The relevance of aortic endograft prosthetic infection.

BACKGROUND: Vascular prosthetic graft infection is a severe complication after open aortic aneurysm repair. Reports of infected endografts are scarce. General treatment consensus with infected graft material is that it should be removed completely. The objective of this study was to describe the incidence of endograft infection after endovascular repair of abdominal (EVAR) and thoracic aortic aneurysm (TEVAR) and to report treatment options and their outcome.

METHODS: A retrospective cohort study was performed of patients endovascularly operated for abdominal and thoracic aortic aneurysm in two large hospitals (one tertiary referral center and one large community hospital) between March 1996 and June 2009. Diagnosis of infected endograft was made based on clinical findings, blood tests and cultures, imaging studies (computed tomography, fludeoxyglucose positron emission tomography), and intraoperative findings at reoperation.

RESULTS: Eleven patients with an infected endograft were identified in 1431 endovascular procedures. One other patient was referred from another hospital. Patients were aged 68 ± 9 years, and all but one were male. The median time from initial TEVAR/EVAR to the diagnosis of infection was 115 days (range, 7-3748 days), with 42% of patients presenting within 3 months after TEVAR/EVAR. Seven patients were diagnosed with endograft infection after elective TEVAR/EVAR and five after emergency TEVAR/EVAR. The incidence was significantly higher in patients that were treated in an emergency setting (0.56% vs 2.79%; P = .002), while there was no significant difference between TEVAR and EVAR procedures (1.37% vs 0.77%). All patients were initially treated with antibiotic therapy, which was complemented with surgical intervention in six patients. In four patients, the infected graft material was completely explanted. Isolated microorganisms included Staphylococcus species (n = 4), Streptococcus species (n = 4), Enterobacter cloacae (n = 1), Escherichia coli (n = 1), Pseudomonas aeruginosa (n = 1), and Listeria monocytogenes (n = 1). Median time of follow-up was 201 days (range, 6-2023 days). During the study period, three out of 12 patients died, of which two were treated conservatively (P = ns). At their last follow-up visit, seven of nine patients still used antimicrobial therapy.

CONCLUSIONS: The incidence of endograft infection is below 1%, with a mortality rate of 25%. Although consensus is that infected graft material should always be removed, this study shows no significant difference in mortality between the conservatively- and the surgically-managed group, possibly related to the small sample size. There may be a role for conservative treatment in selected cases of patients with an infected endograft.