Clinical decision support tools are software that match the input characteristics of an individual patient to an established knowledge base to create patient-specific assessments that support and better inform individualized healthcare decisions. Clinical decision support tools can facilitate better evidence-based care and offer the potential for improved treatment quality and selection, shared decision making, while also standardizing patient expectations. Predict+ is a novel, clinical decision support tool that leverages clinical data from the Exactech Equinoxe shoulder clinical outcomes database, which is composed of >11,000 shoulder arthroplasty patients using one specific implant type from more than 30 different clinical sites using standardized forms. Predict+ utilizes multiple coordinated and locked supervised machine learning algorithms to make patient-specific predictions of 7 outcome measures at multiple postoperative timepoints (from 3 months to 7 years after surgery) using as few as 19 preoperative inputs. Predict+ algorithms predictive accuracy for the 7 clinical outcome measures for each of aTSA and rTSA were quantified using the mean absolute error and the area under the receiver operating curve (AUROC).Introduction
Methods
Acromial and scapular fractures are a rare but difficult complication with reverse total shoulder arthroplasty (rTSA), with an incidence rate reported from 1–10%. The risk factors associated with these fractures types is largely unknown. The goal of this study is to analyze the clinical outcomes, demographic and comorbidity data, and implant sizing and surgical technique information from 4125 patients who received a primary rTSA with one specific prosthesis (Equinoxe, Exactech, Inc) and were sorted based on the radiographic documentation of an acromial and/or scapula fracture (ASF) to identify factors associated with this complication. 4125 patients (2652F/1441M/32 unspecified; mean age: 72.5yrs) were treated with primary rTSA by 23 orthopaedic surgeons. Revision and fracture reverse arthroplasty cases were excluded. The radiographic presence of each fracture was documented and classified using the Levy classification method. 61 patients were identified as having ASF, 10 patients had fractures of the Type 1, 32 patients had Type 2, and 18 patients had Type 3 fractures according to Levy's classification. One fracture was not classifiable. Pre-op and post-op outcome scoring, ROM as well as demographic, comorbidity, implant, and surgical technique information were evaluated for these 61 patients and compared to the larger cohort of patients to identify any associations. A two-tailed, unpaired t-test identified differences (p<0.05).Introduction
Methods
Machine learning is a relatively novel method to orthopaedics which can be used to evaluate complex associations and patterns in outcomes and healthcare data. The purpose of this study is to utilize 3 different supervised machine learning algorithms to evaluate outcomes from a multi-center international database of a single shoulder prosthesis to evaluate the accuracy of each model to predict post-operative outcomes of both aTSA and rTSA. Data from a multi-center international database consisting of 6485 patients who received primary total shoulder arthroplasty using a single shoulder prosthesis (Equinoxe, Exactech, Inc) were analyzed from 19,796 patient visits in this study. Specifically, demographic, comorbidity, implant type and implant size, surgical technique, pre-operative PROMs and ROM measures, post-operative PROMs and ROM measures, pre-operative and post-operative radiographic data, and also adverse event and complication data were obtained for 2367 primary aTSA patients from 8042 visits at an average follow-up of 22 months and 4118 primary rTSA from 11,754 visits at an average follow-up of 16 months were analyzed to create a predictive model using 3 different supervised machine learning techniques: 1) linear regression, 2) random forest, and 3) XGBoost. Each of these 3 different machine learning techniques evaluated the pre-operative parameters and created a predictive model which targeted the post-operative composite score, which was a 100 point score consisting of 50% post-operative composite outcome score (calculated from 33.3% ASES + 33.3% UCLA + 33.3% Constant) and 50% post-operative composite ROM score (calculated from S curves weighted by 70% active forward flexion + 15% internal rotation score + 15% active external rotation). 3 additional predictive models were created to control for the time required for patient improvement after surgery, to do this, each primary aTSA and primary rTSA cohort was subdivided to only include patient data follow-up visits >20 months after surgery, this yielded 1317 primary aTSA patients from 2962 visits at an average follow-up of 50 months and 1593 primary rTSA from 3144 visits at an average follow-up of 42 months. Each of these 6 predictive models were trained using a random selection of 80% of each cohort, then each model predicted the outcomes of the remaining 20% of the data based upon the demographic, comorbidity, implant type and implant size, surgical technique, pre-operative PROMs and ROM measures inputs of each 20% cohort. The error of all 6 predictive models was calculated from the root mean square error (RMSE) between the actual and predicted post-op composite score. The accuracy of each model was determined by subtracting the percent difference of each RMSE value from the average composite score associated with each cohort.Introduction
Methods
Reverse shoulder arthroplasty (rTSA) increases the deltoid abductor moment arm length to facilitate the restoration of arm elevation; however, rTSA is less effective at restoring external rotation. This analysis compares the muscle moment arms associated with two designs of rTSA humeral trays during two motions: abduction and internal/external rotation to evaluate the null hypothesis that offsetting the humerus in the posterior/superior direction will not impact muscle moment arms. A 3-D computer model simulated abduction and internal/external rotation for the normal shoulder, the non-offset reverse shoulder, and the posterior/superior offset reverse shoulder. Four muscles were modeled as 3 lines from origin to insertion. Both offset and non-offset reverse shoulders were implanted at the same location along the inferior glenoid rim of the scapula in 20° of humeral retroversion. Abductor moment arms were calculated for each muscle from 0° to 140° humeral abduction in the scapular plan using a 1.8: 1 scapular rhythm. Rotation moment arms were calculated for each muscle from 30° internal to 60° external rotation with the arm in 30° abduction.Introduction
Methods
The inferior/medial shift in the center of rotation (CoR) associated with reverse shoulder arthroplasty (rTSA) shortens the anterior and posterior shoulder muscles; shortening of these muscles is one explanation for why rTSA often fails to restore active internal/external rotation. This study quantifies changes in muscle length from offsetting the humerus in the posterior/superior directions using an offset humeral tray/liner with rTSA during two motions: abduction and internal/external rotation. The offset and non-offset humeral tray/liner designs are compared to evaluate the null hypothesis that offsetting the humerus in the posterior/superior direction will not impact muscle length with rTSA. A 3-D computer model was developed to simulate abduction and internal/external rotation for the normal shoulder, the non-offset reverse shoulder, and the posterior/superior offset reverse shoulder. Seven muscles were modeled as 3 lines from origin to insertion. Both offset and non-offset reverse shoulders were implanted at the same location along the inferior glenoid rim of the scapula in 20° of humeral retroversion. Muscle lengths were measured as the average of the 3 lines simulating each muscle and are presented as an average length over each arc of motion (0 to 65° abduction with a fixed scapula and 0 to 40° of internal/external rotation with the humerus in 0° abduction) relative to the normal shoulder.Introduction
Methods
