ChatGPT-4 Performs Clinical Information Retrieval Tasks Utilizing Consistently More Trustworthy Resources Than Does Google Search for Queries Concerning the Latarjet Procedure.

PURPOSE: To assess the ability for ChatGPT-4, an automated Chatbot powered by artificial intelligence (AI), to answer common patient questions concerning the Latarjet procedure for patients with anterior shoulder instability and compare this performance to Google Search Engine. METHODS: Using previously validated methods, a Google search was first performed using the query "Latarjet." Subsequently, the top ten frequently asked questions (FAQs) and associated sources were extracted. ChatGPT-4 was then prompted to provide the top ten FAQs and answers concerning the procedure. This process was repeated to identify additional FAQs requiring discrete-numeric answers to allow for a comparison between ChatGPT-4 and Google. Discrete, numeric answers were subsequently assessed for accuracy based on the clinical judgement of two fellowship-trained sports medicine surgeons blinded to search platform. RESULTS: Mean (±standard deviation) accuracy to numeric-based answers were 2.9±0.9 for ChatGPT-4 versus 2.5±1.4 for Google (p=0.65). ChatGPT-4 derived information for answers only from academic sources, which was significantly different from Google Search Engine (p=0.003), which used only 30% academic sources and websites from individual surgeons (50%) and larger medical practices (20%). For general FAQs, 40% of FAQs were found to be identical when comparing ChatGPT-4 and Google Search Engine. In terms of sources used to answer these questions, ChatGPT-4 again used 100% academic resources, while Google Search Engine used 60% academic resources, 20% surgeon personal websites, and 20% medical practices (p=0.087). CONCLUSION: ChatGPT-4 demonstrated the ability to provide accurate and reliable information about the Latarjet procedure in response to patient queries, using multiple academic sources in all cases. This was in contrast to Google Search Engine, which more frequently used single surgeon and large medical practice websites. Despite differences in the resources accessed to perform information retrieval tasks, the clinical relevance and accuracy of information provided did not significantly differ between ChatGPT-4 and Google Search Engine.

Outcomes after débridement, antibiotics, and implant retention for prosthetic joint infection in shoulder arthroplasty.

BACKGROUND: Patients who undergo total shoulder arthroplasty usually have excellent long-term outcomes. However, a subset of patients is diagnosed with a prosthetic joint infection (PJI) requiring revision procedures and prolonged recovery. The purpose of this study was to evaluate rates of recurrent shoulder PJI in patients undergoing débridement, antibiotics, and implant retention (DAIR), single-stage revision, and 2-stage revision. We also sought to compare outcomes and complications across procedures. METHODS: Retrospective chart review was conducted for patients diagnosed with PJI after primary shoulder arthroplasty between January 2010 and August 2021. Patients were included if they underwent treatment with DAIR, single-stage revision, or 2-stage revision. Demographic information, surgical details, complications, laboratory data, postoperative antibiotic regimen, and infectious pathogen were collected. Postoperative patient-reported outcomes were collected: American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, Single Assessment Numeric Evaluation, Shoulder Activity Scale, and PROMIS Upper Extremity. Chi-square, t test, and 1-way analysis of variance were used as appropriate to evaluate each factor. RESULTS: Sixty-five patients were included in the study, 26% treated with DAIR, 9% treated with single-stage revision, and 65% treated with 2-stage revision. There were no significant differences in patient comorbidities. Patients undergoing DAIR were diagnosed significantly earlier than those undergoing single- and 2-stage revision procedures (12.6 ± 22.9 months vs. 49.6 ± 48.4 vs. 25.0 ± 26.6, P = .010). Recurrent PJI was noted in 23.1% of patients: 29.4% of DAIR patients, no single-stage patients, and 23.8% of 2-stage patients (P = .330). Patients undergoing 2-stage revision with treatment failure had a significantly higher Elixhauser Comorbidity Index (0.2 ± 3.7 vs. 3.7 ± 3.9, P = .027). There was no significant difference in patient-reported outcomes across groups. CONCLUSION: Patients undergoing treatment of shoulder PJI with DAIR did not have an increased rate of reinfection compared with single-stage and 2-stage revision procedures. Patients treated with DAIR were diagnosed with PJI significantly earlier than those undergoing single-stage and 2-stage revision procedures. There was no difference in complication rates between groups. This information adds to the body of work detailing outcomes after DAIR for shoulder PJI and provides encouraging data for use in this patient population. Future studies with a larger sample size may be conducted to further investigate specific pathogens, infection timelines, and antibiotic regimens that reduce the risk of treatment failure.

PROMIS physical function and pain perform poorly psychometrically in patients undergoing medial patellofemoral ligament reconstruction.

PURPOSE: The purpose of this study was to assess the validity, floor and ceiling effects, and dimensionality of PROMIS Physical Function (PF) and Pain compared to legacy patient reported outcome (PRO) measures in patients undergoing medial patellofemoral ligament (MPFL) reconstruction. METHODS: Patients who underwent MPFL reconstruction between 2018 to 2020 were retrospectively reviewed. Preoperatively, patients completed the IKDC, VR-12, Kujala, SF-12, KOOS JR, PROMIS PF and Pain surveys. Inter-survey convergence was assessed with Spearman correlations. Psychometric analysis included investigations of inter-survey convergent validity, intra-survey floor and ceiling effects, and Rasch analyses with person-item fit and iterative question elimination model fit testing. RESULTS: A total of 76 patients (mean age: 22.6 ± 8.4 years) who completed preoperative surveys were included (compliance: 91.7-96.2%). Preoperatively, age was significantly associated with both PROMIS PF (coefficient: - 0.291, P = 0.005) and Pain scores (coefficient: 0.294, P = 0.002). PROMIS PF had a Very Good correlation with IKDC and PROMIS Pain had a Very Good correlation with KOOS JR. Other correlations ranged from Poor to Good. No significant floor or ceiling effects were observed for any PRO. On iterative question elimination Rasch modelling, only two questions from PROMIS PF remained after 6 rounds of elimination while PROMIS Pain had no questions remaining after 3 rounds of elimination. CONCLUSION: Preoperative PROMIS PF and Pain provided only Fair to Good correlations with most legacy PROs. Although no significant floor and ceiling effects were observed, PROMIS PF and Pain did not perform well psychometrically in this population of patients undergoing MPFL reconstruction. These results suggest PROMIS questionnaires should be used with caution in this population, especially if used in isolation, when discussing clinical expectations with patients. The authors recommend continued use of legacy PROs specific to assessment of patellar instability and function that have established validity in patellar instability populations. LEVEL OF EVIDENCE: Level IV.

How accurate are positive preoperative aspiration culture results in shoulder periprosthetic joint infection? A concordance study.

BACKGROUND: The concordance between preoperative synovial fluid cultures and intraoperative tissue cultures for identifying pathogenic microorganisms in shoulder periprosthetic joint infection (PJI) remains poorly understood. The purpose of our study was to examine the diagnostic accuracy of positive synovial fluid culture results in early pathogen identification for shoulder PJI. METHODS: A total of 35 patients who met the Musculoskeletal Infection Society criteria for PJI following primary anatomic or reverse arthroplasty and the study inclusion criteria were identified retrospectively from a single institution (multiple surgeons) from January 2011 to June 2021. The inclusion criteria required a positive preoperative intra-articular synovial fluid sample within 90 days analyzed within the same institution and intraoperative tissue cultures at the time of arthrotomy. Concordance was determined when the organism(s) identified from the aspirate correlated with the intraoperative specimens. RESULTS: Overall concordance was identified in 28 of 35 patients (80%), with similar concordance for anatomic (21 of 24, 88%) and reverse (7 of 11, 64%) shoulder arthroplasties (P = .171). Culture discordance occurred in 7 of 35 patients (20%): of these, 5 (14%) had no corresponding intraoperative culture growth whereas 2 (6%) had polymicrobial intraoperative cultures. Monomicrobial Cutibacterium acnes PJI cases were the most common (24 of 35, 69%) and had an overall concordance rate of 79%. Of 5 discordant C acnes patients, 2 had polymicrobial intraoperative cultures and 3 had negative intraoperative culture results; all the patients with negative intraoperative culture results had received antibiotics between the time of aspiration and surgery. Considered separately, concordance in patients who had a positive aspirate finding for C acnes and did not receive antibiotics prior to surgery was 19 of 21 (90%), with a sensitivity of 100% (95% confidence interval, 82%-100%) and a corresponding positive predictive value of 0.91 (95% confidence interval, 58%-93%). CONCLUSION: Preoperative positive aspiration culture results demonstrated favorable sensitivity and specificity when compared with intraoperative tissue cultures in identifying pathogenic microorganisms in shoulder PJI patients. These findings are congruent with literature from hip and knee arthroplasty. Ultimately, confidence in the accuracy of positive preoperative aspiration culture results in shoulder PJI may facilitate the development of early, targeted treatment strategies while directing patient expectations and risk.

Shoulder arthroplasty device clearance: an ancestral network analysis.

BACKGROUND: The US Food and Drug Administration (FDA) oversees medical device regulation and oversight in the United States, and the majority of shoulder arthroplasty devices are cleared via the 510(k) pathway, in which a device demonstrates "substantial equivalence" to a previously cleared predicate. The purpose of this study was to determine an interconnected ancestral network of shoulder arthroplasty devices and determine equivalency ties to devices subsequently recalled by the FDA for design-related issues. METHODS: The FDA 510(k) database was used to identify all legally marketed shoulder arthroplasty devices from May 28, 1976, to July 1, 2021. Direct predicate information obtained via clearance summary documents associated with each device was used to generate an ancestral genealogy network for all shoulder arthroplasty devices cleared between July 1, 2020, and July 1, 2021. FDA design recalls were analyzed, and the number of descendant devices was calculated for each recalled device. RESULTS: An evaluation of all 476 510(k) premarket notification pathway-cleared shoulder devices since 1976 identified 0-313 descendant devices for each. Eighty of these devices (16.8%) have since been recalled, of which 10 recalls were directly related to implant design issues. Furthermore, among 29 of the most recently cleared devices (July 1, 2020-July 1, 2021), 16 (55.2%) claim predicates devices that have subsequently been withdrawn from the market because of design-related failures. CONCLUSIONS: Shoulder arthroplasty devices are linked together via an interconnected FDA 510(k) equivalency approval network dating back to 1976 despite substantive changes in material specifications and device design, many of which have since been recalled. Many of the cleared modern devices claim predicates based on subsequently recalled prostheses.

Artificial intelligence for automated identification of total shoulder arthroplasty implants.

BACKGROUND: Accurate and rapid identification of implant manufacturer and model is critical in the evaluation and management of patients requiring revision total shoulder arthroplasty (TSA). Failure to correctly identify implant designs in these circumstances may lead to delay in care, unexpected intraoperative challenges, increased morbidity, and excess health care costs. Deep learning (DL) permits automated image processing and holds the potential to mitigate such challenges while improving the value of care rendered. The purpose of this study was to develop an automated DL algorithm to identify shoulder arthroplasty implants from plain radiographs. METHODS: A total of 3060 postoperative images from patients who underwent TSA between 2011 and 2021 performed by 26 fellowship-trained surgeons at 2 independent tertiary academic hospitals in the Pacific Northwest and Mid-Atlantic Northeast were included. A DL algorithm was trained using transfer learning and data augmentation to classify 22 different reverse TSA and anatomic TSA prostheses from 8 implant manufacturers. Images were split into training and testing cohorts (2448 training and 612 testing images). Optimized model performance was assessed using standardized metrics including the multiclass area under the receiver operating characteristic curve (AUROC) and compared with a reference standard of implant data from operative reports. RESULTS: The algorithm classified implants at a mean speed of 0.079 seconds (±0.002 seconds) per image. The optimized model discriminated between 8 manufacturers (22 unique implants) with AUROCs of 0.994-1.000, accuracy of 97.1%, and sensitivities between 0.80 and 1.00 on the independent testing set. In the subset of single-institution implant predictions, a DL model identified 6 specific implants with AUROCs of 0.999-1.000, accuracy of 99.4%, and sensitivity >0.97 for all implants. Saliency maps revealed key differentiating features across implant manufacturers and designs recognized by the algorithm for classification. CONCLUSION: A DL model demonstrated excellent accuracy in identifying 22 unique TSA implants from 8 manufacturers. This algorithm may provide a clinically meaningful adjunct in assisting with preoperative planning for the failed TSA and allows for scalable expansion with additional radiographic data and validation efforts.

Factors associated with humeral stem revision in anatomic to reverse total shoulder arthroplasty revision.

BACKGROUND: Recent advances in implant technology have allowed for modular or platform humeral stem insertion during initial anatomic total shoulder arthroplasty (TSA). These systems allow for humeral stem retention during conversion to reverse TSA (RTSA). However, some patients still require humeral stem revision when undergoing revision to RTSA. The purpose of this study was to evaluate the association between patient-specific factors and radiographic parameters with humeral stem revision vs. retention during conversion from TSA to RTSA. METHODS: Retrospective chart review was conducted for patients who underwent a revision TSA to RTSA between January 2010 and May 2022 at a single institution. Patients were included if their prosthesis included a convertible humeral stem. Patient demographic information, surgical details, and postoperative outcomes and complications were collected. Radiographic parameters were measured by 2 graders on radiographs taken prior to the revision procedure. The need for humeral stem revision and American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES) scores (preoperative and 2 years) were also noted. Statistical analysis was performed using chi-square test for categorical variables and t test for continuous variables. RESULTS: One hundred seven patients were included, with 52 undergoing humeral stem revision. Patients were revised an average of 51.0 ± 54 months after primary TSA. Younger patient age (63.6 vs. 68.5 years, P = .017) and use of a lateralized glenosphere (1.6 mm vs. 0.4 mm, P < .001) were significantly associated with need for humeral stem revision. Glenoid to humeral head cut distance (28.3 mm vs. 26.3 mm, P = .076) approached significant association with the need for humeral stem revision. All other measurements were not associated with the need for humeral stem revision. Improvement of ASES scores at 2 years' follow-up was higher in the nonrevised group (increase of 33.4 points) than the revision group (23.3), but this did not reach significance (P = .149). Estimated blood loss and surgical time were significantly higher in the stem revision group than the non-revised group (P = .048 and P < .001, respectively). CONCLUSION: Younger patients and those receiving a lateralized glenosphere were more likely to undergo humeral stem revision during conversion from TSA to RTSA. Glenoid to the humeral head cut distance should be studied further as a potential indication for humeral stem revision, as it correlates with the space available for a revision implant. This information can guide surgeons with preoperative planning for a revision arthroplasty.

Machine-learning model successfully predicts patients at risk for prolonged postoperative opioid use following elective knee arthroscopy.

PURPOSE: Recovery following elective knee arthroscopy can be compromised by prolonged postoperative opioid utilization, yet an effective and validated risk calculator for this outcome remains elusive. The purpose of this study is to develop and validate a machine-learning algorithm that can reliably and effectively predict prolonged opioid consumption in patients following elective knee arthroscopy. METHODS: A retrospective review of an institutional outcome database was performed at a tertiary academic medical centre to identify adult patients who underwent knee arthroscopy between 2016 and 2018. Extended postoperative opioid consumption was defined as opioid consumption at least 150 days following surgery. Five machine-learning algorithms were assessed for the ability to predict this outcome. Performances of the algorithms were assessed through discrimination, calibration, and decision curve analysis. RESULTS: Overall, of the 381 patients included, 60 (20.3%) demonstrated sustained postoperative opioid consumption. The factors determined for prediction of prolonged postoperative opioid prescriptions were reduced preoperative scores on the following patient-reported outcomes: the IKDC, KOOS ADL, VR12 MCS, KOOS pain, and KOOS Sport and Activities. The ensemble model achieved the best performance based on discrimination (AUC = 0.74), calibration, and decision curve analysis. This model was integrated into a web-based open-access application able to provide both predictions and explanations. CONCLUSION: Following appropriate external validation, the algorithm developed presently could augment timely identification of patients who are at risk of extended opioid use. Reduced scores on preoperative patient-reported outcomes, symptom duration and perioperative oral morphine equivalents were identified as novel predictors of prolonged postoperative opioid use. The predictive model can be easily deployed in the clinical setting to identify at risk patients thus allowing providers to optimize modifiable risk factors and appropriately counsel patients preoperatively. LEVEL OF EVIDENCE: III.

Joint and segment sequencing and its relationship to ball velocity and throwing arm kinetics in professional pitchers.

BACKGROUND: Temporal variations during the pitch have demonstrated significant impacts on the kinetic chain, and as such, have implications in injury risk. PURPOSE: To determine the effect of varying chronological orders of maximum joint and segment velocities on ball velocity and upper extremity kinetics. METHODS: Professional baseball pitchers (n = 287) were assessed with 3D-motion capture (480 Hz) while pitching. Pitches were categorized into one of the following groups dependent on the first maximum joint or segment velocity achieved out of chronological order in an inferior to superior direction: knee extension (DscK), pelvis rotation (DscP), trunk rotation (DscT), shoulder rotation (DscS), forearm pronation (DscF), and Proper (for pitchers with the correct temporal sequence), and Total Population, for all pitchers. Ten normalized throwing arm kinetic variables were compared among groups. Regression analysis was conducted on the timing of maximum velocities with ball velocity. RESULTS: The majority of pitches were in the DscK group (64.5%). The DscK group had a significantly slower maximum lead knee extension velocity compared with the Proper group (253°/s vs. 316°/s, P = .017). The Proper group had a significantly faster ball velocity compared with the Total Population (39.0 ± 1.9 m/s vs. 38.3 ± 2.1 m/s, P = .013). The DscP group had a significantly slower maximum pelvis rotation velocity compared with the Proper group (596°/s vs. 698°/s, P < .001). The Proper group had no significant difference in kinetics relative to the population. For every 1 standard deviation delay in attaining maximum lead knee extension velocity, ball velocity increased by 0.38 m/s (B = 3.5, ß = 0.18, P < .001). For every 1 standard deviation delay in timing to achieve maximum pelvis rotation velocity, maximum pelvis rotation velocity and ball velocity increased by 22.5°/s (B = 1107.0, ß = 0.23, P < .001) and 0.48 m/s (B = 23.4, ß = 0.23, P < .001), respectively. CONCLUSION: Pitchers with a discordant sequence of knee extension and pelvis rotation velocity timing had significantly slower corresponding segment/joint velocities. Conversely, pitchers with a proper sequence had the fastest ball velocity with minimal differences in throwing arm kinetics. To maximize ball velocity, professional pitchers should consider achieving maximal velocities in an inferior to superior chronological sequence, with a particular focus on the knee and pelvis.

Biomechanical comparison of 3 latissimus dorsi transfer sites for reverse total shoulder arthroplasty in the absence of teres minor.

BACKGROUND: Reverse total shoulder arthroplasty is often performed to treat rotator cuff tear arthropathy with irreparable rotator cuff tears. Patients with full-thickness tears involving the posterior cuff and teres minor specifically lose active external rotation of the humerus, limiting activities of daily living. A latissimus dorsi transfer has been described as one potential solution, but few studies have compared different latissimus dorsi transfer sites. This study assesses the biomechanics of 3 latissimus dorsi transfer sites, examining external rotation, deltoid muscle forces, and force across the glenoid for specific activities. METHODS: The Newcastle Shoulder Model was modified to include a Delta III reverse shoulder arthroplasty and was used to model the effects of 3 latissimus dorsi transfer sites-anterior, posterolateral, and posterodistal-in the setting of teres minor deficiency. The latissimus dorsi was represented by 5 muscle elements approximating anatomic fascicle divisions. Kinematic data sets representing common activities were input into the model, allowing calculation of muscle forces and glenoid loads. RESULTS: Each of the 3 latissimus dorsi transfer sites demonstrated a change of moment arm from internal to external rotation with increased magnitude. Average maximum deltoid muscle forces and glenoid loading across all motions decreased for each of the 3 transfer sites, with the greatest decreased force noted for the posterior deltoid. This decrease in deltoid force and glenoid loading was significantly greater with the anterior and posterolateral transfer sites, relative to the posterodistal site. DISCUSSION: Latissimus dorsi transfer to all 3 sites in the setting of reverse total shoulder arthroplasty and posterior rotator cuff deficiency resulted in large external rotation moment arms. The transferred latissimus dorsi shared the external rotation load and resulted in decreased deltoid forces and glenoid loading with very small differences between the 3 transfer sites. The posterodistal location reduced deltoid force and glenoid loading to a lower degree compared with other sites, but it also showed that it did not alter the tendon length compared with the native shoulder.

PROMIS Upper Extremity underperforms psychometrically relative to American Shoulder and Elbow Surgeons score in patients undergoing primary rotator cuff repair.

BACKGROUND: A number of patient-reported outcome measure (PROM) instruments are used to assess shoulder pain, function, and postoperative satisfaction. Computer adaptive tests (CATs) have been developed in an effort to tailor question delivery, decrease time to completion and floor/ceiling effects, and increase compliance. Previous investigations have demonstrated excellent correlation between Patient-Reported Outcomes Measurement Information System (PROMIS) Upper Extremity CAT (UE-CAT) and American Shoulder and Elbow Surgeons (ASES) instruments, indicating that the PROMIS UE-CAT may be an acceptable alternative to the ASES. Prior to widespread adoption, however, rigorous psychometric validation must be performed. Thus, the purpose of this study was to evaluate the psychometric properties of PROMIS UE-CAT relative to the ASES score in patients undergoing primary rotator cuff repair (RCR). METHODS: A retrospective study of an institutional RCR registry was performed from July 2018 through March 2019. Preoperative PROMIS UE-CAT and ASES scores were collected. Floor and ceiling effects were determined, and convergent validity was established through Pearson correlations. Rasch partial credit modeling was used for psychometric analysis of the validity of PROMIS UE-CAT and ASES question items, a previously established methodology for assessing shoulder PROMs. Person-item maps were generated to characterize the distribution of question responses along the latent dimension of shoulder disability. RESULTS: Responses from 107 patients (107 shoulders) were included. PROMIS UE-CAT had a strong correlation to ASES (r = 0.684; P < .001), with a 3.7% floor effect and no ceiling effect; ASES had no floor or ceiling effects. PROMIS UE-CAT person-item maps revealed excellent test item coverage between floor and ceiling effects; however, all PROMIS UE-CAT items except "Difficulty putting on a jacket" demonstrated nonsequentially ordered responses, indicating poor reliability of item responses corresponding to shoulder function. Although PROMIS UE-CAT initially demonstrated superior instrument efficiency with fewer test items for overall equivalent coverage of shoulder function assessment, final models after recursive item elimination revealed the ASES instrument to have more well-fitting items over a broader range of shoulder function. CONCLUSION: Despite its overall strong correlation with ASES and equivalent coverage between floor and ceiling effects, the psychometric properties of PROMIS UE-CAT indicate a need for optimization of individual test items to correspond to shoulder function. Until further refinements in the PROMIS UE-CAT instrument are made, it should not replace the ASES instrument in patients undergoing primary RCR.

Biomechanical analysis of anterior stability after 15% glenoid bone loss: comparison of Bankart repair, dynamic anterior stabilization, dynamic anterior stabilization with Bankart repair, and Latarjet.

BACKGROUND: Dynamic anterior shoulder stabilization (DAS) with Bankart repair is a recently described stabilization technique thought to be more robust than an isolated Bankart repair while avoiding many coracoid transfer-related complications and technical demands. DAS involves transfer of the long head biceps through a subscapularis split to the anterior glenoid to create a sling effect. We hypothesize that DAS with Bankart repair will restore anterior stability in a human-cadaveric model with subcritical (15%) glenoid bone loss. METHODS: Eight cadaveric shoulders were tested using an established shoulder simulator to record glenohumeral translations with an accuracy of ±0.2 mm. Shoulders were tested in 5 states-intact soft tissues, Bankart defect with 15% bone loss, isolated Bankart repair, DAS with Bankart repair, isolated DAS, and Latarjet. A 45 N anterior force was applied through the pectoralis major tendon, and translation of the humeral head was recorded and compared with repeated measures analysis of variance. RESULTS: The anterior translation in the intact (native) glenoid was 4.7 mm at neutral position and 4.6 mm at 45° external rotation. Anterior translation significantly increased after introducing a Bankart defect with 15% glenoid bone loss to 9.1 mm (neutral, P = .002) and 9.5 mm (45° external rotation, P < .001). All repair conditions showed a significant decrease in anterior translation relative to Bankart defect. DAS with Bankart repair decreased anterior translation compared with the Bankart defect: 2.7 mm (neutral, P < .001) and 2.1 mm (45° external rotation, P < .001). DAS with Bankart repair significantly decreased anterior translation compared with the isolated Bankart repair (2.7 mm vs. 4.7 mm, P = .023) and the isolated DAS (2.7 mm vs. 4.3 mm, P = .041) in neutral position. The Latarjet procedure resulted in the greatest reduction in anterior translation compared with the Bankart defect: 1.2 mm (neutral, P < .001) and 1.9 mm (45° external rotation, P < .001). CONCLUSIONS: DAS with Bankart repair is a viable alternative to restore anterior glenohumeral stability with a 15% glenoid defect at a greater degree than either DAS or Bankart repair alone. The Latarjet procedure was the most effective in reducing anterior translation but restrained the anterior translation significantly more than the native glenoid.

Backside polyethylene wear in reverse shoulder arthroplasty.

BACKGROUND: Aseptic loosening from implant-associated osteolysis in reverse shoulder arthroplasty (RSA) may contribute to premature implant failure. Although articular side polyethylene (PE) damage has been well documented in the literature, no studies to date have investigated backside wear in RSA. The aims of this investigation were to (1) document and compare the damage between the backside and articular surface in explanted RSA components, (2) assess whether certain quadrants have a greater propensity for damage, and (3) report the most common mode(s) of backside PE damage. METHODS: Twenty-one RSA humeral liners retrieved during revision procedures between 2005 and 2014 were included for analysis. The mean time between implantation and extraction was 16 months (10 days-88 months). Diagnoses at the time of revision included dislocation (10), infection (4), mechanical failure (3), loosening (2), and unknown (2). Liners were examined under light microscopy (×10-30 magnification) and damage on the articular and backside of the liner surface was graded using the modified Hood score. The location and damage modality were compared between the articular side and backside of the implant. RESULTS: Damage was noted on the articular surfaces of all 21 liners and on the backside surface of 20 liners. The total damage in all the quadrants was higher on the articular surface than on the backside of the component, with a mean difference in total quadrant damage scores of 11.74 ± 3.53 (P < .001). There was no difference in damage among the quadrants on the backside (P = .44) or the articular surface (P = .08). The articular side exhibited greater scratching, abrasion, and surface deformation than the backside (P < .001). CONCLUSIONS: This short-term retrieval study demonstrated that backside PE damage occurs on the humeral component of RSA implants. There was greater damage to the articular side of the liner but wear to the backside was present in almost all liners. The clinical importance of backside wear in RSA and its overall contribution to PE particulate disease and osteolysis needs further investigation.

Dynamic estimation of auditory temporal response functions via state-space models with Gaussian mixture process noise.

Estimating the latent dynamics underlying biological processes is a central problem in computational biology. State-space models with Gaussian statistics are widely used for estimation of such latent dynamics and have been successfully utilized in the analysis of biological data. Gaussian statistics, however, fail to capture several key features of the dynamics of biological processes (e.g., brain dynamics) such as abrupt state changes and exogenous processes that affect the states in a structured fashion. Although Gaussian mixture process noise models have been considered as an alternative to capture such effects, data-driven inference of their parameters is not well-established in the literature. The objective of this paper is to develop efficient algorithms for inferring the parameters of a general class of Gaussian mixture process noise models from noisy and limited observations, and to utilize them in extracting the neural dynamics that underlie auditory processing from magnetoencephalography (MEG) data in a cocktail party setting. We develop an algorithm based on Expectation-Maximization to estimate the process noise parameters from state-space observations. We apply our algorithm to simulated and experimentally-recorded MEG data from auditory experiments in the cocktail party paradigm to estimate the underlying dynamic Temporal Response Functions (TRFs). Our simulation results show that the richer representation of the process noise as a Gaussian mixture significantly improves state estimation and capturing the heterogeneity of the TRF dynamics. Application to MEG data reveals improvements over existing TRF estimation techniques, and provides a reliable alternative to current approaches for probing neural dynamics in a cocktail party scenario, as well as attention decoding in emerging applications such as smart hearing aids. Our proposed methodology provides a framework for efficient inference of Gaussian mixture process noise models, with application to a wide range of biological data with underlying heterogeneous and latent dynamics.

Orthopaedic Randomized Controlled Trials Published in General Medical Journals Are Associated With Higher Altmetric Attention Scores and Social Media Attention Than Nonorthopaedic Randomized Controlled Trials.

PURPOSE: To (1) compare the Altmetric Attention Score (AAS) and citation rates between orthopaedic and nonorthopaedic randomized controlled trials (RCTs) from 5 high-impact medical journals and (2) identify general characteristics of these articles associated with greater exposure on social media platforms. METHODS: Articles published in The New England Journal of Medicine (NEJM), Lancet, The Journal of the American Medical Association (JAMA), Annals of Internal Medicine, and Archives of Internal Medicine between January 2011 and December 2016 were analyzed. These journals were selected based on retaining high impact factors with rigorous publication standards and availability of the AAS for their publications. The queried time frame was chosen to balance the inception of the AAS with an optimal period for citation accrual. A total of 14 article characteristics, in addition to number of Tweets, Facebook shares, news mentions, and the AAS, were extracted. Inclusion criteria were orthopaedic RCTs reporting on outcomes after surgical intervention. Linear regression was used to assess the relationship between publication characteristics and the AAS and social media attention. RESULTS: A total of 9 orthopaedic and 59 nonorthopaedic RCTs were included. The mean AASs were significantly different (574 ± 565.7 versus 256.9 ± 222.3, P = .003), whereas citation rate was not (192.2 ± 117.1 versus 382.3 ± 560.3, P = .317). Orthopaedic RCTs had a significantly greater number of mentions on Twitter and Facebook (P < .001). A higher AAS significantly associated with a greater number of citations (ß = 0.75, P = .019) for orthopaedic RCTs. The mean AAS of orthopaedic RCTs favoring nonoperative management (809.6 ± 676.3) was greater than those favoring operative treatment (292.0 ± 248.9) but was not statistically significant (P = .361). CONCLUSION: Orthopaedic RCTs published in 5 high-impact general medical journals had a significantly greater mean AAS relative to nonorthopaedic RCTs, with no differences in citation rates. Additionally, there was a strong association between the AAS and citation rate of orthopaedic RCTs. Orthopaedic RCTs had greater social media exposure on both Twitter and Facebook. CLINICAL RELEVANCE: Orthopaedic surgeons, researchers, and providers who publish RCTs in high-impact medical journals can anticipate extensive social media attention for their articles relative to other nonorthopaedic RCTs in the same journals. Social media attention may be related to operative versus nonoperative management topics. This study provides further evidence for the increasing use of the AAS and its association with citation accrual.

Patient Factors Associated With Clinical Failure Following Arthroscopic Superior Capsular Reconstruction.

PURPOSE: To identify demographic, clinical, and radiographic factors associated with failure after superior capsular reconstruction (SCR). METHODS: Prospectively collected data were analyzed from patients who underwent SCR using a decellularized dermal allograft for an irreparable rotator cuff tear. Demographic characteristics, radiographic findings, concomitant procedures, and patient-reported outcomes (PROs) were recorded. Failure was defined by ≥1 of the following criteria: (1) conversion to reverse total shoulder arthroplasty (RTSA), (2) a decrease in 1-year postoperative shoulder-specific PROs compared with preoperative scores, or (3) patient reports at final follow-up that the shoulder was in a worse condition than before surgery. Preoperative variables were compared between patients meeting the criteria for clinical failure and those who did not. RESULTS: Fifty-four patients (mean age 56.3 ± 5.8 years, range 45 to 70) who underwent SCR, with minimum 1-year follow-up, were included in the analysis. Mean follow-up after surgery was 24 months (range 12 to 53). Eleven patients (20.4%) met criteria for clinical failure. Of the 11, 8 reported decreased American Shoulder and Elbow Surgeons (ASES) or Constant scores or indicated that the operative shoulder was in a worse condition than before surgery. Three patients underwent RTSA in the 6 to 12 months after SCR. Female sex and the presence of a subscapularis tear were associated with failure (P = .023 and P = .029, respectively). A trend toward greater body mass index (BMI), lower preoperative forward flexion, and lower preoperative acromiohumeral distance (AHD) was found in patients with clinical failure (P = .075, P = .088, and P = .081, respectively). No other variable included in the analysis was significantly associated with failure. CONCLUSION: The proportions of female patients and those with subscapularis tear were greater among patients with clinical failure after SCR. Greater BMI, lower preoperative forward flexion, and lower preoperative AHD trended toward association with clinical failure of SCR. LEVEL OF EVIDENCE: 4, case series.

Increased hip arthroscopy operative duration is an independent risk factor for overnight hospital admission.

PURPOSE: The purpose of this study was to determine the association between operative duration and short-term complications as well as overnight hospital admission following hip arthroscopy. METHODS: Hip arthroscopy cases from 2006 to 2016 were retrieved from the National Surgical Quality Improvement Program registry, which prospectively collects 30-day postoperative complications. Patients were stratified into the following groups based on procedure length: group 1 (< 60 min), group 2 (60-120 min), and group 3 (> 120 min). Preoperative characteristics were compared across the cohorts. Multivariate regressions were used to compare complication rates and overnight hospital admission between the three groups. Independent risk factors for overnight hospital admission were characterized. RESULTS: A total of 2129 hip arthroscopy cases were identified. Average operative duration was 99.3 ± 55.7 min. As operative time increased, patients were more likely to be younger, male, and had lower American Society of Anesthesiologists (ASA) class (p < 0.001). Body mass index and comorbidity profiles were similar across the patient cohorts, with the exception of hypertension being more prevalent in the shorter operative time cohort (p < 0.001). Patients in group 3 were more likely to stay overnight in the hospital (26.0%) compared to patients in groups 1 (7.7%) and 2 (10.9%), p < 0.001). All postoperative complication rates were otherwise similar between the cohorts. Independent risk factors for overnight hospital admission included increasing operative time (most notably > 120 min relative to < 60 min, relative risk [RR] = 3.53, 95% CI 2.50-5.00, p < 0.001) and increasing ASA classification (most notably ASA III or IV relative to ASA I, RR = 1.64, 95% CI 1.18-2.27; p = 0.013). CONCLUSIONS: Increasing operative duration was not associated with increased postoperative complications following hip arthroscopy. However, patients were more than three times likely to stay in the hospital overnight if their surgery was longer than 120 min, relative to cases that were less than 60 min. LEVEL OF EVIDENCE: III.

Outcomes are comparable using free bone block autografts versus allografts for the management of anterior shoulder instability with glenoid bone loss: a systematic review and meta-analysis of "The Non-Latarjet".

PURPOSE: Glenoid augmentation using free bone blocks for anterior shoulder instability has been proposed as an alternative to or bail-out for the Latarjet procedure. The purpose of this investigation was to systematically review and compare outcomes of patients undergoing glenoid augmentation using free bone block autografts versus allografts. METHODS: A systematic review using PubMed, MEDLINE, Embase, and the Cochrane Library databases was performed in line with the PRISMA statement. Studies reporting outcomes of patients treated with free bone block procedures for anterior shoulder instability with minimum 2-year follow-up were included. Random effects modelling was used to compare patient-reported outcomes, return to sports, recurrent instability, non-instability related complications, and development of arthritis between free bone block autografts and allografts. RESULTS: Eighteen studies comprising of 623 patients met the inclusion criteria for this investigation. There were six studies reporting on the use of allografts (of these, two used distal tibial, three iliac crest, and one femoral head allograft) in 173 patients and twelve studies utilizing autografts (of these, ten used iliac crest and two used free coracoid autograft) in 450 patients. Mean age was 28.7 ± 4.1 years for the allograft group and 27.8 ± 3.8 years for the autograft group (n.s). Mean follow-up was 98 months in autograft studies and 50.8 months for allograft studies (range 24-444 months, n.s). Overall mean increase in Rowe score was 56.2 with comparable increases between autografts and allografts (n.s). Pooled recurrent instability rates were 3% (95% CI, 1-7%; I2 = 77%) and did not differ between the groups (n.s). Arthritic progression was evident in 11% of autografts (95% CI, 2-27%; I2 = 90%) and 1% (95% CI, 0-8%; I2 = 63%) of allografts (n.s). The overall incidence of non-instability related complications was 5% (95% CI, 2-10%; I2 = 81%) and was similar between the groups (n.s). Pooled return to sports rate was 88% (95% CI, 76-96%; I2 = 76%). CONCLUSION: Glenoid augmentation using free bone block autograft or allograft in the setting of recurrent anterior shoulder instability with glenoid bone loss is effective and safe. Outcomes and complication incidence using autografts and allografts were comparable. Due to the high degree of heterogeneity in the data and outcomes reported in available studies, which consist primarily of retrospective case series, future prospective trials investigating long-term outcomes using free bone block autograft versus allograft for anterior shoulder instability with glenoid bone loss are warranted. LEVEL OF EVIDENCE: IV.

Development of supervised machine learning algorithms for prediction of satisfaction at 2 years following total shoulder arthroplasty.

BACKGROUND: Patient satisfaction after primary anatomic and reverse total shoulder arthroplasty (TSA) represents an important metric for gauging patients' perception of their care and surgical outcomes. Although TSA confers improvement in pain and function for most patients, inevitably some will remain unsatisfied postoperatively. The purpose of this study was to (1) train supervised machine learning (SML) algorithms to predict satisfaction after TSA and (2) develop a clinical tool for individualized assessment of patient-specific risk factors. METHODS: We performed a retrospective review of primary anatomic and reverse TSA patients between January 2014 and February 2018. A total of 16 demographic, clinical, and patient-reported outcomes were evaluated for predictive value. Five SML algorithms underwent 3 iterations of 10-fold cross-validation on a training set (80% of cohort). Assessment by discrimination, calibration, Brier score, and decision-curve analysis was performed on an independent testing set (remaining 20% of cohort). Global and local model behaviors were evaluated with global variable importance plots and local interpretable model-agnostic explanations, respectively. RESULTS: The study cohort consisted of 413 patients, of whom 331 (82.6%) were satisfied at 2 years postoperatively. The support vector machine model demonstrated the best relative performance on the independent testing set not used for model training (concordance statistic, 0.80; calibration intercept, 0.20; calibration slope, 2.32; Brier score, 0.11). The most important factors for predicting satisfaction were baseline Single Assessment Numeric Evaluation score, exercise and activity, workers' compensation status, diagnosis, symptom duration prior to surgery, body mass index, age, smoking status, anatomic vs. reverse TSA, and diabetes. The support vector machine algorithm was incorporated into an open-access digital application for patient-level explanations of risk and predictions, available at https://orthopedics.shinyapps.io/SatisfactionTSA/. CONCLUSION: The best-performing SML model demonstrated excellent discrimination and adequate calibration for predicting satisfaction following TSA and was used to create an open-access, clinical decision-making tool. However, rigorous external validation in different geographic locations and patient populations is essential prior to assessment of clinical utility. Given that this tool is based on partially modifiable risk factors, it may enhance shared decision making and allow for periods of targeted preoperative health-optimization efforts.

Establishing clinically significant outcomes of the Patient-Reported Outcomes Measurement Information System Upper Extremity questionnaire after primary reverse total shoulder arthroplasty.

BACKGROUND AND HYPOTHESIS: Since its introduction, the Patient-Reported Outcomes Measurement Information System Upper Extremity (PROMIS UE) assessment has been increasingly used in shoulder arthroplasty outcome measurement. However, determination of clinically significant outcomes using the PROMIS UE has yet to be investigated following reverse total shoulder arthroplasty (RTSA). We hypothesized that we could establish clinically significant outcomes of the PROMIS UE outcome assessment in patients undergoing primary RTSA and identify significant baseline patient factors associated with achievement of these measures. METHODS: Consecutive patients undergoing primary RTSA between 2018 and 2019 who received preoperative baseline and follow-up PROMIS UE assessments at 12 months after surgery were retrospectively reviewed. Domain-specific anchor questions pertaining to pain and function assessed at 12 months after surgery were used to determine minimal clinically important difference (MCID), substantial clinical benefit (SCB), and patient acceptable symptomatic state (PASS) values for the PROMIS UE using receiver operating characteristic curve and area-under-the-curve (AUC) analysis. Univariate logistic regression analysis was then performed to identify significant patient factors associated with achieving the MCID, SCB, or PASS. RESULTS: A total of 95 patients met all inclusion criteria and were included in the analysis. By use of an anchor-based method, the PASS value was 36.68 (sensitivity, 0.795; specificity, 0.765; AUC, 0.793) and the SCB value was 11.62 (sensitivity, 0.597; specificity, 1.00; AUC, 0.806). By use of a distribution-based method, the MCID value was calculated to be 4.27. Higher preoperative PROMIS UE scores were a positive predictor in achievement of the PASS (odds ratio [OR], 1.107; P = .05), whereas lower preoperative PROMIS UE scores were associated with obtaining SCB (OR, 0.787; P < .001). Greater baseline forward flexion was negatively associated with achievement of the PASS (OR, 0.986; P = .033) and MCID (OR, 0.976, P = .013). Of the patients, 83.2%, 69.5%, and 47.4% achieved the MCID, PASS, and SCB, respectively. CONCLUSION: This study defines the MCID, SCB, and PASS for the PROMIS UE outcome assessment in patients undergoing primary RTSA, of whom the majority achieved meaningful outcome improvement at 12 months after surgery. These values may be used in assessing the outcomes and extent of functional improvement following RTSA.