Magnetic resonance cholangiopancreatography for the diagnostic evaluation of autoimmune pancreatitis

Pancreas. 2010 Nov;39(8):1191-8. doi: 10.1097/MPA.0b013e3181dbf469.

Abstract

Objective: The aim of this study was to determine the role of magnetic resonance cholangiopancreatography (MRCP) for diagnosing autoimmune pancreatitis (AIP) and the accuracy of MRCP in depicting the main pancreatic duct (MPD) morphology of AIP using endoscopic retrograde cholangiopancreatography (ERCP) as the reference standard.

Methods: Thirty-eight AIP patients, 40 pancreatic cancer patients, and 40 patients with normal pancreas were included. MRCP was interpreted in association with cross-sectional magnetic resonance images regarding MPD morphology, pancreatic parenchyma, and extrapancreatic abnormalities. Main pancreatic duct was interpreted as narrowed when a narrowed-appearing segment on MRCP was associated with upstream dilatation or pancreatic parenchymal abnormalities in the same location.

Results: Accuracy of MRCP for depicting MPD morphology of AIP (64.7% [22/34]) was lower than those for pancreatic cancer (88.5% [23/26]) (P = 0.041) or normal pancreas (100% [40/40]) (P < 0.0005). The inaccuracy in AIP was primarily (10/12) due to overestimation of MPD narrowing. Of various differing MRCP findings between AIP and pancreatic cancer, multiple MPD narrowing (AIP vs cancer, 27/38 vs 0/40) and upstream MPD dilatation greater than 5 mm in diameter (AIP vs cancer, 0/38 vs 10/40) could exclude pancreatic cancer and AIP, respectively.

Conclusions: MRCP cannot replace ERCP for the diagnostic evaluation of AIP but may deserve to be used when ERCP has been unsuccessful or is difficult to perform.

MeSH terms

  • Autoimmune Diseases / diagnosis*
  • Cholangiopancreatography, Endoscopic Retrograde / methods
  • Cholangiopancreatography, Magnetic Resonance / methods*
  • Diagnosis, Differential
  • Humans
  • Pancreas / pathology
  • Pancreatic Diseases / diagnosis*
  • Pancreatic Neoplasms / diagnosis*
  • Reproducibility of Results
  • Sensitivity and Specificity