Reply

Abstract

Dr. Miller states that we have succeeded in demonstrating superiority of the rifampin-bonded Dacron graft only in controlling the proliferation of bacteria in contaminated grafts and not in established infection. This was clearly not the case. The method of establishing graft sepsis included direct contamination of the prosthesis with a gross inoculum of Staphylococcus aureus. The presence of an established infection was clearly documented at the time the animals underwent reexploration. There was an exudate, infected surrounding soft tissues, and infection involving the anastomotic sites. Infection in these areas was documented by the presence of positive cultures in the graft, as well as positive cultures in the debrided surrounding tissues. Dr. Miller asks why we simply did not wait longer to see what would happen. That was not an essential element of our experiment; that has been clearly demonstrated in prior experiments showing that anastomotic disruption, graft thrombosis, and continued sepsis will occur. We did not think we needed to prove that again. To the best of my knowledge, there has been no instance in which a graft that has been shown to harbor organisms will spontaneously clear, with or without antibiotics. Clearly, this is the pernicious aspect of an infected foreign body. Although we share Dr. Miller's enthusiasm for polytetrafluoroethylene (PTFE) grafts in the infrainguinal region, their use in the abdominal aorta has not been uniformly accepted. This has nothing to do with the potential for infectibility but rather the handling characteristics in performing an aortic reconstruction. Although there are many articles in the literature suggesting that PTFE is more resistant to infection in contaminated fields, this evidence is largely anecdotal. In a prior study we examined the issue of various graft material and design with respect to bacteremic infectibility. We were not able to demonstrate a difference between PTFE and Dacron in an animal model using a bacteremic challenge to produce vascular graft sepsis.1Roon AJ Malone JM Moore WS et al.Bacteremic infectibility: a function of vascular graft material and design.J Surg Res. 1977; 22: 489-498Abstract Full Text PDF PubMed Scopus (38) Google Scholar We are envious of Dr. Miller's statement that he has never had to remove a PTFE graft for infection. I wish that I could make the same claim. Although the frequency of infection with PTFE grafts in the infrainguinal region is low, it is certainly not zero. I have had the opportunity to treat several PTFE grafts that were infected and exposed. There is a tendency for granulation tissue to cover PTFE grafts; however, in every instance in which that was tried, just as the edges of granulation tissue met, the graft thrombosed. We have also had several instances of chronic draining sinus from infected PTFE grafts that would heal only on removal of the prosthesis. Therefore, although PTFE grafts may well be less susceptible to infection, they are certainly not immune. Finally, Dr. Miller asks why we did not include a PTFE graft arm in the study. Certainly that would have been an interesting part of the experiment. However, to date we have not been able to demonstrate an effective way to bond antibiotics to PTFE graft material. Because the bonding of rifampin to the graft material was the subject of our experiment, the inclusion of a PTFE arm was not possible. 24/41/50241