Prediction Model for Factors Associated With Long-Term Opioid Use Following Total Knee Arthroplasty: A Retrospective
Pin-Hung Yeh1,2,3, Shun-Fa Yang1,4, Jing-Yang Huang1,4
1Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan.
European Journal of Pain (London, England)
|December 25, 2025
Summary
This study identified risk factors for long-term opioid use after knee replacement surgery. A predictive model can help identify high-risk patients early for targeted prevention.
Area of Science:
- Orthopedic Surgery
- Pain Management
- Data Science in Healthcare
Background:
- Knee osteoarthritis affects 10% of the population, with 20% undergoing total knee arthroplasty (TKA).
- A significant proportion of TKA patients develop long-term opioid use.
- Limited comprehensive evaluations of perioperative risk factors for this complication exist.
Purpose of the Study:
- To systematically identify risk factors for long-term opioid use following TKA.
- To develop and evaluate a predictive model for early identification of high-risk patients.
Main Methods:
- A population-based analysis of 15,051 patients undergoing TKA (2002-2018).
- Utilized Least Absolute Shrinkage and Selection Operator (LASSO)-Elastic Net logistic regression for primary analysis.
- Employed hierarchical logistic regression models incorporating demographic, preoperative, in-hospital, and post-discharge variables.
Main Results:
- Identified 1029 long-term opioid users.
- The LASSO-Elastic Net model achieved an Area Under the Curve (AUC) of 0.8213.
- Hierarchical models showed increasing predictive power, with the most comprehensive model reaching an AUC of 0.8251.
Conclusions:
- Risk factors for long-term opioid use were identified across TKA care phases.
- A predictive model demonstrates potential for early identification of patients at high risk for prolonged opioid use.
- Findings aid clinicians in early recognition and targeted preventive strategies for opioid use post-TKA.
