Inferior Glenoid HyperOSTosis (GHOST) lesion development following reverse total shoulder arthroplasty

Patients undergoing reverse total shoulder arthroplasty (rTSA) have been noted on postoperative radiographs to have a curved bony overgrowth on the inferior glenoid neck. This study aims to investigate potential risk factors for and postoperative effects of these ossifications, here called Glenoid HyperOSTosis (GHOST) lesions because of their location.

A retrospective review of patients undergoing rTSA between 2007 and 2020 at a single institution was performed. Predictors, including gender, age, implant sizing, and preoperative diagnosis, were examined using logistic regression analysis. Outcomes including readmission rate, revision rate, preoperative and postoperative range of motion, visual analog scale (VAS) pain scores, and American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES) functional outcome scores were also analyzed using 2-sample t tests.

Preoperative and postoperative radiographs of 170 primary rTSA patients were assessed. One-fourth (25.9%) of rTSAs were identified to have a GHOST lesion. When adjusting for race, age, body mass index (BMI), and preoperative diagnosis, male patients were still associated with 2.28 odds of developing a GHOST lesion compared with female patients (95% confidence interval [CI] 1.08-4.86). Other elements such as age, race, BMI, laterality, preoperative diagnosis, implant manufacturer, and implant sizing demonstrated no statistically significant association with GHOST lesion presence. Postoperatively, GHOST lesion development was not associated with range of motion or ASES score. However, presence of GHOST lesions on radiographs was associated with increased pain scores for patients at 2 months (P = .034) and 12 months (P = .029) postoperatively.

Inferior GHOST lesions are a common and potentially benign finding following rTSA, with unclear etiology. Risk factors for GHOST lesions included male gender, whereas patient demographics, implant type or size, shoulder lateralization, and distalization were not associated with lesion formation. Clinically, greater short-term VAS scores were seen in patients with GHOST lesions. However, there were no differences observed between the 2 groups in ASES scores or postoperative range of motion at later time points. Further research is needed to identify risk factors and assess the clinical implications of GHOST lesions.