Prosthetic joint infection of the knee - arthroscopic biopsy identifies more and different organisms than aspiration alone

BACKGROUND: Prosthetic joint infection (PJI) causes significant morbidity and mortality following knee replacement surgery. Identifying causative organisms and antibiotic sensitivities is critical in increasing the chance of infection eradication. This study investigated whether biopsy alone was superior to aspiration alone for serological diagnosis in PJI following knee replacement. Secondly, we investigated whether biopsy identifies the same or new/different microbiological flora as aspiration. METHODS: Since December 2014, the Exeter Knee Reconstruction Unit (EKRU) has prospectively collated data regarding all PJIs referred from our local/regional network which have been reviewed via our Multi-Disciplinary Team (MDT). We identified and included consecutive patients from this MDT from Dec.2014-Mar.2020 and analysed their electronic records. Statistical analysis was performed using Stata. RESULTS: 65/100 patients studied had both pre-operative aspiration and biopsy. 31/65 (48%) had positive aspiration and biopsies. No aspirate samples were positive with corresponding biopsies negative. In 19/65 (29%) of infection positive patients, biopsy identified new (7) or additional (12) organisms not identified by aspiration. Aspiration had a sensitivity of 70%, specificity of 88%, positive predictive value of 90.3% and negative predictive value of 64.7%. Biopsy had a sensitivity of 97.5%, specificity of 88%, positive predictive value of 92.9% and negative predictive value of 95.7%. CONCLUSION: In 29% of confirmed PJI cases, arthroscopic biopsy identified either additional organisms in a polymicrobial PJI when compared to aspiration, or new positive results when aspiration alone was negative. This study demonstrates the benefits of arthroscopic biopsy for serological diagnosis in cases of knee PJI and aids treatment planning.RD&E staff can access the full-text of this article by clicking on the 'Additional Link' above and logging in with NHS OpenAthens if prompted.Published version, accepted version (12 month embargo