Jinxing Yu MD, Presenter: Nothing to Disclose

Ann S. Fulcher MD, Abstract Co-Author: Nothing to Disclose

Mary Ann Turner MD, Abstract Co-Author: Nothing to Disclose

Haowei Zhang MD, PhD, Abstract Co-Author: Nothing to Disclose

Jennifer M. Hubert MD, Abstract Co-Author: Nothing to Disclose

Ryan Gabriel MD, Abstract Co-Author: Nothing to Disclose

Andrew J. Taylor MD, Abstract Co-Author: Nothing to Disclose

To determine if MRCP characteristics of pancreatic duct (PD) obstruction other than duct-penetrating sign can be used to differentiate pancreatic carcinoma from chronic pancreatitis.

From 1/2004 to 12/2009, 59 patients with chronic pancreatitis and 53 patients with pancreatic carcinoma, all of whom had a portion of PD not visualized at MRCP, were enrolled. Retrospective review of the total of 112 MRCPs was conducted independently by two radiologists blinded to the final diagnosis. MRCP assessment included: appearance of PD upstream to obstruction (abrupt amputation vs. smooth tapering; uniform vs. heterogeneous side branch ectasia; size of the PD); appearance at site of obstruction (ill-defined, discontinuous cystic foci vs. well-defined, continuous cystic foci); appearance of PD downstream to obstruction (normal vs. irregular with side branch ectasia); appearance of common bile duct (CBD) if the lesion arises from the pancreatic head (malignant vs. benign appearing stricture). Results were correlated with clinical, biopsy and surgical findings. The inter-reader agreement was calculated. The sensitivity, specificity, and odds ratio of significant findings and combinations of findings were also calculated.

Pancreatic carcinoma was identified in 90.5 % (48 of 53) of patients by reader 1 and in 92.4 % (49 of 53) by reader 2 (k value 0.75). Six MRCP findings – abrupt amputation, dilatation of PD > 4.2 mm and uniform side branch ectasia of the upstream PD; ill-defined, discontinuous cystic foci at the site of obstruction; normal downstream PD; and malignant appearing CBD stricture were found to be significant for differentiation of pancreatic carcinoma from chronic pancreatitis (P < .05). The inter-reader agreement for each feature ranged from moderate to very good (0.55 to 0.79). When any four of the six findings were used, pancreatic carcinoma could be differentiated from chronic pancreatitis at MRCP with a sensitivity of 98% and specificity of 96%.

MRCP features of PD obstruction other than duct-penetrating sign may be helpful in differentiating pancreatic carcinoma from chronic pancreatitis.

Chronic pancreatitis may mimic pancreatic carcinoma resulting in diagnostic challenge. Careful analysis of MRCP may provide clues for differentiating pancreatic carcinoma from chronic pancreatitis.

Yu, J, Fulcher, A, Turner, M, Zhang, H, Hubert, J, Gabriel, R, Taylor, A, Obstruction of Pancreatic Duct (PD) with Nonvisualization of a Portion of PD-Pancreatic Carcinoma or Pancreatitis: Differentiation on the Basis of MRCP.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11015849.html