To compare the in-hospital mortality rate of the surgical management of endovascular graft infections (iEVAR) and aortic graft infections (iOAR), in patients initially treated for infra-renal aortic abdominal aneurysm.
A retrospective monocentric review was conducted including consecutive patients with infra-renal iEVAR or iOAR, who underwent graft explantation and revascularization from September 2006 to December 2019. The surgical method always consisted of local debridement and in situ reconstruction using a cryopreserved aortic allograft. The primary outcome was the in-hospital mortality. Secondary outcomes were 30-day mortality, overall survival, postoperative morbidity and rate of secondary rupture of the vascular substitute. Survival rates were determined by the Kaplan-Meier method. Univariate analysis attempted to identify mortality predictors. A Propensity score was performed to cope with the initial EVAR vs OAR possible indication bias, and average treatment effect was estimated.
During the study period, 41 patients underwent explantation of an aortic graft infection (20 iEVAR cases and 21 iOAR cases) with in situ reconstruction. The graft-related early mortality rate was 48,7% (55% for iEVAR group and 43% for iOAR group (p=.43). Risk factors for in-hospital mortality were : Age > 65 years (OR 5.54 CI 95% 1.01 - 30.49),the presence of aorto-enteric fistula (OR 4.25 CI 95% 1.05 - 17.20), a short delay between the operative decision and removal of infected material (OR 0.04 CI 95% 0.00-0.5), the occurrence of postoperative revision (OR CI 9.6 95% 2.32-39.95), the number of postoperative revisions (for one revision, OR 11.2 CI 95% 1.73-72.3; for 2 revisions OR 6.40 CI 95% 0.89-45.99; for >2 revisions OR 12.80 CI 95% 1.15-142.58), the occurrence of major postoperative complication (OR 17.10 CI 95% 1.89-154.84). Based on the propensity score and the calculated average treatment effect, if all the patients had received EVAR, the proportion of in-hospital deaths would be increased by 0.20 [CI 95%: -0.09; 0.48] (p=.48).
The EVAR infection does not appear to add excess mortality compared to infections of conventional grafts. This specific risk does not appear to be an initial decision-making criterion. With a propensity score taking into account the initial indication bias EVAR vs OAR, we show a strong but non-significant trend towards excess iEVAR mortality.