Application of Machine Learning for Predicting Clinically Meaningful Outcome After Arthroscopic Femoroacetabular Impingement Surgery

ChatGPT performance on the American Shoulder and Elbow Surgeons maintenance of certification exam

BACKGROUND

While multiple studies have tested the ability of large language models (LLMs), such as ChatGPT, to pass standardized medical exams at different levels of training, LLMs have never been tested on surgical sub-specialty examinations, such as the American Shoulder and Elbow Surgeons (ASES) Maintenance of Certification (MOC). The purpose of this study was to compare results of ChatGPT 3.5, GPT-4, and fellowship-trained surgeons on the 2023 ASES MOC self-assessment exam.

METHODS

ChatGPT 3.5 and GPT-4 were subjected to the same set of text-only questions from the ASES MOC exam, and GPT-4 was additionally subjected to image-based MOC exam questions. Question responses from both models were compared against the correct answers. Performance of both models was compared to corresponding average human performance on the same question subsets. One sided proportional z-test were utilized to analyze data.

RESULTS

Humans performed significantly better than Chat GPT 3.5 on exclusively text-based questions (76.4% vs. 60.8%, P = .044). Humans also performed significantly better than GPT 4 on image-based questions (73.9% vs. 53.2%, P = .019). There was no significant difference between humans and GPT 4 in text-based questions (76.4% vs. 66.7%, P = .136). Accounting for all questions, humans significantly outperformed GPT-4 (75.3% vs. 60.2%, P = .012). GPT-4 did not perform statistically significantly betterer than ChatGPT 3.5 on text-only questions (66.7% vs. 60.8%, P = .268).

DISCUSSION

Although human performance was overall superior, ChatGPT demonstrated the capacity to analyze orthopedic information and answer specialty-specific questions on the ASES MOC exam for both text and image-based questions. With continued advancements in deep learning, LLMs may someday rival exam performance of fellowship-trained surgeons.

Minimal Clinically Important Difference, Substantial Clinical Benefit, and Patient Acceptable Symptom State Values After Hip Arthroscopy for Femoroacetabular Impingement Are Highly Dependent on Their Study Population and Calculation Methods: A Systematic Review

PURPOSE

To provide a summary of available literature on the minimal clinically important difference (MCID), substantial clinical benefit (SCB), and patient acceptable symptom state (PASS) after hip arthroscopy for femoroacetabular impingement (FAI).

METHODS

A systematic review was conducted via the Cochrane Library, PubMed, Ovid MEDLINE, and Embase to identify studies that calculated MCID, SCB, or PASS for patient-reported outcome measures after hip arthroscopy for FAI. The electronic search strategy used was as follows: hip AND arthroscopy AND (MCID OR "minimal clinically important difference" OR SCB OR "substantial clinical benefit" OR PASS OR "patient acceptable symptom state"). Inclusion criteria were English-language studies published from 1980 to 2023 reporting clinical outcome scores and calculated values of MCID, PASS, or SCB for patients undergoing hip arthroscopy for FAI.

RESULTS

Forty-two studies (5 Level II, 19 Level III, and 18 Level IV) met inclusion and exclusion criteria. The most commonly used outcome measures across MCID, SCB, and PASS were the Hip Outcome Score sports-specific subscale and the activities of daily living subscale, the modified Harris Hip Score, and the 12-item international Hip Outcome Tool. The range of MCID values for Hip Outcome Score sports-specific subscale, Hip Outcome Score activities of daily living subscale, modified Harris Hip Score, and 12-item international Hip Outcome Tool were 7.2-15.7, 7.3-15.4, 7.2-16.8, and 8.8-16.2 respectively. Similarly, for SCB the values ranged from 77.9-96.9, 90.4-98.5, 20.0-98.4, and 66.7-87.5, respectively. Lastly, the PASS values ranged from 63.9-80.9, 85.9-99.2, 74.0-97.0, and 59.5-86.0, respectively.

CONCLUSIONS

MCID, SCB, and PASS values for patient-reported outcome measures after hip arthroscopy for the management of FAI are highly dependent on their associated study including study population and calculation methods.

LEVEL OF EVIDENCE

IV, systematic review of Level II-IV studies.

Artificial Intelligence Models Are Limited in Predicting Clinical Outcomes Following Hip Arthroscopy: A Systematic Review

BACKGROUND

Hip arthroscopy has seen a significant surge in utilization, but complications remain, and optimal functional outcomes are not guaranteed. Artificial intelligence (AI) has emerged as an effective supportive decision-making tool for surgeons. The purpose of this systematic review was to characterize the outcomes, performance, and validity (generalizability) of AI-based prediction models for hip arthroscopy in current literature.

METHODS

Two reviewers independently completed structured searches using PubMed/MEDLINE and Embase databases on August 10, 2022. The search query used the terms as follows: (artificial intelligence OR machine learning OR deep learning) AND (hip arthroscopy). Studies that investigated AI-based risk prediction models in hip arthroscopy were included. The primary outcomes of interest were the variable(s) predicted by the models, best model performance achieved (primarily based on area under the curve, but also accuracy, etc), and whether the model(s) had been externally validated (generalizable).

RESULTS

Seventy-seven studies were identified from the primary search. Thirteen studies were included in the final analysis. Six studies (n = 6,568) applied AI for predicting the achievement of minimal clinically important difference for various patient-reported outcome measures such as the visual analog scale and the International Hip Outcome Tool 12-Item Questionnaire, with area under a receiver-operating characteristic curve (AUC) values ranging from 0.572 to 0.94. Three studies used AI for predicting repeat hip surgery with AUC values between 0.67 and 0.848. Four studies focused on predicting other risks, such as prolonged postoperative opioid use, with AUC values ranging from 0.71 to 0.76. None of the 13 studies assessed the generalizability of their models through external validation.

CONCLUSION

AI is being investigated for predicting clinical outcomes after hip arthroscopy. However, the performance of AI models varies widely, with AUC values ranging from 0.572 to 0.94. Critically, none of the models have undergone external validation, limiting their clinical applicability. Further research is needed to improve model performance and ensure generalizability before these tools can be reliably integrated into patient care.

LEVEL OF EVIDENCE

Level IV. See Instructions for Authors for a complete description of levels of evidence.

Machine learning/artificial intelligence in sports medicine: state of the art and future directions

Machine learning (ML) is changing the way health care is practiced and recent applications of these novel statistical techniques have started to impact orthopaedic sports medicine. Machine learning enables the analysis of large volumes of data to establish complex relationships between "input" and "output" variables. These relationships may be more complex than could be established through traditional statistical analysis and can lead to the ability to predict the "output" with high levels of accuracy. Supervised learning is the most common ML approach for healthcare data and recent studies have developed algorithms to predict patient-specific outcome after surgical procedures such as hip arthroscopy and anterior cruciate ligament reconstruction. Deep learning is a higher-level ML approach that facilitates the processing and interpretation of complex datasets through artificial neural networks that are inspired by the way the human brain processes information. In orthopaedic sports medicine, deep learning has primarily been used for automatic image (computer vision) and text (natural language processing) interpretation. While applications in orthopaedic sports medicine have been increasing exponentially, one significant barrier to widespread adoption of ML remains clinician unfamiliarity with the associated methods and concepts. The goal of this review is to introduce these concepts, review current machine learning models in orthopaedic sport medicine, and discuss future opportunities for innovation within the specialty.

ChatGPT Provides Satisfactory but Occasionally Inaccurate Answers to Common Patient Hip Arthroscopy Questions

PURPOSE

To assess the ability of ChatGPT to answer common patient questions regarding hip arthroscopy, and to analyze the accuracy and appropriateness of its responses.

METHODS

Ten questions were selected from well-known patient education websites, and ChatGPT (version 3.5) responses to these questions were graded by 2 fellowship-trained hip preservation surgeons. Responses were analyzed, compared with the current literature, and graded from A to D (A being the highest, and D being the lowest) in a grading scale on the basis of the accuracy and completeness of the response. If the grading differed between the 2 surgeons, a consensus was reached. Inter-rater agreement was calculated. The readability of responses was also assessed using the Flesch-Kincaid Reading Ease Score (FRES) and Flesch-Kincaid Grade Level (FKGL).

RESULTS

Responses received the following consensus grades: A (50%, n = 5), B (30%, n = 3), C (10%, n = 1), D (10%, n = 1). Inter-rater agreement on the basis of initial individual grading was 30%. The mean FRES was 28.2 (± 9.2 standard deviation), corresponding to a college graduate level, ranging from 11.7 to 42.5. The mean FKGL was 14.4 (±1.8 standard deviation), ranging from 12.1 to 18, indicating a college student reading level.

CONCLUSIONS

ChatGPT can answer common patient questions regarding hip arthroscopy with satisfactory accuracy graded by 2 high-volume hip arthroscopists; however, incorrect information was identified in more than one instance. Caution must be observed when using ChatGPT for patient education related to hip arthroscopy.

CLINICAL RELEVANCE

Given the increasing number of hip arthroscopies being performed annually, ChatGPT has the potential to aid physicians in educating their patients about this procedure and addressing any questions they may have.

Artificial Intelligence for Clinically Meaningful Outcome Prediction in Orthopedic Research: Current Applications and Limitations

PURPOSE OF REVIEW

Patient-reported outcome measures (PROM) play a critical role in evaluating the success of treatment interventions for musculoskeletal conditions. However, predicting which patients will benefit from treatment interventions is complex and influenced by a multitude of factors. Artificial intelligence (AI) may better anticipate the propensity to achieve clinically meaningful outcomes through leveraging complex predictive analytics that allow for personalized medicine. This article provides a contemporary review of current applications of AI developed to predict clinically significant outcome (CSO) achievement after musculoskeletal treatment interventions.

RECENT FINDINGS

The highest volume of literature exists in the subspecialties of total joint arthroplasty, spine, and sports medicine, with only three studies identified in the remaining orthopedic subspecialties combined. Performance is widely variable across models, with most studies only reporting discrimination as a performance metric. Given the complexity inherent in predictive modeling for this task, including data availability, data handling, model architecture, and outcome selection, studies vary widely in their methodology and results. Importantly, the majority of studies have not been externally validated or demonstrate important methodological limitations, precluding their implementation into clinical settings. A substantial body of literature has accumulated demonstrating variable internal validity, limited scope, and low potential for clinical deployment. The majority of studies attempt to predict the MCID-the lowest bar of clinical achievement. Though a small proportion of models demonstrate promise and highlight the utility of AI, important methodological limitations need to be addressed moving forward to leverage AI-based applications for clinical deployment.

Development of Machine-Learning Algorithms to Predict Attainment of Minimal Clinically Important Difference After Hip Arthroscopy for Femoroacetabular Impingement Yield Fair Performance and Limited Clinical Utility

PURPOSE

To determine whether machine learning (ML) techniques developed using registry data could predict which patients will achieve minimum clinically important difference (MCID) on the International Hip Outcome Tool 12 (iHOT-12) patient-reported outcome measures (PROMs) after arthroscopic management of femoroacetabular impingement syndrome (FAIS). And secondly to determine which preoperative factors contribute to the predictive power of these models.

METHODS

A retrospective cohort of patients was selected from the UK's Non-Arthroplasty Hip Registry. Inclusion criteria were a diagnosis of FAIS, management via an arthroscopic procedure, and a minimum follow-up of 6 months after index surgery from August 2012 to June 2021. Exclusion criteria were for non-arthroscopic procedures and patients without FAIS. ML models were developed to predict MCID attainment. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC).

RESULTS

In total, 1,917 patients were included. The random forest, logistic regression, neural network, support vector machine, and gradient boosting models had AUROC 0.75 (0.68-0.81), 0.69 (0.63-0.76), 0.69 (0.63-0.76), 0.70 (0.64-0.77), and 0.70 (0.64-0.77), respectively. Demographic factors and disease features did not confer a high predictive performance. Baseline PROM scores alone provided comparable predictive performance to the whole dataset models. Both EuroQoL 5-Dimension 5-Level and iHOT-12 baseline scores and iHOT-12 baseline scores alone provided AUROC of 0.74 (0.68-0.80) and 0.72 (0.65-0.78), respectively, with random forest models.

CONCLUSIONS

ML models were able to predict with fair accuracy attainment of MCID on the iHOT-12 at 6-month postoperative assessment. The most successful models used all patient variables, all baseline PROMs, and baseline iHOT-12 responses. These models are not sufficiently accurate to warrant routine use in the clinic currently.

LEVEL OF EVIDENCE

Level III, retrospective cohort design; prognostic study.

Recommended Requirements and Essential Elements for Proper Reporting of the Use of Artificial Intelligence Machine Learning Tools in Biomedical Research and Scientific Publications

Essential elements required for proper use of artificial intelligence machine learning tools in biomedical research and scientific publications include (1) explanation justifying why a machine learning approach contributes to the purpose of the study; (2) description of the adequacy of the data (input) to produce the desired results (output); (3) details of the algorithmic (i.e., computational) approach including methods for organizing the data (preprocessing); the machine learning computational algorithm(s) assessed; on what data the models were trained; the presence of bias and efforts to mitigate these effects; and the methods for quantifying the variables (features) most influential in determining the results (e.g., Shapley values); (4) description of methods, and reporting of results, quantitating performance in terms of both model accuracy and model calibration (level of confidence in the model's predictions); (5) availability of the programming code (including a link to the code when available-ideally, the code should be available); (6) discussion of model internal validation (results applicable and sensitive to the population investigated and data on which the model was trained) and external validation (were the results investigated as to whether they are generalizable to different populations? If not, consideration of this limitation and discussion of plans for external validation, i.e., next steps). As biomedical research submissions using artificial intelligence technology increase, these requirements could facilitate purposeful use and comprehensive methodological reporting.

Depression Scores Decrease After Hip Arthroscopy for Femoroacetabular Impingement Syndrome

Purpose

To evaluate clinical depression scores and functional outcomes following arthroscopic treatment of femoroacetabular impingement syndrome in patients with elevated preoperative depressive symptoms as defined by Patient-Reported Outcomes Measurement Information System for Depression (PROMIS-D).

Methods

Patients with femoroacetabular impingement syndrome completed the PROMIS-D Computer Adaptive Test and additional patient-reported outcome (PRO) measures preoperatively and at the time of postoperative visits. Patients were categorized into preoperative clinically depressed (CD) and nonclinically depressed (NCD) groups based on preoperative PROMIS-D scores. Scores ≥55 correlate to mild clinical depression, and this cutoff was used to determine preoperative depression status. PROMIS-D scores and functional outcome scores were assessed at 6 months and a minimum of 1-year postoperatively.

Results

In total, 100 patients were included with complete PROs at a minimum of 1-year follow-up. Of those included, 21 (21%) were categorized with preoperative CD. There were no differences in demographic or radiographic variables between the preoperative CD and NCD groups. At 6 months and 12 months postoperatively, the percentage of patients in the preoperative CD group with continued depression was 33.3% and 23.8%, respectively. Overall, 1-year change in PROMIS-D score for the CD group was -9.1 versus -0.8 in the NCD group (P = .001). There was no significant difference in rates of patients achieving patient acceptable symptom state between the preoperative CD and NCD groups.

Conclusions

Patients with symptoms of preoperative CD, as defined by the PROMIS-D score, demonstrated significant improvement in depressive symptoms following hip arthroscopy. In addition, patients with CD preoperatively did not show decreased rates of achieving minimum clinically important difference or patient acceptable symptom state on postoperative PROs compared with patients with NCD.

Level of Evidence

Level IV, therapeutic case series.

Editorial Commentary: Female Patients With Lower Body Mass Index Show the Best Outcomes After Hip Arthroscopy, and Arthroscopic Treatment of Femoroacetabular Impingement in Higher-Body Mass Index Female Patients Results in Improved Outcomes

Which patients will benefit most from hip arthroscopy? Careful patient selection and conservative indications, such as patients with an alpha angle of 60° or greater or a lateral center-edge angle of 40° or greater who fail a trial of conservative treatment, may benefit from hip arthroscopy for femoroacetabular impingement (FAI). In female patients in particular, a lower body mass index (BMI) will predict the most benefit from arthroscopic treatment. That said, patients with a higher BMI can also substantially improve after treatment of FAI. The true art of medicine is determining indications for an individual patient in addition to providing evidence-based counseling and education. We must not forget that sometimes "any improvement" can be a good outcome for a patient who is in pain.

Predicting Outcomes in Hip Arthroscopy for Femoroacetabular Impingement Syndrome

PURPOSE OF REVIEW

Arthroscopic treatment of femoroacetabular impingement syndrome (FAIS) continues to rise in incidence, and thus there is an increased focus on factors that predict patient outcomes. The factors that impact the outcomes of arthroscopic FAIS treatment are complex. The purpose of this review is to outline the current literature concerning predictors of patient outcomes for arthroscopic treatment of FAIS.

RECENT FINDINGS

Multiple studies have shown that various patient demographics, joint parameters, and surgical techniques are all correlated with postoperative outcomes after arthroscopic FAIS surgery, as measured by both validated patient-reported outcome (PRO) scores and rates of revision surgery including hip arthroplasty. To accurately predict patient outcomes for arthroscopic FAIS surgery, consideration should be directed toward preoperative patient-specific factors and intraoperative technical factors. The future of accurately selecting patient predictors for outcomes will only improve with increased data, improved techniques, and technological advancement.

Injections prior to hip arthroscopy are associated with increased risk of repeat hip arthroscopy at 1 and 5 years

INTRODUCTION

Previous studies have shown that intra-articular hip injections prior to hip arthroscopy (HA) can be a helpful diagnostic tool. However, local anesthetic and corticosteroid injections can be chondrotoxic and corticosteroid injections have been shown to increase the risk of infection during subsequent surgical intervention. The purpose of this study was to evaluate whether preoperative injections adversely affect outcomes of HA using a national database.

MATERIALS AND METHODS

The TriNetX database was retrospectively queried. Patients undergoing HA for femoroacetabular impingement with at least 1 year of claims runout were included in the analysis. Patients were grouped by whether they had a hip injection within 1 year prior to HA. The rates of repeat HA, total hip arthroplasty (THA), infection, osteonecrosis, and new onset hip OA at 1- and 5-years postoperatively were compared between groups. Statistical significance was assessed at α = 0.05.

RESULTS

6511 HA patients with previous injection and 1178 HA patients without previous injection were included. Patients with a previous injection were overall younger (32.3 vs. 34.7 years, p < 0.001), more likely to be female (69 vs. 48%, p < 0.001) and had a higher BMI (26.3 vs. 25.7 kg/m2, p = 0.043). At 1 and 5-years postoperatively, patients with any injection were 1.43 (p < 0.001) and 1.89 (p < 0.001) times more likely to undergo repeat HA, respectively. At 1 and 5-years postoperatively, patients who underwent a corticosteroid injection were 2.29 (p < 0.001) and 1.89 (p < 0.001) times more likely to undergo repeat HA than patients with local anesthetic injection only and 1.56 (p < 0.001) and 2.08 (p < 0.001) times more likely to undergo repeat HA than patients with no injection.

CONCLUSIONS

Intraarticular hip injections prior to hip arthroscopy, particularly corticosteroid injections, are associated with increased risk of repeat hip arthroscopy at 1 and 5 years. Additional studies are needed to elucidate this risk.

Intermediate-Term Outcomes of Hip Arthroscopy for Femoroacetabular Impingement Syndrome in Patients With Global Versus Isolated Lateral Acetabular Overcoverage

BACKGROUND

Previous studies evaluating the outcomes of hip arthroscopy for patients with global acetabular overcoverage and focal superolateral acetabular overcoverage suffer from short-term follow-up and inconsistent radiographic criteria when defining these subpopulations of patients with femoroacetabular impingement syndrome (FAIS).

PURPOSE

To evaluate the intermediate-term postoperative outcomes for patients with FAIS in the setting of global acetabular overcoverage, lateral acetabular overcoverage, and normal acetabular coverage.

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

Patients undergoing hip arthroscopy for FAIS were enrolled in a prospective cohort study, and those with a minimum follow-up of 5 years were included in this analysis. Patients were grouped based on type of acetabular coverage: global overcoverage (lateral center-edge angle [LCEA] ≥40°, with coxa profunda), lateral overcoverage (LCEA ≥40°, without coxa profunda), and no overcoverage (LCEA <40°). Functional outcomes (modified Harris Hip Score and Nonarthritic Hip Score) and failure of primary hip arthroscopy were compared between groups.

RESULTS

In total, 94 patients (mean age, 41.9 ± 14.2 years) were included with a mean follow-up duration of 6.1 ± 0.9 years. Of these patients, 40.4% had no acetabular overcoverage, 36.2% had lateral overcoverage, and 23.4% had global overcoverage. There was no difference between groups with respect to percentage of patients who underwent reoperation for either revision arthroscopy or conversion to total hip arthroplasty (28.9% for the normal acetabular coverage group, 29.4% for the lateral overcoverage group, and 31.8% for the global overcoverage group; P = .971). Among patients for whom primary hip arthroscopy did not fail, there was no difference in 5-year functional outcomes between groups. Postoperative LCEA >40° (β = -13.3; 95% CI, -24.1 to -2.6; P = .016), female sex (β = -14.5; 95% CI, -22.7 to -6.2; P = .001), and higher body mass index (β = -1.9; 95% CI, -2.8 to -1.0; P < .001) were associated with worse intermediate-term hip function in terms of modified Harris Hip Score.

CONCLUSION

There was no difference in functional outcomes or rate of reoperation at a minimum of 5 years postoperatively between those with global acetabular overcoverage, those with regional lateral overcoverage, and those with normal acetabular coverage. Provided that an appropriate acetabuloplasty is performed, there is no evidence to suggest that global acetabular overcoverage portends a worse prognosis than other FAIS subtypes.

Stratified care in hip arthroscopy: can we predict successful and unsuccessful outcomes? Development and external temporal validation of multivariable prediction models

OBJECTIVE

Although hip arthroscopy is a widely adopted treatment option for hip-related pain, it is unknown whether preoperative clinical information can be used to assist surgical decision-making to avoid offering surgery to patients with limited potential for a successful outcome. We aimed to develop and validate clinical prediction models to identify patients more likely to have an unsuccessful or successful outcome 1 year post hip arthroscopy based on the patient acceptable symptom state.

METHODS

Patient records were extracted from the Danish Hip Arthroscopy Registry (DHAR). A priori, 26 common clinical variables from DHAR were selected as prognostic factors, including demographics, radiographic parameters of hip morphology and self-reported measures. We used 1082 hip arthroscopy patients (surgery performed 25 April 2012 to 4 October 2017) to develop the clinical prediction models based on logistic regression analyses. The development models were internally validated using bootstrapping and shrinkage before temporal external validation was performed using 464 hip arthroscopy patients (surgery performed 5 October 2017 to 13 May 2019).

RESULTS

The prediction model for unsuccessful outcomes showed best and acceptable predictive performance on the external validation dataset for all multiple imputations (Nagelkerke R2 range: 0.25-0.26) and calibration (intercept range: -0.10 to -0.11; slope range: 1.06-1.09), and acceptable discrimination (area under the curve range: 0.76-0.77). The prediction model for successful outcomes did not calibrate well, while also showing poor discrimination.

CONCLUSION

Common clinical variables including demographics, radiographic parameters of hip morphology and self-reported measures were able to predict the probability of having an unsuccessful outcome 1 year after hip arthroscopy, while the model for successful outcome showed unacceptable accuracy. The externally validated prediction model can be used to support clinical evaluation and shared decision making by informing the orthopaedic surgeon and patient about the risk of an unsuccessful outcome, and thus when surgery may not be appropriate.

Artificial intelligence and ChatGPT in Orthopaedics and sports medicine

Artificial intelligence (AI) is looked upon nowadays as the potential major catalyst for the fourth industrial revolution. In the last decade, AI use in Orthopaedics increased approximately tenfold. Artificial intelligence helps with tracking activities, evaluating diagnostic images, predicting injury risk, and several other uses. Chat Generated Pre-trained Transformer (ChatGPT), which is an AI-chatbot, represents an extremely controversial topic in the academic community. The aim of this review article is to simplify the concept of AI and study the extent of AI use in Orthopaedics and sports medicine literature. Additionally, the article will also evaluate the role of ChatGPT in scientific research and publications.Level of evidence: Level V, letter to review.

Editorial Commentary: Prolonged Delays in Staging Bilateral Hip Arthroscopic Surgery for Femoroacetabular Impingement May Result in Inferior Patient Outcomes: Two Halves of One Problem

Patients with symptomatic femoroacetabular impingement frequently have contralateral symptoms and, thus, alterations to the biomechanics of both hips. It has become increasingly clear that prolonged delays in staging bilateral hip arthroscopic surgery may result in inferior patient outcomes. There is an interchange between primary surgical recovery and altered biomechanics stemming from the untreated hip. At a certain point, the persistence of microinstability and/or femoroacetabular impingement in one hip becomes a limitation to the recovery of the other. Still, individual patient variability remains a critical factor when considering bilateral surgeries. Some patients cannot tolerate 2 surgeries in proximity. The time frame for bilateral surgery should be based on individual patient factors and functional goals.

Editorial Commentary: Legacy Patient-Reported Outcome Measures Are Superior to Patient-Reported Outcomes Measurement Information System for Assessing Function After Hip Arthroscopy

Patient-reported outcome measures (PROMs) are critical tools in hip preservation research for evaluating the efficacy of current treatments, as well as identifying risk factors associated with suboptimal outcomes. These measures have been used for quality improvement, for monitoring of health plan performance, and even for reimbursement models. Over the past 2 decades, legacy hip outcome scores have been developed that are patient-centric and evaluate hip-specific function after surgery. There has been a recent trend in using the Patient-Reported Outcomes Measurement Information System (PROMIS), a tool developed by the National Institutes of Health for evaluating PROMs, in the field of hip arthroscopy. However, on the basis of the evidence in the literature, it is unlikely that PROMIS is superior to legacy PROMs regarding evaluation of hip function, nor is it as responsive to quantifying meaningful changes in function that are important to patients. As such, clinicians and researchers alike should likely continue using legacy PROMs to evaluate patients' functional outcomes after hip arthroscopy while continuing to explore the clinical applications of other PROMIS domains.

Unsupervised machine learning methods and emerging applications in healthcare

Unsupervised machine learning methods are important analytical tools that can facilitate the analysis and interpretation of high-dimensional data. Unsupervised machine learning methods identify latent patterns and hidden structures in high-dimensional data and can help simplify complex datasets. This article provides an overview of key unsupervised machine learning techniques including K-means clustering, hierarchical clustering, principal component analysis, and factor analysis. With a deeper understanding of these analytical tools, unsupervised machine learning methods can be incorporated into health sciences research to identify novel risk factors, improve prevention strategies, and facilitate delivery of personalized therapies and targeted patient care.Level of evidence: I.

Implementation of Machine Learning to Predict Cost of Care Associated with Ambulatory Single-Level Lumbar Decompression

BACKGROUND

With the emergence of the concept of value-based care, efficient resource allocation has become an increasingly prominent factor in surgical decision-making. Validated machine learning (ML) models for cost prediction in outpatient spine surgery are limited. As such, we developed and internally validated a supervised ML algorithm to reliably identify cost drivers associated with ambulatory single-level lumbar decompression surgery.

METHODS

A retrospective review of the New York State Ambulatory Surgical Database was performed to identify patients who underwent single-level lumbar decompression from 2014 to 2015. Patients with a length of stay of >0 were excluded. Using pre- and intraoperative parameters (features) derived from the New York State Ambulatory Surgical Database, an optimal supervised ML model was ultimately developed and internally validated after 5 candidate models were rigorously tested, trained, and compared for predictive performance related to total charges. The best performing model was then evaluated by testing its performance on identifying relationships between features of interest and cost prediction. Finally, the best performing algorithm was entered into an open-access web application.

RESULTS

A total of 8402 patients were included. The gradient-boosted ensemble model demonstrated the best performance assessed via internal validation. Major cost drivers included anesthesia type, operating room time, race, patient income and insurance status, community type, worker's compensation status, and comorbidity index.

CONCLUSIONS

The gradient-boosted ensemble model predicted total charges and associated cost drivers associated with ambulatory single-level lumbar decompression using a large, statewide database with excellent performance. External validation of this algorithm in future studies may guide practical application of this clinical tool.

Supervised machine learning and associated algorithms: applications in orthopedic surgery

Supervised learning is the most common form of machine learning utilized in medical research. It is used to predict outcomes of interest or classify positive and/or negative cases with a known ground truth. Supervised learning describes a spectrum of techniques, ranging from traditional regression modeling to more complex tree boosting, which are becoming increasingly prevalent as the focus on "big data" develops. While these tools are becoming increasingly popular and powerful, there is a paucity of literature available that describe the strengths and limitations of these different modeling techniques. Typically, there is no formal training for health care professionals in the use of machine learning models. As machine learning applications throughout medicine increase, it is important that physicians and other health care professionals better understand the processes underlying application of these techniques. The purpose of this study is to provide an overview of commonly used supervised learning techniques with recent case examples within the orthopedic literature. An additional goal is to address disparities in the understanding of these methods to improve communication within and between research teams.

Criteria for Return to Play After Hip Arthroscopy in the Treatment of Femoroacetabular Impingement: A Systematic Review

BACKGROUND

Femoroacetabular impingement (FAI) is a common pathology in athletes that often requires operative management in the form of hip arthroscopy.

PURPOSE

To systematically review the rates and level of return to play (RTP) and the criteria used for RTP after hip arthroscopy for FAI in athletes.

STUDY DESIGN

Systematic review; Level of evidence, 4.

METHODS

A systematic review of the literature, based on the PRISMA guidelines, was performed using PubMed, Embase, and Scopus databases. Studies reporting outcomes after the use of hip arthroscopy for FAI were included. Outcomes analyzed were RTP rate, RTP level, and criteria used for RTP. Statistical analysis was performed using SPSS software.

RESULTS

Our review found 130 studies, which included 14,069 patients (14,517 hips) and had a mean methodological quality of evidence (MQOE) of 40.4 (range, 5-67). The majority of patients were female (53.7%), the mean patient age was 30.4 years (range, 15-47 years), and the mean follow-up was 29.7 months (range, 6-75 months). A total of 81 studies reported RTP rates, with an overall RTP rate of 85.4% over a mean period of 6.6 months. Additionally, 49 studies reported the rate of RTP at preinjury level as 72.6%. Specific RTP criteria were reported in 97 studies (77.2%), with time being the most commonly reported item, which was reported in 80 studies (69.2%). A total of 45 studies (57.9%) advised RTP at 3 to 6 months after hip arthroscopy.

CONCLUSION

The overall rate of reported RTP was high after hip arthroscopy for FAI. However, more than one-fourth of athletes who returned to sports did not return at their preinjury level. Development of validated rehabilitation criteria for safe return to sports after hip arthroscopy for FAI could potentially improve clinical outcomes while also increasing rates of RTP at preinjury levels.

Relationship of Average Outcomes Scores and Change in Status Requires Further Interpretation Between 1 and 2 Years Following Hip Arthroscopy

BACKGROUND

Previous studies have demonstrated a clinically impactful change in patients between 1 and 2 years after hip arthroscopy. Assessment of differences in patient-specific factors between patients who remain the same and those who change (ie, either improve or decline) could provide valuable outcome information for orthopaedic surgeons treating those patients.

PURPOSE

To identify patients who experienced change in functional status between 1 and 2 years after hip arthroscopy for femoroacetabular impingement syndrome and assess differences in patient-specific factors between those who improved, remained the same, or declined in functional status.

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

Prospectively collected data for patients who underwent hip arthroscopy at 1 of 7 centers were analyzed retrospectively at 1 year and 2 years postoperatively. Patients were categorized as "improved,""remained the same," or "declined" between 1- and 2-year follow-up based on the 12-item International Hip Outcome Tool (iHOT-12) minimal clinically important difference (MCID) value. A 1-way analysis of variance was used to assess differences in iHOT-12 scores, age, body mass index (BMI), alpha angle, and center-edge angle (CEA) between groups. Chi-square analyses were used to assess differences in the proportions of male and female patients in the outcome groups.

RESULTS

The study included 753 patients (515 women and 238 men), whose mean ± SD age was 34.7 ± 12 years. Average 1-year (±1 month) and 2-year (±2 months) iHOT-12 scores for all patients were 73.7 and 74.9, respectively. Based on the calculated MCID of ±11.5 points, 162 (21.5%) patients improved, 451 (59.9%) remained the same, and 140 (18.6%) declined in status between 1- and 2-year follow-up. Those who improved between 1 and 2 years had lower 1-year iHOT-12 scores (P < .0005). We found no difference in age, BMI, alpha angle, CEA, or sex between groups (P > .05).

CONCLUSION

Between 1- and 2-year follow-up assessments, 21.5% of patients improved and 18.6% declined in self-reported functional status. Those with iHOT-12 scores indicating abnormal function at 1 year improved beyond the MCID at 2 years follow-up. Thus, any decisions about the failure or success of arthroscopic hip procedures should not be made until at least the 2-year follow-up. Failing to thrive at 1-year follow-up may not accurately predict outcomes at year 2 or beyond. This could potentially decrease the perceived need for revision surgery in patients who do not thrive before 2-year follow-up.

Using Artificial Intelligence-Enhanced Sensing and Wearable Technology in Sports Medicine and Performance Optimisation

Wearable technologies are small electronic and mobile devices with wireless communication capabilities that can be worn on the body as a part of devices, accessories or clothes. Sensors incorporated within wearable devices enable the collection of a broad spectrum of data that can be processed and analysed by artificial intelligence (AI) systems. In this narrative review, we performed a literature search of the MEDLINE, Embase and Scopus databases. We included any original studies that used sensors to collect data for a sporting event and subsequently used an AI-based system to process the data with diagnostic, treatment or monitoring intents. The included studies show the use of AI in various sports including basketball, baseball and motor racing to improve athletic performance. We classified the studies according to the stage of an event, including pre-event training to guide performance and predict the possibility of injuries; during events to optimise performance and inform strategies; and in diagnosing injuries after an event. Based on the included studies, AI techniques to process data from sensors can detect patterns in physiological variables as well as positional and kinematic data to inform how athletes can improve their performance. Although AI has promising applications in sports medicine, there are several challenges that can hinder their adoption. We have also identified avenues for future work that can provide solutions to overcome these challenges.

Meaningless Applications and Misguided Methodologies in Artificial Intelligence-Related Orthopaedic Research Propagates Hype Over Hope

There exists great hope and hype in the literature surrounding applications of artificial intelligence (AI) to orthopaedic surgery. Between 2018 and 2021, a total of 178 AI-related articles were published in orthopaedics. However, for every 2 original research papers that apply AI to orthopaedics, a commentary or review is published (30.3%). AI-related research in orthopaedics frequently fails to provide use cases that offer the uninitiated an opportunity to appraise the importance of AI by studying meaningful questions, evaluating unknown hypotheses, or analyzing quality data. The hype perpetuates a feed-forward cycle that relegates AI to a meaningless buzzword by rewarding those with nascent understanding and rudimentary technical knowhow into committing several basic errors: (1) inappropriately conflating vernacular ("AI/machine learning"), (2) repackaging registry data, (3) prematurely releasing internally validated algorithms, (4) overstating the "black box phenomenon" by failing to provide weighted analysis, (5) claiming to evaluate AI rather than the data itself, and (6) withholding full model architecture code. Relevant AI-specific guidelines are forthcoming, but forced application of the original Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis guidelines designed for regression analyses is irrelevant and misleading. To safeguard meaningful use, AI-related research efforts in orthopaedics should be (1) directed toward administrative support over clinical evaluation and management, (2) require the use of the advanced model, and (3) answer a question that was previously unknown, unanswered, or unquantifiable.

Machine Learning for Predicting Lower Extremity Muscle Strain in National Basketball Association Athletes

Background

In professional sports, injuries resulting in loss of playing time have serious implications for both the athlete and the organization. Efforts to quantify injury probability utilizing machine learning have been met with renewed interest, and the development of effective models has the potential to supplement the decision-making process of team physicians.

Purpose/Hypothesis

The purpose of this study was to (1) characterize the epidemiology of time-loss lower extremity muscle strains (LEMSs) in the National Basketball Association (NBA) from 1999 to 2019 and (2) determine the validity of a machine-learning model in predicting injury risk. It was hypothesized that time-loss LEMSs would be infrequent in this cohort and that a machine-learning model would outperform conventional methods in the prediction of injury risk.

Study Design

Case-control study; Level of evidence, 3.

Methods

Performance data and rates of the 4 major muscle strain injury types (hamstring, quadriceps, calf, and groin) were compiled from the 1999 to 2019 NBA seasons. Injuries included all publicly reported injuries that resulted in lost playing time. Models to predict the occurrence of a LEMS were generated using random forest, extreme gradient boosting (XGBoost), neural network, support vector machines, elastic net penalized logistic regression, and generalized logistic regression. Performance was compared utilizing discrimination, calibration, decision curve analysis, and the Brier score.

Results

A total of 736 LEMSs resulting in lost playing time occurred among 2103 athletes. Important variables for predicting LEMS included previous number of lower extremity injuries; age; recent history of injuries to the ankle, hamstring, or groin; and recent history of concussion as well as 3-point attempt rate and free throw attempt rate. The XGBoost machine achieved the best performance based on discrimination assessed via internal validation (area under the receiver operating characteristic curve, 0.840), calibration, and decision curve analysis.

Conclusion

Machine learning algorithms such as XGBoost outperformed logistic regression in the prediction of a LEMS that will result in lost time. Several variables increased the risk of LEMS, including a history of various lower extremity injuries, recent concussion, and total number of previous injuries.

Duration of Care and Operative Time Are the Primary Drivers of Total Charges After Ambulatory Hip Arthroscopy: A Machine Learning Analysis

PURPOSE

To develop a machine learning algorithm to predict total charges after ambulatory hip arthroscopy and create a risk-adjusted payment model based on patient comorbidities.

METHODS

A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective hip arthroscopy between 2015 and 2016. Features included in initial models consisted of patient characteristics, medical comorbidities, and procedure-specific variables. Models were generated to predict total charges using 5 algorithms. Model performance was assessed by the root-mean-square error, root-mean-square logarithmic error, and coefficient of determination. Global variable importance and partial dependence curves were constructed to show the impact of each input feature on total charges. For performance benchmarking, the best candidate model was compared with a multivariate linear regression using the same input features.

RESULTS

A total of 5,121 patients were included. The median cost after hip arthroscopy was $19,720 (interquartile range, $12,399-$26,439). The gradient-boosted ensemble model showed the best performance (root-mean-square error, $3,800 [95% confidence interval, $3,700-$3,900]; logarithmic root-mean-square error, 0.249 [95% confidence interval, 0.24-0.26]; R² = 0.73). Major cost drivers included total hours in facility less than 12 or more than 15, longer procedure time, performance of a labral repair, age younger than 30 years, Elixhauser Comorbidity Index (ECI) of 1 or greater, African American race, residence in extreme urban and rural areas, and higher household and neighborhood income.

CONCLUSIONS

The gradient-boosted ensemble model effectively predicted total charges after hip arthroscopy. Few modifiable variables were identified other than anesthesia type; nonmodifiable drivers of total charges included duration of care less than 12 hours or more than 15 hours, operating room time more than 100 minutes, age younger than 30 years, performance of a labral repair, and ECI greater than 0. Stratification of patients based on the ECI highlighted the increased financial risk borne by physicians via flat reimbursement schedules given variable degrees of comorbidities.

LEVEL OF EVIDENCE

Level III, retrospective cohort study.

The minimal clinically important difference for the nonarthritic hip score at 2-years following hip arthroscopy

PURPOSE

The purpose of this study was to determine and establish the MCID for the NAHS at 2 years in patients undergoing hip arthroscopy for femoroacetabular impingement syndrome (FAIS).

METHODS

Patients that underwent primary hip arthroscopy for FAIS between 2010 and 2016 were analyzed for eligibility. Data were collected from a single surgeon's hip arthroscopy database. MCID was calculated for the NAHS utilizing a distribution-based method.

RESULTS

The study included 298 patients (184 females) with an average age of 40.4 ± 13.0 years and average body mass index (BMI) of 25.7 ± 4.2 kg/m². At baseline, the cohort's average NAHS score was 48.7 ± 13.6 and demonstrated an improvement of 36.5 ± 17.0 for NAHS at follow-up. This resulted in MCID values of + 8.5 for NAHS.

CONCLUSION

This is the first study to report the MCID (+ 8.5) for NAHS following primary hip arthroscopy, and as such, is a valuable contribution to future hip arthroscopy research.

LEVEL OF EVIDENCE

IV.

Earlier Treatment Yields Superior Outcomes in Competitive Athletes Undergoing Primary Hip Arthroscopy

PURPOSE

To report minimum 2-year patient-reported outcome scores (PROs) and return to sport (RTS) for competitive athletes undergoing primary hip arthroscopy for femoroacetabular impingement syndrome within 1 year of symptom onset and to compare these results with a propensity-matched control group of competitive athletes with symptoms for over 1 year.

METHODS

Data on professional, collegiate, high-school, and organized amateur athletes who underwent primary hip arthroscopy within 1 year of symptom onset between April 2008 and November 2017 were collected. RTS and minimum 2-year PROs were collected for the modified Harris Hip Score (mHHS), Non-arthritic Hip Score (NAHS), Hip Outcome Score-Sport Specific Subscale (HOS-SSS), International Hip Outcome Tool (iHOT-12), and visual analog pain scale (VAS). Rates of achieving minimal clinically important difference (MCID) were also evaluated. These patients were propensity-matched to a control group of competitive athletes with symptoms for over one year for comparison.

RESULTS

Fifty competitive athletes (51 hips, 54.9% female) were included in the study group with a mean follow-up of 70.9 ± 29.1 months and age of 23.6 ± 11.3 years. They demonstrated significant improvement from preoperative to latest follow-up for all recorded PROs (P < .001) and RTS at a rate of 72.9%. When outcomes were compared to the control group, the study group demonstrated similar preoperative scores for all PROs but significantly better minimum 2-year postoperative scores for NAHS (93.8 vs 85.1, P = .0001), HOS-SSS (89.1 vs 77.2, P = .001), iHOT-12 (87.7 vs 76.4, P = 0.011), and VAS (1.5 vs 2.4, P = 0.027). Rates of achieving MCID for HOS-SSS and mHHS were comparable between groups. Further, RTS rates were similar between groups (P = .301).

CONCLUSION

Competitive athletes undergoing primary hip arthroscopy with symptoms for less than 1 year demonstrated superior 2-year PROs compared to a propensity-matched control group with symptoms for over 1 year, but the rates achieving MCID and RTS were similar between groups.

LEVEL OF EVIDENCE

Level III, retrospective comparative study.

Machine Learning Algorithms Predict Achievement of Clinically Significant Outcomes After Orthopaedic Surgery: A Systematic Review

PURPOSE

To determine what subspecialties have applied machine learning (ML) to predict clinically significant outcomes (CSOs) within orthopaedic surgery and to determine whether the performance of these models was acceptable through assessing discrimination and other ML metrics where reported.

METHODS

The PubMed, EMBASE, and Cochrane Central Register of Controlled Trials databases were queried for articles that used ML to predict achievement of the minimal clinically important difference (MCID), patient acceptable symptomatic state (PASS), or substantial clinical benefit (SCB) after orthopaedic surgical procedures. Data pertaining to demographic characteristics, subspecialty, specific ML algorithms, and algorithm performance were analyzed.

RESULTS

Eighteen articles met the inclusion criteria. Seventeen studies developed novel algorithms, whereas one study externally validated an established algorithm. All studies used ML to predict MCID achievement, whereas 3 (16.7%) predicted SCB achievement and none predicted PASS achievement. Of the studies, 7 (38.9%) concerned outcomes after spine surgery; 6 (33.3%), after sports medicine surgery; 3 (16.7%), after total joint arthroplasty (TJA); and 2 (11.1%), after shoulder arthroplasty. No studies were found regarding trauma, hand, elbow, pediatric, or foot and ankle surgery. In spine surgery, concordance statistics (C-statistics) ranged from 0.65 to 0.92; in hip arthroscopy, 0.51 to 0.94; in TJA, 0.63 to 0.89; and in shoulder arthroplasty, 0.70 to 0.95. Most studies reported C-statistics at the upper end of these ranges, although populations were heterogeneous.

CONCLUSIONS

Currently available ML algorithms can discriminate the propensity to achieve CSOs using the MCID after spine, TJA, sports medicine, and shoulder surgery with a fair to good performance as evidenced by C-statistics ranging from 0.6 to 0.95 in most analyses. Less evidence is available on the ability of ML to predict achievement of SCB, and no evidence is available for achievement of the PASS. Such algorithms may augment shared decision-making practices and allow clinicians to provide more appropriate patient expectations using individualized risk assessments. However, these studies remain limited by variable reporting of performance metrics, CSO quantification methods, and adherence to predictive modeling guidelines, as well as limited external validation.

LEVEL OF EVIDENCE

Level III, systematic review of Level III studies.

Personalized Medicine Using Predictive Analytics: A Machine Learning-Based Prognostic Model for Patients Undergoing Hip Arthroscopy

BACKGROUND

Personalized medicine models to predict outcomes of orthopaedic surgery are scarce. Many have required data that are only available postoperatively, mitigating their usefulness in preoperative decision making.

PURPOSE

To establish a method for predictive modeling to enable individualized prognostication and shared decision making based on preoperative patient factors using data from a prospective hip preservation registry.

STUDY DESIGN

Cohort study (Prognosis); Level of evidence, 2.

METHODS

Preoperative data of 2415 patients undergoing hip arthroscopy for femoroacetabular impingement syndrome between February 2008 and November 2017 were retrospectively analyzed. Two machine-learning analyses were evaluated: Tree-structured survival analysis (TSSA) and Cox proportional hazards modeling for predicting time to event and for computing hazard ratios for survivorship after hip arthroscopy. The Fine-Gray model was similarly used for repeat hip arthroscopy. Variables were selected for inclusion using the minimum Akaike Information Criterion (AIC). The stepwise selection was used for the Cox and Fine-Gray models. A web-based calculator was created based on the final models.

RESULTS

Prognostic models were successfully created using Cox proportional hazards modeling and Fine-Gray models for survivorship and repeat hip arthroscopy, respectively. The Harrell C-statistics of the Cox model calculators for survivorship after hip arthroscopy and the Fine-Gray model for repeat hip arthroscopy were 0.848 and 0.662, respectively. Using the AIC, 13 preoperative variables were identified as predictors of survivorship, and 6 variables were identified as predictors for repeat hip arthroscopy. In contrast, the TSSA model performed poorly, resulting in a Harrell C-statistic <0.6, rendering it inaccurate and not interpretable. A web-based calculator was created based on these models.

CONCLUSION

This study successfully created an institution-specific machine learning-based prognostic model for predictive analytics in patients undergoing hip arthroscopy. Prognostic models to predict survivorship and the need for repeat surgeries were both adapted into web-based tools to assist the physician with shared decision making. This prognostic model may be useful at other institutions after performing external validation. Additionally, this study may serve as proof of concept for a methodology for the development of patient-specific prognostic models. This methodology may be used to create other predictive analytics models in different realms of orthopaedic surgery, contributing to the evolution from evidence-based medicine to personalized medicine.

Sports Medicine and Artificial Intelligence: A Primer

Artificial intelligence (AI) represents the fourth industrial revolution and the next frontier in medicine poised to transform the field of orthopaedics and sports medicine, though widespread understanding of the fundamental principles and adoption of applications remain nascent. Recent research efforts into implementation of AI in the field of orthopaedic surgery and sports medicine have demonstrated great promise in predicting athlete injury risk, interpreting advanced imaging, evaluating patient-reported outcomes, reporting value-based metrics, and augmenting the patient experience. Not unlike the recent emphasis thrust upon physicians to understand the business of medicine, the future practice of sports medicine specialists will require a fundamental working knowledge of the strengths, limitations, and applications of AI-based tools. With appreciation, caution, and experience applying AI to sports medicine, the potential to automate tasks and improve data-driven insights may be realized to fundamentally improve patient care. In this Current Concepts review, we discuss the definitions, strengths, limitations, and applications of AI from the current literature as it relates to orthopaedic sports medicine.

Association Between Preoperative Patient Factors and Clinically Meaningful Outcomes After Hip Arthroscopy for Femoroacetabular Impingement Syndrome: A Machine Learning Analysis

BACKGROUND

The International Hip Outcome Tool 12-Item Questionnaire (IHOT-12) has been proposed as a more appropriate outcome assessment for hip arthroscopy populations. The extent to which preoperative patient factors predict achieving clinically meaningful outcomes among patients undergoing hip arthroscopy for femoroacetabular impingement syndrome (FAIS) remains poorly understood.

PURPOSE

To determine the predictive relationship of preoperative imaging, patient-reported outcome measures, and patient demographics with achievement of the minimal clinically important difference (MCID), Patient Acceptable Symptom State (PASS), and substantial clinical benefit (SCB) for the IHOT-12 at a minimum of 2 years postoperatively.

STUDY DESIGN

Case-control study; Level of evidence, 3.

METHODS

Data were analyzed for consecutive patients who underwent hip arthroscopy for FAIS between 2012 and 2018 and completed the IHOT-12 preoperatively and at a minimum of 2 years postoperatively. Fifteen novel machine learning algorithms were developed using 47 potential demographic, clinical, and radiographic predictors. Model performance was evaluated with discrimination, calibration, decision-curve analysis and the brier score.

RESULTS

A total of 859 patients were identified, with 685 (79.7%) achieving the MCID, 535 (62.3%) achieving the PASS, and 498 (58.0%) achieving the SCB. For predicting the MCID, discrimination for the best-performing models ranged from fair to excellent (area under the curve [AUC], 0.69-0.89), although calibration was excellent (calibration intercept and slopes: -0.06 to 0.02 and 0.24 to 0.85, respectively). For predicting the PASS, discrimination for the best-performing models ranged from fair to excellent (AUC, 0.63-0.81), with excellent calibration (calibration intercept and slopes: 0.03-0.18 and 0.52-0.90, respectively). For predicting the SCB, discrimination for the best-performing models ranged from fair to good (AUC, 0.61-0.77), with excellent calibration (calibration intercept and slopes: -0.08 to 0.00 and 0.56 to 1.02, respectively). Thematic predictors for failing to achieve the MCID, PASS, and SCB were presence of back pain, anxiety/depression, chronic symptom duration, preoperative hip injections, and increasing body mass index (BMI). Specifically, thresholds associated with lower likelihood to achieve a clinically meaningful outcome were preoperative Hip Outcome Score-Activities of Daily Living <55, preoperative Hip Outcome Score-Sports Subscale >55.6, preoperative IHOT-12 score ≥48.5, preoperative modified Harris Hip Score ≤51.7, age >41 years, BMI ≥27, and preoperative α angle >76.6°.

CONCLUSION

We developed novel machine learning algorithms that leveraged preoperative demographic, clinical, and imaging-based features to reliably predict clinically meaningful improvement after hip arthroscopy for FAIS. Despite consistent improvements after hip arthroscopy, meaningful improvements are negatively influenced by greater BMI, back pain, chronic symptom duration, preoperative mental health, and use of hip corticosteroid injections.

Predictors of Achieving the Patient Acceptable Symptomatic State at Minimum 5-Year Follow-up Following Primary Hip Arthroscopy in the Adolescent Athlete

BACKGROUND

Predictors of outcomes in adolescent athletes undergoing hip arthroscopy have not been established. The purpose of this study was to identify factors correlated with achieving the Patient Acceptable Symptomatic State (PASS) for the Hip Outcome Score-Sports Specific Subscale (HOS-SSS) at a minimum 5-year follow-up in adolescent athletes undergoing primary hip arthroscopy.

METHODS

Data were reviewed on adolescent (below 18 y old) athletes who underwent primary hip arthroscopy for femoroacetabular impingement syndrome between March 2008 and October 2015. Patients were included if they participated in sport within 1 year before surgery and had preoperative, 2-year, and minimum 5-year postoperative patient-reported outcome (PRO) scores for the modified Harris Hip Score, HOS-SSS, Visual Analog Scale for pain (VAS), and Non-Arthritic Hip Score (NAHS). Patients were divided into 2 groups based on whether they achieved PASS for HOS-SSS. Demographics, intraoperative findings, radiographic variables, surgical treatments, and PRO were compared. Multivariate logistic regression with corresponding odds ratios (ORs) quantified the correlation between variables and achievement of the PASS.

RESULTS

A total of 123 athletes with a mean age of 16.2±1.1 years were included. These athletes demonstrated significant imrpovement from preoperative to minimum 5-year follow-up for all recorded PROs (P<0.001). The multivariate logistic regression model identified preoperative NAHS (P=0.019, OR: 1.033), 2-year postoperative HOS-SSS (P=0.014, OR: 1.037), and 2-year postoperative VAS (P=0.003, OR: 0.590) as statistically significantly correlated with achieving the PASS. Athletes with a 2-year postoperative VAS pain score ≤2 achieved PASS at a rate of 81.9%, while those with a score >2 achieved PASS at a rate of 24.1% (P<0.001, OR: 14.2, 95% confidence interval: 5.23-38.7).

CONCLUSIONS

Favorable outcome were achieved at mid-term follow-up in adolescent athletes undergoing primary hip arthroscopy. Preoperative NAHS, 2-year postoperative HOS-SSS, and 2-year postoperative VAS pain scores were correlated with achieving the PASS for HOS-SSS at a minimum 5-year follow-up. Patients with 2-year postoperative VAS ≤2 were significantly more likely to achieve the PASS at 5-year follow-up than those with scores >2.

LEVEL OF EVIDENCE

Level III-case-control study.

Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction

Purpose

To develop and internally validate a machine-learning algorithm to reliably predict cost after anterior cruciate ligament reconstruction (ACLR).

Methods

A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective ACLR from 2015 to 2016. Features included in initial models consisted of patient characteristics (age, sex, insurance status, income, medical comorbidities as classified by the Clinical Classifications Software diagnosis code) as well as intraoperative variables (type of anesthesia and procedure-specific factors). Models were generated to predict total charges using 4 algorithms: random forest, extreme gradient boost, elastic net penalized regression, and support vector machines with radial kernels. Training was performed with 10-fold cross-validation followed by internal validation via 0.632 bootstrapping. Model discriminative performance was assessed by area under the receiver operating characteristic curve, calibration, and the Brier score. Decision curve analysis was performed to demonstrate the net benefit of using the final model in practice.

Results

In total, 7,311 patients undergoing ambulatory ACLR were included. The random forest model demonstrated the best performance assessed via internal validation (area under the curve = 0.85), calibration, and the Brier score (0.208). Cost incurred was influenced by anesthesia type, operating room time, and number of chronic comorbidities. Decision curve analysis revealed a net benefit for use of the random forest model and the model was integrated into a web-based open-access application.

Conclusions

The random forest model predicted cost after ambulatory ACLR using a large, statewide database with good performance. The top variables found to predict increased charges were general anesthesia, operating room time, meniscal repair, self-pay insurance, patient neighborhood characteristics, and number of chronic conditions.

Level of Evidence

III, retrospective cohort study.

Repeat Revision Hip Arthroscopy Outcomes Match That of Initial Revision But Not That of Primary Surgery for Femoroacetabular Impingement Syndrome

PURPOSE

To (1) report on pre- and postoperative patient-reported outcome (PRO) scores for patients undergoing repeat revision surgery in short-term follow-up and (2) compare minimal clinically important difference (MCID) and patient acceptable symptomatic state achievement between primary, revision, and repeat revision hip arthroscopy cohorts.

METHODS

Data from consecutive patients undergoing revision hip arthroscopy from January 2012 to February 2019 were retrospectively reviewed. Hips that underwent 2 revision hip arthroscopic surgeries were identified and matched 1:3 to patients undergoing revision surgery and 1:3 to patients undergoing primary surgery by age, sex, and body mass index. Baseline demographic data, surgical indications, and hip-specific PROs were collected were obtained preoperatively and at minimum 1-year follow-up. MCID was calculated individually for each cohort.

RESULTS

Twenty patients who underwent repeat revision were matched to 60 patients who underwent revision and 60 primary patients. Patients who underwent repeat revision achieved MCID on all investigated PROs at a similar rate to patients undergoing primary surgery (90.0% vs 91.7%, P = .588) and at a greater rate than patients undergoing first-time revision surgery (90.0% vs 71.7%, P = .045). Patients who underwent repeat revision achieved patient acceptable symptomatic state on all investigated PROs at a similar rate to patients who underwent first-time revision (30.0% vs 55.0%, P = .053) but at a significantly lower rate than primary patients (30.0% vs 76.7%, P < .001). However, patients undergoing repeat revision surgery had significantly lower preoperative PROs (P < .001 for all) and no significant difference in PROs at minimum 1-year follow-up compared with patients undergoing revision (P > .05). Compared with the primary cohort, patients who underwent repeat revision had significantly lower Hip Outcome Score-Activities of Daily Living (77.3 ± 16.7 vs 86.1 ± 14.4; P = .034), Hip Outcome Score-Sports Subscale (60.6 ± 27.2 vs 76.1 ± 23.8; P < .001), and modified Harris Hip Score (69.2 ± 19.3 vs 81.7 ± 16.1; P = .048) at a minimum of 1-year follow-up.

CONCLUSIONS

Second-time revision hip arthroscopy, which often requires advanced procedures, results in clinically significant improvement in PROs; however, outcomes for repeat revision cases are similar to first-time revision cases but inferior to those obtained following primary surgeries.

LEVEL OF EVIDENCE

Level III, retrospective case-control study.

Determining Clinically Meaningful Thresholds for the Nonarthritic Hip Score in Patients Undergoing Arthroscopy for Femoroacetabular Impingement Syndrome

PURPOSE

The purpose of this study was to improve the interpretability of the Nonarthritic Hip Score (NAHS) by determining the minimal clinically important difference (MCID), patient acceptable symptomatic state (PASS), and substantial clinical benefit (SCB) after hip arthroscopy for femoroacetabular impingement. The secondary aim was to identify variables associated with achievement of the thresholds.

METHODS

Patients who underwent hip arthroscopy for femoroacetabular impingement and completed postoperative questionnaires between August 2019 and March 2020 were included. Patients were excluded if they underwent previous ipsilateral hip surgery, underwent gluteus medius repair, or had a previous hip condition. The MCID, PASS, and SCB thresholds were calculated for the NAHS at minimum 1-, 2-, and 5-year follow-up. Distribution- and anchor-based methods with receiver operating characteristic analysis were used to determine the thresholds. Multivariate logistic regression was performed to determine predictors of achieving the MCID and PASS.

RESULTS

The study included 343 hips with an average follow-up period of 48 months. On the basis of the distribution-based approach, the MCID was 8.7 using a method in which the standard deviation of the baseline score was divided by 2. By use of a method in which the standard deviation of the change score was divided by 2, MCID scores of 9.1, 8.3, and 12.6 resulted at a minimum of 1, 2, and 5 years, respectively. The PASS thresholds for these time points were 81.9, 85.6, and 81.9. The absolute SCB thresholds were 91.9, 94.4, and 93.1 and the change score thresholds were 30.6, 24.4, and 29.3 for a minimum of 1, 2, and 5 years, respectively. The preoperative NAHS was positively associated with achievement of the PASS and inversely related to achievement of the MCID.

CONCLUSIONS

This study provides important clinical thresholds for the NAHS. These thresholds were determined for minimum 1-, 2-, and 5-year time points. The MCID was determined as 8.7, the PASS ranged between 81.9 and 85.6, and the absolute SCB value ranged from 91.9 to 94.4. The preoperative NAHS was found to be positively associated with achievement of the PASS and inversely related to achievement of the MCID.

LEVEL OF EVIDENCE

Level IV, retrospective case series.

Psychological Healthcare Burden Lessens After Hip Arthroscopy for Those With Comorbid Depression or Anxiety

Purpose

In this study, we investigated whether patients undergoing arthroscopic treatment of femoral acetabular impingement syndrome (FAIS) seek health care for treatment of comorbid depression and anxiety in the year following hip arthroscopy.

Methods

Using the Truven Health Marketscan database, FAIS patients who underwent hip arthroscopy between January 2009 and December 2016 were identified. Claims related to depression or anxiety filed during the year before surgery were required for inclusion. Using claims for pharmaceutical and psychological therapy treatments for mental health disorders, four groups of patients were analyzed on the basis of preoperative anxiety/depression treatment: medication only, therapy only, medication + therapy, and no treatment. Number of opioid pain prescriptions within 180 days prior to surgery and >90 days after hip arthroscopy were also compared.

Results

Depression and anxiety claims were identified in 5,208/14,830 (35.1%) patients. Preoperative treatment for depression and anxiety included medication only (n = 648, 12.4%), therapy only (n = 899, 17.3%), medication + therapy (n = 252, 4.8%), and no treatment (n = 3,409, 65.5%). Of the 900 patients who filled an anxiety/depression-related prescription prior to surgery, 422 (46.9%) patients did not fill a similar prescription in the postoperative year. Of the 1,151 patients receiving anxiety/depression-related therapy prior to surgery, 549 (47.7%) did not receive therapy in the postoperative year. Preoperative opioid prescriptions were filled for 393 patients (60.6%) in medication-only group, 275 (30.6%) in therapy-only group, 156 (61.9%) in medication + therapy group, and 1,059 (31.1%) in the group receiving no treatment. Opioid prescriptions >90 days postoperatively were filled for 330 (50.9%), 225 (25.0%), 120 (47.6%), and 861 (25.3%) patients, respectively.

Conclusion

Hip arthroscopy for FAIS is associated with a decreased postoperative use of health care resources for the treatment of depression and anxiety. Clinicians should also be aware of the potential interplay between preoperative psychotropic medication use and prolonged opioid use when counseling patients.

Level of Evidence

IV, therapeutic case series.

Predicting the Risk of Subsequent Hip Surgery Before Primary Hip Arthroscopy for Femoroacetabular Impingement Syndrome: A Machine Learning Analysis of Preoperative Risk Factors in Hip Preservation

BACKGROUND

The number of patients requiring reoperation has increased as the volume of hip arthroscopy for femoroacetabular impingement syndrome (FAIS) has increased. The factors most important in determining patients who are likely to require reoperation remain elusive.

PURPOSE

To leverage machine learning to better characterize the complex relationship across various preoperative factors (patient characteristics, radiographic parameters, patient-reported outcome measures [PROMs]) for patients undergoing primary hip arthroscopy for FAIS to determine which features predict the need for future ipsilateral hip reoperation, namely, revision hip arthroscopy, total hip arthroplasty (THA), hip resurfacing arthroplasty (HRA), or periacetabular osteotomy (PAO).

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

A cohort of 3147 patients undergoing 3748 primary hip arthroscopy procedures were included from an institutional hip preservation registry. Preoperative computed tomography of the hip was obtained for each patient, from which the following parameters were calculated: the alpha angle; the coronal center-edge angle; the neck-shaft angle; the acetabular version angle at 1, 2, and 3 o'clock; and the femoral version angle. Preoperative PROMs included the modified Harris Hip Score (mHHS), the Hip Outcome Score (HOS)-Activities of Daily Living subscale (HOS-ADL) and the Sport Specific subscale, and the international Hip Outcome Tool (iHOT-33). Random forest models were created for revision hip arthroscopy, the THA, the HRA, and the PAO. Area under the curve (AUC) for the receiver operating characteristic curve and accuracy were calculated to evaluate each model.

RESULTS

A total of 171 patients (4.6%) underwent subsequent hip surgery after primary hip arthroscopy for FAIS. The AUC and accuracy, respectively, were 0.77 (fair) and 76% for revision hip arthroscopy (mean, 26.4-month follow-up); 0.80 (good) and 81% for THA (mean, 32.5-month follow-up); 0.62 (poor) and 69% for HRA (mean, 45.4-month follow-up); and 0.76 (fair) and 74% for PAO (mean, 30.4-month follow-up). The most important factors in predicting reoperation after primary hip arthroscopy were higher body mass index (BMI) and lower preoperative HOS-ADL for revision hip arthroscopy, greater age and lower preoperative iHOT-33 for THA, increased BMI for HRA, and larger neck-shaft angle and lower preoperative mHHS for PAO.

CONCLUSION

Despite the low failure rate of hip arthroscopy for FAIS, our study demonstrated that machine learning has the capability to identify key preoperative risk factors that may predict subsequent ipsilateral hip surgery before the index hip arthroscopy. Knowledge of these demographic, radiographic, and patient-reported outcome data may aid in preoperative counseling and expectation management to better optimize hip preservation.

Machine Learning Algorithms Predict Functional Improvement After Hip Arthroscopy for Femoroacetabular Impingement Syndrome in Athletes

BACKGROUND

Despite previous reports of improvements for athletes following hip arthroscopy for femoroacetabular impingement syndrome (FAIS), many do not achieve clinically relevant outcomes. The purpose of this study was to develop machine learning algorithms capable of providing patient-specific predictions of which athletes will derive clinically relevant improvement in sports-specific function after undergoing hip arthroscopy for FAIS.

METHODS

A registry was queried for patients who had participated in a formal sports program or athletic activities before undergoing primary hip arthroscopy between January 2012 and February 2018. The primary outcome was achieving the minimal clinically important difference (MCID) in the Hip Outcome Score-Sports Subscale (HOS-SS) at a minimum of 2 years postoperatively. Recursive feature selection was used to identify the combination of variables, from an initial pool of 26 features, that optimized model performance. Six machine learning algorithms (stochastic gradient boosting, random forest, adaptive gradient boosting, neural network, support vector machine, and elastic-net penalized logistic regression [ENPLR]) were trained using 10-fold cross-validation 3 times and applied to an independent testing set of patients. Models were evaluated using discrimination, decision-curve analysis, calibration, and the Brier score.

RESULTS

A total of 1,118 athletes were included, and 76.9% of them achieved the MCID for the HOS-SS. A combination of 6 variables optimized algorithm performance, and specific cutoffs were found to decrease the likelihood of achieving the MCID: preoperative HOS-SS score of ≥58.3, Tönnis grade of 1, alpha angle of ≥67.1°, body mass index (BMI) of >26.6 kg/m2, Tönnis angle of >9.7°, and age of >40 years. The ENPLR model demonstrated the best performance (c-statistic: 0.77, calibration intercept: 0.07, calibration slope: 1.22, and Brier score: 0.14). This model was transformed into an online application as an educational tool to demonstrate machine learning capabilities.

CONCLUSIONS

The ENPLR machine learning algorithm demonstrated the best performance for predicting clinically relevant sports-specific improvement in athletes who underwent hip arthroscopy for FAIS. In our population, older athletes with more degenerative changes, high preoperative HOS-SS scores, abnormal acetabular inclination, and an alpha angle of ≥67.1° achieved the MCID less frequently. Following external validation, the online application of this model may allow enhanced shared decision-making.

Editorial Commentary: Personalized Hip Arthroscopy Outcome Prediction Using Machine Learning-The Future Is Here

Machine learning and artificial intelligence are increasingly used in modern health care, including arthroscopic and related surgery. Multiple high-quality, Level I evidence, randomized, controlled investigations have recently shown the ability of hip arthroscopy to successfully treat femoroacetabular impingement syndrome and labral tears. Contemporary hip preservation practice strives to continually refine and improve the value of care provision. Multiple single-center and multicenter prospective registries continue to grow as part of both United States-based and international hip preservation-specific networks and collaborations. The ability to predict postoperative patient-reported outcomes preoperatively holds great promise with machine learning. Machine learning requires massive amounts of data, which can easily be generated from electronic medical records and both patient- and clinician-generated questionnaires. On top of text-based data, imaging (e.g., plain radiographs, computed tomography, and magnetic resonance imaging) can be rapidly interpreted and used in both clinical practice and research. Formidable computational power is also required, using different advanced statistical methods and algorithms to generate models with the ability to predict individual patient outcomes. Efficient integration of machine learning into hip arthroscopy practice can reduce physicians' "busywork" of data collection and analysis. This can only improve the value of the patient experience, because surgeons have more time for shared decision making, with empathy, compassion, and humanity counterintuitively returning to medicine.

Clinical and Research Medical Applications of Artificial Intelligence

Artificial intelligence (AI), including machine learning (ML), has transformed numerous industries through newfound efficiencies and supportive decision-making. With the exponential growth of computing power and large datasets, AI has transitioned from theory to reality in teaching machines to automate tasks without human supervision. AI-based computational algorithms analyze "training sets" using pattern recognition and learning from inputted data to classify and predict outputs that otherwise could not be effectively analyzed with human processing or standard statistical methods. Though widespread understanding of the fundamental principles and adoption of applications have yet to be achieved, recent applications and research efforts implementing AI have demonstrated great promise in predicting future injury risk, interpreting advanced imaging, evaluating patient-reported outcomes, reporting value-based metrics, and augmenting telehealth. With appreciation, caution, and experience applying AI, the potential to automate tasks and improve data-driven insights may be realized to fundamentally improve patient care. The purpose of this review is to discuss the pearls, pitfalls, and applications associated with AI.

Development and Internal Validation of Supervised Machine Learning Algorithms for Predicting Clinically Significant Functional Improvement in a Mixed Population of Primary Hip Arthroscopy

PURPOSE

To (1) develop and validate a machine learning algorithm to predict clinically significant functional improvements after hip arthroscopy for femoroacetabular impingement syndrome and to (2) develop a digital application capable of providing patients with individual risk profiles to determine their propensity to gain clinically significant improvements in function.

METHODS

A retrospective review of consecutive hip arthroscopy patients who underwent cam/pincer correction, labral preservation, and capsular closure between January 2012 and 2017 from 1 large academic and 3 community hospitals operated on by a single high-volume hip arthroscopist was performed. The primary outcome was the minimal clinically important difference (MCID) for the Hip Outcome Score (HOS)-Activities of Daily Living (ADL) at 2 years postoperatively, which was calculated using a distribution-based method. A total of 21 demographic, radiographic, and patient-reported outcome measures were considered as potential covariates. An 80:20 random split was used to create training and testing sets from the patient cohort. Five supervised machine learning algorithms were developed using 3 iterations of 10-fold cross-validation on the training set and assessed by discrimination, calibration, Brier score, and decision curve analysis on an independent testing set of patients.

RESULTS

A total of 818 patients with a median (interquartile range) age of 32.0 (22.0-42.0) and 69.2% female were included, of whom 74.3% achieved the MCID for the HOS-ADL. The best-performing algorithm was the stochastic gradient boosting model (c-statistic = 0.84, calibration intercept = 0.20, calibration slope = 0.83, and Brier score = 0.13). Of the initial 21 candidate variables, the 8 most important features for predicting the MCID for the HOS-ADL included in model training were body mass index, age, preoperative HOS-ADL score, preoperative pain level, sex, Tönnis grade, symptom duration, and drug allergies. The algorithm was subsequently transformed into a digital application using local explanations to provide customized risk assessment: https://orthoapps.shinyapps.io/HPRG_ADL/.

CONCLUSIONS

The stochastic boosting gradient model conferred excellent predictive ability for propensity to gain clinically significant improvements in function after hip arthroscopy. An open-access digital application was created, which may augment shared decision-making and allow for preoperative risk stratification. External validation of this model is warranted to confirm the performance of these algorithms, as the generalizability is currently unknown.

LEVEL OF EVIDENCE

IV, Case series.

Evaluation of outcome reporting trends for femoroacetabular impingement syndrome- a systematic review

PURPOSE

The aim of this systematic review was to evaluate the trends in the literature regarding surgical treatment for femoroacetabular impingement syndrome (FAIS) and to present which patient-reported outcome-measures (PROMs) and surgical approaches are included.

METHODS

This systematic review was conducted with the PRISMA guidelines. The literature search was performed on PubMed and Embase, covering studies from 1999 to 2020. Inclusion criteria were clinical studies with surgical treatment for FAIS, the use of PROMs as evaluation tool and studies in English. Exclusion criteria were studies with patients < 18 years, cohorts with < 8 patients, studies with primarily purpose to evaluate other diagnoses than FAIS and studies with radiographs as only outcomes without using PROMs. Data extracted were author, year, surgical intervention, type of study, level of evidence, demographics of included patients, and PROMs.

RESULTS

The initial search yielded 2,559 studies, of which 196 were included. There was an increase of 2,043% in the number of studies from the first to the last five years (2004-2008)-(2016-2020). There were 135 (69%) retrospective, 55 (28%) prospective and 6 (3%) Randomized Controlled Trials. Level of evidence ranged from I-IV where Level III was most common (44%). More than half of the studies (58%) originated from USA. Arthroscopic surgery was the most common surgical treatment (85%). Mean follow-up was 27.0 months (± 17 SD), (range 1.5-120 months). Between 1-10 PROMs were included, and the modified Harris Hip Score (mHHS) was most commonly used (61%).

CONCLUSION

There has been a continuous increase in the number of published studies regarding FAIS with the majority evaluating arthroscopic surgery. The mHHS remains being the most commonly used PROM.

Machine Learning Algorithms Predict Clinically Significant Improvements in Satisfaction After Hip Arthroscopy

PURPOSE

To develop machine learning algorithms to predict failure to achieve clinically significant satisfaction after hip arthroscopy.

METHODS

We queried a clinical repository for consecutive primary hip arthroscopy patients treated between January 2012 and January 2017. Five supervised machine learning algorithms were developed in a training set of patients and internally validated in an independent testing set of patients by discrimination, Brier score, calibration, and decision-curve analysis. The minimal clinically important difference (MCID) for the visual analog scale (VAS) score for satisfaction was derived by an anchor-based method and used as the primary outcome.

RESULTS

A total of 935 patients were included, of whom 148 (15.8%) did not achieve the MCID for the VAS satisfaction score at a minimum of 2 years postoperatively. The best-performing algorithm was the neural network model (C statistic, 0.94; calibration intercept, -0.43; calibration slope, 0.94; and Brier score, 0.050). The 5 most important features to predict failure to achieve the MCID for the VAS satisfaction score were history of anxiety or depression, lateral center-edge angle, preoperative symptom duration exceeding 2 years, presence of 1 or more drug allergies, and Workers' Compensation.

CONCLUSIONS

Supervised machine learning algorithms conferred excellent discrimination and performance for predicting clinically significant satisfaction after hip arthroscopy, although this analysis was performed in a single population of patients. External validation is required to confirm the performance of these algorithms.

LEVEL OF EVIDENCE

Level III, therapeutic case-control study.

Large Heterogeneity Among Minimal Clinically Important Differences for Hip Arthroscopy Outcomes: A Systematic Review of Reporting Trends and Quantification Methods

PURPOSE

To perform a systematic review of reporting trends and quantification methods for the minimal clinically important difference (MCID) within the hip arthroscopy literature.

METHODS

Cochrane, PubMed, and OVID/MEDLINE databases were queried for hip arthroscopy articles that reported the MCID. Studies were classified as (1) calculating new MCID values for their specific study-population or (2) referencing previously established MCID values. Data pertaining to patient demographics, study characteristics, outcome measures, method of MCID quantification, MCID value, anchor questions, measurement error, and study from which referenced MCID values were obtained were extracted.

RESULTS

A total of 59 articles with 18,830 patients (19,867 hips) was included. A total of 19 unique outcome measures was reported. A total of 33 (n = 55.9%) studies (follow-up range 6-60 months) used previously established MCID values to assess their study population (MCID values established at a follow-up range 6-31 months). The remaining 26 studies (44.1%) performed new MCID calculations. The MCID values were inconsistent and varied widely (Hip Outcome Score-Activities of Daily Living: 5.0-15.4; Hip Outcome Score-Sports Subscale: 6-25; modified Harris hip score: 2.4-20.9). Among the 33 studies that used previously established MCID values, 10 different studies were cited as the reference. Among the remaining 26 studies that calculated a new MCID value, the most common method was 0.5 standard deviation method (n = 21, 80.8%). Only 3 of 26 (11.5%) studies reported a measurement of error in conjunction with their MCID values.

CONCLUSIONS

Inconsistencies in MCID reporting and quantification methods led to a wide range of MCID values for commonly administered outcome measures within the hip arthroscopy literature-even for the same outcome measures. The majority of studies referenced previously established MCID values with variable ranges of follow-up and applied those values to assess their specific study population at varying follow-ups.

LEVEL OF EVIDENCE

IV, systematic review.

Machine-learning algorithm that can improve the diagnostic accuracy of septic arthritis of the knee

PURPOSE

Prompt diagnosis and treatment of septic arthritis of the knee is crucial. Nevertheless, the quality of evidence for the diagnosis of septic arthritis is low. In this study, the authors developed a machine learning-based diagnostic algorithm for septic arthritis of the native knee using clinical data in an emergency department and validated its diagnostic accuracy.

METHODS

Patients (n = 326) who underwent synovial fluid analysis at the emergency department for suspected septic arthritis of the knee were enrolled. Septic arthritis was diagnosed in 164 of the patients (50.3%) using modified Newman criteria. Clinical characteristics of septic and inflammatory arthritis were compared. Area under the receiver-operating characteristic (ROC) curve (AUC) statistics was applied to evaluate the efficacy of each variable for the diagnosis of septic arthritis. The dataset was divided into independent training and test sets (comprising 80% and 20%, respectively, of the data). Supervised machine-learning techniques (random forest and eXtreme Gradient Boosting: XGBoost) were applied to develop a diagnostic model using the training dataset. The test dataset was subsequently used to validate the developed model. The ROC curves of the machine-learning model and each variable were compared.

RESULTS

Synovial white blood cell (WBC) count was significantly higher in septic arthritis than in inflammatory arthritis in the multivariate analysis (P = 0.001). In the ROC comparison analysis, synovial WBC count yielded a significantly higher AUC than all other single variables (P = 0.002). The diagnostic model using the XGBoost algorithm yielded a higher AUC (0.831, 95% confidence interval 0.751-0.923) than synovial WBC count (0.740, 95% confidence interval 0.684-0.791; P = 0.033). The developed algorithm was deployed as a free access web-based application ( www.septicknee.com ).

CONCLUSION

The diagnosis of septic arthritis of the knee might be improved using a machine learning-based prediction model.

LEVEL OF EVIDENCE

Diagnostic study Level III (Case-control study).

Development of Machine Learning Algorithms to Predict Being Lost to Follow-up After Hip Arthroscopy for Femoroacetabular Impingement Syndrome

Purpose

To determine factors predictive of patients who are at risk for being lost to follow-up after hip arthroscopy for femoroacetabular impingement syndrome (FAIS).

Methods

A prospective clinical repository was queried between January 2012 and October 2017 and all patients who underwent hip arthroscopy for primary or revision FAIS with minimum 2-year follow-up were included. A total of 27 potential risk factors for loss to follow-up were available and tested for predictive value. An 80:20 random sample split of all patients was performed to create training and testing sets. Cross-validation, minimum Bayes information criteria, and adaptive machine-learning algorithms were used to develop the predictive model. The model with the best predictive performance was selected based off of the lowest postestimation deviance between the training and testing samples. The c-statistic is a measure of discrimination. It ranges from 0.5 to 1.0, with 1.0 being perfect discrimination and 0.5 indicating the model is no better than chance. A log-likelihood χ² test was used to evaluate the goodness-of-fit of the logistic regression model.

Results

A total of 2113 patients were included. Inference of minimum Bayes information criteria model indicated that male sex (odds ratio [OR] 1.82, P = .028), non-white race (African American OR 2.41, P = .013; other non-white OR 1.42, P = .042), smoking (OR 1.07, P = .021), and failure to provide a phone number (OR 1.78, P = .032) increased the risk for being lost to follow-up. Furthermore, greater preoperative International Hip Outcome Tool 12-item component questionnaire (OR 1.03, P = .004), and modified Harris Hip Score (OR 1.05, P = .014) scores increased the risk of being lost to follow-up. The c-statistic was 0.76 (95% confidence interval 0.701-0.848). The log-likelihood indicated that the regression model as a whole was statistically significant (P = .002).

Conclusions

Patients who are male, non-white, smokers, fail to provide a telephone number, and have greater preoperative modified Harris Hip Score and International Hip Outcome Tool 12-item component questionnaire scores are at an increased risk for being lost to follow-up 2 years after hip arthroscopy for FAIS.

Level of Evidence

Level III, case control study.

Radiographic Indices Are Not Predictive of Clinical Outcomes Among 1735 Patients Indicated for Hip Arthroscopic Surgery: A Machine Learning Analysis

BACKGROUND

The relationship between the preoperative radiographic indices for femoroacetabular impingement syndrome (FAIS) and postoperative patient-reported outcome measure (PROM) scores continues to be under investigation, with inconsistent findings reported.

PURPOSE

To apply a machine learning model to determine which preoperative radiographic indices, if any, among patients indicated for the arthroscopic correction of FAIS predict whether a patient will achieve the minimal clinically important difference (MCID) for 1- and 2-year PROM scores.

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

A total of 1735 consecutive patients undergoing primary hip arthroscopic surgery for FAIS were included from an institutional hip preservation registry. Patients underwent preoperative computed tomography of the hip, from which the following radiographic indices were calculated by a musculoskeletal radiologist: alpha angle, beta angle, sagittal center-edge angle, coronal center-edge angle, neck shaft angle, acetabular version angle, and femoral version angle. PROM scores were collected preoperatively, at 1 year postoperatively, and at 2 years postoperatively for the modified Harris Hip Score (mHHS), the Hip Outcome Score (HOS)-Activities of Daily Living (HOS-ADL) and -Sport Specific (HOS-SS), and the International Hip Outcome Tool (iHOT-33). Random forest models were created for each PROM at 1 and 2 years' follow-up, with each PROM's MCID used to establish clinical meaningfulness. Data inputted into the models included ethnicity, laterality, sex, age, body mass index, and radiographic indices. Comprehensive and separate models were built specifically to assess the association of the alpha angle, femoral version angle, coronal center-edge angle, McKibbin index, and hip impingement index with respect to each PROM.

RESULTS

As evidenced by poor area under the curves and P values >.05 for each model created, no combination of radiographic indices or isolated index (alpha angle, coronal center-edge angle, femoral version angle, McKibbin index, hip impingement index) was a significant predictor of a clinically meaningful improvement in scores on the mHHS, HOS-ADL, HOS-SS, or iHOT-33. The mean difference between 1- and 2-year PROM scores compared with preoperative values exceeded the respective MCIDs for the cohort.

CONCLUSION

In patients appropriately indicated for FAIS corrective surgery, clinical improvements can be achieved, regardless of preoperative radiographic indices, such as the femoral version angle, coronal center-edge angle, and alpha angle. No specific radiographic parameter or combination of indices was found to be predictive of reaching the MCID for any of the 4 studied hip-specific PROMs at either 1 or 2 years' follow-up.