Arthrography of the equine shoulder joint.

Techniques and normal radiographic anatomy for positive and double contrast shoulder arthrography in horses were evaluated. General anaesthesia was used for most radiographic projections of the shoulder. The mediolateral projection provided the most information during arthrography, although the supinated mediolateral view occasionally allowed better definition of the cartilage surfaces on the medial aspects of the humeral head. The craniocaudal mediolateral oblique and caudocranial projections provided limited additional information. Water soluble non-ionic contrast agents, such as metrizamide and iohexol, were suitable for shoulder arthrography; iohexol resulted in less synovitis and lameness. Arthrography in cases of osteochondrosis and osteochondritis dissecans allowed better evaluation of cartilage attachment to subchondral bone, better evaluation of the length and depth of cartilage lesions and more accurately defined the site and shape of osteocartilaginous free bodies. Cartilage thickening without detachment from the subchondral bone could only be determined by arthrography. Although these thick cartilage regions may later dissect from the subchondral bone, most cases where the cartilage was firmly adherent were not candidates for surgical debridement and carried a favourable prognosis. The determination of a free flap by arthrography indicated the need for surgery. Extensive humeral and glenoid cavity lesions were better defined by arthrography, allowing a rational decision between surgical debridement or euthanasia. Using arthrography, evaluation of the size and patency of the communicating canal to a subchondral cystic defect better separated cases with long, narrow and poorly patent canals for conservative rather than surgical therapy.