Search

Background: We present an automated image ingestion pipeline for a knee radiography registry, integrating a multilabel image-semantic classifier with conformal prediction-based uncertainty quantification and an object detection model for knee hardware., Methods: Annotators retrospectively classified 26,000 knee images detailing presence, laterality, prostheses, and radiographic views. They further annotated surgical construct locations in 11,841 knee radiographs. An uncertainty-aware multilabel EfficientNet-based classifier was trained to identify the knee laterality, implants, and radiographic view. A classifier trained with embeddings from the EfficientNet model detected out-of-domain images. An object detection model was trained to identify 20 different knee implants. Model performance was assessed against a held-out internal and an external dataset using per-class F1 score, accuracy, sensitivity, and specificity. Conformal prediction was evaluated with marginal coverage and efficiency., Results: Classification Model with Conformal Prediction: F1 scores for each label output > 0.98. Coverage of each label output was > 0.99 and the average efficiency was 0.97. Domain Detection Model:The F1 score was 0.99, with precision and recall for knee radiographs of 0.99. Object Detection Model:Mean average precision across all classes was 0.945 and ranged from 0.695 to 1.000. Average precision and recall across all classes were 0.950 and 0.886., Conclusions: We present a multilabel classifier with domain detection and an object detection model to characterize knee radiographs. Conformal prediction enhances transparency in cases when the model is uncertain., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Background: Soft tissue management in total hip arthroplasty includes appropriate restoration and/or alteration of leg length (LL) and offset to re-establish natural hip biomechanics. The purpose of this study was to evaluate the effect of LL and offset-derived variables in a multivariable survival model for dislocation., Methods: Clinical, surgical, and radiographic data was retrospectively acquired for 12,582 patients undergoing primary total hip arthroplasty at a single institution from 1998 to 2018. There were twelve variables derived from preoperative and postoperative radiographs related to LL and offset that were measured using a validated automated algorithm. These measurements, as well as other modifiable and nonmodifiable surgical, clinical, and demographic factors, were used to determine hazard ratios for dislocation risk., Results: None of the LL or offset variables conferred significant risk or protective benefit for dislocation risk. By contrast, all other variables included in the multivariable model demonstrated a statistically significant effect on dislocation risk with a minimum effect size of 28% (range 0.72 to 1.54) (sex, surgical approach, acetabular liner type, femoral head size, neurologic disease, spine disease, and prior spine surgery)., Conclusions: Contrary to traditional teaching and our hypothesis, operative changes in LL and offset did not demonstrate any clinically or statistically significant effect in this large and well-characterized cohort. This does not imply that these variables are not important in individual cases, but rather suggests the overall impact of LL and offset changes is relatively minor for dislocation risk compared to other variables that were found to be highly clinically and statistically significant in this population. These results may also suggest that surgeons do a good job of restoring native LL and offset for patients, which may mitigate their analyzed impact., (Copyright © 2024 The Author(s). Published by Elsevier Inc. All rights reserved.)

Background: Previous studies have suggested that wound complications may differ by surgical approach after total hip arthroplasty (THA), with particular attention toward the direct anterior approach (DAA). However, there is a paucity of data documenting wound complication rates by surgical approach and the impact of concomitant patient factors, namely body mass index (BMI). This investigation sought to determine the rates of wound complications by surgical approach and identify BMI thresholds that portend differential risk., Methods: This multicenter study retrospectively evaluated all primary THA patients from 2010 to 2023. Patients were classified by skin incision as having a laterally based approach (posterior or lateral approach) or DAA (longitudinal incision). We identified 17,111 patients who had 11,585 laterally based (68%) and 5,526 (32%) DAA THAs. The mean age was 65 years (range, 18 to 100), 8,945 patients (52%) were women, and the mean BMI was 30 (range, 14 to 79). Logistic regression and cut-point analyses were performed to identify an optimal BMI cutoff, overall and by approach, with respect to the risk of wound complications at 90 days., Results: The 90-day risk of wound complications was higher in the DAA group versus the laterally based group, with an absolute risk of 3.6% versus 2.6% and a multivariable adjusted odds ratio of 1.5 (P < .001). Cut-point analyses demonstrated that the risk of wound complications increased steadily for both approaches, but most markedly above a BMI of 33., Conclusions: Wound complications were higher after longitudinal incision DAA THA compared to laterally based approaches, with a 1% higher absolute risk and an adjusted odds ratio of 1.5. Furthermore, BMI was an independent risk factor for wound complications regardless of surgical approach, with an optimal cut-point BMI of 33 for both approaches. These data can be used by surgeons to help consider the risks and benefits of approach selection., Level of Evidence: Level III., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Background: Intraprosthetic dissociation (IPD) is a complication unique to dual mobility (DM) implants where the outer polyethylene head dissociates from the inner femoral head. Increasing reports of IPD at the time of closed reduction of large head DM dislocations prompted this biomechanical study evaluating the assembly and dissociation forces of DM heads., Methods: We tested 17 polyethylene DM heads from 5 vendors. Of the heads, 12 were highly cross-linked polyethylene (4 vendors) and 5 were infused with vitamin E (2 vendors). Heads were between 46 and 47 mm in diameter, accepting a 28 mm-inner ceramic head. Implants were assembled and disassembled using a servohydraulic machine that recorded the forces and torques applied during testing. Dissociation was tested via both axial pull-out and lever-out techniques, where lever-out simulated stem-on-acetabular component impingement., Results: The initial maximum assembly force was significantly different between all vendors (P < .01) and decreased for all implants with subsequent assembly. Vendor 4-E (Link with vitamin E) heads required the highest assembly force (1,831.9 ± 81.95 N), followed by Vendor 3 (Smith & Nephew), Vendor 5 (DePuy Synthes), Vendor 1-E (Zimmer Biomet with vitamin E), Vendor 2 (Stryker), and Vendor 1 (Zimmer Biomet Arcom). Vendor 4-E implants showed the greatest dissociation resistance in both pull-out (2,059.89 N, n = 1) and lever-out (38.95 ± 2.79 Nm) tests. Vendor 1-E implants with vitamin E required higher assembly force, dissociation force, and energy than Vendor 1 heads without vitamin E., Conclusions: There were notable differences in DM assembly and dissociation forces between implants. Diminishing force was required for assembly with each additional trial across vendors. Vendor 4-E DM heads required the highest assembly and dissociation forces. Vitamin E appeared to increase the assembly and dissociation forces. Based on these results, DM polyethylene heads should not be reimplanted after dissociation, and there may be a role for establishing a minimum dissociation energy standard to minimize IPD risk., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Background: Our previously reported randomized clinical trial of direct anterior approach (DAA) versus mini-posterior approach (MPA) total hip arthroplasty showed slightly faster initial recovery for patients who had a DAA and no differences in complications or clinical or radiographic outcomes beyond 8 weeks. The aims of the current study were to determine if early advantages of DAA led to meaningful clinical differences beyond 5 years and to identify differences in midterm complications., Methods: Of the 101 original patients, 93 were eligible for follow-up at a mean of 7.5 years (range, 2.1 to 10). Clinical outcomes were compared with Harris Hip, 12-Item Short Form Health Survey, and Hip Disability and Osteoarthritis Outcomes Scores (HOOS) scores and subscores, complications, reoperations, and revisions., Results: Harris Hip scores were similar (95.3 ± 6.0 versus 93.5 ± 10.3 for DAA and MPA, respectively, P = .79). The 12-Item Short Form Health Survey physical and mental scores were similar (46.2 ± 9.3 versus 46.2 ± 10.6, P = .79, and 52.3 ± 7.1 versus 55.2 ± 4.5, P = .07 in the DAA and MPA groups, respectively). The HOOS scores were similar (97.4 ± 7.9 versus 96.3 ± 6.7 for DAA and MPA, respectively, P = .07). The HOOS quality of life subscores were 96.9 ± 10.8 versus 92.3 ± 16.0 for DAA and MPA, respectively (P = .046). No clinical outcome met the minimally clinically important difference. There were 4 surgical complications in the DAA group (1 femoral loosening requiring revision, 1 dislocation treated closed, and 2 wound dehiscences requiring debridement), and 6 surgical complications in the MPA group (3 dislocations, 2 treated closed, and 1 revised to dual mobility; 2 intraoperative fractures treated with a cable; and 1 wound dehiscence treated nonoperatively)., Conclusions: At a mean of 7.5 years, this randomized clinical trial demonstrated no clinically meaningful differences in outcomes, complications, reoperations, or revisions between DAA and MPA total hip arthroplasty., Level of Evidence: IV., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Background: Revision total hip arthroplasty (THA) requires preoperatively identifying in situ implants, a time-consuming and sometimes unachievable task. Although deep learning (DL) tools have been attempted to automate this process, existing approaches are limited by classifying few femoral and zero acetabular components, only classify on anterior-posterior (AP) radiographs, and do not report prediction uncertainty or flag outlier data., Methods: This study introduces Total Hip Arhtroplasty Automated Implant Detector (THA-AID), a DL tool trained on 241,419 radiographs that identifies common designs of 20 femoral and 8 acetabular components from AP, lateral, or oblique views and reports prediction uncertainty using conformal prediction and outlier detection using a custom framework. We evaluated THA-AID using internal, external, and out-of-domain test sets and compared its performance with human experts., Results: THA-AID achieved internal test set accuracies of 98.9% for both femoral and acetabular components with no significant differences based on radiographic view. The femoral classifier also achieved 97.0% accuracy on the external test set. Adding conformal prediction increased true label prediction by 0.1% for acetabular and 0.7 to 0.9% for femoral components. More than 99% of out-of-domain and >89% of in-domain outlier data were correctly identified by THA-AID., Conclusions: The THA-AID is an automated tool for implant identification from radiographs with exceptional performance on internal and external test sets and no decrement in performance based on radiographic view. Importantly, this is the first study in orthopedics to our knowledge including uncertainty quantification and outlier detection of a DL model., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: This study introduces THA-Net, a deep learning inpainting algorithm for simulating postoperative total hip arthroplasty (THA) radiographs from a single preoperative pelvis radiograph input, while being able to generate predictions either unconditionally (algorithm chooses implants) or conditionally (surgeon chooses implants)., Methods: The THA-Net is a deep learning algorithm which receives an input preoperative radiograph and subsequently replaces the target hip joint with THA implants to generate a synthetic yet realistic postoperative radiograph. We trained THA-Net on 356,305 pairs of radiographs from 14,357 patients from a single institution's total joint registry and evaluated the validity (quality of surgical execution) and realism (ability to differentiate real and synthetic radiographs) of its outputs against both human-based and software-based criteria., Results: The surgical validity of synthetic postoperative radiographs was significantly higher than their real counterparts (mean difference: 0.8 to 1.1 points on 10-point Likert scale, P < .001), but they were not able to be differentiated in terms of realism in blinded expert review. Synthetic images showed excellent validity and realism when analyzed with already validated deep learning models., Conclusion: We developed a THA next-generation templating tool that can generate synthetic radiographs graded higher on ultimate surgical execution than real radiographs from training data. Further refinement of this tool may potentiate patient-specific surgical planning and enable technologies such as robotics, navigation, and augmented reality (an online demo of THA-Net is available at: https://demo.osail.ai/tha&#95;net)., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Implementing tools that identify cost-saving opportunities for ambulatory orthopaedic surgeries can improve access to value-based care. We developed and internally validated a machine learning (ML) algorithm to predict cost drivers of total charges after ambulatory unicompartmental knee arthroplasty (UKA)., Methods: We queried the New York State Ambulatory Surgery and Services database to identify patients who underwent ambulatory, defined as <24 hours of care before discharge, elective UKA between 2014 and 2016. A total of 1,311 patients were included. The median costs after ambulatory UKA were $14,710. Patient demographics and intraoperative parameters were entered into 4 candidate ML algorithms. The most predictive model was selected following internal validation of candidate models, with conventional linear regression as a benchmark. Global variable importance and partial dependence curves were constructed to determine the impact of each input parameter on total charges., Results: The gradient-boosted ensemble model outperformed all candidate algorithms and conventional linear regression. The major differential cost drivers of UKA identified (in decreasing order of magnitude) were increased operating room time, length of stay, use of regional and adjunctive periarticular analgesia, utilization of computer-assisted navigation, and routinely sending resected tissue to pathology., Conclusion: We developed and internally validated a supervised ML algorithm that identified operating room time, length of stay, use of computer-assisted navigation, regional primary anesthesia, adjunct periarticular analgesia, and routine surgical pathology as essential cost drivers of UKA. Following external validation, this tool may enable surgeons and health insurance providers optimize the delivery of value-based care to patients receiving outpatient UKA., Level of Evidence: III., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Electronic health records have facilitated the extraction and analysis of a vast amount of data with many variables for clinical care and research. Conventional regression-based statistical methods may not capture all the complexities in high-dimensional data analysis. Therefore, researchers are increasingly using machine learning (ML)-based methods to better handle these more challenging datasets for the discovery of hidden patterns in patients' data and for classification and predictive purposes. This article describes commonly used ML methods in structured data analysis with examples in orthopedic surgery. We present practical considerations in starting an ML project and appraising published studies in this field., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Total joint arthroplasty is becoming one of the most common surgeries within the United States, creating an abundance of analyzable data to improve patient experience and outcomes. Unfortunately, a large majority of this data is concealed in electronic health records only accessible by manual extraction, which takes extensive time and resources. Natural language processing (NLP), a field within artificial intelligence, may offer a viable alternative to manual extraction. Using NLP, a researcher can analyze written and spoken data and extract data in an organized manner suitable for future research and clinical use. This article will first discuss common subtasks involved in an NLP pipeline, including data preparation, modeling, analysis, and external validation, followed by examples of NLP projects. Challenges and limitations of NLP will be discussed, closing with future directions of NLP projects, including large language models., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: In this work, we applied and validated an artificial intelligence technique known as generative adversarial networks (GANs) to create large volumes of high-fidelity synthetic anteroposterior (AP) pelvis radiographs that can enable deep learning (DL)-based image analyses, while ensuring patient privacy., Methods: AP pelvis radiographs with native hips were gathered from an institutional registry between 1998 and 2018. The data was used to train a model to create 512 × 512 pixel synthetic AP pelvis images. The network was trained on 25 million images produced through augmentation. A set of 100 random images (50/50 real/synthetic) was evaluated by 3 orthopaedic surgeons and 2 radiologists to discern real versus synthetic images. Two models (joint localization and segmentation) were trained using synthetic images and tested on real images., Results: The final model was trained on 37,640 real radiographs (16,782 patients). In a computer assessment of image fidelity, the final model achieved an "excellent" rating. In a blinded review of paired images (1 real, 1 synthetic), orthopaedic surgeon reviewers were unable to correctly identify which image was synthetic (accuracy = 55%, Kappa = 0.11), highlighting synthetic image fidelity. The synthetic and real images showed equivalent performance when they were assessed by established DL models., Conclusion: This work shows the ability to use a DL technique to generate a large volume of high-fidelity synthetic pelvis images not discernible from real imaging by computers or experts. These images can be used for cross-institutional sharing and model pretraining, further advancing the performance of DL models without risk to patient data safety., Level of Evidence: Level III., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Background: Automatic methods for labeling and segmenting pelvis structures can improve the efficiency of clinical and research workflows and reduce the variability introduced with manual labeling. The purpose of this study was to develop a single deep learning model to annotate certain anatomical structures and landmarks on antero-posterior (AP) pelvis radiographs., Methods: A total of 1,100 AP pelvis radiographs were manually annotated by 3 reviewers. These images included a mix of preoperative and postoperative images as well as a mix of AP pelvis and hip images. A convolutional neural network was trained to segment 22 different structures (7 points, 6 lines, and 9 shapes). Dice score, which measures overlap between model output and ground truth, was calculated for the shapes and lines structures. Euclidean distance error was calculated for point structures., Results: Dice score averaged across all images in the test set was 0.88 and 0.80 for the shape and line structures, respectively. For the 7-point structures, average distance between real and automated annotations ranged from 1.9 mm to 5.6 mm, with all averages falling below 3.1 mm except for the structure labeling the center of the sacrococcygeal junction, where performance was low for both human and machine-produced labels. Blinded qualitative evaluation of human and machine produced segmentations did not reveal any drastic decrease in performance of the automatic method., Conclusion: We present a deep learning model for automated annotation of pelvis radiographs that flexibly handles a variety of views, contrasts, and operative statuses for 22 structures and landmarks., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Identifying ambulatory surgical candidates at risk for adverse surgical outcomes can optimize outcomes. The purpose of this study was to develop and internally validate a machine learning (ML) algorithm to predict contributors to unexpected hospitalizations after ambulatory unicompartmental knee arthroplasty (UKA)., Methods: A total of 2,521 patients undergoing UKA from 2006 to 2018 were retrospectively evaluated. Patients admitted overnight postoperatively were identified as those who had a length of stay ≥ 1 day were analyzed with four individual ML models (ie, random forest, extreme gradient boosting, adaptive boosting, and elastic net penalized logistic regression). An additional model was produced as a weighted ensemble of the four individual algorithms. Area under the receiver operating characteristics (AUROC) compared predictive capacity of these models to conventional logistic regression techniques., Results: Of the 2,521 patients identified, 103 (4.1%) required at least one overnight stay following ambulatory UKA. The ML ensemble model achieved the best performance based on discrimination assessed via internal validation (AUROC = 87.3), outperforming individual models and conventional logistic regression (AUROC = 81.9-85.7). The variables determined most important by the ensemble model were cumulative time in the operating room, utilization of general anesthesia, increasing age, and patient residency in more urban areas. The model was integrated into a web-based open-access application., Conclusion: The ensemble gradient-boosted ML algorithm demonstrated the highest performance in identifying factors contributing to unexpected hospitalizations in patients receiving UKA. This tool allows physicians and healthcare systems to identify patients at a higher risk of needing inpatient care after UKA., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Studies developing predictive models from large datasets to risk-stratify patients under going revision total hip arthroplasties (rTHAs) are limited. We used machine learning (ML) to stratify patients undergoing rTHA into risk-based subgroups., Methods: We retrospectively identified 7,425 patients who underwent rTHA from a national database. An unsupervised random forest algorithm was used to partition patients into high-risk and low-risk strata based on similarities in rates of mortality, reoperation, and 25 other postoperative complications. A risk calculator was produced using a supervised ML algorithm to identify high-risk patients based on preoperative parameters., Results: There were 3,135 and 4,290 patients identified in the high-risk and low-risk subgroups, respectively. Each group significantly differed by rate of 30-day mortalities, unplanned reoperations/readmissions, routine discharges, and hospital lengths of stay (P < .05). An Extreme Gradient Boosting algorithm identified preoperative platelets < 200, hematocrit > 35 or < 20, increasing age, albumin < 3, international normalized ratio > 2, body mass index > 35, American Society of Anesthesia class ≥ 3, blood urea nitrogen > 50 or < 30, creatinine > 1.5, diagnosis of hypertension or coagulopathy, and revision for periprosthetic fracture and infection as predictors of high risk., Conclusion: Clinically meaningful risk strata in patients undergoing rTHA were identified using an ML clustering approach. Preoperative labs, demographics, and surgical indications have the greatest impact on differentiating high versus low risk., Level of Evidence: III., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Image data has grown exponentially as systems have increased their ability to collect and store it. Unfortunately, there are limits to human resources both in time and knowledge to fully interpret and manage that data. Computer Vision (CV) has grown in popularity as a discipline for better understanding visual data. Computer Vision has become a powerful tool for imaging analytics in orthopedic surgery, allowing computers to evaluate large volumes of image data with greater nuance than previously possible. Nevertheless, even with the growing number of uses in medicine, literature on the fundamentals of CV and its implementation is mainly oriented toward computer scientists rather than clinicians, rendering CV unapproachable for most orthopedic surgeons as a tool for clinical practice and research. The purpose of this article is to summarize and review the fundamental concepts of CV application for the orthopedic surgeon and musculoskeletal researcher., (Copyright © 2023 Elsevier Inc. All rights reserved.)

The growth of artificial intelligence combined with the collection and storage of large amounts of data in the electronic medical record collection has created an opportunity for orthopedic research and translation into the clinical environment. Machine learning (ML) is a type of artificial intelligence tool well suited for processing the large amount of available data. Specific areas of ML frequently used by orthopedic surgeons performing total joint arthroplasty include tabular data analysis (spreadsheets), medical imaging processing, and natural language processing (extracting concepts from text). Previous studies have discussed models able to identify fractures in radiographs, identify implant type in radiographs, and determine the stage of osteoarthritis based on walking analysis. Despite the growing popularity of ML, there are limitations including its reliance on "good" data, potential for overfitting, long life cycle for creation, and ability to only perform one narrow task. This educational article will further discuss a general overview of ML, discussing these challenges and including examples of successfully published models., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: The hip-spine relationship is increasingly recognized as critical for optimizing stability following total hip arthroplasty (THA). However, these measurements are not routinely obtained during THA workup. It has been suggested that insight can be gained from supine antero-posterior pelvis radiograph, measuring the distance from the superior border of the pubic symphysis to the sacro-coccygeal joint (PSCD). This study assessed the correlation between PSCD and lateral lumbar radiographic metrics in a cohort of preoperative THA patients., Methods: We retrospectively evaluated 250 consecutive patients who underwent THA with preoperative supine antero-posterior pelvis and lateral lumbar radiographs. The mean age was 68 years (range, 42 to 89), 61% were women, and the mean body mass index was 30 kg/m 2 (range, 19 to 52). Two reviewers measured PSCD, pelvic tilt (PT), sacral slope (SS), pelvic incidence (PI), and lumbar lordosis (LL). Inter-observer reliability was calculated for all measurements, and correlation coefficients were calculated for PSCD with respect to PT, SS, PI, and LL., Results: Correlations between PSCD and lumbar radiographic metrics were all statistically significant, except for PI in men but graded as "weak" or "very weak" for men and women, respectively, as follows: PT = -0.30 (P < .01) and -0.46 (P < .01); SS = 0.27 (P < .01) and 0.22 (P < .01); PI = -0.04 (P = .70) and -0.19 (P = .02); and LL = 0.45 (P < .01) and 0.30 (P < .01). Inter-observer reliability was graded as "strong" for every metric., Conclusion: The PSCD was weakly correlated with all evaluated lateral lumbar radiographic metrics in both sexes, despite strong inter-observer reliability. Therefore, PSCD cannot reliably serve as a proxy for evaluating the hip-spine relationship., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Direct anterior approach (DAA) total hip arthroplasty (THA) has been associated with higher rates of superficial incisional dehiscence. However, limited data are available regarding the outcomes following initial treatment of this complication. This study aimed to evaluate patient risk factors, reoperations, and revisions in those who developed superficial wound dehiscence following DAA THA., Methods: We identified 3,687 patients who underwent a primary DAA THA between 2010 and 2019 from our enterprise total joint registry. Of these, 98 (2.7%) patients developed a superficial wound dehiscence requiring intervention [irrigation and debridement (n = 42) or wound care with or without antibiotics (n = 56)]. Dehiscence was noted at a median of 27 (range, 2-105) days. These patients were compared to patients who did not have a superficial wound complication (n = 3,589). Landmark survivorship analysis was performed to account for immortal time bias with a 45-day landmark time., Results: Patients who had superficial wound dehiscence compared to those who did not, were more often women (64 versus 53%, P = .02) and had increased mean body mass index (33 versus 29, P < .001). There was no difference in 4-year survivorship free from any revision between cohorts (97 versus 98%, respectively, P = .14). There were 2 (2.0%) revisions in the superficial dehiscence group: 1 for periprosthetic joint infection and 1 for aseptic femoral loosening., Conclusion: Superficial wound dehiscence following DAA THA was associated with higher body mass index and was more common in women. Fortunately, with proper index management, the risk of revision THA and periprosthetic joint infection was not increased for these patients., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Spinal anesthesia is increasingly used in complex patient populations including revision total hip arthroplasties (THAs). This study aimed to investigate the pain control, length of stay (LOS), and complications associated with spinal versus general anesthesia in a large institutional series of revision THAs., Methods: We retrospectively identified 4,767 revision THAs (4,533 patients) from 2001 to 2016 using our institutional total joint registry. Of these cases, 86% had general and 14% had spinal anesthesia. Demographics between groups were similar with mean age of 66 years, 52% women, and mean body mass index of 29. Complications including all-cause rerevisions and reoperations were studied. Data were analyzed using an inverse probability of treatment weighted model based on propensity score that accounted for patient and surgical factors. The mean follow-up was 7 years., Results: Patients treated with spinal anesthesia required fewer postoperative oral morphine equivalents (P < .001) and had lower numeric pain rating scale scores (P < .001). Spinal anesthesia had a decreased LOS (4.2 versus 4.8 days; P = .007), fewer cases of altered mental status (odds ratio (OR) 3.1, P = .001), fewer blood transfusions (OR 2.3, P < .001), fewer intensive care unit admissions (OR 2.3, P < .001), fewer rerevisions (OR 1.6, P = .04), and fewer reoperations (OR 1.5, P = .02)., Conclusion: Spinal anesthesia was associated with lower oral morphine equivalent use and reduced LOS in this large cohort of revision THAs. Furthermore, spinal anesthesia was associated with fewer cases of altered mental status, transfusion, intensive care unit admission, rerevision, and reoperation after accounting for numerous patient and operative factors., Level of Evidence: Level III, Retrospective Comparative Study., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Many risk factors have been described for periprosthetic femur fracture (PPFFx) following total hip arthroplasty (THA), yet a patient-specific risk assessment tool remains elusive. The purpose of this study was to develop a high-dimensional, patient-specific risk-stratification nomogram that allows dynamic risk modification based on operative decisions., Methods: We evaluated 16,696 primary nononcologic THAs performed between 1998 and 2018. During a mean 6-year follow-up, 558 patients (3.3%) sustained a PPFFx. Patients were characterized by individual natural language processing-assisted chart review on nonmodifiable factors (demographics, THA indication, and comorbidities), and modifiable operative decisions (femoral fixation [cemented/uncemented], surgical approach [direct anterior, lateral, and posterior], and implant type [collared/collarless]). Multivariable Cox regression models and nomograms were developed with PPFFx as a binary outcome at 90 days, 1 year, and 5 years, postoperatively., Results: Patient-specific PPFFx risk based on comorbid profile was wide-ranging from 0.4-18% at 90 days, 0.4%-20% at 1 year, and 0.5%-25% at 5 years. Among 18 evaluated patient factors, 7 were retained in multivariable analyses. The 4 significant nonmodifiable factors included the following: women (hazard ratio (HR) = 1.6), older age (HR = 1.2 per 10 years), diagnosis of osteoporosis or use of osteoporosis medications (HR = 1.7), and indication for surgery other than osteoarthritis (HR = 2.2 for fracture, HR = 1.8 for inflammatory arthritis, HR = 1.7 for osteonecrosis). The 3 modifiable surgical factors were included as follows: uncemented femoral fixation (HR = 2.5), collarless femoral implants (HR = 1.3), and surgical approach other than direct anterior (lateral HR = 2.9, posterior HR = 1.9)., Conclusion: This patient-specific PPFFx risk calculator demonstrated a wide-ranging risk based on comorbid profile and enables surgeons to quantify risk mitigation based on operative decisions., Level of Evidence: Level III, Prognostic., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: The advent of highly porous ingrowth surfaces and highly crosslinked polyethylene has been expected to improve implant survivorship in revision total hip arthroplasty. Therefore, we sought to evaluate the survival of several contemporary acetabular designs following revision total hip arthroplasty., Methods: Acetabular revisions performed from 2000 to 2019 were identified from our institutional total joint registry. We studied 3,348 revision hips, implanted with 1 of 7 cementless acetabular designs. These were paired with highly crosslinked polyethylene or dual-mobility liners. A historical series of 258 Harris-Galante-1 components, paired with conventional polyethylene, was used as reference. Survivorship analyses were performed. For the 2,976 hips with minimum 2-year follow-up, the median follow-up was 8 years (range, 2 to 35 years)., Results: Contemporary components with adequate follow-up had survivorship free of acetabular rerevision of ≥95% at 10-year follow-up. Relative to Harris-Galante-1 components, 10-year survivorship free of all-cause acetabular cup rerevision was significantly higher in Zimmer Trabecular Metarevision (hazard ratio (HR) 0.3, 95% confidence interval (CI) 0.2-0.45), Zimmer Trabecular MetaModular (HR 0.34, 95% CI 0.13-0.89), Zimmer Trilogy (HR 0.4, 95% CI 0.24-0.69), DePuy Pinnacle Porocoat (HR 0.24, 95% CI 0.11-0.51), and Stryker Tritanium revision (HR 0.46, 95% CI 0.24-0.91) shells. Among contemporary components, there were only 23 rerevisions for acetabular aseptic loosening and no rerevisions for polyethylene wear., Conclusion: Contemporary acetabular ingrowth and bearing surfaces were associated with no rerevisions for wear and aseptic loosening was uncommon, particularly with highly porous designs. Therefore, it appears that contemporary revision acetabular components have dramatically improved upon historical results at available follow-up., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: The role of medications to prevent arthrofibrosis following total knee arthroplasty (TKA) remains unclear. We investigated the effect of common oral medications with reported antifibrotic properties on preventing arthrofibrosis and manipulation under anesthesia (MUA) following primary TKA., Methods: Using our total joint registry, 9,771 patients (12,735 knees) who underwent TKA with cemented, posterior-stabilized, and metal-backed tibial components from 2000 to 2016 were identified. Arthrofibrosis, defined as range of motion (ROM) ≤90° for ≥12 weeks postoperatively or as ROM ≤90° requiring MUA, was diagnosed in 454 knees (4%) and matched 1:2 to controls. Mean age was 62 years (range, 19 to 87) and 57% were women. The majority of operative diagnoses were osteoarthritis. Perioperative use of 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitors (statins), angiotensin converting enzyme inhibitors (ACE inhibitors), angiotensin II receptor blockers (ARBs), oral corticosteroids, antihistamines, and nonsteroidal anti-inflammatory drugs (NSAIDs) were manually confirmed. Medication effect in preventing arthrofibrosis and MUA was assessed using adjusted multivariable analyses. Mean follow-up was 8 years (range, 2 to 20)., Results: Reduced risk of arthrofibrosis was associated with perioperative NSAID use (odds ratio (OR) 0.67, P = .045). A similar trend was observed with perioperative corticosteroids (OR 0.52, P = .098). Corticosteroids were associated with reduced risk of MUA (OR 0.26, P = .036), and NSAIDs trended towards reducing MUA (OR 0.69, P = .11)., Conclusion: This investigation determined that perioperative NSAID use was associated with reduced risk of arthrofibrosis and trended towards reduced risk of subsequent MUA. Similarly, oral corticosteroids were associated with reduced risk of MUA and trended towards reduced risk of arthrofibrosis., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Interest in spinal anesthesia utilization in revision total knee arthroplasties (TKAs) is rising. This study investigated the pain control, length of stay (LOS), and complications associated with spinal versus general anesthesia in a single institution series of revision TKAs., Methods: We identified 3,711 revision TKAs (3,495 patients) from 2001 to 2016 using our institutional total joint registry. There were 66% who had general anesthesia and 34% who had spinal anesthesia. Mean age, sex, and BMI were similar between groups at 67 years, 53% women, and 32, respectively. Data were analyzed using inverse probability of treatment weighted models based on propensity scores that accounted for patient and operative factors. Mean follow-up was 6 years (range, 2 to 17)., Results: Patients treated with spinal anesthesia required fewer postoperative oral morphine equivalents (OMEs) (P < .0001) and had lower numeric pain rating scale scores (P < .001). Spinal anesthesia was associated with shorter LOS (4.0 versus 4.6 days; P < .0001), less cases of altered mental status (AMS; Odds Ratio (OR) 2.0, P = .004), less intensive care unit (ICU) admissions (OR 1.6, P = .02), fewer re-revisions (OR 1.7, P < .001), and less reoperations (OR 1.4, P < .001). There was no difference in the incidence of VTE (P = .82), 30-day readmissions (P = .06), or 90-day readmissions (P = .18) between anesthetic techniques., Conclusion: We found that spinal anesthesia for revision TKAs was associated with significantly lower pain scores, reduced OME requirements, and decreased LOS. Furthermore, spinal anesthesia was associated with fewer cases of AMS, ICU admissions, and re-revisions even after accounting for numerous patient and operative factors., Level of Evidence: Level III, Retrospective Comparative Study., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Perioperative medical management during total hip arthroplasty (THA) is continuously improving, allowing an increasing number of medically complex patients to undergo total joint arthroplasty. This study examined mortalities, medical complications, implant survivorships, and clinical outcomes of THA in patients who have pulmonary hypertension (HTN)., Methods: We identified 638 patients who had pulmonary HTN and underwent 508 primary THAs and 191 revision THAs from 2000 to 2016 at a tertiary care center. Patients were followed up at regular intervals until death, revision surgery, or last clinical follow-up. Perioperative medical complications were individually reviewed. The risk of death was examined by calculating standardized mortality ratios and Cox proportional hazards regression models. Cumulative incidence analyses were used for reporting mortality, reoperation, and revision with death as a competing risk., Results: The 90-day mortality was 1.8% and 3.1% for primary and revision THAs, respectively. The risk of death was approximately two-fold higher compared to primary (hazard ratio 2.69) and revision (hazard ratio 2.04) THA patients who did not have pulmonary HTN. Rate of medical complications within 90 days from surgery were 6.2% and 13.1% in primary and revision THAs, respectively. The 10-year cumulative incidence of any revision was 9% and 14% following primaries and revisions, respectively., Conclusion: Patients who had pulmonary HTN undergoing primary and revision THAs had an increased risk of death and experienced a high rate of medical complications within 90 days of surgery. Counseling of risks, medical optimization, and referral to medical centers expert at managing complex medical problems should be considered., Level of Evidence: Level IV., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Background: Failed stemmed total knee arthroplasty (TKA) components present with varying degrees of bone loss and technical challenges. A classification system has been proposed based upon metaphyseal bone loss and diaphyseal cortical integrity. A validation study was performed to determine interobserver and intraobserver reliability at multiple institutions and with different levels of training., Methods: An online survey with digital anteroposterior and lateral radiographs was sent to 5 arthroplasty surgeons and 5 adult reconstruction fellows. The survey included 62 cases with stemmed femoral and tibial components, considered failures and pending revision, and scored by each reviewer independently using the classification system. Each case was scored in 2 separate sessions. Interobserver and intraobserver reliability was assessed using the intraclass correlation coefficient (ICC)., Results: Interobserver grading for both the femur (0.69) and tibia (0.72) showed strong reliability among the attendings and fellows, with slightly stronger reliability in tibia cases. The intraclass correlation coefficient (ICC) for attendings and fellows was similar overall, demonstrating consistency of the grading regardless of training level. Intraobserver comparisons showed a strong ICC for attendings and fellows in femoral cases, while fellows had near-perfect ICC in tibia cases. Across all reviewers there was on average 93% agreement within 1 grade per case with the majority of the discrepancy occurring at the metaphyseal-diaphyseal junction., Conclusion: This classification demonstrated overall strong interobserver and intraobserver reliability, with 93% agreement within 1 grade of bone loss. With further education, this classification system can ultimately be used to standardize the degree of bone loss in failed stemmed components and help with preoperative planning., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: Optimal management of morbid obesity before total hip arthroplasty (THA) remains debated. Recently, bariatric procedures have become more common with advancements in surgical techniques. We hypothesized that bariatric surgery prior to primary THA would mitigate acute postoperative complications and improve implant survivorship., Methods: A retrospective review from 1995 to 2020 identified 88 primary THA procedures in 71 unique patients who previously underwent bariatric surgery (73% Roux-en-Y). This cohort was matched 1:1:1 for age, gender, surgical year, American Society of Anesthesiologists score, and Charlson Comorbidity Index to cohorts of patients with body mass index (BMI) <40 kg/m 2 and BMI ≥40 kg/m 2 . Revisions, reoperations, and acute complications were compared. Subgroup analysis then evaluated historical (pre-2012) relative to contemporary (2012 and after) bariatric procedures., Results: Revision rates for bariatric patients were higher relative to controls with low (hazard ratio [HR] 19, P < .01) and high BMI (HR 8, P < .01). Reoperation rates showed a similar increase for bariatric patients when compared to low (HR 9, P < .01) and high BMI (HR 4, P = .01) patients. Moreover, bariatric patients had an increased dislocation risk compared to the low (HR 7, P = .03) and high BMI (HR 17, P < .01) patients. Contemporary bariatric techniques had similar complications, revisions, and reoperations relative to historical procedures., Conclusion: Morbidly obese patients undergoing THA have increased risks of certain complications, but it is unclear if bariatric surgery improves this risk. This study found that patients undergoing bariatric surgery have worse implant survivorship and higher dislocation rates compared to patients with naturally low and high BMIs. Further investigation into the post-bariatric metabolic state is warranted., Level of Evidence: Prognostic Level IV., (Copyright © 2022 The Authors. Published by Elsevier Inc. All rights reserved.)

Background: No prior studies have examined outcomes based on approach concordance between primary and revision total hip arthroplasty (THA). There is theoretical concern that performing surgery through multiple planes could potentiate dislocation risk. This study aimed to assess the impact of utilizing concordant vs discordant surgical approaches between primary and revision THA on incidence of dislocation, re-revision, reoperation, and nonoperative complications., Methods: Between 2000 and 2018, 705 revision THAs were retrospectively identified in patients who underwent primary THA at the same academic center. Surgical approach was determined for primary and revision THA from operative notes with dislocations, re-revisions, reoperations, and complications determined from our total joint registry. Complication rates were compared between those with concordant and discordant surgical approaches. Mean age was 65 years, 50% were female, mean body mass index was 31 kg/m 2 , and mean follow-up was 4 years., Results: Surgical approach discordance occurred in 97 cases (14%), which was more frequent when the direct anterior approach was used for primary THA (72%, P < .001) compared to lateral (12%) or posterior (10%) approaches. There were no statistically significant differences in the incidence of dislocations, re-revisions, reoperations, and nonoperative complications among those with concordant and discordant approaches for the overall cohort and when analyzed by primary approach (P > .05 for all)., Conclusion: Comparable dislocation and complication rates were observed among revision THAs with concordant and discordant approaches between primary and revision THA. These data provide reassurance that changing vs maintaining the surgical approach from primary to revision THA does not significantly increase dislocation or re-revision risk., Level of Evidence: IV., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Background: Recent advancements in bariatric surgical techniques have increased its utilization for the management of morbid obesity prior to total knee arthroplasty (TKA). We hypothesized that bariatric surgery prior to primary TKA would mitigate postoperative complications and improve implant survivorship., Methods: A retrospective review from 1992-2020 identified 205 bariatric surgery patients with subsequent primary TKA. This cohort was matched 1:1:1 to patients without bariatric surgery and with BMI <40 kg/m 2 and BMI ≥40 kg/m 2 . Revisions, reoperations, and 90-day complications were evaluated. Subgroup analysis evaluated bariatric patients with BMI >40 kg/m 2 at TKA, the time between surgeries, and compared historical to contemporary bariatric techniques., Results: Bariatric patients demonstrated higher revision rates than low (HR 4, P < .01) and high BMI (HR 9, P < .01) controls, and increased reoperations when compared to the low (HR 2, P < .01) and high BMI (HR 6, P < .01) groups. Reoperation for instability was more common in bariatric patients than low (HR 15, P = .01) and high BMI (HR 17, P < .01) groups. Reoperation for infection was higher in bariatric patients relative to the high BMI (HR 6, P = .03), but not the low BMI cohort (HR 3, P = .06). There was no difference in 90-day complications (P = .33). Bariatric patients with high BMI and contemporary bariatric procedures did not significantly impact complications or survivorship, but bariatric surgery >2 years before TKA was associated with higher revision rates (P = .01)., Conclusion: This study found that bariatric surgery patients who undergo primary TKA have worse implant survivorship, mostly related to infection and instability. Further investigation into perioperative optimization is warranted., Level of Evidence: Prognostic Level IV., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Background: Several studies have investigated the distribution of hip-knee-ankle (HKA) angle in healthy populations; however, few have evaluated this metric in patients undergoing total knee arthroplasty (TKA). The purpose of this study is to compare HKA angle distribution in early and advanced knee osteoarthritis (OA) patients., Methods: Full limb radiographs were used to measure HKA angle for 983 subjects from the Osteoarthritis Initiative (OAI) cohort and 4,901 pre-TKA patients from an institutional cohort. Measurements were made using a previously validated deep learning algorithm. Linear regression models were used to determine the association of HKA alignment angle with patient characteristics., Results: The mean ± standard deviation HKA angle was -1.3° ± 3.2° in the OAI cohort and -4.1° ± 6.1° in the pre-TKA cohort. In the OAI cohort, normal alignment (64%) was the most common knee alignment followed by varus (29%), and valgus (7%). In pre-TKA patients, the most common alignment was varus (62%), followed by normal (27%) and valgus (11%). In pre-TKA patients, mean HKA angle in primary knee OA, post-traumatic knee OA, and rheumatoid arthritis patients were -4.3° ± 6.1°, -3.2° ± 6.4°, and -2.9° ± 6.1°, respectively. HKA angle was strongly associated (P < .001) with gender and body mass index., Conclusion: TKA patients have a wider alignment distribution and more severe varus and valgus alignment than individuals "at risk" for knee OA from the OAI cohort. These epidemiologic findings improve our understanding of HKA angle distribution and its correlation with demographic characteristics in early and late-stage arthritis., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Background: Although perioperative medical management during total knee arthroplasty (TKA) has improved, there is limited literature characterizing outcomes of patients with pulmonary hypertension (HTN). This study examined mortality, medical complications, implant survivorship, and clinical outcomes in this medically complex cohort., Methods: We identified 887 patients with pulmonary HTN who underwent 881 primary TKAs and 228 revision TKAs from 2000 to 2016 at a tertiary care center. Patients were followed up at regular intervals until death, revision surgery, or last clinical follow-up. Perioperative medical complications were individually reviewed. The risk of death was examined by calculating standardized mortality ratios and Cox proportional hazards regression models. Cumulative incidence analysis was used for reporting mortality, revision, and reoperation with death as a competing risk., Results: The 90-day mortality was 0.7% and 4.8% for primary and revision TKAs, respectively. The risk of death was 2-fold higher compared to primary (hazard ratio 2.54, 95% confidence interval [CI] 2.12-3.05) and revision (hazard ratio 2.16, 95% CI 1.78-2.62) TKA patients without pulmonary HTN. Rate of medical complications within 90 days from surgery was 6.5% and 14% in primary and revision TKAs. The 10-year cumulative incidence of any revision was 5% and 16% in primaries and revisions, respectively., Conclusion: Patients with pulmonary HTN undergoing primary and revision TKAs had excess risk of death and experience a high rate of medical complications within 90 days of surgery. Counseling of risks, medical optimization, and referral to tertiary centers should be considered., Level of Evidence: Level IV., (Copyright © 2021 Elsevier Inc. All rights reserved.)

Background: Adverse local tissue reactions (ALTR) are associated with total hip arthroplasty (THA) failure in patients with metal-on-metal bearings and/or taper corrosion, which can also occur in metal-on-polyethylene articulations. Patients are monitored with blood cobalt (Co) and chromium (Cr) concentrations which do not always correlate with the degree of soft tissue reaction at revision THA. The purpose of this study was to determine how the blood and prosthetic hip synovial fluid levels of Co and Cr correlate with one another, and determine which concentration is more predictive of ALTR., Methods: Synovial fluid and blood samples were collected at the time of revision THA in patients with (n = 26) and without ALTR (n = 27). Whole blood, serum, and synovial fluid metal ion concentrations were correlated with one another and clinical findings., Results: The ratio of synovial fluid to whole blood Co concentration in ALTR and non-ALTR hips was 120:1 and 18:1 (P = .006). The mean ratio of synovial fluid to whole blood Cr concentration in ALTR and non-ALTR hips was 414:1 and 24:1 (P = .01). The mean synovial fluid Co/Cr concentrations were elevated in patients with ALTR compared with those without (2007.3 vs. 12.4 ppb, P = .007) and (3188.2 vs. 10.3 ppb, P = .02). The synovial fluid Co concentration was the most accurate test for detecting ALTR (cut off: 19.75 PPB, AUC 0.973)., Conclusion: In patients with ALTR, synovial fluid Co and Cr levels were 120 times higher and 414 times higher than whole blood Co and Cr levels. Synovial fluid Co ion concentration was the most accurate in predicting ALTR., Level of Evidence: Diagnostic level II., (Copyright © 2021 Elsevier Inc. All rights reserved.)

Background: Dislocation after total hip arthroplasty (THA) is among the most common causes of early revision in contemporary practice. Abnormal spinopelvic alignment increases risk for dislocation, but methods to identify such are limited and can be complex. We sought to determine the effect of pelvic tilt, using a novel radiographic measurement, on dislocation risk by evaluating those with and without a history of dislocation., Methods: Using our institutional total joint registry, we identified 10,082 primary THAs performed between 2006 and 2015. Postoperatively, 177 dislocated (1.7%). Dislocators were matched 1:1 to control patients who did not dislocate. Pelvic tilt was calculated using the pubic symphysis to sacrococcygeal junction distance (PSCD) from a supine anteroposterior pelvis radiograph both preoperatively and postoperatively. The association between dislocation and both pelvic tilt and PSCD was then evaluated by logistic regression. Mean follow-up was 3 years., Results: Patients who dislocated had more posterior pelvic tilt (mean pelvic tilt of 57° vs 60°; P = .02) and smaller PSCDs (mean 41 mm vs 46 mm; P = .04) than controls. Patients with a PSCD <0 mm (symphysis above sacrococcygeal junction) had 9-fold odds of dislocation compared to those with a PSCD >50 (odds ratio 9; P = .006)., Conclusion: Patients who dislocated following primary THA had more posterior pelvic tilt. Additionally, those with a PSCD <0 had 9-fold odds of dislocation. Assessing the PSCD can alert a surgeon of increased risk for dislocation and identification of a negative PSCD should encourage further investigation or optimization of the preoperative plan to minimize dislocation risk., Level of Evidence: Level IV, case-control study., (Copyright © 2021. Published by Elsevier Inc.)

Background: Consensus on whether low-dose (81 mg) or regular-dose (325 mg) aspirin (ASA) is more effective for venous thromboembolism (VTE) chemoprophylaxis in primary total joint arthroplasties (TJAs) is not reached. The goal of this study is to evaluate the efficacy of low-dose and regular-dose ASA for VTE chemoprophylaxis in primary total hip arthroplasties and total knee arthroplasties., Methods: We retrospectively identified 3512 primary TJAs (2344 total hip arthroplasties and 1168 total knee arthroplasties) with ASA used as VTE chemoprophylaxis between 2000 and 2019. Patients received ASA twice daily for 4-6 weeks after surgery with 961 (27%) receiving low-dose ASA and 2551 (73%) receiving regular-dose ASA. The primary endpoint was 90-day incidence of symptomatic VTEs. Secondary outcomes were gastrointestinal (GI) bleeding events and mortality. The mean age at index TJA was 66 years, 54% were female, and mean body mass index was 31 kg/m 2 . The mean Charlson Comorbidity Index was 3.5. Mean follow-up was 3 years., Results: There was no difference in 90-day incidence of symptomatic VTEs between low-dose and regular-dose ASA (0% vs 0.1%, respectively; P = .79). There were no GI bleeding events in either group. There was no difference in 90-day mortality between low-dose and regular-dose ASA (0.3% vs 0.1%, respectively; P = .24)., Conclusion: In 3512 primary TJA patients treated with ASA, we found a cumulative incidence of VTE <1% at 90 days. Although this study is underpowered, it appears that twice daily low-dose ASA was equally effective to twice daily regular-dose ASA for VTE chemoprophylaxis, with no difference in risk of GI bleeds or mortality., Level of Evidence: III, retrospective cohort study., (Copyright © 2021 Elsevier Inc. All rights reserved.)

Background: Ankylosing spondylitis (AS) is a common inflammatory spondyloarthropathy with hip involvement in 40% of patients. With the renewed interest in the hip-spine interplay, this study aimed to define long-term outcomes of primary total hip arthroplasty (THA) in the setting of AS., Methods: We identified 309 hips (219 patients) with AS treated with primary THA from 1969 to 2018. Mean age was 49 years, 80% were males, and mean body mass index was 28 kg/m 2 . Cumulative incidences of any revision, reoperation, and dislocation were calculated utilizing a competing risk analysis. Harris Hip Scores and complications were also reported. Mean follow-up was 16 years., Results: The cumulative incidence of any revision after primary THA was 2.3% at 5 years and 17.5% at 20 years. The most common reasons for revision (n = 73) were aseptic loosening (41%), osteolysis/polyethylene (PE) wear (30%, all with conventional PE), and femoral component fracture (8%). The cumulative incidence of dislocation was 1.9% at 5 years and 2.9% at 20 years. Younger age was associated with increased risk of revision (hazard ratio (HR) = 1.3, P < .01) and reoperation (HR = 1.2, P < .01), but not dislocation (HR = 0.7, P = .1). Twenty-eight hips (9%) experienced a postoperative complication not requiring reoperation. The mean Harris Hip Score improved from 51 to 76 after THA (P < .001)., Conclusion: In this series of 309 primary THAs in patients with AS, the 20-year cumulative incidence of any revision after primary THA was 17.5%. Aseptic loosening, osteolysis/PE wear, and femoral component fracture were the most common reasons for revision. Notably, the cumulative incidence of dislocation at 20 years was only 2.9%., Level of Evidence: Level IV., (Copyright © 2021 Elsevier Inc. All rights reserved.)

Background: Inappropriate acetabular component angular position is believed to increase the risk of hip dislocation after total hip arthroplasty. However, manual measurement of these angles is time consuming and prone to interobserver variability. The purpose of this study was to develop a deep learning tool to automate the measurement of acetabular component angles on postoperative radiographs., Methods: Two cohorts of 600 anteroposterior (AP) pelvis and 600 cross-table lateral hip postoperative radiographs were used to develop deep learning models to segment the acetabular component and the ischial tuberosities. Cohorts were manually annotated, augmented, and randomly split to train-validation-test data sets on an 8:1:1 basis. Two U-Net convolutional neural network models (one for AP and one for cross-table lateral radiographs) were trained for 50 epochs. Image processing was then deployed to measure the acetabular component angles on the predicted masks for anatomical landmarks. Performance of the tool was tested on 80 AP and 80 cross-table lateral radiographs., Results: The convolutional neural network models achieved a mean Dice similarity coefficient of 0.878 and 0.903 on AP and cross-table lateral test data sets, respectively. The mean difference between human-level and machine-level measurements was 1.35° (σ = 1.07°) and 1.39° (σ = 1.27°) for the inclination and anteversion angles, respectively. Differences of 5⁰ or more between human-level and machine-level measurements were observed in less than 2.5% of cases., Conclusion: We developed a highly accurate deep learning tool to automate the measurement of angular position of acetabular components for use in both clinical and research settings., Level of Evidence: III., (Copyright © 2021 Elsevier Inc. All rights reserved.)

Background: Dislocation is a common complication following total hip arthroplasty (THA), and accounts for a high percentage of subsequent revisions. The purpose of this study is to illustrate the potential of a convolutional neural network model to assess the risk of hip dislocation based on postoperative anteroposterior pelvis radiographs., Methods: We retrospectively evaluated radiographs for a cohort of 13,970 primary THAs with 374 dislocations over 5 years of follow-up. Overall, 1490 radiographs from dislocated and 91,094 from non-dislocated THAs were included in the analysis. A convolutional neural network object detection model (YOLO-V3) was trained to crop the images by centering on the femoral head. A ResNet18 classifier was trained to predict subsequent hip dislocation from the cropped imaging. The ResNet18 classifier was initialized with ImageNet weights and trained using FastAI (V1.0) running on PyTorch. The training was run for 15 epochs using 10-fold cross validation, data oversampling, and augmentation., Results: The hip dislocation classifier achieved the following mean performance (standard deviation): accuracy = 49.5 (4.1%), sensitivity = 89.0 (2.2%), specificity = 48.8 (4.2%), positive predictive value = 3.3 (0.3%), negative predictive value = 99.5 (0.1%), and area under the receiver operating characteristic curve = 76.7 (3.6%). Saliency maps demonstrated that the model placed the greatest emphasis on the femoral head and acetabular component., Conclusion: Existing prediction methods fail to identify patients at high risk of dislocation following THA. Our radiographic classifier model has high sensitivity and negative predictive value, and can be combined with clinical risk factor information for rapid assessment of risk for dislocation following THA. The model further suggests radiographic locations which may be important in understanding the etiology of prosthesis dislocation. Importantly, our model is an illustration of the potential of automated imaging artificial intelligence models in orthopedics., Level of Evidence: Level III., (Copyright © 2021 Elsevier Inc. All rights reserved.)

Background: Our institution previously initiated a perioperative surgical home initiative to improve quality and efficiency across the hospital arc of care of primary total knee arthroplasty and total hip arthroplasty patients. Phase II of this project aimed to (1) expand the perioperative surgical home to include revision total hip arthroplasties and total knee arthroplasties, hip preservation procedures, and reconstructions after oncologic resections; (2) expand the project to include the preoperative phase; and (3) further refine the perioperative surgical home goals accomplished in phase I., Methods: Phase II of the Orthopedic Surgery and Anesthesiology Surgical Improvement Strategies project ran from July 2018 to July 2019. The evaluated arc of care spanned from the preoperative surgical consult visit through 90 days postoperative in the expanded population described above., Results: Mean length of stay decreased from 2.2 days to 2.0 days (P < .001), 90-day readmission decreased from 3.0% to 1.6% (P < .001), and Press-Ganey scores increased from 77.1 to 79.2 (97th percentile). Mean and maximum pain scores and opioid consumption remained unchanged (lowest P = .31). Annual surgical volume increased by 10%. Composite changes in surgical volume and cost reductions equaled $5 million., Conclusion: Application of previously successful health systems engineering tools and methods in phase I of Orthopedic Surgery and Anesthesiology Surgical Improvement Strategies enabled additional evolution of an orthopedic perioperative surgical home to encompass more diverse and complex patient populations while increasing system-wide quality, safety, and financial outcomes. Improved process and outcomes metrics reflected increased efficiency across the episode of care without untoward effects., Level of Evidence: III Therapeutic., (Copyright © 2021 Elsevier Inc. All rights reserved.)