Aims. The management of periprosthetic joint infection (PJI) after total knee arthroplasty (TKA) is challenging. The correct antibiotic management remains elusive due to differences in epidemiology and resistance between countries, and reports in the literature. Before the efficacy of surgical treatment is investigated, it is crucial to analyze the bacterial strains causing PJI, especially for patients in whom no organisms are grown. Methods. A review of all revision TKAs which were undertaken between 2006 and 2018 in a tertiary referral centre was performed, including all those meeting the consensus criteria for PJI, in which organisms were identified. Using a cluster analysis, three chronological time periods were created. We then evaluated the antibiotic resistance of the identified bacteria between these three clusters and the effectiveness of our antibiotic regime. Results. We identified 129 PJIs with 161 culture identified bacteria in 97 patients. Coagulase-negative staphylococci (CNS) were identified in 46.6% cultures, followed by Staphylococcus aureus in 19.8%. The overall resistance to antibiotics did not increase significantly during the study period (p = 0.454). However, CNS resistance to teicoplanin (p < 0.001), fosfomycin (p = 0.016), and tetracycline (p = 0.014) increased significantly. Vancomycin had an 84.4% overall sensitivity and 100% CNS sensitivity and was the most effective agent. Conclusion. Although we were unable to show an overall increase in antibiotic resistance in organisms that cause PJI after TKA during the study period, this was not true for CNS. It is concerning that resistance of CNS to new antibiotics, but not vancomycin, has increased in a little more than a decade. Our findings suggest that referral centres should continuously monitor their bacteriological analyses, as these have significant implications for prophylactic treatment in both primary arthroplasty and revision arthroplasty for PJI. Cite this article: Bone Joint J 2021;103-B(6 Supple A):171–176

Introduction. Management of periprosthetic joint infection (PJI) after total knee arthroplasty (TKA) is a challenging task. Antibiotic management remains elusive due to differences in epidemiology and resistance between countries and reports. Increasing resistance of bacteria is a concern. The aim of this study was to investigate the development of antibiotic resistance of main causative bacteria in a single center. Methods. A retrospective review of all revision TKA between 2006 and 2018 in a tertiary referral center was performed. Included were cases meeting the consensus criteria for PJI with at least two positive cultures. Three chronological groups were created using a two-step cluster analysis. Results. In 129 cases of PJI there were 161 different bacteria identified by growth in culture. Of all bacteria 76.4% were gram positive, 23.6% were gram negative. Coagulase-negative Staphylococci (CNS) were diagnosed in 46.6% cultures, followed by Staphylococcus aureus in 19.8% of cultures. Overall antibiotic resistance (p=0.454) and Staphylococcus aureus resistance (p=0.788) have not increased during the study period. CNS resistance to teicoplanin (4% to 44%; p<0.001), fosfomycin (16% to 44%, p=0.016) and tetracycline (0% to 28%, p=0.014) increased significantly. Vancomycin had an 84.4% overall sensitivity and 100% CNS sensitivity and remained the most effective agent. Conclusion. The increasing resistance to several antibiotics over a little more than a decade is concerning. This study suggests that referral centres should continuously monitor obtained cultures as this has significant implications for both prophylactic treatment in primary as well as empirical antibiotic treatment in PJI. Resistance profiles may differ between centres and are subject to dynamic changes