Treatment and Outcomes of Aortic Graft Infections Using a Decision Algorithm.

Background Aortic graft infection (AGI) is a rare but devastating complication requiring both explant of the infected prosthesis and lower extremity revascularization. Despite a variety of methods to treat AGI, there is a paucity of evidence that describes comparative outcomes. Moreover, controversy exists surrounding what the optimal repair strategy is with limited descriptions of how these techniques should be employed in this complex group of patients. Therefore, the purpose of this analysis was to review our experience with AGI management while highlighting a practice philosophy that can achieve acceptable outcomes. Methods All AGI patients between 2002-2019 were reviewed. The primary end-point was 30-day mortality. Secondary end-points included complications, re-infection, unplanned re-operation and all-cause mortality. Kaplan-Meier methodology was used to estimate time to events. Cox regression models were employed to identify association between patient factors and operative strategy with survival. Subgroup analysis included outcome comparison among four different operative approaches(extra-anatomic bypass with aortic ligation [EAB] and in-situ reconstruction [ISR] using either NAIS, cryopreserved allograft [Cryo], or antibiotic-soaked prosthetic grafts [Other]). Results 142 patients (male-69%, mean age 67 ± 11 years) were reviewed. Median time to AGI presentation was 52 (IQR 16-128) months. ISR was performed in 70% (n = 99)[ISR: NAIS-49% (n = 49), Cryo, 33% (n = 33) and Other-23% (n = 23)]. EAB was used in 26% (n = 37), of which 57% (n = 21) were staged repairs[no reconstruction, 4%: intraoperative death-2, AGI removal without reconstruction-2]. A graft enteric erosion/fistula was identified in 39% (n = 55). Mean follow-up time was 14 ± 27 (median 2.2[IQR 0.1-16]) months. Overall, 30-day mortality was 21% and 69% (n = 98) experienced a complication. The most common complications were pulmonary (35%;n = 50), vascular (28%;n = 39), gastrointestinal (22%;n = 31) and renal (21%;n = 30). Freedom from re-infection at one and three years was 78 ± 5% and 73 ± 6% while freedom from unplanned re-operation was 50 ± 5% and 40 ± 6%, respectively. Corresponding one- and five-year freedom from all-cause mortality was 67 ± 4% and 53 ± 4%. When stratified by the four different repair strategies, unadjusted rates of postoperative complications and mortality were not different. However, EAB patients had more renal complications. All-cause mortality predictors included age (HR 1.04, 95%CI 1.01–1.1; P = 0.003), CHF (HR 2.7, 1.3–5.7; P = 0.01), and graft enteric erosion/fistula (HR 2.2, 1.3–3.8;P = 0.005) while total graft excision was protective (HR 0.34, 0.2–0.7; P = 0.003). Conclusions AGI repair, regardless of operative strategy, results in significant early morbidity, and mortality. The need for unplanned re-operation is common; however, long-term survival is acceptable in appropriately selected patients. Re-infection risk mandates life-long surveillance and consideration of indefinite anti-microbial suppression in certain subgroups. Due to the complexity and intensity of care, all AGI should be treated, when possible, at centers performing high-volume aortic surgery.