MR arthrography of the shoulder with gadopentetate dimeglumine has been proved to be valuable in the diagnosis of injuries of the rotator cuff and glenoid labrum. To date, no standard protocol for the concentration of gadopentetate dimeglumine and the volume of iodinated contrast material used for intraarticular injection exists. This study compared the intraarticular signal intensities achieved with three different MR arthrography protocols with respect to the volume of iodinated contrast material, the concentration of gadopentetate dimeglumine, and the interval between injection of contrast material and the MR study.

MR arthrography was performed with three different protocols in 38 patients (26 men, 12 women). Patients examined with protocol 1 received an intraarticular injection of 5 ml of iotrolan and 5 ml of a 5 mmol/l solution of gadopentetate dimeglumine, resulting in a gadopentetate dimeglumine concentration of 2.5 mmol/l. Patients in protocol 2 received only 0.5 ml of iotrolan plus 12 ml of a 10 mmol/l solution of gadopentetate dimeglumine, leading to a 10 mmol/l concentration. The interval between injection and MR imaging for patients in protocols 1 and 2 was 30-45 min. Patients in protocol 3 received an intraarticular injection of 10 ml iotrolan and 1 ml of a 500 mmol/l solution of gadopentetate dimeglumine, resulting in an intraarticular concentration of gadopentetate dimeglumine of 45 mmol/l. MR examinations were performed between 20 and 360 min after administration of contrast material. Patients in protocol 3 had additional conventional and CT arthrography within 30 min after injection. The contrast-to-noise ratio of the intraarticular signal intensity was calculated for all protocols.

The contrast-to-noise ratio (mean +/- SD) was 22.7 +/- 7.4 for the 2.5 mmol/l solution (protocol 1), 47.5 +/- 11.9 for the 10 mmol/l solution (protocol 2), and 9.9 +/- 3.1 for the 45 mmol/l solution (protocol 3) within 1 hr after injection. For protocol 3, an increase in contrast-to-noise ratio was observed after 90-180 min (53.5 +/- 12.7), followed by a decrease in contrast-to-noise ratio after 180 min (17.1 +/- 8.2). The image quality of conventional, CT, and MR arthrography (90-180 min after injection) was good in all cases of protocol 3.

For MR arthrography performed within 1 hr after injection, protocol 2 is sufficient. The optimal time frame for protocol 3 is between 1.5 and 3 hr after joint puncture. Protocol 3 allows the prior performance of conventional and CT arthrography. Therefore, the addition of 1 ml of a 500 mmol/l solution of gadopentetate dimeglumine to 10 ml of iodinated contrast material is advisable if a subsequent MR study is anticipated.