Abstract
Purpose: The success of humeral head replacement following fracture is reliant on several factors, one of which is version. The correct humeral version (HV) is highly variable, and is patient and side dependent. In the setting of fracture, there is no intra-operative landmark to guide the surgeon as to the anatomic version. This study has examined computed tomography (CT) of the shoulder and compared the HV to the metaphyseal version (MV) to evaluate reliability in predicting the anatomic version.
Method: A retrospective review of 50 shoulder CT scans was carried out. Patients were excluded if the anatomy prevented HV or MV evaluation. The HV and MV was measured by 2 independent evaluators. Inter and intra-rater reliability was performed.
Results: There were 27 right and 23 left shoulder CT’s reviewed. The mean age of patients was 45.3 (range 13–85). The difference between the MV and HV was approximately 2.8 (95% CI 0.63–5.1). Inter and intra-rater reliability was 0.966 and 0.984, respectively.
Conclusion: Determining the version of the humeral head in the setting of fracture is difficult and highly inaccurate. The biceps groove has been previously cited as a landmark for arthroplasty position, however, given the anatomic variability, version may be miscalculated. We have demonstrated the medial calcar of the proximal humerus is within 3 degrees of the actual humeral head version. This CT guided approach is novel, reproducible and demonstrates excellent reliability. It is both accurate and consistent and may be successfully utilized in the setting where normal anatomic landmarks are absent, such as fracture.
Correspondence should be addressed to: COA, 4150 Ste. Catherine St. West Suite 360, Westmount, QC H3Z 2Y5, Canada. Email: meetings@canorth.org
