From the Journals

Gastrointestinal complications are a key risk factor for death after AEF repair

View on the News

Manage the GI component to aid in success

While secondary aortoenteric fistulas fortunately are a rare occurrence after open aortic repair (less than 1%), the reported results of treatment are disappointing (up to 75% mortality). The authors have demonstrated significant improvement in overall mortality in their series by paying more attention to the GI component of the procedure and postoperative management of the patient.

Dr. Murray L. Shames  is professor of surgery and radiology and chief of the division of vascular surgery at the University of South Florida, Tampa, and director of the Tampa General Hospital aortic program.

Dr. Murray L. Shames

Our group also has found the management of the GI component to be an important predictor of success. We recently reviewed our own experience with AEFs between 2002 and 2015. Of the 44 patients treated, 80% had extra-anatomic reconstruction and aortic ligation with 51% of the patients having a single-staged procedure. Our overall in-hospital mortality was 30%. Vascular surgeons performed the GI reconstruction in 61% of cases (56% primary repair). GI surgeons performed more complex or multisegment repairs 67% of the time. GI complications occurred in 30% of cases when vascular surgeons performed the repair, and 18% when the repair was performed by GI surgeons. As in the current report, we identified that a GI complication significantly increased the risk of mortality. This finding is of particular importance as we graduate integrated residents into the workforce, a multidisciplinary approach to management of AEFs is critical to improved outcomes.

When we published our institutions experience with AEFs from 1991 to 2004, the study included 29 patients with one-third of patients presenting in shock. Extra-anatomic repair was performed in 86% (68% single stage). Mortality was 24%, with shock, blood transfusions, and suprarenal clamping associated with a worse outcome. Our 5-year survival was 61% and freedom from recurrent infection or amputation was 86% and 88% respectively. While our experience favors an extra-anatomic reconstruction, there has been increased utilization of in-situ repair in recent years. The current authors have extensive experience with the Neoaortoiliac System (NAIS) procedure and have had minimal complications associated with the extensive vein harvest necessary for NAIS; however, this is not universal. Wound complications, compartment syndrome, and chronic venous insufficiency are not uncommon complications. Improved results using cryopreserved aortoiliac allografts (CAA) were reported by The Vascular Low-Frequency Disease Consortium. In a 2014 publication on 220 patients receiving CAA for aortic reconstruction for aortic graft infection, freedom from graft-related complications, graft explant, and limb loss was 80%, 88%, and 97%, respectively, at 5 years. This compares favorably with extra-anatomic reconstruction and other techniques. In suprarenal reconstructions, allografts have been favored since the natural branches can be used to simplify multiple vessel reconstruction. In cases of low-grade infection (Staphylococcus epidermidis), excellent results have been demonstrated with wide debridement, in-situ replacement with rifampin-bonded grafts and omental coverage.

AEF can also occur after endovascular abdominal aortic aneurysm (AAA) repair and a management strategy for removing the endograft should be considered in the planning of these cases. The authors describe their technique, which involves constraining the endograft using an umbilical tape. Another technique that I prefer uses a 20-cc syringe with the tip cut off. The endograft is constrained by advancing the syringe proximally over the endograft. The syringe collapses the device and can be used to constrain the proximal fixation stents of a suprarenal graft, simplifying removal of the stent graft. Infected EVAR will likely become more common as an etiology as we continue to expand EVAR utilization for AAA repair. Another Low-Frequency Consortium paper reviewed the treatment and outcome of EVAR infection. In that report of 206 infected EVAR (endovascular aneurysm repair) and TEVAR (thoracic endovascular aortic repair) patients, 90% had in situ replacement with a 30-day mortality of 11% and morbidity of 35%.

As in previous reports on AEF, the authors highlight the importance of high clinical suspicion in making an expeditious diagnosis. Many imaging modalities can be used, but often operative exploration is required for a definitive diagnosis. Complete graft excision and wide debridement are critical to minimize the risk of recurrent infection. Optimal revascularization techniques should be determined by the experience of the operator, current experience demonstrating nearly equivalent outcomes with extra-anatomic and in situ replacement. Careful GI reconstruction and post-operative nutrition, culture-specific antibiotics, and ICU care are likely more important than the mode of reconstruction. Lifelong surveillance to detect recurrent infections also is recommended.

Murray L. Shames, MD , is professor of surgery and radiology and chief of the division of vascular surgery at the University of South Florida, Tampa, and director of the Tampa General Hospital aortic program.



Aortoenteric fistulas (AEFs) are an uncommon but lethal form of aortic graft infection with morbidity and mortality rates reported in the literature to range from 14% to 75%. Over a 20-year period, researchers found that nearly half of their patients undergoing repair of their aortoenteric fistulas died within 60 days. The presence of gastrointestinal complications increased the risk of mortality more than threefold, according to the results of a single-center retrospective review of consecutive AEF repairs.

The researchers assessed 50 patients who presented with AEF and had repair during 1995-2014. Sixty percent of the patients were men, and the overall median age was 70 years. The median follow-up for the entire cohort was 14 months. The duodenum was the most common location of the enteric defect, found in 80% of the infections. Overall, 23 patients (46%) died by day 60, according to the report published in the July Journal of the American College of Surgeons.

Univariate analysis showed that advanced age, chronic renal insufficiency, any complications, and GI complications in particular (occurring in 26% of patients) were all associated with an increase in overall mortality (P less than .05). But upon multivariate analysis, gastrointestinal complications (hazard ratio, 3.23; P = .015) and advanced age (HR, 1.07; P = .01) were the only independent predictors of mortality, Atish Chopra, MD, of the division of vascular surgery, Oregon Health & Science University, Portland, and his colleagues wrote.

The institution changed operative procedures in 2007, based upon an earlier assessment of the importance of GI complications performed by the researchers, with greater emphasis placed on ensuring a viable GI reconstruction, and early intervention for mesenteric ischemia. In addition, they surmised that, after 2007, there was improved adherence to achieving wide debridement of nonviable and infected tissue, and to creating a tension-free anastomosis to healthy tissue edges while optimizing nutritional, medical, and antibiotic therapy, according to the researchers.


Recommended Reading

Some measures to control HAI sound better than they perform
MDedge Surgery
Strict OR attire policy had no impact on SSI rate
MDedge Surgery
Type of headwear worn during surgery had no impact on SSI rates
MDedge Surgery
Single-port laparoscopy has few complications but BMI matters
MDedge Surgery
Potential postthyroidectomy quality improvement metrics arise from study
MDedge Surgery