Paraduodenal pancreatitis: benign and malignant mimics at MRI

Pardeep Kumar Mittal, Peter Harri, Sadhna Nandwana, Courtney C. Moreno, Takashi Muraki, Volkan Adsay, Kelly Cox, Burcin Pehlivanoglu, Lauren F. Alexander, Argha Chatterjee, Frank H. Miller

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


Paraduodenal pancreatitis, also known as groove pancreatitis, is a rare form of chronic pancreatitis that masquerades as pancreatic adenocarcinoma affecting the pancreaticoduodenal groove, a potential space between the head of the pancreas, duodenum, and common bile duct. Two forms of groove pancreatitis have been described. The segmental form involves the pancreatic head with development of scar tissue within the groove, whereas the pure form affects the groove only, sparing the pancreatic head. Imaging findings of groove pancreatitis often overlap with primary duodenal, ampullary, or pancreatic neoplasms, which often results in a diagnostic challenge. In addition, paraduodenal pancreatitis can be mistaken for cystic pancreatic lesions, especially when there is involvement of the duodenal wall. Preoperative recognition of this entity is very important in order to avoid unnecessary procedures, although surgery, such as pancreaticoduodenectomy, may still be required to relieve obstructive symptoms. In this article, the pathophysiology and magnetic resonance imaging characteristics of paraduodenal pancreatitis and important benign and malignant mimics are discussed.

Original languageEnglish (US)
Pages (from-to)2652-2674
Number of pages23
JournalAbdominal Radiology
Issue number11
StatePublished - Nov 1 2017
Externally publishedYes


  • Magnetic resonance imaging (MRI)
  • Pancreatic adenocarcinoma
  • Paraduodenal pancreatitis (PDP)

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Gastroenterology
  • Urology


Dive into the research topics of 'Paraduodenal pancreatitis: benign and malignant mimics at MRI'. Together they form a unique fingerprint.

Cite this