| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (F.C.S., V.V., G.B., M.P.V., Y.M.), Gastroenterology (P.H.), and Hepatobiliary and Digestive Surgery (A.S.), Hôpital Beaujon, Clichy, France. Received August 7, 2002; revision requested October 2; revision received October 27; accepted January 15, 2003. Address correspondence to G.B., Department of Radiology, University of Palermo, Via Villaermosa 29, 90139 Palermo, Italy (e-mail: gbranca@yahoo.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: CT findings in 33 patients (24 women, nine men; age range, 18–84 years; mean age, 41 years) with unilocular pancreatic lesions (macrocystic serous cystadenoma, n = 12; mucinous cystadenoma, n = 11; pseudocyst, n = 10) were retrospectively and jointly reviewed by two blinded observers. Twenty-three patients underwent helical CT, which included pancreatic and portal venous phase imaging with delays of 40 seconds and 65 seconds, respectively, after contrast material injection. Ten patients underwent conventional (nonhelical) CT. The number, size, location, and contour of lesions were reviewed, along with wall thickness and enhancement and other signs of pancreatic and peripancreatic involvement. Diagnosis was based on lesion resection (n = 22) or on a combination of cytologic findings, biochemical markers, and tumor markers (n = 11). The Fisher exact test was used to analyze the results.
RESULTS: Three of four CT findings were independently specific for macrocystic serous cystadenoma: location in the pancreatic head, lobulated contour, and absence of wall enhancement. When two of these four criteria were used in combination, 83% (10 of 12) of patients with unilocular macrocystic serous cystadenoma were identified. When three or four of these criteria were used, a specificity of 100% was achieved. Location in the pancreatic head (P < .05), lobulated contour (P < .005), and lack of wall enhancement (P < .005) were specific for macrocystic serous cystadenoma in comparison with mucinous cystic tumor. Lobulated contours (P < .005) were specific for macrocystic serous cystadenoma in comparison with pseudocyst. Other CT findings were not helpful in distinguishing between the three types of lesions.
CONCLUSION: A combination of CT findings is helpful in making the diagnosis of pancreatic unilocular macrocystic serous cystadenoma.
© RSNA, 2003
Index terms: Pancreas, CT, 770.1211, 770.12112, 770.12115 • Pancreas, cysts, 770.3122, 770.3123 • Pancreas, neoplasms, 770.31
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Since serous cystadenomas may demonstrate macroscopic variation in locule size, the World Health Organization has subclassified them (4) as serous microcystic adenomas and serous oligocystic adenomas. At examination of the cut surface, however, serous oligocystic adenomas may occasionally reveal only one macroscopically visible cyst. This corresponds to a lesion described in 1992 by Lewandrowski et al (5) that is the unilocular variant of macrocystic serous cystadenoma. Since then, many authors have reported cases of macrocystic serous cystadenoma that were misdiagnosed as pseudocysts or mucinous cystadenoma and were therefore inappropriately managed (3,6–12). All of these authors concluded that computed tomography (CT) is too inaccurate to aid in the differential diagnosis of cystic lesions of the pancreas, especially in differentiating macrocystic serous tumors from mucinous cystic tumors. In these reports, the most challenging differential diagnosis was the unilocular variant of macrocystic serous cystadenoma.
Although some articles have been published on unilocular macrocystic serous cystadenoma (5,13–17), no series to our knowledge has demonstrated the imaging features. The lack of a radiologic study may be a possible explanation for the poor understanding of the imaging findings and, therefore, the difficulties in diagnosis.
We suspected that by reviewing the cases of a relatively large number of patients with unilocular macrocystic serous cystadenoma and comparing the findings in this cohort with findings in two groups of patients who had the two other most commonly encountered unilocular pancreatic cystic lesions, we might be able to detect specific signs that could be of use in the differential diagnosis of cystic pancreatic lesions. Thus, the purpose of our study was to compare the CT appearance of pancreatic unilocular macrocystic serous cystadenoma, of mucinous cystadenoma, and of pseudocyst to determine if there are CT findings that may assist in the differential diagnosis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
There were 12 patients with unilocular macrocystic serous cystadenoma (10 women, two men; age range, 35–64 years; mean age, 50 years). In each patient, the mass was detected because of abdominal pain (n = 9) or was an incidental finding (n = 3). There were 11 women with mucinous cystadenoma (age range, 18–74 years; mean age, 46 years). The lesion was detected because of abdominal pain (n = 6) or was an incidental finding (n = 5). None of the patients with cystic neoplasms had a history of pancreatitis or diabetes mellitus. Finally, there were 10 patients with pseudocysts (seven men, three women; age range, 28–74 years; mean age, 46 years). Seven of these patients had a history of alcoholic pancreatitis. All 10 patients were symptomatic (abdominal pain, n = 9; thoracic pain, n = 1).
CT Protocols
In all patients, CT examinations of the pancreas included nonenhanced and contrast material–enhanced imaging. Helical CT was performed in 23 patients; imaging was performed during a late arterial phase referred to as the pancreatic phase (18) (section thickness, 2.5 mm; pitch, 1.5) and during the portal venous phase (section thickness, 5.5 mm; pitch, 1.5), with delays of 40 seconds and 65 seconds, respectively, after initiation of the intravenous bolus injection of contrast material. In our institution we routinely perform this protocol for the evaluation of all pancreatic masses. The reconstruction overlap was 3 mm in the pancreatic phase and 6 mm in the portal phase; the latter enabled the exploration of the entire upper abdomen. Ten patients underwent only conventional (nonhelical) CT.
Nonenhanced CT was followed by contrast-enhanced CT. Contiguous 5-mm-thick scans were obtained at both nonenhanced and portal phase imaging. Patients received 120 mL of iodinated contrast material (iohexol, Omnipaque; Nycomed, Cork, Ireland; 350 mg of iodine per milliliter) intravenously. For multiphasic helical CT examinations, contrast material was administered at a rate of 3 mL/sec with use of a power injector (Medrad, Pittsburgh, Pa). For conventional CT examinations, contrast material was administered at a rate of 2 mL/sec. All CT examinations were performed with a Twin Flash (Marconi Medical Systems, Cleveland, Ohio) or DRG1 (Siemens, Erlangen, Germany) scanner.
Image Analysis
The CT scans were selected by a study coordinator who was not involved in the retrospective blinded interpretation of hard-copy images that was to be performed by consensus of two experienced abdominal radiologists (M.P.V. and V.V., 17 and 15 years of experience, respectively). These radiologists were provided the diagnosis of "cystic lesion of the pancreas" but were blinded to the specific diagnosis and clinical information. The number, size, and location (head, body or tail, or diffuse if all parts of the pancreas were involved) of the cystic lesions were evaluated. The nature of the lesion contour (round or ovoid, or lobulated) was noted. Lobulation was defined as the presence of rounded contours that could not be described as the borders of the same circle. The wall of the cyst was considered thin if it was 2 mm or less, and it was considered thick if it was more than 2 mm in diameter for at least 25% of the lesion circumference. Enhancement of the wall was regarded as absent or present on images obtained after contrast enhancement.
Images obtained in the pancreatic phase and portal venous phase were evaluated together. Subjective visual criteria were used rather than region-of-interest measurements. The content of the cyst was classified as homogeneous or heterogeneous. The presence of calcifications of the wall, mural nodules, or septa was noted. The common bile duct was considered abnormally dilated if the diameter was more than 6 mm. The presence and distribution of pancreatic intraductal calcifications (defined as areas of hyperattenuation on nonenhanced images) were recorded. The main pancreatic duct was considered abnormally dilated if the diameter was more than 3 mm at the pancreatic head and 2 mm in the body and tail. Peripancreatic abnormalities that were evaluated included lymphadenopathy and varices. Peripancreatic lymphadenopathy was diagnosed when ovoid or round extravisceral masses were identified that had short-axis diameters of 10 mm or more and attenuation less than or equal to that of skeletal muscle. The peripancreatic adipose tissue was considered infiltrated if visually perceptible increased attenuation relative to that of subcutaneous fat was present. Varices were defined as abnormally enlarged collateral veins.
Histopathologic Analysis
Color slides of the cut surface of the resected gross specimen and/or paraffin-embedded tissue section were reviewed by an experienced pathologist. In patients who did not undergo resection, diagnosis was based on analysis of the cyst fluid, which included analysis of enzyme levels, tumor markers, and cytologic findings (19,20). Fluid was collected by using fine-needle aspiration with a 19-gauge needle and CT guidance. After centrifugation, the aliquots of cyst fluid were frozen at -30°C until analysis. Biochemical markers and tumor markers were measured in cyst fluids. The upper limit of the normal range for amylase was 5,000 U/L (19). The maximal serum reference values for carcinoembryonic antigen and CA 72-4 were 5 ng/mL (5 µg/L) and <40 U/mL, respectively (19,20). Cytologic examination was performed in all cases. Presence of acute inflammation and histiocytes was suggestive of pseudocyst. Cuboidal epithelial cells containing cytoplasmic glycogen were characteristic of unilocular macrocystic serous cystadenoma. Epithelial cells with cytoplasmic mucin led to the diagnosis of mucinous cystadenoma (19). Combination of typical features of biochemical analysis, tumor markers, and cytologic analysis was considered sufficient to establish a specific diagnosis.
Statistical Analysis
Sensitivity and specificity values of the CT criteria were calculated. Differences in numbers between the three groups were compared by using the Fisher exact test. Numbers used in this statistical analysis were numbers of tumors. A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Imaging Findings
Table 1 summarizes the different imaging features observed in patients with unilocular macrocystic serous cystadenoma, mucinous cystadenoma, and pseudocyst. All patients had solitary cystic lesions.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
To our knowledge, the research that currently exists regarding the role of CT in the diagnosis of unilocular macrocystic serous cystadenoma is limited to a few case reports (14–17). An "irregularly shaped" unilocular serous cystadenoma was reported by Jin et al (14), and an "irregular external surface" was described by Fujiwara et al (16). Moreover, most of the unilocular macrocystic serous cystadenomas reported had thin walls (14–17) and were located in the head of the pancreas (13,15). However, the question of whether these findings or their combination were of any value in the diagnosis of unilocular macrocystic serous cystadenoma remained largely unanswered, because there was a lack of formal study results describing unilocular macrocystic serous cystadenomas and directly correlating them with other unilocular cystic lesions. Therefore, neither the CT discriminatory criteria nor their clinical usefulness was assessed. These difficulties in diagnosis are reflected by the large number of surgery studies in which, on the basis of CT scans, unilocular macrocystic serous cystadenoma had been preoperatively diagnosed as mucinous cystadenoma or pseudocyst because of its macrocystic appearance (3,5–12).
Some investigators have evaluated the performance of different readers using cross-sectional imaging to evaluate cystic lesions of the pancreas; the results have been controversial. Johnson et al (21) assessed the ability of CT to enable discrimination between serous and mucinous cystadenomas. They reported that by using the number (ie, six) or the size (2 cm in diameter at greatest dimension) of cysts as a cutoff, reviewers suggested the correct tumor type in 93% and 95% of cases, respectively. Conversely, Curry et al (22) reported that CT was insensitive in enabling differentiation of cystic pancreatic tumors. Reviewers correctly identified serous cystadenoma in only 23%–41% of cases (22). The greatest difficulty was identifying serous tumors that contained cysts larger than 2 cm, which mostly correspond to the macrocystic variant of serous cystadenoma (22). In the study by Curry et al (22), differentiation between serous and mucinous tumors was based on multiple criteria, including presence and location of calcifications and adenopathy. Using criteria similar to those adopted in the studies by Johnson et al (21) and Curry et al (22), Procacci et al (23) conducted a multireader blinded retrospective study to investigate the accuracy of CT in the characterization of cystic neoplasms. They found that only 60% of cystic tumors demonstrate imaging findings that allow accurate characterization, while a macrocystic unilocular pattern was "absolutely non-specific" and characterization was "almost impossible" in cases of unilocular cystic masses. These statements alone emphasize the need to establish reliable imaging criteria for the diagnosis of unilocular macrocystic serous cystadenoma. The criteria adopted by these authors (21–23) did not prove useful for lesion differentiation in our study. We were interested in the observation that other CT findings (lobulated contour, thin capsule, absence of wall enhancement, and location in the pancreatic head) were more frequently observed with unilocular macrocystic serous cystadenoma. When at least three of these four signs were present, diagnosis could be made with a specificity of 100%. We believe that the results of our study are encouraging, although further research with a larger series of patients is required to test if the findings we have noted can be successfully applied to similar patient populations.
Our study differs from previous research because we investigated CT findings that could facilitate differentiation of pancreatic cystic lesions that were unilocular. The inclusive criteria of our study made a bias in case selection unavoidable. However, we decided to focus on unilocular cystic lesions because these represent the most challenging cases to diagnose correctly, both in our experience and in the experience of other investigators (3,5–12).
None of the previous investigators included pseudocysts in a multireader blinded evaluation. Pseudocysts represent the most commonly encountered pancreatic cystic lesions (6), and misdiagnosis and mismanagement still occur frequently (6). Since there are patients (although very few) with pancreatitis who do not have abdominal pain (24) or an increase in enzyme levels (25), pseudocyst should be considered in the differential diagnosis of unilocular cystic pancreatic lesions; therefore, we decided to include pseudocyst in the design of our study.
Our purpose was to determine the CT imaging features of unilocular macrocystic serous cystadenoma, not to assess the ability of different phases of enhancement in the evaluation of cystic lesions of the pancreas; therefore, our readers reviewed both nonenhanced images and images obtained in different phases of enhancement together, as they would have done in a clinical setting. The pancreatic phase allowed us to evaluate the pancreatic parenchyma in the phase of maximal enhancement, while the portal venous phase allowed us to judge any involvement of adjacent vessels or to rule out metastatic liver disease.
Apart from the intrinsic limits of any retrospective study, several other limitations should be emphasized. First, we might have improved the validity of our results if we had relied on objective region-of-interest determinations rather than on subjective visual assessment of wall enhancement. However, the rarity of this lesion would not make a prospective study feasible for those working in an environment such as ours, in which evaluation must rely only on hard-copy film images. Second, we included what are by far the most frequently occurring cystic lesions of the pancreas, but we did not include lesions such as undifferentiated pancreatic carcinomas, cystic islet cell tumors, or congenital cysts (26). However, it is only recently that Bergin et al (27) described what, to their knowledge, is the first report in the radiology literature that evaluates three patients with congenital cysts. Moreover, in a retrospective review of 398 cystic pancreatic neoplasms, Le Borgne et al (11) showed cystic islet cell tumors to be only 2.5% of the total number of neoplasms, and the authors did not report any cases of undifferentiated pancreatic carcinomas or epithelial cysts in their series. Since our aim was to determine the imaging findings in a lesion that is virtually unknown to radiologists and to compare these findings with those in the two other most common pancreatic cystic lesions, it would be outside of the scope of this article to describe other rarely encountered lesions. Third, some of the cystic lesions in our series were not resected, and the diagnosis was established instead by analyzing a combination of biochemical markers, tumor markers, and cytologic findings. Although none of these represent a standard of reference, we and other investigators (19,20,28,29) believe that the combination of these criteria represent a compelling confirmation of diagnosis. Fourth, it could be argued that our study population is small. However, we are not aware of any study that focuses on the radiologic findings in a series of patients with unilocular macrocystic serous cystadenoma and that tries to define criteria that could allow differential diagnosis between the lesions we included in this study. Moreover, our results are statistically significant.
In conclusion, the results of our study show that certain findings (lobulated contour, absence of wall enhancement, and location in the pancreatic head) are observed more often in unilocular macrocystic serous cystadenoma than in mucinous cystadenoma and pseudocyst. The presence of a combination of CT findings is highly suggestive of macrocystic serous cystadenoma and can therefore be helpful in diagnosis.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-Y. Choi, M.-J. Kim, J. Y. Lee, J. S. Lim, J. J. Chung, K. W. Kim, and H. S. Yoo Typical and Atypical Manifestations of Serous Cystadenoma of the Pancreas: Imaging Findings With Pathologic Correlation Am. J. Roentgenol., July 1, 2009; 193(1): 136 - 142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Macari, M. E. Finn, G. L. Bennett, K. C. Cho, E. Newman, C. H. Hajdu, and J. S. Babb Differentiating Pancreatic Cystic Neoplasms from Pancreatic Pseudocysts at MR Imaging: Value of Perceived Internal Debris Radiology, April 1, 2009; 251(1): 77 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Macari, T. Lee, S. Kim, S. Jacobs, A. J. Megibow, C. Hajdu, and J. Babb Is Gadolinium Necessary for MRI Follow-Up Evaluation of Cystic Lesions in the Pancreas? Preliminary Results Am. J. Roentgenol., January 1, 2009; 192(1): 159 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Kim, D. H. Lee, Y. T. Ko, J. W. Lim, H. C. Kim, and K. W. Kim CT of Serous Cystadenoma of the Pancreas and Mimicking Masses Am. J. Roentgenol., February 1, 2008; 190(2): 406 - 412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D'Onofrio, A. J. Megibow, N. Faccioli, R. Malago, P. Capelli, M. Falconi, and R. P. Mucelli Comparison of Contrast-Enhanced Sonography and MRI in Displaying Anatomic Features of Cystic Pancreatic Masses Am. J. Roentgenol., December 1, 2007; 189(6): 1435 - 1442. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Visser, B. M. Yeh, A. Qayyum, L. W. Way, C. E. McCulloch, and F. V. Coakley Characterization of Cystic Pancreatic Masses: Relative Accuracy of CT and MRI Am. J. Roentgenol., September 1, 2007; 189(3): 648 - 656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Katz, D. M. Friedel, D. Kho, N. Georgiou, and J. J. Hines Relative Accuracy of CT and MRI for Characterization of Cystic Pancreatic Masses Am. J. Roentgenol., September 1, 2007; 189(3): 657 - 661. [Full Text] [PDF]
|
|
|
|
|
|
S. Y. Kim, J. M. Lee, S. H. Kim, K.-S. Shin, Y. J. Kim, S. K. An, C. J. Han, J. K. Han, and B. I. Choi Macrocystic neoplasms of the pancreas: CT differentiation of serous oligocystic adenoma from mucinous cystadenoma and intraductal papillary mucinous tumor. Am. J. Roentgenol., November 1, 2006; 187(5): 1192 - 1198. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. V. Sahani, R. Kadavigere, A. Saokar, C. Fernandez-del Castillo, W. R. Brugge, and P. F. Hahn Cystic Pancreatic Lesions: A Simple Imaging-based Classification System for Guiding Management RadioGraphics, November 1, 2005; 25(6): 1471 - 1484. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Aberle, C. Chiles, C. Gatsonis, B. J. Hillman, C. D. Johnson, B. L. McClennan, D. G. Mitchell, E. D. Pisano, M. D. Schnall, and A. G. Sorensen Imaging and Cancer: Research Strategy of the American College of Radiology Imaging Network Radiology, June 1, 2005; 235(3): 741 - 751. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. H. Kim, S. Saini, D. Sahani, P. F. Hahn, P. R. Mueller, and Y. H. Auh Imaging Diagnosis of Cystic Pancreatic Lesions: Pseudocyst versus Nonpseudocyst RadioGraphics, May 1, 2005; 25(3): 671 - 685. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Behl and W. C. Mundell 73-Year-Old Man With Vomiting and Fever Mayo Clin. Proc., October 1, 2004; 79(10): 1301 - 1304. [PDF]
|
|
|
|
 |
|
 M. D'Onofrio, S. Caffarri, G. Zamboni, M. Falconi, and G. Mansueto Contrast-Enhanced Ultrasonography in the Characterization of Pancreatic Mucinous Cystadenoma J. Ultrasound Med., August 1, 2004; 23(8): 1125 - 1129. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
