Infection after total knee replacement, which is a serious and expensive complication, often represent a diagnostic and therapeutic problem. The current incidence of infection after the primary procedure is 1 to 3%, depending on the published series. A correct and timely diagnosis, classification between early and delayed infection, and which microorganisms are involved, are crucial steps in defining prevention and treatment strategies. Determination of the annual and three years incidence of infection after primary total knee replacement; evaluation of the microorganisms involved and its resistance patterns; assessment of treatment – surgical approach and selection of antibiotics. Collection of clinical and laboratorial data of all patients who underwent primary total knee arthroplasty between 2011 and 2013 in our hospital; definition of periprosthesic infection cases following the Musculoskeletal Infection Society (MSIS) criteria. During the study period, 526 primary knee replacements were performed in 521 patients; with 41 patients having bilateral replacements. The mean follow-up period was 30 months; 5 patients had no follow up and 1 died in the post operatory. We reported 9 prosthetic infections, of which 2 did not reached the MSIS criteria, but were also considered based on high clinical suspicion. The majority of the cases (6) were delayed infections. The calculated 3 year incidence of infection after primary knee replacement was 1,6%, with annual rates of 3,0% (2011), 1,7% (2012) and 0,9% (2013). The microorganisms isolated were as follows: Staphylococcus aureus and coagulase-negative staphylococci, resistant to penicillin; Streptococcus agalactiae and one isolate of Serratia marcescens, both showing multiple antibiotic resistances. Only one case was treated with surgical debridement and conservation of prosthesis, in the other 8 cases a two-stage implant revision procedure was performed. The antibiotics selected were vancomycin, fluoroquinolones and association of gentamicin and clindamycin. Our local infection rates are in line with the published series from reference surgical centers. The annual incidence is decreasing, probably because the majority of our infection cases are delayed (recent years, shorter follow up period) and our preventive measures are improving. The microorganisms identified are also in agreement with published data, and our antibiotic resistance pattern is a valuable information to consider in a first empirical approach. Treatment options suitable to each case, and antibiotic protocols need to be improved in our local practice. Preventive measures in delayed infections are still under debate, and represent another future challenge.
