255. Utility of Magnetic Resonance Imaging in Predicting Failure of DAIR (Debridement, Antibiotics, and Implant Retention) for Treatment of Periprosthetic Joint Infection. (4th December 2021)

• Record Type: Journal Article
• Authors: Kapadia, Milan; Burge, Alissa; Bogner, Eric; Sculco, Peter; Carli, Alberto V
• Is Part Of: Open forum infectious diseases. Volume 8(2021)Supplement 1
• Page Start: S234
• Page End: S234
• Publication Date: 2021-12-04
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofab466.457
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Debridement, antibiotics, and implant retention (DAIR) is commonly utilized to treat acute periprosthetic joint infection (PJI) where thorough debridement of affected tissues is considered essential. Small case series describe occurrences where PJI spreads into adjacent tissues (iliopsoas recess in hips; neurovascular bundle in knees) and bone (osteomyelitis). Surgeons often cite adjacent tissue/bone infiltration as a poor predictor for DAIR. We sought to evaluate if the presence of adjacent tissue/bony lesions on preoperative magnetic resonance imaging (MRI) was associated with poorer DAIR outcomes. Methods: MSIS criteria-positive hip (n=22) and knee (n=12) PJI cases in our institution from 2010-2020, with preoperative MRI prior to DAIR treatment, were evaluated. Demographics, microbiology, chronicity, and host grade were recorded. MRIs were assessed by two board-certified radiologists blinded to treatment outcomes, scoring images based on the presence of 18 distinct findings. Inter-rater reliability was calculated using bias adjusted Kappa scores. Failure was defined as repeat surgery for PJI. Univariate analysis and logistic regression were used to determine predictors of DAIR success at 90 days and 2 years. Results: When comparing successful and non-successful hip PJI cases, the presence of a psoas recess fluid collection on MRI was significantly predictive of a higher rate of treatment failure at 2 years (odds ratio=0.12; p = 0.045), with a moderate adjusted kappa score of 0.5. With regard to knee PJI cases, capsular disruption (40% [2/5] vs 100% [7/7], p= 0.046) and patellar tendon disruption (25% [1/4] vs 100% [8/8], p = 0.018) were independently associated with higher 90-day failure. However, knee MRI findings were not predictive of failure using regression. Conclusion: In this preliminary study, preoperative MRI findings anecdotally linked with PJI treatment failure could be reliably identified. However, few predicted DAIR failure. Further studies are needed to clarify the role of MRI in predicting PJI treatment success. Disclosures: All Authors : No reported disclosures