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.

Validation and Generalizability of Machine Learning Models for the Prediction of Discharge Disposition Following Revision Total Knee Arthroplasty.

BACKGROUND: Postoperative discharge to facilities account for over 33% of the $ 2.7 billion revision total knee arthroplasty (TKA)-associated annual expenditures and are associated with increased complications when compared to home discharges. Prior studies predicting discharge disposition using advanced machine learning (ML) have been limited due to a lack of generalizability and validation. This study aimed to establish ML model generalizability by externally validating its prediction for nonhome discharge following revision TKA using national and institutional databases. METHODS: The national and institutional cohorts comprised 52,533 and 1,628 patients, respectively, with 20.6 and 19.4% nonhome discharge rates. Five ML models were trained and internally validated (five-fold cross-validation) on a large national dataset. Subsequently, external validation was performed on our institutional dataset. Model performance was assessed using discrimination, calibration, and clinical utility. Global predictor importance plots and local surrogate models were used for interpretation. RESULTS: The strongest predictors of nonhome discharge were patient age, body mass index, and surgical indication. The area under the receiver operating characteristic curve increased from internal to external validation and ranged between 0.77 and 0.79. Artificial neural network was the best predictive model for identifying patients at risk for nonhome discharge (area under the receiver operating characteristic curve = 0.78), and also the most accurate (calibration slope = 0.93, intercept = 0.02, and Brier score = 0.12). CONCLUSION: All five ML models demonstrated good-to-excellent discrimination, calibration, and clinical utility on external validation, with artificial neural network being the best model for predicting discharge disposition following revision TKA. Our findings establish the generalizability of ML models developed using data from a national database. The integration of these predictive models into clinical workflow may assist in optimizing discharge planning, bed management, and cost containment associated with revision TKA.

Immediate and short-term biomechanical adaptation of habitual barefoot runners who start shod running.

This study investigated the immediate and short-term effects of minimalist shoes (MS) and traditional running shoes (TRS) on vertical loading rates, foot strike pattern and lower limb kinematics in a group of habitual barefoot runners. Twelve habitual barefoot runners were randomly given a pair of MS or TRS and were asked to run with the prescribed shoes for 1 month. Outcome variables were obtained before, immediate after and 1 month after shoe prescription. Average and instantaneous vertical loading rates at the 1-month follow-up were significantly higher than that at the pre-shod session (P < 0.034, η2p > 0.474). Foot strike angle in the TRS group was significantly lower than that in the MS group (P = 0.045, η2p = 0.585). However, there was no significant time nor shoe effect on overstride, knee and ankle excursion (P > 0.061). Habitual barefoot runners appeared to land with a greater impact during shod running and they tended to have a more rearfoot strike pattern while wearing TRS. Lower limb kinematics were comparable before and after shoe prescription. Longer period of follow-up is suggested to further investigate the footwear effect on the running biomechanics in habitual barefoot runners.

A new footwear technology to promote non-heelstrike landing and enhance running performance: Fact or fad?

This study sought to compare the kinetics and kinematics data in a group of habitual shod runners when running in traditional running shoes and newly designed minimalist shoes with lug platform. This novel footwear design claims to simulate barefoot running and reduce energy loss during impact. We compared footstrike angle (FSA), vertical average (VALR) and instantaneous (VILR) loading rates, energy loss and initial vertical stiffness between two shoe conditions. Runners demonstrated a decreased FSA while running in minimalist shoes with lug platform than traditional shoes (P = 0.003; Cohen's d = 0.918). However, we did not observe a landing pattern transition. VALR and VILR between two footwear conditions showed no significant difference (P = 0.191-0.258; Cohen's d = 0.304-0.460). Initial vertical stiffness (P = 0.032; Cohen's d = 0.671) and energy loss (P = 0.044; Cohen's d = 0.578) were greater when running in minimalist shoes with lug platform. The results show that minimalist shoes with lug platform reduce the FSA but may not lead to a landing pattern switch or lower vertical loading rates. Interestingly, the new shoe design leads to a greater energy loss than traditional running shoes, which could be explained by a higher initial vertical stiffness.

Comparison of the correlations between impact loading rates and peak accelerations measured at two different body sites: Intra- and inter-subject analysis.

BACKGROUND: High average (VALR) and instantaneous vertical loading rates (VILR) during impact have been associated with many running-related injuries. Peak acceleration (PA), measured with an accelerometer, has provided an alternative method to estimate impact loading during outdoor running. This study sought to compare both intra- and inter-subject correlations between vertical loading rates and PA measured at two body sites during running. METHODS: Ground reaction force data were collected from 10 healthy adults (age=23.6±3.8 years) during treadmill running at different speeds and inclination surfaces. Concurrently, PAs at the lateral malleoli and the distal tibia were measured using synchronized accelerometers. RESULTS: We found significant positive intra-subject correlation between loading rates and PA at the lateral malleoli (r=0.561-0.950, p<0.001) and the distal tibia (r=0.486-0.913, p<0.001). PA measured at the lateral malleoli showed stronger correlation with loading rates (p=0.004) than the measurement at the distal tibia. On the other hand, inter-subject variances were observed in the association between PA and vertical loading rates. The inter-subject variances at the distal tibia were 3.88±3.09BW/s and 5.69±3.05BW/s in VALR and VLIR respectively. Similarly, the inter-subject variances in the measurement at lateral malleoli were 5.24±2.85BW/s and 6.67±2.83BW/s in VALR and VLIR respectively. CONCLUSIONS: PA measured at lateral malleoli has stronger correlation with VALR or VILR than the measurement at distal tibia. Caution is advised when using PA to conduct inter-subject comparisons of vertical loading rates during running.