Comparison of needle arthroscopy, traditional arthroscopy, and computed tomography for the evaluation of medial coronoid disease in the canine elbow.

Abstract

Objective: To evaluate the diagnostic value of still images of needle arthroscopy (SNAR), still images of traditional arthroscopy (STAR), and computed tomography (CT) to diagnose medial coronoid process (MCP) pathology.

Study design: Prospective clinical trial.

Animals: Dogs (n = 17) presented for evaluation of elbow dysplasia.

Methods: For each case, two SNAR and STAR images of the MCP were reviewed independently and in random order by three board-certified surgeons. Computed tomographic images were reviewed by one board-certified radiologist. Reviewers were blinded to surgical and clinical findings. Surgical findings from real-time TAR with palpation were used as the gold standard. Receiver operating characteristic (ROC) curves and concordance statistics tests for the diagnostic accuracy of MCP fissure, MCP fragment, medial compartment condition, and cartilage score were calculated.

Results: Images of 27 elbows joints were reviewed. For MCP fissure detection, areas under the ROC curves for CT (0.84), STAR (0.73), and SNAR (0.57) did not differ. For the detection of MCP fragment, STAR had a larger area under the ROC curve (0.93) compared with SNAR (0.74, P = .015) and CT (0.54, P < .001). Still images of TAR and SNAR had comparable concordance for cartilage score (0.80 and 0.77, respectively) and medial compartment pathology (0.80 and 0.73, respectively).

Conclusion: Still images of NAR, STAR, and CT had similar diagnostic value to identify MCP fissures. Still images of TAR was superior to SNAR and CT to identify MCP fragments.

Clinical significance: The diagnostic accuracy of SNAR varied on the basis of the coronoid lesion being evaluated.