Limitations in the use of rifampicin-gelatin grafts against virulent organisms.

OBJECTIVE: Efficacy and duration of antibacterial activity of rifampicin-gelatin grafts against virulent organisms were evaluated in an animal model. MATERIALS AND METHODS: Rifampicin-gelatin grafts were prepared with impregnation of Gelseal (Vascutek Ltd, Scotland) graft in 1 mg/mL rifampicin solution. Rifampicin-gelatin grafts (6 cm long; n = 24) and plain Gelseal grafts as controls (n = 4) were implanted into the canine abdominal aorta with inoculation of Staphylococcus epidermidis, Escherichia coli, or methicillin-resistant Staphylococcus aureus (MRSA), and the rifampicin-gelatin grafts were retrieved after 1 to 4 weeks. Disks cut from the retrieved rifampicin-gelatin grafts were placed on agar plates streaked with one of the organisms, and the graft antibacterial activity was assessed with the width of the inhibition zone. RESULTS: In in vitro tests, initial inhibition zones (inhibition zone of 24 hours after incubation) of rifampicin-gelatin grafts against S epidermidis, MRSA, and E coli were 40.0 +/- 0.3 mm, 36.0 +/- 0.2 mm, and 11.8 +/- 0.1 mm, respectively. In the implantation, S epidermidis -inoculated rifampicin-gelatin grafts had no findings of graft infection, and no colony growth was recognized on the plates streaked with the perigraft fluids. Initial inhibition zones of S epidermidis -inoculated rifampicin-gelatin grafts retrieved at 1 or 2 weeks were 20.1 +/- 1.1 mm and 7.6 +/- 1.0 mm, respectively. In E coli -inoculated and MRSA-inoculated rifampicin-gelatin grafts, all of the eight animals had perigraft abscess, and blood culture test results probed septicemia in five animals with patent grafts at death. Inhibition zones against E coli or MRSA were not formed on the plates streaked with the same organism, whereas initial inhibition zones of E coli -inoculated and MRSA-inoculated rifampicin-gelatin grafts on S epidermidis -streaked plates were 8.0 +/- 0.2 mm and 18.5 +/- 0.5 mm, respectively. In the MRSA group, however, recolonization of high minimal inhibitory concentration strains developed within the inhibition zones as early as 24 hours. Histologically, neither organisms nor inflammatory cells were found in S epidermidis -inoculated rifampicin-gelatin grafts and tissue ingrowth was recognized at 2 to 4 weeks, whereas E coli -inoculated and MRSA-inoculated rifampicin-gelatin grafts had aggressive neutrophil infiltration into the graft interstices, revealing establishment of uncontrollable graft infection. CONCLUSION: These results suggested that rifampicin-gelatin grafts are clearly valid for S epidermidis infection, whereas no efficacy was recognized against either MRSA or E coli graft infection because of early development of high minimal inhibitory concentration MRSA strains or poor susceptibility.