Prediction of suitable outpatient candidates following revision total knee arthroplasty using machine learning

Abstract

AimsTo identify variables independently associated with same-day discharge (SDD) of patients following revision total knee arthroplasty (rTKA) and to develop machine learning algorithms to predict suitable candidates for outpatient rTKA.MethodsData were obtained from the American College of Surgeons National Quality Improvement Programme (ACS-NSQIP) database from the years 2018 to 2020. Patients with elective, unilateral rTKA procedures and a total hospital length of stay between zero and four days were included. Demographic, preoperative, and intraoperative variables were analyzed. A multivariable logistic regression (MLR) model and various machine learning techniques were compared using area under the curve (AUC), calibration, and decision curve analysis. Important and significant variables were identified from the models.ResultsOf the 5,600 patients included in this study, 342 (6.1%) underwent SDD. The random forest (RF) model performed the best overall, with an internally validated AUC of 0.810. The ten crucial factors favoring SDD in the RF model include operating time, anaesthesia type, age, BMI, American Society of Anesthesiologists grade, race, history of diabetes, rTKA type, sex, and smoking status. Eight of these variables were also found to be significant in the MLR model.ConclusionThe RF model displayed excellent accuracy and identified clinically important variables for determining candidates for SDD following rTKA. Machine learning techniques such as RF will allow clinicians to accurately risk-stratify their patients preoperatively, in order to optimize resources and improve patient outcomes.Cite this article: Bone Jt Open 2023;4(6):399–407.

Enabling Personalized Medicine in Orthopaedic Surgery Through Artificial Intelligence

“…Importantly, anticipating a patient's specific postoperative outcome and clinical trajectory requires simultaneous consideration of several different patient-specific factors. Applications of AI in preoperative evaluation include determining candidacy for surgery and identification of factors associated with achieving meaningful, patient-specific postoperative outcomes 34-60 (Table I), which may currently be the most frequent application of AI within orthopaedics. Because the current value-driven health care system places importance on clinically significant outcome improvements, PROMs serve an important role in the quantification of patient-perceived improvement and therefore help determine the quality of care that patients receive 61-65 .…”

What’s New in Adult Reconstructive Knee Surgery

“…The random-forest model had the best accuracy, and variables of importance were joint-line change, postoperative femorotibial angle, and hemoglobin A1c 64 . Yeramosu et al 65 compared a multivariable logistic regression model and diverse machine learning techniques and found that 1 of the models (random forest) had an accuracy (AUC) of 0.810 for identifying candidates for same-day discharge after revision TKA. According to that model, the factors of importance for same-day discharge, in decreasing order, were operative time, anesthesia type, age, BMI, American Society of Anesthesiologists (ASA) class, race, diabetes, revision TKA type, sex, and smoking.…”

A leap towards personalized orthopaedic surgery and the prediction of spinopelvic mechanics in total hip arthroplasty

“…Its applications range from the identification of fractures and surgical planning to predicting length of stay and postoperative complications. 25,26 However, while the promise is immense, there are inherent challenges in integrating AI into clinical practice. It is vital that AI models are developed with transparent methodologies and high-quality data, and that they undergo systematic external validation.…”