Surgical management of abdominal aortic graft infection: network meta-analysis

Abstract

Background A paucity of evidence exists regarding the optimal management for abdominal aortic graft infection. The aim of this paper was to assess short- and long-term outcomes following different surgical options in aortic graft infection patients. Methods Medline, Embase and the Cochrane Library were searched from inception to February 2023. Network meta-analysis was performed using a frequentist method. Patients were divided into four treatment groups: complete graft removal with in situ repair, complete graft removal with extra-anatomic repair, partial graft removal with in situ repair and partial graft removal with extra-anatomic repair. The mortality rate at 30-days and 1-year was the primary outcome. Secondary outcomes were longer-term mortality rate, primary patency and reinfections. For included RCTs, the Cochrane risk-of-bias tool was utilized to assess the risk of bias. The methodological quality of cohort studies was evaluated using the Newcastle–Ottawa scale. Results Among 4559 retrieved studies, 22 studies with 1118 patients (11 multi-arm and 11 single-arm studies) were included. Patients received complete graft removal with in situ repair (N = 852), partial graft removal with in situ repair (N = 36), complete graft removal with extra-anatomic repair (N = 228) and partial graft removal with extra-anatomic repair (N = 2). Both network meta-analysis results and pooled results of multi- and single-arm cohorts indicated that partial graft removal with in situ repair has the lowest 30-day and 1-year mortality rates (0% and 6.1% respectively), followed by complete graft removal with in situ repair (11.9% and 23.8% respectively) and complete graft removal with extra-anatomic repair (16.6% and 41.4% respectively). In addition, complete graft removal with in situ repair had a lower 3-year (complete graft removal with in situ repair versus complete graft removal with extra-anatomic repair: 32.1% versus 90%) and 5-year (complete graft removal with in situ repair versus complete graft removal with extra-anatomic repair: 45.6% versus 67.9%) mortality rate when compared with complete graft removal with extra-anatomic repair. Patients in the complete graft removal with in situ repair group had the lowest reinfections (8%), followed by partial graft removal with in situ repair (9.3%) and complete graft removal with extra-anatomic repair (22.4%). Conclusion Partial graft removal with in situ repair was associated with lower 30-day and 1-year mortality rates when compared with complete graft removal with in situ repair and complete graft removal with extra-anatomic repair. Partial graft removal with in situ repair might be a feasible treatment for specific aortic graft infection patients.