Preoperative Risk Factors and Phenotypic Clustering in Periprosthetic Joint Infection: A Matched Case-Control Study.

Area of Science:

• Orthopedic Surgery
• Infectious Disease
• Biostatistics

Background:

• Periprosthetic joint infection (PJI) is a severe complication following total joint arthroplasty.
• Effective preoperative risk stratification is crucial for PJI prevention.

Purpose of the Study:

• To identify preoperative predictors of PJI.
• To explore data-driven patient phenotypes associated with PJI using unsupervised learning.

Main Methods:

• Retrospective case-control study of 182 patients undergoing hip or knee arthroplasty.
• Multivariable logistic regression with LASSO regularization for risk factor identification.
• K-means clustering for patient phenotype discovery.

Main Results:

• Preoperative anemia (OR 5.91), higher ASA score (OR 1.77), and surgical delay (OR 1.67) independently predicted PJI.
• A five-variable preoperative model achieved an AUC of 0.718.
• Three distinct patient phenotypes were identified with varying PJI rates.

Conclusions:

• Anemia, comorbidity burden, and surgical delay are significant, modifiable targets for PJI prevention.
• An integrative approach combining prediction and phenotype discovery aids clinical risk stratification.
• Findings highlight the need for preoperative optimization to reduce PJI risk.