Acute Traumatic Aortic Injury: ImagingEvaluationandManagementmsrads.web.unc.edu/files/2018/08/Traumatic-Aortic-Injury.pdf · prevalence of serious thoracic trauma and aortic injuries.

Acute Traumatic Aortic Injury:Imaging Evaluation and Management1

Scott D. Steenburg, MDJames G. Ravenel, MDJohn S. Ikonomidis, MD, PhDClaudio Schonholz, MDScott Reeves, MD

Despite recent advances in prehospital care, multidetectorcomputed tomographic (CT) technology, and rapid defini-tive therapy, trauma to the aorta continues to be a sub-stantial source of morbidity and mortality in patients withblunt trauma. The imaging evaluation of acute aortic inju-ries has undergone radical change over the past decade,mostly due to the advent of multidetector CT. Regardlessof recent technologic advances, imaging of the aorta in thetrauma setting remains a multimodality imaging practice,and thus broad knowledge by the radiologist is essential.Likewise, the therapy for acute aortic injuries has changedsubstantially. Though open surgical repair continues to bethe mainstay of therapy, percutaneous endovascular re-pair is becoming commonplace in many trauma centers.Here, the historical and current status of imaging andtherapy of acute traumatic aortic injuries will be reviewed.

� RSNA, 2008

1 From the Department of Radiology (S.D.S., J.G.R., C.S.),Department of Surgery, Division of Cardiothoracic Surgery(J.S.I.), and Department of Anesthesia and PerioperativeMedicine (S.R.), Medical University of South Carolina, POBox 250322, 169 Ashley Ave, Charleston, SC 29425. Re-ceived August 9, 2007; revision requested October 12;revision received January 16, 2008; accepted February27; final version accepted March 17; final review byJ.G.R. April 10. Address correspondence to J.G.R.(e-mail: [email protected] ).

� RSNA, 2008

REVI

EWS

AND

COM

MEN

TARY

�ST

ATE

OFTH

EAR

T

748 Radiology: Volume 248: Number 3—September 2008

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

Traumatic aortic rupture was firstdescribed in 1557 by Vesalius (1).However, acute traumatic aortic

injuries (ATAIs) remained rare until theadvent of high-speed motor vehicles inthe mid 1900s. Of all cases of blunttrauma resulting in substantial injury tothe thorax, motor vehicle collisions ac-count for a majority, followed by fallsfrom height, pedestrian-automobile col-lisions, and crush injuries. ATAI fromblunt trauma is a substantial cause ofmorbidity and mortality, occurring inapproximately 0.5%–2% of all nonlethalmotor vehicle collisions (2,3) and 10%–20% of all high-speed deceleration fatal-ities (3–10). An overwhelming majorityof major trauma patients are young, andif they survive, face a life-time of mor-bidity (11).

The morbidity and mortality of thisinjury are high, and injury is immedi-ately lethal in 80%–90% of cases (3,4).With improved in-field care and rapiddetection and treatment of ATAI, themorbidity and mortality have improved,and patients who initially survive aremore likely than ever to undergo suc-cessful repair (2,5,12,13). It is thus par-amount that the radiologist be aware ofthe wide range of presentations and thevarious imaging findings of ATAI. This

review will be broad in scope, focusingon the pathophysiologic, clinical, radio-logic, and surgical perspectives. The im-portance of recent diagnostic and thera-peutic advances will be emphasized.

Motor Vehicle Collision Profile

It is estimated that 75%–80% of tho-racic aortic injuries are a result of high-speed motor vehicle collisions, withmost ATAIs occurring after rapid decel-eration as a result of head-on or side-impact collisions above 50 km/h (14–16). Head-on collisions have historicallybeen thought to play a predominant rolein the mechanism of ATAI (4,6). Suchstudies have led to vehicle modificationsand improved traffic laws, such as seatbelt requirements, airbags, collapsiblesteering columns, and energy-absorbingbumpers. There has only been recentattention to vehicle safety modificationsto protect occupants from lateral impactcollisions.

Interestingly, case studies detailingthe motor vehicle collision profile thatresults in thoracic trauma and ATAIsuggest that only about 50% are due tohead-on collisions (16–18). In reality,side-impact collisions are associatedwith a substantially higher mortalityrate than head-on collisions, regardlessof the injury severity score (17). In fact,mounting evidence suggests that victimsof serious lateral impact crashes may beat greater risk of ATAI than those ofnonlateral impact collisions, with onestudy (7) showing that 73% of all aorticisthmus injuries resulted from side-im-pact collisions. Evidence also suggeststhat occupant restraint mechanisms(excluding side airbags) are largely inef-fective in curtailing ATAI in side-impactcollisions (19). It remains to be seen ifside curtain airbags will decrease theprevalence of serious thoracic traumaand aortic injuries.

Pathophysiology and Mechanism ofInjury

The extent and morphology of aorticinjuries vary widely, ranging from in-timal hemorrhage to complete tran-section. Aortic injury most commonly

results from transverse tears and canbe segmental (55%) or circumferen-tial (45%) and may be partial (65%)or transmural (35%). Spiral and irreg-ular tears are very rare (4,7). Partiallacerations usually involve only the in-ner two vessel wall layers, resulting ina contained rupture. The adventitiamay be injured in up to 40% of cases,and adventitial injury is almost univer-sally fatal because of rapid exsangui-nation. Temporary tamponade may beachieved by surrounding mediastinalsoft tissues (4).

Despite extensive research, the ex-act mechanism of ATAI has not beenprecisely determined. A majority ofATAIs result from violent deceleration,most commonly as a result of a motorvehicle collision, especially head-on andside-impact collisions. Proposed mecha-nisms contributing to ATAI includeshearing forces, rapid deceleration, hy-drostatic forces, and the osseous pinch(Fig 1) (7820–23).

Rapid deceleration in the antero-posterior and lateral directions hasbeen shown to be sufficient to result incardiac displacement, resulting in tor-sion and shearing forces against theaorta at levels of relative immobility,mainly the ligamentum arteriosum, aor-tic root, and diaphragm (7,8). A combi-nation of compression and upwardthrust of the heart, which also involveshear and torsion, has been suggested(14). Lateral compression can result insevere internal chest deformation, re-sulting in anterior displacement of theheart and thus shearing forces at theaortic isthmus (16,17).

Increased intravascular pressurecan exceed 2000 mm Hg following di-rect compression of the aorta and hasbeen termed the water-hammer effect.The pressures that can be created by

Published online10.1148/radiol.2483071416

Radiology 2008; 248:748–762

Abbreviations:ATAI � acute traumatic aortic injuryTEA � transesophageal echoaortography

Authors stated no financial relationship to disclose.

Essentials

� With improvements in prehospi-tal care, more survivors withacute traumatic aortic injury(ATAI) will reach trauma centers.

� Rapid diagnosis and treatment ofATAI is the standard of care.

� Multidetector CT is the diagnostictest of choice for detection ofATAI in the stable patient; whenunequivocal findings are present,no further imaging is necessary.

� Transesophageal echoaortographyis well suited for evaluation of theunstable patient who must pro-ceed to the operating room priorto multidetector CT imaging.

� Conventional angiography, intra-vascular US, and transesophagealechoaortography are valuable ad-juncts in the evaluation of equivo-cal multidetector CT cases.

STATE OF THE ART: Acute Traumatic Aortic Injury Steenburg et al

Radiology: Volume 248: Number 3—September 2008 749

this mechanism have been shown to re-sult in mainly transverse tears at thelevel of the isthmus (23), but can alsotravel retrograde resulting in injury atthe aortic root (22).

The osseous pinch results from di-rect compression of the aorta betweenthe anterior chest wall and the thoracicspine (20,21). An animal model studyby Crass et al (21) showed that antero-posterior compression of the chest con-sistently results in transverse lacera-tions to the aortic isthmus when com-pressed between the anterior chest(particularly the manubrium and medialclavicles) posteriorly against the tho-racic spine. This is supported with in

vivo data (20). These studies also ex-plain the predilection for concomitantinjuries to the branch vessels (20,21).Direct injury of the thoracic aorta mayalso occur due to penetrating injuryfrom rib and thoracic vertebral bodyfractures (24,25).

Whatever the dominant mecha-nism, it is safe to say that the patho-physiology of aortic injury appears com-plex and is likely due to interplay of acombination of the above mechanisms.Computer models and cadaveric simu-lations have been proposed and willlikely be helpful in the future (26,27).

Signs, Symptoms, and ClinicalPresentation

Clinical signs and symptoms are non-specific and insensitive for the diagno-sis and exclusion of ATAI. A majorityof patients with ATAI have no clinicalsigns of aortic injury until the suddenonset of hemodynamic instability(10,28,29). Symptoms of ATAI arethought to be due to stretching of themediastinal connective tissues by me-diastinal blood and include retroster-

nal pain, referred interscapular pain,dyspnea, hoarseness, and cough(10,30). Clinical signs of ATAI are ab-sent in up to one-third of patients butwhen present include upper limb hyper-tension and lower limb hypotensionwith diminished femoral pulses due to“pseudocoarctation syndrome,” systolicmurmur, external chest wall injuries,paraplegia, and initial chest tube outputgreater than 750 mL (31).

On the basis of the presence of cer-tain clinical variables, a risk assessmentfor aortic injury can also be made.Blackmore et al (32) found that sevencriteria were predictors of ATAI. Theseinclude age older than 50 years, beingunrestrained, hypotension with systolicblood pressure of less than 90 mm Hg,thoracic injury, abdominopelvic injuryrequiring emergent laparotomy or withfractures of the lumbar spine and pelvis,long bone fractures, and major head in-jury. Not surprisingly, patients meetingmore criteria had a greater chance ofhaving ATAI and those meeting four cri-teria or more had a 30% chance of sus-taining ATAI in that study (32). In amore recent re-evaluation (33) of these

Figure 1

Figure 1: Schematic depicts various mecha-nisms of injury resulting in ATAI. The osseouspinch is shown by posterior translation of the ster-num (large arrows), which traps the aorta at thenarrowest point of the thorax. Torsion is depictedin the ascending aorta (curved arrows) resulting intwisting above the fixed aortic valve. Compressionof lumen forces blood inferiorly toward the aorticvalve resulting in the water hammer effect (arrowsin aortic root). At ligamentum arteriosum, the aortais affected by both bending and shearing stresses(small arrows) augmented by the anatomicchanges from the osseous pinch against the tho-racic spine and pressure changes from the waterhammer effect. The combination of forces is con-centrated at the fixed point of the descending tho-racic aorta, the ligamentum arteriosum (parallellines inferior to transverse aortic arch).

Figure 2

Figure 2: Findings of ATAI. Supine anteroposterior chest radiograph following motor vehicle collisionshows widening of the mediastinum with deviation of the trachea (T) to the right, depression of the left main-stem bronchus (LM), convexity in the aortopulmonary window (arrow), and left apical cap (�) due to mediasti-nal hematoma from ATAI.



clinical predictors, only four factors (ab-dominopelvic injury, thoracic injury, hy-potension, and being unrestrained)were found to be predictive. Moreover,

the combination of three or four factorsresulted in a 2% chance of ATAI (33).The differences presumably reflectchanges in patient populations and in-

jury detection over the years. In thatstudy, chest radiographs were onlymarginally efficient in increasing diag-nostic yield.

Morbidity, Mortality, and Outcome

Aortic injuries carry a high mortalityrate and are immediately fatal in an es-timated 80%–90% of all cases(4,7,8,15). In Parmley et al classic se-ries (4), only 20% of patients with ATAIinitially survived more than 1 hour. Ofthose, there is an estimated mortality of30% within the first 6 hours, 49%within the first 24 hours, 72% by 8 days,and 90% at 4 months if undetected anduntreated. Williams et al (14) suggestedan even more dismal outcome, with94% mortality within 1 hour and up to99% mortality at 24 hours if untreated.Blunt aortic injury rarely occurs in iso-lation (only 3.2% of autopsy cases)(15), and there is a substantial increasein mortality following ATAI in patientswith co-existing head, thoracic, and ab-dominal injuries (34–36). Overall mor-tality following ATAI therefore dependson several factors, including high injuryseverity score, prolonged transport totrauma center, and hemodynamic insta-bility at presentation (2,13,35,36).Overall survival is worse with increasingage, with mortality in excess of 80% in

Figure 3

Figure 3: (a) Upright posteroanterior chest radiograph in young male patient following high-velocity mo-tor vehicle collision initially interpreted as normal. While the mediastinum is not wide, the usual concavity isabsent in the aortopulmonary window (arrow). (b) Transverse contrast-enhanced multidetector CT scanshows traumatic disruption of the aorta (arrow).

Figure 4

Figure 4: Mediastinal hematoma with preserved fat plane around aorta at transverse contrast-enhancedmultidetector CT. (a) Image at the level of aortic arch shows hemorrhage in the anterior mediastinum (H) withintervening fat plane between hemorrhage and aortic wall (arrows). (b) Image at the level of thoracic inlet re-veals intimal tear in left subclavian artery (arrow) accounting for source of hemorrhage.

Figure 5

Figure 5: Periaortic hematoma without directinjury. Transverse contrast-enhanced multidetec-tor CT scan reveals hemorrhage in the mediasti-num adjacent to the medial wall of descendingthoracic aorta (�) due to traumatic tear of azygousvein (AZ, arrow).



patients older than 55 years (36). If de-tected in a timely manner, it is esti-mated that 60%–80% of patients withATAI who reach the hospital alive willsurvive following definitive therapy(13,37,38). Therefore, prompt recogni-tion and treatment of these injuries arecritical for long-term survival.

Radiographic Findings in ATAI

Chest RadiographyIn the acute trauma setting, the supinechest radiograph is still frequently ob-tained. While the predominant ratio-nale is the inclusion or exclusion of im-mediate life-threatening lesions requir-ing immediate treatment (massivehemithorax or tension pneumothorax),it also provides some data to guide sus-picion of ATAI. Evaluation for mediasti-nal hematoma, and by inference a majorvascular injury, is the main goal of theinitial chest radiograph (Fig 2). Whilemediastinal widening greater than 8 cmand/or 25% of the width of the thorax isthe most frequent observation, it is notnecessarily the most sensitive finding(39–41). More discriminating findingsinclude any abnormality of the trans-verse aortic arch or loss of the aortopul-monary window (42,43). Therefore,even in those cases where the medias-tinum is not widened, obscuration of thelung interface with the transverse or de-scending thoracic aorta should still beviewed with suspicion. It should also benoted that in up to 7% of ATAI cases,the chest radiograph may be normal ordeceptively underwhelming (Fig 3).Thus, in the setting of a rapid decelera-tion force or high clinical suspicion, fur-ther evaluation is warranted regardlessof chest radiographic findings (39).Other signs of ATAI on chest radio-graphs include rightward tracheal,esophageal, and/or nasogastric tube de-viation; left mainstem bronchus depres-sion; and a left apical cap (41–50).

Given that disruption of the aorta re-quires high-force trauma, other injuriesare often present in the lung (pulmonarycontusion), pleura (hemithorax and pneu-mothorax), diaphragm, and bony thorax.Although fractures of the first and second

ribs are clearly markers of severe bluntforce trauma, they are not by themselvesindependent predictors of ATAI, norshould the absence of skeletal abnormal-ities be taken as a reassuring sign (44).Regardless, the greater the degree ofmanifestation of blunt force trauma, thehigher the suspicion should be for ATAI.

The abnormal (even minimally so)supine anteroposterior radiograph intrauma should always be evaluated with

further imaging. In cases where the riskbased on the mechanism of injury is low,this can be done with an upright radio-graph (preferably a posteroanteriorview), which is best suited for level 2 or 3trauma patients. It should be emphasizedthat the most important observation is aclear visualization of the aortic arch, notabsence of mediastinal widening, as a dis-criminating feature. However, if the forcevector or mechanism of injury is suffi-

Figure 6

Figure 6: Retrocrural hematoma secondary to ATAI at transverse contrast-enhanced multidetector CT.(a) Image at aortic hiatus shows periaortic hematoma (arrow). (b) Image at the level of aortic isthmus showsATAI (arrow) with extensive mediastinal hematoma.

Figure 7

Figure 7: Small tear as manifestation of ATAI. (a) Transverse contrast-enhanced multidetector CT scanshows a small intimal tear along lateral border of proximal descending aorta (arrow). (b) Sagittal oblique con-trast-enhanced multidetector CT reformation shows craniocaudal extent of the tear (arrow).



cient, if the patient is unconscious orintoxicated, or if the patient is alreadyundergoing a contrast material– en-hanced multidetector CT of the abdomenand pelvis then multidetector CT withcontrast material is the most appropriateexamination. If the patient must go to theoperating room immediately because oflife-threatening abdominal or neurologic

injuries, intraoperative transesophagealechocardiography may be used in someinstitutions to evaluate the aorta.

CT ScanningBeginning in the early to mid 1990s, CThas gained an increasingly importantrole, first in screening and now in diag-nosis of ATAI, such that multidetector

CT is the diagnostic test of choice forATAI (45–48). The diagnostic sensitiv-ity routinely exceeds 98% for ATAI,while specificity depends on the defini-tion of a positive result (46–49). Whenthe presence of mediastinal hematomaalone is included as a diagnostic crite-rion, false-positive findings are high andtherefore should not be used as a crite-rion for definitive injury, while if onlydirect signs are used specificity ap-proaches 100% (46,49,50). In many in-stitutions, multidetector CT has re-placed the need for conventional an-giography prior to surgery in themajority of cases. It should be notedthat this practice, even today, is notuniversally held (51).

Indirect findings of ATAI generallyconsist of the presence of mediastinalhematoma. The importance of hema-toma depends on the location. Bloodwithin the mediastinum with a pre-served fat plane around the thoracicaorta is not from ATAI (Fig 4). Thesource is usually small veins within themediastinum and generally does not re-quire further evaluation for ATAI (52–54). Depending on location, careful at-tention should be paid to intercostal ar-teries, internal mammary arteries, andarch branch vessels as they may be thesource of bleeding. On the other hand,periaortic hematoma is in direct conti-nuity with the aortic wall (Fig 5). Thebleeding presumably originates from ei-ther small veins immediately adjacent tothe aorta or from the vasa vasorum(41). From this standpoint, periaortichematoma may represent an occult inti-mal injury requiring further evaluationwith intravascular ultrasonography (US)or transesophageal echoaortography(TEA). In these cases, the role of con-ventional angiography alone is limited.Authors of a recent study (55) foundthat none of 24 cases (albeit with a rela-tively wide confidence interval of 0% to10.8%) with periaortic hematoma with-out direct signs were found to haveATAI at conventional angiography. Instable patients with low clinical suspi-cion, it may be appropriate to follow upwith multidetector CT in 48–72 hours.In patients who have undergone abdom-inal multidetector CT alone, the pres-

Figure 8

Figure 8: Aortic transection with contained pseudoaneurysm. (a) Transverse contrast-enhanced multide-tector CT scan shows focal outpouching at the level of ligamentum arteriosum (arrow). (b) Left anterioroblique reformation reveals circumferential nature of the tear (arrows). � � left subclavian artery.

Figure 9

Figure 9: Aortic transection with associated injuries. (a) Transverse contrast-enhanced multidetector CTscan reveals transection at the level of ligamentum arteriosum (arrow). E � pleural effusion. (b) Coronal refor-mation shows aortic injury (arrow) in association with left diaphragmatic rupture (arrowhead) containing her-niated stomach (St) and splenic laceration (Sp) with perisplenic hematoma (�).



ence of a retrocrural hematoma (56)(Fig 6) or a small caliber aorta may bean indication of thoracic aortic injury.These findings should be followed witheither multidetector CT of the chest orTEA, depending on the patient’s clinicalstatus and need for surgical interven-tion.

Direct signs of ATAI include pres-ence of an intimal flap, traumatic pseu-doaneurysm, contained rupture, in-traluminal mural thrombus, abnormalaortic contour, and sudden change inaortic caliber (aortic “pseudocoarcta-tion”) (Figs 7–9). These should be inter-preted as definitive positive findings,and unless there are extenuating cir-cumstances, no additional imaging isnecessary. Active extravasation of con-trast material in practice is exceedinglyrare as it often portends impending ex-sanguination (Fig 10). Rarely, a true dis-section may occur. In the setting of di-rect signs and an unstable patient, sur-gery should be performed expeditiouslyand confirmation with conventional an-giography can be a dangerous, if notfatal, waste of time.

Multidetector CTIt is clear that multidetector CT is thediagnostic modality for the initial evalu-ation and accurate diagnosis of thoracictrauma. However, protocols need to beoptimized to generate the greatestamount of data in the least amount oftime. We use a standardized traumaprotocol for contrast-enhanced multide-tector CT of the chest, abdomen, andpelvis in a single scan with injection ofapproximately 140 mL of contrast me-dia at 3–4 mL/sec with a 75-secondprescan delay. Images are acquired atthinnest possible multidetector arrayand are generally reconstructed at atransverse section thickness of 2–2.5mm, with sagittal and coronal reforma-tions for review at a picture archivingand communication system station. Wehave found this to be a good compro-mise between speed, data, and the needfor imaging various organs. Despite notusing a classic CT angiographic tech-nique, we have found that our techniqueproduces reasonable three-dimensionalimages when needed. In difficult cases,

we will also use interactive manipula-tion and interpretation of the data set togenerate multiplanar reformations invarious obliquities of the coronal andsagittal plane for optimal display of vas-cular injuries and morphology and loca-tion with respect to adjacent vascularstructures, such as the left subclavianartery.

While it is true that direct and indi-rect findings of ATAI are usually imme-diately apparent on conventional trans-verse images, the exact morphology andextent of vascular injury, particularly atthe aortic isthmus, may not be fully ap-preciated until viewed in another plane(Fig 11). The isotropic data sets ac-quired with multidetector CT can beused to generate multiplanar reforma-tions, which can be used to aid in surgi-cal planning. It is our experience thatthe sagittal oblique plane (which simu-lates the standard projection obtainedduring angiography of the aorta) best

displays the thoracic aorta along its longaxis. Three-dimensional volume-ren-dered images are also useful to the sur-geon, as these images display the in vivoanatomy with its relationship to the ad-jacent structures, show the distance ofthe isthmus injury from the subclavianartery, and display the exact morphol-ogy of the injury (Fig 12). Some authorsadvocate the use of endovascular viewsfor the evaluation of intimal morphologyin difficult cases (57). In equivocalcases, it is often on these postprocessedimages that the diagnosis is established.

Isthmus injuries.—The aortic isth-mus, within 2 cm of the origin of the leftsubclavian artery, is the most commonlocation for ATAI as established bymany large autopsy series (4,7,8,15).The predilection of this injury for thislocation is thought to me to be due to itsrelatively immobile position within thethorax, being tethered by the ligamen-tum arteriosum (4,7,8,15). The severity

Figure 10

Figure 10: Aortic transection with active extrava-sation. (a) Transverse contrast-enhanced multidetec-tor CT scan shows active hemorrhage (�) extendinganteriorly from ATAI (arrow). Repair was successfuland patient survived injury. (b) Sagittal oblique viewreveals complex nature of injury (arrow) and allowsaccurate assessment of the distance from left subcla-vian artery to the proximal extent of the tear. (c) Vol-ume rendered image from superior perspective givesadditional information on the relationship of superioraspect of the tear (arrow) to left subclavian artery(LSC).



of injury at this location may range fromminimal intimal injury (see below) tofrank rupture with active extravasation.On axial images, these isthmus injuriesare seen most commonly along the me-dial curvature of the arch at the level ofthe left pulmonary artery and left main-stem bronchus and should be the firstlocation the radiologist looks at whenencountering a large mediastinal hema-toma. However, it must be noted thataortic injuries may occur at any locationalong the aorta.

Minimal aortic injuries.—Minimalaortic injuries, only affecting the intimaof the vessel wall (Fig 13), are estimatedto occur in 10% of patients with ATAI(58). These lesions pose a substantialproblem because (a) they are being en-countered at an increased frequencyand (b) very little data exist on the opti-mal management. The apparent in-creased incidence is likely secondary tothe improved spatial resolution of multi-detector CT (58). Confirmation with di-rect conventional angiography may notbe possible, as nearly half of the casesshow no demonstrable abnormality atangiography, and intravascular US maybe required to confirm the diagnosis(58). Some suggest that these minimalaortic injuries may not need any inter-vention, as with imaging follow-up themajority remain stable or resolve (58).

Aortic root and ascending aorta.—Traumatic injuries to the ascendingaorta were seen in approximately 5%–14% of ATAIs in several large autopsyseries (7,8); however, at multidetectorCT, these lesions are extremely rarepresumably because of their intrinsi-cally lethal nature (59–61). There arethree main patterns of injuries to theascending aorta: (a) a laceration of theascending aorta between the aortic rootand the brachiocephalic artery, (b) inju-ries to the aortic root that involves theaortic valve, and (c) injuries to the aor-tic root that do not involve the aorticvalve (Fig 14). It is intuitive that injuriesto the aortic root should be associatedwith hemopericardium, but it has beenshown to be an inconsistent finding inascending aortic injuries (60,61).Therefore, the absence of hemopericar-dium should not be used to exclude as-cending aortic injury. Injuries of theaortic root may also co-exist with inju-ries at the isthmus. Problem solvingwith cardiac-gated acquisition can beperformed if there are equivocal abnor-malities of the ascending aorta at rou-tine contrast-enhanced chest multide-tector CT, but the efficacy of this hasnot be objectively evaluated.

Aortic arch and branch vessel inju-ries.—Injuries to the aortic arch andbranch vessels are less common but po-

tentially fatal. The incidence of aorticarch injuries is rare, estimated to occurin less than 4% of blunt chest traumapatients, as seen in a large, multi-centerprospective evaluation of 274 patientswith aortic injuries (39). The incidenceof aortic branch vessel injuries varieswidely in the radiologic and surgical lit-erature. Isolated aortic branch injuriesare most common, but may be seencombined with blunt ATAI in 0%–45%of patients (62–66). The brachioce-phalic and common carotid arteries aremost commonly injured, seen in 66%–90% of patients with branch vessel inju-ries (63,64). The extent of branch ves-sel injuries, like aortic injuries, rangesfrom subtle intimal injuries to completetransection and contained rupture(63,64).

At multidetector CT, a carefulsearch for a branch vessel injury shouldbe performed in the presence of medi-astinal hematoma when there is no di-rect sign of aortic injury as well as incases of ATAI. This has important impli-cations for surgical approach, as bra-chiocephalic artery and proximal com-mon carotid artery injuries require me-dian sternotomy for proper exposure. Itis intuitive that the volumetric data ac-quired at multidetector CT should makethe diagnosis less problematic.

Mid and distal descending thoracic

Figure 11

Figure 11: Segmental ATAI. (a) Transverse contrast-enhanced CT image reveals periaortic hematoma and contour abnormality of the proximal descending thoracicaorta (arrows). (b) Transverse CT image inferior to a reveals continued periaortic hematoma and intimal flap projecting into the lumen (arrow). (c) Sagittal oblique viewshows segmental nature of injury (arrows) and relationship to left subclavian artery.



aorta.—Trauma to the mid and distaldescending aorta (Fig 15) is discoveredin approximately 1%–12% of autopsieswith aortic injuries (7,8). The distal de-scending aorta is tethered to the adja-cent spine by the crux of the diaphragmand therefore injuries here are thoughtto occur as a result of shear forces ap-plied at this location. Injuries to thissegment of the thoracic aorta can beassociated with diaphragm injury in10% of cases and with adjacent com-pression fractures of the thoracic spine(25,67).

Imaging pitfalls.—Diagnostic pitfallscan be divided into two categories, ana-tomic and technical. Anatomic pitfallsinclude venous, arterial, and pulmonary(68). The left superior intercostal veinruns adjacent to the transverse aorticarch, and the hemiazygous vein may beseen posterolateral to the descendingaorta (Fig 16). With both the aorta andthe vein opacified with contrast mate-rial, the adjacent vessel walls can ap-proximate the appearance of an intimalflap. This is usually not a diagnostic di-lemma by scrolling images up and downand tracing out the vein back to its in-sertion into the left subclavian vein. Oc-casionally, the take-off of bronchial andintercostal arteries can have small in-fundibula that may give the impressionof a small pseudoaneurysm. Infundibulaare typically conical in shape and theartery can be seen at the apex of theoutpouching. Ductus remnants maypresent as either bumps or diverticula(Figs 17, 18). Similar to infundibula,these may simulate injury and can bevexing due to their location, particularlywhen mediastinal hematoma is present.Clues to the diagnosis include continuitywith the aortic wall, with obtuse marginsand occasionally calcification. When amore acute angle is present, it canpresent a substantial diagnostic di-lemma. It has been suggested that en-doluminal views may be helpful byshowing the absence of an intimal flap(57). Finally, enhancement of the col-lapsed lung adjacent to the aorta mayalso simulate an intimal flap. Carefulanalysis often reveals pulmonary vesselsand bronchi entering the collapsed lungand confirms the diagnosis. Occasion-

ally, normal postisthmic aortic dilata-tion (or aortic spindle) may be seenwhen viewed in the sagittal obliqueplane. This contour change is a normalfinding and should not be misinter-preted as an aortic injury.

Technical issues may arise from pa-tient motion, breathing, and cardiac pul-sation. In the absence of mediastinal hem-orrhage, these can often be reliably dis-missed. In the presence of hemorrhage,evaluation of lung windows can often de-lineate the presence of breathing artifactas motion within the lung parenchyma.Depending on the location, the artifactcan either be dismissed or reimaged witheither multidetector CT or conventionalangiography. Cardiac pulsation is mostprevalent at the aortic root and ascendingaorta. The best clue to cardiac pulsation is

that the apparent intimal flap projectsacross the lumen and into the mediastinalfat (Fig 16). Cardiac pulsation artifactscan sometimes be distinguished from trueintimal flaps by evaluating the sharpnessof their interface with the high-densitycontrast material in the aortic lumen.Typically vascular flaps have a distinct in-terface, whereas a pulsation artifact isusually fuzzy. Care should be used withthis rule when other artifacts, such as mo-tion artifact, are present as well. If thearea remains questionable, alternativestrategies include performing a 180° re-construction of data that may reveal achange in orientation of the artifact orrescanning, preferably using cardiac gat-ing to suppress motion. For stable pa-tients, it is reasonable to rescan in 24–48hours for reassessment.

Figure 12

Figure 12: Complex ATAI. (a, b) Transverse contrast-enhanced images reveal ATAI with saccular aneu-rysm medial (a) and lateral (b) to aorta (arrow). (c, d) Volume rendered images from medial (c) and lateral (d)perspective show relationship of each component of the injury (arrow) to the left subclavian artery.



MR ImagingWhile magnetic resonance (MR) an-giography has excellent test characteris-tics for the detection of ATAI (69), itsutility in the trauma patient is limiteddue to location remote from the traumabay, scan time, room in the bore of themagnet for support devices, and a myr-iad of other logistical issues. MR imag-ing, however, may have a role in thefollow-up when delayed surgery is con-

templated and minimal or equivocal in-timal injuries are present, particularlyas a strategy for radiation dose reduc-tion in young trauma victims (69,70).

Conventional AngiographyThe value of conventional aortographyfor the evaluation of ATAI is well estab-lished, with sensitivity of nearly 100%,specificity of more than 98%, and accu-racy of more than 99%, and thus for

decades it has been considered the“gold standard” examination (28). Theimaging findings of ATAI may rangefrom subtle contour irregularity to frankcontrast material extravasation (10,29,

Figure 13

Figure 13: “Minimal” aortic injury at transverse contrast-enhanced multidetector CT. (a) Image showsperiaortic hematoma surrounding mid descending thoracic aorta (arrows). (b) Image obtained 72 hours laterreveals small thrombus at site of previously occult intimal injury (arrow).

Figure 14

Figure 14: Ascending aorta ATAI. Transversecontrast-enhanced CT scan at the level of left maincoronary artery shows aortic injury just above theright coronary cusp (arrow) with associated hemo-pericardium (H).

Figure 15

Figure 15: Complex traumatic aortic injury with tears in proximal and mid descending aorta at transversecontrast-enhanced multidetector CT. (a) Image at the level of left pulmonary artery shows contour abnormality(arrow) and periaortic hematoma (H). (b) Image at the level of left atrium shows second tear (arrow) and moreextensive periaortic and mediastinal hematoma.

Figure 16

Figure 16: Artifacts that may mimic ATAI.Transverse contrast-enhanced multidetector CTscan shows pulsation artifact at the ascendingaorta (black arrows). The linear hypodensity con-tinues into mediastinal fat. The hemiazygous veinmimics a contour abnormality along the postero-lateral surface of descending thoracic aorta (whitearrow).



71). Pseudoaneurysm formation, in-traluminal filling defect, and intimal ir-regularity are common findings in vesselwall injury.

Historically, the role of aortographywas to (a) identify or exclude ATAI and, ifpresent, (b) determine the exact locationof the injury with respect to the branchvessels and (c) evaluate for co-existingbranch vessel injuries (72–74). In thepre-CT era, approximately 80%–90% ofaortograms obtained for suspected ATAIbased on the mechanism and initial chestradiograph were normal (30,74). Con-ventional angiography for the diagnosis ofATAI has taken on a substantially limitedrole in the multidetector CT era. Addi-tionally, disruption or delay in definitivecare to perform angiography can be acritical indirect cause of patient morbidityand mortality. Despite this, the value ofconventional aortography remains an im-portant problem-solving tool in the stablepatient, for planning prior to endovascu-lar stent graft therapy, and in some casesfor detection of branch vessel injury.

Intravascular USEndoluminal or intravascular US is an-other useful adjunctive imaging modal-ity that can be used to provide high-resolution cross-sectional images of

the vessel wall and the surroundingtissues. The findings of aortic injury atintravascular US include vessel walldisruption, intimal flap, focal pseudo-aneurysm, intramural and periaortic

hematoma, and complete transection(Fig 19) (75). Although these findingsare considered to be specific for ATAI,false-positive results have been de-scribed (10,75).

Figure 17

Figure 17: Atypical ductus bump. (a) Transverse contrast-enhanced multidetector CT scan shows apparent intimal flap at the level of ligamentum arteriosum (ar-rows). There is no periaortic hematoma. Note also anterior mediastinal hematoma (H) due to rib fractures. (b) Left anterior oblique maximum intensity projection showssmall outpouching at the level of ligamentum arteriosum with obtuse angle superiorly and acute angle inferiorly (arrow). (c) Conventional angiogram in left anterioroblique projection shows similar finding (arrow) to b.

Figure 18

Figure 18: Ductus diverticulum. (a) Transverse contrast-enhanced multidetector CT scan at the level ofligamentum arteriosum shows well-defined contrast material– containing structure extending toward leftpulmonary artery (arrow). Note anterior mediastinal hematoma (arrowhead) due to sternal fracture. E � effu-sion, � � mediastinal hematoma. (b) Left anterior oblique reformation shows extension from aorta (Ao) to-ward left pulmonary artery (LPA) along typical course of ductus arteriosus (arrow). Conventional angiography(not shown) performed because of mediastinal hematoma confirmed finding of ductus diverticulum.



The advantages of intravascularUS mainly involve its problem-solvingcapabilities. It can be performed con-currently with conventional aortogra-phy and has been shown to be a usefulcomplementary modality (10). Intra-vascular US is an operator- and expe-rience-dependent invasive procedure,requiring arterial puncture, and com-plete evaluation of the aorta can betime-consuming and may not allowcomplete visualization of the brachio-cephalic artery (10). Technical limita-tions of intravascular US, such asdepth of penetration, also exist andare inherent in the size and frequencyof the ultrasound transducer (75).

TEA ExaminationThough TEA is usually performed by ananesthesiologist, a cardiologist, or athoracic surgeon, radiologists may beasked to offer alternative examinationsfor the evaluation of the aorta in equiv-ocal multidetector CT cases or in criti-cally ill and unstable patients. TEA has alimited role in routine screening forATAI and should not be performed atthe expense of evaluation for other co-morbid injuries, which is best accom-plished with multidetector CT. In a re-view of the literature (76), the sensitiv-

ity of TEA for evaluation of ATAI ranged56%–99% and specificity ranged 89%–99%. In a single study (77) in whichTEA was compared with helical CT in95 patients, all surgical ATAI caseswere detected with both techniques. Inaddition, TEA can be helpful in evaluat-ing equivocal CT or conventional angio-graphic findings for both the inclusionand exclusion of ATAI (77).

TEA is widely available, relativelynoninvasive, and can be performedquickly at the bedside or in the operat-ing room, which are seen as its mainadvantages. Moreover, since imaging isperformed in real-time, the aortic valve,sinotubular junction, and ascendingaorta can be evaluated to a much betterextent with TEA than with multidetec-tor CT. Thus, unlike the other imagingtechniques, TEA can be used intraoper-atively to immediately affect surgicaland anesthesia decisions. Finally, TEAcan help evaluate the myocardium forwall motion abnormalities, as well asassess the physiologic consequence ofany pericardial fluid. Disadvantages in-clude variable skill and experience ofthe operator, lack of 24-hour availabil-ity, and relatively poor visualization ofthe distal ascending aorta and proximalarch (78).

Treatment of ATAI: The Surgeon’sPerspective

Open Thoracotomy

Acute traumatic aortic injuries are con-sidered to be surgical emergencies, andmost patients should undergo repair im-mediately. In the current era of percuta-neous intervention, however, these“higher risk” patients may perhaps bebetter served with endovascular stent-grafting than with open surgery.

For open repair of descending tho-racic aortic tears, a generous left pos-terolateral thoracotomy is made at thefourth or fifth intercostal space. It isabsolutely essential that complete prox-imal control of the aortic arch be ob-tained prior to placement of the cross-clamp. Decision-making with regard tothe location of proximal clamp place-ment in order to avoid clamping acrossthe tear has been greatly aided by usingtwo-dimensional and three-dimensionalreformations (Figs 10–12).

The two basic strategies for the re-pair of descending traumatic tears areno distal perfusion adjuncts (simpleclamp and sew) and perfusion adjuncts(left heart bypass and cardiopulmonarybypass). The “clamp and sew” tech-nique has been used successfully bymany authors, but distal spinal cord is-chemia and paraplegia rates becomeunacceptably high after more than 30minutes of clamp time (79,80). There-fore, most authors will use some form ofdistal perfusion adjunct.

A meta-analysis of mortality andrisk of paraplegia in the repair of trau-matic aortic rupture in 1492 patientsshowed an overall postoperative para-plegia rate of 9.9%. Among patientstreated with simple aortic cross-clamp-ing, the hospital mortality was 16% andincidence of paraplegia was 19.2%; forpassive shunts, mortality was 12.3%and paraplegia rate was 11.1%; andwith active perfusion, paraplegia ratewas 2.3% (79,80). In another study (5),the incidence of paraplegia was 3.2%with cardiopulmonary bypass, 29%with simple aortic cross-clamping, andnone with nonheparinized bypass. Thesefindings have swayed the weight of evi-

Figure 19

Figure 19: Stent-graft for treatment of ATAI. (a) Transverse contrast-enhanced multidetector CT scanshows complex tear (arrow) of proximal descending thoracic aorta. (b) Conventional angiogram in left anterioroblique projection following stent-graft deployment shows exclusion of transection from aortic lumen.



dence toward the use of perfusion ad-juncts for these procedures. The actualreconstruction is usually performedwith resection of the injured segmentand placement of a Dacron graft.

Endovascular Stent-Graft RepairEndovascular stent-grafting of nontrau-matic thoracic and abdominal aortic an-eurysms was first described, to ourknowledge, in 1991 (81) and is now awell-established method of treatment.However, it was not until 1997 that thistechnique was adapted for the treat-ment of acute thoracic aortic injuries(82). There are now more than 500 doc-umented cases of endovascular treat-ment of ATAI in the radiologic and sur-gical literature (83). The premise is toprevent further rupture of the vascularinjury by exclusion from the systemicblood pressure of the thoracic aorta andthus reduce the risk of rupture (Fig 19).There are clear advantages of thismethod over open repair, includingavoidance of single lung ventilation, aorticcross-clamping, cardiopulmonary bypass,systemic anticoagulation, reduced bloodloss, and reduced surgical time (84).Endovascular repair may be performedin an immediate or delayed setting.Even in acutely injured patients withmultiple co-morbid injuries, stent-graft-ing is feasible (85–87).

Endovascular stent-grafting plan-ning with multidetector CT is critical fortechnical success. Dimensions to docu-ment are (a) the caliber of the aortaproximal and distal to the injury, (b) thedistance from the left subclavian arteryto the injury, (c) the length of the vascu-lar injury, and (d) any anatomic variants(83). These can be easily determined byusing sagittal oblique multiplanar refor-mations or curved reformatted imagesgenerated on a three-dimensional work-station or volume viewer. Technical suc-cess, defined as complete exclusion ofthe vascular injury, approaches 100%(84,88–90). Short-term complicationsinclude stroke, puncture-site complica-tions, device collapse, and recurrent la-ryngeal nerve damage, with morbidityranging 3%–36% (84,91–93). One limi-tation of this technique is the lack ofsmall-caliber devices for use in young

patients or those with a small aorta.Mortality ranges 0%–20%, comparedwith 15%–50% with emergent thora-cotomy (82,89,93).

Very little data are available in theliterature concerning long-term out-comes and complications of endovas-cular repair of thoracic aortic injuries,but data to date appear promising(83,84,88,89,92–94). At the time of thisreview, the Food and Drug Administra-tion has not yet approved the use ofstent-graft devices for the treatment oftraumatic or nontraumatic thoracic aor-tic aneurysms.

Future of Aortic Trauma Imaging

Cardiac gating in the setting of thoracictrauma has been entertained since theinception of 64-detector CT technology.The presumed advantage would be the“freezing” of motion of the ascendingaorta and a decrease in pulsation arti-fact within the descending aorta. How-ever, to our knowledge, clinical data inthe setting of acute trauma have yet tobe presented. Performing cardiac-gatedchest multidetector CT in all patientswith thoracic trauma may not be practi-cal in the acute trauma setting for anumber of reasons, including the addi-tional time of preparing the patient forcardiac gating, potential patient hemo-dynamic instability, tachycardia, andlimited availability of adequately trainedtechnologists. Finally, the longer breathhold may result in unintended artifactsfrom breathing that may mitigate theadvantage of cardiac gating. As such, itis likely to remain predominantly aproblem-solving tool.

Summary

The imaging evaluation, diagnosis, andmanagement of ATAI have undergonerapid change over the past 20 years.Previously considered a screening mo-dality for the detection of mediastinalhematoma, multidetector CT is now theaccepted initial imaging modality forsuspected ATAI, and in those patientswith unequivocal evidence of aortic in-jury, no further diagnostic imaging eval-uation is necessary. Angiography, TEA,

and intravascular US remain useful inequivocal multidetector CT cases andfor problem solving in unusual presenta-tions of ATAI. While urgent or emer-gent repair remains the standard ofcare, the treatment of ATAI is in theprocess of undergoing a similar shift,with endovascular stent-grafting tech-niques playing a larger role in initialmanagement.

Acknowledgment: The authors thank MicheleC. Ravenel, DDS, for her assistance with illustra-tion in Figure 1.

References1. Vesalius A, Bonetus T. Sepulchretum sive

anatomia practica ex cadaveribus morbodenatis. Sect 2. Geneva, Switzerland: 1700;290.

2. Frick EJ, Cipolle MD, Pasquale MD, et al.Outcome of blunt thoracic aortic injury in alevel 1 trauma center: an 8-year review.J Trauma 1997;43:844–851.

3. Gray L Jr, Kirsh M. A new roentgeno-graphic finding in acute traumatic rupture ofthe aorta. J Thorac Cardiovasc Surg 1975;70:86–88.

4. Parmley LF, Mattingly TW, Manion WC,Jahnke EJ Jr. Nonpenetrating traumatic in-jury of the aorta. Circulation 1958;17:1086–1101.

5. Kodali S, Jamieson WR, Leia-Stephens M,Miyagishima RT, Janusz MT, Tyers GF.Traumatic rupture of the thoracic aorta: a20-year review—1969–1989. Circulation1991;84(5 suppl):III40–III46.

6. Greendyke RM. Traumatic rupture of aorta:special reference to automobile accidents.JAMA 1966;195:527–530.

7. Feczko JD, Lynch L, Pless JE, Clark MA,McClain J, Hawley DA. An autopsy casereview of 142 nonpenetrating (blunt) inju-ries of the aorta. J Trauma 1992;33:846–849.

8. Burkhart HM, Gomez GA, Jacobson LE,Pless JE, Broadie TA. Fatal blunt aorticinjuries: a review of 242 autopsy cases.J Trauma 2001;50:113–115.

9. Dart CH Jr, Braitman HE. Traumatic rup-ture of thoracic aorta: diagnosis and man-agement. Arch Surg 1976;111:697–702.

10. Williams DM, Simon HJ, Marx MV, StarkeyTD. Acute traumatic aortic rupture: intra-vascular US findings. Radiology 1992;182:247–249.

11. National Highway Traffic Safety Administra-tion National Center for Statistics and Anal-ysis. Traffic safety facts: comparison analy-sis of fatality trend by age group—1996 to2005. Washington, DC: U.S. Department ofTransportation, 2007.

12. Hartford JM, Fayer RL, Shaver TE, et al.Transection of the thoracic aorta: assess-ment of a trauma system. Am J Surg 1986;151:224–229.

13. Cowley RA, Turney SZ, Hankins JR, Rodri-guez A, Attar S, Shankar BS. Rupture of thethoracic aorta caused by blunt trauma: a



fifteen-year experience. J Thorac Cardio-vasc Surg 1990;100:652–660; discussion660–661.

14. Williams JS, Graff JA, Uku JM, Steinig JP.Aortic injury in vehicular trauma. Ann Tho-rac Surg 1994;57:726–730.

15. Dosios TJ, Salemis N, Angouras D, NonasE. Blunt and penetrating trauma of the tho-racic aorta and aortic arch branches: anautopsy study. J Trauma 2000;49:696–703.

16. Katyal D, McLellan BA, Brenneman FD,Boulanger BR, Sharkey PW, Waddell JP.Lateral impact motor vehicle collisions: sig-nificant cause of blunt traumatic rupture ofthe thoracic aorta. J Trauma 1997;42:769–772.

17. Dischinger PC, Cushing BM, Kerns TJ. In-jury patterns associated with directionimpact: drivers admitted to trauma centers.J Trauma 1993;35:454–458; discussion458–459.

18. Loo GT, Siegel JH, Dischinger PC, et al.Airbag protection versus compartment in-trusion effect determines the pattern of inju-ries in multiple trauma motor vehiclecrashes. J Trauma 1996;41:935–951.

19. Shkrum MJ, McClafferty KJ, Green RN,Nowak ES, Young JG. Mechanisms of aorticinjury in fatalities occurring in motor vehiclecollisions. J Forensic Sci 1999;44:44–56.

20. Cohen AM, Crass JR, Thomas HA, FisherRG, Jacobs DG. CT evidence for the “osse-ous pinch” mechanism of traumatic aorticinjury. AJR Am J Roentgenol 1992;159:271–274.

21. Crass JR, Cohen AM, Motta AO, TomashefskiJF Jr, Wiesen EJ. A proposed new mecha-nism of traumatic aortic rupture: the osse-ous pinch. Radiology 1990;176:645–649.

22. Creasy JD, Chiles C, Routh WD, Dyer RB.Overview of traumatic injury of the thoracicaorta. RadioGraphics 1997;17:27–45.

23. Lundervall J. The mechanism of traumaticrupture of the aorta. Acta Pathol MicrobiolScand 1964;62:34–46.

24. Marco JV, Gregory JS. Posterior fracture ofthe left sixth rib causing late aorticlaceration: case report. J Trauma 1997;42:736–737.

25. Murakami R, Tajima H, Ichikawa K, et al.Acute traumatic injury of the distal descend-ing aorta associated with thoracic spine in-jury. Eur Radiol 1998;8:60–62.

26. Richens D, Field M, Neale M, Oakley C. Themechanism of injury in blunt traumatic rup-ture of the aorta. Eur J Cardiothorac Surg2002;21:288–293.

27. Baque P, Serre T, Cheynel N, et al. An ex-perimental cadaveric study for a better un-derstanding of blunt traumatic aortic rup-ture. J Trauma 2006;61:586–591.

28. Wintermark M, Wicky S, Schnyder P. Imag-ing of acute traumatic injuries of the tho-racic aorta. Eur Radiol 2002;12:431–442.

29. Patel NH, Stephens KE Jr, Mirvis SE,Shanmuganathan K, Mann FA. Imaging ofacute thoracic aortic injury due to blunttrauma: a review. Radiology 1998;209:335–348.

30. Merrill WH, Lee RB, Hammon JW Jr, FristWH, Stewart JR, Bender HW Jr. Surgical

treatment of acute traumatic tear of the tho-racic aorta. Ann Surg 1988;207:699–706.

31. Kram HB, Wohlmuth DA, Appel PL,Shoemaker WC. Clinical and radiographicindications for aortography in blunt chesttrauma. J Vasc Surg 1987;6:168–176.

32. Blackmore CC, Zweibel A, Mann FA. Deter-mining risk of traumatic aortic injury: howto optimize imaging strategy. AJR Am JRoentgenol 2000;174:343–347.

33. Kirkham JR, Blackmore CC. Screening foraortic injury with chest radiography andclinical factors. Emerg Radiol 2007;14:211–217.

34. Sturm JT, McGee MB, Luxenberg MG. Ananalysis of risk factors for death at the scenefollowing traumatic aortic rupture.J Trauma 1988;28:1578–1580.

35. Cernaianu AC, Cilley JH Jr, Baldino WA,Spence RK, DelRossi AJ. Determinants ofoutcome in lesions of the thoracic aorta inpatients with multiorgan system trauma.Chest 1992;101:331–335.

36. Lee RB, Stahlman GC, Sharp KW. Treat-ment priorities in patients with traumaticrupture of the thoracic aorta. Am Surg1992;58:37–43.

37. Orron DE, Porter DH, Kim D, Tortella B.False-positive aortography following bluntchest trauma: case report. Cardiovasc Inter-vent Radiol 1988;11:132–135.

38. Turney SZ, Attar S, Ayella R, Cowley RA,McLaughlin J. Traumatic rupture of theaorta: a five-year experience. J Thorac Car-diovasc Surg 1976;72:727–734.

39. Fabian TC, Richardson JD, Croce MA, et al.Prospective study of blunt aortic injury:Multicenter Trial of the American Associa-tion for the Surgery of Trauma. J Trauma1997;42:374–380; discussion 380–383.

40. Marnocha KE, Maglinte DD, Woods J,Goodman M, Peterson P. Mediastinal-width/chest-width ratio in blunt chesttrauma: a reappraisal. AJR Am J Roentge-nol 1984;142:275–277.

41. Woodring JH, Dillon ML. Radiographicmanifestations of mediastinal hemorrhagefrom blunt chest trauma. Ann Thorac Surg1984;37:171–178.

42. Mirvis SE, Bidwell JK, Buddemeyer EU,Diaconis JN, Pais SO, Whitley JE. Imagingdiagnosis of traumatic aortic rupture: a re-view and experience at a major trauma cen-ter. Invest Radiol 1987;22:187–196.

43. Sefczek DM, Sefczek RJ, Deeb ZL. Radio-graphic signs of acute traumatic rupture ofthe thoracic aorta. AJR Am J Roentgenol1983;141:1259–1262.

44. Lee J, Harris JH Jr, Duke JH Jr, WilliamsJS. Noncorrelation between thoracic skele-tal injuries and acute traumatic aortic tear.J Trauma 1997;43:400–404.

45. Dyer DS, Moore EE, Ilke DN, et al. Thoracicaortic injury: how predictive is mechanismand is chest computed tomography a reli-able screening tool?—a prospective study of1,561 patients. J Trauma 2000;48:673–682;discussion 682–683.

46. Fabian TC, Davis KA, Gavant ML, et al.Prospective study of blunt aortic injury: he-lical CT is diagnostic and antihypertensivetherapy reduces rupture. Ann Surg 1998;227:666–676; discussion 676–677.

47. Gavant ML, Menke PG, Fabian T, Flick PA,Graney MJ, Gold RE. Blunt traumatic aorticrupture: detection with helical CT of thechest. Radiology 1995;197:125–133.

48. Mirvis SE, Shanmuganathan K, Miller BH,White CS, Turney SZ. Traumatic aorticinjury: diagnosis with contrast-enhancedthoracic CT—5-year experience at a majortrauma center. Radiology 1996;200:413–422.

49. Melton SM, Kerby JD, McGiffin D, et al.The evolution of chest computed tomogra-phy for the definitive diagnosis of blunt aor-tic injury: a single-center experience.J Trauma 2004;56:243–250.

50. Fishman JE, Nunez D Jr, Kane A, Rivas LA,Jacobs WE. Direct versus indirect signs oftraumatic aortic injury revealed by helicalCT: performance characteristics and inter-observer agreement. AJR Am J Roentgenol1999;172:1027–1031.

51. Bruckner BA, DiBardino DJ, Cumbie TC,et al. Critical evaluation of chest computedtomography scans for blunt descending tho-racic aortic injury. Ann Thorac Surg 2006;81:1339–1346.

52. Ellis JD, Mayo JR. Computed tomographyevaluation of traumatic rupture of the tho-racic aorta: an outcome study. Can AssocRadiol J 2007;58:22–26.

53. Mirvis SE, Shanmuganathan K, Buell J,Rodriguez A. Use of spiral computed tomog-raphy for the assessment of blunt traumapatients with potential aortic injury.J Trauma 1998;45:922–930.

54. Scaglione M, Pinto A, Pinto F, Romano L,Ragozzino A, Grassi R. Role of contrast-enhanced helical CT in the evaluation ofacute thoracic aortic injuries after bluntchest trauma. Eur Radiol 2001;11:2444–2448.

55. Sammer M, Wang E, Blackmore CC, BurdickTR, Hollingworth W. Indeterminate CT an-giography in blunt thoracic trauma: is CTangiography enough? AJR Am J Roentgenol2007;189:603–608.

56. Wong H, Gotway MB, Sasson AD, JeffreyRB. Periaortic hematoma at diaphragmaticcrura at helical CT: sign of blunt aortic in-jury in patients with mediastinal hematoma.Radiology 2004;231:185–189.

57. Mirvis SE, Shanmuganathan K. Diagnosis ofblunt traumatic aortic injury 2007: still anemesis. Eur J Radiol 2007;64:27–40.

58. Malhotra AK, Fabian TC, Croce MA,Weiman DS, Gavant ML, Pate JW. Minimalaortic injury: a lesion associated with ad-vancing diagnostic techniques. J Trauma2001;51:1042–1048.

59. Pretre R, LaHarpe R, Cheretakis A, et al.Blunt injury to the ascending aorta: threepatterns of presentation. Surgery 1996;119:603–610.

60. Steenburg SD, Ravenel JG, Ikonomidis JS.Blunt traumatic injury of the ascendingaorta: multidetector CT findings in twocases. Emerg Radiol 2007;13:217–221.

61. Symbas PJ, Horsley WS, Symbas PN. Rup-ture of the ascending aorta caused by blunttrauma. Ann Thorac Surg 1998;66:113–117.

62. Chen MY, Miller PR, McLaughlin CA, KortesisBG, Kavanagh PV, Dyer RB. The trend of



using computed tomography in the detec-tion of acute thoracic aortic and branch ves-sel injury after blunt thoracic trauma: sin-gle-center experience over 13 years.J Trauma 2004;56:783–785.

63. Chen MY, Regan JD, D’Amore MJ, RouthWD, Meredith JW, Dyer RB. Role of an-giography in the detection of aortic branchvessel injury after blunt thoracic trauma.J Trauma 2001;51:1166–1172.

64. Holdgate A, Dunlop S. Review of branchaortic injuries in blunt chest trauma. EmergMed Australas 2005;17:49–56.

65. Mirvis SE. Thoracic vascular injury. RadiolClin North Am 2006;44:181–197, vii.

66. Gavant ML, Flick P, Menke P, Gold RE. CTaortography of thoracic aortic rupture. AJRAm J Roentgenol 1996;166:955–961.

67. Rizoli SB, Brenneman FD, Boulanger BR,Maggisano R. Blunt diaphragmatic and tho-racic aortic rupture: an emerging injurycomplex. Ann Thorac Surg 1994;58:1404–1408.

68. Fisher RG, Sanchez-Torres M, WhighamCJ, Thomas JW. “Lumps” and “bumps” thatmimic acute aortic and brachiocephalic ves-sel injury. RadioGraphics 1997;17:825–834.

69. Fattori R, Celletti F, Bertaccini P, et al. De-layed surgery of traumatic aortic rupture:role of magnetic resonance imaging. Circu-lation 1996;94:2865–2870.

70. Fattori R, Celletti F, Descovich B, et al.Evolution of post-traumatic aortic aneurysmin the subacute phase: magnetic resonanceimaging follow-up as a support of the surgi-cal timing. Eur J Cardiothorac Surg 1998;13:582–586; discussion 586–587.

71. Fisher RG, Hadlock F. Laceration of the tho-racic aorta and brachiocephalic arteries byblunt trauma: report of 54 cases and reviewof the literature. Radiol Clin North Am1981;19:91–110.

72. Ahrar K, Smith DC. Trauma to the aortaand aortic arch branches. Curr Opin Cardiol1998;13:355–368.

73. Ahrar K, Smith DC, Bansal RC, Razzouk A,

Catalano RD. Angiography in blunt thoracicaortic injury. J Trauma 1997;42:665–669.

74. Sturm JT, Hankins DG, Young G. Thoracicaortography following blunt chest trauma.Am J Emerg Med 1990;8:92–96.

75. Uflacker R, Horn J, Phillips G, Selby JB.Intravascular sonography in the assessmentof traumatic injury of the thoracic aorta.AJR Am J Roentgenol 1999;173:665–670.

76. Cinnella G, Dambrosio M, Brienza N, TulloL, Fiore T. Transesophageal echocardiogra-phy for diagnosis of traumatic aortic injury:an appraisal of the evidence. J Trauma2004;57:1246–1255.

77. Patel NH, Hahn D, Comess KA. Blunt chesttrauma victims: role of intravascular ultra-sound and transesophageal echocardiogra-phy in cases of abnormal thoracic aorto-gram. J Trauma 2003;55:330–337.

78. Konstadt SN, Reich DL, Quintana C, LevyM. The ascending aorta: how much doestransesophageal echocardiography see?Anesth Analg 1994;78:240–244.

79. von Oppell UO, Dunne TT, De Groot KM,Zilla P. Spinal cord protection in the ab-sence of collateral circulation: meta-analysisof mortality and paraplegia. J Card Surg1994;9:685–691.

80. von Oppell UO, Dunne TT, De Groot MK,Zilla P. Traumatic aortic rupture: twenty-year metaanalysis of mortality risk of para-plegia. Ann Thorac Surg 1994;58:585–593.

81. Parodi JC, Palmaz JC, Barone HD. Trans-femoral intraluminal graft implantation forabdominal aortic aneurysms. Ann Vasc Surg1991;5:491–499.

82. Kato N, Dake MD, Miller DC, et al. Trau-matic thoracic aortic aneurysm: treatmentwith endovascular stent-grafts. Radiology1997;205:657–662.

83. Rousseau H, Midulla M, Marcheix B, et al.Indications for endovascular treatment:which treatment for acute traumatic aorticdissections? In: Amor M, Bergeron P, In-glese L, Ischinger T, Mathias K, Raithel D,eds. Thoracic combo from theory topractice: thoracic aortic dissections andtheir endovascular treatment. Marseille,France: Com, 2007;192:–202.

84. Tehrani HY, Peterson BG, Katariya K, et al.Endovascular repair of thoracic aortic tears.Ann Thorac Surg 2006;82:873–877; discus-sion 877–878.

85. Agostinelli A, Saccani S, Borrello B, NicoliniF, Larini P, Gherli T. Immediate endovascu-lar treatment of blunt aortic injury: ourtherapeutic strategy. J Thorac CardiovascSurg 2006;131:1053–1057.

86. Bortone AS, Schena S, D’Agostino D, et al.Immediate versus delayed endovasculartreatment of post-traumatic aortic pseudoa-neurysms and type B dissections: retrospec-tive analysis and premises to the upcomingEuropean trial. Circulation 2002;106(12suppl 1):I234–I240.

87. Lobato AC, Quick RC, Phillips B, et al. Im-mediate endovascular repair for descendingthoracic aortic transection secondary toblunt trauma. J Endovasc Ther 2000;7:16–20.

88. Amabile P, Collart F, Gariboldi V, Rollet G,Bartoli JM, Piquet P. Surgical versus endo-vascular treatment of traumatic thoracicaortic rupture. J Vasc Surg 2004;40:873–879.

89. Semba CP, Kato N, Kee ST, et al. Acuterupture of the descending thoracic aorta:repair with use of endovascular stent-grafts.J Vasc Interv Radiol 1997;8:337–342.

90. Garzon G, Fernandez-Velilla M, Marti M,Acitores I, Ybanez F, Riera L. Endovascularstent-graft treatment of thoracic aortic dis-ease. RadioGraphics 2005;25(suppl 1):S229–S244.

91. Ott MC, Stewart TC, Lawlor DK, Gray DK,Forbes TL. Management of blunt thoracicaortic injuries: endovascular stents versusopen repair. J Trauma 2004;56:565–570.

92. Malloy PC, Richard HM 3rd. Thoracic an-giography and intervention in trauma. Ra-diol Clin North Am 2006;44:239–249, viii.

93. Kasirajan K, Heffernan D, Langsfeld M.Acute thoracic aortic trauma: a comparisonof endoluminal stent grafts with open repairand nonoperative management. Ann VascSurg 2003;17:589–595.

94. Scheinert D, Krankenberg H, Schmidt A,et al. Endoluminal stent-graft placement foracute rupture of the descending thoracicaorta. Eur Heart J 2004;25:694–700.



Acute Traumatic Aortic Injury: ImagingEvaluationandManagementmsrads.web.unc.edu/files/2018/08/Traumatic-Aortic-Injury.pdf · prevalence of serious thoracic trauma and aortic injuries.

Documents