¹Klinik für Orthopädie und Unfallchirurgie, Muskuloskelettales Universitätszentrum München (MUM), LMU Klinikum, München, Deutschland

Text

Objectives and questions: Periprosthetic joint infections (PJIs) remain a significant complication following total hip arthroplasty, impacting both patient outcomes and healthcare costs. Accurate classification of PJIs is crucial for guiding treatment decisions and optimizing patient management. The TNM classification system, commonly used in oncology, has been adapted for PJIs to improve the assessment of infection severity. This study aims to investigate the application of the PJI-TNM classification by assessing its interobserver and intraobserver reliability and evaluating its ability to predict outcomes, including re-infection, complications, and the need for re-revision surgery.

Material and methods: All patients who underwent revision at our institution between January 2013 and June 2024 due to diagnosed PJI following THA were retrospectively identified. A total of 83 acute and 101 chronic PJIs were included. Each case was classified according to the PJI-TNM framework. Classification accuracy was assessed separately for each subcategory. Interobserver reliability was evaluated using Fleiss’ kappa, while intraobserver reliability was determined using Cohen’s kappa. The predictive value of the preoperative classification for clinical outcomes, including re-infection, complications, and the need for re-revision surgery, was analyzed using logistic regression. Correlation between classification and outcomes was examined using Spearman’s rank correlation. Predictive performance was further assessed through receiver operating characteristic curve analysis. Statistical significance was defined as p < 0.05.

Results: In the acute PJIs, the majority was T0 (n=76), indicating a stable standard or revision implant without significant soft tissue defects, followed by T1 (n=4) with a loosened implant and T2 (n=3) with severe soft tissue defects. The most common N classification was N0 (n=66), representing infections without mature biofilm, followed by N2 (n=9) with mature biofilm and difficult-to-treat or polymicrobial infections, and N1 (n=8) with mature biofilm but without difficult-to-treat bacteria or culture-negative infections. Regarding comorbidity status, the majority of patients were classified as M2 (n=36), indicating severe comorbidities (Charlson Comorbidity Index (CCI) 4–5), followed by M1 (n=30) with moderate comorbidities (CCI 2–3) and M0 (n=17) with mild or no comorbidities (CCI 0–1). Classification of the chronic group as well as, intraobserver and interobserver reliability, and correlations between the PJI-TNM classification and clinical outcomes, are currently being analyzed, with results expected within a couple of weeks.

Discussion and conclusions: The PJI-TNM classification offers a detailed understanding of periprosthetic infections, improving communication among healthcare providers and supporting research into PJI outcomes. The upcoming study results will clarify the classification’s predictive value and clinical utility.