Related Experiment Video
Updated: Oct 7, 2025
14:56
An Experimental Paradigm for the Prediction of Post-Operative Pain PPOP
Published on: January 27, 2010
21.5K
Machine learning algorithms predict extended postoperative opioid use in primary total knee arthroplasty.
Christian Klemt1, Michael Joseph Harvey1, Matthew Gerald Robinson1
1Bioengineering Laboratory, Department of Orthopaedic Surgery, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Boston, MA, 02114, USA.
Summary
Machine learning accurately predicts extended opioid use after total knee arthroplasty (TKA). This helps identify at-risk patients for tailored pain management and reduced complications.
Area of Science:
- Orthopedic Surgery
- Pain Management
- Machine Learning in Healthcare
Background:
- Effective pain management is crucial for recovery after total knee arthroplasty (TKA).
- Prolonged opioid use post-TKA is a significant concern, linked to adverse outcomes.
- Predicting extended opioid use is essential for personalized patient care.
Purpose of the Study:
- To develop and validate machine learning models for predicting extended opioid use following primary TKA.
- To identify key predictors of prolonged postoperative opioid use in TKA patients.
Main Methods:
- Evaluation of 8873 patients undergoing primary TKA, with 7.2% experiencing extended opioid use (>90 days).
- Manual review of electronic health records to extract demographic and surgical variables.
- Development and assessment of five machine learning algorithms using discrimination, calibration, and decision curve analysis.
Main Results:
- Strongest predictors identified: preoperative opioid duration (100%), drug abuse (54%), and depression (47%).
- All five machine learning models demonstrated excellent predictive performance (AUC > 0.83).
- Machine learning models showed superior net benefits compared to standard management strategies.
Conclusions:
- Machine learning models show high accuracy in predicting extended opioid use after TKA.
- These models can aid in pre-operative identification of at-risk individuals.
- Enables tailored peri-operative counseling and pain management to mitigate opioid-related complications.
