Angiology: Open Access
Open Access

Abstract

A Meta-Analysis of Outcomes of In Situ Reconstruction after Total or Partial Removal of Infected Abdominal Aortic Grafts?

Michel Batt, Fabrice Camou, Amandine Coffy, Patrick Feugier, Eric Senneville, Jocelyne Caillon, Brigitte Calvet, Christian Chidiac, Frederic Laurent, Matthieu Revest and Jean Pierre Daures

Objective: To determine the most appropriate indication for total resection (TR) or partial resection(PR) of infected aortic grafts.
Methods: A meta-analysis was conducted on the rates of early/late mortality, amputations and reinfection. A meta-regression was performed with 8 variables: patient age, male prevalence, presence of prosthetic duodenal fistula (PDF), presence of virulent or nonvirulent organisms, urgency, omentoplasty and follow-up.
Results: For TR and PR, the rates of early mortality and reinfection were 16.8% and 10.5%, 11% and 27% respectively.
For TR, the rates of early mortality and reinfection for in situ reconstruction with allografts and veins were 23% and 10.7%, 10.5% and 2.8% respectively.
For TR, urgency, male gender and omentoplasty all had a negative correlation with early mortality. Omentoplasty had a positive correlation with late mortality. Male gender, PDF and virulent organisms had a negative correlation with reinfection, whereas nonvirulent organisms had a positive correlation. Presence of PDF and follow-up had a negative correlation with amputation, whereas male gender had a positive correlation. For PR, no statistical correlation was analyzable except a negative correlation between PDF and reinfection.
Conclusion: For TR, operative mortality increases in males, and in emergencies, so total removal of the infected aortic graft seems not advisable, and alternative interventions should be considered. With regard to the risk of reinfection, TR appears recommendable in the presence of nonvirulent infectious organisms, but there is increased risk in the presence of virulent organisms and PDF, and so other optimal options may be discussed.
For PR results were not analyzable except for a negative correlation in the presence of PDF.