If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Editor's Choice – European Society for Vascular Surgery (ESVS) 2020 Clinical Practice Guidelines on the Management of Vascular Graft and Endograft Infections
b ESVS Guidelines Committee: Gert J. de Borst (Chair) (Utrecht, The Netherlands), Frederico Bastos Gonçalves (Lisbon, Portugal), Stavros K. Kakkos (Patras, Greece), Philippe Kolh (Liège, Belgium), Riikka Tulamo (Helsinki, Finland), Melina Vega de Ceniga (Review coordinator) (Bizkaia, Spain).
ESVS Guidelines Committee
Footnotes
b ESVS Guidelines Committee: Gert J. de Borst (Chair) (Utrecht, The Netherlands), Frederico Bastos Gonçalves (Lisbon, Portugal), Stavros K. Kakkos (Patras, Greece), Philippe Kolh (Liège, Belgium), Riikka Tulamo (Helsinki, Finland), Melina Vega de Ceniga (Review coordinator) (Bizkaia, Spain).
c Document Reviewers: Regula S. von Allmen (Gallen, Switzerland), Jos C. van den Berg (Bern, Switzerland), E. Sebastian Debus (Hamburg-Eppendorf, Germany), Mark J.W. Koelemay (Amsterdam, The Netherlands), Jose P. Linares-Palomino (Granada, Spain), Gregory L.L. Moneta (Portland, OR, USA), Jean-Baptiste Ricco (Poitiers, France), Anders Wanhainen (Uppsala, Sweden).
Document Reviewers
Footnotes
c Document Reviewers: Regula S. von Allmen (Gallen, Switzerland), Jos C. van den Berg (Bern, Switzerland), E. Sebastian Debus (Hamburg-Eppendorf, Germany), Mark J.W. Koelemay (Amsterdam, The Netherlands), Jose P. Linares-Palomino (Granada, Spain), Gregory L.L. Moneta (Portland, OR, USA), Jean-Baptiste Ricco (Poitiers, France), Anders Wanhainen (Uppsala, Sweden).
a Writing Committee: Nabil Chafké (Chair)* (Strasbourg, France), Holger Diener (Co-Chair) (Hamburg, Germany), Anne Lejay (Strasbourg, France), Ojan Assadian (Vienna, Austria), Xavier Berard (Bordeaux, France), Jocelyne Caillon (Nantes, France), Inge Fourneau (Leuven, Belgium), Andor W.J.M. Glaudemans (Groningen, The Netherlands), Igor Koncar (Belgrade, Serbia), Jes Lindholt (Odense, Denmark), Germano Melissano (Milan, Italy), Ben R. Saleem (Groningen, The Netherlands), Eric Senneville (Tourcoing, France), Riemer H.J.A. Slart (Groningen, The Netherlands), Zoltan Szeberin (Budapest, Hungary), Omke Teebken (Peine, Germany), Maarit Venermo (Helsinki, Finland), Frank Vermassen (Ghent, Belgium), Thomas R. Wyss (Bern, Switzerland). b ESVS Guidelines Committee: Gert J. de Borst (Chair) (Utrecht, The Netherlands), Frederico Bastos Gonçalves (Lisbon, Portugal), Stavros K. Kakkos (Patras, Greece), Philippe Kolh (Liège, Belgium), Riikka Tulamo (Helsinki, Finland), Melina Vega de Ceniga (Review coordinator) (Bizkaia, Spain). c Document Reviewers: Regula S. von Allmen (Gallen, Switzerland), Jos C. van den Berg (Bern, Switzerland), E. Sebastian Debus (Hamburg-Eppendorf, Germany), Mark J.W. Koelemay (Amsterdam, The Netherlands), Jose P. Linares-Palomino (Granada, Spain), Gregory L.L. Moneta (Portland, OR, USA), Jean-Baptiste Ricco (Poitiers, France), Anders Wanhainen (Uppsala, Sweden). d European Association of Nuclear Medicine e Groupe de Recherche sur les Infections de Prothèses
After studying medicine at Hanover Medical School and philosophy and social psychology at the Leibniz University Hanover, Omke E. Teebken joined the Christian Albrechts University in Kiel at the end of the 1990s as a research fellow at the Clinic for Cardiovascular Surgery headed by Professor Dr Axel Haverich, whom Omke E. Teebken later followed back to Hanover.
In Hanover, besides working as a clinician, Omke E.Teebken was particularly active scientifically, contributing to the establishment of the then newly founded Leibniz Laboratories for Biotechnology and Artificial Organs (LEBAO). His work focused on regenerative medicine and tissue engineering, and subsequently he wrote his habilitation thesis in in this field. After basic training in cardiac surgery, he specialised clinically in vascular surgery and played a pioneering role in the development of this field. Before being appointed director of the Clinic for Vascular Surgery – Endovascular Surgery at the Peine Clinic in 2016, Omke E. Teebken headed the Vascular Surgery – Endovascular Surgery Division of the Department of Cardiothoracic, Thoracic, Transplantation and Vascular Surgery at Hanover Medical School. Professor Teebken was a highly appreciated, committed, and competent colleague and teacher.
On 8 April 2019, Professor Teebken passed away after a short and severe illness. He was member and author of the ESVS guideline writing committee, an esteemed colleague, and friend.
Prof. Dr. med. Omke Enno Teebken 21.8.1968 – 8.4.2019
1. Introduction and General Aspects
1.1 Purpose of the guidelines
Guidelines driven by scientific societies on vascular graft/endograft infection (VGEI) have not been published. The European Society for Vascular Surgery (ESVS) has developed clinical practice guidelines for the care of patients with VGEI. The aim of this document is to assist physicians involved in the diagnosis and treatment of patients with VGEI in selecting the best management strategy in different scenarios. The potential users of this guideline include angiologists, vascular, cardiovascular and general surgeons, infectious disease physicians, and radiologists, and the target population comprises patients with VGEI in the supra-aortic trunks, thoracic and/or abdominal aorta, and peripheral arteries.
Guidelines have the purpose of promoting a standard of care according to specialists in the field, in this case represented by members of the ESVS. However, under no circumstances should these guidelines be seen as the legal standard of care in all patients. As the word guidelines states in itself, the document is a guiding principle, but the care given to a single patient is always dependent on the individual (symptom variability, comorbidities, age, etc.) and treatment setting (techniques available, local expertise).
1.2 Methods
1.2.1 The writing committee
The members of this guidelines Writing Committee (WC) were selected by the ESVS, the European Association of Nuclear Medicine (EANM), and the Groupe de Recherche sur les Infections de Prothèses, to represent physicians involved in the management of patients with VGEI. They include vascular surgeons, radiologists, and infectious disease specialists. WC members have provided disclosure statements of all relationships that might be perceived as real or potential sources of conflicts of interest, which are kept on file at the ESVS headquarters. No ESVS reviewers or individual WC members received any financial support from third parties in direct or indirect relation to this guideline, and all WC members and reviewers signed declarations of interest.
1.2.2 Evidence collection
1.2.2.1 Search strategy
The purpose, list of topics, and tasks and methods regarding the construction of the guidelines were agreed and distributed among the WC members in an initial meeting held in Strasbourg on 30 June 2017.
1.2.2.2 Literature search and selection
All WC members performed a systematic literature search strategy for each of their assigned sections, carried out in PubMed, Scopus, Cardiosource Clinical Trials Database, and the Cochrane Library databases, first from January 1997 to November 2017, with a later update to February 2019 for relevant papers published in English. Reference checking and a hand search added other relevant literature. Abstracts were excluded. Single case reports or case series were included if they were of paramount importance to these guidelines to enlighten the manuscript.
Selection of the literature was performed based on information provided in the title and abstract of the retrieved studies. Only peer reviewed published literature and studies presenting pre-defined outcomes were considered. The selection process followed the pyramid of evidence, with aggregated evidence at the top of the pyramid (systematic reviews, meta-analysis), followed by randomised controlled trials (RCTs), then observational studies, leaving expert opinion at the bottom. The level of evidence per section in the guidelines is dependent on the level of evidence available on the specific subject.
1.2.2.3 Evidence and recommendation grading criteria
To define the current guidelines, members of the WC reviewed and summarised the selected literature. Conclusions were drawn based on the availability and quality of the scientific evidence, and recommendations for the evaluation and treatment of patients with VGEI were formulated based on the analysis of the evidence and through consensus when evidence was scarce.
The European Society of Cardiology (ESC) grading system was used for evidence and recommendation rating. The letter A, B, or C reflects the level of current evidence (Table 1), and weighing the level of evidence and expert opinion, each recommendation is graded as class I, IIa, IIb, or III (Table 2). For those recommendations tables of evidence were built and are available as supplementary material.
Table 1Level of evidence
Level of evidence A
Data derived from multiple randomized clinical trials or meta analyses.
Level of evidence B
Data derived from a single randomized clinical trials or large non-randomized studies.
Level of evidence C
Consensus of opinion of the experts and/or small studies, retrospective studies, registries.
The goals behind patient participation in healthcare decision making can be categorised as democratisation and increased quality of decisions. Patient engagement improves the validity of clinical guidelines and is encouraged by international and national groups. In order to better understand patient feedback, European patients were interviewed: representatives of patient associations in the field of aortic dissection and infectious diseases; and patients treated for abdominal VGEI (patients operated on by surgeons of the WC). The main questions that arose from discussions were: (1) Did you feel your physician provided enough information about the risk of infection at the time of the initial procedure? (2) What did you think about the management once the diagnosis of VGEI was made? and (3) Did you think that your physician provided enough information on the risks related to VGEI? Patients were interviewed with a focus on these three open questions.
1.2.3 The revision process
The guidelines document, merged and harmonised by the co-chairmen of the WC, underwent internal review. Once approved by every WC member, it moved on to external revision by the ESVS Guidelines Committee (GC) members and chosen external experts in the field. Each draft was revised by the WC and the final document, approved by all WC and GC members and external reviewers, was submitted to the European Journal of Vascular and Endovascular Surgery on 20 July 2019.
1.2.4 The update plan
As technology and disease knowledge in this field changes rapidly, current recommendations can become outdated. It is an aim of the ESVS to revise the guidelines when important new insights in the evaluation and management of VGEI become available or every five years at the latest.
2. General Considerations
2.1 Definition of incisional surgical site infection
Studies dealing with VGEI are mostly case series rather than randomised studies. Diagnosis of VGEI is usually related to clinical findings, imaging studies, and microbiological examinations.
Criteria for incisional surgical site infections (SSI), which can be both superficial and deep, have been described by the Centers for Disease Control and Prevention (CDC) and can be applied to the description of VGEI (Table 3).
For diagnosis of SSI, diagnostic criteria 1, 2, and 3 must all be true.
1
Infection occurs within 30 days after the operative procedure
Infection occurs within 30 days after the operative procedure if no implant is left in place, or within one year if implant is in place and the infection appears to be related to the operative procedure
and 2
Infection involves only skin and subcutaneous tissue of the incision
Infection involves deep soft tissues (e.g., fascia and muscle layers) of the incision
and 3
Patient has at least one of the following:
Patient has at least one of the following:
•
Purulent drainage from the superficial incision
•
Purulent drainage from the deep incision but not from the organ/space component of the surgical site
•
Organisms isolated from an aseptically obtained culture of fluid or tissue from the superficial incision
•
A deep incision spontaneously dehisces or is deliberately opened by a surgeon and is culture positive or not cultured when the patient has at least one of the following signs or symptoms: fever (>38°C), or localised pain or tenderness. A culture negative finding does not meet this criterion
•
At least one of the following signs or symptoms of infection: pain or tenderness, localised swelling, redness or heat, and superficial incision is deliberately opened by surgeon and is culture positive or not cultured. A culture negative finding does not meet this criterion
•
An abscess or other evidence of infection involving the deep incision is found on direct examination, during re-operation, or by histopathological or radiological examination
•
Diagnosis of superficial incisional SSI by a surgeon or attending physician
•
Diagnosis of a deep incisional SSI by a surgeon or attending physician
Types
Incisional primary
A superficial incisional SSI that is identified in the primary incision in a patient who has had an operation with one or more incisions
A deep incisional SSI that is identified in a primary incision in a patient who has had an operation with one or more incisions
Incisional secondary
A superficial incisional SSI that is identified in the secondary incision in a patient who has had an operation with >1 incision (e.g., donor site [leg] incision to harvest autologous veins for in situ reconstruction of an abdominal vascular graft infection)
A deep incisional SSI that is identified in the secondary incision in a patient who has had an operation with >1 incision (e.g., donor site [leg] incision to harvest autologous veins for in situ reconstruction of an abdominal vascular graft infection)
Reporting instructions
Do not report a skin suture abscess with minimal inflammation and discharge confined to the points of suture penetration, as an infection
Classify infection that involves both superficial and deep incision sites as deep incisional SSI
Do not report a localised stab wound infection as SSI; instead, report as skin or soft tissue infection, depending on its depth
If the incisional site infection involves or extends into the fascial and muscle layers, report as a deep incisional SSI
∗ For diagnosis of SSI, diagnostic criteria 1, 2, and 3 must all be true.
differentiate between superficial and deep incisional SSIs without placing emphasis on vascular grafts (VGs), the Szilagyi classification and the Samson classification specifically also consider VG involvement, while the extent of graft involvement can be described using the Bunt classification (Table 4).
Furthermore, aortic VGEI can also be divided into early (< 4 months) or late (> 4 months) onset, which, in many cases, is also extrapolated to other VGEl.
However, the clinical relevance of differentiation between early and late infections remains a matter of debate.
Table 4Classifications for wound and vascular graft infections with respect to wound infection (Szilagyi, Samson) and to the extent of graft involvement (Bunt)
Grade III: infection involving the vascular prosthesis
Samson classification:
Group 1: no deeper than dermis
Group 2: subcutaneous tissue, no direct contact with the graft
Group 3: body of graft but not anastomosis
Group 4: exposed anastomosis, no bleeding, no bacteraemia
Group 5: anastomosis involved, bleeding, bacteraemia
Extent of graft involvement (Bunt classification modified)
Peripheral graft infection:
P0 graft infection: infection of a cavitary graft (e.g., aortic arch; abdominal and thoracic aortic interposition; aorto-iliac, aortofemoral, iliofemoral graft infections)
P1 graft infection: infection of a graft whose entire anatomical course is non-cavitary (e.g., carotid–subclavian, axillo-axillary, axillofemoral, femorofemoral, femorodistal, dialysis access bridge graft infections)
P2 graft infection: infection of the extracavitary portion of a graft whose origin is cavitary (e.g., infected groin segment of an aortofemoral or thoracofemoral graft, cervical infection of an aortocarotid graft)
P3 graft infection: infection involving a prosthetic patch angioplasty (e.g., carotid and femoral endarterectomies with prosthetic patch closure)
Graft-enteric erosion
Graft-enteric fistula
Aortic stump sepsis after excision of an infected aortic graft
2.3 Definition of vascular graft/endograft infection
To overcome the numerous shortcomings of current classifications, the Management of Aortic Graft Infection (MAGIC) group has developed a list of major and minor criteria with respect to clinical, surgical, radiological, and laboratory findings (Table 5).
Once VGEI is suspected, an exhaustive evaluation of the clinical status, signs of infection, and comorbidities of the patient according to the MAGIC criteria is recommended.
Blood culture(s) positive and no apparent source except graft infection
Fever ≥38°C with graft infection as most likely cause
Abnormally elevated inflammatory markers with graft infection as most likely cause, e.g., erythrocyte sedimentation rate, C reactive protein, white cell count
According to the MAGIC criteria, VGEI is suspected in the presence of one major or two minor criteria of the three different categories, and VGEI is diagnosed when there is at least a single major criterion and any other criterion from another category. For example, a fever ≥ 38°C is considered non-specific for VGEI and therefore it is required that no other clinical cause is apparent. Sepsis and systemic inflammatory response syndrome may be caused by something other than VGEI and is defined as combinations of different findings. Anorexia, lethargy, and malaise may accompany aortic graft and endograft (EG) infection, but are also considered insufficiently specific.
Intra-operative fluids around a graft can represent pus, but despite a yellowish or cloudy appearance may be present for non-infective reasons and microbiological culture will be negative. Therefore, pus cells must be proven by direct microscopy to be considered a major criterion. Furthermore, a direct communication between non-sterile sites and a prosthesis indicates graft infection: aorto-enteric fistula (AEnF), aorto-bronchial fistula (ABF), deployment of a stent graft in an already infected field (e.g., infected aneurysm), and exposed grafts in deep open wounds.
2.4 Epidemiology
2.4.1 Incidence
VGEI are usually multifactorial and result from the complex involvement of patient, surgical, and environmental factors, making the real incidence difficult to assess. Reported incidences of VGEI by type and anatomical location will be developed in specific sections.
2.4.2 Risk factors
Multiple risk factors contribute to VGEI and are listed in Table 6.
Table 6Risk factors for vascular graft/endograft infection
The pathogenesis of VGEI is multifactorial. Presumably, early VGEI are mostly caused by a breach in sterility during implantation or the presence of bacteria in the aneurysmal thrombus, while late VGEI are mostly caused by haematogenous seeding from a bacteraemia (mostly arising from the urinary or respiratory tract), or from bacterial translocation or iatrogenic contamination during catheterisation.
The pathogenesis of AEnF, aorto-oesophageal (AEsF), and ABF remains unclear. Ischaemia of the visceral wall due to occlusion of the feeding arteries, and mechanical erosion by the aneurysm or of a suture line pseudo-aneurysm, especially when still under pressure due to presence of an endoleak, have all been suggested. Fistula can occur as a result of direct trauma related to surgical injury, poor tunnelling, erosion by direct contact, or by the penetration of an oversised EG. Previous adjacent or remote infection in any site is considered to be a causative or contributing factor.
The quality of material incorporation related to tissue ingrowth and healing also plays a role, explaining that VEGI might even be more frequent than VG infection (VGI), as there is no tissue ingrowth in the wall of the EG fabric that is surrounded only by thrombotic material, contrary to VG.
As mentioned in the MAGIC criteria, the clinical presentation of patients with VGEI varies between mild symptoms (redness of the skin, non-purulent effusion from a wound) to severe and evident symtoms such as sepsis or anastomotic rupture with hypovolaemic shock.
Fever of unclear origin and an unexplained leukocytosis with concomitant increase of C reactive protein (CRP) and fever may be the only clinical or laboratory sign of VGEI. In other cases the clinical manifestations may include abscess, mass, septic embolisation, septic shock, bleeding, melaena, haematemesis, haematuria, ileus, or abdominal distension. When VGEI is suspected, a complete clinical and biochemical evaluation of the patient is required in order to provide a sufficient analytical overview.
Post-implantation syndrome, characterised by transitory fever associated with elevated leukocytes and CRP may be observed following endograft implantation, but might also be distinguishing from an actual infection.
Micro-organism identification is a key issue in order to provide the patient with the best treatment. Using the different available sampling techniques, micro-organisms can be isolated in about 75% – 98% of cases.
Responsible pathogens are Gram positive bacteria in up to 58% of VGEI (including enterococci, Staphylococcus aureus, and coagulase negative staphylococci); Gram negative bacteria account for about 34% of VGEIs and and anaerobes 8%.
Staphylococcus aureus, Enterobacteriaceae, Pseudomonas aeruginosa, and beta haemolytic streptococci were classified as virulent, while bacteria belonging to the skin colonising flora such as Staphylococcus epidermidis, corynebacterial, and Cutibacterium acnes were classified as non-virulent agents. The results of this meta-analysis established that virulent organisms were significantly associated with an increased risk of re-infection.
Antimicrobial resistance of the causative bacteria is another factor that may reduce the chance of healing, but this relationship has not been clearly established in the setting of VGEIs.
The susceptibility of bacteria to the few antibiotics that exhibit a sustained activity in the environment of a biofilm (e.g., rifampicin combinations for staphylococcal implant infections) is another element that may lead to re-infection in patients treated for VGEIs.
Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. Foreign-Body Infection (FBI) Study Group.
Microbiological samples may support establishing the diagnosis of a VGEI. Ideally samples should be harvested before the start of antimicrobial therapy. However the accuracy and relevance of microbiological tests depend on whether specimens were collected without contamination, and in an adequate quantity.
A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2013 Recommendations by the Infectious Diseases Society of America (IDSA) and the American Society for Microbiology (ASM).
Moreover, samples should be forwarded quickly to the microbiology laboratory. If they cannot be forwarded immediately they should be stored at +4°C.
2.7.2.1 Directly obtained specimens
Meaningful results will be achieved with specimens obtained directly from the suspected infection site. These may include surgically explanted prosthetic materials, intra-operatively obtained tissue and graft biopsies from the infected area, or at least three samples from perigraft fluid collection.
Paediatric anaerobic tubes, which require very small amounts of material, can be used.
Aspirated specimens obtained under ultrasound (US) or computed tomography (CT) guidance provide material for an accurate microbiological diagnosis. The presence of graft incorporation into tissue reliably excluded the presence of bacteria in cultures in 97% of investigated grafts, whereas the finding of graft disincorporation accurately predicted a positive culture in 89% of all positive VGEI cases.
In general, tissue specimens or a portion of the graft material are superior to swab specimens of infected sites, even when collected using a sterile technique intra-operatively. At least three direct specimens should be collected in sterile containers.
Swabs should be avoided because they do not allow differentiation of colonising micro-organisms from true pathogens and may lead to overprescription of broad spectrum antibiotics. Swabs have an inherent difficulty transferring bacteria or fungi from the swab fibres onto culture media, and because the inoculum from the swab often is not uniformly distributed across several different agar plates.
Evaluation of the effects of storage in two different swab fabrics and under three different transport conditions on recovery of aerobic and anaerobic bacteria.
If swabs are used, the type of swab should be selected on basis of its ability to collect micro-organisms. Polyethylene terephthalate (PET) swabs should be used.
A new innovative specimen collection technology (microDTTect) could help in the future, as it allows for contamination free sampling, and also it can dislodge bacteria embedded in a biofilm from prosthetic surfaces.
Indirect specimens might also be meaningful, especially when direct specimens are not collected in cases when redo surgery is not performed. Such speciments include blood cultures, specimens obtained from a superficial wound, a draining sinus, or otherwise close anatomical structures.
Despite being an indirect microbiological sampling method, blood cultures may yield supportive information, as pre-operative blood cultures have been found to be positive in about 35% of cases and both pre- and peri-operative samples are positive for the same micro-organism in about 22% – 30% of cases.
However, other indirectly obtained specimens using swabs, biopsy samples or aspirates obtained from a superficial wound, a draining sinus, or otherwise close anatomical structures always contain skin flora or colonisation, and might not accurately reflect the causative micro-organism of a VGEI.
Bacteriological investigation of negative pressure wound therapy (NPWT) foams should not be performed to support the diagnosis of a VGEI because of the low sensitivity and specificity.
Therefore, results from indirectly obtained specimens should be considered with caution.
2.7.3 Microbiological sample processing
Specimens may be investigated using different techniques such as direct streaking specimens on agar plates, placing specimens into broth culture, homogenisation of tissue or graft specimens with serial dilution techniques, sonication of a harvested graft, or vortex mixing tissue samples in order to enhance the recovery of biofilm forming micro-organisms.
Enhanced sample processing techniques such as vortex mixing specimens or sonication improve the detection rate of microorganisms attached to graft material.
Importantly, the high energy levels of direct ultrasonic disruption can decrease the number of viable Gram negative bacteria, and vortex agitation consistently produced the lowest bacterial numbers among the three methods tested. An ultrasonic bath treatment of one to five minutes duration of infected VG at a frequency of 25 – 40 KHz may be the optimal preparation method for causative bacteria detection.
Adding broad range polymerase chain reaction (PCR) detection to sonicated fluid cultures may even increase the detection rate of bacteria attached to graft material.
Various imaging techniques are used in the diagnostic work up when VGEI is suspected. Conventional imaging techniques such as US, CT, CT with angiography (CTA), and magnetic resonance with angiography (MRA) are used most frequently. Other available imaging tools are nuclear medicine techniques, such as 18F-fluoro-D-deoxyglucose positron emission tomography (18F-FDG-PET) with or without diagnostic contrast enhanced CT (18F-FDG-PET/CT), and white blood cell scintigraphy (WBCS), that can be combined with single photon emission computed tomography (SPECT/CT) for better localisation of the infection.
US is the most common, non-invasive, low cost imaging modality to identify findings associated with VGEI.
US characteristics of VGEI are the presence of pseudo-aneurysm, sustained presence of gas (if still present after > 7 weeks), and purely anechoic fluid collections (if still present >3 months after surgery).
It can differentiate between haematoma or abscess formation, which makes it a good primary imaging screening modality, especially for superficial peripheral VG. However, the absence of peri-prosthetic collections on US does not allow ruling out of a VGEI. US also allows investigating for graft thrombosis, which can be the first sign of VGEI, and it can guide puncture for bacteriological purposes.
However, US has a high interoperator variability and the predictive value is limited in the case of a centrally located graft due to overlying bowel gas or obesity.
Therefore, the sensitivity of US for the diagnosis of VGEI is considered as low, and additional investigations are often needed to obtain more detailed information on VG status.
2.8.2.2 Computed tomography angiography
CTA has been considered the reference imaging standard in diagnosing VGEI for a long time, as it is able to visualise the characteristic features of VGEI.
The use of intravenous contrast, with images acquired in the arterial phase, may show certain signs such as ectopic gas, fluid, soft tissue enhancement, pseudo-aneurysm, focal bowel thickening, and discontinuation of the aneurysmal wall, all of which can all be used as criteria to increase the likelihood of a VGEI.
In a systematic review and meta-analysis of patients with suspected VGEI, the pooled sensitivity of CTA in diagnosing VGEI was 0.67 (95% confidence interval [CI] 0.57 – 0.75) and the pooled specificity was 0.63 (95% CI 0.48 – 0.76). This meta-analysis showed that an isolated CTA does not provide enough evidence to establish the diagnosis of VGEI (Fig. 1).
Standalone CTA can confirm the diagnosis of VGEI, but a second imaging modality such as 18F-FDG-PET/CT or WBCS combined with SPECT/CT may be useful to map the extent of the infection.
2.8.2.3 Magnetic resonance angiography
MRA has not been evaluated as extensively as CTA for the diagnosis of VGEI, but several studies have suggested that MRA offers better anatomical and functional information than CTA, including tissue characterisation.
Simultaneous or sequential acquisition of 18F-FDG-PET with MRA provides additional quantitative molecular functional information concerning the inflammatory lesion, and accurate localisation, as well as anatomical changes with motion correction. After six post-operative weeks, the presence of collections with a hypo-intense signal in T1 and a hyperintense signal in T2 strongly suggests a VGEI.
In a series of patients with suspected aortic VGEI, the sensitivity of MRA was 0.68 (95% CI 0.50 – 0.86), and the specificity 0.97 (95% CI 0.91 – 1.00).
However, owing to low availability and long acquisition times resulting in motion artefacts, MRA is currently not used as a first line diagnostic modality if VGEI is suspected.
2.8.3 Nuclear imaging techniques
Nuclear medicine imaging techniques, such as 18F-FDG-PET combined with (low dose or contrast enhanced) CT and WBCS combined with SPECT/CT, incorporate anatomical and metabolic information at the same time and are able to differentiate between VGEI, soft tissue infection, and, in some cases, inflammation by pattern recognition, heterogeneity, and intensity of uptake with FDG-PET,
and by increase in size or intensity with time with WBCS.40
2.8.3.1 Positron emission tomography
18F-FDG-PET imaging is based on the uptake of radioactive labelled glucose in cells/tissue with enhanced glucose metabolism, such as inflammatory cells and micro-organisms such as bacteria or fungi. This diagnostic method may differentiate between peri-prosthetic collection and involvement of the graft material but should be combined with low dose CT for anatomical correlation. Nowadays, 18F-FDG-PET is mainly performed in hybrid mode with FDG-PET/CT, which has an established role in the assessment of suspected VGEI, providing accurate anatomical localisation of the site of infection.
The EANM and Society of Nuclear Medicine and Molecular Imaging published procedural guidelines on how to perform a 18F-FDG-PET scan for infectious purposes.
As the administered dose of 18F-FDG and time interval between the scan acquisition may cause heterogeneity between studies, the EANM launched a strategy to harmonise 18F-FDG-PET/CT studies (EANM Research Limited, EARL).
There are different ways to analyse and interpret 18F-FDG-PET/CT studies. The main interpretation criteria are the calculated maximum standardised uptake value (SUVmax), the tissue to background ratio, the pattern of uptake (focal/diffuse), or the visual grading scale.
It is suggested that SUVmax > 8 in the perigraft area is the cut off value for distinguishing infected grafts from non-infected grafts, but this is based on a small number of patients. It is also considered that linear, diffuse, and homogeneous uptake with projection of the vessel is highly suggestive of infection. Although in the past diabetes and use of antibiotics were supposed to degrade image quality, two recent studies demonstrated that diagnostic accuracy was not affected.
Do hyperglycemia and diabetes affect the incidence of false-negative 18F-FDG PET/CT studies in patients evaluated for infection or inflammation and cancer? A comparative analysis.
In a meta-analysis, the sensitivity of single 18F-FDG PET without combined low dose or contrast enhanced CT in diagnosing VGEI in patients with a suspected VGEI was 0.94 (95% CI 0.88 – 0.98), with a specificity of 0.70 (95% CI 0.59 – 0.79).
18F-FDG-PET combined with CT (adding low dose or contrast enhanced CT) showed even better results, with a sensitivity of 0.95 (95% CI 0.87 – 0.99) and a specificity of 0.80 (95% CI 0.69– 0.89) (Fig. 1).
WBCS detects infected sites by visualizing the increase of accumulation of radiolabelled white blood cells over time. Recently, procedural guidelines for the labelling of the white clood cells, and for the correct acquisition and interpretation criteria for WBCS were published.
Clinical indications, image acquisition and data interpretation for white blood cells and anti-granulocyte monoclonal antibody scintigraphy: an EANM procedural guideline.
The diagnosis of VGEI infection is based on the presence of pathological accumulation of labelled white blood cells at the site of infection. At least two sets of images are required (2 – 4 and 20 – 24 hours after injection) and an increase in intensity or size with time is considered positive for an infection. When positive, SPECT/CT images are mandatory for exact localisation of the infection (soft tissue only, graft, or extension).
Clinical indications, image acquisition and data interpretation for white blood cells and anti-granulocyte monoclonal antibody scintigraphy: an EANM procedural guideline.
WBCS is a very specific method, but it has some limitations. The procedure is time consuming and labour intensive, as the imaging needs to be performed at least at two different time points (preferably 2 – 4 and 20 – 24 hours after injection) and in a laboratory specifically equipped to perform leukocyte labelling. Furthermore, the diagnostic accuracy of WBCS depends on the region of the body in which the images are performed. The accuracy is somewhat lower in the central parts of the body than in peripheral parts (so in case of aortic VGEI) as the tracer is eliminated via the intestinal tract and physiologically taken up in the bone marrow, leading to a difficult interpretation of the aorta. Using antigranulocyte antibody scintigraphy as an alternative does not demand laboratory labelling, but does require dual time point imaging and is hampered by physiological uptake in bone marrow and excretion in the intestinal tract.
Clinical indications, image acquisition and data interpretation for white blood cells and anti-granulocyte monoclonal antibody scintigraphy: an EANM procedural guideline.
Clinical indications, image acquisition and data interpretation for white blood cells and anti-granulocyte monoclonal antibody scintigraphy: an EANM procedural guideline.
The estimated sensitivity of WBCS (without SPECT/CT) in diagnosing VGEI in the most recent meta-analysis was 0.90 (95% CI 0.85 – 0.94) with a specificity of 0.88 (95% CI 0.81 – 0.94).
When WBCS was combined with SPECT/CT, the sensitivity increased to 0.99 (95% CI 0.92 – 1.00), with a specificity of 0.82 (95% CI 0.57 – 0.96) (Fig. 1).
It is, however, not recommended as the first imaging modality in diagnosing VGEI because of the aforementioned limitations and limited availability (Fig. 2).
Tabled
1
Recommendation 1
Once vascular graft/endograft infection is suspected, exhaustive evaluation of clinical status, signs of infection and patient comorbidities according to the MAGIC criteria is recommended.
A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2013 Recommendations by the Infectious Diseases Society of America (IDSA) and the American Society for Microbiology (ASM).
To obtain microbiological proof of vascular graft/endograft infection, the yield of at least three deep rather than superficial samples should be considered.
A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2013 Recommendations by the Infectious Diseases Society of America (IDSA) and the American Society for Microbiology (ASM).
Microbiological investigation of negative pressure wound therapy foams should not be performed in order to support the diagnosis of vascular graft/endograft infection.
For patients with a clinical suspicion of vascular graft/endograft infection and with non-convincing findings on CTA, the use of 18F-FDG-PET combined with low dose CT is recommended as an additional imaging modality to improve diagnositc accuracy.
In patients with a clinical suspicion of peripheral vascular graft/endograft infection, single photon emission computed tomography, if available, is recommended as an additional imaging modality to improve diagnostic accuracy.
Figure 2Imaging workflow if vascular graft/endograft infection (VGEI) is suspected, divided into thoracic/abdominal and limb grafts. CTA = computed tomography angiography; MRA = magnetic resonance angiography; 18F-FDG-PET/CT = 18F-fluoro-D-deoxyglucose positron emission tomography/computed tomography; WBCS = white blood cell scintigraphy; SPECT/CT = single photon emission computed tomography/computed tomography. *18F-FDG PET/CT can add more information, particularly in inconclusive CT. In some high grade infection cases a second imaging modality as 18F-FDG PET/CT and/or WBCS combined with SPECT/CT may be useful to map the extent of the infection. †WBCS can be applied if available otherwise, 18F-FDG PET/CT can be used.
3. Strategies to Prevent Graft or Endograft Infection
3.1 Raw materials
Currently, vascular devices are mainly made of two different polymers, PET or expanded polytetrafluoroethylene (ePTFE) for soft materials, and different alloys, nitinol being the most used, for stents. The choice of these polymers and alloys is mainly related to their chemical and mechanical stability, rather than their properties for inhibiting micro-organism colonisation. They can be implanted as raw materials or associated with different adjuncts such as matrices of impregnation or surface treatment. There is no strong evidence on differences of susceptibility to infection of synthetic vascular raw biomaterials. Differences of susceptibility to infection between PET and ePTFE have been evaluated in vitro and in vivo with conflicting results. In vivo studies did not find differences in infectability between PET and ePTFE materials,
Although all bacteria are able to adhere to an inert support, some bacteria, such as coagulase negative staphylococci (e.g., S. epidermidis) or streptococcus viridans (e.g., Streptococcus mitis oralis), show a high propensity for adhesion to foreign materials.
No significant difference has been found in vivo between PET and glutaraldehyde treated bovine pericardium, used as aortic patches, to resist bacterial infection.
The high prevalence of nasal carriage of S. aureus in the general population and its role in potentially severe VGEIs raises the question of the beneficial effect of its decolonisation in patients undergoing vascular surgery.
In another prospective study, patients undergoing aorto-iliac surgery were screened for S. aureus nasal carriage and, if positive, were treated with mupirocin nasal ointment and chlorhexidine body washes, and compared with a historical control group of patients who tested positive but received no treatment. The incidence of S. aureus SSI was significantly lower in patients who were screened positive and who were treated for methicillin resistant S. aureus (MRSA) nasal carriage compared with patients negative for nasal MRSA carriage (0% vs. 13.6%).
Of note, S. aureus eradication in this setting was associated with a decrease in S.aureus related SSIs but not in the SSIs due to other bacteria, which may be explained by the competing behaviour of bacteria causing SSI.
In a meta-analysis, there was no evidence of any benefit from a pre-operative bathing or shower regimen with antiseptic agents over unmedicated bathing.
A number of measures have been found to reduce the risk of SSI of 51% in patients undergoing open and endovascular elective surgery or elective lower limb amputation: peri-operative normothermia; hair removal the day before surgery; and discipline in aseptic care in the operating room.
In a meta-analysis, antimicrobial prophylaxis with broad spectrum systemic antibiotics significantly reduced the risk of wound infection and early graft infection in arterial reconstructions (relative risk [RR] 0.25, 95% CI 0.17 – 0.38; and RR 0.31, 95% CI 0.11 – 0.85, respectively).
In all patients undergoing open or endovascular abdominal aortic aneurysm repair, therefore peri-operative systemic antimicrobial prophylaxis is recommended.
Antimicrobial prophylaxis for vascular surgery should cover the bacteria most likely to be responsible for SSIs and achieve adequate tissue levels at the time of incision and throughout the procedure to prevent any bacterial colonisation of injured skin/soft tissue and implant. First or second generation cephalosporins are the most widely used agents owing to their profile of tolerance and antibacterial spectrum that cover methicillin susceptible staphylococci (i.e., S. aureus and coagulase negative staphyloccoci), streptococci, and some Gram negative bacilli. A meta-analysis of 22 RCTs concluded that prophylactic systemic antibiotics for patients undergoing peripheral arterial reconstruction reduced the risk of SSI (RR 0.25, 95% CI 0.17 – 0.38) and early VGI (RR 0.31, 95% CI 0.11 – 0.85).
In the same study, no difference in the protective effect on SSI rate was noted between first or second generation cephalosporins, penicillins with lactamase inhibitors, aminoglycosides, or vancomycin.
The coverage of MRSA and/or coagulase negative staphyloccoci may be considered according to the local prevalence of these strains, even though no significant difference concerning SSI rates using cefazolin plus vancomycin or daptomycin vs. cefazolin alone has been found.
Antibiotic prophylaxis has the best efficacy when administered before the incision (ideally within 30 min) with re-injection for longer interventions according to the half life of the compound (i.e., two hours for cefuroxime and four hours for cefazolin).
In a retrospective study including all SSIs after lower extremity revascularisation procedures between 2012 and 2016, meticulous wound closure with a monofilament absorbable suture has been shown to be superior to staples in decreasing SSI rates.
3.3 Antibiotic prophylaxis during dental extraction
Beyond the peri-operative risk of SSI, the implant can be infected at any time after the intervention, especially in the presence of bacteraemia (i.e., secondary haematogenous related to SSI). Analogous to prosthetic cardiac valves, antibiotic prophylaxis after VG for secondary infection may follow the recommendations of the ESC and American College of Cardiology/American Heart Association.
In their most recent guidelines, antimicrobial prophylaxis was recommended for patients with a prosthetic cardiac valve for high risk procedures such as dental procedures involving the manipulation of the gingival or peri-apical region of teeth or perforation of the oral mucosa, including scaling and root canal procedures.
Antibiotic prophylaxis has therefore been proposed recently for those patients with an aortic prosthesis, whether placed by open surgical repair or endovascular aneurysm repair (EVAR), before any dental procedure involving the manipulation of the gingival or peri-apical region of teeth or perforation of the oral mucosa, including scaling and root canal procedures.
In every case where a vascular graft/endograft is implanted, antimicrobial prophylaxis to cover the first 24 hours, by intravenous administration of a first/second generation cephalosporin or vancomycin in the event of penicillin allergy, is recommended.
Antimicrobial prophylaxis to prevent vascular graft/endograft infection should be considered before any dental procedure involving the manipulation of the gingival or peri-apical region of teeth or perforation of the oral mucosa, including scaling and root canal procedures for patients previously operated on with a vascular/endovascular graft.
In view of the rarity of VGEI, the complexity of diagnosis and treatment, and the difficulty of the interventions and severity of complications, centralisation of the patients suffering from VGEI is clearly indicated. Accordingly, patients should be transferred to specialised high volume centres with multidisciplinary experience in VGEI, including angiologists, vascular, cardiovascular and general surgeons, microbiologists and radiologists.
4.1 Antimicrobial therapy
4.1.1 Choice of antimicrobial therapy
Antimicrobial therapy is an integral part of VGEI treatment. In the acute phase intensive antimicrobial therapy with broad spectrum antibiotics or antibiotics directed against the most likely infecting organisms is indicated to control infection and sepsis. In the choice of antimicrobial therapy the fact that the graft material may be covered with a biofilm and also the local epidemiology of resistance patterns have to be considered. In some specific situations, the addition of antifungal agents should be considered, especially in visceral fistula cases. Once the responsible infecting organisms are known the spectrum should be narrowed if possible.
4.1.2 Duration of treatment
There is no consensus on the optimal length of antimicrobial therapy for VGEI. If prosthetic material can be removed and a thorough debridement of all infected tissue can be performed, a minimum of two weeks of intravenous therapy, if possible, followed by an oral regimen for another two to four weeks is indicated. If the infected material is replaced by a new VG, four to six weeks of intensive antimicrobial therapy is usually proposed to prevent recurrent infection. Many authors favour a total treatment time of three to six months in this situation, and some even advocate one year of treatment. In those patients in whom general conditions preclude any surgery, lifelong treatment should be considered.
This can be an option in patients at higher risk of surgery, especially in low grade infections with less virulent infecting organisms, susceptible to suitable antibiotics, and without other complications. In some cases, the infection cannot be totally eradicated but kept under control by year long or even lifelong therapy.
Because of the complexity of interpreting microbiological tests results and the permanent evolution of antimicrobial therapies and micro-organism resistance, antimicrobial therapy management must be done by an infectious diseases specialist within a multidisciplinary team, including vascular surgeons, radiologists, microbiologists, anaesthetists, and gastrointestinal and pulmonary specialists for cases with a concomitant fistula.
Patients with multidrug resistant (MDR) bacteria (such as MRSA, but not methicillin resistant coagulase negative staphylococci, extended spectrum beta lactamase producing Enterobacteriaceae, or glycopeptide resistant enterococi) should be isolated and should remain isolated during the hospital stay as carriage is prolonged, especially in patients receiving antibiotics. The wearing of gloves is required for the manipulation of any biological material, but this is not specific to MDR bacteria (gown for any contact with the patient, mask if pulmonary colonisation or infection). Both colonised and infected patients with these bacteria should be managed in a single patient room to reduce the risk of spread to other patients.
ESCMID guidelines for the management of the infection control measures to reduce transmission of multidrug-resistant Gram-negative bacteria in hospitalized patients.
Historically, total removal of the infected VG or EG, debridement and rinsing with antiseptic solution of the infected area, and extra-anatomic reconstruction (EAR) outside the infected field was considered as the gold standard to avoid recurrent infection. This procedure should be performed in two stages when possible. However, this approach is not always feasible, often not easy, and increases the risk of complications like stump blowout. Therefore, most authors now prefer an in situ reconstruction (ISR) with infection resistant material combined with removal of the infected graft material, aggressive debridement of the arterial bed and targeted antimicrobial therapy.
In most situations the results with ISR are at least equivalent to extra-anatomic repair.
In thoracic and abdominal procedures especially, it is recommended that any VG and anastomosis should be covered with viable tissue such as omentum, muscle, or pericardial patch.
Direct contact with viscera or organs should also be avoided. If no viable tissue is available, a bovine pericardial patch can be used. Anastomoses or suture lines can be reinforced with fascia or pledgets.
Tabled
1
Recommendation 14
Antimicrobial therapy is recommended in every patient with an infected graft/endograft.
For the diagnosis and treatment of vascular graft/endograft infection it is recommended that the patient be transferred to specialised high volume centre with multidisciplinary experience in this pathology.
The exact incidence of VGEI in the supra-aortic trunks (SAT) is unknown, but probably extremely low. SAT VGEI includes prosthetic patch, bypass, and stent graft infections.
However, infection rates might be underestimated, owing to lack of recognition and under reporting, as in other locations.
Over the last three decades, a total of 140 cases of SAT patch/bypass infections have been reported, mostly involving carotid patches. An overall incidence of 0.25% – 0.5% was reported in a systematic review of carotid endarterectomies involving PET patches.
Because of its infrequent occurrence, it is difficult to identify the aetiology of SAT EG or endograft infection (EGI), but haematoma could be a risk factor in promoting the development of early infection.
A systematic review of the literature identified only 12 patients with SAT stent graft infections in the last three decades: eight patients presented with infected carotid stent graft, three with infected subclavian stent graft, and one with an infected tandem brachiocephalic and subclavian artery stent graft.
Half of all reported infections occurred within the first four post-operative months and involved carotid patches. The most common clinical presentation for early infection is abscess, neck mass, and haemorrhage. When infection presents later, the main symptom is a draining sinus. Patients with SAT stent graft infection typically present with fever, malaise, and pain. SAT stent graft infection is mostly encountered after stent graft implantation for carotid blowout syndrome.
In the setting of early post-operative infection, S. aureus is the most commonly encountered micro-organism, while S. epidermidis is the predominant pathogen in patients who present with late infections.
US allows the evaluation of the patency of the revascularisation, the presence of a collection, and its characterisation. It has been highlighted that carotid patch corrugation on US might be an early warning sign of VGI.
5.2 Treatment options for supra-aortic trunk vascular graft/endograft infection
5.2.1 Conservative treatment
Conservative treatment of SAT VGEI is not recommended in patients fit for intervention because of the risk of suture line rupture for patch/bypass and vascular wall necrosis for stent graft, potentially leading to uncontrollable major bleeding in the chest and tracheal compression in the neck. However, successful conservative treatment of the infected stent graft using parenteral antimicrobial therapy alone has been described.
Endovascular treatment is also an option for SAT VGEI. It is mostly used in life threatening presentations in order to control a major bleeding related to VGEI.
For acute bleeding related to blowout syndrome, open surgery in an irradiated area may be challenging, and emergency operative ligation may be associated with high rates of major morbidity, meaning that an endovascular approach is usually preferred. A systematic review and meta-analysis including 559 patients demonstrated that both coil embolisation and reconstruction with stent grafts may be safe treatment options for carotid blowout syndrome: peri-operative mortality was 3% for patients treated by carotid embolisation and 12% for patients treated with covered stent grafts, while the peri-operative stroke rate was 1% in both groups.
The stent graft first strategy can also be considered as a bridge to definitive therapy in unstable patients. This strategy allows a controlled and semi-elective graft revision, including excision and reconstruction. Patients should receive intravenous broad spectrum or microbiology based antimicrobial therapy, followed by long term treatment.
The endo vacuum assisted closure (VAC) technique is a hybrid approach that has been used in 10 cases of infected SAT reconstructions. The EndoVAC technique is a three step procedure: relining of the infected reconstruction with a stent graft; removal of the infected VG without clamping; and use of NPWT to permit granulation.
This technique provided good results but was performed in few patients selected on an individual basis for the treatment as they had severe comorbidity and adverse anatomy, providing limited evidence.
5.2.3 Reconstruction
5.2.3.1 Graft material
A surgical approach with total explantation of infected foreign material is recommended in the elective setting only. Replacement of the explanted bypass/patch or arterial segment is usually mandatory to avoid cerebral ischaemia or infarction. However, primary ligation of the vessel may be considered in emergency life threatening situations, if the infected reconstruction is already thrombosed without neurological symptoms or with an already completed cerebral infarct, to avoid revascularisation syndrome and cerebral haemorrhage.
Autologous material is usually considered as the first line option for reconstruction. Because of the typically short length of these reconstructions, autologous saphenous vein reconstruction (bypass or patch) can be performed in the majority of cases.
A direct surgical approach with total explantation of the foreign material is mostly performed in non-emergency conditions in an attempt to avoid emergency procedures for life threatening haemorrhage in the neck or chest. Usually, bypasses and arterial segments with infected patches or stent grafts are short, except bypasses starting from the ascending aorta.
Consequently total explantation is usually performed. Obtaining proximal arterial control is mandatory, even using an occlusion balloon or at an unscarred site through a sternotomy or a thoracotomy, in order safely to enter the cervical phlegmon and minimise peri-operative complications.
Management of carotid Dacron patch infection: a case report using median sternotomy for proximal common carotid artery control and in situ polytetrafluoroethylene grafting.
The approach that involves proximal arterial control is even more relevant in cases with active bleeding due to infection associated arterial wall breakdown.
Management of carotid Dacron patch infection: a case report using median sternotomy for proximal common carotid artery control and in situ polytetrafluoroethylene grafting.
Indeed, primary arterial control through the same cervical incision may be difficult, with the risk of significant blood loss and increased inadvertent peripheral nerve injuries.
Partial explantation with local wound debridement is less often performed.
5.2.3.3 Adjunctive therapy
A muscle flap may be considered as a possible adjunctive option for SAT VGEI infection. Sternocleidomastoid or pectoralis major muscle flaps have been reported in 11 cases following SAT patch/bypass infections. Of these 11, seven flaps were following complete removal of the infected material and arterial reconstruction, while four flaps were combined with local wound debridement.
None of the 11 patients died from a related cause. Of the seven patients who underwent complete removal of the infected material and arterial reconstruction, one stroke and two transient nerve injuries were reported. Of the four patients who underwent local wound debridement, one pseudo-aneurysm occurred requiring the placement of a covered stent.
5.3 Follow up and prognosis
Over the last three decades, 140 cases of SAT patch/bypass infections have been reported and 138 treatment modalities described: total removal of infected material and arterial reconstruction in 86 cases; phlegmon excision in two cases; the EndoVAC technique in 10; ligation in seven; endovascular treatment using covered stent grafts in seven; and conservative treatment in 26 cases.
Of the 12 cases of SAT stent graft infections reported over the last three decades, 11 treatment modalities were described: stent graft removal and arterial reconstruction in six cases; stent graft removal without arterial reconstruction in two cases; carotid embolisation in two cases; and conservative treatment in one case. Peri-operative complications were described in 54.6% of cases. Median follow up was 4.5 months, and five patients died from related causes (Fig. 4).
When patch corrugation is found on ultrasound follow up after carotid endarterectomy further investigations may be considered to exclude a vascular graft infection.
For patients with supra-aortic trunk vascular graft/endograft infection, total removal of infected material followed by reconstruction with autologous material is recommended.
The EndoVAC technique may be considered as a treatment option in selected patients with supra-aortic trunk vascular graft/endograft infection when neither total removal of infected material nor when usual conservative VAC therapy are considered feasible or safe.
In the emergency setting with active bleeding in patients with supra-aortic trunk vascular graft/endograft infection, a combined endovascular and surgical approach may be considered.
Conservative treatment, including antimicrobial therapy without reconstruction, for supra-aortic trunk vascular graft/endograft infection may be considered for patients unfit for surgery.
Together with the increasing number of procedures performed on the thoracic aorta, including thoracic EVAR (TEVAR), the incidence of thoracic VGEI is also increasing. Additionally, thoracic VGEI is often associated with AEsF, ABF, or aortopulmonary fistula (APF), which makes treatment more complicated, with the need (besides the aortic reconstruction) for adjunctive surgical procedures to repair the oesophageal or bronchial lesion.
New insights regarding the incidence, presentation and treatment options of aorto-oesophageal fistulation after thoracic endovascular aortic repair: the European Registry of Endovascular Aortic Repair Complications.
Aorto-bronchial and aorto-pulmonary fistulation after thoracic endovascular aortic repair: an analysis from the European Registry of Endovascular Aortic Repair Complications.
A systematic review identified 43 studies reporting on 233 patients with 49 VGI and 184 EGI. Only four were multicentre studies, which included 107 patients, all with EGI. The remaining 39 single centre studies included 49 patients with VGI and 77 with EGI.
New insights regarding the incidence, presentation and treatment options of aorto-oesophageal fistulation after thoracic endovascular aortic repair: the European Registry of Endovascular Aortic Repair Complications.
In a systematic review, the association with AEF was significantly more frequent (60% vs. 31%) and the time interval from index procedure to infection was significantly shorter (17 ± 21 months vs. 32 ± 61 months) with EG compared with VG, respectively.
The clinical symptoms of thoracic aortic VGEI can range from unexplained fever, as observed in post-implantation syndrome, to sepsis, massive bleeding, and shock. Owing to the depth of the thoracic aorta, visible local signs of infection are mostly absent. Septic emboli can cause secondary loci of infection or even abscesses. For AEsF or ABF, haematemesis or haemoptysis may be the first symptom. This bleeding may be massive, especially for AEsF but is often preceded by self limiting “herald bleeding”.
6.1.3 Specific diagnostic modalities
Definite diagnosis mostly demands a CTA showing perigraft fluid, air in the aneurysm sac or surrounding it, or abscess formation in the surrounding tissues.
When an AEsF is present, the prosthetic material may be seen protruding in the oesophagus on oesophagoscopy. In case of an ABF the defect in the bronchus can only be seen when it is centrally located, e.g., in the left main bronchus. Diagnostic examinations should be performed without delay when infection of a thoracic aortic VGEI is suspected.
6.2 Thoracic vascular graft/endograft infection without fistula
6.2.1 Conservative treatment
Although surgical principles dictate control of sepsis, removal of all infected prosthetic material, and reconstruction in a clean field, this is not always achievable because it places a huge burden on an already sick patient (Fig. 5). Therefore, staged repair or conservative treatment are the only options.
Figure 5Proposed algorithm for the management of thoracic aortic graft/endograft infection. ∗ In a single or staged prodecure. † Materials that can be used are cryoprerved allografts, treated PET grafts or biological xenografts. VGEI = vascular graft/endograft infection; PET = polyethylene terephthalate.
In the presence of perigraft fluid collections or abscesses, percutaneous drainage under ultrasound or CT guidance can be performed in combination with antimicrobial therapy. A 10 – 14 F pigtail catheter or a 12 – 20 F drain is inserted percutaneously and left in place until the collection is totally or sufficiently drained.
6.2.1.2 Irrigation
Irrigation with saline or an antiseptic solution can be used in order to dilute the bacterial burden in prosthetic and peri-prosthetic tissues. It can be performed through percutaneous drains or after a surgical procedure with aortic reconstruction.
In a systematic review of single centre series, conservative treatment (antimicrobial therapy with or without percutaneously inserted drainage of fluid collections or flushing) was performed in 2% of patients with VGI (n = 1/49) and 17% with EGI (n = 13/77). The mortality rate was 100% at 30 days in VGI group; and 38% at 30 days, 75% at one year, and 100% at five years in the EGI group.
Removal of the infected graft material, aggressive debridement of the arterial bed, and arterial reconstruction with suturing in healthy non-infected tissue using infection resistant material constitute the basis of this treatment modality
6.2.2.1 Specific techniques
The operative technique largely depends on the VG or EG location. If the prosthesis extends into the aortic arch, a median sternotomy or a clamshell incision is indicated, and the intervention needs to be performed under total cardiopulmonary bypass, circulatory arrest, and selective cerebral perfusion. If the proximal extent of the VG or EG is distal to the left subclavian artery (LSCA), the procedure can be performed through a left thoracotomy with single lung ventilation and left heart bypass. Dissection of the proximal neck, usually between the left common carotid artery (LCCA) and the LSCA, can be difficult owing to the inflammation caused by the infection or the previous intervention. Care should be taken not to damage adjacent structures, like the lung, vagus nerve, or oesophagus. If extensive reconstructions need to be performed, measures like cerebrospinal fluid drainage may be considered in order to reduce the risk of spinal cord ischaemia. Intra-operative evaluation using motor evoked or somatosensory evoked potentials may also be used. If the VG or EG extends to the level of the LSCA and clamping between the LCCA and the LSCA is not possible, hypothermic circulatory arrest is needed to allow complete prosthetic material excision.
When exposure of the thoraco-abdominal aorta is required, a thoracophrenolaparotomy is the preferred approach. As in primary procedures, the splanchnic arteries can be perfused with normothermic blood and the renal arteries with cold crystalloids during cross clamping.
6.2.2.2 Graft materials
Cryopreserved aortic allografts have been proposed for the replacement of infected thoracic VGEI.
As a biological material, allografts have demonstrated a higher resistance to infection than synthetic VGs, but are exposed to the subsequent risk of degeneration, rupture, and bleeding when the infection is caused by necrotising organisms, such as P. aeruginosa or Candida spp.
Long term durability results, especially with regard to the development of calcification and aneurysms when used in the thoracic aorta, are still lacking.
Treated PET VGs, such as rifampicin soaked and silver coated (with or without triclosan) VGs, have been proposed in order to decrease the risk of early infection.
In a series including mainly explanted infected abdominal endografts, treated PET VGs have been shown to provide better results in terms of prevention of re-infection and five year overall survival than standard PET VGs (53% vs. 12%) These results can probably be extrapolated to the TEVAR setting. Bovine pericardium has been reported as a technical option for ISR of thoracic aortic VGEI, tailoring a custom made tube by sewing pericardial sheets.
While promoted in case series, this technique still needs further studies and longer follow up.
6.2.2.3 Adjunctive therapy
The recommended VG coverage to avoid its direct contact with surrounding organs like lung or oesophagus using the surrounding tissues is often not possible. Therefore, it is advised to cover the VG with other viable tissue. Intercostal flap coverage can be used, as well as pericardial or omental flaps. An intercostal flap has limited volume and is best prepared at the moment of thoracotomy to avoid damage caused by the retractor. When a pericardial flap is used, the pericardial defect may need to be repaired with synthetic material. Omentum can be prepared by laparoscopic access and routed through the diaphragm via the aortic hiatus, to cover the VG and fill a dead space after mediastinal debridement.
If there is no viable tissue available, use of a bovine pericardial patch is suggested.
6.2.3 Extra-anatomic reconstruction
To avoid reconstruction in a contaminated field and recurrent infection, EAR outside the infected field and secondary aortic ligation with removal of the infected VG or EG can be performed in one or two stages.
6.2.3.1 Technique
To restore distal perfusion after aortic ligation, axillo-bifemoral or bilateral axillofemoral bypasses can be performed, but retrograde blood flow to the visceral organs under all these circumstances may be insufficient.
The most commonly used EAR is the so called ventral aorta, consisting of a retrosternally placed VG that originates from the ascending aorta, the distal anastomosis being on the supracoeliac abdominal aorta or more distally, on the infrarenal aorta or iliac arteries.
32532-8&id=fx1.jpg){kind=link}
32532-8&id=gr1.jpg){kind=link}
32532-8&id=gr2.jpg){kind=link}
32532-8&id=gr3.jpg){kind=link}
32532-8&id=gr4.jpg){kind=link}
32532-8&id=gr5.jpg){kind=link}