Coracoid graft positioning in the Latarjet procedure.

Purpose: The success of shoulder stabilization with the Latarjet procedure depends on the correct positioning of the coracoid graft at the glenoid. The aim of this study was to assess intra-observer reproducibility and inter-observer reproducibility of a new standardized CT scan analysis for coracoid graft positioning in the axial plane after the Latarjet procedure and to assess the positioning in the study group.

Methods: A consecutive series of 27 patients (22 men, 5 woman, 26.1 ± 6.4 years-13 right, 14 left shoulders) were followed up with CT scans between 2010 and 2012. The analysis of the CT scans (2.4 ± 0.7 months postoperatively) was performed with Osirix™ software. The assessment included two criteria in the axial plane: relation of the graft to the articular surface of the glenoid and impingement of the graft with the maximal humeral head circumference. Grafts were judged to be lateral, congruent, flush or medial. The strength of intra-observer agreement and inter-observer agreement was measured by the Kappa coefficient.

Results: The Kappa coefficient for intra-observer agreement was "substantial" (K = 0.64 ± 0.14, z = 4.6) to "almost perfect" (K = 0.81 ± 0.14, z = 5.7). The Kappa coefficient for inter-observer variability was "substantial" (K = 0.59 ± 0.14, z = 4.3) to "almost perfect" (K = 0.89 ± 0.14, z = 6.0). In our study, in the axial plane, 3 (11 %) transplants were lateral; 6 (22 %) transplants were congruent; 16 (60 %) flush and 2 (7 %) medial.

Conclusion: This standardized CT scan analysis after Latarjet procedure has shown to accurately describe graft positioning in the axial plane with both good intra-observer reproducibility and inter-observer reproducibility.

Level Of Evidence: Case series, treatment study, Level IV.