Simple Preoperative Radiographic and CT Measurements Predict Adequate Bone Quality for Stemless Total Shoulder Arthroplasty.

INTRODUCTION:
While there is increased utilization of stemless humeral implants in anatomic total shoulder arthroplasty (TSA), there are inadequate objective metrics to evaluate bone quality sufficient for fixation. Our goals are to 1) compare patient characteristics in patients that had plans for stemless TSA but received stemmed TSA due to intraoperative assessments and 2) propose threshold values of bone density, using the deltoid tuberosity index (DTI) and proximal humerus Hounsfield units (HU), on preoperative X-ray and computed tomography (CT) to allow for preoperative determination of adequate bone stock for stemless TSA.

METHODS:
Observational study at an academic institution from 2019 to 2021, including consecutive primary TSAs templated to undergo stemless TSA based on 3D CT preoperative plans. Final implant selection was determined by intraoperative assessment of bone quality. Preoperative X-ray and CT images were assessed to obtain DTI and proximal humeral bone density in HU, respectively. Receiver operative curve (ROC) was used to analyze the potential of preoperative X-ray and CT to classify patients as candidates for stemless TSA.

RESULTS:
61 planned stemless TSAs were included, with 56 (91.8%) undergoing stemless TSA and 5 (8.2%) undergoing stemmed TSA after intraoperative assessment determined the bone quality was inadequate for stemless fixation. There were no significant differences between the two groups in terms of gender (p=0.640), BMI (p=0.296), race (p=0.580). The stem cohort was significantly older (mean age 69 +/- 12 years vs. 59 +/- 10 years, p=0.029), had significant lower DTI (1.45 +/- 0.13 vs. 1.68 +/- 0.18, p=0.007), and significantly less proximal humeral HU (-1.4 +/- 17.7 vs. 78.8 +/- 52.4, p=0.001). ROC for DTI had an area under the curve (AUC) of 0.86 and bone density in HU had an AUC of 0.98 in its ability to distinguish patients that underwent stemless TSA versus short stem TSA. A threshold cut-off of 1.41 for DTI resulted in sensitivity of 98% and specificity of 60%, and a cut-off value of 14.4 Hounsfield units resulted in a sensitivity of 95% and specificity of 100%.

CONCLUSION:
Older age, lower DTI, and less proximal humeral bone density in HU was associated with the requirement to switch from stemless to short stem humeral fixation in primary TSA. Preoperative DTI had good ability (AUC of 0.86) and preoperative HU had excellent ability (AUC of 0.98) to categorize patients as appropriate for stemless TSA. This can help surgeons adequately plan humeral fixation using standard preoperative imaging data.