Local delivery of vancomycin for the prophylaxis of prosthetic device-related infections.

PURPOSE: To evaluate the in vivo efficacy and pharmacokinetics of vancomycin delivered from glycerylmonostearate (GMS) implants in a prosthetic-device based biofilm infection model.

METHODS: A biofilm infection model was developed in male Sprague-Dawley rats by implanting a vascular graft on the dorsal side of each rat and infecting it with 1.5 x 10(8) cfu/ml Staphylococcus epidermidis. The rats were divided into 3 groups of 6 rats each: 1) the control group that received no antibiotics, 2) the IM group that received multiple IM injections of vancomycin at a dose of 25 mg/kg every 6 h for a total of 12 doses, and 3) the implant group that received GMS implants designed to deliver vancomycin at a total dose of 300 mg/kg for a period of 4 days. The pharmacokinetics of vancomycin was determined from IM and implant groups by analyzing for vancomycin in blood using HPLC. In vivo efficacy was studied by evaluation of the wound site and the prosthetic device upon excision, for evidence of infection in the form of purulent discharge at the wound site and yellowish discoloration of the prosthetic device and inflammation as sign of biofilm formation. Microbiological evaluation on the wound site and the prosthetic device was performed by culturing the swabs at the wound site and the prosthetic device in sterile tryptic soy broth for 36-48 h at 37 degrees C.

RESULTS: Vancomycin was successfully delivered in a sustained manner for 100 h from GMS implants and the resulting plasma profile showed that the concentrations, after an initial burst, plateaued at about of 4.77 +/- 1.43 mug/ml with less fluctuations than the IM group in which the plasma concentrations fluctuated between 2.73 +/- 0.94 mug/ml and 19.26 +/- 3.67 mug/ml. Upon excision of the wound site, all the animals in the control group developed infection in the form of purulent discharge and yellowish discoloration of the prosthetic device. However, none of the rats in the implant group showed evidence of infection clearly demonstrating the efficacy of the local delivery system in preventing infection. Systemically delivered vancomycin by IM injections failed to prevent infection in four out of six rats. Microbiological evaluation of the wound site and prosthetic device resulted in isolation of biofilm-producing organisms such as Staphylococcus epidermidis, Enterococcus faecalis, and Staphylococcus aureus. These organisms were isolated in greater number of animals in the control group compared to the IM and implant groups.

CONCLUSIONS: The GMS implants as a delivery system for vancomycin were successful in preventing infection in all the animals compared to the IM and control groups demonstrating the efficacy of a local delivery system in a prosthetic device related biofilm infection model.