Key Words:
blunt trauma , thoracic aortic injury , screening , diagnosis , open repair , endovascular repair , device-related complications , mortality , paraplegia
Introduction
The screening, diagnosis, method, and timing of definitive management of blunt thoracic aortic injuries (BTAI) have undergone revolutionary changes over the past decade. Routine chest computed tomography (CT) scan has replaced plain x-rays for screening purposes, and contrast-enhanced CT angiography (CTA) has replaced formal angiography as a method of definitive diagnosis. Semi-elective definitive repair of BTAI instead of emergency repair has now become the new standard, and endovascular stent grafts have largely replaced open operative repair. All of these changes have resulted in a significant reduction in early mortality and complications.
Historical Background
The first case of blunt thoracic aortic injury was reported by the anatomist Andreas Vesalius in a man who fell from a horse in 1557. The first reported repair of an acute repair of a BTAI occurred in the late 1950s. In the 1970s there was the development and widespread use of various shunting techniques and graft materials. In the 1990s, we see the first reports supporting routine use of CT scan as a screening method in patients with a suspicious mechanism of injury; and soon afterward contrast-enhanced CTA is advocated as the preferred method of definitive diagnosis of BTAI. In 1997, the first endovascular repair of a patient with BTAI was reported; and in the 2000s, endovascular aortic repair (EVAR) became the new preferred therapeutic approach.
Epidemiology
It is estimated that 8000 to 9000 blunt-trauma victims suffer thoracic aortic injury every year in the United States. The majority of these are due to motor-vehicle crashes (about 70%), followed by motorcycle crashes (13%), falls from heights (7%), automobile–pedestrian accidents (7%), and other mechanisms. The overall incidence of thoracic aortic injury in patients reaching the hospital alive is less than 0.5%. In a series of 5838 pedestrian injuries reaching hospital care, the incidence of BTAI was 0.3%. Likewise, in a study based on National Trauma Data Bank (NTDB), Arthurs et al identified 3144 blunt thoracic aortic injuries among 1.1 million trauma admissions for an overall incidence of 0.3%. In another study of 1613 admissions following high-level falls, the incidence of BTAI was 0.1%. The presence of a pelvic fracture is a marker of an associated thoracic aortic injury. In analysis of 1450 pelvic fractures, aortic injury was diagnosed in 1.4%. However, it seems that patients with this injury pattern who reach the hospital alive are only the tip of the iceberg, and the real incidence of BTAI is much higher. The vast majority of patients with this injury dies at the scene and never reach hospital care. The incidence of aortic injuries in fatal traffic injuries is very high. In a recent analysis of 304 deaths due to blunt trauma in the county of Los Angeles, 102 patients (33%) had a rupture of the thoracic aorta. About 80% of the deaths due to a free rupture of the aorta occurred at the scene, and only 20% occurred in the hospital ( Fig. 10-1 ). In another autopsy analysis of 25 fatalities in a recent train crash in Los Angeles, thoracic aortic rupture was found in 8 cases (33%). All mortalities occurred at the scene.
The incidence of aortic trauma increases with age, and it is rare to find this injury in the pediatric population. In a NTDB analysis, the incidence of thoracic aortic injury in children younger than 16 years old was 7 times lower than in adults (0.03% vs. 0.21%). In an analysis of 5838 automobile–pedestrian injuries, there were no aortic injuries in the age group 14 years and younger. The incidence increased to 0.2% in the group 15 to 65 years, to 0.5% in the group 56 to 65 years, and to 1.5% in the group older than 65 years.
About 40% of patients with aortic rupture have severe associated injuries (body area Abbreviated Injury Scale [AIS] score ≥ 4), the most common being the head and abdomen. The mean injury severity score (ISS) of 40 is a strong indicator of the grave condition of patients with pattern of vascular trauma.
Site and Type of Aortic Injury
The most common anatomic location of aortic injury is the medial aspect of the lumen, distal to the left subclavian artery ( Fig. 10-2 ). In a prospective analysis of 185 cases of thoracic aortic injury, the rupture involved the isthmus in 75%, followed by the descending thoracic aorta in 22% and the ascending aorta in 4%. Computer simulation and cadaver studies have shown that the combination of increased intraaortic pressure (mean 1149 mm Hg) and rotational forces exerts a highly focused stress at the isthmus. In addition, the tensile strength at the isthmus was found to be only 63% of that of the proximal aorta. The most common type of injury is a false aneurysm (58%), followed by dissection (25%) and intimal tear (20%) ( Fig. 10-3 ).
Natural History of Blunt Thoracic Aortic Injuries
The majority of patients with BTAI die at the scene (before reaching hospital care). In analysis of 242 fatal BTAI, Burkhart et al reported that 57% of the deaths occurred at the scene or on arrival to the hospital, 37% died within first 4 hours of admission, and 6% died at a time point greater than 4 hours after admission. In another autopsy study of 102 patients with BTAI, about 80% of the deaths occurred at the scene and only 20% in the hospital. In a recent NTDB-based study, 68% of patients suffering BTAI never underwent an attempt at treatment due to early death or severity of associated injuries.
Screening and Diagnosis
Supine chest x-ray has been used as the primary screening tool for diagnosis of BTAI. Numerous radiographic findings have been described as markers of aortic trauma, including the presence of a widened upper mediastinum (greater than 8 cm on an anteroposterior supine chest film at the level of the aortic knob) ( Fig. 10-4 ), obliteration of the aortic contour, and loss of the paravertebral pleural stripe. Additionally, depression of the left mainstem bronchus; rightward deviation of the nasogastric tube; left apical pleural hematoma (i.e., apical cap); left hemothorax; and fracture of the sternum, scapula, upper ribs, or clavicle have been shown to be associated with this pattern of vascular trauma. The presence of a widened mediastinum is the most common finding, but this finding still has a low sensitivity and specificity. Many conditions, such as fracture of the sternum or thoracic spine or simply the supine position in an obese patient may cause a widened appearance of the mediastinum on chest x-ray. The most specific signs are loss of the aortic knob, an abnormality of the aortic arch and deviation of the nasogastric tube. However, the sensitivity of these findings is still quite low.
Traditionally, a normal chest x-ray had been considered reliable in excluding BTAI. However, numerous studies have now shown that basic chest x-ray is a poor screening tool for aortic trauma and that a significant number of injuries may not show any mediastinal abnormalities. Given this understanding of the limitations of chest x-ray, many centers now use contrast-enhanced CT scan of the chest as the primary screening tool for BTAI, irrespective of x-ray findings. The sensitivity and negative predictive value of the CT scan in the diagnosis of BTAI approaches 100%.
Recently, Starnes and colleagues proposed an insightful classification of BTAI based on high-quality, contrast-enhanced CT imaging. These authors stratified aortic injury into four categories related to the presence or absence of an abnormality in the external contour of the wall of the descending thoracic aorta. The injuries without external contour alterations include small intimal tears (less than 10 mm; Category I) and more extensive intimal flaps (greater than 10 mm; Category II). Injuries with changes in external-wall contour include pseudoaneurysms (Category III) and rupture of the aorta (Category IV) ( Fig. 10-5 ). These investigators also suggested readily applicable treatment guidelines related to this useful radiographic classification.
Catheter-based arch aortography remained the gold standard for the definitive diagnosis of BTAI until the late 1990s. However, it is invasive, takes time, and carries with it a small risk of stroke. In the last few years the contrast-enhanced CT scan has replaced formal aortography in the definitive diagnosis of BTAI. The new generation multislice CT scanners with 3-D reconstruction have been shown to have almost 100% sensitivity and specificity and 90%-positive and 100%-negative predictive values. In these studies, the overall diagnostic accuracy of contrast-enhanced CT imaging has been shown to be greater than 99% ( Fig. 10-6 ). Catheter-based arch aortography still has a limited diagnostic role in the rare cases where the CT scan findings are suspicious but not diagnostic. Aortography with or without intravascular ultrasound (IVUS) may also be performed as the initial diagnostic or therapeutic step in the endovascular treatment of BTAI (i.e., stent-graft repair).
Transesophageal echocardiography (TEE) is another diagnostic modality in the evaluation of suspected BTAI. The initial enthusiasm surrounding this modality has been replaced by skepticism and failure to gain popularity because of conflicting reports about accuracy and concerns regarding its ready availability. The dramatic shifting from formal aortography and TEE to contrast-enhanced CT scanning in the diagnosis of BTAI is demonstrated by a multicenter study sponsored by the American Association for the Surgery of Trauma (AAST). The use of angiography and TEE for the diagnosis of thoracic aortic injuries decreased from 87% and 12%, respectively, in 1997 to only 8% and 1% in 2007 ( Table 10-1 ). Other diagnostic modalities such as magnetic resonance imaging (MRI) or IVUS may be useful in rare patients where the CTA findings are not definitive.
| AAST 1 | AAST 2 | p -value | |
|---|---|---|---|
| Number | 253 | 193 | |
| Aortogram | 207 (87%) | 16 (8.3%) | <0.001 |
| CT scan | 88 (34.8%) | 180 (93.3%) | <0.001 |
| TEE | 30 (11.9%) | 2 (1.0%) | <0.001 |
In summary, the new generation of scanners has made contrast-enhanced CT scanning the standard diagnostic modality for screening and diagnosis of BTAI. Traditional catheter-based aortography may have a diagnostic role in patients undergoing arteriography for other injuries such as pelvic fracture or complex liver injury or as an initial step to endovascular repair of BTAI. TEE might be useful in critically ill patients in the intensive care unit who cannot be transferred safely to the radiology suite for CT scan.
Initial Management of Thoracic Aortic Injuries
Prompt diagnosis and early appropriate treatment remain the primary tenets of survival for patients with BTAI. Stabilization and prevention of free rupture of a contained aortic injury until definitive repair can be performed is the most urgent priority. The risk of free rupture is highest in the first few hours after the injury, with more than 90% of ruptures occurring within the first 24 hours. In an AAST multicenter study by Fabian et al, 24 (8.8%) of the 274 patients in the study population progressed to free rupture. However, rigorous blood pressure control reduces the risk of rupture to about 1.5%. Blood pressure control is best achieved with a combination of fluid restriction and pharmacologic intervention. The systolic blood pressure should be kept as low as tolerated, in most patients at about 90 mm Hg to 110 mm Hg. In elderly patients, the optimal systolic pressure may be slightly higher. Cautious restriction of intravenous fluids and administration of beta-blocker therapy in the form of an esmolol drip are the most commonly used modalities for blood pressure control.
Timing of Definitive Management
Untreated, the risk of rupture of a BTAI is highest in the first 24 hours after injury. In the AAST multicenter study by Fabian et al, 24 (8.8%) of patients progressed to free rupture. Ninety-two percent of patients with ruptures died within 24 hours of injury, 1 at 30 hours and 1 at 6 days. In the group of 13 free ruptures in which the time of rupture was known, 46% occurred within 4 hours and another 38% took place within 8 hours after injury. For these reasons, the definitive management of TAI had been considered as an emergency, and this policy remained the standard for many years. However, subsequent studies demonstrated that the early initiation of pharmacologic blood pressure control with restrictive fluid resuscitation decreases wall stress in the region of the injury and reduces the risk of rupture to approximately 1.5%. Patients with contained aortic injury who survive beyond 4 hours after injury with this medical management rarely progress to rupture and death. The successful management of these injuries therefore hinges on early diagnosis and careful blood pressure control.
In the late 1990s and early 2000s, some studies suggested that select patients with BTAI and major associated injuries could be managed safely with delayed aortic repair. This approach was demonstrated to be safe and allowed for stabilization of other major associated injuries provided that the patient’s blood pressure was adequately controlled. The concept of delayed repair was subsequently adopted more liberally and was shown to be safe in patients with no severe associated injuries or other major comorbidities.
The safety of delayed BTAI repair and its effect on outcomes was a matter of controversy until recently. Most clinical studies of this scenario included only patients with major associated injuries and reported contradictory results. Some studies showed improved outcome with delayed repair while others failed to show benefit. Wahl et al, in a retrospective review of 48 cases, reported that delayed (greater than 24 hours following injury) aortic repair was safe but was associated with longer hospital stay and was costlier than early repair. A similar study of 78 patients by Hemmila et al reported a higher complication rate and a longer hospital stay in the group of patients who underwent delayed (greater than16 hours following injury) repair.
However, other studies suggested that delayed repair was associated with improved outcome. A recent AAST multicenter, prospective study analyzed outcomes in 178 patients with BTAI according to timing of repair. In this study, early repair was classified as being within 24 hours of injury, and delayed repair was classified as that performed after 24 hours. In this study, the two groups were similar with regard to injury severity major associated injuries, type of aortic injury, and type of aortic repair (operative versus endovascular). The mean time from injury to repair was 10.2 hours in the early group and 126.2 hours in the delayed group. In the study, the overall mortality in the delayed-repair group was lower than that of the early-repair group (5.8% versus 16.5%, p = 0.034). A multivariate analysis adjusting for injury severity, severe extrathoracic injuries, Glasgow Coma Score (GCS), hypotension on admission, advanced age, and method of aortic injury repair showed an increased risk of death in the early-repair group (adjusted OR [95% CI] 7.78 [1.69 to 35.70], adjusted p-value = 0.008). The survival benefits in the delayed-repair group were confirmed in the subanalysis of the groups with or without major associated injuries. The incidence of paraplegia was similar in the two groups (early repair 1.8%; delayed repair 1.4%). In a more recent trauma registry study of 145 BTAI repairs, delayed repair was identified as the only independent factor shown to be associated with lower mortality.
These recent studies provide the strongest evidence to date that with adequate medical management delayed repair is not only safe but may be preferable to emergent repair in select patients. This approach allows for optimizing patient risk factors and physiologic condition and ensures that other more life-threatening injuries can be prioritized. The optimal time from injury or admission to repair is unknown and should be individualized, taking into account factors such as the presence of other severe injuries or comorbid conditions, the physiologic status of the patient, and the type and severity of the aortic injury. Delayed repair should not be attempted in cases with active leaking from the aortic injury ( Fig. 10-7 ). Urgent repair, within a few hours of diagnosis, remains advisable in cases with large contained injuries.
Definitive Management of Thoracic Aortic Injuries
Open operative repair of BTAI was for many decades the only standard definitive management. However, in the last decade there has been a dramatic increase in the use of endovascular stent grafts to treat this injury pattern. This shift is clearly demonstrated by two large prospective studies by the AAST in 1997 and in 2007. In the 1997 study, all 207 cases of BTAI were managed with open operative repair; while, in the 2007 study, 65% of the 193 cases of BTAI were managed with endovascular stent-graft repair and only 35% with open operative repair ( Table 10-2 ). Currently, the only absolute indications for open repair are a small aorta or an injury involving the aortic arch where placement of an endograft might be technically difficult or impossible. A third and evolving therapeutic option for select cases with minor aortic injuries is observation combined with medical therapy.
