Scapular notching in reverse shoulder arthroplasty: validation of a computer impingement model.

PURPOSE: The purpose of this study is to validate a reverse shoulder computer impingement model and quantify the impact of implant position on scapular impingement by comparing it to that of a radiographic analysis of 256 patients who received the same prosthesis and were followed postoperatively for an average of 22.2 months. METHODS: A geometric computer analysis quantified anterior and posterior scapular impingement as the humerus was internally and externally rotated at varying levels of abduction and adduction relative to a fixed scapula at defined glenoid implant positions. These impingement results were compared to radiographic study of 256 patients who were analyzed for notching, glenoid baseplate position, and glenosphere overhang. RESULTS: The computer model predicted no impingement at 0° humeral abduction in the scapular plane for the 38 mm, 42 mm, and 46 mm devices when the glenoid baseplate cage peg is positioned 18.6 mm, 20.4 mm, and 22.7 mm from the inferior glenoid rim (of the reamed glenoid) or when glenosphere overhang of 4.6 mm, 4.7 mm, and 4.5 mm was obtained with each size glenosphere, respectively. When compared to the radiographic analysis, the computer model correctly predicted impingement based upon glenoid base- plate position in 18 of 26 patients with scapular notching and based upon glenosphere overhang in 15 of 26 patients with scapular notching. CONCLUSIONS: Reverse shoulder implant positioning plays an important role in scapular notching. The results of this study demonstrate that the computer impingement model can effectively predict impingement based upon implant positioning in a majority of patients who developed scapular notching clinically. This computer analysis provides guidance to surgeons on implant positions that reduce scapular notching, a well-documented complication of reverse shoulder arthroplasty.