Cite this article as: Martin C, Thony F, Rodiere M, Bouzat P, Lavagne P, Durand M et al. Long-term results following emergency stent graft repair for traumatic rupture
of the aortic isthmus. Eur J Cardiothorac Surg 2017;51:767–72.
Long-term results following emergency stent graft repair for
traumatic rupture of the aortic isthmus†
Cecile Martin
a,
*, Frederic Thony
b
, Mathieu Rodiere
b
,PierreBouzat
c
, Pierre Lavagne
c
,MichelDurand
d
and
Olivier Chavanon
a,e
a
Thoracic and Vascular Centre, Cardiac Surgery Clinic, Grenoble University Hospital, Grenoble, France
b
Central Radiology and Medical Imaging Service, Grenoble University Hospital, Grenoble, France
c
Surgical Polyvalent Intensive Care Unit, Grenoble University Hospital, Grenoble, France
d
Cardiovascular and Thoracic Intensive Care Unit, Grenoble University Hospital, Grenoble, France
e
HP2 Laboratory, Universite´ Grenoble Alpes, Grenoble, France
* Corresponding author. Service de Chirurgie Cardiaque CHU Grenoble Hoˆ pital Nord, CS 10217, 38043 Grenoble Cedex 9, France. Tel: +33-4-76765462;
fax: +33-4-76765281; e-mail: cmartin2@chu-grenoble.fr (C. Martin).
Received 7 April 2016; received in revised form 18 September 2016; accepted 10 October 2016
Abstract
OBJECTIVES: Endovascular repair of traumatic injury of the aortic isthmus is a safe technique that has shown good short-term results.
However, the future of these stent grafts remains unexamined, especially in relation to young patients.
METHOD: Between January 2000 and December 2014, 60 patients were treated with endovascular aortic stent graft for injury of the aortic
isthmus. Follow-up was done by computed tomography scans with intravenous contrast or magnetic resonance imaging associated with a
chest X-ray in order to control the stent graft.
RESULTS: In total, 48 men (80%) were included; the average age was 43 ± 17 years [17; 79]. The median time between the accident and
endovascular repair was 6 h. Endovascular repair was successful in all cases with no cerebrovascular or paraplegia after treatment.
Seventeen patients (27.3%) received a total coverage of the left subclavian artery; one of them received a subclavian carotid bypass. Mean
follow-up was 5 years with a maximum of 14 years. There was no repeat surgery related to the aorta during follow-up. No stent graft fail-
ure, neurological or ischaemic event related to the stent graft was noted. One patient had a type 1 endoleak without any reintervention.
The survival rate was 86.5% in 1 year, 81.6% in 5 years and 75.3% in 10 years.
CONCLUSIONS: Treatment of injuries of the aortic isthmus with stent graft seems to be a safe long-term technique; we did not notice any
event related to the stent graft during the follow-up.
Keywords: Aorta • Trauma • Blunt • Endovascular procedure • Outcomes
INTRODUCTION
Traumatic injury of the aortic isthmus is a serious lesion occur-
ring primarily in a context of multiple traumas associated with le-
sions that can all individually threaten a patient’s life [1]. The
complexity of the injury and the haemodynamic precariousness
of these patients complicate medical care. Endovascular aortic
stent grafts have quickly found their place in this indication as
they make the treatment of these patients easier [2]. Many teams
including ours [3, 4] have also published their results which show
a lower mortality rate compared with open surgical repair,
decreased morbidity and especially a lower rate of medullar
ischaemia. Recent publications provided data on the long-term
outcomes and durability of aortic stent grafts in this population
[5, 6]. But more results are needed and questions persist on the
long-term results of these aortic endografts, in particular for
younger patients with aortic diameters that are smaller than the
aortic diameter of patients with aneurysms usually treated by
endovascular repair. This study is one of the largest single-centre
cohorts of patients with traumatic aortic injury treated by endo-
graft. The aim of this study is to report the long-term outcomes
of patients with traumatic aortic injury treated by endograft.
MATERIALS AND METHODS
We collected data concerning patients presenting with trauma
and traumatic aortic injury. We retrospectively reviewed pro-
spective data that were collected on patients admitted to our in-
stitution for traumatic injury of the aortic isthmus, between
January 2000 and December 2014. This study received the
†Presented at the 68th Annual Meeting of the French Society of Cardiothoracic
and Vascular Surgery, Marseille, France, 10–13 June 2015.
ENDOVASCULAR
AORTIC SURGERY
V
C
The Author 2016. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
European Journal of Cardio-Thoracic Surgery 51 (2017) 767–772 ORIGINAL ARTICLE
doi:10.1093/ejcts/ezw369 Advance Access publication 30 December 2016
Downloaded from https://academic.oup.com/ejcts/article/51/4/767/2770805 by guest on 20 August 2022
approval of the local ethics committee. Management and follow-
up was performed according to the standard of care [7].
On admission, after clinical evaluation by the medical and sur-
gical team, patients had a whole-body computed tomography
(CT) scan with intravenous contrast if their haemodynamic status
so allowed (diagnosis confirmation, measurement of aortic diam-
eter and further assessment of the lesion). Aortic segments
studied are defined as following: segment 1 from the sinotubular
junction to the proximal brachiocephalic artery, segment 2 from
the proximal brachiocephalic artery to the distal left subclavian
artery (LSA) and segment 3 from the distal LSA to the aorta at the
level of the diaphragmatic hiatus. The traumatic aortic lesion was
analysed according to the classification described by Azizzadeh
et al. [8]. In case of haemodynamic instability and a high suspi-
cion of isthmic injury, patients were transferred directly to the
angiography room for diagnosis and therapeutic care.
Once the diagnosis of aortic injury of the isthmus was con-
firmed, a mixed surgical and radiological team first performed
endovascular repair of the aortic lesion. Targeted systolic blood
pressure before treatment was <_120 mmHg. Because of a fre-
quent bleeding state, systemic heparin was not used. Patients
were placed under general anaesthesia and one of their femoral
arteries was surgically exposed. A graduated pigtail catheter
(Boston Scientific) was inserted in the ascending aorta through
the left humeral artery for angiographic control and to mark out
the LSA ostium as previously described [3]. If the distance be-
tween the lesion and the distal LSA ostium was <1 cm, the LSA
was covered with the stent graft. If the distance between the le-
sion and the distal LSA ostium was >1 cm, the LSA was not cov-
ered with the stent graft.
Endografts used were nitinol self-expandable devices,
Excluder
V
R
(W.L. Gore and Associates, Flagstaff, AZ, USA) for 5 pa-
tients, and Talent
V
R
or Valiant
V
R
(Medtronic, Santa Rosa, CA, USA)
for, respectively, for 17 and 38 patients. The stent graft was over-
sized by 10–20%. In case of an endoleak, the endograft was
moulded with a compliant balloon. After endovascular repair, pa-
tients were given antiplatelet treatment (aspirin 160 mg/day) after
post-traumatic bleeding was controlled. The duration of antipla-
telet treatment was 1 year.
A follow-up CT-scan was performed after 1 week and after 3, 6
and 12 months, and then on a yearly basis. A clinical follow-up
was also performed during these follow-up CT-scan appoint-
ments (clinical examination, blood pressure measurement on
both arms). Patients who had left our region were contacted
again by phone and imaging examinations performed outside
our centre were viewed and analysed. Patients who were lost to
follow-up were contacted by phone and clinical evaluation ques-
tionnaires were sent to referring physicians.
Results are expressed in mean ± standard deviation [extremes]
or numbers (percentage). Quantitative data were compared
by the Wilcoxon test. The significance threshold was set for a
P-value <0.05. Statistics were performed using R software (3.1.0).
RESULTS
Of the 63 patients admitted for traumatic injury of the aortic isth-
mus, 3 died in the hospital before endovascular treatment and
were not included. A total of 60 patients underwent endovascular
stent graft repair. General characteristics are described in Table 1.
In our cohort, road accidents represent more than half of the
aetiologies (56.7%), followed by accidental or deliberate falls
(23.3%) and mountain accidents (11.7%). The severity degree of
patients is specified in Table 2. All patients treated for traumatic
aortic injury suffered from multiple traumas. We did not find any
isolated traumatic aorta injury. The most frequently associated le-
sions were brain injuries followed by bone fractures and abdom-
inal injuries (liver, spleen or mesentery). Among the patients with
brain injuries, 19% had severe brain trauma with a Glasgow
score <7 initially. Furthermore, although the aortic lesion was
treated as a priority, we performed a total of 9 non-aortic sur-
geries before endovascular repair in 7 patients. In fact, 1 of these
patients also received three interventions before aortic endovas-
cular repair. These involved life-threatening injuries (visceral hae-
morrhage in 5 cases), or the functional prognosis of a limb in
Table 1: Patients’ characteristics (n = 60)
Variables Number of
patients (%)
Demographic
Male/female 48/12 (80/20)
Age (years) 43 ± 17 [17; 79]
Systemic arterial hypertension 8 (13.3)
Accident type
Car 16 (26.7)
Motorbike 16 (26.7)
Fall 14 (23.3)
Mountain 7 (11.7)
Bike 2 (3.3)
Pedestrian 2 (3.3)
Unknown 3 (5)
Table 2: Patient’s injury severity
Variables Number of
patients (%)
Trauma severity scores
Glasgow score
>_11 42 (70)
>_7–11 6 (10)
<7 9 (15)
Glasgow score not specified 3 (5)
Injury severity score 38 ± 17 [18; 75]
Associated injuries
Abdominal injury 26 (43.3)
Haemothorax 32 (53.3)
Pulmonary contusion 36 (60)
Myocardial contusion 17 (28.3)
Pelvic injury 23 (38.3)
Bone fractures 33 (55)
Haemodynamics and transfusion
Haemorrhagic shock 19 (31.6)
Red blood cell (number per patient) 8.7 ± 6 [0; 20]
Fresh frozen plasma (number per patient) 6.1 ± 7 [0; 20]
Platelet concentrates (number per patient) 0.9 ± 1 [0; 3]
Nonaortic operations before
endovascular repair
9 (15)
Hip dislocation reduction or open fracture 3 (33.3)
Splenectomy 3 (33.3)
Right thoracotomy for haemothorax 1 (11.1)
Bladder suture 1 (11.1)
Amputation 1 (11.1)
768 C. Martin et al. / European Journal of Cardio-Thoracic Surgery
Downloaded from https://academic.oup.com/ejcts/article/51/4/767/2770805 by guest on 20 August 2022
patients with stable aortic injury (hip dislocation reduction or
open fracture in 3 cases and limb amputation in 1 case).
Early results of aortic stent graft repair are specified in Table 3.
The median time between the accident and the endovascular
treatment was 6 h. Of the entire cohort, only 3 patients were
treated beyond 7 days: 1 showed a Grade 3 aortic lesion moni-
tored initially because of a severe coma and the aortic lesion was
treated on Day 44. The second had a dissecting haematoma,
which was initially monitored. It became progressive on imaging
control and was treated on Day 39. The last patient showed a
Grade 1 lesion that was medically treated at first and a stent graft
was placed later because of an increase in diameter observed on
the CT-scan at 6 months. Note that Grade 3 aortic lesions are the
most frequently found in our series. Only 1 patient had a Grade 4
aortic lesion and presented with severe brain and liver damage.
This patient was haemodynamically unstable and required a
transfusion of 18 packed red blood cells. Quick endovascular re-
pair controlled his haemorrhagic state. Unfortunately, the patient
died of head injuries.
The endovascular procedure was a technical success in all pa-
tients and there was no surgical conversion. Mean diameter of
stent grafts was 26.5 mm ± 3.5 [22; 34] (median = 26 mm). We had
no evidence of neurological or vascular complication. Complete
coverage of the LSA was required in <30% of the cases without
acute ischaemia or emergency revascularization of the left arm.
There were 5 patients with minimal type 1 endoleaks at dis-
charge. Follow-up showed regression of these endoleaks on CT-
scans without any treatment and only 1 of these showed a per-
sistent type 1 endoleak after 1 year.
Hospital mortality was 13.3%, i.e. 8 patients. Half of these pa-
tients died from brain injury (brain death in 4 patients). Two (2)
died in a context of uncontrolled haemorrhagic shock (splenic
bleeding and cardiac arrest and resuscitation before stent graft
treatment for one), 1 died in a context of multiple organ failure
and the last one died from septic shock in a context of gastro-
intestinal ischaemia (mesenteric tear). Except for the patient who
had a cardiac arrest, all the other patients had their aortic lesion
controlled and there was no death directly related to the aortic
stent graft.
Long-term results are described in Table 4. Mean clinical
follow-up was 5 years with a maximum follow-up of 14 years
(minimum 1 year) and median follow-up was 4 years. Mean
radiographic follow-up was 4 years with a maximum of 14 years
and median radiographic follow-up was 3 years. There were 16
patients whose age at the time of treatment was <30 years
(26.7% of the cohort). Mean follow-up in this sub-group was 4
years with a maximum follow-up of 9.5 years. In addition, the
youngest patient was 17 years old at the time of the accident.
Follow-up for this patient is now more than 8 years. More than
half of the patients lost to follow-up were in this sub-group (5
patients).
A total of 9 patients were lost to follow-up. Among these pa-
tients, 2 were foreign patients from whom we have had no news
since repatriation. Three patients died in the follow-up period.
There was no autopsy but causes of death were clearly identified
by questioning family or the patient’s general practitioner. The
first one died 2 years after stent graft repair (75 years old). The
patient had had severe brain trauma 10 years before and suffered
from neurological after-effects. In addition, he suffered from
chronic alcoholism. The cause of death was related to recurrent
falls in the context of these neurological after-effects and chronic
alcoholism. In this case, a stent graft did not cover the LSA. The
second patient died 9 years after stent graft repair (88 years old).
The cause of death was unknown but no aortic event was re-
ported. The LSA was partially covered by a stent graft and was
permeable on CT-scan control. The last patient suffered from se-
vere depression and aortic trauma occurred in a context of a de-
liberate fall. This patient died 4 years after stent graft repair (62
years old). The cause of death was suicide. No secondary event
related to the aorta or any neurological or ischaemic event
related to the aortic stent graft was noted. We did not note any
stent graft failure or collapse. However, 1 patient showed a min-
imal thrombus at the distal portion of the stent graft. This patient
prematurely discontinued antiplatelet therapy. The thrombus was
medically treated with oral anticoagulants. Monitoring by imag-
ing showed a stable thrombus.
We noted the occurrence of systemic arterial hypertension
shortly after endograft repair in 7 patients (5 male/2 female). The
Table 3: Early results of aortic stent graft repair (n = 60)
Variables Number of
patients (%)
Aortic injury classification
Grade 1 1 (1.7)
Grade 2 0
Grade 3 50 (83.3)
Grade 4 1 (1.3)
Aortic lesion not specified 8 (13.3)
Short-term outcomes
Median time between accident
and endovascular repair (h)
6
Implantation success 60 (100)
Conversion 0
Coverage of the left subclavian artery 26 (43.3)
Complete 17 (28.3)
Partial 9 (15)
Paraplegia 0
Cerebrovascular accident 0
Repeat surgery 0
Type 1 endoleak 5 (8.3)
SAPS II 37 ± 17 [8; 84]
Hospital Mortality 8 (13.3)
ICU length of stay (days) 10 ± 10 [0; 53]
Total hospital stay (days) 21 ± 17 [0; 73]
SAPS II: simplified acute physiology score.
Table 4: Long-term results after endovascular repair of trau-
matic aortic injury (n = 52)
Variables Number of
patients (%)
Number of patients who died in
the follow-up period
3 (5.8)
Patients lost to follow-up 9 (17.3)
Type 1 endoleak 1 (1.9)
Partial thrombosis 1 (1.9)
Decrease in left radial pulse 4 (23.5)
Left arm claudication 2 (11.8)
Carotid–subclavian bypass 1 (5.9)
Systemic arterial hypertension 7 (13.5)
ENDOVASCULAR
AORTIC SURGERY
769C. Martin et al. / European Journal of Cardio-Thoracic Surgery
Downloaded from https://academic.oup.com/ejcts/article/51/4/767/2770805 by guest on 20 August 2022
mean age of this sub-group was 49.9 years ± 16 [32; 73] and the
median age was 42 years. None of these 7 patients had a history
of hypertension before the accident.
In addition, we compared the diameters of anatomical aortic
segments 1–3 (downstream to the stent graft) between preopera-
tive imaging (when preoperative or immediate postoperative
imaging was available) and the later imaging conducted as a part
of follow-up. We did not find any significant increase in the
diameters of segments 1 and 3. In comparison, there was a sig-
nificant increase in the segment 2 diameters (25.6 vs 27.2 mm ± 3,
P = 0.019).
A Kaplan–Meier survival analysis of patients treated with endo-
vascular stent graft for aortic injury of the isthmus is presented in
Fig. 1. Survival was 86.5% at 1 year, 81.6% at 5 years and 75.3%
10 years after trauma.
DISCUSSION
The demographic and general characteristics of our cohort are
comparable with other selected series in a recent meta-analysis
[9]. The severity of associated injuries, especially brain damage,
and the natural history of traumatic aortic injury with a risk of
spontaneous rupture (2% [10]–87.5% [11]) led us to quickly pro-
ceed with endovascular repair. Of course, medical treatment of
traumatic aortic injury is fundamental and based on control of
both blood pressure (target <120 mmHg systolic blood pressure
[9] or <80 mmHg average blood pressure [7]) and heart rate [12].
Indeed, medical treatment has reduced the rate of spontaneous
rupture of the aorta [11, 13]. Even if some authors found it bene-
ficial to delay aortic injury treatment [10, 14], or to have a con-
servative attitude towards traumatic aortic lesion [12, 13], their
data showed that death rate by spontaneous rupture is not equal
to zero. In our opinion, even if it is a small risk, it is hardly ac-
ceptable especially since it is often a young population. For ex-
ample, Caffarelli et al. [12] or Paul et al. [13] were in favour of
conservative treatment for traumatic aortic lesion. But their data
concerned only few patients with small aortic lesion or patients
with a low rate of brain damage. In the specific case of brain
damage, low systolic blood pressure may worsen brain lesion.
Even if they had good results with conservative treatment, there
were no clinical or radiological criteria allowing them to affirm
which patient should benefit from this conservative treatment or
not.
Another argument in favour of early treatment of aortic injury
is the deleterious role of hypotension on cerebral perfusion pres-
sure in patients with severe traumatic brain injury [15–17]. Faster
endovascular repair allows the adaptation of the mean arterial
pressure, the use of vasopressors and loading necessary to main-
tain good systemic perfusion and sufficient cerebral perfusion
pressure to avoid spontaneous rupture of the aorta.
Based on our previous results that compare endovascular and
surgical treatment [3] we have been treating all patients with aor-
tic injury of the isthmus by stent graft repair since the early
2000s. Mortality in our series was significantly lower than the re-
ported mortality (33%) in recent rounds of surgical treatment
[18]. Simple installation in supine position, fast stent graft repair
and the possible absence of any anticoagulant therapy can treat
the majority of multiple trauma patients with stent graft.
The trend to use this technique was controversial because of
the lack of data on long-term outcomes of aortic stent grafts, es-
pecially for young patients with aortic diameters smaller than
those of the patients usually treated for aneurysmal pathologies
[19]. In addition, traumatic injury of the aorta is rare (47/7112
chest trauma in 5 years 0.66% [13]) and publications are mostly
small single-centre series, retrospective views from which it is dif-
ficult to draw formal conclusions. For instance, in a meta-analysis
of 139 studies [9], the median sample size was 15 patients and
the median follow-up 2 years. Despite the scarceness of long-
term data, some teams have recently published good encourag-
ing long-term results of endovascular repair of aortic transection
[2, 20]. These data reinforce our strategy to treat aortic transec-
tion with endografts, even in young patients.
Another argument against aortic stent graft is the risk of radi-
ation exposure with CT-scan follow-up, in particular in young pa-
tients. But angio resonance magnetic imaging is an alternative to
decrease radiation exposure. In that case, a chest X-ray is needed
to control the stent graft and aortic diameter is measured with
resonance magnetic imaging.
Regarding the management of LSA revascularization in case of
complete coverage of the LSA in acute endovascular repair, recom-
mendations suggest an individual decision according to the pa-
tient’s clinical status [21]. We have never performed LSA
revascularization before endovascular repair. Firstly, blunt aortic
trauma is a life-threatening lesion and in that case patients need ur-
gent endovascular repair. Secondly, blunt aortic lesions require a
shorter endograft than a thoracic aortic aneurysm or dissection
(mean endograft length (mm) was 112 ± 13 [100; 150]). In that case,
the risk of paraplegia is low. Thirdly, we have never found any ana-
tomical risk factor that exposes patients to the possibility of vital
organ hypoperfusion. Our data showed that complete coverage of
the LSA had little impact on the long-term outcome in patients pre-
senting with aortic trauma treated by endograft. Similar to
Antonello et al. [22], patients with initial symptoms see their func-
tional impact decrease gradually until it completely disappears so
they can return to their activities without functional impairment. In
our series, only 1 patient required a carotid-to-subclavian bypass a
few months after stent graft repair. Klocker et al. [23]andMcBride
et al. [24] also studied ischaemic and functional outcomes on the
left arm after coverage of the LSA without finding significant differ-
ences in terms of left arm perfusion or functional symptoms.
We have a conservative approach to type 1 endoleaks
observed immediately after the procedure (Fig. 2). Of the 5
Figure 1: Kaplan–Meier survival curve after aortic trauma treated by stent graft
repair.
770 C. Martin et al. / European Journal of Cardio-Thoracic Surgery
Downloaded from https://academic.oup.com/ejcts/article/51/4/767/2770805 by guest on 20 August 2022
patients with an endoleak, only 1 had a persistent endoleak,
which was monitored by imaging. For the other 4 patients,
the addition image was no longer visible on the control CT-scan
(Fig. 3). One hypothesis is that the addition image identified as
an endoleak was a contrast agent that remained sequestered in
the aneurysmal pocket and was excluded after deployment of
the stent graft. Another hypothesis is related to the specific char-
acteristics of the nitinol stent graft that may continue to expand
after stent graft deployment, permanently closing secondary and
minimal endoleaks.
We noted arterial hypertension in 7 patients after endograft
through the follow-up. Surprisingly, this hypertension appeared
shortly after stent graft repair and we did not find any aetiology
of a secondary arterial hypertension (such as renal artery stenosis,
surrenal pathology). Three of these patients were under 40 years
of age and the probability of an undiagnosed hypertension is
low. In fact, the impact of thoracic aortic endografts is unknown
and we wondered whether there is a link between thoracic aortic
endografts and systemic arterial hypertension. But we did not
find any published data supporting this hypothesis. More results
are needed to confirm whether or not there is a link between
aortic stent grafts and hypertension.
Finally, we noted a significant increase of the aortic diameter
in anatomical segment 2, which is precisely upstream to the
prosthesis. If this augmentation is statistically significant, it is only
1.6 mm over a period of an average of 4 years. Within this con-
text, we do not consider it to be clinically relevant but longer
follow-up is needed.
Limitations of our study are the monocentric and retrospective
characters of the cohort, and the small size of our sample. Of
course, the number of patients lost to follow-up is a major bias,
but the population with trauma is younger and less compliant to
follow-up. Another team reported the same problem with
follow-up, for example, Azizzadeh et al. [8] who showed that only
56% of patients were fully compliant to follow-up imaging.
Kidane et al. [25] clearly showed a significant loss to follow-up in
studies comparing treatments of traumatic aortic injuries.
However, data collected from the remaining patients provide
precious information about a rare pathology. These results are
encouraging and highlight the necessity of carrying out
multicentric studies for long-term periods as well as maintaining
records to decrease the impact of these different biases.
CONCLUSION
Stent graft repair for traumatic injury of the descending thoracic
aorta is a simple treatment, safe and quick for life-threatened pa-
tients. Long-term follow-up in our study has shown no major
complications related to the stent, the aorta or to the intentional
coverage of the LSA. Arterial hypertension was noted after stent
graft repair in a few patients. The mechanism of this hypertension
must be clarified by other more advanced studies. These good
remote results along with good early results reinforce the validity
of this therapeutic approach.
Conflict of interest: none declared.
REFERENCES
[1] Fox N, Schwartz D, Salazar JH, Haut ER, Dahm P, Black JH et al.
Evaluation and management of blunt traumatic aortic injury: a practice
management guideline from the Eastern association for the surgery of
trauma. J Trauma Acute Care Surg 2015;78:136–46.
[2] Canaud L, Marty-Ane´ C, Ziza V, Branchereau P, Alric P. Minimum 10-
year follow-up of endovascular repair for acute traumatic transection of
the thoracic aorta. J Thorac Cardiovasc Surg 2015;149:825–9.
[3] Broux C, Thony F, Chavanon O, Bach V, Hacini R, Sengel C et al. Emergency
endovascular stent graft repair for acute blunt thoracic aortic injury: a retro-
spective case control study. Intensive Care Med 2006;32:770–4.
[4] Estrera AL, Miller CC, Guajardo-Salinas G, Coogan S, Charlton-Ouw K,
Safi HJ et al. Update on blunt thoracic aortic injury: fifteen-year single-in-
stitution experience. J Thorac Cardiovasc Surg 2013;145:S154–8.
[5] Steuer J, Bjo¨rck M, Sonesson B, Resch T, Dias N, Hultgren R et al. Editor’s
choice – durability of endovascular repair in blunt traumatic thoracic
aortic injury: long-term outcome from four tertiary referral centers. Eur J
Vasc Endovasc Surg 2015;50:460–5.
[6] Khashram M, He Q, Oh TH, Khanafer A, Wright IA, Vasudevan TM et al.
Late radiological and clinical outcomes of traumatic thoracic aortic injury
managed with thoracic endovascular aortic repair. World J Surg
2016;40:1763–70.
Figure 2: Type 1 endoleak on CT-scan 1 week after endovascular repair of trau-
matic aortic injury.
Figure 3: Total regression of type 1 endoleak on CT-scan 2 months after endo-
vascular repair.
ENDOVASCULAR
AORTIC SURGERY
771C. Martin et al. / European Journal of Cardio-Thoracic Surgery
Downloaded from https://academic.oup.com/ejcts/article/51/4/767/2770805 by guest on 20 August 2022