Predicting Primary Stability of Stemless TSA From Preoperative Bone Density and Surgical Factors

Abstract

This study investigated how preoperative bone quality, patient and implant factors can predict the primary stability of a stemless humeral component in total shoulder arthroplasty (TSA). Thirty‑one cadaveric humeri were CT‑scanned with a calibration phantom, and trabecular bone density within the metaphysis was quantified. Specimens were implanted and subjected to cyclic loading at 500 N and 870 N along physiologically representative directions and implant‑to‑bone micromotion was measured using motion capture. Generalized linear models were developed to predict implant micromotion based on patient and implant size factors, including age, height, weight, gender, ethnicity, cancer history, chemotherapy status, apparent peri-implant density, anchor size, head size, resection angle, and load level. Micromotion was significantly larger at 870 N than at 500 N, and males exhibited significantly lower micromotion than females, consistent with the associated higher bone density and larger implant sizes. Peri‑implant density was the strongest individual predictor explaining 34% of micromotion variation. Model predictions went up 49% of micromotion variation if load level was accounted for. Adding anchor and head sizes markedly improved this predictive power to 71%, which outperformed models incorporating patient height and gender (67%). Moderate improvement (73%) was seen when density, load level, height, gender, and anchor size were included after automatic backward selection. These findings demonstrate that bone density only partially predicts primary stability of anatomic stemless shoulder arthroplasty, and that incorporating implant sizing substantially improves predictive accuracy. This preoperative information can provide a more objective and consistent identification of patients with adequate bone quality for stemless fixation.