LAUSR

BACKGROUND Preoperative planning with commercially available imaging software in shoulder arthroplasty may allow for improved decision making and more accurate placement of the glenoid component. METHODS 81 consecutive shoulder computed tomography (CT) scans obtained for preoperative planning purposes for shoulder arthroplasty were analyzed by commercially available software from four companies (Blueprint - Wright Medical, Memphis, TN; GPS - Exactech, Gainesville, FL, USA; Materialise - DJO, Vista, CA, USA; and VIP - Arthrex, Naples, FL, USA), and by 5 fellowship trained sports medicine/shoulder surgeons. Inclination, version and subluxation of the humerus were measured in a blinded fashion on axial and coronal sequences at the mid-glenoid. Surgeon measurements were analyzed for agreement, and were compared to the 4 commercial programs. RESULTS Surgeon reliability was acceptable for version (intraclass correlation coefficient (ICC): 0.876), inclination (ICC:0.84), and subluxation (ICC: 0.523). Significant differences were found between surgeon and commercial software measurements in version (p=0.03), inclination (p=0.023) and subluxation (p<0.001). Software measurements tended to be more superiorly inclined (average -2° to 2° greater), more retroverted (average 2°-5° greater) and more posteriorly subluxed (average 7°-10° greater) than surgeon measurements. In comparing imaging software measurements only Blueprint produced significantly different version measurements than surgeon measurements (p=0.02). CONCLUSION Preoperative planning software for shoulder arthroplasty has limited agreement in measures of version, inclination and subluxation measurements while surgeons have high inter-reliability. Surgeons should be cautious when using commercial software planning systems and when comparing publications that use different planning systems to determine preoperative glenoid deformity measurements.