Shoulder magnetic resonance arthrography: a prospective randomized study of anterior and posterior ultrasonography-guided contrast injections

Abstract

Magnetic resonance (MR) arthrography is an accurate imaging method for internal shoulder derangements and rotator cuff pathologies. Both anterior and posterior contrast injection techniques, under palpatory, fluoroscopic, or ultrasonographic guidance have been described in the literature. However, clinical comparisons of the injection techniques remain few. To compare the performance of anterior and posterior ultrasonography (US)-guided arthrography injections of the shoulder regarding patient discomfort and influence on diagnostic MR reading, and to illustrate the typical artifacts resulting from contrast leakage in the respective techniques. 43 MR arthrographies were prospectively randomized into anterior and posterior US-guided contrast injections and performed by two radiologists, with the study of artifacts from contrast leakage. Pain from the injections was assessed by a survey utilizing a 100-mm visual analogue scale (VAS). Of the 23 anterior injections, nine caused contrast artifacts in the subscapular tendon, and in three the leakage extended further anteriorly. Of the 20 posterior injections, 12 showed injection artifacts of the rotator cuff, extending outside the cuff in seven. Two of the anterior and none of the posterior artifacts compromised diagnostic quality. In posterior injections, the leakage regularly occurred at the caudal edge of the infraspinatus muscle and was easily distinguishable from rotator cuff tears. All patients completed the pain survey. Mean VAS scores were 25.0 (median 18, SD 22) for anterior, and 25.4 (median 16, SD 25) for posterior injections. The two radiologists achieved different mean VAS scores but closely agreed as to anterior and posterior VAS scores. Arthrography injections were fairly simple to perform under US guidance. Patient discomfort for anterior and posterior injections was equally minor. A tailored approach utilizing anterior or posterior injections, depending on anticipated shoulder pathology, is recommended. Because in posterior injections all artifacts were posterior and readily recognizable, it seems especially suitable for suspected anterior rotator cuff, joint capsule, and labral pathologies.