Using machine learning to identify risk factors for short-term complications following thumb carpometacarpal arthroplasty

Rohan M Shah; Rushmin Khazanchi; Anitesh Bajaj; Krishi Rana; Saaz Malhotra; Jennifer Moriatis Wolf

doi:10.1016/j.jham.2024.100156

Using machine learning to identify risk factors for short-term complications following thumb carpometacarpal arthroplasty

J Hand Microsurg. 2024 Sep 20;16(5):100156. doi: 10.1016/j.jham.2024.100156. eCollection 2024 Dec.

Authors

Rohan M Shah¹, Rushmin Khazanchi¹, Anitesh Bajaj¹, Krishi Rana², Saaz Malhotra¹, Jennifer Moriatis Wolf³

Affiliations

¹ Northwestern University Feinberg School of Medicine, Chicago, IL, USA.
² Northwestern University, Evanston, IL, USA.
³ Department of Orthopaedic Surgery, University of Chicago Pritzker School of Medicine, Chicago, IL, USA.

PMID: 39669732
PMCID: PMC11632740 (available on 2025-09-20)
DOI: 10.1016/j.jham.2024.100156

Abstract

Background: Thumb carpometacarpal (CMC) joint osteoarthritis is among the most common degenerative hand diseases. Thumb CMC arthroplasty, or trapeziectomy with or without tendon augmentation, is the most frequently performed surgical treatment and has a strong safety profile. Though adverse outcomes are infrequent, the ability to predict risk for complications has substantial clinical benefits. In the present study, we evaluated a well-known surgical database with machine learning (ML) techniques to predict short-term complications and reoperations after thumb CMC arthroplasty.

Methods: A retrospective study was conducted using data from the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) years 2005-2020. Outcomes were 30-day wound and medical complications and 30-day return to the operating room. We used three ML algorithms - a Random Forest (RF), Elastic-Net Regression (ENet), and Extreme Gradient Boosted Tree (XGBoost), and a deep learning Neural Network (NN). Feature importance analysis was performed in the highest performing model for each outcome to identify predictors with the greatest contributions.

Results: We included a total of 7711 cases. The RF was the best performing algorithm for all outcomes, with an AUC score of 0.61±0.03 for reoperations, 0.55±0.04 for medical complications, and 0.59±0.03 for wound complications. On feature importance analysis, procedure duration was the highest weighted predictor for reoperations. In all outcomes, procedure duration, older age, and female sex were consistently among the top five predictors.

Conclusions: We successfully developed ML algorithms to predict reoperations, wound complications, and medical complications. RF models had the highest performance in all outcomes.

Keywords: Deep learning; Machine learning; Trigger finger release.