Vascular prosthetic graft infection: epidemiology, bacteriology, pathogenesis and treatment

Vascular prosthetic graft infection remains a major surgical challenge. Prevention of risk factors and antibiotic therapy can reduced but not eradicate it. Management of infected vascular grafts depends on several factors, including the location of the infected prosthesis, the extent of infection, and the underlying micro-organism. Classic treatment consists of extra-anatomic bypass grafting. The disappointing results due to the high mortality and amputation rate have kindled interest in alternative approaches, such as in situ reconstruction with antibiotic-bonded prostheses, autogenous veins or arterial allografts.

We focused on the treatment of aortic graft infection by means of both fresh and cryopreserved arterial allograft. Here, the experience of the Italian Collaborative Vascular Homograft Group is reported.

Between March 1994 and December 2000 seventy-nine patients with aortic graft infection were treated. The results of 68 patients are analysed. Eleven patients were treated with fresh, and 57 with cryopreserved homograft. Emergency surgical procedures were performed in 12 patients (17%). Aortoenteric fistula was diagnosed in 22 patients. The mean interval between the first procedure and the insertion of a homograft for patients with infected aortic graft was 3 years (range 1-15). The mean duration of follow-up was 30 months (range 1-68). Clinical and duplex scanning evaluation were routinely performed. Computer tomography (CT), magnetic resonance (MR), or arteriography were performed on the basis of duplex scanning results.

The analysis was performed on 68 cases for which there were sufficient reliable data. Eleven deaths occurred during the early postoperative period (30 days), a mortality rate of 16%. There were also seventeen late deaths, a mortality rate of 25%. Eleven patients had graft occlusion; six cases were successfully treated with thrombectomy. In three cases leg amputation was necessary. The results of fresh and cryopreserved homografts were compared. No significant differences of early postoperative mortality, late mortality, homograft-related mortality, graft failure were observed. The presence of aortoenteric fistula is a negative predicting factor of perioperative early mortality, which causes a rapid decline in the survival curve. Thirty-six months after the surgery the actuarial survival of the patients was 57% and the actuarial patency of the allograft was 41%.

No significant difference in terms of clinical outcome was observed when using fresh, rather than cryopreserved homografts. The only factor that significantly influenced the survival rate appeared to be the aorto-enteric fistula.