Reinfection after resection and revascularization of infected infrarenal abdominal aortic grafts

Abstract

Objective: Despite advances in perioperative care, long-term and amputation-free survival rates are poor after resection of infected abdominal aortic grafts. We reviewed our cases to determine the rate of reinfection and risk factors for mortality and limb loss.

Methods: We reviewed cases with infrarenal aortic graft infection from 1999 to 2013. Cases requiring graft excision were included for analysis. Thoracic and thoracoabdominal aortic grafts were excluded. Reconstruction types included both extra-anatomic and in situ grafts. Patient comorbidities, surgical outcomes, and known reinfection rates were assessed. Univariate and Kaplan-Meier analysis were performed.

Results: Twenty-eight patients had resection of infected infrarenal abdominal aortic grafts during the study period. Most patients (26/28; 93%) had infected aortoiliac or aortofemoral prosthetic bypass grafts, but two of 28 patients had infected endovascular aortoiliac stent grafts. The median age was 69 years (range, 46-86 years), with 68% men and 32% women. Aortoenteric fistulae or graft-enteric erosions were noted in 12 of 28 (43%) patients at operation. There were 79% of patients who had in situ reconstruction, including 4 (14%) with polyester, 1 (4%) with polytetrafluoroethylene, 3 (11%) with cadaveric homograft, 3 (11%) with composite grafts, and 11 (39%) with native femoropopliteal vein grafts. Five (18%) patients had extra-anatomic bypass and one had excision without revascularization. In-hospital mortality after initial graft excision and revascularization occurred in two (7%) patients. Seven (25%) patients had evidence of reinfection after a median of 20 months, of whom five underwent reintervention with two additional in-hospital deaths. All in-hospital deaths occurred in patients with graft-enteric contamination. Overall limb salvage and survival at a mean follow-up of 2.5 years were 82% and 46%, respectively, and did not differ among revascularization types (P = .85 and .74). One-year amputation-free survival was 47% overall. Three patients with native femoropopliteal vein graft repair required amputation in follow-up. Diabetes was the only observed risk factor for amputation (P = .05). Risks for mortality included history of cerebrovascular disease (P = .05) and shock on presentation (P = .04). No other comorbid condition, type of revascularization, or perioperative complication was associated with limb loss or mortality on univariate analysis.

Conclusions: Revascularization after excision of infected abdominal aortic grafts can be done with acceptable in-hospital morbidity and mortality. Reinfection is problematic, regardless of revascularization conduit, and is associated with limb loss and death. New and aggressive local anti-infective strategies are warranted.