Objective: To determine the usefulness of 5 radiographic projections of the elbow joint for identification of fragmented coronoid process (FCP) and associated secondary degenerative changes in dogs.
Design: Longitudinal clinical study.
Animals: 19 dogs suspected of having FCP (7 dogs, 1 elbow joint; 12 dogs, both elbow joints).
Procedure: 5 radiographic projections were obtained before and after surgery for all elbow joints on which exploratory arthrotomy was performed because of suspected FCP. Radiographs obtained before surgery were compared with radiographs obtained after surgery and radiographs taken of the excised fragments. Each projection was evalated for its utility in definitively identifying FCP and 7 specific degenerative changes. Interevaluator agreement, use of each radiographic projection for definitively identifying a feature, and ability of the 4 evaluators to definitively identify a feature were analyzed.
Results: The craniolateral-caudomedial oblique (Cr15L-CdMO) projection had a significantly higher sensitivity for definitively identifying FCP, compared with the other 4 projections. Interevaluator and kappa agreement for the 5 projections did not differ significantly among the 7 degenerative changes evaluated. The radiographic projection with the highest sensitivity and specificity for detection of a particular degenerative change varied by feature.
Clinical implications: Of the 5 radiographic projections evaluated, the Cr15L-CdMO projection provided the highest sensitivity and was the best projection to use for definitively identifying FCP. The craniocaudal, Cr15L-CdMO, and mediolateral projections most reliably allowed for identification of secondary degenerative changes.