722018

research-article2017
JETXXX10.1177/1526602817722018Journal of Endovascular TherapyArgyriou et al

A SAGE Publication

Meta-analysis

Journal of Endovascular Therapy

Endograft Infection After Endovascular

1­–10
© The Author(s) 2017
Reprints and permissions:
Abdominal Aortic Aneurysm Repair: A sagepub.com/journalsPermissions.nav
DOI: 10.1177/1526602817722018
https://doi.org/10.1177/1526602817722018

Systematic Review and Meta-analysis www.jevt.org

Christos Argyriou, MD, PhD1, George S. Georgiadis, MD, PhD1,

Miltos K. Lazarides, MD, PhD, FEBVS1, Efstratios Georgakarakos, MD, PhD, MSc1,
and George A. Antoniou, MD, PhD, MSc, FEBVS2

Abstract
Purpose: To report a meta-analysis of the published evidence on the outcomes of aortic endograft infection after
endovascular aneurysm repair (EVAR). Methods: A search of electronic information sources (PubMed/MEDLINE,
SCOPUS, CENTRAL) and bibliographic reference lists identified 12 studies reporting on 362 patients (mean age 72 years;
279 men). The methodological quality of the selected studies was assessed using the Newcastle-Ottawa scale. Endpoints
were 30-day/in-hospital mortality and follow-up mortality. Pooled estimates are reported with the 95% confidence interval
(CI). The review was registered at the International Prospective Register of Systematic Reviews in Health and Social Care
(CRD42016034166). Results: The incidence of graft infection after EVAR was 0.6% (95% CI 0.4% to 0.8%). The time
from implantation to diagnosis ranged from 1 to 128 months (mean 25). The majority of patients (293, 81%) underwent
surgical treatment (95% CI 77% to 83%); 9 (2.5%) patients (95% CI 21% to 43%) received conservative treatment. Aortic
replacement with a prosthetic graft was performed in 58% (95% CI 52% to 62%), whereas cryopreserved allografts and
autologous grafts were used in 31% (95% CI 28% to 33%) and 11% (95% CI% 8 to 14%), respectively. Less than half of the
patients (40%) had emergency surgery. The pooled estimate of 30-day/in-hospital mortality was 26.6% (95% CI 16.9% to
39.2%). The pooled 30-day/in-hospital mortality for 9 patients treated conservatively was 63.3% (95% CI 30.7% to 87.0%).
The pooled overall follow-up mortality was 45.7% (95% CI 36.4% to 55.4%) vs 58.6% (95% CI 28.8% to 83.3%) for the
9 patients receiving conservative treatment. Conclusion: Aortic endograft infection is a rare complication after EVAR.
Surgical treatment with complete explantation of the infected endograft seems to be the optimal management in selected
patients. Supportive medical treatment without surgical intervention has a significant associated mortality.

Keywords
abdominal aortic aneurysm, antibiotics, endograft infection, endovascular aneurysm repair, explantation, extra-anatomical
bypass, in-situ reconstruction, mortality, open conversion, stent-graft

Introduction secondary interventions after EVAR are in the form of an

Endovascular aneurysm repair (EVAR) is currently consid-
ered the predominant treatment for abdominal aortic aneu-
rysms (AAAs), especially in high-risk patients,1 providing
low early morbidity and mortality compared with open
repair. In particular, operative mortality after open repair is
4.6% vs 1.2% for EVAR; combining operative mortality
and severe complications produced rates of 9.8% vs 4.7%,
respectively.2,3 Only a small proportion of EVAR patients
develop complications, which range from 0.7% related to
local-vascular or implant-related complications to 11% for
systemic complications.4 Endoleak, endotension, stent-graft
migration, stent-graft thrombosis, and graft infection requir-
ing early or late open conversion are not infrequent.5–8 Most
endovascular procedure; however, in a minority of patients
(reported as <5%), a conversion to open repair is required.9–12
1
Department of Vascular and Endovascular Surgery, University General
Hospital of Alexandroupolis, “Democritus” University of Thrace,
Alexandroupolis, Greece
2
Department of Vascular and Endovascular Surgery, The Royal Oldham
Hospital, Pennine Acute Hospitals NHS Trust, Manchester, UK
Corresponding Author:
George S. Georgiadis, Department of Vascular and Endovascular
Surgery, University General Hospital of Alexandroupolis, “Democritus”
University of Thrace, 7 Alexandrou Papanastasiou Street,
Alexandroupolis, 68131 Greece.
Email: ggeorgia@med.duth.gr
2 Journal of Endovascular Therapy 00(0)
Endograft infection is a rare but potentially lethal com-
plication after EVAR, especially if left untreated.4 A post-
EVAR graft infection represents both a diagnostic and
therapeutic challenge and is considered an absolute indica-
tion for late open conversion.13,14 Abdominal aortic endo-
graft explantation for infection is high risk and associated
with graft-enteric fistula in one-third of the cases.15
Endograft explantation is considered the gold standard
approach, whereas a nonoperative management or “bridg-
ing” therapy could be considered for patients with high sur-
gical risk.16,17 Although the incidence of graft infection after
open AAA repair is thought to be similar to that after
EVAR,16 aortic endograft infection remains an underre-
ported condition in the literature.18 Our objective was to
perform a systematic review of the literature and accumu-
late the evidence on the management and outcomes of
endograft infection after previous EVAR.
Methods
Study Design and Search Strategies
The review conformed to the Preferred Reporting Items for
Systematic Reviews and Meta-Analyses (PRISMA) state-
ment standards.19 The objectives, selection criteria, out-
come measures, and methods of analysis were specified in
a study protocol, which was registered at the International
Prospective Register of Systematic Reviews in Health and
Social Care (CRD42016034166).20
Studies were identified by searching electronic biblio-
graphic databases and scanning reference lists of articles.
The following electronic bibliographic sources were
searched: PubMed/MEDLINE, SCOPUS, and the Cochrane
Central Register of Controlled Trials (CENTRAL). Search
strategies included MESH terms and key words: ((Aortic
Aneurysm, Abdominal) OR (Aortic Aneurysm)) AND
((endovascular abdominal aortic aneurysm repair) OR
(endovascular aneurysm repair) OR (EVAR) OR (stent-
graft) OR (endograft)) AND ((infection) OR (infected stent-
graft)). Thesaurus headings, search operators, and limits in
each of the above databases were adapted accordingly. The
last search was run in October 2016. No language con-
straints were applied.
Eligibility Criteria, Study Selection, and Data
Management
The study included observational studies reporting out-
comes of endograft infection after standard EVAR for an
infrarenal or aortoiliac aneurysm using conventional endo-
grafts, either inside or outside the instructions for use. Case
reports or series with <5 patients and articles providing
inadequate information on the management and outcomes
of endograft infection after EVAR were excluded. Studies
reporting endograft infection in patients with previous
EVAR or thoracic endovascular aortic repair (TEVAR)
were included only if they reported separate data for the
EVAR subgroup. Also excluded were studies involving
complex EVAR procedures, such as fenestrated and
branched repair or the chimney technique.
The search identified 1301 records of which 326 studies
were duplicates and another 700 were irrelevant to the topic
(Figure 1). Among the remaining 275 articles assessed for
eligibility, 263 were excluded due to their study design or
because they reported mixed or incomplete data. Finally, 12
studies8,13,15,16,21–28 published between 2007 and 2016 on
362 patients with infected endografts were included in the
qualitative and quantitative analysis.
One author (C.A.) extracted relevant information includ-
ing study design and year of publication; baseline demo-
graphics and clinical characteristics of the study populations,
such as presenting symptoms, methods of treatment (exci-
sion and in situ revascularization or extra-anatomical
bypass, endovascular repair, conservative treatment),
microbiology results, and outcome data.
Risk of Bias Assessment
The methodological quality of observational cohort studies
was evaluated by 2 authors (C.A. and G.G.) using the
Newcastle-Ottawa scale (NOS).29 Using the tool, each
study was judged on 8 items, categorized into 3 groups: the
selection of the study groups, the comparability of the
groups, and the ascertainment of outcome of interest. Stars
were awarded for each quality item up to a maximum of 9
stars.
Endpoints
The primary endpoint was mortality calculated at both 30
days from the diagnosis of graft infection and during fol-
low-up. Secondary outcomes were complete or partial
endograft explantation with in situ or extra-anatomical
reconstruction and adjunctive endovascular procedures to
treat a complication of endograft infection, such as profuse
bleeding as a result an aortoenteric fistula (AEF). Outcomes
after other forms of treatment (eg, antibiotics or supportive
therapies) were also included.
Statistical Analysis
Standard descriptive statistics, reported as the mean and
95% confidence interval (CI), were used to summarize
baseline clinical information. A meta-analysis was per-
formed on 30-day/in-hospital mortality and overall (follow-
up) mortality. The pooled proportion was calculated as the
back transformation of the weighted mean of the trans-
formed proportions using the random effects model
Argyriou et al 3
Figure 1.  Study flow diagram according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.
proposed by DerSimonian-Laird. Heterogeneity across
studies was evaluated using the I2 statistic. Heterogeneity
and robustness of pooled proportions were explored by con-
ducting sensitivity analyses. Publication bias was assessed
both visually evaluating the funnel plots and mathemati-
cally using the Egger regression intercept. The meta-analy-
sis was conducted using Comprehensive Meta-Analysis
software (version 2.0; Biostat, Englewood, NJ, USA).
Results
Patient Population
The data presented in the 12 articles were derived from
single and multicenter experiences or from national regis-
tries (Table 1). A total of 362 patients (mean age 72 years;
279 men) were treated for infected endografts after previ-
ous standard EVAR with an overall reported incidence of
0.6% (95% CI 0.4% to 0.8%). Hypertension was present in
80% (95% CI 77% to 82%) of patients, whereas coronary
artery disease and tobacco use were reported in 60% (95%
CI 57% to 63%) and 50% (95% CI 47% to 53%) of the total
population, respectively. Presenting symptoms/signs and
imaging/laboratory findings were pain, fever, and leukocy-
tosis in 71% (95% CI 68% to 73%) of patients. Weight loss
and fatigue or generalized weakness in 30% (95% CI 27%
to 32%); infection/abscess in 22% (95% CI 20% to 23%);
and bleeding complications in 10% (95% CI 7% to 12%)
were common. Ten percent (95% CI 7% to 12%) of patients
were asymptomatic, with the infection being detected dur-
ing routine imaging follow-up. Other baseline demographic
and clinical characteristics of the study populations are out-
lined in Table 1.
Diagnostic Methods
The time from implantation to diagnosis of infection
ranged from 1 to 128 months (mean 25). Thirteen differ-
ent commercial endografts were involved, as shown in
Table 2; in 23%, the type of endograft was not reported.
Diagnostic modalities included computed tomography
(CT) in 92% (95% CI 88% to 93%) of patients, followed
by white blood cell (WBC) count in 37% (95% CI 34% to
40%), magnetic resonance imaging (MRI) in 12% (95%
4 Journal of Endovascular Therapy 00(0)

Table 1.  Baseline Clinical Characteristics.

First Author, Study Interval to

Year N Age, ya Type Infection, moa Symptomsb Microbiologyb Positive Imagingb
Chaufour, 201615 33 69 (57–87) MC 1 (1–26) Pain/fever (64); groin/psoas G+ (55) CT (100)
abscess (18); general
infection (18)
Smeds, 201621 180 68 (35–88) MC 22 (0.2–158) Pain (~66); fever/chills (~66) G+ (~22); G− CT (~95);
(~13); fungi (~5); WBC (~34);
polymicrobial MRI (~4)
(~35)
Davila, 201522 36 69 (54–80) MC 20 (2–82) Leukocytosis (63); pain (58); G+ (67); G− (17); CT (93);
fever (56) fungi (14); MRI (33);
anaerobes (17) WBC (75)
Capoccia, 201523 26 NR MC 20 (1–72) Fever (81); weight loss (88); G+ (20); G− (23); CT (100); WBC
weakness (92); pulsating fungi (16) (31); PET (8);
mass (19); melena (19); US (8); GI
hematemesis (4) endoscopy
(12); MRI (4)
Turney, 201424 13 75 (50–93) SC NR NR NR NR
Menna, 201425 12 75 (69–82) SC 40 (5–93) NR NR NR
Lyons, 201326 13 76 (66–88) SC NR Fever/rigor; leukocytosis; NR NR
chest/abdominal pain;
bleeding/rupture; anorexia
Arya, 201313 12 70 SC 32 Fever (33); back pain (42); G+ (25); G− (33); CT (92);
infection (17) WBC (42);
MRA/MRI (17)
Laser, 201116 9 71 (54–84) NR 33 (6–80) Back pain (44); fever/chills G+ (33); G− (44); CT (89); WBC
(33); GI hemorrhage (11); fungi (11) (33); MRI (22);
infection/abscess (22); endoscopy (11)
asymptomatic (11)
Phade, 20118 6 73 SC 29 NR NR NR
Heyer, 200927 5 67 (57–70) SC 20 Fever/chills; malaise; weight G+ (100) CT (100)
loss; back/abdominal pain;
psoas abscess
Sharif, 200728 6 73 (70–77) 2 Center 18 (5–61) Psoas abscess (33); G+ (67) CT (100);
asymptomatic (33); WBC (17)
persistent fever (17);
AEF (17)

Abbreviations: AEF, aortoenteric fistula; CT, computed tomography; G+, gram-positive cocci; G−, gram-negative cocci; GI, gastrointestinal; MC,
multicenter; SC, single center; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; NR, not reported; PET, positron emission
tomography; US, ultrasound; WBC, white blood cell.
a
Data are presented as the means with range in parentheses.
b
Data are presented as the percentage in parentheses.

CI 10% to 13%), and positron emission tomography Type of Treatment

(PET) CT in <1%. In 10% (95% CI 8% to 12%) of
patients, other adjunctive diagnostic procedures not spe-
cifically stated in the studies were employed to confirm
the diagnosis. Gram-positive cocci were isolated from the
explanted endografts in 170 (47%) patients (95% CI 42%
to 51%), Gram-negative cocci were identified in 108
(30%) patients (95% CI 27% to 32%), and fungi were
found in 37 (10%) patients (95% CI 8% to 12%); in the
remaining 47 (13%) patients (95% CI 8% to 12%), the
infection was polymicrobial.
Of the 362 patients with endograft infection, 293 (81%)
patients (95% CI 77% to 83%) underwent surgical treat-
ment. Among the surgical patients, 233 (64%) patients
(95% CI 61% to 67%) underwent in situ reconstruction, 58
(16%) patients (95% CI 13% to 18%) were treated with
extra-anatomic bypass, whereas 2 (10.5%) patients under-
went endovascular treatment (95% CI 0.5% to 12%).
Overall, conservative treatment was applied in 9 (2.5%)
patients (95% CI 21% to 43%), whereas 1 patient did not
Argyriou et al 5

Table 2.  Type of Aortic Endograft Used at the Initial ranging from 0.4 to 60.8 months (mean 4.7) based on the
Endovascular Aneurysm Repair.a results of culture and sensitivities.
Zenithb 82 (23)
Excluderc 74 (20) Outcome Synthesis
AneuRXd 33 (9)
TAGc 21 (5) Meta-analysis of the studies revealed that the pooled esti-
Talentd 14 (4) mate for 30-day/in-hospital mortality (Figure 2A) was
Ancuree 13 (4) 26.6% (95% CI 16.9% to 39.2%). Significant between-study
Endurantd 11 (3) heterogeneity was identified (I2=71%, p<0.001). The likeli-
Endologixf 10 (3) hood of publication bias was low (p=0.357; Figure 2B).
Medtronic (unknown type) 7 (2) Three studies23,26,28 reported separate 30-day mortality data
Anacondag 7 (2) for a total of 9 patients managed conservatively, with a
Viabahnc 3 (1) pooled estimate of 63.3% (95% CI 30.7% to 87.0%); the
Homemade endografts 2 (0.5) statistical heterogeneity was low (I2=0%, p=0.676). Capoccia
Ovationh 1 (0.5) et al23 reported outcomes in 2 patients undergoing endovas-
Not reported 84 (23) cular treatment for infected aortic endograft with a 30-day
a
Data are presented as the counts (percentages).
mortality of 50% (95% CI 47% to 52%).
b
Cook, Bloomington, IN, USA. The pooled estimate for overall mortality (Figure 3A)
c
W. L. Gore, Flagstaff, AZ, USA. during follow-up was 45.7% (95% CI 36.4% to 55.4%).
d
Medtronic, Minneapolis, MN, USA. Moderate statistical heterogeneity was found (I2=44%,
e
Guidant, Indianapolis, IN, USA.
f
Endologix, Irvine, CA, USA. p=0.064), and the likelihood of publication bias was low
g
Vascutek, Ltd, Inchinnan, UK. (p=0.439; Figure 3B). Three studies21,23,28 reported overall
h
Trivascular, Santa Rosa, CA, USA. (follow-up) mortality data for patients treated conserva-
tively, with a pooled estimate of 58.6% (95% CI 28.8% to
83.3%); the statistical heterogeneity was low (I2=0%,
Table 3.  Methods of Treating Infected Endografts in the Patient p=0.507). No data on secondary outcome parameters were
Cohort.a reported for meta-analysis.
In situ reconstruction 233 (64)
Extra-anatomical bypass 58 (16) Risk of Bias Assessment
Conservative treatment 9 (2.5)
Endovascular treatment 2 (1) All studies were regarded as moderate quality studies since
No treatment 1 (0.5) they were awarded with NOS scores of 5 stars (Table 4).
Not reported 59 (16)
a
Data are given as counts (percentage). Discussion
Since the first case of endograft infection reported by
receive any treatment at all because he did not consent to Chalmers et al,30 the incidence of graft-related sepsis has
any therapies. In 59 (16%) patients (95% CI 13% to 17%) been increasing, presumably because of the widespread
the method of treatment was not reported (Table 3). Aortic application of EVAR. The evidence mostly comes from
replacement with prosthetic graft was used in 58% (95% small single-center case series31 and a few multicenter
CI 52% to 62%) of patients, whereas cryopreserved studies.15,21 This is because a randomized controlled trial
allografts and autologous grafts were used in 31% (95% CI involving the treatment of infected endografts cannot be
28% to 33%) and 11% (95% CI 8% to 14%) of patients, carried out due to the ethical concerns.
respectively. There were not enough data available con- The incidence of endograft infection after EVAR is dif-
cerning whether the prosthetic grafts were rifampicin- ficult to estimate because most data derive from single-cen-
soaked, silver-coated, or triclosan-coated. Only 2 articles ter studies reporting open conversion in a limited number of
referred to either standard Dacron or rifampicin-soaked patients.5–17,21,28,32–35 It is estimated to range from 0.2% to
Dacron grafts22 and polyester silver grafts.15 Surgical treat- 8% of the total EVAR population,6,22,36 which is similar to
ment was undertaken in the emergency setting in 134 the reported incidence of graft infection after open AAA
(39%) patients (95% CI 37% to 43%),13,15,16,21,22 with no repair.37 In this meta-analysis, the incidence of endograft
further reported data considering the immediate postopera- infection was 0.6% at a mean 25 months after the index
tive and follow-up periods. Postoperatively, all patients procedure. Similar figures were reported in recent large-
received broad-spectrum antibiotic therapy for a period scale studies from other institutions.16,21,23,27,34
6 Journal of Endovascular Therapy 00(0)

Figure 2.  (A) The pooled estimate for 30-day/in-hospital mortality was 26.6% (95% CI 16.9% to 39.2%), and (B) the likelihood of
publication bias was low (p=0.357). CI, confidence interval.

Graft infection seems to be a multifactorial process.
Several risk factors have been identified, such as urgent or
emergency EVAR, endograft implantation in the radiology
suite (as opposed to the operating or hybrid theatre), periop-
erative infective complications, and adjunctive procedures
following EVAR.15,18,26 Furthermore, interval vascular pro-
cedures (eg, translumbar embolization for type II endoleak)
prior to the diagnosis of endograft infection or nonvascular
procedures (eg, spine operations or urological interven-
tions) that patients underwent after the EVAR may poten-
tially seed the endograft.21
A high index of suspicion is required for the early
detection and diagnosis of endograft infection. Patient-
and procedure-related risk factors should be considered
when EVAR-related sepsis is suspected.21 Endografts are
considered to be safe in terms of infection since the stent-
graft is delivered to the target vessel covered in its delivery
sheath. On the other hand, foreign materials provide a nidus
for microbial apposition, either by direct contact or through
hematogenous or lymphatic seeding.38 Other risk factors
representing possible sources/predisposing factors for
infection include trapped thrombus/hematoma inside the
Argyriou et al 7

Figure 3.  (A) The pooled estimate for overall mortality during follow-up was 45.7% (95% CI 36.4% to 55.4%) and (B) the likelihood
of publication bias was low (p=0.439). CI, confidence interval.

aneurysm sac, graft erosion with secondary bacterial trans-
location, and secondary endovascular or open surgical
interventions after EVAR.39,40
Several studies have shown that the initial presentation
of an infected aortic endograft is often insidious.26,27
Constitutional findings in our series included pain, fever, and
leukocytosis in nearly three-quarters. Weight loss, fatigue,
generalized weakness, infection/abscess, and bleeding com-
plications were common, whereas some patients were asymp-
tomatic. The presenting symptoms are not specific, and they
should be interpreted in the context of predisposing factors.
Surgical explantation of the endograft, performed in
more than three-quarters of the patients, is considered the
standard of care in selected cases. It is followed by recon-
struction of the aorta using autologous tissue when possi-
ble.41 Even though supraceliac clamping followed by graft
removal may be technically challenging, it is thought to be
advantageous in minimizing the risk of damaging the aor-
tic endothelium and subsequent anastomotic bleed.7,42
Restoration of blood flow can be achieved with in-situ
reconstruction or an extra-anatomical bypass.5,39 In our
study, two-thirds of patients underwent in situ reconstruc-
tion, whereas extra-anatomical bypass was performed in
<20%. Endovascular treatment was undertaken in only 2
patients, one of whom died during the perioperative
period.23 Nine patients received conservative treatment. In
8 Journal of Endovascular Therapy 00(0)

Table 4.  Newcastle-Ottawa Assessment Scale. compared with other included meta-analysis studies (39%15
vs 45.7%). Even though neoaortoiliac reconstruction with
First Author, Year Selection Comparability Outcome
autologous materials is considered the standard of care for
Chafour, 2016 15
2 0 3 the treatment of aortic graft infection,45 only 11% of patients
Smeds, 201621 2 0 3 had their aorta reconstructed with autologous vein.
Davila, 201522 2 0 3 Cryopreserved grafts were used in 23% of all reconstruc-
Capoccia, 201523 2 0 3 tions. These grafts have been shown to provide superior
Turney, 201424 2 0 3 patency and lower reinfection rates to prosthetic grafts at the
Menna, 201425 2 0 3 cost of aneurysmal degeneration and an increased risk of
Lyons, 201326 2 0 3 limb thrombosis.49,50 Gram-positive infections seem to pre-
Arya, 201313 2 0 3 dominate in patients with infected endografts, with
Laser, 201116 2 0 3
Staphylococcus aureus, Staphylococcus epidermidis, and
Phade, 20118 2 0 3
Streptococci most often isolated. Following in order of fre-
Heyer, 200927 2 0 3
quency, G-negative microbes, anaerobes, and fungi are also
Sharif, 200728 2 0 3
found in the endograft or in the aortic aneurysm. Polymicrobial
infection constitutes at least one-third of total endograft
infections. Nearly all patients included in our review received
this group of patients, the pooled estimates of 30-day and
empirical antibiotic therapy at the time of diagnosis, the dura-
overall mortality (63.3% and 58.6%, respectively) were
tion of which varied across the studies.
high compared with the mortality figures in patients under-
going surgical treatment (26.6% and 45.7%, respectively).
These findings are consistent with other reports,31 which Limitations
suggest that nonoperative treatment is associated with a sig- Despite attempts to include all reported cases of AAA endo-
nificant mortality that is considerably higher in cases of an graft infection, it is possible that some have escaped consid-
AEF. However, due to scarce data, safe conclusions regard- eration. Specific inclusion criteria were set and data
ing the number and morbidity or mortality of patients with collected only from studies focusing on patients with endo-
AEFs cannot be drawn. graft infection after previous EVAR, providing a standard
An interesting finding in the study by Chaufour et al15 set of baseline information on which to base our analysis.
was that AEF was present in a third of patients with infected Some of the sporadic cases of infection identified in obser-
endografts, thus highlighting the fact the AEF could be a vational studies primarily reporting clinical efficacy of
more common finding than initially thought. Interestingly, EVAR were not eligible for inclusion due to a lack of mini-
the graft-enteric erosion was located in the aneurysm wall mum relevant information. Individual patient data from
and not on the central part of the endograft containing either other large observational studies, registries, and random-
hooks or barbs, a finding possibly leading to an etiology of ized clinical trials were not sought, which limits the com-
an inflammatory process of the aortic wall. There was also pleteness of the present review.
a nonsignificant trend in survival in the first year of patients An inevitable fact associated with this pooled dataset is
with no AEF compared with those who had AEF, a finding the heterogeneity of reported data in terms of type of endo-
that possibly makes AEF more common than we already graft, cause of infection, modes and setting of treatment,
presume.15 Procedures performed in the emergency setting and reported outcomes. Comparisons between treatments
have been reported to have higher morbidity and mortality were not possible. In particular, the considerably higher
risk than elective procedures.8,11 mortality in patients undergoing conservative treatment
Interestingly, synthetic grafts were used for aortic recon- compared to the surgical arm is questionable since they rep-
struction in more than half of the patients. Autologous vein resent only 2.5% of the study population. Furthermore, no
reconstruction has been shown to be associated with a longer further analysis could be performed to investigate predic-
operative procedure and venous morbidity.24,43,44 Impregnating tive factors of outcomes due to the lack of relevant informa-
the prosthetic graft with rifampicin decreases the risk of rein- tion provided by the individual studies. The methodological
fection compared to standard synthetic grafts; however, quality score was moderate or low for several of the studies
reconstruction with prosthetic grafts confers inferior results included in the review.
to autologous vein or cryopreserved allografts.21,45–48 These Potential biases in the review process should be also
data contradict the study by Chaufour et al,15 in which the being considered; the study selection and data extraction of
use of rifampicin-soaked synthetic grafts were used only in the included studies were performed by a single author.
almost 17% of patients with endograft infection, though Only 3 medical electronic bibliographic databases were
their more extensive use of nonsynthetic grafts probably searched, and the “gray literature” or major journals in vas-
offered a survival benefit at least in the first month cular and endovascular surgery were not searched. No
Argyriou et al 9
assistance from a clinical information specialist was pro-
vided because of lack of funding.
Conclusion
There is insufficient low-quality evidence on the manage-
ment and outcomes of stent-graft infection after EVAR.
Aortic endograft infection is a rare complication after
EVAR, with a considerable associated mortality. Surgical
treatment with complete explantation of the infected endo-
graft seems to be the optimal management in selected
patients. Despite the relatively small number of patients in
the conservative group, supportive medical treatment with-
out surgical intervention has a significant associated mor-
tality. Further research from national and international
registries and prospective multicenter studies is required to
define the role and outcomes of specific surgical treatments
and predictive factors of outcomes in the management of
post-EVAR stent-graft infection.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
References
1. Jean-Baptiste E, Hassen-Khodja R, Bouillanne PJ, et al.
Endovascular repair of infrarenal abdominal aortic aneurysms
in high-risk-surgical patients. Eur J Vasc Endovasc Surg.
2007;34:145–151.
2. Sicard GA, Zwolak RM, Sidawy AN, et al; Society for
Vascular Surgery Outcomes Committee. Endovascular
abdominal aortic aneurysm repair: long-term outcome mea-
sures in patients at high-risk for open surgery. J Vasc Surg.
2006;44:229–236.
3. Greenhalgh RM, Brown LC, Kwong GP, et al, EVAR trial
participants. Comparison of endovascular aneurysm repair
with open repair in patients with abdominal aortic aneurysm
(EVAR trial 1), 30-day operative mortality results: ran-
domised controlled trial. Lancet. 2004;364:843–848.
4. Harris PL, Vallabhaneni SR, Desgranges P, et al. Incidence
and risk factors of late rupture, conversion, and death after
endovascular repair of infrarenal aortic aneurysms: the
EUROSTAR experience. European Collaborators on Stent/
graft techniques for aortic aneurysm repair. J Vasc Surg.
2000;32:739–749.
5. Numan F, Gulsen F, Solak S, et al. Management of endograft
infections. J Cardiovasc Surg (Torino). 2011;52:205–223.
6. Verzini F, Cao P, De Rango P, et al. Conversion to open
repair after endografting for abdominal aortic aneurysm:
causes, incidence and results. Eur J Vasc Endovasc Surg.
2006;31:136–142.
7. Jimenez JC, Moore WS, Quinones-Baldrich WJ. Acute and
chronic open conversion after endovascular aortic aneurysm
repair: a 14-year review. J Vasc Surg. 2007;46:642–647.
8. Phade SV, Keldahl ML, Morasch MD, et al. Late abdominal
aortic endograft explants: indications and outcomes. Surgery.
2011;150:788–795.
9. Botsios S, Bausback Y, Piorkowski M, et al. Late open conver-
sion after endovascular aneurysm repair. Interact Cardiovasc
Thorac Surg. 2014;19:622–626.
10. Kelso RL, Lyden SP, Butler B, et al. Late conversion of aortic
stent grafts. J Vasc Surg. 2009;49:589–595.
11. Brinster CJ, Fairman RM, Woo EY, et al. Late open conver-
sion and explantation of abdominal aortic stent grafts. J Vasc
Surg. 2011;54:42–46.
12. Kouvelos G, Koutsoumpelis A, Lazaris A, et al. Late open
conversion after endovascular abdominal aortic aneurysm
repair. J Vasc Surg. 2015;61:1350–1356.
13. Arya S, Coleman DM, Knepper J, et al. Outcomes after late
explantation of aortic endografts depend on indication for
explantation. Ann Vasc Surg. 2013;27:865–873.
14. Ducasse E, Calisti A, Speziale F, et al. Aortoiliac stent graft
infection: current problems and management. Ann Vasc Surg.
2004;18:521–526.
15. Chaufour X, Gaudric J, Goueffic Y, et al; AURC (French
University Surgeons Association) collaborators. A multi-
center experience with infected abdominal aortic endograft
explantation. J Vasc Surg. 2017;65:372–380.
16. Laser A, Baker N, Rectenwald J, et al. Graft infection after
endovascular abdominal aortic aneurysm repair. J Vasc Surg.
2011;54:58–63.
17. Moulakakis KG, Mylonas SN, Antonopoulos CN, et al.
Comparison of treatment strategies for thoracic endograft
infection. J Vasc Surg. 2014;60:1061–1071.
18. Hobbs SD, Kumar S, Gilling-Smith GL. Epidemiology and
diagnosis of endograft infection. J Cardiovasc Surg (Torino).
2010;51:5–14.
19. Moher D, Liberati A, Tetzlaff J, et al, PRISMA Group.
Preferred reporting items for systematic reviews and meta-
analyses: the PRISMA statement. BMJ. 2009;339:b2535.
20. Tursunbayeva A, Pagliari C, Bunduchi R, et al. Human
resource information systems in health care: protocol for a
systematic review. JMIR Res Protoc. 2015;4:e135.
21. Smeds MR, Duncan AA, Harlander-Locke MP, et al.
Treatment and outcomes of aortic endograft infection. J Vasc
Surg. 2016;63:332–340.
22. Davila VJ, Stone W, Duncan AA, et al. A multicenter experi-
ence with the surgical treatment of infected abdominal aortic
endografts. J Vasc Surg. 2015;62:877–883.
23. Capoccia L, Speziale F, Menna D, et al. Preliminary results
from a national enquiry of infection in abdominal aortic endo-
vascular repair (Registry of Infection in EVAR–R.I.EVAR).
Ann Vasc Surg. 2016;30:198–204.
24. Turney EJ, Steenberge SP, Lyden SP, et al. Late graft explants
in endovascular aneurysm repair. J Vasc Surg. 2014;59:
886–893.
25. Menna D, Capoccia L, Sirignano P, et al. Infective etiology
affects outcomes of late open conversion after failed endovas-
cular aneurysm repair. J Endovasc Ther. 2015;22:110–115.
10 Journal of Endovascular Therapy 00(0)
26. Lyons OT, Patel AS, Saha P, et al. A 14-year experience with
aortic endograft infection: management and results. Eur J
Vasc Endovasc Surg. 2013;46:306–313.
27. Heyer KS, Modi P, Morasch MD, et al. Secondary infections
of thoracic and abdominal aortic endografts. J Vasc Interv
Radiol. 2009;20:173–179.
28. Sharif MA, Lee B, Lau LL, et al. Prosthetic stent graft infec-
tion after endovascular abdominal aortic aneurysm repair. J
Vasc Surg. 2007;46:442–448.
29. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa
Scale (NOS) for assessing the quality if nonrandomized stud-
ies in meta-analyses. http://www.ohri.ca/programs/clinical_
epidemiology/oxford.htm. Accessed June 25, 2017.
30. Chalmers N, Eadington DW, Gandanhamo D, et al. Case
report: infected false aneurysm at the site of an iliac stent. Br
J Radiol. 1993;66:946–948.
31. Moulakakis KG, Sfyroeras GS, Mylonas SN, et al. Outcome
after preservation of infected abdominal aortic endografts. J
Endovasc Ther. 2014;21:448–455.
32. Casey K, Hernandez-Boussard T, Mell MW, et al. Differences
in readmissions after open repair versus endovascular aneu-
rysm repair. J Vasc Surg. 2013;57:89–95.
33. Vogel TR, Dombrovskiy VY, Carson JL, et al. Infectious
complications after elective vascular surgical procedures. J
Vasc Surg. 2010;51:122–129.
34. Cernohorsky P, Reijnen MM, Tielliu IF, et al. The rele-
vance of aortic endograft prosthetic infection. J Vasc Surg.
2011;54:327–333.
35. Vogel TR, Dombrovskiy VY, Graham AM, et al. The
impact of hospital volume on the development of infec-
tious complications after elective abdominal aortic sur-
gery in the Medicare population. Vasc Endovascular Surg.
2011;45:317–324.
36. Verhoeven EL, Tielliu IF, Prins TR, et al. Frequency and out-
come of re-interventions after endovascular repair for abdom-
inal aortic aneurysm: a prospective cohort study. Eur J Vasc
Endovasc Surg. 2004;28:357–3364.
37. O’Hara PJ, Hertzer NR, Beven EG, et al. Surgical manage-
ment of infected abdominal aortic grafts: review of a 25-year
experience. J Vasc Surg. 1986;3:725–731.
38. Murphy EH, Szeto WY, Herdrich BJ, et al. The manage-
ment of endograft infections following endovascular tho-
racic and abdominal aneurysm repair. J Vasc Surg. 2013;58:
1179–1185.
39. Setacci C, De Donato G, Setacci F, et al. Management of
abdominal endograft infection. J Cardiovasc Surg (Torino).
2010;51:33–41.
40. Fiorani P, Speziale F, Calisti A, et al. Endovascular graft
infection: preliminary results of an international enquiry. J
Endovasc Ther. 2003;10:919–927.
41. Blanch M, Berjon J, Vila R, et al. The management of aortic
stent-graft infection: endograft removal versus conservative
treatment. Ann Vasc Surg. 2010;24:554.e1–544.e5.
42. Forbes TL, Harrington DM, Harris JR, et al. Late conversion
of endovascular to open repair of abdominal aortic aneu-
rysms. Can J Surg. 2012;55:254–258.
43. Bockler D, Probst T, Weber H, et al. Surgical conversion
after endovascular grafting for abdominal aortic aneurysms.
J Endovasc Ther. 2002;9:111–118.
44. Ali AT, Modrall JG, Hocking J, et al. Long-term results of the
treatment of aortic graft infection by in situ replacement with
femoral popliteal vein grafts. J Vasc Surg. 2009;50:30–39.
45. Chung J, Clagett GP. Neoaortoiliac System (NAIS) procedure
for the treatment of the infected aortic graft. Semin Vasc Surg.
2011;24:220–226.
46. Lew W, Moore W. Antibiotic-impregnated grafts for aortic
reconstruction. Semin Vasc Surg. 2011;24:211–219.
47. Chaar CI, Eid R, Park T, et al. Delayed open conversions after
endovascular abdominal aortic aneurysm repair. J Vasc Surg.
2012;55:1562–1569.e1.
48. Gambardella I, Blair PH, McKinley A, et al. Successful
delayed secondary open conversion after endovascular repair
using partial explantation technique: a single-center experi-
ence. Ann Vasc Surg. 2010;24:646–654.
49. Harlander-Locke MP, Harmon LK, Lawrence PF, et al. The
use of cryopreserved aortoiliac allograft for aortic reconstruc-
tion in the United States. J Vasc Surg. 2014;59:669–674.
50. Touma J, Cochennec F, Parisot J, et al. In situ reconstruction in
native and prosthetic aortic infections using cryopreserved arte-
rial allografts. Eur J Vasc Endovasc Surg. 2014;48:292–299.