Machine learning approach to predict venous thromboembolism among patients undergoing multi-level spinal posterior instrumented fusion.

Background: The absence of consensus for prophylaxis of venous thromboembolism (VTE) in spine surgery underscores the importance of identifying patients at risk. This study incorporated machine learning (ML) models to assess key risk factors of VTE in patients who underwent posterior spinal instrumented fusion. Methods: Data was collected from the IBM MarketScan Database [2009-2021] for patients ≥18 years old who underwent spinal posterior instrumentation (3-6 levels), excluding traumas, malignancies, and infections. VTE incidence (deep vein thrombosis and pulmonary embolism) was recorded 90-day post-surgery. Risk factors for VTE were investigated and compared through several ML models including logistic regression, linear support vector machine (LSVM), random forest, XGBoost, and neural networks. Results: Among the 141,697 patients who underwent spinal fusion with posterior instrumentation (3-6 levels), the overall 90-day VTE rate was 3.81%. The LSVM model demonstrated the best prediction with an area under the curve (AUC) of 0.68. The most important features for prediction of VTE included remote history of VTE, diagnosis of chronic hypercoagulability, metastatic cancer, hemiplegia, and chronic renal disease. Patients who did not have these five key risk factors had a 90-day VTE rate of 2.95%. Patients who had an increasing number of key risk factors had subsequently higher risks of postoperative VTE. Conclusions: The analysis of the data with different ML models identified 5 key variables that are most closely associated with VTE. Using these variables, we have developed a simple risk model with additive odds ratio ranging from 2.80 (1 risk factor) to 46.92 (4 risk factors) over 90 days after posterior spinal fusion surgery. These findings can help surgeons risk-stratify their patients for VTE risk, and potentially guide subsequent chemoprophylaxis.

Database analysis comparing incidence and complication rates between inpatient and outpatient laminotomies for lumbar disc herniation.

Background: Lumbar disc herniation (LDH) is a common condition that can be characterized with disabling pain. While most patients recover without surgery, some still require operative intervention. The epidemiology and trends of laminotomy for LDH have not been recently studied, and current practice patterns might be different from historical norms. This study aimed to investigate the trends of inpatient and outpatient laminotomies for LDH and compare complication rates between these two sites of service. Methods: A large, national database was utilized to identify patients > 8 years old who underwent a laminotomy for LDH between 2009 and 2019. Two cohorts were created based on site of surgery: inpatient versus outpatient. The outpatient cohort was defined as patients who had a length of stay less than 1 day without any associated hospitalization. Epidemiologic analyses for these cohorts were performed by demographics. Patients in both groups were then 1:1 propensity-score matched based on age, sex, insurance type, geographic region, and comorbidities. Ninety-day postoperative complications were compared between cohorts utilizing multivariate logistic regressions. Results: The average incidence of laminotomy for LDH was 13.0 per 10,000 persons-years. Although the national trend in incidence had not changed from 2009 to 2019, the proportion of outpatient laminotomies significantly increased in this time period (p=.02). Outpatient laminotomies were more common among younger and healthier patients. Patients with inpatient laminotomies had significantly higher rates of surgical site infections (odds ratio [OR] 1.61, p<.001), venous thromboembolism (VTE) (OR 1.96, p<.001), hematoma (OR 1.71, p<.001), urinary tract infections (OR 1.41, p<.001), and acute kidney injuries (OR 1.75, p=.001), even when controlling for selected confounders. Conclusions: Our study demonstrated an increasing trend in the performance of laminotomy for LDH toward the outpatient setting. Even when controlling for certain confounders, patients requiring inpatient procedures had higher rates of postoperative complications. This study highlights the importance of carefully evaluating the advantages and disadvantages of performing these procedures in an outpatient versus inpatient setting.

Usefulness of Current Patient-Reported Outcome Scales for ACL Injury: A Mixed-Methods Evaluation of Stakeholder-Perceived Utility of Specific Constructs and Items Across the Rehabilitation Timeline.

Background: Numerous patient-reported outcome measures (PROMs) have been used in patients with anterior cruciate ligament reconstruction (ACLR), often with overlapping constructs of interest and limited content validity. Inefficient scale application increases burden and diminishes overall usefulness for both the patient and practitioner. Purpose: To isolate specific PROM items across a diverse set of constructs that patients and practitioners perceive as having the greatest value at various stages of recovery and return to sport (RTS) in patients after ACLR. Study Design: Cross-sectional study. Methods: A combined 77 stakeholders participated in this 2-phase mixed-methods investigation. In phase 1, a total of 27 patients and 21 practitioners selected individual PROM items from various constructs that had the greatest utility or importance. In phase 2, the highest rated items were further tested in a head-to-head comparison with 29 stakeholders who attended the 2022 ACL Injury Research Retreat. In addition to the utility assessment, practitioners answered other questions related to importance and timing of PROM assessments. Results: In phase 1, both patients and practitioners shared the same top item in 6 of the 8 (75%) constructs assessed. In phase 2, the construct of psychological burden was rated as "extremely important" by 59% of respondents, followed by physical function (54%), symptoms (35%), and donor site issues (10%). The PROM items of confidence, perceived likelihood of reinjury, and difficulty stopping quickly were rated by a respective 93%, 89%, and 86% of the sample as either "very useful" or "extremely useful." All constructs except donor site issues were rated by most stakeholders to be absolutely necessary to evaluate treatment progress and RTS readiness at the 6-month postoperative time and at RTS. Conclusion: Overall, psychological burden, with specific items related to confidence and reinjury likelihood, were rated as most important and useful by both patients and practitioners. The second most important and useful PROM item was related to higher intensity function (eg, decelerating or jumping/landing activities during sports).

Early Versus Late Periprosthetic Joint Infection After Total Knee Arthroplasty: Do Patient Differences Exist?

BACKGROUND: Periprosthetic joint infection (PJI) is a devastating complication following total knee arthroplasty (TKA). Little evidence exists comparing those with early versus late PJI. The purpose of the study was to determine comorbidity profile differences between patients developing early and late PJI. METHODS: There were 72,659 patients undergoing primary TKA from 2009 to 2021, who were identified from a commercial claims and encounters database. Subjects diagnosed with PJI were categorized as either 'early' (within 90 days of index procedure) or 'late' (> 2 years after index arthroplasty). Non-infected patients within these periods served as control groups following 4:1 propensity score matching on other extraneous variables. Logistic regression analyses were performed comparing comorbidities between groups. RESULTS: Patients were significantly younger in the late compared to the early infection group (58.1 versus 62.4 years, P < .001). When compared to those with early PJI, patients who had chronic kidney disease (13.3 versus 4.1%; OR [odds ratio] 5.17, P = .002), malignancy (20.4 versus 10.5%; OR 2.53, P = .009), uncomplicated diabetes (40.8 versus 30.6%; OR 2.00, P = .01), rheumatoid arthritis (9.2 versus 3.3%; OR 2.66, P = .046), and hypertension (88.8 versus 81.6%; OR 2.17, P = .04), were all significant predictors of developing a late PJI. CONCLUSIONS: When compared to patients diagnosed with early PJI following primary TKA, the presence of chronic kidney disease, malignancy, uncomplicated diabetes, rheumatoid arthritis, and hypertension, were independent risk factors for the development of late PJI. Younger patients who have these comorbidities may be targets for preoperative optimization interventions that minimize the risk of PJI.

Outcomes of Unicompartmental Knee Arthroplasty in Patients Receiving Long-Term Anticoagulation Therapy: A Propensity-Matched Cohort Study.

BACKGROUND: Although total knee arthroplasty has been considered the gold-standard treatment for severe osteoarthritis of the knee, unicompartmental knee arthroplasty (UKA) has become an increasingly favorable alternative for single-compartment osteoarthritis of the knee. Few studies have examined potential high-risk populations undergoing this procedure. The purpose of this study was to investigate the outcomes of UKA in patients receiving long-term anticoagulation therapy. METHODS: In this study, a large administrative database was queried to identify patients undergoing UKA between 2009 and 2019, who were then divided into a cohort receiving long-term anticoagulation and a control cohort. Propensity scores were utilized to match these patients. Multivariable logistic regression was utilized to compare 90-day and 2-year complication rates between cohorts. RESULTS: Patients who were on long-term anticoagulation had significantly increased odds of extended length of stay, surgical site infection, wound complication, transfusion, deep vein thrombosis, pulmonary embolism, and readmission at 90-day follow-up. The long-term anticoagulation cohort also experienced significantly higher odds of periprosthetic joint infection and mechanical complications at 2-year follow-up; however, odds of conversion to total knee arthroplasty were not increased. CONCLUSIONS: This study demonstrated that long-term anticoagulation use was associated with poorer medical and surgical outcomes at both 90 days and 2 years postoperatively in patients undergoing UKA, even after rigorous adjustment for confounders.

Upper Extremity Fractures in the Emergency Department: A Database Analysis of National Trends in the United States.

BACKGROUND: Upper extremity (UE) fractures are a common reason for emergency department (ED) visits, but recent data on their epidemiology are lacking. This study aimed to describe the incidence, demographics, patient characteristics, and associated health care factors of UE fractures, hypothesizing that they would remain prevalent in the ED setting. METHODS: Using the Nationwide ED Sample database, patients presenting to the ED with UE fractures in 2016 were identified, and population estimates were used to calculate incidence rates. Data on insurance status, trauma designation, cost, and teaching status were analyzed. RESULTS: The study identified 2 118 568 patients with UE fractures, representing 1.5% of all ED visits in 2016. Men accounted for 54.2% of UE fractures, with phalangeal fractures being most common. Distal radius and/or ulna fractures were most common in women (30.4%). The greatest proportion of UE fractures (23.2%) occurred in patients aged 5 to 14 years (1195.5 per 100 000). Nontrauma centers were the most common treating institutions (50.4%), followed by level I (19.5%), II (15.3%), and III (12.8%) centers. The greatest proportion of fractures (38.3%) occurred in the southern United States. Emergency department cost of treatment was almost 2-fold in patients with open UE fractures compared with closed. CONCLUSION: This study provides important epidemiological information on UE fractures in 2016. The incidence rate of UE fractures in the ED has remained high, with most occurring in the distal radius, phalanges, and clavicle. In addition, UE fractures were most common in younger patients, men, and those in the southern United States during the summer. These findings can be useful for health care providers and policymakers when evaluating and treating patients with UE fractures.

Should All Patients Receive Extended Oral Antibiotic Prophylaxis? Defining Its Role in Patients Undergoing Primary and Aseptic Revision Total Joint Arthroplasty.

BACKGROUND: Prior studies have demonstrated reduced periprosthetic joint infection (PJI) rates following extended oral antibiotics (EOAs) for high-risk patients undergoing primary total joint arthroplasty (TJA). This study compared 3-month PJI rates in all patients undergoing primary or aseptic revision TJA with or without EOA prophylaxis. METHODS: In total, 2,982 consecutive primary (n = 2,677) and aseptic revision (n = 305) TJAs were performed by a single, fellowship-trained arthroplasty surgeon from 2016 to 2022 were retrospectively reviewed. Beginning January 2020, all patients received 7 days of 300 mg oral cefdinir twice daily immediately postoperatively. Rates of PJI at 3 months were compared between patients who received or did not receive EOA. RESULTS: Rates of PJI at 3 months in patients undergoing primary and aseptic revision TJA were significantly lower in those receiving EOA prophylaxis compared to those who did not (0.41 versus 1.13%, respectively; P = .02). After primary TJA, lower PJI rates were observed with EOA prophylaxis utilization (0.23 versus 0.74%, P = .04; odds ratio [OR] 3.85). Following aseptic revision TJA, PJI rates trended toward a significant decrease with the EOA compared to without (1.88 versus 4.83%, respectively; P = .16; OR 2.71). CONCLUSIONS: All patients undergoing primary or aseptic revision TJA who received EOA prophylaxis were 3.85 and 2.71 times less likely, respectively, to develop PJI at 3 months compared to those without EOA. Future studies are needed to determine if these results are maintained at postoperative time periods beyond 3 months following primary TJA. LEVEL OF EVIDENCE: III, Retrospective review.

Chronic Anticoagulation is Associated With Increased Risk for Postoperative Complications Following Aseptic Revision Total Hip Arthroplasty.

BACKGROUND: With an aging global population, the incidence of revision total hip arthroplasty (rTHA) is expected to increase markedly. While patients undergoing primary total hip arthroplasty who require chronic anticoagulation (AC) have been associated with increased postoperative complications, less is known about the impact of chronic AC status on postoperative complications in the rTHA setting. This study sought to compare complication rates following aseptic rTHA between patients who were on chronic AC and those who were not. METHODS: A large national database was utilized to retrospectively identify 9,421 patients who underwent aseptic rTHA between 2014 and 2019. Patients were divided into 2 cohorts: 1,790 patients (19.0%) were in the chronic AC cohort (ie, having an AC prescription filled within 6 months prior to and following rTHA), and 7,631 patients (81.0%) were not on chronic AC. Postoperative complications at 90-days and 2-years were compared between cohorts utilizing univariate and multivariate analyses, controlling for sex, age, and comorbidities. RESULTS: At 90-days, chronic AC patients had increased odds of prosthetic joint infections (PJIs) (odds ratio [OR] 3.2, P < .001), surgical site infections (OR 3.6, P < .001), and mechanical prosthetic complications (OR 3.5, P < .001), which included any aseptic loosening, implant dislocation, or broken prosthetic. At 2-years, chronic AC patients had increased odds of PJI (OR 3.3, P < .001) as well as mechanical prosthetic complications (OR 3.2, P < .001). Chronic AC patients were also at increased risk for reoperation within 2 years after initial aseptic rTHA (OR 1.9, P < .001). CONCLUSIONS: Patients on chronic AC have significantly higher odds of 90-day and 2-year complications after aseptic rTHA. This includes increased odds of PJI, surgical site infection, and mechanical prosthetic complications. Patients receiving chronic AC who undergo rTHA should be counseled on the risk-benefit ratio of their chronic AC status in a multidisciplinary setting to optimize their postoperative outcomes.

Trends in Costs and Professional Reimbursements for Revision Total Hip and Knee Arthroplasty.

BACKGROUND: With increasing numbers of revision total hip and total knee arthroplasties (rTHAs and rTKAs), understanding trends in related out-of-pocket (OOP) costs, overall costs, and provider reimbursements is critical to improve patient access to care. METHODS: A large database was used to identify 92,116 patients who underwent rTHA or rTKA between 2009 and 2018. The OOP costs associated with the surgery and related inpatient care were calculated as the sum of copayment, coinsurance, and deductible payments. Professional reimbursement was calculated as total payments to the principal physician. All monetary data were adjusted to 2018 dollars. Multivariate regressions evaluated the associations between costs and procedure type, insurance type, and region of service. RESULTS: From 2009 to 2018, overall costs for rTHA significantly increased by 35.0% and overall costs for rTKA significantly increased by 32.3%. The OOP costs for rTHA had no significant changes, while OOP costs for rTKA increased by 20.1%, with patients on Medicare plans having the lowest OOP costs. Professional reimbursements, when measured as a percentage of overall costs, decreased significantly by 4.4% for rTHA and 4.0% for rTKA, with the lowest reimbursements from Medicare plans. CONCLUSION: From 2009 to 2018, total costs related to rTHA and rTKA significantly increased. The OOP costs significantly increased for rTKA, and professional reimbursements for both rTHA and rTKA decreased relative to total costs. Overall, these trends may combine to create greater financial burden to patients and the healthcare system, as well as further limit patients' access to revision arthroplasty care.

Complications Rates Among Revision Total Knee Arthroplasty Patients Diagnosed With COVID-19 Postoperatively.

BACKGROUND: The COVID-19 pandemic introduced a new set of challenges for the arthroplasty community, including the management of patients diagnosed with COVID-19 following revision total knee arthroplasty (rTKA) and its potential impact on postoperative recovery. This study sought to characterize the risks of postoperative COVID-19 infection among rTKA patients. METHODS: A large national database was utilized to query 8,022 total patients who underwent rTKA between 2018 and 2021, of which 60 had a COVID diagnosis within 90 days after surgery (rTKA/COVID positive). These patients were 1:10 propensity-score matched to 600 rTKA patients who did not have a 90-day postoperative COVID diagnosis (rTKA/COVID negative) and 600 COVID positive patients who did not undergo rTKA. Controlling for potential confounders, multivariate logistic regressions were utilized to compare 90-day postoperative complications between groups. RESULTS: Compared to rTKA/COVID negativepatients, the rTKA/COVID positive cohort had significantly higher rates of pneumonia (odds ratio [OR] = 6.1, P < .001), pulmonary embolism (PE) (OR = 32.4, P < .001), deep venous thrombosis (DVT) (OR = 32.4, P < .001), and 90-day readmissions (OR = 2.1, P = .02). Similarly, the rTKA/COVID positive cohort had significantly higher rates of pneumonia (OR = 4.3, P = .001), PE (OR = 36.8, P < .001), and DVT (OR = 36.8, P < .001) compared to COVID positive patients who did not undergo rTKA. CONCLUSIONS: Revision total knee arthroplasty patients diagnosed with COVID-19 postoperatively had increased rates of thromboembolic events, pneumoniae, and 90-day readmissions. Risk mitigation efforts would suggest extending the prophylactic anticoagulation period for rTKA patients diagnosed with postoperative COVID-19.

Periprosthetic Joint Infection Risk After Primary Total Knee Arthroplasty: Are All Preoperative Corticosteroid Injections the Same?

BACKGROUND: Previous evidence has demonstrated an increased risk of periprosthetic joint infection (PJI) following primary total knee arthroplasty (TKA) in patients receiving corticosteroid injection (CSI) within 3 months of surgery. The study aimed to determine if PJI risk after TKA varied among different corticosteroid agents. METHODS: A total of 85,073 patients undergoing primary TKA from 2009 to 2019 were identified from a large national database. Of these, 1,092 (1.3%) received an ipsilateral, intra-articular CSI within 90 days of TKA. These patients were compared to those not receiving CSI using multivariate logistic regressions following 1:4 propensity score matching, with PJI development as the primary outcome. RESULTS: Patients given an injection of any corticosteroid within 90 days of TKA had significantly higher PJI rates compared to controls (1.6 versus 0.41%; P < .001). This finding was driven by patients receiving methylprednisolone acetate (n = 543) or betamethasone (n = 153), with prevalence rates of 1.7 and 2.6%, respectively (P = .003 and P = .01, respectively). No significant increase in the rate of PJI was observed for patients receiving triamcinolone (1.2%; P = .08; n = 342) or dexamethasone (0.0%; P = 1; n = 54) within 90 days preceding TKA. PJI risk for all agents, administered more than 90 days preoperatively normalized to control levels (0.51 versus 0.34%). CONCLUSIONS: These results suggest that PJI risk varies with CSI type. In this large database study, only patients given methylprednisolone acetate or betamethasone injections within 90 days of surgery had significantly higher PJI rates compared to controls.