Does bacteremia pose a direct threat to synthetic vascular grafts?

ABSTRACT

This study was undertaken to determine the significance of graft lumen exposure to blood-borne organisms in the development of graft infection. Three groups of dogs were studied. In group I (n = 20), the infrarenal aorta was dissected from surrounding tissue, divided, and reconstructed with a Dacron tube interposition graft. In group II (n = 9) the aorta was similarly isolated, but Dacron graft material was wrapped around the intact aorta. In group III (n = 13) the infrarenal aorta was isolated, but no graft material was placed. All dogs were given intravenous 1 X 10(7) Staphylococcus aureus at the completion of surgery. Group I grafts were harvested 8 hours, 1 day, or 21 days after bacterial challenge. Group II and III grafts were harvested 1 day or 21 days after infusion. At the time of harvest, selective cultures of the periaortic tissue (PAT), periaortic graft (PAG), and interposition graft lumen (GL) were taken. The overall infection rates were similar, with 17 of 20 (85%) dogs in group I, 6 of 9 (67%) in group II, and 11 of 13 (85%) in group III found to be culture positive. In group I, 3 of 3 dogs at 8 hours, 2 of 2 on day 1, and 12 of 15 on day 21 had positive PAT cultures. Only 4 of 15 on day 21 had positive GL cultures. In group II, 4 of 5 dogs on day 1 and 2 of 4 on day 21 had positive PAT and PAG cultures. In group III, 9 of 9 animals on day 1 and 2 of 4 on day 21 had positive PAT cultures. All aortic lumen cultures were negative in groups II and III. The difference between GL and PAT cultures was statistically significant in all groups (I, p = 0.01; II, p = 0.05; III, p = 0.01). Serial quantitative blood cultures revealed a mean bacterial load of 10.5 +/- 4.5 CFU/ml at 15 minutes postinfusion, which fell steadily until no bacteria were detected at 3.5 hours. Lymphangiography demonstrated periaortic pooling of lymph in the immediate postoperative period. These data suggest that the bacteremia in this model is transient and rapidly clears. Periaortic tissues quickly sequester bacteria, possibly because of lymphatic leakage. The GL appears to be secondarily infected.