Predictive factors for length of hospital stay following primary total knee replacement in a total joint replacement centre in Hong Kong

To Predict the Length of Hospital Stay After Total Knee Arthroplasty in an Orthopedic Center in China: The Use of Machine Learning Algorithms

Background and Objectives: Total knee arthroplasty (TKA) is widely performed to improve mobility and quality of life for symptomatic knee osteoarthritis patients. The accurate prediction of patients' length of hospital stay (LOS) can help clinicians for rehabilitation decision-making and bed assignment planning, which thus makes full use of medical resources.

Methods: Clinical characteristics were retrospectively collected from 1,298 patients who received TKA. A total of 36 variables were included to develop predictive models for LOS by multiple machine learning (ML) algorithms. The models were evaluated by the receiver operating characteristic (ROC) curve for predictive performance and decision curve analysis (DCA) for clinical values. A feature selection approach was used to identify optimal predictive factors.

Results: The areas under the ROC curve (AUCs) of the nine models ranged from 0.710 to 0.766. All the ML-based models performed better than models using conventional statistical methods in both ROC curves and decision curves. The random forest classifier (RFC) model with 10 variables introduced was identified as the best predictive model. The feature selection indicated the top five predictors: tourniquet time, distal femoral osteotomy thickness, osteoporosis, tibia component size, and post-operative values of Hb within 24 h.

Conclusions: By analyzing clinical characteristics, it is feasible to develop ML-based models for the preoperative prediction of LOS for patients who received TKA, and the RFC model performed the best.

Modifiable, Postoperative Risk Factors for Delayed Discharge Following Total Knee Arthroplasty: The Influence of Hypotension and Opioid Use

BACKGROUND

We sought to identify independent modifiable risk factors for delayed discharge after total knee arthroplasty (TKA) that have been previously underrepresented in the literature, particularly postoperative opioid use, postoperative laboratory abnormalities, and the frequency of hypotensive events.

METHODS

Data from 1033 patients undergoing TKA for primary osteoarthritis of the knee between June 2012 and August 2014 at an academic orthopedic specialty hospital were reviewed. Patient demographics, comorbidities, inpatient opioid medication, postoperative hypotensive events, and abnormalities in laboratory values, all occurring on postoperative day 0 or 1, were collected. Multivariate logistic regression analysis was performed to identify independent risk factors for a prolonged length of stay (LOS) >3 days.

RESULTS

The average age of patients undergoing primary TKA in our cohort was 65.9 (standard deviation, 9.1) years, and 61.7% were women. The mean LOS for all patients was 2.64 days (standard deviation, 1.14; range, 1-9). And 15.3% of patients had a LOS >3 days. On multivariate logistic regression analysis, nonmodifiable risk factors associated with a prolonged LOS included nonwhite race (odds ratio [OR], 2.01), single marital status (OR, 1.53), and increasing age (OR, 1.47). Modifiable risk factors included every 5 postoperative hypotensive events (OR, 1.31), 10-mg increases in oral morphine equivalent consumption (OR, 1.04), and postoperative laboratory abnormalities (hypocalcemia: OR, 2.15; low hemoglobin: OR, 2.63).

CONCLUSION

This study identifies potentially modifiable factors that are associated with increased LOS after TKA. Doubling down on efforts to control the narcotic use and to use opioid alternatives when possible will likely have efficacy in reducing LOS. Attempts should be made to correct laboratory abnormalities and to be cognizant of patient opioid use, age, and race when considering potential avenues to reduce LOS.