharma A K, Kumar N, Ghumman S, Puri S K. Pancreatic tumors: Prospective evaluation using MR imaging with MR cholangiography and MR angiography. Indian J Radiol Imaging 2006;16:515-21

How to cite this URL:
Chhibber S, Sharma A K, Kumar N, Ghumman S, Puri S K. Pancreatic tumors: Prospective evaluation using MR imaging with MR cholangiography and MR angiography. Indian J Radiol Imaging [serial online] 2006 [cited 2021 Jan 18];16:515-21. Available from:

   Introduction Top

A wide variety of neoplasm arises in the exocrine and endocrine tissue of the pancreas and may be benign or malignant in behavior. The challenges in imaging these tumors lie in early and accurate diagnosis, precise localization, determination of the extent and the potential for resectability.

About 90% of pancreatic tumors are adenocarcinomas with a ductal phenotype. About 70% of adenocarcinomas arise in the head, neck and uncinate process; 20% arise in the body and 5 to 10% in the tail.

Ductal adenocarcinoma of the pancreas is typically an unencapsulated solid tumor with an associated desmoplastic reaction. Adenocarcinomas of the pancreas tend to constrict or obstruct the adjacent duct or blood vessel [1].

Distinguishing between pancreatic carcinoma and chronic pancreatitis can prove to be a considerable diagnostic dilemma on any imaging modality.

Recent publications have documented the specific indications of MR in staging of pancreatic carcinoma to avoid under staging on basis of CT examination.

The purpose of our study was to prospectively evaluate the pre-operative accuracy of MR imaging, MRCP and MR angiography in evaluation of patients suspected of having pancreatic tumors.

   Materials and Methods Top

Out of the total 30 patients of pancreatic tumors who had got MRI done at GB Pant Hospital, New Delhi from 2001-2004, 21 underwent surgery. 19 patients were males and 11 were females with a mean age of 55 years. Histological diagnosis was adenocarcinoma in 18 cases, while there was one case each of neuroendocrine tumor, mesenchymal tumor [Figure - 4]

cystic tumor [Figure - 5],, lymphoma, papillary tumor and periampullary carcinoma. Benign findings included chronic pancreatitis (n=4) [Figure - 6] and````` other benign pancreatic adenomas (n=2).

Imaging Techniques

All MR studies were performed on 1.5 (Gyroscan; Philips) system with a standard phased array surface coil .
" T2 TSE axial [TE-60, TR-776, Slice thickness-7mm ]
" T1FFE axial [TE- 4.6 , TR-183, Flip angle- 80, Slice thickness-7mm ]
" B-TFE transverse , coronal [TE-1.68, TR-3.4, Flip angle-80, Slice thickness-8mm ]
" THRIVE transverse [TE-2.4, TR-4.8, Flip angle-10, Slice thickness-2mm]

For these sequences the FOV was 375 mm; and matrix, 256 x 256

MRCP was performed with T2 weighted turbo spin-echo sequence [TE-800, TR-8000, Turbo factor 205; matrix, 256x 256; section thickness 40mm ] in twelve coronal oblique orientations (-30 through +30)

Image analysis

MR images were prospectively evaluated. All components - (MR imaging, MRCP, and MR Angiography) were evaluated. The size and location of the tumor was documented. In addition the signal intensity pattern of the lesion for the different sequences was determined.

Evaluation included the assessment of loco-regional tumor extent. Peripancreatic tumor growth was diagnosed with irregular signal of the peripancreatic fat or infiltration of the adjacent structures (e.g. duodenum, colon, stomach). Lymph nodes with short axis diameter greater than 1cm were considered metastatic. Presence of peritoneal deposits and mesenteric implants were assessed. In addition, focal liver lesion were documented and categorized as benign or metastatic.

For assessment of the pancreaticobiliary system, MRCP images were interpreted and assessed for the presence of pathologic morphologic features. The presence of dilatation, stenosis or obstruction was determined.

Vascular infiltration included identification of the celiac axis, hepatic artery, splenic artery, superior mesenteric artery, the portal vein, the superior mesenteric vein, and the splenic vein. Vascular tumor infiltration was assumed on the basis
" Vessel encasement (>90 degrees)
" Vessel occlusion
" Vessel caliber reduction
" Vessel flattening

Final diagnosis of the tumor were categorized as

Definitely benign: No delineable solid tumor, focal mass lesion with relatively high signal on T1, intrapancreatic and peripancreatic fluid collections, dilated pancreatic duct with intraductal calculi, MRCP images- irregular dilatation of the MPD, side branches with intraductal calculi, duct penetrating sign -dilated ducts seen to penetrate the mass lesion.

Indeterminate: findings could not be classified as benign or malignant.

Malignant: A delineable solid lesion distorting the lobular contour of the gland, upstream dilation of the pancreatic duct, metastasis, vascular involvement, lymphadenopathy, invasion of adjacent organs, MRCP images were consistent with pancreaticobiliary duct stenosis or obstruction.

Finally the tumors were assessed for there resectability.

Criteria for unresectability
" Metastasis
" Peripancreatic arterial and or venous involvement >90 degrees
" Lymphadenopathy (short axis diameter>1cm)

Involvement of the adjacent organs was not considered criteria for unresectability.

On the basis of these findings, tumors were classified as Resectable or Unresectable.

All the patients with pathologically proven carcinoma were subjected to surgery except those patients who had metastasis to liver or definite vascular involvement greater than 180.

MR results were correlated with results from surgery and histopathological analysis.

   Result Top

A total of 30 patients were included in the study. 19 cases were males and 11 were females with an age range of patient between 21-80 years.

Based on imaging findings a malignant diagnosis was determined in 24 and benign diagnosis in 6 patients. Histopathological diagnosis was obtained in all 30 patients. 21 of the 24 patients with histologically proven malignancy underwent surgery.

Overall assessment of pancreatic diseases in 30 patients resulted in 27 correct diagnoses (accuracy of 90%). Among patients with final histopathological diagnosis of malignancy MR imaging yielded correct diagnosis in 23 of 24 patients.

1 histopathological proven patient of adenocarcinoma was misdiagnosed as a case of chronic pancreatitis on MR imaging.

2 patient with histologically confirmed chronic pancreatitis was misdiagnosed as malignant.

All 24 pancreatic tumors were hypo/isointense on T1 GRE images and fat suppressed THRIVE images.

The tumor appeared hypointense in majority 75 % of the cases, but 6 cases demonstrated tumoral hyper intensity on T2WI.

21 of the 24 histologically proven patients who underwent surgery, 8 patients had surgically proven vascular infiltration, and 6 of these patients were correctly depicted on MR imaging. [Figure - 1],[Figure - 2]

MR underestimated vascular involvement in 2 case which subsequently proved to be unresectable during surgery due to vascular involvement.

MR overestimated vascular involvement in 1case .

5 of the 8 patients were diagnosed to have lymphadenopathy preoperatively on MR imaging.

Duodenal invasion was correctly detected by MRI in 5 of the 6 cases that showed involvement at surgery with a sensitivity of 83 %

Per operatively hepatic metastasis were seen in 5 patients and MR imaging correctly detected 3 cases. [Figure - 3]


MR overestimated resectability in 3* cases, missing subcentimeter hepatic metastasis, misdiagnosing a case of adenocarcinoma as chronic pancreatitis, and underestimating vascular involvement in 1 case each.


   Discussion Top

Ductal adenocarcinoma represents the most common pancreatic malignancy with a poor prognosis. In patients suspected of having pancreatic tumors, various investigators have addressed the value of different imaging modalities like US, CT, MR imaging, Angiography, ERCP, PET for tumor detection and evaluation of resectability.

The current study was initiated to prospectively determine the accuracy of a single non invasive modality in the clinical work up of patients suspected of having pancreatic tumors with regard to lesion detectability, characterization of lesion status ( malignant vs. benign), and determination of resectability of the lesion.

In our study, the overall accuracy in characterization of lesion status was calculated at 90%. As previously described, the presence of chronic pancreatitis represents an extremely difficult challenge in differential diagnosis of pancreatic cancer. In the present study 2 cases of histologically proven chronic pancreatitis were misdiagnosed as malignant on imaging.

One case of adenocarcinoma was misdiagnosed as a case of chronic pancreatitis. These findings were consistent with previous studies [2],[3],[4] describing the accuracy of MR imaging in characterizing the lesion.

The signal intensity characteristics of pancreatic adenocarcinomas were in concordance with earlier studies [3], where the tumor appeared predominantly hypointense on T1FFE and T2WI. However, a small proportion, 25 % of patients with pancreatic adenocarcinoma exhibited a mildly hyperintense or mixed signal intensity on T2WI. Our results were in agreement with Jenkins et al [5] who concluded that there is no statistically significance difference in T1 and T2 relaxation time between chronic pancreatitis and pancreatic cancer.

All the tumors were found to be hypointense to the pancreas on Fat suppressed (THRIVE) images. Peripancreatic fat planes appeared less distinct on Fat suppressed images and therefore, evaluation of vascular invasion was difficult as compared to T1WI GRE images. [6]

On Post Gadolinium administration tumors showed weak enhancement due to their associated desmoplastic reaction, thus appearing hypointense to the enhancing pancreas.

We obtained contrast enhanced T1WI GRE images after the acquisition of MR Angiographic sequence , thus it remained unanswered whether or not a different MR imaging approach would have substantially contributed to a correct diagnosis in the misdiagnosed cases on imaging.

Pamela T Johnson et al [7] where compared the pancreatic carcinoma with chronic pancreatitis on dynamic MR imaging.

The majority of the patients with pancreatic carcinoma had ductal stenosis or obstruction with consequential dilatation of the pancreatic duct in the body or tail, with and without dilatation of the common bile duct. In cases of chronic pancreatitis there was dilatation of the main pancreatic duct along with dilatation of the side branches with few cases showed intraductal calculi. The dilated pancreatic duct was seen to "penetrate" through the inflammatory mass in cases with chronic pancreatitis as described by Ichikawa et al [8].

In the present study the overall accuracy for determining vascular involvement was 86% (18 out of 21). Sensitivity was 75% (6 of 8); specificity was 92% (12 of 13). These results were consistent with previous studies [2],[3],[9],[10],[11] assessing the accuracy of MR imaging in determining vascular involvement.

In agreement with the results of other investigators[12]; most MRI errors were due to "underestimation" of vascular involvement. The problem of underestimation of vascular tumor invasion was also found with other imaging modalities. Muller et al [13].

The actual degree of vascular invasion by pancreatic cancer is rarely "overstaged" with MRI, in our study this happened with one case. This suggests that MRI does not stage resecatble tumors as unresectable, which potentially denies the patient of curative treatment.

Overall accuracy for determining tumor nonresectabilty was 81 % (17 out of 21) and specificity was 92% (11 of 12). Similar results were obtained by other researchers[3],[7],[14]on reviewing the literature.

However in prediction of tumor non resectability, the sensitivity was only 67% (6 of 9), which indicates the limitations on diagnosing lymphadenopathy, detection of subcentimeter metastasis, peritoneal deposits and invasion of the mesentery. Fischer U et al [15] also had similar limitation in there comparative study in staging of patients with a suspected pancreatic mass with preoperative MRI

In conclusion, the results from this MR study indicate that a comprehensive combination of MR imaging sequences with MRCP and MR angiography offers excellent potential to depict, characterize and preoperatively determine the resectability of pancreatic lesions. In addition , when a lesion is deemed resectable , then the patient might be spared an unnecessary ERCP and stent placement along with attendant complications

Whereas the findings in minor proportion of patients with chronic pancreatitis provided diagnostic difficulties, local regional extension, metastatic dissemination and resectability of most pancreatic tumors were correctly classified

   References Top

1.Scott Gazelle , Sanjay Saini Hepatobiliary and Pancreatic Radiology 1998:784-785  Back to cited text no. 1    
2.Catalano C, Pavone P, Laghi A, Panebianco V, Scipioni A, Fanelli F, Brillo R, Passariello R. Eur Radiol. 1998;8(3):428-34  Back to cited text no. 2    
3.Lopez Hanninen E, Amthauer H, Hosten N, Ricke J, Bohmig M, Langrehr J, Hintze R, Neuhaus P, Wiedenmann B, Rosewicz S, Felix R. Radiology. 2002 Jul;224(1):34-41.  Back to cited text no. 3    
4.Tomakoi Ichikawa, Hiroki Hardome, Junichi Hachiya, Toshiaki Nitatori ET al.1997. Radiology 202:655-662  Back to cited text no. 4    
5.Jenkins JP, Braganza JM, Hickey DS, Isherwood I, Machinin M. 1987. Radiology 60:333-341  Back to cited text no. 5    
6.Gabata T, Matsui O, Kadoya M, Yoshikawa J, Miyayama S, Takashima T, Nagakawa T, Kayahara M, Nonomura A. Radiology. 1994 Dec;193(3):683-8  Back to cited text no. 6    
7.Hochwald SN, Rofsky NM, Dobryansky M, Shamamian P, Marcus SG. J Gastrointest Surg. 1999 Sep-Oct;3(5):506-11.  Back to cited text no. 7    
8.Ichikawa T, Sou H, Araki T, Arbab AS, Yoshikawa T, Ishigame K, Haradome H, Hachiya J. Radiology. 2001 Oct;221(1):107-16.  Back to cited text no. 8    
9.Gaa J, Wendl K, Tesdal IK, Meier-Willersen HJ, Lehmann KJ, Bohm C, Mockel R, Richter A, Trede M, Georgi M. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr. 1999 Jun;170(6):528-33.  Back to cited text no. 9    
10.Sironi S, De Cobelli F, Zerbi A, Angeli E, Balzano G, Taccagni G, Di Carlo V, Del Maschio A. AJR Am J Roentgenol. 1995 Oct;167(4):997-1001.  Back to cited text no. 10    
11.Arslan A, Buanes T, Geitung JT. Eur J Radiol. 2001 May;38(2):151-9.  Back to cited text no. 11    
12.Freeny PC, Marks WM, Ryan JA, Traverso LW.1988. Radiology 166:125-133  Back to cited text no. 12    
13.Muller MF , Meyenberger C, Bertschinger P, Schaer R et al 1994. Radiology 190:745-751  Back to cited text no. 13    
14.Schima W, Fugger R, Schober E, Oettl C, Wamser P, Grabenwoger F, Ryan JM, Novacek G. AJR Am J Roentgenol. 2002 Sep;179(3):717-24.  Back to cited text no. 14    
15.Fischer U, Vosshenrich R, Horstmann O, Becker H, Salamat B, Baum F, Grabbe E. Eur Radiol. 2002 Feb;12(2):296-303. Epub 2001 Oct 30.  Back to cited text no. 15    

Correspondence Address:
A K Sharma
Dept Of Radiodiagnosis, G.B Pant Hospital , New Delhi
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0971-3026.32260

Rights and Permissions


  [Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5], [Figure - 6]

  [Table - 1], [Table - 2]

This article has been cited by
1 Adult Bile Duct Strictures: Role of MR Imaging and MR Cholangiopancreatography in Characterization
Venkata S. Katabathina,Anil K. Dasyam,Navya Dasyam,Keyanoosh Hosseinzadeh
RadioGraphics. 2014; 34(3): 565
[Pubmed] | [DOI]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  

    Materials and Me...
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded374    
    Comments [Add]    
    Cited by others 1    

Recommend this journal