Endovascular management of acute blunttraumatic thoracic aortic injury: A singlecenter experienceClare L. Bent, FRCR,a Matthew B. Matson, MRCP, FRCR,a Mo Sobeh, MS, FRCS,c

Ian Renfrew, MRCP, FRCR,a Rakesh Uppal, BSc, FRCS (CTh),b Michael Walsh, MS, FRCS,c,d

Karim Brohi, FRCS, FRCA,c,d and Constantinos Kyriakides, MD, FRCS,c,e London, United Kingdom

Background: Traumatic injury of the thoracic aorta is a life-threatening complication in patients who sustain decelerationor crush injuries. The magnitude of force necessary to cause blunt thoracic aortic injury results in a high proportion ofconcomitant injuries, posing a significant challenge for prioritizing management. Open surgical mortality is increased inthe presence of coexisting head, lung, and abdominal injuries. Spinal cord ischemia may occur following aorticcross-clamping and operative hypotension. Endovascular stent-graft placement offers a safe, effective, and timelytreatment option. The aim of this study was to assess our single center experience of endovascular repair following acuteblunt traumatic aortic injury.Methods: Data from thirteen consecutive patients (mean age, 43.2 years; range, 16 to 84 years) with acute blunt traumaticaortic injury treated by endovascular stent-graft insertion between October 2001 and March 2007 was prospectivelycollected. Demographics, injury characteristics, technique, and complications were recorded. Follow-up data consisted ofcomputed tomographic angiography and plain chest radiography at regular intervals. Mean and median follow-up afterstent-graft implantation were 28.9 and 29 months, respectively.Results: All patients underwent endovascular repair within a median of 9 hours from hospital presentation. Two patientsunderwent carotico-carotid bypass immediately prior to endovascular stenting during a single anesthetic. Stent-graftimplantation was technically successful in all patients. No patient required conversion to open surgical repair of the acuteblunt traumatic aortic injury. Procedure-related paraplegia was zero. Complications included proximal migration ofinitial stent-graft in one patient and iliac artery avulsion in another patient with consequent ilio-femoral bypass. Themedian hospital stay was 17 days. There were no in-hospital deaths.Conclusion: Endovascular repair is evolving as the procedure of choice for acute blunt traumatic aortic injury. Treatment oflesions that extend into the aortic arch is feasible with extra-anatomical bypass. In our study, endovascular repair of blunt

traumatic aortic injury is a safe procedure with low morbidity and a mortality rate of zero. (J Vasc Surg 2007;46:920-7.)

Thoracic aortic injury following chest trauma is poten-tially life-threatening. Immediate on-scene mortality is be-tween 80% to 90% irrespective of mechanism, be it pene-trating or blunt trauma.1-3 Of those who survive the initialinsult, 32% die within 24 hours and 74% within 2 weeks.4

Prompt diagnosis and treatment is therefore essential.Traditionally, surgical repair is the gold standard

method of treatment.5 However, complication rates are

From the Department of Radiology,a Department of Cardiothoracic Sur-gery,b Department of Vascular Surgery,c and Department of TraumaSurgery,d Barts and The London NHS Trust and Queen Mary School ofMedicine and Dentistrye.

Competition of interest: none.Abstract presented in Bent CL, Matson M, Renfrew I, Walsh M, Sobeh M,

Kyriakides C. Traumatic injury of the thoracic aorta: an endovascularapproach. BJS 2006;93(S1):95-6.

Presented in part as “Early Results of Traumatic Aortic Injury Repair” at theAssociation of Surgeons of Great Britain and Ireland (ASGBI), AnnualScientific Meeting, Edinburgh, United Kingdom, May 3, 2006.

Reprint requests: Constantinos Kyriakides, MD, FRCS, Department ofVascular Surgery, Barts and The London NHS Trust, Royal LondonHospital, London, E1 1BB UK (e-mail: constantinos.kyriakides@bartsandthelondon.nhs.uk).

CME article0741-5214/$32.00Copyright © 2007 by The Society for Vascular Surgery.

doi:10.1016/j.jvs.2007.07.032

920

high. In addition, the mechanism of injury involved fre-quently results in coexisting nonaortic injuries complicat-ing open thoracic surgery.

Controversy persists regarding the management ofacute and chronic blunt traumatic aortic injury, involvingthe indications, timing of intervention, and long-termfollow-up.3,6 The purpose of this study was to assess thefeasibility and outcomes of endovascular repair of acuteblunt traumatic aortic injury (ABTAI) in a single center.

METHODS

We prospectively collected data from all patients withan ABTAI treated in a tertiary referral center with endovas-cular stent-graft repair between October 2001 and March2007.

During the study interval, 13 patients with an ABTAItreated with endovascular repair were identified. Theirmean age was 43.2 years (range, 16 to 84 years) and 11 ofthe 13 patients (85%) were male.

The cause of aortic injury involved a motor vehiclecollision in 10 patients, motorcycle crash in two patientsand a fall of approximately 20 feet in the final patient.

Two additional patients presented with an undiag-nosed chronic BTAI and treated with endovascular repair

during the study period. Both attended hospital acutely

JOURNAL OF VASCULAR SURGERYVolume 46, Number 5 Bent et al 921

with chest pain and demonstrated a calcified thoracic pseu-doaneurysm located at the aortic isthmus on radiologicalimaging. A previous history of a significant chest traumafollowing a road traffic accident with hospital admissionwas elicited in both patients; one 30 years7 and the other 15years previously. Due to their chronicity, these patientswere excluded from our study. A further patient died oftraumatic aortic injury on arrival to hospital. Diagnosis wasconfirmed at post-mortem.

All patients were initially reviewed and stabilized by thetrauma surgery service in concordance with the AdvancedTrauma Life Support guidelines.8 Chest radiography dem-onstrated a mediastinal abnormality in all patients. Aorticinjury was diagnosed with computed tomographic (CT)angiography. Aortography was performed in two patientsprior to endovascular repair, as CT findings were equivocal.

Stent-graft insertion was accepted as the first-line treat-ment option for ABTAI. The decision to treat with endo-vascular repair is dependent upon morphology of aortic

Table I. Patient demographics

Patient Age (y) Sex Mode of injury

1 16 F RTA Pulmonary contusionBilateral hemothoraxFluid in abdomen duPelvic and bilateral fe

2 41 M RTA Bilateral hemothoraxMultiple rib, right scD10 burst fracture w

3 27 M RTA Frontal lobe contusioBilateral hemothoraxMultiple rib fracturesSplenic lacerationTibia and fibula fract

4 22 M RTA Right pneumo-hemoMultiple rib fracturesRight scapula fractur

5 56 M RTA Rib fractures6 53 M RTA Frontal lobe contusio

Bilateral hemothoraxRight pneumothoraxMultiple right rib fraRight scapula fractur

7 42 M RTA Cerebral edemaRight hemothoraxRight tension pneumMultiple right rib fra

8 56 M Fall D12 vertebral fractur9 66 F RTA Right clavicular fract

10 17 M RTA Bilateral femoral frac11 84 M RTA Temporal lobe contu

Vertebral and rib fracHemothorax

12 43 M RTA No additional injury13 39 M RTA Extradural and subdu

Right hemopneumotMultiple right rib fraC2, T12 and pelvic fDuodenal/adrenal anOpen femoral/extrem

F, Female; M, male; RTA, road traffic accident.

injury, presence of concomitant injuries complicating open

repair, availability of stent-grafts, and the operator’s prefer-ence. No patients underwent open surgical repair duringthe study period.

Concomitant injuries were documented in 12 out ofthe 13 patients with an ABTAI (92.3%) and are listed inTable I. Patients with coexisting chest injuries had ribfractures and pulmonary contusions (n � 10). Five patientshad coexisting head injuries. Vertebral, pelvic, and extrem-ity fractures were identified in four patients. Three patientshad additional intra-abdominal injuries: two had a duode-nal laceration, one of which also had a renal contusion, andone patient had a splenic contusion.

Injuries were treated in order of threat to life. Nine of13 patients underwent treatment of acute life-threateninginjuries prior to endovascular intervention. Procedures in-cluded chest drain insertion (n � 8), intracranial pressurebolt insertion (n � 5), laparotomy (n � 3), and orthopedicintervention (n � 1). Two adjunctive carotico-carotid by-pass procedures were performed prior to endovascular re-

ditional injuries Interventional delay (h)

uodenal injuryl fractures

26

and clavicular fracturesrd trauma

10

d subdural hemorrhage 6

x 25

6d subarachnoid hemorrhage

s

27

raxs, right scapula and clavicular fractures

7

2758

and subarachnoid hemorrhage 81

9emorrhage

sesney contusionsractures

4

Ad

s

e to dmora

apulaith con an

uresthora

e

n an

cturee

othoctureeureturessiontures

notedral hhoraxctureracturd kidity f

pair during a single anesthetic. One took place in the main

JOURNAL OF VASCULAR SURGERYNovember 2007922 Bent et al

surgical theaters prior to transfer into the interventionalradiology suite for stent-graft insertion; the second patientunderwent both carotid bypass and stent-graft insertion inthe interventional radiology suite. A further four orthope-dic procedures were carried out following completion ofendovascular repair.

Endovascular stent-graft procedures were performed inthe interventional radiology suite under fluoroscopic an-giographic control (Phillips Integris Allura, Phillips, Eind-hoven, the Netherlands) by a team of dedicated interven-tionalists including radiologists, and vascular andcardiothoracic surgeons.

Four different self-expanding commercially availableendovascular stent-grafts were used: Talent LPS and Val-iant (Medtronic Vascular; Sunrise, Fla), Excluder (WLGore and Associates; Sunnyvale, Calif) and Relay (BoltonMedical; Sunrise, Fla). Informed consent was obtained forendovascular treatment from the patient or consultant incharge of their care.

The dimensions of the stent-graft used were deter-mined by the anatomical configuration of the aorta asdemonstrated on contrast-enhanced helical CT images andangiographic images. Aortic arch morphology, aortic vesseldiameter along with the length, and location of lesion allcontributed to the selection process. For optimal fixation,all stent-grafts were oversized by 10% to 15% comparedwith the aortic diameter at landing zone sites. Dependingon type of thoracic stent-graft selected, aortic diameterswere measured from outer to outer wall (Medtronic Vas-cular; Sunrise, Fla), or inner to inner wall (WL Gore andAssociates; Sunnyvale, Calif; Bolton Medical; Sunrise, Fla)as per recommendations published by stent-graft manufac-turer. The younger patients within our cohort, four ofwhom were under 27 years of age, had small aortic diame-ters (18 to 20 mm). In these patients, the stent-graft sizechosen was based upon the smallest size available.

Patients were treated under general anesthesia (n � 11)

Table II. Procedural information

Patient Type of access Stent type No. of stents

1 Surgical cutdown Medtronic 12 Percutaneous Medtronic 23 Percutaneous Medtronic 14 Percutaneous Gore 15 Surgical cutdown Gore 16 Surgical cutdown Medtronic 17 Percutaneous Gore 4

8 Percutaneous Gore 2

9 Surgical cutdown BVM 110 Percutaneous Gore 111 Surgical cutdown Medtronic 112 Surgical cutdown Medtronic 113 Surgical cutdown Medtronic 1

LSA, Left subclavian artery.

or regional anesthesia with sedation (n � 2). Lumbar drains

were not used. All patients received intravenous antibioticsfor prophylaxis prior to stent-graft deployment.

A standard angiographic pigtail catheter was insertedvia percutaneous puncture (5F) into the common femoralartery or brachial artery to maintain vascular access andangiographic control throughout the procedure. Surgicalexposure (n � 6) or percutaneous puncture (n � 6) of acommon femoral artery was performed to allow introduc-tion of the stent-graft delivery device. One patient requiredaccess via the left common iliac artery during explorativelaparotomy due to small caliber vessels unable to accom-modate the stent-graft delivery device. For those patientsundergoing percutaneous repair, two Perclose suture de-vices (Abbott Laboratories; Redwood City, Calif) weredeployed at this site using a previously described “preclose”technique.9

Initially, angiographic evaluation of the aortic lesionwas performed using the pigtail catheter positioned in theascending thoracic aorta. Digital subtraction angiography(DSA) was then used to visualize exact aortic arch anatomy,location of ABTAI, and the location of the left subclavianartery (LSA). Using the planned access site, a stiff wire,either a Meier (Boston Scientific; Miami, Fla), Lunderquist(Cook Inc; Bloomington, Ind), or Amplatz (Boston Scien-tific; Miami, Fla), was then advanced into the ascendingaorta for guidance of the selected endovascular device.Technical information is listed in Table II.

Prior to device insertion, eight of the 13 patients weresystemically heparinized during the procedure with a singledose of 5000 international units of heparin; five patientshad significant head injuries contraindicating heparin ad-ministration.

The stent-graft was then introduced and position veri-fied by DSA immediately before stent-graft deployment. Atdeployment, no pharmacological methods to reduce sys-tolic arterial blood pressure were used. Completion angiog-

Size of stent Cover of LSA origin LSA patent

26 � 112 mm No Yes28 � 120 mm (2) Complete No26 � 115 mm Partial Yes26 � 100 mm No Yes34 � 200 mm Complete No28 � 120 mm Partial Yes34 � 200 mm Complete No

1/34/37 � 150 mm28 � 100 mm No Yes34 � 200 mm38 � 145 mm No Yes26 � 100 mm No Yes34 � 110 mm Partial Yes24 � 115 mm Complete No24 � 115 mm Partial Yes

3

raphy was then performed in all patients to check stent-

JOURNAL OF VASCULAR SURGERYVolume 46, Number 5 Bent et al 923

graft position and to confirm complete injury exclusionwithout evidence of an endoleak.

Two patients required carotico-carotid bypass prior toendovascular repair. In one case, this was due to extensionof the ABTAI into the aortic arch and in the second, due toa common origin of the LCCA and LSA. In the remaining11 patients, two required complete coverage of the LSAorigin, and four had partial coverage with the uncoveredmetal struts of the stent-graft (Talent LPS or Valiant) toensure optimal apposition of the stent-graft with the vesselwall. Five patients had stent-graft implantation distal to theLSA origin.

Persistence of blood flow within the LSA could beobserved in all patients. Of the four patients with completecoverage of the LSA origin (two with a carotico-carotidbypass due to LCCA and LSA coverage and two withcoverage of the LSA origin by the covered portion of thestent-graft), blood flow to the distal LSA occurred viaretrograde filling from the ipsilateral vertebral artery.

A single stent-graft was required in 10 patients, twostent-grafts for two patients and four stent-grafts for onepatient. This patient sustained a traumatic aortic dissection,requiring four stent-grafts due to refilling of the false lumendespite entry-tear coverage.

Following removal of procedural equipment, the accesssite underwent either formal surgical arterial closure orclosure with the Perclose sutures in situ, if a percutaneoustechnique had been used. No further anticoagulation wasadministered.

Follow-up examinations were performed using contrastenhanced computed tomography angiography at 3, 6, and9 months after implantation. In addition, plain chest radi-ography was performed on an annual basis to monitor forstent fractures.

RESULTS

Endovascular repair was performed within a median of9 hours and mean of 18.5 hours from presentation tohospital and was technically successful in all cases. Nopatient required conversion to open surgical repair of theABTAI. No procedure-related paraplegia or stroke oc-curred. Procedural complications occurred in two patients.The intraoperative and 30-day mortality was zero.

Morbidity following ABTAI occurred in 10 out of the13 patients due to respiratory insufficiency and orthopedicinjuries requiring external fixation devices. Procedure-related morbidity occurred in two patients as a result ofcomplications during stent-graft implantation.

In one patient, the external iliac artery was avulsed onremoval of the stent sheath following successful deploy-ment of the stent-graft. Bleeding was immediately con-trolled by the introduction of a proximal occlusion balloonvia a femoral artery and an ilio-femoral bypass was per-formed. The patient was female with a relatively smallexternal iliac artery diameter. However, preprocedure CThad demonstrated vessel diameters sufficient to accommo-

date the delivery system (�7.5 mm).

In a second patient, a proximal endoleak was identifiedduring completion angiography. The proximal stent-graft toaortic wall seal was felt to be suboptimal due to incompletegraft self-expansion and possible under-sizing. To rectify this,an attempt was made to balloon this segment. On doing so,the stent-graft itself became dislodged and migrated into theascending aorta proximal to the arch vessels, maintainingsupra-aortic vessel perfusion. A second graft of a larger diam-eter was subsequently implanted in the correct position with agood radiological result (Fig 1, a b, and c). To minimize risk ofdistal migration of the initial stent-graft this was retrievedthrough an ascending aortic arteriotomy via a mini-sternot-omy incision without the need to use cardiac bypass (thispatient had a concomitant D12 vertebral fracture).

There were no cases of groin infection or hematoma.Of the four patients with intentional occlusion of LSA

during stent-graft insertion, three experienced intermittentsymptoms of discoloration and paresthesia of the left upperlimb in the immediate postoperative period. Despite this,none has subsequently developed vertebrobasilar or leftarm ischemia necessitating revascularisation via carotid-subclavian bypass or transposition.

Median hospital stay was 17 days (range: 4 to 58 days).Procedural-related morbidity did not lengthen in-hospitalstay. Both patients with procedural complications sufferedorthopedic injuries (one patient with pelvic fractures andbilateral femoral fractures; one patient with a D12 vertebralfracture and spinal cord injury) requiring extensive rehabilita-tion.

During follow-up (mean: 28.9 months; median: 29months; range: 4 to 67 months) all patients were alivewithout endoleak, stent migration, stent collapse, falseaneurysm expansion, or rupture.

One patient underwent an additional surgical proce-dure 7 months following ABTAI due to the identificationof a coexisting traumatic aortic root-right ventricular fistula(Fig 2) identified during the endovascular procedure,which was confirmed on transoesophageal echocardiogra-phy. This finding has previously been reported10 followingblunt chest trauma.

DISCUSSION

Patients diagnosed with traumatic rupture of the tho-racic aorta typically present with multiple concomitantinjuries complicating traditional surgical management.Mortality rates for emergent open surgical repair forABTAI range from 15% to 30% in contemporary studies.11-13

Operative requirements including lateral positioning in pa-tients with vertebral or spinal cord injuries (with paraplegiarates ranging from 2.3% to 25.5%);14,15 thoracotomy andsingle lung ventilation in the presence of lung contusions,frequently (24% to 65%) resulting in prolonged respiratoryinsufficiency and infectious complications;2,16,17 and finallyhigh level systemic heparinization in those with multipleconcomitant injuries, all increase the risk of morbidity.

The development of endovascular techniques has led toa number of small studies examining the technical feasibility

and outcome of endovascular repair in ABTAI.18,19 The

JOURNAL OF VASCULAR SURGERYNovember 2007924 Bent et al

majority have demonstrated lower morbidity and mortalityrates in comparison to open surgical techniques advocatingits use.20 Despite this, endovascular treatment for trauma is

Fig 1. A 56-year-old male with an acute blunt trauangiography demonstrates a pigtail catheter traversing thduring stent-graft deployment. A guidewire with an undeb, Aortography demonstrates migration of the deployedc, Supra-aortic vessel perfusion is seen. A second undepready for deployment.

logistically as well as technically challenging requiring ex-

peditious imaging, multidisciplinary input, and an availablestock of equipment.

Standardized sizes of thoracic stent-grafts are currently

aortic injury following a fall from height. a, Aortict subclavian artery (LSA) allowing angiographic controld stent-graft is positioned at the level of the LSA origin.-graft to a position proximal to the supra-aortic vessels.stent-graft is inserted and positioned at the LSA origin

matice lefployestent

loyed

available for emergencies; however, these sizes remain lim-

JOURNAL OF VASCULAR SURGERYVolume 46, Number 5 Bent et al 925

ited. With a younger patient population presenting follow-ing trauma in comparison to those with degenerative aorticdisease, the tighter curvature of the aortic arch and smallercaliber of both the aorta and iliac arteries present significantchallenges to the design and engineering of current stent-graft devices.

The tighter curvature of the aortic arch affects endovascu-lar repair in two ways. The anatomical angulation may be tooacute to accommodate stiff wires such as the Lunderquist. Thelack of degenerative vascular disease elsewhere allows alterna-tives such as the Amplatz wire to be used without difficulty.Second, the tight curvature increases the risk of an inadequateproximal seal. Concern centers on the longitudinal flexibilityof the semi-rigid stent-graft design. As a result, many stent-grafts are not flexible enough to conform to an acutely angu-lated distal aortic arch and proximal descending aorta. Such aninability to gain optimal apposition with the vessel wall maylead to the development of an endoleak or “wind sock” effect

Fig 2. A 16-year-old female with an acute blunt traumatic aorticinjury. Early perfusion of the pulmonary arteries is seen duringaortography secondary to a traumatic aortic root-right ventricularfistula which was repaired electively 6 months following injury.

promoting distal migration.21

With the smallest aortic stent-graft in manufacturemeasuring 22 mm (Bolton Medical; Sunrise, Fla), signifi-cant over-sizing occurs in young patients with smalleraortic diameters. Case reports have highlighted that over-sizing can result in stent-graft collapse – a complicationwhich resulted in the voluntary withdrawal of the GoreTAG endoprosthesis in 2001.22-24 This alone may compli-cate endovascular repair as an acute treatment option.

The majority of patients sustain ABTAI located withinthe proximal descending segment, with 90% to 95% at theaortic isthmus.25 Cover of the LSA origin may therefore benecessary to lengthen the proximal neck sealing zone.Despite many studies reporting LSA origin coverage duringendovascular repair, it is not without complication.26 Con-sequently, if time permits, optimal preprocedural work-upshould include Duplex assessment of the extra-cranial cir-culation to ensure antegrade blood flow in the right verte-bral artery. In the presence of retrograde flow, it would beprudent to perform left carotid-subclavian bypass prior tostent-grafting to minimize the risk of an infarct involvingthe vertebrobasilar and posterior circulation.

In the absence of head injury or bleeding diathesis, allpatients at our institution undergo intra-arterial adminis-tration of 5000 international units of heparin prior tostent-graft insertion. During endovascular repair, the largecaliber stent-graft delivery sheath almost occludes the ar-teries leading to the aorta. In accordance with nontrau-matic endovascular aortic repair, the purpose of heparinadministration is therefore to prevent thrombus formationin these arteries, in addition to the lumen of the sheath.27

This in turn will minimize the risk of embolism. During thestudy, no thromboembolic complications were observed.

A major complication experienced in our cohort in-volved iliac artery avulsion and occurred in one patient(7.7%). This correlates with previously published litera-ture.20,28 The delivery devices used for introduction ofaortic stent-grafts are of large profile; therefore, preproce-dural assessment of the iliac arteries is mandatory to ensurethe vessel caliber is sufficient (minimum 7.5 mm) to accom-modate the delivery system. One possible explanation forsuch a complication may involve vascular spasm secondaryto irritation of the vessel wall during guidewire and cathetermanipulation. A hypothesis based on the initial successfuldelivery of the stent-graft system, followed by avulsionupon removal, when the external iliac artery was observedto be tightly contracted around the delivery device. Thiscomplication is frequently observed in young femaleswhere the vessel wall is soft and more compliant withlongitudinal pulling forces, resulting in an increased risk ofrupture. Accepting that the vessels may appear of smallcaliber in the presence of hypovolemia, when measured tobe less than 7.5 mm, use of an iliac or aortic conduit may beconsidered via a retroperitoneal incision.

In the event of iliac artery avulsion, insertion of aproximal occlusion balloon via either femoral artery is alife-saving maneuver, allowing control of hemorrhage untililio-femoral bypass grafting is completed. With this in

mind, it is vital that a range of stent-graft sizes is available,

JOURNAL OF VASCULAR SURGERYNovember 2007926 Bent et al

along with equipment to deal with procedural complica-tions within the emergency endovascular setting.

The second major type of complication involved proximalmigration of the stent-graft as a result of under-sizing andballooning of the proximal stent-graft seal. In the traumasetting, we oversize the stent-graft used by 10% to 15% of thediameter measured on CT imaging or aortography, as op-posed to a 15% to 20% oversize for aneurysmal disease. Thisdiscrepancy is due to a less aggressive approach recommendedin trauma patients, where the thoracic aorta is of normalcaliber and the intima friable following ABTAI. However intrauma, the potential for stent-graft under-sizing is increaseddue to hypovolemia causing a relatively smaller aortic caliberand sympathetic overdrive resulting in vasoconstriction – anobservation contrasting to published reports where extensiveover-sizing (�50%) has been noted as a potential complica-tion.29 These factors all need to be considered when selectingan appropriate stent-graft. At our institution, ballooning isreserved for malposition or type 1 endoleaks to minimize therisk of further aortic injury, which could result in catastrophicretrograde dissection.

Long-term follow-up remains necessary to assess bothstent-graft durability and progression of aortic disease. With ayounger population involved in trauma, follow-up protocolsmay need refinement to minimize risk of radiation exposure.Current recommendations involve annual plain chest radiog-raphy combined with interval CT scanning to monitor forstent fractures and endoleaks from the day the stent-graft isinserted. Consequently, cumulative radiation exposure is po-tentially hazardous. Although the usage of endovascular stent-grafts in ABTAI offers significant advantages, further develop-ment into smaller sizes, lower profiles and stent-grafts thatgenerate fewer artefacts during magnetic resonance imagingassessment would be beneficial.

CONCLUSION

Endovascular repair is rapidly becoming the procedureof choice for ABTAI. Even in the presence of extension ofinjury into the aortic arch, endovascular management re-mains a feasible option with extra-anatomical bypass. As aless invasive therapeutic option to standard surgical tech-niques, it is highly advantageous for polytrauma patients.Procedure-related morbidity remains low with no pub-lished case of paraplegia in trauma. In those patients withfew concomitant injuries, such a procedure could also allowearlier ambulation post trauma and a reduction in hospitalstay. In the absence of long-term follow-up regardingstent-graft durability and in those patients requiring defin-itive open thoracic surgery, endovascular repair could act asan intermediate measure until the patient is stable enoughto undergo such a procedure.

AUTHOR CONTRIBUTIONS

Conception and design: CB, MM, CKAnalysis and interpretation: CB, MM, MS, IR, RU, MW,

KB, CKData collection: CB

Writing the article: CB, KB, CK

Critical revision of the article: CB, MM, MS, IR, RU, MW,KB, CK

Final approval of the article: CKStatistical analysis: Not applicableObtained funding: Not applicableOverall responsibility: CK

REFERENCES

1. Pierangeli A, Turinetto B, Galli R, Calderara L, Fattori R, Gavelli G.Delayed treatment of isthmic aortic rupture. Cardiovasc Surg 2000;8:280-3.

2. Gammie JS, Shah AS, Hattler BG, Kormos RL, Peitzman AB, GriffithBP, et al. Traumatic aortic rupture: diagnosis and management. AnnThorac Surg 1998;66:1295-300.

3. Galli R, Pacini D, Di Bartolomeo R, Fattori R, Turinetto B, Grillone G,et al. Surgical indications and timing of repair of traumatic ruptures ofthe thoracic aorta. Ann Thorac Surg 1998;65:461-4.

4. Ferrari E, Tozzi P, von Segesser L. Thoracic aorta emergencies: is theendovascular treatment the new gold standard? Interact CardioVascThorac Surg 2006;5:730-4.

5. Creasy JD, Chiles C, Routh WD, Dyer RB. Overview of traumatic injuryof the thoracic aorta. Radiographics 1997;17:27-45.

6. Lebl DR, Dicker RA, Spain DA, Brundage SI. Dramatic shift in theprimary management of traumatic thoracic aortic rupture. Arch Surg2006;141:177-80.

7. Tai N, Renfrew I, Kyriakides C. Chronic pseudoaneurysm of the tho-racic aorta due to trauma: 30-year delay in presentation and treatment.Injury Extra 2005;36:475-8.

8. American College of Surgeons. Advanced Trauma and Life SupportCourse For Physicians. 7th ed. Chicago: Committee on Trauma, Amer-ican College of Surgeons, 2004.

9. Morasch MS, Kibbe MR, Evans ME, Meadows WS, Eskandari MK,Matsumura JS, et al. Percutaneous repair of abdominal aortic aneurysm.J Vasc Surg 2004;40:12-6.

10. Siavelis HA, Marsan R, Marshall WJ, Maull K. Aortoventricular fistulasecondary to blunt trauma: a case report and review of the literature.J Trauma 1997;43:713-5.

11. Fabian TC, Richardson JD, Croce MA, Smith JS, Rodman G, KearneyPA, et al. Prospective study of blunt aortic injury: multicenter trial of theAmerican Association for the Surgery of Trauma. J Trauma 1997;42:374-83.

12. Turney SZ, Attar S, Ayella R, Cowley RA, McLaughlin J. Traumaticrupture of the aorta: a five-year experience. J Thorac Cardiovasc Surg1976;72:727-32.

13. Crowley RA, Turney SZ, Hankins JR, Rodriguez A, Attar S, ShanskarBS. Rupture of the thoracic aorta caused by blunt trauma. J ThoracCardiovasc Surg 1990;100:652-61.

14. Attar S, Cardarelli MG, Downing SW, Rodriguez A, Wallace DC, WestRS, et al. Traumatic aortic rupture: recent outcome with regard toneurologic deficit. Ann Thorac Surg 1999;67:959-64.

15. von Oppell UO, Dunne TT, De Groot MK, Zilla P. Traumatic aorticrupture: 20-year meta-analysis of mortality and risk of paraplegia. AnnThorac Surg 1994;58:585-93.

16. Jahromi AS, Kazemi K, Safar HA, Doobay B, Cina CS. Traumaticrupture of the thoracic aorta: cohort study and systematic review. J VascSurg 2001;34:1029-34.

17. Razzouk AJ, Gundry SR, Wang N, del Rio MJ, Varnell D, Bailey LL.Repair of traumatic aortic rupture: a 25-year experience. Arch Surg2000;135:913-8.

18. Wellons ED, Milner R, Solis M, Levitt A, Rosenthal D. Stent-graftrepair of traumatic thoracic aortic disruptions. J Vasc Surg 2004;40:1095-100.

19. Orford VP, Atkinson NR, Thomson K, Milne PY, Campbell WA,Roberts A, et al. Blunt traumatic aortic transection: the endovascularexperience. Ann Thorac Surg 2003;75:106-12.

20. Stone DH, Brewster DC, Kwolek CJ, LaMuraglia GM, Conrad MF,Chung TK, et al. Stent-graft versus open surgical repair of the thoracic

aorta: mid-term results. J Vasc Surg 2006;44:1188-97.

JOURNAL OF VASCULAR SURGERYVolume 46, Number 5 Bent et al 927

21. Hoffer EK, Karmy-Jones R, Bloch RD, Meissner MH, Borsa JJ, Ni-cholls SC, et al. Treatment of acute thoracic aortic injury with commer-cially available abdominal aortic stent-grafts. J Vasc Interv Radiol 2002;13:1037-41.

22. Neschis DG, Moaine S, Gutta R, Charles K, Scalea TM, Flinn WR, et al.Twenty consecutive cases of endografts repair of traumatic aortic dis-ruption: lessons learned. J Vasc Surg 2007;45:487-92.

23. Idu MM, Reekers JA, Balm R, Ponsen KJ, de Mol BA, Legemate DA.Collapse of stent-graft following treatment of traumatic thoracic aorticrupture. J Endovasc Ther 2005;12:503-7.

24. Steinbauer MG, Stehr A, Pfister K, Herold T, Zorger N, Topel I, et al.Endovascular repair of proximal endograft collapse after treatment ofthoracic aortic disease. J Vasc Surg 2006;43:609-12.

25. Akins CW, Buckley MJ, Dagget W, McIlduff JB, Austen WG. Acutetraumatic aortic disruption of the thoracic aorta: a 10-year experience.

26. Riesenman PJ, Farber MA, Mendes RR, Marston WA, Fulton JJ, KeagyBA. Coverage of the left subclavian artery during thoracic endovascularaortic repair. J Vasc Surg 2007;45:90-4.

27. Noriyuki K, Dake M, Craig Miller D, Semba C, Scott Mitchell R, RazaviM, et al. Traumatic thoracic aortic aneurysm: treatment with endovas-cular stent-grafts. Radiology 1997;205:657-62.

28. White RA, Donayre CE, Walot I, Lippmann M, Woody J, Lee J.Endovascular exclusion of descending thoracic aortic aneurysms andchronic dissections: initial clinical results with the AneuRex device.J Vasc Surg 2001;33:927-34.

29. Hoornweg LL, Dinkelman MK, Goslings JC, Reekers JA, Verhagen H,Verhoeven Elm, et al. Endovascular management of traumatic rupturesof the thoracic aorta: a retrospective multicenter analysis of 28 cases inThe Netherlands. J Vasc Surg 2006;43:1096-102.

Ann Thorac Cardiovasc Surg 1981;31:305-9. Submitted May 9, 2007; accepted Jul 24, 2007.