Defining juxtapapillary diverticulum with 3D segmented trueFISP MRCP: comparison with conventional MRCP sequences with an oral negative contrast agent

Abstract

To compare three-dimensional segmented true fast imaging with steady-state precession magnetic resonance cholangiopancreatography (3D-trueFISP-MRCP) to conventional MRCP sequences with an oral negative contrast agent for diagnosing juxtapapillary diverticulum. A total of 42 patients with (n = 21) and without (n = 21) juxtapapillary diverticulum confirmed by endoscopic retrograde cholangiopancreatography (ERCP) were evaluated. Three MRCP sequences, 3D-trueFISP-MRCP, two-dimensional rapid acquisition with relaxation enhancement MRCP (2D-RARE-MRCP), and 3D T(2)-weighted turbo spin-echo MRCP (3D-TSE-MRCP), obtained after administering an oral negative contrast agent (FerriSeltz; Otsuka Pharmaceutical, Tokushima, Japan) were compared. Two radiologists independently and blindly interpreted the presence or absence of juxtapapillary diverticulum and its positional relationship against the papilla. The detectability of juxtapapillary diverticulum of each sequence as decided by consensus was then compared. Kappa statistics were used to measure interobserver agreement of the classifications. The sensitivity, specificity, and accuracy for detecting juxtapapillary diverticulum of 3D-trueFISPMRCP (61.9%, 85.7%, and 73.8%) were substantively higher than those of 2D-RARE-MRCP (0.0%, 100%, and 50.0%) and 3D-TSE-MRCP (9.5%, 100%, and 54.8%). Interobserver agreement of the position of juxtapapillary diverticulum by 3D-trueFISP-MRCP was good (kappa = 0.55). 3D-trueFISP-MRCP can define juxtapapillary diverticulum even with an oral negative contrast agent, whereas conventional MRCP sequences cannot.