Three-Dimensional CT Planning for Anatomic Shoulder Arthroplasty: Reliability of Version Correction With Short-Term...

Anatomic shoulder arthroplasty for end-stage glenohumeral arthritis offers excellent outcomes for many patients. Computed tomography (CT) has become recognized as a valuable preoperative planning tool for assessing acquired bone deficiencies, such as excessive posterior glenoid version. It remains uncertain whether three-dimensional (3D) CT technology helps surgeons reliably restore native glenoid version and if version correction offers superior short-term clinical outcomes.  Methods: Fifty consecutive patients undergoing anatomic shoulder arthroplasty were enrolled in this study. A preoperative CT scan with 3D planning software was used to restore a normal range of glenoid version (0 to 10 degrees retroversion). All patients received a pegged cemented glenoid polyethylene component and underwent a postoperative CT scan to evaluate component alignment and glenoid version. Outcomes were measured by the ASES (American Shoulder and Elbow Surgeons) shoulder score and PROMIS (Patient-Reported Outcomes Measurement Information System) Global Health survey at baseline, 14 weeks, one year, and two years.  Results: Using the conventional radiological reading method, 76% of patients had their glenoid version corrected to 0-10°. Nearly all postoperative measures significantly improved at two-year follow-up, including pain, pain medication consumption, and range of motion. There were no statistically significant differences in patient outcomes when comparing glenoid version within versus outside the normal glenoid version correction range.  Conclusion: Restoring near-anatomic glenoid version is believed to be an important step to maximize prosthetic longevity. Three-dimensional CT planning software offers ways to recognize and correct abnormal versions. In the study cohort, approximately three in four patients were corrected to normal levels. There was no statistical difference in short-term surgical outcomes within or outside the desired version range.  Level of evidence: Level IV.