Paraduodenal Pancreatitis: Clinical Performance of MR Imaging in Distinguishing from Carcinoma
Abstract
Purpose
To evaluate the diagnostic performance of contrast material–enhanced magnetic resonance (MR) imaging for distinguishing paraduodenal pancreatitis (PDP) from pancreatic head duct adenocarcinoma (CA) in patients with diagnoses confirmed by histopathologic analysis.
Materials and Methods
This retrospective study was approved by the institutional review board and is HIPAA compliant. Between July 2007 and July 2010, 47 patients who underwent Whipple procedure and MR imaging less than 60 days before surgery were identified retrospectively. Two relatively inexperienced fellowship trainees with 9 months of body fellowship training were asked to record the presence or absence of three MR imaging features: focal thickening of the second portion of the duodenum; abnormal enhancement of the second portion of the duodenum; and cystic focus in the expected region of the accessory pancreatic duct. Strict criteria for diagnosis of PDP included presence of all three imaging features. Any case that did not fulfill the criteria was classified as CA. Sensitivity, specificity, positive predictive value, and negative predictive value for characterization of PDP was calculated for each reader with 95% confidence intervals. A κ test assessed level of agreement between readers.
Results
Each reader correctly categorized 15 of 17 (88.2%) PDP cases when all three imaging criteria were met. Alternatively, 26 of 30 (86.7%) pancreatic duct CA were correctly categorized as inconsistent with PDP. Four patients with histopathologic diagnosis of CA were incorrectly classified as PDP by each reader. Agreement between the two readers showed substantial κ agreement for the diagnosis of PDP and differentiation from pancreatic duct CA.
Conclusion
Contrast-enhanced MR imaging may help accurately identify PDP and distinguish it from CA when strict diagnostic criteria are followed.
© RSNA, 2013
Supplemental material: S1
Introduction
Chronic pancreatitis is a cause of abdominal pain that may occur as sequelae of multiple factors, including biliary stone disease, alcohol consumption, malignancy, metabolic disorders, and various genetic and environmental insults, including trauma (1). Paraduodenal pancreatitis (PDP), also known as groove pancreatitis (2,3), is a specific type of focal pancreatitis centered in the pancreaticoduodenal groove, which also involves the adjacent second portion of the duodenum and the pancreatic head (4). PDP reflects an inflammatory process that may result from obstruction of the accessory pancreatic duct as it drains into the second portion of the duodenum through the minor ampulla (4–6). Cystic change is a prominent feature of this process, likely related to accessory duct obstruction, and is frequently located in the expected region of the pancreatic accessory duct (7).
It is challenging to differentiate PDP from pancreatic head duct adenocarcinoma (CA), and published work on this subject emphasizes this point (7–9). However, optimum therapy for both entities may be similar and require pancreaticoduodenectomy (ie, Whipple procedure) as a treatment endpoint. Casetti et al (10) reported excellent clinical results in patients with PDP who underwent Whipple procedure, with resolution of pain in 76% of patients. Similar clinical outcome data were reported from our center: patients who underwent Whipple procedure for PDP reported improved quality of life scores after intervention (11). While surgical therapy may ultimately be required in both patient groups, accurate preoperative diagnosis is important to guide preoperative management and patient counseling.
Needle tissue samples are challenging to interpret; the relatively small sample volume may not adequately exclude the presence of neoplasm (8,12). Computed tomography (CT) provides superior spatial resolution, but lacks sufficient contrast resolution to reliably distinguish between cancer and focal inflammatory processes of the pancreas. The soft-tissue capabilities of magnetic resonance (MR) imaging may be preferred for the evaluation of the pancreas, specifically for delineation of inflammation and analysis of pancreatic cyst content and architecture (13–16).
Based on clinical experience at our center, we have observed imaging features on dynamic contrast agent–enhanced MR imaging that may reflect underlying histopathologic findings that characterize PDP and that are related to accessory duct obstruction. We postulate that these imaging features may be useful in differentiation of PDP from CA. Some authors have previously reported imaging findings that are suggestive of PDP (7,17), and they have emphasized the overlap of imaging features with carcinoma. Few studies (9,18) have systematically evaluated the imaging features of PDP compared with CA. The purpose of our study was to evaluate the diagnostic performance of contrast-enhanced MR imaging for distinguishing PDP from pancreatic head duct CA in patients with diagnoses confirmed by histopathologic analysis.
Materials and Methods
Patients
This retrospective study was approved by our institutional review board and was compliant with the Health Insurance Portability and Accountability Act, and all patients provided written informed consent for participation in the retrospective records review at the time of MR imaging. A search of our histopathologic electronic medical record database was performed by using the search terms “Whipple and/or pancreatectomy” to identify patients who had a Whipple pancreatectomy procedure between July 2007 and July 2010. Patients were identified and categorized into the following two cohorts: patients with a pathologic diagnosis of PDP (without evidence of carcinoma) and patients with a pathologic diagnosis of pancreatic duct CA. Inclusion criteria were patients with confirmed diagnosis of PDP or cancer based on histopathologic evaluation of the surgical specimen, and a preoperative MR examination performed at our institution within 60 days of surgery. Exclusion criteria were lack of institutional MR examination within 60 days of surgery. Nine patients (six patients from CA cohort and three patients from PDP cohort) were excluded for this reason (five men [mean age, 64 years; age range 53–71 years] and four women [mean age, 58 years; age range, 52–64 years]). Each patient’s social history regarding smoking (recorded as number of pack-years) and the presence of alcohol abuse (defined as >2 drinks per day, $5 days per week) were obtained by interview during the patient’s office visit before surgery.
MR Imaging
All MR examinations were performed on a 1.5-T MR system (Magnetom Avanto, Siemens Medical Solutions, Erlangen, Germany; or Signa HDxt, GE Healthcare, Milwaukee, Wis) by using corresponding multichannel torso phased-array surface receiver coil; details of pulse sequence parameters are in Table 1. The weight of each patient was recorded, and gadobenate dimeglumine (MultiHance; Bracco Diagnostics, Princeton, NJ) was administered at a dose of 0.05 mmol/kg and a rate of 2 mL/sec, followed by a 30-mL saline flush at 2 mL/sec with a dual chamber power injector (Spectris; Medrad, Warrendale, Pa). Arterial phase images were acquired by using a real-time bolus-tracking method, with breath-hold instructions and image acquisition initiated when the contrast agent bolus arrived at the celiac trunk. This method produces reproducible arterial phase images that are less sensitive to hemodynamic status of the patient. The time delay between bolus trigger point and acquisition initiation was 8 seconds by using methodologic processes previously reported (19). Venous and delayed phase imaging was initiated at 70 seconds and 180 seconds after the bolus trigger point.
 |
MR Imaging Analysis
All MR imaging data sets were made anonymous and transferred for review to a separate workstation (ClearCanvas Workstation v 2.0; ClearCanvas, Toronto, Canada). The image analysis protocol was designed to determine the ability of readers with limited experience to differentiate PDP from CA by using directed feature templates for PDP. The two readers (D.G. and S.H.E.) were board-certified radiologists who were within 1 year of certification and who had 9 months of abdominal fellowship training. Both reviewers were given lectures on the MR imaging features of PDP by two abdominal MR imaging experts (D.R.M. and B.K., 14 and 5 years of experience, respectively). Six MR examinations (three PDP, three CA), documented by using pathologic analysis, were shown to the readers as part of the training process; these studies were subsequently discarded and were not included in the final study population. Review sessions were performed by the two blinded reviewers independently and on different days. The reviewers understood that patients had either PDP or CA, but were blinded to the number of patients of each diagnosis and also the individual pathologic results in each case. Each reader was asked to record the presence or absence of the following three distinct imaging features: (a) focal thickening (>3 mm) of the second portion of the duodenum on coronal and axial T2-weighted images; (b) increased abnormal enhancement of the second portion of the duodenum (relative to proximal jejunal loops in the upper abdomen) on postgadolinium-chelate enhanced T1-weighted three-dimensional gradient-recalled echo images; and (c) a fluid cavity in the expected region of the accessory pancreatic duct or contained within the wall of the second portion of the duodenum at the orifice of the accessory duct. For testing diagnostic specificity, the presence of all three imaging features classified the case as PDP (Fig 1). Any case that did not have all three criteria was classified as CA for statistical analysis. These three imaging criteria were selected based on previously published findings (17,20) and from extensive experience with PDP at our center.
Figure 1a: PDP in a 45-year-old woman with abdominal pain. (a) Precontrast, (b) arterial, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) coronal single-shot T2-weighted MR images and (e) single-shot T2-weighted coronal cholangiopancreatographic coronal slab. Images show the three criteria evaluated in association with PDP: abnormal thickening of the second portion of the duodenum (a–e, arrow); abnormal enhancement of the second portion of the duodenum (a–c, arrow); and complex fluid cavity in the pancreatic head in the expected location of the pancreatic accessory duct (a–e, arrowhead).
Figure 1b: PDP in a 45-year-old woman with abdominal pain. (a) Precontrast, (b) arterial, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) coronal single-shot T2-weighted MR images and (e) single-shot T2-weighted coronal cholangiopancreatographic coronal slab. Images show the three criteria evaluated in association with PDP: abnormal thickening of the second portion of the duodenum (a–e, arrow); abnormal enhancement of the second portion of the duodenum (a–c, arrow); and complex fluid cavity in the pancreatic head in the expected location of the pancreatic accessory duct (a–e, arrowhead).
Figure 1c: PDP in a 45-year-old woman with abdominal pain. (a) Precontrast, (b) arterial, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) coronal single-shot T2-weighted MR images and (e) single-shot T2-weighted coronal cholangiopancreatographic coronal slab. Images show the three criteria evaluated in association with PDP: abnormal thickening of the second portion of the duodenum (a–e, arrow); abnormal enhancement of the second portion of the duodenum (a–c, arrow); and complex fluid cavity in the pancreatic head in the expected location of the pancreatic accessory duct (a–e, arrowhead).
Figure 1d: PDP in a 45-year-old woman with abdominal pain. (a) Precontrast, (b) arterial, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) coronal single-shot T2-weighted MR images and (e) single-shot T2-weighted coronal cholangiopancreatographic coronal slab. Images show the three criteria evaluated in association with PDP: abnormal thickening of the second portion of the duodenum (a–e, arrow); abnormal enhancement of the second portion of the duodenum (a–c, arrow); and complex fluid cavity in the pancreatic head in the expected location of the pancreatic accessory duct (a–e, arrowhead).
Figure 1e: PDP in a 45-year-old woman with abdominal pain. (a) Precontrast, (b) arterial, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) coronal single-shot T2-weighted MR images and (e) single-shot T2-weighted coronal cholangiopancreatographic coronal slab. Images show the three criteria evaluated in association with PDP: abnormal thickening of the second portion of the duodenum (a–e, arrow); abnormal enhancement of the second portion of the duodenum (a–c, arrow); and complex fluid cavity in the pancreatic head in the expected location of the pancreatic accessory duct (a–e, arrowhead).
Additional imaging features were also categorized and recorded for each patient. These features included: the presence or absence of pancreatic duct dilatation (>2 mm), extrahepatic biliary duct dilatation (>6 mm), stenosis of the main portal vein or portal venous confluence (>30% luminal stenosis), and the presence or absence of a fluid-filled cavity identified in the region of the accessory duct or duodenal wall.
Pathologic Analysis
Surgical specimens from Whipple procedure were fixed in 10% formalin solution and embedded in paraffin. Specimens were sectioned in 5-mm-thick sections in the coronal plane and stained with hematoxylin and eosin. Gross specimens were reviewed for features suggestive of PDP or CA. Features of PDP included duodenal scarring, cystic changes in the region of the minor ampulla, and extension of chronic inflammatory changes in the head of the pancreas. Specific attention was paid in each case to any histologic features suggestive of PDP, including changes of myoadenomyomatosis in the duodenal wall, myoid proliferation in the region of the minor ampulla, and Brunner gland hyperplasia. Features of CA included a proliferation of small tubular glands lined by cuboidal epithelium that appeared to be malignant within a desoplastic stroma. Each set of slides was retrospectively reviewed by a gastrointestinal pathologist (N.V.A., 15 years of experience specializing in hepatobiliary and pancreatic diseases). A final diagnosis of PDP or CA was rendered through a combination of gross and histologic analysis.
Statistical Analysis
Statistical software (SAS version 9.2; SAS Institute, Cary, NC) was used for all data analysis. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of MR imaging for the diagnosis of PDP were estimated as percentages. Corresponding 95% confidence intervals were calculated according to the efficient-score method (corrected for continuity) (21). The agreement level between readers was measured by using κ coefficient and was tested for discrepancy by using χ2 test. We defined κ values for level of agreement as follows: 0.81–0.99, almost perfect agreement; 0.61–0.80, substantial agreement; 0.41–0.60, moderate agreement; 0.21–0.40, fair agreement; and 0.01–0.20, slight agreement (22). A t test was used to compare differences between PDP and CA patients for a history of alcohol abuse, and for the presence or absence of the following imaging features: a cystic focus in the expected location of the pancreatic accessory duct, extrahepatic bile duct dilatation, pancreatic duct dilatation, and stenosis of the portal vein.
Results
Patients
Between July 2007 and July 2010, 47 patients (mean age, 60 years; age range, 25–85 years) were identified who fulfilled study inclusion criteria, and the following were included in the blinded image review: 28 men (mean age, 56 years; age range, 31–81 years) and 19 women (mean age, 63 years; age range, 35–85 years). Of these patients, 17 had a final histopathologic diagnosis of PDP (13 men and four women; mean age, 50 years; age range, 31–62 years), and 30 had a final histopathologic diagnosis of pancreatic duct CA (15 men and 15 women; mean age, 65 years; age range, 25–81 years); no patients had coexistent PDP and CA on histopathologic analysis. Average smoking consumption was 26 pack-years in the PDP subgroup versus 15 pack-years in the CA subgroup. In the PDP subgroup, a clinical history of alcohol abuse was elicited in 17 of 17 (100%) of patients, while in the CA subgroup this history was present in only three of 30 (10.0%) of patients (P < .001).
MR Imaging Analysis
Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of MR imaging for the diagnosis of PDP are shown in Table 2, with associated 95% confidence intervals. Overall, each reader correctly categorized 15 of 17 (88.2%) cases of PDP when all three imaging criteria were present by using the provided diagnostic templates (Fig 2). Alternatively, 26 of 30 (86.7%) patients with CA were correctly categorized as not consistent with PDP. Four patients with a histopathologic diagnosis of CA were incorrectly classified as PDP by each reader; however, only two of these patients were misclassified by both readers as having PDP (Fig 3).
 |
Figure 2a: PDP in a 50-year-old man with abdominal pain. (a) Precontrast, (b) arterial phase T1-weighted three-dimensional gradient-recalled echo, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) fat-suppressed single-shot T2-weighted MR images. Arrow shows duodenal thickening with abnormal enhancement. This patient also has a small complex fluid focus in the expected location of the accessory duct orifice that is most conspicuous on T1-weighted three-dimensional gradient-recalled echo images (arrowhead). The fat-suppressed single-shot T2-weighted image shows inflammation and edema in and around the second portion of the duodenum.
Figure 2b: PDP in a 50-year-old man with abdominal pain. (a) Precontrast, (b) arterial phase T1-weighted three-dimensional gradient-recalled echo, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) fat-suppressed single-shot T2-weighted MR images. Arrow shows duodenal thickening with abnormal enhancement. This patient also has a small complex fluid focus in the expected location of the accessory duct orifice that is most conspicuous on T1-weighted three-dimensional gradient-recalled echo images (arrowhead). The fat-suppressed single-shot T2-weighted image shows inflammation and edema in and around the second portion of the duodenum.
Figure 2c: PDP in a 50-year-old man with abdominal pain. (a) Precontrast, (b) arterial phase T1-weighted three-dimensional gradient-recalled echo, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) fat-suppressed single-shot T2-weighted MR images. Arrow shows duodenal thickening with abnormal enhancement. This patient also has a small complex fluid focus in the expected location of the accessory duct orifice that is most conspicuous on T1-weighted three-dimensional gradient-recalled echo images (arrowhead). The fat-suppressed single-shot T2-weighted image shows inflammation and edema in and around the second portion of the duodenum.
Figure 2d: PDP in a 50-year-old man with abdominal pain. (a) Precontrast, (b) arterial phase T1-weighted three-dimensional gradient-recalled echo, (c) delayed phase T1-weighted three-dimensional gradient-recalled echo, and (d) fat-suppressed single-shot T2-weighted MR images. Arrow shows duodenal thickening with abnormal enhancement. This patient also has a small complex fluid focus in the expected location of the accessory duct orifice that is most conspicuous on T1-weighted three-dimensional gradient-recalled echo images (arrowhead). The fat-suppressed single-shot T2-weighted image shows inflammation and edema in and around the second portion of the duodenum.
Figure 3a: Pancreatic duct CA in an 80-year-old woman with abdominal pain and jaundice, inaccurately diagnosed as PDP by both blinded readers. (a) Precontrast axial T1-weighted three-dimensional gradient-recalled echo, (b) delayed enhanced axial T1-weighted three-dimensional gradient-recalled echo, and (c) single-shot T2-weighted without and (d) with spectral selection attenuated inversion recovery fat-suppression MR images. These images demonstrate features suggestive of PDP, including abnormal thickening and enhancement of the second portion of the duodenum (arrow), with a complex cystic focus associated with the minor ampulla (arrowhead). Pathologic specimen analysis demonstrated invasive duct CA centered at the accessory ampulla.
Figure 3b: Pancreatic duct CA in an 80-year-old woman with abdominal pain and jaundice, inaccurately diagnosed as PDP by both blinded readers. (a) Precontrast axial T1-weighted three-dimensional gradient-recalled echo, (b) delayed enhanced axial T1-weighted three-dimensional gradient-recalled echo, and (c) single-shot T2-weighted without and (d) with spectral selection attenuated inversion recovery fat-suppression MR images. These images demonstrate features suggestive of PDP, including abnormal thickening and enhancement of the second portion of the duodenum (arrow), with a complex cystic focus associated with the minor ampulla (arrowhead). Pathologic specimen analysis demonstrated invasive duct CA centered at the accessory ampulla.
Figure 3c: Pancreatic duct CA in an 80-year-old woman with abdominal pain and jaundice, inaccurately diagnosed as PDP by both blinded readers. (a) Precontrast axial T1-weighted three-dimensional gradient-recalled echo, (b) delayed enhanced axial T1-weighted three-dimensional gradient-recalled echo, and (c) single-shot T2-weighted without and (d) with spectral selection attenuated inversion recovery fat-suppression MR images. These images demonstrate features suggestive of PDP, including abnormal thickening and enhancement of the second portion of the duodenum (arrow), with a complex cystic focus associated with the minor ampulla (arrowhead). Pathologic specimen analysis demonstrated invasive duct CA centered at the accessory ampulla.
Figure 3d: Pancreatic duct CA in an 80-year-old woman with abdominal pain and jaundice, inaccurately diagnosed as PDP by both blinded readers. (a) Precontrast axial T1-weighted three-dimensional gradient-recalled echo, (b) delayed enhanced axial T1-weighted three-dimensional gradient-recalled echo, and (c) single-shot T2-weighted without and (d) with spectral selection attenuated inversion recovery fat-suppression MR images. These images demonstrate features suggestive of PDP, including abnormal thickening and enhancement of the second portion of the duodenum (arrow), with a complex cystic focus associated with the minor ampulla (arrowhead). Pathologic specimen analysis demonstrated invasive duct CA centered at the accessory ampulla.
Agreement between the two readers, tested by using the Cohen κ statistic, showed substantial agreement (22) for the diagnosis of PDP and differentiation from pancreatic duct CA by contrast-enhanced MR imaging (κ = 0.735).
Additional imaging features were categorized and recorded for each patient (Table 3). In 16 of 17 (94.1%) patients with PDP, a complex fluid cavity structure was identified either in the region of the pancreatic accessory duct or within the medial wall of the duodenum adjacent to the accessory duct orifice (Fig E1 [online]). However, such a finding in this region was noted in only four of 30 (13.3%) patients with CA, and this difference was statistically significant (P < .001).
 |
Discussion
Our results show that there are identifiable imaging features on contrast-enhanced MR imaging that can characterize PDP. These features include the following: (a) focal thickening of the second portion of the duodenum; (b) abnormal increased enhancement of the second portion of the duodenum; and (c) cystic change in the region of the pancreatic accessory duct. It is also interesting that our case series shows that patients who presented with all three of these MR imaging features rarely had pancreatic duct CA, despite the overall appearance of an enlarged mass in the pancreatic head. Additionally, we showed that the MR imaging features that characterize PDP were detected consistently by inexperienced radiologists. This point helps support the conclusion that the MR imaging features associated with PDP may be useful in routine clinical practice where variable levels of experience may exist in the radiologic community. Of note was a high correlation between PDP and alcohol abuse, as noted previously (4). Absent portal vein involvement was found in similar frequency between PDP and CA patients, but that is largely due to presurgery selection of resectable disease.
Previous studies have reported imaging findings that are suggestive of PDP, including inflammatory changes in the pancreatic head, pancreatic or duodenal wall cyst formation, and duodenal wall thickening and stenosis. However, overlap of these imaging features with carcinoma has been noted (7,17). Few studies have systematically evaluated the ability of imaging to distinguish between PDP and CA. Ishigami et al (9) described the utility of patchy portal venous enhancement on multidetector-row CT and MR imaging as an indicator of PDP, and Johnson et al (18) demonstrated no statistically significant difference in enhancement patterns between chronic pancreatitis and CA.
Differentiation of PDP from invasive carcinoma is an important unmet clinical need. While new data (10,11) suggests that Whipple procedure may be the optimum therapeutic intervention for both entities, patients with CA may need preoperative chemotherapy. Accurate preoperative diagnosis of PDP may reduce the need for an additional diagnostic work-up, which may reduce overall cost and risk in patient care by reducing the delay to therapy and reducing the number of follow-up imaging tests or invasive biopsy procedures. In addition, long-term prognosis is markedly different between PDP and CA, with dismal 5-year survival rates for even surgically resectable pancreatic duct CA (23).
A limitation of our study is its retrospective design; every patient was preselected for having one or the other disease of investigation. Our methodologic analysis did not ask reviewers to identify the presence of cancer because of the assumption that all patients without PDP had cancer. We also recognize that there is a high proportion of PDP in our study population; while reflective of the referral patterns at our institution during the time period of this study, this is a potential source of bias. However, the readers were blinded to the pathologic diagnosis in each case and were not given an indication of the proportion of PDP in the study population. In addition, we specifically limited the scope of our study to the differentiation of PDP from CA, and did not include other potential neoplastic or inflammatory etiologies of a pancreatic head mass. Our study also only evaluated the performance of contrast-enhanced MR imaging in identifying PDP, and did not provide comparisons to other imaging modalities or to long term outcomes analysis in these patients.
In summary, our investigation suggested that MR imaging is accurate in identifying PDP and for differentiation of this entity from ductal carcinoma of the pancreatic head. Our findings may affect presurgical management of patients with a pancreatic head mass.
Advance in Knowledge
■ By using three strict diagnostic criteria for paraduodenal pancreatitis (PDP) (focal thickening of the second portion of the duodenum; abnormal increased enhancement of the second portion of the duodenum; and cystic change in the region of the pancreatic accessory duct), distinction from pancreatic duct adenocarcinoma (CA) can be achieved with a diagnostic accuracy of 87.2% (41 of 47 patients), and a diagnosis of cancer can be excluded with a negative predictive value of 92.9% (26 of 28 cases).
Implication for Patient Care
■ In patients with a pancreatic head mass, MR imaging may be used to identify the presence of PDP and help distinguish it from CA, which aids in optimization of therapeutic decisions, including the decision of whether or not to use preoperative chemotherapy.
Author Contributions
Author contributions: Guarantors of integrity of entire study, B.K., D.R.M., Z.C.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, B.K., D.R.M., Z.C., N.V.A.; clinical studies, B.K., D.R.M., J.M.S., D.G., E.B.T., Z.C.; experimental studies, Z.C.; statistical analysis, E.B.T., Z.C.; and manuscript editing, B.K., D.R.M., J.M.S., S.H.E., Z.C., N.V.A.
References
- 1. . Inflammatory, infectious, and other non-neoplastic disorders of the pancreas. In: Odze R, Goldblum J, eds. Surgical pathology of the GI tract, liver, biliary tract and pancreas. Philadelphia, Pa: Saunders Elsevier, 2009; 881–897. Crossref, Google Scholar
- 2. . Groove pancreatitis. Int J Pancreatol 1991;10(3-4):173–182. Medline, Google Scholar
- 3. . A special form of segmental pancreatitis: “groove pancreatitis”. Hepatogastroenterology 1982;29(5):198–208. Medline, Google Scholar
- 4. . Paraduodenal pancreatitis: a clinico-pathologically distinct entity unifying “cystic dystrophy of heterotopic pancreas”, “para-duodenal wall cyst”, and “groove pancreatitis”. Semin Diagn Pathol 2004;21(4):247–254. Crossref, Medline, Google Scholar
- 5. Groove pancreatitis and pancreatic heterotopia in the minor duodenal papilla. Pancreas 2005;30(4):e92–e95. Crossref, Medline, Google Scholar
- 6. Segmental groove pancreatitis accompanied by protein plugs in Santorini’s duct. J Gastroenterol 1998;33(2):289–294. Crossref, Medline, Google Scholar
- 7. . Groove pancreatitis: a diagnostic challenge. Eur Radiol 2009;19(7):1736–1743. Crossref, Medline, Google Scholar
- 8. . Chronic pancreatitis and the differential diagnosis versus pancreatic cancer. Arch Pathol Lab Med 2009;133(3):382–387. Crossref, Medline, Google Scholar
- 9. Differential diagnosis of groove pancreatic carcinomas vs. groove pancreatitis: usefulness of the portal venous phase. Eur J Radiol 2010;74(3):e95–e100. Crossref, Medline, Google Scholar
- 10. “Paraduodenal” pancreatitis: results of surgery on 58 consecutives patients from a single institution. World J Surg 2009;33(12):2664–2669. Crossref, Medline, Google Scholar
- 11. . Quality of life outcomes after the Whipple procedure for paraduodenal pancreatitis. Presented at the Pancreas Club, New Orleans, LA, May 2010. Google Scholar
- 12. Chronic pancreatitis or pancreatic ductal adenocarcinoma? Semin Diagn Pathol 2004;21(4):268–276. Crossref, Medline, Google Scholar
- 13. . MR imaging of cystic lesions of the pancreas. RadioGraphics 2009;29(6):1749–1765. Link, Google Scholar
- 14. . Pancreatic fluid collections prior to intervention: evaluation with MR imaging compared with CT and US. Radiology 1997;203(3):773–778. Link, Google Scholar
- 15. Differentiation of intraductal papillary mucinous neoplasms from other pancreatic cystic masses: comparison of multirow-detector CT and MR imaging using ROC analysis. J Magn Reson Imaging 2007;26(1):86–93. Crossref, Medline, Google Scholar
- 16. CT vs MRCP: optimal classification of IPMN type and extent. J Gastrointest Surg 2008;12(1):101–109. Crossref, Medline, Google Scholar
- 17. . MRI features of groove pancreatitis. AJR Am J Roentgenol 2007;189(1):73–80. Crossref, Medline, Google Scholar
- 18. . Pancreatic carcinoma versus chronic pancreatitis: dynamic MR imaging. Radiology 1999;212(1):213–218. Link, Google Scholar
- 19. Optimization of single injection liver arterial phase gadolinium enhanced MRI using bolus track real-time imaging. J Magn Reson Imaging 2011;33(1):110–118. Crossref, Medline, Google Scholar
- 20. . Groove pancreatitis: MRI and pathologic findings. Abdom Imaging 2008;33(3):342–348. Crossref, Medline, Google Scholar
- 21. . Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med 1998;17(8):857–872. Crossref, Medline, Google Scholar
- 22. . Understanding interobserver agreement: the kappa statistic. Fam Med 2005;37(5):360–363. Medline, Google Scholar
- 23. : Pancreas Cancer Fact Sheet. National Cancer Institute (NCI) web site. http://seer.cancer.gov/statfacts/html/pancreas.html. Published 2011. Accessed February 15, 2011. Google Scholar
Article History
Received September 26, 2011; revision requested November 18; revision received January 23, 2013; accepted February 25; final version accepted May 13.Published online: Nov 2013
Published in print: Nov 2013
