Reply

Abstract

We thank Dr. Bergamini for his comments and appreciate his interest in our work. Although the animal model used by Bergamini et al. 1Bergamini TM Bandyk DF Govostis D Kaebnick HW Towne JB. Infection of vascular prostheses caused by bacterial biofilms.J VASC SURG. 1988; 7: 21-30PubMed Scopus (90) Google Scholar seems to be similar in nature to ours, there are important differences with respect to bacterial inoculum and methodology. In our study, we intended to develop a model of invasive Staphylococcus epidermidis graft infection by soaking the grafts in a high density (109 colony forming units [CFU]) bacterial suspension. In vitro tests confirmed that the number of viable bacteria deposited on the graft material was higher than 107 CFU per cm2, that is, higher than 108 CFU for a 5 cm graft. As shown by White et al. 2White JV Nessel CC Whang K Differential effect of type of bacteria on peripheral graft infections.in: Management of infected arterial grafts. Quality Medical Publishing, St Louis1994: 25-42Google Scholar in a study comparing the size inoculum and pathogenesis of Staphylococcus aureus and S. epidermidis graft infections, 2White JV Nessel CC Whang K Differential effect of type of bacteria on peripheral graft infections.in: Management of infected arterial grafts. Quality Medical Publishing, St Louis1994: 25-42Google Scholar once a critical number of bacteria is reached in the perigraft environment, both types of staphylococci aggressively invade the graft, causing the signs and symptoms of overt infection. In the model used by Bergamini et al., 1Bergamini TM Bandyk DF Govostis D Kaebnick HW Towne JB. Infection of vascular prostheses caused by bacterial biofilms.J VASC SURG. 1988; 7: 21-30PubMed Scopus (90) Google Scholar the inoculating solution consisted of 107 CFU/ml S. epidermidis, and the bacterial concentration of the graft biofilm was 106 CFU/cm2 graft material. The grafts were immersed in the inoculating solution for 24 hours at room temperature, which may have influenced the viability of bacteria. These methodologic differences may account for the variations in the mode of presentation of S. epidermidis graft infection between our model and the model used by Bergamini et al. 1Bergamini TM Bandyk DF Govostis D Kaebnick HW Towne JB. Infection of vascular prostheses caused by bacterial biofilms.J VASC SURG. 1988; 7: 21-30PubMed Scopus (90) Google Scholar We did not obtain leucocyte counts, but all the infected animals were febrile. On the contrary, the animals treated with a rifampin-bonded graft were afebrile and healthy at the time of graft explantation. With regard to the identification of bacterial species, we used the commercially available standardized micromethod of conventional and chromographic tests to confirm that the S. epidermidis study strain was recovered. Moreover, because the study strain was resistant to streptomycin, backplating was done on triptycase soy agar containing streptomycin. We also verified the absence of contaminants. In summary, the differences in the clinical presentation of S. epidermidis graft infection between our study and that of Bergamini et al. 1Bergamini TM Bandyk DF Govostis D Kaebnick HW Towne JB. Infection of vascular prostheses caused by bacterial biofilms.J VASC SURG. 1988; 7: 21-30PubMed Scopus (90) Google Scholar probably involves the density of the bacterial inoculum. Our study demonstrates that antibiotic-bonded grafts may play a role in the treatment of acute graft infections. Additional studies with other bacterial species are surely warranted.