Neighbors-based prediction of physical function after total knee arthroplasty

Area of Science:

• Orthopedics
• Rehabilitation Medicine
• Biostatistics

Background:

• Total Knee Arthroplasty (TKA) is a common procedure for knee osteoarthritis.
• Predicting functional recovery post-TKA is crucial for patient management and setting expectations.
• Current prediction methods may lack personalization and accuracy.

Purpose of the Study:

• To develop and validate a novel neighbors-based prediction approach for functional recovery after TKA.
• To assess the accuracy and calibration of these personalized predictions.

Main Methods:

• Utilized data from 397 TKA patients for methodology development and 202 patients for validation.
• Employed a neighbors-based approach, matching patients based on preoperative TUG, age, sex, and BMI.
• Assessed physical function using the Timed Up and Go (TUG) Test.

Main Results:

• Identified an optimal match number (m=35) balancing bias, coverage, and precision.
• Predictions demonstrated good calibration in out-of-sample testing.
• The method showed potential for accurate, personalized functional recovery predictions.

Conclusions:

• The developed neighbors-based prediction method offers personalized insights into TKA functional recovery.
• These predictions can inform clinical decisions before and after TKA surgery.
• This approach represents a step towards more tailored patient care in orthopedic surgery.