Purpose: Newer approaches to the patient with an infected aortic graft are available. We reviewed a recent 10-year experience with a more traditional approach to evaluate its outcome in the 1990s.
Methods: From January 1983 to January 1993, 27 patients with an aortic graft infection were treated at our institution. There were 18 paraprosthetic infections, eight graft enteric erosions, and one aortoduodenal fistula. The involved bypasses included 20 aortofemoral (74%), five aortoiliac (18%), and two aortic tube grafts (8%). Nineteen aortic replacements were done originally for aneurysmal disease (70%). We reviewed the outcome of each patient treated as it related to the method of management. The therapy for graft infection consisted of aortic graft removal and axillofemoral bypass in 20 patients (74%), treatment by an in situ method in four patients (15%), excision of an aortofemoral limb and extraanatomic bypass in two patients (7%), and extraanatomic bypass alone in one patient (4%). In the group treated by graft removal and extraanatomic bypass, four patients (20%) had staged operations (extraanatomic bypass followed by interval aortic graft removal), nine (45%) had single operations with extraanatomic bypass preceding graft removal, and seven (35%) had single operations with graft removal preceding extraanatomic bypass.
Results: The 30-day operative mortality rate was 3.7%. There were no instances of aortic stump blowout. The 3-year primary patency rate for axillofemoral bypass limbs was 80.2%, and the secondary patency rate was 87.4%. No limbs were lost as a result of ischemic complications. There was one late amputation for an unrelated problem.
Conclusions: The results of alternate approaches to the management of patients with infected aortic grafts were equivalent both in terms of perioperative mortality and morbidity rates in this group of patients. Complete excision of the aortic graft with axillofemoral bypass provided a satisfactory long-term outcome and remains the standard with which other approaches must be compared.