Improved perioperative mortality after secondary aortoenteric fistula repair and lessons learned from a 20-year experience.

BACKGROUND: Secondary aortoenteric fistulas (SAEFs) are rare but represent one of the most challenging and devastating problems for vascular surgeons. Several issues surrounding SAEF treatment remain unresolved, including optimal surgical reconstruction and conduit choice. We performed an audit of our experience with SAEFs and highlight aspects of care that have affected outcomes over time with the intent to identify factors associated with best outcomes. METHODS: We performed a single center, retrospective review of all consecutive SAEF repairs (1999-2019), defined as presence of a false communication between an enteric structure and pre-existing aortic graft. The primary endpoint was 30-day mortality. Secondary endpoints included incidence of complications and overall survival. Time-dependent outcome comparison was performed. Cox proportional hazards modeling and life-table analysis estimated risk and freedom from endpoints. RESULTS: A total of 57 patients (63% male; n = 36) presented with SAEF (median age, 69 years; interquartile range [IQR], 61-74 years). Median follow-up time was 10 months (interquartile range, 3-21 months. The most common presenting symptoms were gastrointestinal bleeding (60%; n = 34) and abdominal pain (56%; n= 3 2). For the overall cohort, 30% (n = 17) underwent extra-anatomic bypass with aortic ligation, 30% (n = 17) rifampin-soaked Dacron graft, 26% (n = 15) femoral vein (eg, neoaortoiliac system), and 14% (n = 8) cryopreserved aortic allograft. The enteric communication involved the duodenum in 85% (n = 48), and a double-layer hand-sewn primary repair was most commonly employed (61%; n = 35). Thirty-day mortality was 35% (n = 20) with no significant difference between 90 days (39%; n = 22) and 180 days (42%; n = 24). Morbidity was 70% (n = 40), with gastrointestinal (30%; n = 17; leak [9%]), pulmonary (25%; n = 14), and renal (21%) complications being most common. Incidence of reoperation for any vascular and/or gastrointestinal-related complication was 56% (n = 32). One-year and 3-year survival was 54% ± 6% and 48% ± 8%, respectively. Over time, 30- and 90-day mortality improved (odds ratio, 0.1; 95% confidence interval, 0.4-0.5; P = .002) despite no change in patient factors, operative strategy, conduit choice, or morbidity rate. Prehospital history of gastrointestinal bleeding was associated with worse survival (hazard ratio, 2.0; 95% confidence interval, 1.0-3.9; P = .06); however, reconstruction strategy (in-situ vs extra-anatomic bypass), postoperative gastrointestinal and/or vascular complication, omental flap use, and preoperative endovascular aneurysm repair history were not associated with outcome. CONCLUSIONS: In conclusion, we observed improved short-term mortality despite no significant change in patient presentation or postoperative complications. This highlights increasing institutional experience in selecting the optimal surgical strategy and improved ability to rescue patients experiencing adverse postoperative events. An individualized approach to reconstruction and conduit choice can lead to best outcomes after SAEF management when patients are treated at a high-volume aortic surgery center.

Aortoduodenal fistulas after endovascular abdominal aortic aneurysm repair and open aortic repair.

OBJECTIVE: In the present study, we have reported and compared aortoduodenal fistulas (ADFs) after endovascular abdominal aortic aneurysm repair (EVAR) vs after open aortic repair (OAR). METHODS: We retrospectively analyzed the data from patients treated for ADFs from January 2015 to May 2020 in our hospital. The clinical data, diagnostic procedures, and surgical options were evaluated. The primary endpoints of the present study were 30-day and 1-year mortality. The secondary endpoints were major postoperative complications. RESULTS: A total of 24 patients (20 men; median age, 69 years; range, 53-82 years) were admitted with ADFs after EVAR (n = 9) or OAR (n = 15). These patients accounted for ∼4.3% of all abdominal aortic aneurysm repairs in our hospital. The median interval from the initial aortic repair and the diagnosis of ADF was 68 months (range, 6-83 months) for the ADF-EVAR group and 80 months (range, 1-479 months) for the ADF-OAR group. Three patients in the ADF-EVAR group had refused surgical treatment owing to their high surgical risk. One patient in the ADF-OAR group had undergone removal of the aortic prosthesis without replacement. Of the remaining 20 patients, 12 (ADF-EVAR group, n = 4; ADF-OAR group, n = 8) had undergone in situ replacement of the aorta and 8 (ADF-EVAR group, n = 2; ADF-OAR group, n = 6) had undergone extra-anatomic reconstruction with aortic ligation. After a mean follow-up of 26 months, no patient had experienced early limb loss. However, one case of rupture of the venous graft (ADF-EVAR), one case of aortic stump blowout (ADF-OAR), and one case of a ureteroarterial fistula with a homograft (ADF-OAR) had occurred. Overall, the incidence of postoperative complications was significantly greater after ADF-OAR (93% vs 33%; P = .036). The most frequent bacteria involved in the blood cultures were Escherichia coli (25% of patients), and Candida spp. (61%) were the predominant pathogens found on intra-abdominal smears. The in-hospital mortality rates for the ADF-EVAR and ADF-OAR group were 22% and 13%, respectively. The corresponding 1 -year mortality rates were 22% and 33%. CONCLUSIONS: Patients with ADFs after EVAR or OAR have limited overall survival. In addition to the similar therapeutic approaches, we found no significant differences in postoperative mortality between these two uncommon pathologic entities. In our study, the overall postoperative morbidity seemed greater for the ADF-OAR group.

Evaluation of Endovascular Intervention for Tracheo-Innominate Artery Fistula: A Systematic Review.

BACKGROUND: While the standard of care for suspected tracheo-innominate artery fistula (TIF) necessitates sternotomy, perioperative mortality remains high. Endovascular interventions have been attempted, but reports have been anecdotal. The aim of this systematic review was to evaluate the outcomes of endovascular management of TIF by pooling the existing evidence. METHODS: An electronic database search of Ovid MEDLINE, Scopus, Cumulative Index to Nursing and Allied Health Literature, and Cochrane Controlled Trials Register was performed to identify all studies examining endovascular treatment of TIF. Patients greater than 14 years of age who underwent endovascular intervention for TIF were included. 25 studies consisting of 27 patients met the inclusion criteria. RESULTS: 48.1% (13/27) of patients were male and median age was 39.0 [IQR 16.0, 47.5] years. Tracheostomy was present in 96.3% (26/27) of cases. Median duration from tracheostomy to TIF presentation was 2.2 months [0.5, 42.5]. On presentation, 84.6% (22/26) had tracheal hemorrhage, and 22.8% (6/27) were hemodynamically unstable. 96.3% (26/27) underwent covered stent graft placement while 1 patient (3.8%) had coil embolization. 18.5% (5/27) of patients required repeat endovascular intervention for recurrent bleeding, while 11.1% (3/27) required rescue sternotomy. Median hospital length of stay was 30 days [16.0, 46.5], and overall mortality was 29.6% (8/27) with a median follow-up time of 5 months [1.2, 11.5]. CONCLUSION: While uncommon, endovascular treatment of TIF may be a feasible alternative to sternotomy. The approach may be useful in those who are unable to undergo surgery or are likely to have adhesions from prior chest operations.

First Result of a Tailored Progressive Multistep Approach for the Treatment of Aorto-esophageal Fistulae.

BACKGROUND: Aorto-esophageal fistulae (AEFs) are a rare but serious and life-threatening disease of the mediastinum. Especially, AEF in the presence of infected stent grafts, for example, after thoracic endovascular aortic repair (TEVAR) is only curable by a multistage interdisciplinary surgical approach. This study presents the results of our four-stage approach consisting of bridging TEVAR, esophagectomy, complete stent removal followed by total bovine tube aortic replacement (TBTAR), and finally esophageal reconstruction. METHODS: A case series of four patients from our department receiving a four-stage treatment of AEF is presented in this study. Retrospective database analysis focusing on overall survival, duration of intensive care unit and total hospital stay until discharge, complications, surgical time frame, and completion of chosen surgical treatment course was performed. RESULTS: Overall, four patients surgically treated for AEF since May 2015 were included. A 30-day mortality was 0%, and overall survival at 1 year was 75%. All patients survived more than 5 months and could be discharged after TEVAR and esophagectomy. TBTAR could be performed in two of four patients (50%). Esophageal reconstruction was completed in all patients. Average follow-up was 20.3 ± 1.7 months or until death. CONCLUSION: The acute management of AEF using this approach seems satisfactory, especially for reducing acute short-term mortality. Complete restoration of the circulatory system and digestive tract remains challenging and is associated with high morbidity. We support the application of bridging TEVAR with a staggered approach to further surgical treatment individually tailored to the patient.

In situ bypass and extra-anatomic bypass procedures result in similar survival in patients with secondary aortoenteric fistulas.

OBJECTIVE: The optimal revascularization modality in secondary aortoenteric fistula (SAEF) remains unclear in the literature. The purpose of this investigation was to determine the revascularization approach associated with the lowest morbidity and mortality using real-world data in patients with SAEF. METHODS: A retrospective, multi-institutional study of SAEF from 2002 to 2014 was performed using a standardized database. Baseline demographics, comorbidities, and operative and postoperative variables were recorded. The primary outcome was long-term mortality. Descriptive statistics, Kaplan-Meier survival analysis, and univariate and multivariate analyses were performed. RESULTS: During the study period, 182 patients at 34 institutions from 11 countries presented with SAEF (median age, 72 years; 79% male). The initial aortic procedures that resulted in SAEF were 138 surgical grafts (76%) and 42 endografts (23%), with 2 unknown; 102 of the SAEFs (56%) underwent complete excision of infected aortic graft material, followed by in situ (in-line) bypass (ISB), including antibiotic-soaked prosthetic graft (53), autogenous femoral vein (neoaortoiliac surgery; 17), cryopreserved allograft (28), and untreated prosthetic grafts (4). There were 80 patients (44%) who underwent extra-anatomic bypass (EAB) with infected graft excision. Overall median Kaplan-Meier estimated survival was 319 days (interquartile range, 20-2410 days). Stratified by EAB vs ISB, there was no significant difference in Kaplan-Meier estimated survival (P = .82). In comparing EAB vs ISB, EAB patients were older (74 vs 70 years; P = .01), had less operative hemorrhage (1200 mL vs 2000 mL; P = .04), were more likely to initiate dialysis within 30 days postoperatively (15% vs 5%; P = .02), and were less likely to experience aorta-related hemorrhage within 30 days postoperatively (3% aortic stump dehiscence vs 11% anastomotic rupture; P = .03). There were otherwise no significant differences in presentation, comorbidities, and intraoperative or postoperative variables. Multivariable Cox regression showed that the duration of antibiotic use (hazard ratio, 0.92; 95% confidence interval, 0.86-0.98; P = .01) and rifampin use at time of discharge (hazard ratio, 0.20; 95% confidence interval, 0.05-0.86; P = .03) independently decreased mortality. CONCLUSIONS: These data suggest that ISB does not offer a survival advantage compared with EAB and does not decrease the risk of postoperative aorta-related hemorrhage. After repair, <50% of SAEF patients survive 10 months. Each week of antibiotic use decreases mortality by 8%. Further study with risk modeling is imperative for this population.

Midterm Results after Open Surgical and Endovascular Management of Arterioureteral Fistula.

BACKGROUND: Arterioureteral fistula refers to the anomalous fistulous connection between the iliac artery and the ureter. It is often associated with pelvic malignancy, abdominal surgery, and radiation. As it is a potentially life-threatening condition, prompt diagnosis and management is essential. METHODS: We performed a retrospective analysis of patients treated for arterioureteral fistula in a single-vascular institution from January 2013 to March 2019. Preoperative assessment included physical and laboratory examinations and medical history, with diagnosis established through computed tomography angiography, digital subtraction angiography, or ureteroscopy. Parameters analyzed included perioperative mortality and morbidity as well as treatment durability during midterm follow-up. RESULTS: Nine patients with ten arterioureteral fistulas were included in the study. Macroscopic hematuria was the main presenting symptom, with 2 patients admitted due to hemorrhagic shock. Endovascular treatment was carried out in 6 patients. In 4 cases, single stent-graft deployment inside the common iliac artery was performed, in one case in combination with plugging of the internal iliac artery. One patient underwent implantation of an iliac-branched device, whereas in another patient coiling of the internal iliac artery sufficed for management of the fistula. Open surgical repair was carried out in three cases. Perioperative mortality was zero; one patient had prolonged hospital stay due to superficial wound infection. Recurrent hematuria and stent-graft infection were observed during follow-up in three patients after endovascular repair, all of them treated through open surgery with no further complications. One patient developed an enterocutaneous fistula after open repair during follow-up and required redo surgery. DISCUSSION: Arterioureteral fistula is a challenging clinical scenario demanding prompt diagnosis and management. Open surgery remains the treatment of choice in cases of preexisting vascular reconstruction or manifest infection. Endovascular techniques offer a viable solution in significantly comorbid patients or in patients presenting with acute, life-threatening bleeding. Rigorous follow-up is required regardless of treatment modality due to the considerable rate of reinterventions.

Management of Tumor- and Nontumor-related Aorto-esophageal and Aorto-bronchial Fistulas.

BACKGROUND: This study aims to review and compare the clinical presentation, management, and outcome in patients with tumor-related (TR) and nontumor-related (NTR) aorto-esophageal fistula (AEF) and aorto-bronchial fistula (ABF) with particular focus on the thoracic endovascular aortic repair. METHODS: We retrospectively reviewed a series of 16 consecutive patients with TR (n = 8) and NTR (n = 8), ABF (n = 6), and AEF (n = 10) admitted to our hospital from 2011 to 2019. RESULTS: The median age was 62 years (range 46-81), with 11 men. The most common predisposing factor was esophageal or gastric cardia cancer (n = 6), followed by open repair of the thoracic aorta (n = 5). Endoluminal vacuum therapy (Endo-SPONGE®) accounted for 3 cases of AEFs. Thoracic endovascular aortic repair (TEVAR) was applied in 13 patients (4 with ABFs and 9 with AEFs). The primary technical success of the TEVARs was 100%. One patient (8%) was complicated with postoperative middle cerebral artery syndrome and left-sided hemiparesis. The respective in-hospital, 6-month, and 1-year mortality rates were 0% (n = 0), 25% (n = 2), and 25% (n = 2) for the NTR group and 63% (n = 5), 88% (n = 7), and 100% (n = 8) for the TR group. After a mean period of 13 months, 5 (31%) patients were still alive, and one patient lost to follow-up after 11 months. The survivors (n = 5) had all nontumor-related ABF. Progression of underlying cancer and hemodynamic shock were the most common causes of death. CONCLUSIONS: TEVAR represents a reliable option in the treatment of NTR ABFs. In the cases of TR fistulas and NTR AEFs, TEVAR should be applied more selectively. The associated mortality remains very high.

Endovascular and open surgical options in the treatment of uretero-arterial fistulas.

OBJECTIVE: To report and analyze the indications and results of endovascular and open surgical treatment for uretero-arterial fistula. METHODS: We retrospectively reviewed the clinical data of 25 consecutive patients with uretero-arterial fistulas admitted to our hospital from 2011 to 2020. Endpoints were technical success, freedom from open conversion, stent-graft/graft-related complications, and 30-day and one-year mortality. RESULTS: The study included 25 patients (68% female, n = 17) with 27 uretero-arterial fistulas by bilateral pathologies in two patients. The mean age was 61 ± 11 years (range 35-80). The most common predisposing factors for uretero-arterial fistula were history of pelvic operations for malignancy in 21 patients (84%), radiotherapy in 21 patients (84%), previous pelvic vascular bypass in 2 patients (8%), and iliac aneurysms in 2 patients (8%). On average, the period between the primary pelvic surgery and the diagnosis of uretero-arterial fistulas was 46 months (range 7-255). Twenty patients (80%) underwent endovascular treatment of the uretero-arterial fistulas. The primary technical success of the endovascular treatment was 95%, and the freedom from open conversion was 40% at six months and 30% at one year. Thirteen uretero-arterial fistulas (48%) underwent delayed open conversion due to recurrent bleeding in six cases (46%), stent-graft infection in three cases (23%), or pelvic abscess in four cases (31%). Primary open surgery was applied for five (20%) patients. After a mean follow-up of 34 months, early (<30 days) mortality was 8% (2/25), one-year mortality 16% (4/25), and overall mortality was 24% (6/25). CONCLUSIONS: Uretero-arterial fistula is a late complication of prior pelvic surgery, radiation, and indwelling ureteral stents. Endovascular treatment remains an effective and less invasive modality in controlling the related life-threatening arterial bleeding of the uretero-arterial fistula. Open surgical treatment is still required for patients with local sepsis, previously failed endovascular treatment or infected stent-grafts.

Thoracic endovascular aortic repair for esophageal cancer invading the thoracic aorta: a questionnaire survey study.

BACKGROUND: Locally advanced esophageal cancer occasionally invades the aorta, and hemorrhage from the esophagoaortic fistula can cause sudden death. Thoracic endovascular aortic repair (TEVAR) enables hemostasis in such cases, and prophylactic TEVAR can prevent fatal hemorrhagic events during treatment. However, its efficacy in Japan has not been evaluated. This study aimed to clarify the clinical significance of TEVAR in esophageal cancer patients. METHODS: The Japan Esophageal Society conducted a questionnaire survey targeting authorized or semi-authorized member institutes of the Authorized Institutes for Board Certified Esophageal Surgeons. Patients who underwent TEVAR for esophageal cancer were identified from 19 institutes. Data on patient demographics, treatment performed, and survival rate were obtained using the questionnaire. The Kaplan-Meier method was used for survival analysis and to compare differences in survival rates between those who underwent TEVAR for hemorrhage and those for preoperative prophylaxis. RESULTS: Of the 41 patients identified, 20 patients underwent TEVAR for hemorrhage or impending hemorrhage from the esophagoaortic fistula, while 21 patients underwent TEVAR as preoperative prophylaxis. The median survival time after TEVAR was 135 days in the hemorrhage or impending hemorrhage group and 378 days in the preoperative prophylaxis group. Eighteen patients underwent esophagectomy after TEVAR. No hemorrhagic event was observed during the perioperative period. The median survival time of the patients who underwent esophagectomy was 373 days. Some patients who achieved R0 resection obtained long-term survival. CONCLUSION: TEVAR is an efficacious modality to control a life-threatening hemorrhage from esophagoaortic fistula and helps to prolong the survival of patients with locally advanced esophageal cancer invading the aorta.

Endovascular Management of Ureteroarterial Fistula: Single Institution Experience and Systematic Literature Review.

BACKGROUND: Ureteroarterial fistula (UAF) represents an uncommon complication after urological surgery; however, this is a well-documented condition in patients with predisposing risk factors. The aim of the present study is to report and analyze the endovascular management of a series of patients with UAF, treated in authors' hospital, and to report and analyze the same data concerning patients retrieved from a systematic literature review. METHODS: Authors conducted a retrospective analysis of prospectively collected data and a systematic literature review. The research was carried out through PubMed database searching the following keywords: "uretero arterial fistula" and "uretero iliac fistula." It includes only articles reporting the endovascular management. RESULTS: Forty-six articles were included in the present study for a total of 94 patients. Risk factors were as follows: chronic indwelling ureteral stents, pelvic surgery, radiotherapy, iliac artery pseudo-aneurysm, and chemotherapy. All patients had gross hematuria at presentation. Stent graft placement was performed in 89 patients, embolization in 5 patients, and iliac internal artery embolization combined with stent graft placement was performed in 24 patients. Four postprocedural complications were observed (4.2%). During a median follow-up of 8 months, 10 complications related to UAF were observed (10.6%): rebleeding (7 cases) and stent thrombosis (3 cases). Two patients died for causes related to UAF (2.1%): rebleeding (1) and retroperitoneal abscess (1). CONCLUSION: Based on the present data, endovascular treatment is feasible and safe with low postprocedural complications and mortality rate. Considering the increase in surgery and radiotherapy performed, UAF should be always debated in patients with massive hematuria.

High mortality in patients with tracheoarterial fistulas: clinical experience and treatment recommendations.

OBJECTIVES: A tracheoarterial fistula (TAF) is an uncommon but a critical complication of tracheostomy and represents a surgical emergency. Surgical therapy with ligation of the brachiocephalic trunk can be considered as the first choice of treatment to control massive and life-threatening haemorrhage. METHODS: We describe 3 cases of TAF in patients who had long-term placement of a tracheostomy tube and the occurrence of a severe massive haemorrhage caused by an injured brachiocephalic trunk. All cases required emergent surgical revision. Different surgical techniques were used. In addition to the 3 case reports, we present a review of the literature of published TAF cases, summarize the different measures to control bleeding and compare the operative procedures used in the treatment of TAF. RESULTS: The occurrence of TAF is rare and constitutes a surgical emergency. The mortality rate of tracheoarterial erosion is 100% without surgical intervention for active bleeding into the airway. Therefore, rapid control of bleeding (via digital compression and an overinflated cuff) is the most important and first step of therapy. Subsequent emergency surgery with ligation and resection of the TAF and covering of the trachea should be considered to finally control the massive haemorrhage. In addition, cardiopulmonary bypass and circulatory arrest can be useful for surgical treatment of cases with uncontrollable bleeding. CONCLUSIONS: The mortality rate after resection of the fistula remains high. Finally, the most important factors for patient survival and outcomes are quick diagnosis of the TAF and immediate surgical control of bleeding.

Role of Endovascular Aortic Repair in the Treatment of Infected Aortic Aneurysms Complicated by Aortoenteric or Aortobronchial Fistulae.

BACKGROUND: The aim of this study was to compare outcomes and identify factors related to increased mortality of open surgical and endovascular aortic repair (EVAR) of primary mycotic aortic aneurysms complicated by aortoenteric fistula (AEF) or aortobronchial fistula (ABF). METHODS: Patients with primary mycotic aortic aneurysms complicated by an AEF or ABF treated by open surgery or endovascular repair between January 1993 and January 2014 were retrospectively reviewed. Outcomes were compared between the open surgery and endovascular groups, and a Cox's proportional hazard model was used to determine factors associated with mortality. RESULTS: A total of 29 patients included 14 received open surgery and 15 received endovascular repair. Positive initial bacterial blood culture results included Salmonella spp., oxacillin-resistant Staphylococcus aureus, and Klebsiella pneumoniae. Mortality within 1 month of surgery was higher in the open surgery than in the endovascular group (43 vs. 7%, respectively, p = 0.035). Shock, additional surgery to repair gastrointestinal (GI) or airway pathology, and aneurysm rupture were associated with a higher risk of death. Compared with patients without resection surgery, the adjusted hazard ratio of death within 4 years in patients with resection for GI/bronchial disease was 0.25. Survival within 6 months was better in the endovascular group (p = 0.016). CONCLUSION: The results of this study showed that EVAR/thoracic EVAR (TEVAR) is feasible for the management of infected aortic aneurysms complicated by an AEF or ABF, and results in good short-term outcomes. However, EVAR/TEVAR did not benefit long-term survival compared with open surgery.

Surgical strategy for the treatment of aortoesophageal fistula.

OBJECTIVE: To present a surgical strategy for aortoesophageal fistula (AEF). METHODS: From October 1999 to May 2017, 27 patients with AEF were treated at Kobe University Hospital. After 9 patients with malignancies or fish bone penetration were excluded, 18 patients who had AEF secondary to aortic lesions were investigated. The mean age was 67.2 ± 10.4 years, and the male/female ratio was 16:2. Twelve patients had a nondissection thoracic aneurysm, and 6 patients had a chronic aortic dissection. Six patients were in shock. Seven patients had a previous thoracic endovascular aortic repair (TEVAR) in the descending aorta, 2 patients had descending aorta replacement, 1 had hemiarch replacement, and 2 had total arch replacement. As the first treatment for AEF, 3 patients underwent TEVAR as destination therapy, 3 patients had a bridge TEVAR to open surgery, 1 patient had an extra-anatomical bypass from the ascending aorta to the abdominal aorta, and 11 patients had an in situ reconstruction of the descending aorta. The esophagus was resected in 16 patients, and an omental flap was installed in 16 patients. Additional procedures were extra-anatomical bypass in 2 patients and in situ reconstruction of the aorta in 3 patients. RESULTS: Hospital mortality was noted in 4 patients (22.2% persistent sepsis, n = 3: pneumonia, n = 1). However, since 2007, only 1 of 13 patients has died (pneumonia). Late death occurred in 5 patients, due to pneumonia, cerebral bleeding, diarrhea, sudden death, and persistent infection. Actuarial survival was 42.4 ± 12.8% at 5 years and freedom from aorta-related death was 59.4 ± 13.5% at 5 years. Nine patients achieved completed reconstruction of the esophagus 172 ± 57 days after initial surgery. CONCLUSIONS: Although a comparative study was not performed, 1-stage surgery consisting of resection of an aneurysm and esophagus, in situ reconstruction of the descending aorta, and omental flap installation provided a better outcome in the treatment for AEF. Bridging TEVAR to the open surgery is a useful adjunct in patients with AEF with hemorrhagic shock. Later reconstruction of the esophagus can be performed in the survivors.

Current status of endovascular treatment of aortoenteric fistula.

Aortoenteric fistula (AEF) is one of the most challenging diagnostic and therapeutic entities in vascular surgery. AEF can occur either primarily involving the aorta and the gastrointestinal tract or, more commonly, secondary to previous aortic reconstructive surgery. Traditionally, the treatment of AEF includes graft excision and extra-anatomic bypass surgery or in situ graft replacement. However, recently endovascular repair has emerged as an alternative therapeutic option. In this article, we present published and current evidence for endovascular repair of primary and secondary AEF. When endovascular treatment is applied where appropriate, early outcomes seem to be superior compared to open surgery. This benefit may be lost during long-term follow-up, implying that a staged approach with early conversion to in situ grafting may realize the best patient survival and morbidity. Lifelong administration of antibiotics is associated with a reduction in re-infection. An endovascular approach used as a bridging procedure in unstable patients is recommended, followed by definitive open therapy, if feasible, in patients with good life expectancy.

Secondary Aortoenteric Fistula After Abdominal Aortic Graft Implementation in Our Own Material.

BACKGROUND: Secondary aortoenteric fistula is a rare but severe complication occurring after abdominal aortic graft implementation. OBJECTIVES: The aim of our study was to review the clinical presentation, diagnosis and postoperative course of patients with aortoenteric fistulas following vascular operations on the abdominal aorta in the years 2000-2014. MATERIAL AND METHODS: Among all the patients treated in our center, during a 10-year period, aortoenteric fistulas were observed in 24 cases. The first symptoms occurred between 4 weeks and 8 years after the procedure (3.4 years on average). The most common clinical presentation was gastrointestinal bleeding in 16 cases (66.7%). All patients underwent surgical repair of an aortoenteric fistula, with graft removal and replacement in situ using silver impregnated prosthesis. RESULTS: The 30-day mortality rate after surgical treatment of SAEF was 37.5% (9 patients). Causes of death included: hemorrhagic shock, multi-organ failure and myocardial infarction. The early complications after the surgical repair of an aortoenteric fistula occurred in 19 (79.2%) patients. Mortality during the one-year follow-up period was 38.5% (5 patients) and the one-year complication rate was 69.2%. CONCLUSIONS: Each patient with a history of abdominal aortic graft implementation and presenting symptoms of gastrointestinal bleeding requires careful diagnosing for aortoenteric fistula. New methods of surgical treatment offering lower mortality should be developed.

Editor's Choice - Management of Secondary Aorto-enteric and Other Abdominal Arterio-enteric Fistulas: A Review and Pooled Data Analysis.

OBJECTIVES: To compare management strategies for secondary abdominal arterio-enteric fistulas (AEFs). METHODS: This study is a review and pooled data analysis. Medline and Scopus databases were searched for studies published between 1999 and 2015. Particular emphasis was given to short- and long-term outcomes in relation to AEF repair type. RESULTS: Two hundred and sixteen publications were retrieved, reporting on 823 patients. In-hospital mortality was 30.7%. Open surgery had higher in-hospital mortality (246/725, 33.9%), than endovascular methods (7/98, 7.1%, p < .001, OR 6.7, 95% CI 3-14.7, including staged endovascular to open surgery, 0/13, 0%). In-hospital mortality after graft removal/extra-anatomical bypass grafting was 31.2% (66/226), graft removal/in situ repair 34% (137/403), primary closure of the arterial defect 62.5% (10/16), and for miscellaneous open procedures 41.3% (33/80), p = .019. Among the subgroups of in situ repair, homografts were associated with a higher mortality than impregnated prosthetic grafts (p = .047). There was no difference in recurrent AEF-free rates between open and endovascular procedures. Extra-anatomical bypass/graft removal and in situ repair had a lower AEF recurrence rate than primary closure and homografts. Late sepsis occurred more often after endovascular surgery (2-year rates 42% vs. 19% for open, p = .001). The early survival benefit of endovascular surgery was blunted during follow-up, although it remained significant (p < .001). Within the in situ repair group, impregnated prosthetic grafts were associated with the worst overall and AEF related mortality free rates and vein grafts with the best. No recurrence, sepsis, or mortality was reported following staged endograft placement to open repair after a mean follow-up of 16.8 months (p = .18, p = .22, and p = .006, respectively, compared with patients in other groups). CONCLUSIONS: Endovascular surgery, where appropriate, is associated with better early survival than open surgery for secondary AEFs. Most of this benefit is lost during long-term follow-up, implying that a staged approach with early conversion to in situ vein grafting may achieve the best results in selected patients.

Late Fate of Cryopreserved Arterial Allografts.

OBJECTIVE: Initial enthusiasm for use of cryopreserved arterial allografts was subsequently tempered by suboptimal long-term outcome. Thrombosis, anastomotic pseudo-aneurysm, allograft disruption, aneurysmal degeneration, recurrent intestinal fistulization, and persistent infection are commonly reported in series with long-term follow-up. The authors reviewed their experience over the past 15 years with the use of cryopreserved arterial allografts as a vascular substitute for vascular prosthetic infection or for primary arterial infection, to investigate allograft-related complications. MATERIAL AND METHODS: A retrospective analysis of prospectively collected data was conducted for 103 cryopreserved arterial allografts inserted in 96 patients between July 2000 and July 2015. There were 78 patients with infected vascular prosthesis (IVP), nine patients with an aorto-enteric fistula (AEF), and nine patients with primary arterial infection (PAI). RESULTS: The in-hospital mortality was eight out of 78 (9%) IVP patients, three out of nine AEF patients, and zero out of nine PAI patients. Median follow-up was 49 months. Allograft-related re-interventions were necessary in 29% of the patients with IVP and four of the patients with AEF, but none of the patients with PAI. Five-year survival for the IVP, AEF, and PAI patients was 53%, 44%, and 90%, respectively. CONCLUSION: This series highlights some shortcomings of cryopreserved arterial allografts in the long term, including suboptimal outcome-results and shortage of material. The authors discuss the allograft-related complications and suggest some tricks to minimize their risk.

The Importance of Early Diagnosis and Treatment of Patients with Aortoenteric Fistulas Presenting with Herald Bleeds.

BACKGROUND: Massive bleeding in patients with aortoenteric fistula (AEF) may be preceded by minor, intermittent gastrointestinal (GI) blood loss, termed the "herald bleed." The aims of this retrospective study were to: (i) analyze the interval between the herald bleed and onset of major GI hemorrhage and/or diagnosis of AEF and (ii) to evaluate the diagnostic roles of endoscopy and computed tomography imaging. METHODS: Analysis of all patients diagnosed with AEF or iliac-enteric fistulas between 1994 and 2013 in a single institution. RESULTS: In 31 of a total of 34 fistula cases, GI bleeding was the presenting symptom. Of these, 17 of 31 presented with herald bleed while 14 of 31 presented with massive GI bleeding. In patients with a herald bleed, median time from first bleeding to diagnosis was 14 (2-137) days. In 5/17 patients, herald bleeding preceded major hemorrhage with a median of 6 (4-92) days before a diagnosis of AEF was made or intervention could be initiated. CT angiography (CTA) showed abnormalities associated with a fistula in 27 (79%) cases, of which in 12 (35%) cases a fistula was actually identified. Esophagogastroduodenoscopy (EGD) demonstrated a fistula in 8 (25%) patients, while 50% of EGDs were completely normal. CONCLUSIONS: Any patient with history of aortic surgery and GI bleeding should be considered to have an AEF until proven otherwise. The sensitivity of CTA for detecting AEF is substantially greater than that of EGD. The time interval between herald bleed and subsequent massive hemorrhage associated with AEF is unpredictable but may be as short as 4 days.

Results from the Multicenter Study on Aortoenteric Fistulization After Stent Grafting of the Abdominal Aorta (MAEFISTO).

OBJECTIVE: This study investigated the frequency, clinical features, therapeutic options, and results of aortoenteric fistulas (AEFs) developing after endovascular abdominal aortic repair (EVAR). METHODS: Eight Italian centers with an EVAR program participated in this retrospective multicenter study and collected data on AEFs that developed after a previous EVAR. RESULTS: A total of 3932 patients underwent EVAR between 1997 and 2013 at the participating centers. During the same period, 32 patients presented with an AEF during EVAR follow-up, 21 with original EVAR performed for atherosclerotic aneurysmal disease (ATS group) and 11 with the original EVAR performed for a postsurgical pseudoaneurysm (PSA group). The incidence of AEF development after EVAR was 0.46% in the ATS group and 3.9% in the PSA group. Anastomotic PSA as the indication to EVAR (P < .0001) and urgent/emergency EVAR (P = .01) were significantly associated with AEF development. Median time between EVAR and the AEF diagnosis was 32 months (interquartile range, 11-75 months) for the ATS group and 14 months (interquartile range, 10.5-21.5 months) for the PSA group. Among five AEF patients treated conservatively, two (40%) died, at 7 and 15 months, and the remaining three were alive at a median follow-up of 12 months. The AEF was treated surgically in 27 patients, including aortic stent graft explantation in all cases, in situ aortic reconstruction in 14 (52%), and extra-anatomic bypass in 13 (48%). Perioperative mortality was 37% (10 of 27). No additional aortic-related death was recorded in operated-on patients at a median follow-up of 28 months. CONCLUSIONS: Late AEFs rarely occur during EVAR follow-up, but the risk is significantly increased when EVAR is performed for PSA after previous aortic surgery and EVAR is performed as an emergency. Conservative and surgical treatment of post-EVAR AEF are both associated with high mortality. However, beyond the perioperative period, surgical correction of AEFs appears to be durable at midterm follow-up.

Infective etiology affects outcomes of late open conversion after failed endovascular aneurysm repair.

PURPOSE: To retrospectively review all patients undergoing late open conversion (LOC) after endovascular aneurysm repair (EVAR) in order to identify any clinical or technical predictors of poor outcome. METHODS: Twenty-six consecutive patients (24 men; mean age 74.7 ± 8.3 years) underwent LOC between June 2006 and April 2013 at our institution. The mean interval from index EVAR to LOC was 40.4 ± 29.2 months (range 5-93 months). The indication for LOC was endoleak in 14 (54%) patients and infection in 12 (46%): 2 (8%) patients with endoleak had a ruptured aneurysm and 6 (23%) patients with infection had a recurrent secondary aortoesophageal fistula (sAEF). RESULTS: In all 12 cases of infection and in 12 of 14 endoleaks, the entire endograft was explanted. A rifampin-soaked Dacron silver graft was implanted in all patients with infection. Patients with any infection and with recurrent AEF required more blood units than patients with endoleak (6.40 vs. 1.86, p = 0.045; 6.76 vs. 1.86, p = 0.0036, respectively). Compared with endoleak, the duration of conversions in the setting of infection (274 vs. 316 minutes, p = 0.42) and recurrent sAEF (274 vs. 396 minutes, p = 0.021) was longer. All patients with recurrent sAEF died at a mean 3.0 ± 2.5 days after LOC from proximal anastomosis disruption and hemorrhagic shock (n = 2), myocardial infarction (n = 2), acute stroke (n = 1), or persistent sepsis (n = 1). Perioperative mortality was significantly higher in patients with endograft infection (6/12, p = 0.002) and in cases of supraceliac cross-clamping (4/6, p = 0.003). The association of infection with supraceliac cross-clamping was a strong predictor for perioperative mortality (p < 0.001). CONCLUSION: In our experience, endograft infection led to greater perioperative mortality after LOC. Recurrent aortoenteric fistula in association with supraceliac cross-clamping is a strong predictor of poor outcome. Patients surviving the perioperative period may have good chances of long-term survival.