RETURN TO ABSTRACT LISTING


Objective:
To improve MR cholangiopancreatography (MRCP) image quality, some institutions administer T2-weighted (T2W) signal suppressing oral contrast agents such as fruit juices with high manganese (Mn) content to minimize nonbiliary bright T2 signals in the bowel. Novel oral dark borosilicate contrast materials (DBCMs) are being increasingly developed. We sought to assess the effect of these oral agents on ghost artifacts in MRCP through a biliary phantom.

Materials and Methods:
A saline-filled biliary phantom was constructed to simulate the gallbladder, common bile duct, and first and second order bile duct branches. A 7-cm diameter gastric chamber was serially filled with saline, pineapple juice (high and low Mn), and novel DBCM at regular and half-strength, then imaged twice each using 3D T2W MRCP sequence (TE = 894 ms, TR = 2152 ms, slice thickness = 1 mm, matrix = 320 x 200, field of view = 380 mm, number of averages = 2, echo train length = 160) on a 3T MR scanner (Premier Signa, General Electric Healthcare). On 40-cm slab maximum intensity projection coronal images, regions of interest (ROI) were placed on the gallbladder, stomach, gallbladder ghost, stomach ghost, and away from the ghost artifacts to record mean (SD) signal intensity and standard deviation in consistent locations. Values were normalized to that of the gallbladder and averaged for each stomach lumen material.

Results:
The normalized stomach lumen signal was significantly lower when filled with the high or low [Mn] pineapple juice or the full or half-strength DBCM (0.028, 0.062, 0.040, and 0.069, respectively) compared to saline (0.590) (<em>p</em> < 0.05). Similarly, the signal of the stomach ghost was lower for the high and low [Mn] pineapple juices and the full and half-strength DBCMs (0.022, 0.026, 0.024, and 0.028, respectively) compared to saline-filled stomach (0.182) (<em>p</em> < 0.05). Paradoxically, the normalized signal of the gallbladder ghost was worsened for the high [Mn] and low [Mn] pineapple juice and the full and half-strength DBCM-filled stomach scans (0.044, 0.026, 0.043, and 0.059, respectively) compared to saline-filled stomach (0.016) (<em>p</em> < 0.05).

Conclusion:
Compared to stomachs filled with bright T2W signal fluid, stomachs filled with signal-suppressing oral contrast material substantially reduce stomach lumen and stomach ghost artifact signals in 3D MRCP. However, the use of signal suppressing oral contrast agents may slightly worsen ghost artifact severity from other nonbowel fluid-filled structures.