Prevention of graft infection by use of prostheses bonded with a rifampin/collagen release system

Abstract

The objective of this study was to test the efficacy of bonding rifampin to double-velour Dacron grafts with collagen to prevent graft sepsis. Fifty 6.0 mm Dacron grafts (length 5.0 cm) impregnated with either collagen (control) or collagen plus rifampin (experimental) were implanted in dogs end-to-end into the infrarenal aorta. The dogs were divided into four groups (each with an experimental and control subdivision) as a function of time between grafting and bacterial challenge. At 2, 7, 10, or 12 days after graft implantation, sequential groups were challenged with 1.2 × 108 colony forming units of Staphylococcus aureus (clinical isolate) intravenously suspended in 250 ml normal saline. Three weeks after hematogenous seeding, the grafts were sterilely harvested. One-tailed Fischer's exact test was used to compare the patency and culture-proven infection of control and antibiotic coated grafts as a function of implantation time before bactermic challenge. In the 2-day group, four of six control grafts were infected compared with zero of six experimental grafts (p < 0.030). In the 7-day group, five of six control grafts were infected with S. aureus versus zero of six in the experimental group (p < 0.008). In the 10-day group, one of six experimental grafts was infected, but the control group had only two of six graft infections. In the 12-day group two of six experimental grafts and one of five control grafts were infected. These results indicate that rifampin bonded with collagen to knitted Dacron grafts will protect the graft from bacteremic infection for 7 days after implantation in a highly challenging model. After 7 days accelerated healing of the collagen graft surface may, in itself, protect against delayed bacterial seeding.