Low density barium and bentonite mixture versus high density barium: a comparative study to optimize negative gastrointestinal contrast agents for MRI.

Abstract

The purpose of this study was to compare the patient tolerance and efficacy, as magnetic resonance imaging negative oral contrast agents, of a mixture of clay compound bentonite and low density barium sulfate suspension with that of higher density barium sulfate. Twenty patients were randomized into two groups: 10 patients receiving a mixture of low concentration 60% w/v barium sulfate plus 2.5% w/v bentonite, and 10 patients receiving 220% w/v barium sulfate Liuqid-HD (E-Z-EM, Westbury, NY). Post-contrast Spin-echo (SE) T1- and T2-weighted images (WI) were obtained on a 1.0T magnet. Two independent readers scored the overall intraluminal signal intensity and delineation of the gastrointestinal tract and adjacent organs. Patient acceptance was evaluated via a short questionnaire, by recording spontaneous comments and documenting the quantity of contrast agent ingested. There was greater intraluminal bowel signal reduction and organ delineation with 220% w/v barium than with the barium-bentonite mixture on both SE T1WI (p = 0.03) and SE T2WI (p = 0.42). With both agents there was greater signal reduction on SE T2WI than SE T1WI. Higher scores for organ delineation for both contrast agents were seen with SE T1WI. With 220% w/v barium, there was significantly better delineation of the pancreatic body (p = 0.02) and pancreatic tail (p = 0.02) on T1WI compared with SE T2WI. With the barium-bentonite mixture, SE T1WI showed improved delineation of jejunum compared with SE T2WI (p = 0.03). There were no statistically significant differences between the volume of contrast ingested in the two groups. Abdominal cramps were recorded for one patient in each group. These results suggest that barium-bentonite mixture, although useful as a negative gastro-intestinal contrast agent, is not as effective as 220% w/v barium. Further studies with a larger patient population and concentration optimization studies are needed.