Abstract
Orthopaedic surgical site infections, especially when a hardware is involved, are associated with biofilm formation. Clinical strategies for biofilm eradication still fall short. The present study used a novel animal model of long-bone fixation with vancomycin- or gentamicin-controlled release and measured the levels of antibiotic achieved at the site of release and in the surrounding tissue. Then, using fluids that contain serum proteins (synovial fluid or diluted serum), the levels of vancomycin or gentamicin required to substantially reduce colonising bacteria were measured in a model representative of either prophylaxis or established biofilms. In the in vivo model, while the levels immediately adjacent to the antibiotic release system were up to 50× the minimal inhibitory concentration in the first 24 h, they rapidly dropped. At peripheral sites, values never reached these levels. In the in vitro experiments, Staphylococcus aureus biofilms formed in serum or in synovial fluid showed a 5-10 fold increase in antibiotic tolerance. Importantly, concentrations required were much higher than those achieved in the local delivery systems. Finally, the study determined that the staged addition of vancomycin and gentamicin was not more efficacious than simultaneous vancomycin and gentamicin administration when using planktonic bacteria. On the other hand, for biofilms, the staged addition seemed more efficacious than adding the antibiotics simultaneously. Overall, data showed that the antibiotics' concentrations near the implant in the animal model fall short of the concentrations required to eradicate biofilms formed in either synovial fluid or serum.