Abstract
Background:
Abdominal aortic stent graft infection (AAGI) is a severe complication. The optimal management of AAGI remains unclear. This study provides updated results of bilateral axillofemoral bypasses (AFBs) for patients with AAGI.
Patients and Methods:
In total, 31 patients (25 men; mean age, 67.1 years) with AAGI treated using AFB between January 2006 and April 2020 were included. Overall, the mean follow-up duration was 24 months (range, 1–72). In the 23 patients who survived the post-operative period, the mean follow-up duration was 32 months (range, 12–72).
Results:
Thirty-day and in-hospital mortality rates was 16% and 26%, respectively. The 12-month primary and secondary patency rates for the AFB graft were both 91%. In total, seven (30%) patients received re-interventions such as thrombectomy and balloon angioplasty. No amputation was required during follow-up. Culture results were positive in 87% of pre-operative cultures and 84% of intra-operative cultures. Staphylococcus aureus was the most prevalent pathogen, with four cases of methicillin-resistant Staphylococcus aureus and one each of vancomycin-resistant enterococci, carbapenem-resistant Klebsiella pneumoniae, and carbapenem-resistant Enterobacteriaceae. In-hospital mortality rate was 57% in patients with drug-resistant pathogens.
Conclusions:
Reconstruction with bilateral AFB and stent graft removal in patients with AAGI is a feasible treatment modality and provided an acceptable patency rate and low amputation rate. Additional studies investigating long-term results and the optimal treatment of AAGI are required.
Abdominal aortic stent graft infection (AAGI) is an extremely complex complication with a reported early mortality rate of 17%–24% [1,2]. Its prevalence was reported to be 0.2%–4%, with an anticipated upward trend following the growing demand for vascular graft procedures [3–5]. Risk factors for AAGIs include comorbidities such as chronic kidney disease, diabetes, immunosuppression, and bacteremia [6]. Abdominal pain and fever were the most common presentations of AAGI. However, AAGI may present with no obvious physical findings to suggest infection [7].
Managing AAGI has always been a challenge for vascular surgeons. Extra-anatomic bypass (EAB) has been the treatment of choice since its introduction in the 1970s [8]. However, its role as the standard treatment has been challenged because of its disadvantages including unsatisfactory patency rate, high amputation rate, long operation time, and risk of stump rupture [7]. Over the past decades, the number of patients receiving in situ reconstructions (ISR) has increased considerably. Several meta-analyses have demonstrated the inferiority of EAB compared with ISR with respect to patency, amputation, and mortality [2,9,10]. However, EAB theoretically has a lower risk of infection because there is no reconstruction in the infected area. Hence, EAB remains the treatment of choice for virulent pathogens such as methicillin-resistant Staphylococcus aureus (MRSA), Pseudomonas, or multi-drug–resistant pathogens [7].
The low incidence and complex nature of AAGI makes randomized study designs difficult to implement. Meta-analyses of different case series remain the most common method for comparing different treatment options. This study was designed to provide updated results of bilateral axillofemoral bypass (AFB) reconstructions for AAGI to facilitate further investigation.
Patients and Methods
Study design
In total, 31 patients treated for AAGI through stent graft removal and AFB reconstruction between January 2006 and April 2020 were included in our study, which was approved by the Institutional Review Board of our institute. Patients were identified retrospectively, and their medical records were reviewed. Abdominal aortic stent graft infection diagnosis was confirmed mostly through computed tomography (CT). Nuclear medicine scans such as 18F-fluorodeoxyglucose positron emission tomography (PET)/CT were used if regular CT could not reveal concrete radiological evidence when clinical presentations strongly suggested graft infection. Vascular surgeons with similar training backgrounds and high degree of consensus at periodical meetings on treatment policies executed all preoperative planning, operative procedures, and follow-ups.
Surgical details
During surgery, all patients received bilateral AFB with 8- or 10-mm ring-reinforced grafts, followed by insertion of an aortic balloon into the abdominal aorta around the superior mesenteric artery level from the left axillary artery for endovascular aortic occlusion. After aortic balloon inflation and clamping of the bilateral iliac arteries, the abdominal aorta was excised, the stent graft was dismantled, and the infected tissue was cleaned. The aorta and bilateral iliac stumps were suture ligated. Finally, the abdominal cavity was washed with normal saline and the infected area was covered with the omentum. All patients received post-operative intravenous antibiotic agents for at least four weeks.
Follow-up
After discharge, patients received regular outpatient follow-up every three months. Axillofemoral bypass graft patency was assessed through clinical examination and non-invasive vascular examinations, such as sonography or Doppler segmental pressure measurement. In patients with acceptable renal function and uremic patients under regular hemodialysis, contrast CT was performed three, six, and 12 months after treatment and then annually. Aspirin was used routinely by our patients. Axillofemoral bypass graft occlusions were managed through thrombectomy and balloon angioplasty in all patients.
Statistical analysis
Patient demographics and characteristics are presented as means and standard deviations or ranges for continuous variables and as numbers (percentages) for categorical variables. Statistical analyses were performed using Microsoft Excel 2016 (Microsoft, Redmond, WA).
Results
Patient demographics and surgical details
The demographic data and comorbidities of the 31 patients are listed in Table 1. The mean patient age at AAGI diagnosis was 67.1 ± 9.9 years. Moreover, 25 (81%) of the included patients were men. The mean time between abdominal aortic stent graft implantation and stent graft infection diagnosis was 14.7 ± 5.1 months. The explanted stents were Zenith Flex (Cook Inc., Bloomington, IN) in 11 (36%) patients; Endurant (Medtronic Vascular, Santa Rosa, CA) in three (10%); and Excluder (WL Gore & Associates, Flagstaff, AZ) in 17 (55%).
Patient Characteristics
Survival, complications, and patency rate
Outcomes are listed in Table 2. The mean follow-up duration was 24 (range, 1–72) months, and the mean 30-day and in-hospital mortality rates were 16% and 26%, respectively. Only one patient, who presented with aorto-enteric fistula, died within 30 days. Five patients had peri-operative duodenal perforation, and in-hospital mortality occurred in three (60%). Aortic stump rupture occurred in two patients, both of whom required re-operation and one of whom died in the hospital. Other peri-operative complications included acute kidney failure requiring hemodialysis in eight (26%) patients, acute myocardial infarction in three (10%), pneumonia in four (13%), stroke in one (3%), and wound complication in six (19%).
Survival, Complications and Patency Rate
In the 23 patients who survived the post-operative period, 12-month primary and secondary patency rates of the AFB graft were both 91%. During the long-term follow-up of 32 (range, 12–72) months, seven (30%) patients experienced graft failure and thus received re-interventions such as thrombectomy and balloon angioplasty. No amputation was required during their follow-up.
Microbiology
Table 3 presents the microbiologic results of our patients. Pre-operative blood cultures were positive in 27 (87%) patients, with only one (3%) polymicrobial culture. Gram-positive cocci were identified in 14 (45%) patients, with Staphylococcus aureus being the most prevalent bacterium: methicillin-susceptible Staphylococcus aureus (MSSA) in seven cases and MRSA in four cases.
Microbiology
MSSA = methicillin-susceptible Staphylococcus aureus; MRSA = methicillin-resistant Staphylococcus aureus; CRKP = carbapenem-resistant Klebsiella pneumoniae; VRE = vancomycin-resistant enterococci; CRE = carbapenem-resistant Enterobacteriaceae.
Intra-operative aortic cultures were positive in 26 (84%) patients, with no polymicrobial cultures. Gram-positive cocci were identified in 14 (45%) patients, with Staphylococcus aureus remaining the most prevalent pathogen. In addition to the four MRSA cases, drug-resistant pathogens such as vancomycin-resistant enterococci (VRE), carbapenem-resistant Klebsiella pneumoniae (CRKP), and carbapenem-resistant Enterobacteriaceae (CRE) were each isolated from one of the intra-operative cultures. The in-hospital mortality rate was 57% in patients with drug-resistant pathogens.
Discussion
In our patients, the 30-day and in-hospital mortality rates were 16% and 26%, respectively, comparable to previous data. Batt et al. [2] reviewed 36 studies published between January 1994 and July 2014, with 16 studies including results regarding EAB. Early mortality rates were 24% in patients treated with EAB and 15% in patients treated with ISR. In their meta-analysis, O'Connor et al. [9] found the early mortality rates to be 18% for EAB and 7%–14% for ISR. The inferiority of EAB in these studies should be interpreted with caution because of biases associated with meta-analyses. In addition, patients in poorer conditions and with virulent pathogens are more likely to receive EAB rather than ISR.
In the current study, the lower limb amputation rate during a mean follow-up duration of 32 months was 0%. Moreover, the 12-month primary and secondary patency rates were both 91%. Most of the previous meta-analyses have revealed an amputation rate of 8%–11% for EAB and 0%–8% for ISR [2,9,10]. Regarding patency, the results have been presented in various ways, with the length of follow-up duration varying greatly among published studies. In a case series presented by Batt et al. [11], the overall 12-month patency rate was 89% in 82 patients with aortic graft infection, without differences between the EAB and ISR groups. In summary, here, we noted low amputation and patency rates comparable to the published data.
The duodenum was the most common site of bowel erosion in patients with AAGI, likely because of its retroperitoneal fixation and proximity to the aorta [12,13]. In our study, duodenal repair was required in six patients, five of whom had peri-operative duodenal perforation and one had aortoenteric fistula diagnosed before surgery. Bowel erosion is associated with a disastrous survival rate. An analysis of 50 patients with aortoenteric fistula reported a 60-day mortality rate of approximately 50% [13]. In the current study, peri-operative duodenal perforation was associated with in-hospital mortality rate of 60%.
The current results for pre-operative blood cultures and intra-operative aorta cultures revealed negative culture rates of 13% and 16%, respectively. Gram-positive cocci were isolated from 45% of both pre-operative and intra-operative aorta cultures. In prior case series and reviews, negative culture rates were 18%–37%, with gram-positive cocci being the most commonly isolated organisms [14,15]. However, polymicrobial cultures were reported at 26%–52% in published reviews [1,11], but they were only 0%–3% in our series. This result may be explained by the lower proportion of patients with aortoenteric fistula in our study. Several drug-resistant pathogens such as MRSA, VRE, CRKP, and CRE were isolated from our intra-operative cultures, and these were associated with a 57% in-hospital mortality rate. Isolation of MRSA has been reported in two relevant case series [11,14]. However, other drug-resistant pathogens and the prognoses associated with such organisms have not been discussed thus far, and thus, may require further investigation.
A limitation of our study is its single-center retrospective design, with a relatively short follow-up duration. Furthermore, its small sample size may limit the generalizability of its results. Moreover, our data did not compare different management strategies such as ISR or non-operative treatments. Further studies are warranted to overcome these limitations.
Conclusions
Our study revealed acceptable results for bilateral AFB regarding patency and amputation rates for patients receiving stent graft removal after AAGI. Long-term follow-up is warranted, and additional studies investigating the optimal treatment of AAGI are warranted.
Footnotes
Acknowledgment
We thank the Taiwan Association of Cardiovascular Surgery Research for their help. This manuscript was edited by Wallace Academic Editing.
Funding Information
No funding was received.
Author Disclosure Statement
The authors declare that they have no conflict of interest.