This study analyzes the muscle moment arms of three different reverse shoulder design philosophies using a previously published method. Digital bone models of the shoulder were imported into a 3D modeling software and markers placed for the origin and insertion of relevant muscles. The anatomic model was used as a baseline for moment arm calculations. Subsequently, three different reverse shoulder designs were virtually implanted and moment arms were analyzed in abduction and external rotation. The results indicate that the lateral offset between the joint center and the axis of the humerus specific to one reverse shoulder design increased the external rotation moment arms of the posterior deltoid relative to the other reverse shoulder designs. The other muscles analyzed demonstrated differences in the moment arms, but none of the differences reached statistical significance. This study demonstrated how the combination of variables making up different reverse shoulder designs can affect the moment arms of the muscles in different and statistically significant ways. The role of humeral offset in reverse shoulder design has not been previously reported and could have an impact on external rotation and stability achieved post-operatively. ß

Update This article was updated on TK because of a previous error, which was discovered after the preliminary version of the article was posted online. In Table VII, the fracture rate in the study by Walch et al. that had read “4.6% (21 of 457)” now reads “0.9% (4 of 457).” Background: Acromial and scapular fractures after reverse total shoulder arthroplasty (rTSA) are rare and challenging complications, and little information is available in the literature to identify patients who are at risk. This study analyzes risk factors for, and compares the outcomes of patients with and without, acromial and scapular fractures after rTSA with a medialized glenoid/lateralized humeral implant. Methods: Four thousand one hundred and twenty-five shoulders in 3,995 patients were treated with primary rTSA with 1 design of reverse shoulder prosthesis by 23 orthopaedic surgeons. Sixty-one of the 4,125 shoulders had radiographically identified acromial and scapular fractures. Demographic characteristics, comorbidities, implant-related data, and clinical outcomes were compared between patients with and without fractures to identify risk factors. A multivariate logistic regression, 2-tailed unpaired t test, and chi-square test or Fisher exact test identified significant differences (p < 0.05). Results: After a minimum duration of follow-up of 2 years, the rate of acromial and scapular fractures was 1.77%, with the fractures occurring at a mean (and standard deviation) of 17.7 ± 21.1 months after surgery. Ten patients had a Levy Type-1 fracture, 32 had a Type-2 fracture, 18 had a Type-3 fracture, and 1 fracture could not be classified. Patients with acromial and scapular fractures were more likely to be female (84.0% versus 64.5% [p = 0.004]; odds ratio [OR] = 2.75 [95% confidence interval (CI) = 1.45 to 5.78]), to have rheumatoid arthritis (9.8% versus 3.3% [p = 0.010]; OR = 3.14 [95% CI = 1.18 to 6.95]), to have rotator cuff tear arthropathy (54.1% versus 37.8% [p = 0.005]; OR = 2.07 [95% CI = 1.24 to 3.47]), and to have more baseplate screws (4.1 versus 3.8 screws [p = 0.017]; OR = 1.53 [95% CI = 1.08 to 2.17]) than those without fractures. No other implant-related differences were observed in the multivariate analysis. Patients with fractures had significantly worse outcomes than patients without fractures, and the difference in mean improvement between these 2 cohorts exceeded the minimum clinically important difference for the majority of measures. Conclusions: Acromial and scapular fractures after rTSA are uncommon, and patients with these fractures have significantly worse clinical outcomes. Risk factors, including female sex, rheumatoid arthritis, cuff tear arthropathy, and usage of more baseplate screws were identified on multivariate logistic regression analysis. Consideration of these findings and patient-specific risk factors may help the orthopaedic surgeon (1) to better inform patients about this rare complication preoperatively and (2) to be more vigilant for this complication when evaluating patients postoperatively. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Objectives: To evaluate tuberosity union rate and clinical outcome after 3- and 4-part proximal humerus fractures in the elderly. Design: Retrospective, multicenter database cohort study. Setting: Level I and Level II trauma centers. Patients: Fifty-five patients older than 65 years had insertion of reverse shoulder arthroplasty (RTSA) for OTA/AO 11-B and 11-C proximal humerus fractures. Intervention: Treatment with RTSA using a dedicated low profile onlay fracture stem using variable tuberosity fixation. Main Outcome Measures: Constant score, the American Shoulder and Elbow Surgeons score, Shoulder Pain and Disability Index score, University of California at Los Angeles score, Simple Shoulder Test score, visual analog pain score, shoulder function score, active range of motion, external rotation (ER)-specific tasks and position, rate of greater tuberosity healing, effect of tuberosity healing on overall clinical metrics, incidence of humeral lucency, and scapular notching. Results: Eighty-three percent of the greater tuberosities that were repaired united. Greater tuberosity union resulted in greater active ER (P = 0.0415). There was a statistically significant difference in the ability to do ER-type activities between the 2 cohorts reflected in the ability to position one's hand behind their head with the elbow forward (P = 0.002) and comb their hair (P < 0.001). Conclusion: The use of a low profile onlay fracture stem in RTSA for acute 3- and 4-part proximal humerus fractures in the elderly can result in a high tuberosity union rate. Greater tuberosity healing significantly influences ER and ER-type activities that are not apparent by analysis of the overall metrics studied. Level of Evidence: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

No abstract

Background Machine learning techniques can identify complex relationships in large healthcare datasets and build prediction models that better inform physicians in ways that can assist in patient treatment decision-making. In the domain of shoulder arthroplasty, machine learning appears to have the potential to anticipate patients’ results after surgery, but this has not been well explored. Questions/purposes (1) What is the accuracy of machine learning to predict the American Shoulder and Elbow Surgery (ASES), University of California Los Angeles (UCLA), Constant, global shoulder function, and VAS pain scores, as well as active abduction, forward flexion, and external rotation at 1 year, 2 to 3 years, 3 to 5 years, and more than 5 years after anatomic total shoulder arthroplasty (aTSA) or reverse total shoulder arthroplasty (rTSA)? (2) What is the accuracy of machine learning to identify whether a patient will achieve clinical improvement that exceeds the minimum clinically important difference (MCID) threshold for each outcome measure? (3) What is the accuracy of machine learning to identify whether a patient will achieve clinical improvement that exceeds the substantial clinical benefit threshold for each outcome measure? Methods A machine learning analysis was conducted on a database of 7811 patients undergoing shoulder arthroplasty of one prosthesis design to create predictive models for multiple clinical outcome measures. Excluding patients with revisions, fracture indications, and hemiarthroplasty resulted in 6210 eligible primary aTSA and rTSA patients, of whom 4782 patients with 11,198 postoperative follow-up visits had sufficient preoperative, intraoperative, and postoperative data to train and test the predictive models. Preoperative clinical data from 1895 primary aTSA patients and 2887 primary rTSA patients were analyzed using three commercially available supervised machine learning techniques: linear regression, XGBoost, and Wide and Deep, to train and test predictive models for the ASES, UCLA, Constant, global shoulder function, and VAS pain scores, as well as active abduction, forward flexion, and external rotation. Our primary study goal was to quantify the accuracy of three machine learning techniques to predict each outcome measure at multiple postoperative timepoints after aTSA and rTSA using the mean absolute error between the actual and predicted values. Our secondary study goals were to identify whether a patient would experience clinical improvement greater than the MCID and substantial clinical benefit anchor-based thresholds of patient satisfaction for each outcome measure as quantified by the model classification parameters of precision, recall, accuracy, and area under the receiver operating curve. Results Each machine learning technique demonstrated similar accuracy to predict each outcome measure at each postoperative point for both aTSA and rTSA, though small differences in prediction accuracy were observed between techniques. Across all postsurgical timepoints, the Wide and Deep technique was associated with the smallest mean absolute error and predicted the postoperative ASES score to ± 10.1 to 11.3 points, the UCLA score to ± 2.5 to 3.4, the Constant score to ± 7.3 to 7.9, the global shoulder function score to ± 1.0 to 1.4, the VAS pain score to ± 1.2 to 1.4, active abduction to ± 18 to 21°, forward elevation to ± 15 to 17°, and external rotation to ± 10 to 12°. These models also accurately identified the patients who did and did not achieve clinical improvement that exceeded the MCID (93% to 99% accuracy for patient-reported outcome measures (PROMs) and 85% to 94% for pain, function, and ROM measures) and substantial clinical benefit (82% to 93% accuracy for PROMs and 78% to 90% for pain, function, and ROM measures) thresholds. Conclusions Machine learning techniques can use preoperative data to accurately predict clinical outcomes at multiple postoperative points after shoulder arthroplasty and accurately risk-stratify patients by preoperatively identifying who may and who may not achieve MCID and substantial clinical benefit improvement thresholds for each outcome measure. Clinical Relevance Three different commercially available machine learning techniques were used to train and test models that predicted clinical outcomes after aTSA and rTSA; this device-type comparison was performed to demonstrate how predictive modeling techniques can be used in the near future to help answer unsolved clinical questions and augment decision-making to improve outcomes after shoulder arthroplasty.