A machine-learning model using preoperative characteristics slightly improved the identification of patients at high risk of medical complications after fast-track hip and knee arthroplasty compared to a logistic regression model (AUROC 76.3% vs 74.7%).
Cohort (n=21,926)
Yes
Does a machine-learning model improve the prediction of postoperative medical morbidity leading to prolonged length of stay or readmission in patients undergoing fast-track total hip and knee arthroplasty compared to logistic regression?
A machine-learning model using preoperative characteristics slightly improved the identification of patients at high risk for medical complications after fast-track hip and knee arthroplasty compared to traditional logistic regression.
Absolute Event Rate: 76.3% vs 74.7%
BACKGROUND: Machine-learning models may improve prediction of length of stay (LOS) and morbidity after surgery. However, few studies include fast-track programs, and most rely on administrative coding with limited follow-up and information on perioperative care. This study investigates potential benefits of a machine-learning model for prediction of postoperative morbidity in fast-track total hip (THA) and knee arthroplasty (TKA). METHODS: Cohort study in consecutive unselected primary THA/TKA between 2014-2017 from seven Danish centers with established fast-track protocols. Preoperative comorbidity and prescribed medication were recorded prospectively and information on length of stay and readmissions was obtained through the Danish National Patient Registry and medical records. We used a machine-learning model (Boosted Decision Trees) based on boosted decision trees with 33 preoperative variables for predicting "medical" morbidity leading to LOS > 4 days or 90-days readmissions and compared to a logistical regression model based on the same variables. We also evaluated two parsimonious models, using the ten most important variables in the full machine-learning and logistic regression models. Data collected between 2014-2016 (n:18,013) was used for model training and data from 2017 (n:3913) was used for testing. Model performances were analyzed using precision, area under receiver operating (AUROC) and precision recall curves (AUPRC), as well as the Mathews Correlation Coefficient. Variable importance was analyzed using Shapley Additive Explanations values. RESULTS: Using a threshold of 20% "risk-patients" (n:782), precision, AUROC and AUPRC were 13.6%, 76.3% and 15.5% vs. 12.4%, 74.7% and 15.6% for the machine-learning and logistic regression model, respectively. The parsimonious machine-learning model performed better than the full logistic regression model. Of the top ten variables, eight were shared between the machine-learning and logistic regression models, but with a considerable age-related variation in importance of specific types of medication. CONCLUSION: A machine-learning model using preoperative characteristics and prescriptions slightly improved identification of patients in high-risk of "medical" complications after fast-track THA and TKA compared to a logistic regression model. Such algorithms could help find a manageable population of patients who may benefit most from intensified perioperative care.
Michelsen et al. (Wed,) conducted a cohort in Primary total hip (THA) and knee arthroplasty (TKA) (n=21,926). Machine-learning model (Boosted Decision Trees) vs. Logistic regression model was evaluated on Prediction of medical morbidity leading to length of stay > 4 days or 90-day readmission (AUROC). A machine-learning model using preoperative characteristics slightly improved the identification of patients at high risk of medical complications after fast-track hip and knee arthroplasty compared to a logistic regression model (AUROC 76.3% vs 74.7%).