Predicting prolonged length of stay following revision total knee arthroplasty: A national database analysis using machine learning models.

BACKGROUND: As the number of revision total knee arthroplasty (TKA) continues to rise, close attention has been paid to factors influencing postoperative length of stay (LOS). The aim of this study is to develop generalizable machine learning (ML) algorithms to predict extended LOS following revision TKA using data from a national database. METHODS: 23,656 patients undergoing revision TKA between 2013 and 2020 were identified using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database. Patients with missing data and those undergoing re-revision or conversion from unicompartmental knee arthroplasty were excluded. Four ML algorithms were applied and evaluated based on their (1) ability to distinguish between at-risk and not-at-risk patients, (2) accuracy, (3) calibration, and (4) clinical utility. RESULTS: All four ML predictive algorithms demonstrated good accuracy, calibration, clinical utility, and discrimination, with all models achieving a similar area under the curve (AUC) (AUCLR=AUCRF=AUCHGB=0.75, AUCANN=0.74). The most important predictors of prolonged LOS were found to be operative time, preoperative diagnosis of sepsis, and body mass index (BMI). CONCLUSIONS: ML models developed in this study demonstrated good performance in predicting extended LOS in patients undergoing revision TKA. Our findings highlight the importance of utilizing nationally representative patient data for model development. Prolonged operative time, preoperative sepsis, BMI, and elevated preoperative serum creatinine and BUN were noted to be significant predictors of prolonged LOS. Knowledge of these associations may aid with patient-specific preoperative planning, discharge planning, patient counseling, and cost containment with revision TKA.

Medical Malpractice Litigation Trends Following Primary Total Hip and Knee Joint Arthroplasty: An Updated Nationwide Analysis.

BACKGROUND: Total joint arthroplasty (TJA) is the most common procedure associated with malpractice claims within orthopaedic surgery. Although prior research has assessed prevalent causes and outcomes of TJA-related lawsuits before 2018, the dynamic healthcare environment demands regular re-evaluations. This study aimed to provide an updated analysis of the predominant causes and outcomes of TJA-related malpractice lawsuits and analyze the outcomes of subsequent appeals following initial jury verdicts. METHODS: A legal database was queried for cases between 2018 and 2022 involving primary hip and knee TJA in the United States. Cases were listed as original rulings or appeals and reviewed for the alleged negligence, damages incurred, demographics, and verdicts. Appeals were further assessed for appellant details, preliminary judgment, and outcomes. The findings were compared to previous litigation data using descriptive statistics. RESULTS: The final cohort comprised 59 cases: 33 (56%) total knee arthroplasty (TKA) and 26 (44%) total hip arthroplasty (THA) from 2018 to 2022. The TKA cases primarily cited pain (24%), while the THA cases cited nerve injuries (31%). Negligence largely stemmed from procedural error (47%), postsurgical error (27%), and failure to inform (14%). Case outcomes were in favor of the defense in 66% of cases. Overall, 90% of primary verdicts led to appeals, with 71% by the plaintiff. Initial rulings were upheld in 87% of plaintiff appeals, whereas only 53% of defendant appeals retained the initial judgment. CONCLUSION: The primary causative factor of litigation shifted from infection to ongoing/worsening pain postoperatively in TKA cases over time. While nerve injury TKA cases have decreased, it remains the most cited damage in THA cases. Defense verdicts are common, but there is an increasing number of verdicts against defendants. Plaintiffs are more likely to appeal, but are less successful in appellate courts. These findings allow surgeons and policymakers to address emerging litigation trends in TJA to enhance patient care, mitigate risks, and improve the overall quality of TJA.

Reactive Oxygen Species Mediate the Bactericidal Activity of Chlorine Against Salmonella.

Chlorine and its derivatives have been used as an antibacterial agent to reduce Salmonella contamination in poultry meat during processing. We evaluated the survival of 4 different Salmonella serotypes (Typhimurium, Enteritidis, Heidelberg, and Gaminara) in the presence of 50 ppm sodium hypochlorite (NaOCl) alone or with the addition of thiourea (radical scavenger) or Dip (iron chelator) to determine the contribution of reactive oxygen species (ROS) in the bactericidal activity of NaOCl. The result showed that for all four serotypes the addition of thiourea or Dip significantly increased the % survival as compared to the respective NaOCl treatment groups, while it was significantly higher with thiourea as compared to Dip (P < 0.05). We also evaluated the survival of 11 deletion mutants of S. Typhimurium, which were demonstrated to increase (∆atpC, ∆cyoA, ∆gnd, ∆nuoG, ∆pta, ∆sdhC, and ∆zwf) or decrease the production of ROS (∆edd, ∆fumB, ∆pykA, and ∆tktB) in Escherichia coli (E. coli), in the presence of 50 ppm. The results showed that only two (∆sdhC and ∆zwf) out of 7 ROS-increasing mutants showed reduced % survival as compared to the wild-type (P < 0.05), while all four deletion ROS-decreasing mutants showed significantly higher % survival as compared to the wild-type (P < 0.05). This work suggests that the production of ROS is a major component of the bactericidal activity of NaOCl against Salmonella serotypes and there might be a significant difference in the metabolic pathways involved in ROS production between Salmonella and E. coli.

Predicting extended hospital stay following revision total hip arthroplasty: a machine learning model analysis based on the ACS-NSQIP database.

INTRODUCTION: Prolonged length of stay (LOS) following revision total hip arthroplasty (THA) can lead to increased healthcare costs, higher rates of readmission, and lower patient satisfaction. In this study, we investigated the predictive power of machine learning (ML) models for prolonged LOS after revision THA using patient data from a national-scale patient repository. MATERIALS AND METHODS: We identified 11,737 revision THA cases from the American College of Surgeons National Surgical Quality Improvement Program database from 2013 to 2020. Prolonged LOS was defined as exceeding the 75th value of all LOSs in the study cohort. We developed four ML models: artificial neural network (ANN), random forest, histogram-based gradient boosting, and k-nearest neighbor, to predict prolonged LOS after revision THA. Each model's performance was assessed during training and testing sessions in terms of discrimination, calibration, and clinical utility. RESULTS: The ANN model was the most accurate with an AUC of 0.82, calibration slope of 0.90, calibration intercept of 0.02, and Brier score of 0.140 during testing, indicating the model's competency in distinguishing patients subject to prolonged LOS with minimal prediction error. All models showed clinical utility by producing net benefits in the decision curve analyses. The most significant predictors of prolonged LOS were preoperative blood tests (hematocrit, platelet count, and leukocyte count), preoperative transfusion, operation time, indications for revision THA (infection), and age. CONCLUSIONS: Our study demonstrated that the ML model accurately predicted prolonged LOS after revision THA. The results highlighted the importance of the indications for revision surgery in determining the risk of prolonged LOS. With the model's aid, clinicians can stratify individual patients based on key factors, improve care coordination and discharge planning for those at risk of prolonged LOS, and increase cost efficiency.

Socioeconomic area deprivation index is not associated with postoperative complications following revision total hip and knee joint arthroplasty.

Introduction: Revision hip and knee total joint arthroplasty (TJA) carries a high burden of postoperative complications, including surgical site infections (SSI), venous thromboembolism (VTE), reoperation, and readmission, which negatively affect postoperative outcomes and patient satisfaction. Socioeconomic area-level composite indices such as the area deprivation index (ADI) are increasingly important measures of social determinants of health (SDoH). This study aims to determine the potential association between ADI and SSI, VTE, reoperation, and readmission occurrence 90 days following revision TJA. Methods: 1047 consecutive revision TJA patients were retrospectively reviewed. Complications, including SSI, VTE, reoperation, and readmission, were combined into one dependent variable. ADI rankings were extracted using residential zip codes and categorized into quartiles. Univariate and multivariate logistic regressions were performed to analyze the association of ADI as an independent factor for complication following revision TJA. Results: Depression (p = 0.034) and high ASA score (p < 0.001) were associated with higher odds of a combined complication postoperatively on univariate logistic regression. ADI was not associated with the occurrence of any of the complications recorded following surgery (p = 0.092). ASA remained an independent risk factor for developing postoperative complications on multivariate analysis. Conclusion: An ASA score of 3 or higher was significantly associated with higher odds of developing postoperative complications. Our findings suggest that ADI alone may not be a sufficient tool for predicting postoperative outcomes following revision TJA, and other area-level indices should be further investigated as potential markers of social determinants of health.

Are social determinants of health associated with an increased length of hospitalization after revision total hip and knee arthroplasty? A comparison study of social deprivation indices.

INTRODUCTION: Length of stay (LOS) has been extensively assessed as a marker for healthcare utilization, functional outcomes, and cost of care for patients undergoing arthroplasty. The notable patient-to-patient variation in LOS following revision hip and knee total joint arthroplasty (TJA) suggests a potential opportunity to reduce preventable discharge delays. Previous studies investigated the impact of social determinants of health (SDoH) on orthopaedic conditions and outcomes using deprivation indices with inconsistent findings. The aim of the study is to compare the association of three publicly available national indices of social deprivation with prolonged LOS in revision TJA patients. MATERIALS AND METHODS: 1,047 consecutive patients who underwent a revision TJA were included in this retrospective study. Patient demographics, comorbidities, and behavioral characteristics were extracted. Area deprivation index (ADI), social deprivation index (SDI), and social vulnerability index (SVI) were recorded for each patient, following which univariate and multivariate logistic regression analyses were performed to determine the relationship between deprivation measures and prolonged LOS (greater than five days postoperatively). RESULTS: 193 patients had a prolonged LOS following surgery. Categorical ADI was significantly associated with prolonged LOS following surgery (OR = 2.14; 95% CI = 1.30-3.54; p = 0.003). No association with LOS was found using SDI and SVI. When accounting for other covariates, only ASA scores (ORrange=3.43-3.45; p < 0.001) and age (ORrange=1.00-1.03; prange=0.025-0.049) were independently associated with prolonged LOS. CONCLUSION: The varying relationship observed between the length of stay and socioeconomic markers in this study indicates that the selection of a deprivation index could significantly impact the outcomes when investigating the association between socioeconomic deprivation and clinical outcomes. These results suggest that ADI is a potential metric of social determinants of health that is applicable both clinically and in future policies related to hospital stays including bundled payment plan following revision TJA.

Prediction of 30-Day Mortality Following Revision Total Hip and Knee Arthroplasty: Machine Learning Algorithms Outperform CARDE-B, 5-Item, and 6-Item Modified Frailty Index Risk Scores.

BACKGROUND: Although risk calculators are used to prognosticate postoperative outcomes following revision total hip and knee arthroplasty (total joint arthroplasty [TJA]), machine learning (ML) based predictive tools have emerged as a promising alternative for improved risk stratification. This study aimed to compare the predictive ability of ML models for 30-day mortality following revision TJA to that of traditional risk-assessment indices such as the CARDE-B score (congestive heart failure, albumin (< 3.5 mg/dL), renal failure on dialysis, dependence for daily living, elderly (> 65 years of age), and body mass index (BMI) of < 25 kg/m2), 5-item modified frailty index (5MFI), and 6MFI. METHODS: Adult patients undergoing revision TJA between 2013 and 2020 were selected from the American College of Surgeons National Surgical Quality Improvement Program database and randomly split 80:20 to compose the training and validation cohorts. There were 3 ML models - extreme gradient boosting, random forest, and elastic-net penalized logistic regression (NEPLR) - that were developed and evaluated using discrimination, calibration metrics, and accuracy. The discrimination of CARDE-B, 5MFI, and 6MFI scores was assessed individually and compared to that of ML models. RESULTS: All models were equally accurate (Brier score = 0.005) and demonstrated outstanding discrimination with similar areas under the receiver operating characteristic curve (AUCs, extreme gradient boosting = 0.94, random forest = NEPLR = 0.93). The NEPLR was the best-calibrated model overall (slope = 0.54, intercept = -0.004). The CARDE-B had the highest discrimination among the scores (AUC = 0.89), followed by 6MFI (AUC = 0.80), and 5MFI (AUC = 0.68). Albumin < 3.5 mg/dL and BMI (< 30.15) were the most important predictors of 30-day mortality following revision TJA. CONCLUSIONS: The ML models outperform traditional risk-assessment indices in predicting postoperative 30-day mortality after revision TJA. Our findings highlight the utility of ML for risk stratification in a clinical setting. The identification of hypoalbuminemia and BMI as prognostic markers may allow patient-specific perioperative optimization strategies to improve outcomes following revision TJA.

Neighborhood socioeconomic disadvantages associated with prolonged length of stay and non-home discharge following revision total hip and knee joint arthroplasty.

Background: Discharge disposition and length of stay (LOS) are widely recognized markers of healthcare utilization patterns of total hip and knee joint arthroplasty (TJA). These markers are commonly associated with increased postoperative complications, patient dissatisfaction, and higher costs. Area deprivation index (ADI) has been validated as a composite metric of neighborhood-level disadvantage. This study aims to determine the potential association between ADI and discharge disposition or extended LOS following revision TJA. Methods: This study conducted a retrospective analysis of a consecutive series of revision hip and knee TJA patients from a single tertiary institution. Univariate and multivariate regression analysis was used to determine the association between ADI and discharge disposition or LOS, adjusting for patient demographics and comorbidities. Results: 1047 consecutive revision TJA patients were identified across 463 different neighborhoods. 193 (18.4 %) had an extended LOS, and 334 (31.9 %) were discharged to non-home facilities. Compared with Q1 (least deprived cohort), Q2 (odds ratio [OR] = 1.63; p = 0.030) and Q4 (most deprived cohort: OR = 2.04; p = 0.002) cohorts demonstrated higher odds of non-home discharge. Patients in the highest ADI quartile (most deprived cohort) were associated with increased odds of prolonged LOS following revision TJA compared to those in the lowest ADI quartile (OR = 2.63; p < 0.001). Conclusion: This study suggests that higher levels of neighborhood-level disadvantage may be associated with higher odds of non-home discharge and prolonged LOS following revision TJA. Development of interventions based on the area deprivation index may improve discharge planning and reduce unnecessary non-home discharges in patients living in areas of socioeconomic deprivation.

Generalizability of machine learning models predicting 30-day unplanned readmission after primary total knee arthroplasty using a nationally representative database.

Unplanned readmission after primary total knee arthroplasty (TKA) costs an average of US $39,000 per episode and negatively impacts patient outcomes. Although predictive machine learning (ML) models show promise for risk stratification in specific populations, existing studies do not address model generalizability. This study aimed to establish the generalizability of previous institutionally developed ML models to predict 30-day readmission following primary TKA using a national database. Data from 424,354 patients from the ACS-NSQIP database was used to develop and validate four ML models to predict 30-day readmission risk after primary TKA. Individual model performance was assessed and compared based on discrimination, accuracy, calibration, and clinical utility. Length of stay (> 2.5 days), body mass index (BMI) (> 33.21 kg/m2), and operation time (> 93 min) were important determinants of 30-day readmission. All ML models demonstrated equally good accuracy, calibration, and discriminatory ability (Brier score, ANN = RF = HGB = NEPLR = 0.03; ANN, slope = 0.90, intercept = - 0.11; RF, slope = 0.93, intercept = - 0.12; HGB, slope = 0.90, intercept = - 0.12; NEPLR, slope = 0.77, intercept = 0.01; AUCANN = AUCRF = AUCHGB = AUCNEPLR = 0.78). This study validates the generalizability of four previously developed ML algorithms in predicting readmission risk in patients undergoing TKA and offers surgeons an opportunity to reduce readmissions by optimizing discharge planning, BMI, and surgical efficiency.

Development of real-time PCR assay for quantitative detection of Clostridium septicum.

Cellulitis is an important disease in commercial turkey farms associated with significant economic loss. Although the etiology of cellulitis is not fully elucidated, Clostridium septicum (C. septicum) is one of the main causes of this infectious disease. In this study, we report the development of a quantitative real-time PCR (qRT PCR) assay targeting the alpha-toxin gene (csa), which involves a prior 15-cyle PCR using a nested pair of primers to increase the detection sensitivity. Additionally, the TaqMan probe was employed to increase the target-specificity of the assay. The performance of our nested qRT-PCR assay was evaluated using Clostridium isolates from turkey farms, representing both septicum and non-septicum species, as well as sponge swab samples from turkey farms. Our step-by-step development of the assay showed that the csa gene is a suitable target for specific detection of C. septicum strains and that the inclusion of nested PCR step significantly increased the detection sensitivity of the final qRT PCR assay. The performance of the assay was also validated by a high correlation of the threshold cycle numbers of the qRT PCR assay with the relative abundance of C. septicum read counts in 16S rRNA gene microbiota profiles of the C. septicum-containing samples from turkey farms.

Predicting 30-day unplanned hospital readmission after revision total knee arthroplasty: machine learning model analysis of a national patient cohort.

Revision total knee arthroplasty (TKA) is associated with a higher risk of readmission than primary TKA. Identifying individual patients predisposed to readmission can facilitate proactive optimization and increase care efficiency. This study developed machine learning (ML) models to predict unplanned readmission following revision TKA using a national-scale patient dataset. A total of 17,443 revision TKA cases (2013-2020) were acquired from the ACS NSQIP database. Four ML models (artificial neural networks, random forest, histogram-based gradient boosting, and k-nearest neighbor) were developed on relevant patient variables to predict readmission following revision TKA. The length of stay, operation time, body mass index (BMI), and laboratory test results were the strongest predictors of readmission. Histogram-based gradient boosting was the best performer in distinguishing readmission (AUC: 0.95) and estimating the readmission probability for individual patients (calibration slope: 1.13; calibration intercept: -0.00; Brier score: 0.064). All models produced higher net benefit than the default strategies of treating all or no patients, supporting the clinical utility of the models. ML demonstrated excellent performance for the prediction of readmission following revision TKA. Optimization of important predictors highlighted by our model may decrease preventable hospital readmission following surgery, thereby leading to reduced financial burden and improved patient satisfaction.

Do preoperative PROMIS scores independently predict 90-day readmission following primary total knee arthroplasty?

INTRODUCTION: The rising demand for total knee arthroplasty (TKA) is expected to increase the total number of TKA-related readmissions, presenting significant public health and economic burden. With the increasing use of Patient-Reported Outcomes Measurement Information System (PROMIS) scores to inform clinical decision-making, this study aimed to investigate whether preoperative PROMIS scores are predictive of 90-day readmissions following primary TKA. MATERIALS AND METHODS: We retrospectively reviewed a consecutive series of 10,196 patients with preoperative PROMIS scores who underwent primary TKA. Two comparison groups, readmissions (n = 79; 3.6%) and non-readmissions (n = 2091; 96.4%) were established. Univariate and multivariate logistic regression analyses were then performed with readmission as the outcome variable to determine whether preoperative PROMIS scores could predict 90-day readmission. RESULTS: The study cohort consisted of 2170 patients overall. Non-white patients (OR = 3.53, 95% CI [1.16, 10.71], p = 0.026) and patients with cardiovascular or cerebrovascular disease (CVD) (OR = 1.66, 95% CI [1.01, 2.71], p = 0.042) were found to have significantly higher odds of 90-day readmission after TKA. Preoperative PROMIS-PF10a (p = 0.25), PROMIS-GPH (p = 0.38), and PROMIS-GMH (p = 0.07) scores were not significantly associated with 90-day readmission. CONCLUSION: This study demonstrates that preoperative PROMIS scores may not be used to predict 90-day readmission following primary TKA. Non-white patients and patients with CVD are 3.53 and 1.66 times more likely to be readmitted, highlighting existing racial disparities and medical comorbidities contributing to readmission in patients undergoing TKA.

[ 11 C]-PBR28 positron emission tomography signal as an imaging marker of joint inflammation in knee osteoarthritis.

ABSTRACT: Although inflammation is known to play a role in knee osteoarthritis (KOA), inflammation-specific imaging is not routinely performed. In this article, we evaluate the role of joint inflammation, measured using [ 11 C]-PBR28, a radioligand for the inflammatory marker 18-kDa translocator protein (TSPO), in KOA. Twenty-one KOA patients and 11 healthy controls (HC) underwent positron emission tomography/magnetic resonance imaging (PET/MRI) knee imaging with the TSPO ligand [ 11 C]-PBR28. Standardized uptake values were extracted from regions-of-interest (ROIs) semiautomatically segmented from MRI data, and compared across groups (HC, KOA) and subgroups (unilateral/bilateral KOA symptoms), across knees (most vs least painful), and against clinical variables (eg, pain and Kellgren-Lawrence [KL] grades). Overall, KOA patients demonstrated elevated [ 11 C]-PBR28 binding across all knee ROIs, compared with HC (all P 's < 0.005). Specifically, PET signal was significantly elevated in both knees in patients with bilateral KOA symptoms (both P 's < 0.01), and in the symptomatic knee ( P < 0.05), but not the asymptomatic knee ( P = 0.95) of patients with unilateral KOA symptoms. Positron emission tomography signal was higher in the most vs least painful knee ( P < 0.001), and the difference in pain ratings across knees was proportional to the difference in PET signal ( r = 0.74, P < 0.001). Kellgren-Lawrence grades neither correlated with PET signal (left knee r = 0.32, P = 0.19; right knee r = 0.18, P = 0.45) nor pain ( r = 0.39, P = 0.07). The current results support further exploration of [ 11 C]-PBR28 PET signal as an imaging marker candidate for KOA and a link between joint inflammation and osteoarthritis-related pain severity.

The Use of Artificial Intelligence for the Prediction of Periprosthetic Joint Infection Following Aseptic Revision Total Knee Arthroplasty.

Periprosthetic joint infection (PJI) following revision total knee arthroplasty (TKA) for aseptic failure is associated with poor outcomes, patient morbidity, and high health care expenditures. The aim of this study was to develop novel machine learning algorithms for the prediction of PJI following revision TKA for patients with aseptic indications for revision surgery. A single-institution database consisting of 1,432 consecutive revision TKA patients with aseptic etiologies was retrospectively identified. The patient cohort included 208 patients (14.5%) who underwent re-revision surgery for PJI. Three machine learning algorithms (artificial neural networks, support vector machines, k-nearest neighbors) were developed to predict this outcome and these models were assessed by discrimination, calibration, and decision curve analysis. This is a retrospective study. Among the three machine learning models, the neural network model achieved the best performance across discrimination (area under the receiver operating characteristic curve = 0.78), calibration, and decision curve analysis. The strongest predictors for PJI following revision TKA for aseptic reasons were prior open procedure prior to revision surgery, drug abuse, obesity, and diabetes. This study utilized machine learning as a tool for the prediction of PJI following revision TKA for aseptic failure with excellent performance. The validated machine learning models can aid surgeons in patient-specific risk stratifying to assist in preoperative counseling and clinical decision making for patients undergoing aseptic revision TKA.

Patient-Reported Outcomes Measurement Information System Physical Function and Global Physical Health Subscale Strongly Correlate and Perform Similarly to the Knee Injury and Osteoarthritis Outcome Score-Physical Function Short Form.

BACKGROUND: Arthroplasty surgeons use a variety of patient-reported outcome measures (PROMs) to assess functional well-being, including the Knee Injury and Osteoarthritis Outcome Score (KOOS) Physical Function short form (KOOS-PS), Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function Short Form 10a (PROMIS PF SF 10a), and PROMIS Global-10 Physical Health subscale. However, there is a paucity of literature assessing their concurrent validity and performance. METHODS: Between June 2016 and December 2020, patient visits at an arthroplasty clinic for knee concerns were identified. Patients who completed KOOS-PS, PROMIS PF SF 10a, and PROMIS Global-10, including its physical and mental health subscales, at the same visit were identified. Spearman rho (ρ) correlations were calculated and ceiling and floor effects identified. Overall, 5,303 patient encounters were included. RESULTS: Among physical function domains, strong correlation existed between the KOOS-PS and PROMIS PF SF 10a (ρ = 0.76, P < .001), KOOS-PS and PROMIS Global Physical Health (ρ = 0.71, P < .001), and PROMIS PF SF 10a and PROMIS Global Physical Health (ρ = 0.78, P < .001). No physical function-focused PROM had an appreciable floor effect (ie, at or more than 1%). The KOOS-PS had a small but measurable ceiling effect (n = 105 [2.0%]). CONCLUSIONS: All of the examined PROMs are acceptable to measure the functional status of patients with knee pathology, with the PROMIS Global-10 also being able to capture elements of mental health too. The PROMIS Global-10 may be of most value of the PROMs assessed, as the United States Centers for Medicare and Medicaid Services already incorporate the mental health component into new alternative payment models.

Can machine learning models predict prolonged length of hospital stay following primary total knee arthroplasty based on a national patient cohort data?

INTRODUCTION: The total length of stay (LOS) is one of the biggest determinators of overall care costs associated with total knee arthroplasty (TKA). An accurate prediction of LOS could aid in optimizing discharge strategy for patients in need and diminishing healthcare expenditure. The aim of this study was to predict LOS following TKA using machine learning models developed on a national-scale patient cohort. METHODS: The ACS-NSQIP database was queried to acquire 267,966 TKA cases from 2013 to 2020. Four machine learning models-artificial neural network (ANN), random forest, histogram-based gradient boosting, and k-nearest neighbor were trained and tested on the dataset for the prediction of prolonged LOS (LOS exceeded the 75th of all values in the cohort). The model performance was assessed by discrimination (area under the receiver operating characteristic curve [AUC]), calibration, and clinical utility. RESULTS: ANN delivered the best performance among the four models. ANN distinguished prolonged LOS in the study cohort with an AUC of 0.71 and accurately predicted the probability of prolonged LOS for individual patients (calibration slope: 0.82; calibration intercept: 0.03; Brier score: 0.089). All models demonstrated clinical utility by generating positive net benefits in decision curve analyses. Operation time, pre-operative transfusion, pre-operative laboratory tests (hematocrit, platelet count, and white blood cell count), and BMI were the strongest predictors of prolonged LOS. CONCLUSION: ANN demonstrated modest discrimination capacity and excellent performance in calibration and clinical utility for the prediction of prolonged LOS following TKA. Clinical application of the machine learning models has the potential to improve care coordination and discharge planning for patients at high risk of extended hospitalization after surgery. Incorporating more relevant patient factors may further increase the models' prediction strength.

Trends and Risk Factors for Surgical Site Infection after Treatment of the Ankle Fracture: National Cohort Study.

Surgical site infection (SSI) is a major complication after the surgical treatment of ankle fractures that can result in catastrophic consequences. This study aimed to determine the incidence of SSI in several cohorts from national insurance databases over the past 12 years and identify its predictors. The claimed data for patients (n = 1,449,692) with ankle fractures between 2007 and 2019 were investigated, and a total of 41,071 patients were included in the final analysis. The covariates included were age, sex, season, fracture type (closed vs. open), type of surgical fixation procedure, and comorbidities of each patient. All subjects were divided into two groups according to the SSI after the surgical fixation of the ankle fracture (no infection group vs. infection group). The number of SSIs after the surgical treatment of ankle fractures was 874 (2.13%). Open fractures [odds ratio, (OR) = 4.220] showed the highest risk for SSI, followed by the male sex (OR = 1.841), an increasing number of comorbidities (3-5, OR = 1.484; ≥6, OR = 1.730), a history of dementia (OR = 1.720) or of myocardial infarction (OR = 1.628), and increasing age (OR = 1.010). The summer season (OR = 1.349) showed the highest risk among the four seasons for SSI after ankle fracture surgery.

Validation of Machine Learning Model Performance in Predicting Blood Transfusion After Primary and Revision Total Hip Arthroplasty.

BACKGROUND: The rates of blood transfusion following primary and revision total hip arthroplasty (THA) remain as high as 9% and 18%, respectively, contributing to patient morbidity and healthcare costs. Existing predictive tools are limited to specific populations, thereby diminishing their clinical applicability. This study aimed to externally validate our previous institutionally developed machine learning (ML) algorithms to predict the risk of postoperative blood transfusion following primary and revision THA using national inpatient data. METHODS: Five ML algorithms were trained and validated using data from 101,266 primary THA and 8,594 revision THA patients from a large national database to predict postoperative transfusion risk after primary and revision THA. Models were assessed and compared based on discrimination, calibration, and decision curve analysis. RESULTS: The most important predictors of transfusion following primary and revision THA were preoperative hematocrit (<39.4%) and operation time (>157 minutes), respectively. All ML models demonstrated excellent discrimination (area under the curve (AUC) >0.8) in primary and revision THA patients, with artificial neural network (AUC = 0.84, slope = 1.11, intercept = -0.04, Brier score = 0.04), and elastic-net-penalized logistic regression (AUC = 0.85, slope = 1.08, intercept = -0.01, and Brier score = 0.12) performing best, respectively. On decision curve analysis, all 5 models demonstrated a higher net benefit than the conventional strategy of intervening for all or no patients in both patient cohorts. CONCLUSIONS: This study successfully validated our previous institutionally developed ML algorithms for the prediction of blood transfusion following primary and revision THA. Our findings highlight the potential generalizability of predictive ML tools developed using nationally representative data in THA patients.

Machine Learning Models Based on a National-Scale Cohort Identify Patients at High Risk for Prolonged Lengths of Stay Following Primary Total Hip Arthroplasty.

BACKGROUND: Existing machine learning models that predicted prolonged lengths of stay (LOS) following primary total hip arthroplasty (THA) were limited by the small training volume and exclusion of important patient factors. This study aimed to develop machine learning models using a national-scale data set and examine their performance in predicting prolonged LOS following THA. METHODS: A total of 246,265 THAs were analyzed from a large database. Prolonged LOS was defined as exceeding the 75th percentile of all LOSs in the cohort. Candidate predictors of prolonged LOS were selected by recursive feature elimination and used to construct four machine learning models-artificial neural network, random forest, histogram-based gradient boosting, and k-nearest neighbor. The model performance was assessed by discrimination, calibration, and utility. RESULTS: All models exhibited excellent performance in discrimination (area under the receiver operating characteristic curve [AUC] = 0.72 to 0.74) and calibration (slope: 0.83 to 1.18, intercept: -0.01 to 0.11, Brier score: 0.185 to 0.192) during both training and testing sessions. The artificial neural network was the best performer with an AUC of 0.73, calibration slope of 0.99, calibration intercept of -0.01, and Brier score of 0.185. All models showed great utility by producing higher net benefits than the default treatment strategies in the decision curve analyses. Age, laboratory tests, and surgical variables were the strongest predictors of prolonged LOS. CONCLUSION: The excellent prediction performance of machine learning models demonstrated their capacity to identify patients prone to prolonged LOS. Many factors contributing to prolonged LOS can be optimized to minimize hospital stay for high-risk patients.

Effects of prior antibiotic use on clarithromycin resistance in Helicobacter pylori.

BACKGROUND: Macrolide antibiotics are widely used to treat various infections such as pneumonia and sinusitis, and previous exposure to macrolides is presumed to be a risk factor for standard triple therapy failure in Helicobacter pylori (H. pylori) eradication. We aimed to determine whether previous use of macrolide antibiotics could affect clarithromycin resistance of H. pylori. MATERIALS AND METHODS: From the Korea National Health Insurance Service (NHIS2021-1-775) database, a total of 46,160 patients who were tested for clarithromycin resistance of H. pylori from 2016 to 2019 in Korea were identified. Their history of antibiotics in the past 10 years and history of respiratory comorbidity in the past 1 year were investigated. RESULTS: Clarithromycin resistance rate of H. pylori in Korea was 16.2%. A multivariate analysis revealed that female sex (OR: 1.472, p < .001), age > 50 years (OR: 1.340, p < .001), previous use of macrolide antibiotics (clarithromycin, OR: 2.902, p < .001; azithromycin, OR: 1.930, p < .001; erythromycin, OR: 2.060, p = .001; roxithromycin, OR: 2.022, p < .001), and history of respiratory comorbidity (sinusitis, OR: 1.271, p < .001; laryngopharyngitis, OR: 1.135, p = .032; bronchitis, OR: 1.245, p = .001; pneumonia, OR: 1.335, p = .026) were independent risk factors of clarithromycin resistance in H. pylori. CONCLUSIONS: The use of macrolide antibiotics and a recent diagnosis of respiratory disease might increase clarithromycin resistance of H. pylori.