RSNA 2014 

Abstract Archives of the RSNA, 2014


Comparative Performance of MRCP with and without Contrast for Suspected Choledocholithiasis in Hospitalized Patients

Scientific Papers

Presented on December 5, 2014
Presented as part of SST06: Gastrointestinal (Gallbladder and Biliary Imaging)


Laura Heacock MS, MD, Presenter: Nothing to Disclose
Ankur Doshi MD, Abstract Co-Author: Nothing to Disclose
Justin Michael Ream MD, Abstract Co-Author: Nothing to Disclose
Danny C. Kim MD, Abstract Co-Author: Nothing to Disclose
James S. Babb PhD, Abstract Co-Author: Nothing to Disclose
Stella Kang MD, Abstract Co-Author: Nothing to Disclose


Magnetic resonance cholangiopancreatography (MRCP) is widely used as a noninvasive, accurate test for suspected choledocholithiasis. Current ACR guidelines recommend contrast-enhanced MRCP, but in hospitalized patients the full protocol may increase costs, scan time, and patient discomfort with questionable benefits. We compared performance of a potential short protocol with non-contrast MRI/ HASTE MRCP to contrast-enhanced MRI/3D MRCP.


We retrospectively evaluated 69 standard contrast-enhanced abdominal MRI/MRCP for suspected bile duct stones in inpatients. Two radiologists first used only non-contrast sequences including 2D coronal/axial HASTE, followed by the entire exam with post-contrast sequences and 3D MRCP. Readers noted perceived need for contrast, presence of common bile duct (CBD) stone, CBD dilatation, cholangitis, or other causes of acute biliary obstruction. Reader agreement and confidence were assessed. We also tested clinical factors predicting need for contrast in biliary assessment. ERCP, intraoperative cholangiogram or documented clinical resolution served as reference standard.


In 69 patients, 21 had confirmed choledocholithiasis, 4 had acute hepatitis, and 2 had clinical cholangitis. Both noncontrast and contrast-enhanced image sets resulted in high accuracy for bile duct stone (88-91% vs 87-90%); there was no significant difference in accuracy, sensitivity, specificity, NPV, PPV for either reader for any feature assessed with or without contrast (p>0.6). Reader agreement was excellent for non-contrast and contrast-enhanced detection of CBD stones (k=0.84, 0.77) and CBD dilatation (k=0.71, 0.61). 1 reader reported increased confidence (p<0.001) with inclusion of contrast-enhanced images. There was no significant association between perceived need by readers for intravenous contrast in biliary evaluation and fever, or risk factors for choledocholithiasis in American Society of Gastroenterology guidelines.


For hospitalized patients with suspected choledocholithiasis, performance of noncontrast MRI with HASTE MRCP may be equal to contrast-enhanced MRI/3D MRCP; a shorter test would offer potentially increased patient tolerability and reduced hospital costs.


For inpatients with suspected choledocholithiasis, an abbreviated non-contrast MRI with HASTE MRCP may offer a faster exam with no compromise in diagnostic test performance.

Cite This Abstract

Heacock, L, Doshi, A, Ream, J, Kim, D, Babb, J, Kang, S, Comparative Performance of MRCP with and without Contrast for Suspected Choledocholithiasis in Hospitalized Patients.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL.