Experimental colonization of vascular grafts with Staphylococcus aureus

In an attempt to solve the problem of infection of arterial grafts, the authors designed an experimental model to reproduce, in vitro, hematogenous seeding of grafts with Staphylococcus aureus. The inoculum, containing an average of 10(7) viable bacterial cells per millilitre, was circulated through grafts of various types. Normal dog aortas were used as controls. They entrapped a mean of 8 bacteria/cm2. The prosthetic grafts, previously placed in dogs as femorofemoral arteriovenous bypasses for 2 hours, trapped many more cells: expanded polytetrafluoroethylene, 23 cells/cm2; bovine heterograft, 607 cells/cm2 and Dacron velour, 2801 cells/cm2. All cell counts were significantly (p less than 0.001) different from control values. Thoracoabdominal aortic bypass grafts implanted in dogs 2 months previously gave the following mean numbers of trapped bacteria: expanded polytetrafluoroethylene, 19 122 cells/cm2; bovine heterograft, 863 cells/cm2 and Dacron velour, 3500 cells/cm2. Polytetrafluoroethylene had significantly (p less than 0.001) higher numbers of trapped bacteria than any other type of prosthesis. The bacteria were located mainly on irregular fibrin strands and on surface defects of the grafts. The addition of cefazolin during the seeding process at concentrations 10 to 25 times the minimal inhibitory concentration did not decrease the numbers of bacteria in any graft. Bacterial colonization of prosthetic arterial grafts depends on the graft material and on the duration of implantation, but this study provided no answer to the controversial question of how to prevent arterial graft infections with antimicrobial agents in patients who undergo vascular procedures that expose them to bacteremia.