The trusted source for
healthcare information and
Author: Ademola Adewale, MD, Attending Emergency Medicine Physician, Florida Emergency Physicians, Department of Emergency Medicine, Florida Hospital, Orlando.
Peer Reviewer: Stephen Crabtree, DO, FACEP, Center for Operational Medicine, Tactical Medicine Section, Department of Emergency Medicine, Medical College of Georgia, Augusta.
Trauma to the thoracic cavity is responsible for approximately 10-25% of all trauma-related deaths,1,2 with the majority of these deaths occurring after arrival at the emergency department (ED). The mortality for isolated chest injury is relatively low (less than 5%); however, with multiple organ system involvement, the mortality approaches 30%.1 This article dissects the critical aspects of thoracic trauma and highlights acute care management strategies.— The Editor
Thoracic trauma may be either blunt or penetrating. Compression, deceleration forces, and direct impact are mechanisms that result in the injury patterns seen in thoracic trauma. Unlike blunt trauma, penetrating trauma results from extrinsic violation of the integrity of the thoracic cavity. These injury mechanisms may result in pneumothorax, hemothorax, pulmonary contusion, or injuries to the mediastinal structures. With the proliferation of firearms, penetrating chest injuries are becoming more prevalent, and nearly all result in development of a pneumothorax, with hemothorax occurring in almost 75% of the cases.1 The use of scoring scales such as injury severity scale, abbreviated injury severity scale, or the thoracic trauma severity score may assist the ED physician and trauma team in algorithmic decision-making. Although controversies abound regarding management of traumatic arrest patients, the decision to perform thoracotomy should be individualized. Moreover, fewer than 10% of blunt chest injuries and 15-30% of penetrating injuries require thoracotomy.2
Morbidity and mortality from chest injuries results from the injury pattern itself and physiologic derangements such as hypoxia, hypercarbia, and acidosis. The pathophysiology of these clinical entities arises from inadequate tissue perfusion from hypovolemia (secondary to blood loss), ventilation and perfusion mismatch from pulmonary contusion, or change in intrathoracic pressure from either tension or open pneumothorax.2
The presence of severe respiratory distress is associated with a high mortality, with 10% of these patients requiring intubation at the scene or immediately on arrival in the ED.1 The most common associated risk factors for respiratory distress include multiple rib fractures, shock, pneumothorax, hemopneumothorax, and coma.1
Thoracic cage injuries result from the direct effect of trauma to the chest wall or the thoracic cage. They include flail chest and rib, sternal, thoracic spine, and scapular fractures.
Rib Fractures and Flail Chest. The ribs are the most commonly injured structures in the thorax.2 This injury should be suspected in patients with localized chest wall pain or contusion over one or more rib segments after trauma. The upper ribs, first through third, are well protected by the bony framework of the upper limb; protective structures include the scapula, humerus, and clavicle, along with their muscular attachments.2 Fracture of the first, second, or third rib requires a significant amount of force and should lead to high index of suspicion for injuries involving the head, neck, spinal cord, tracheobronchial tree, lungs, or the great vessels.1,2 The mortality associated with these fractures is approximately 15-30%.1,2
The middle ribs, four through nine, are most commonly affected by blunt trauma.2 A direct force striking anterior-to-posterior on the thoracic cage fractures these ribs along their shafts and tends to drive the ends of the bones into the thoracic cavity. This subsequently leads to intrathoracic injuries such as pneumothorax or lung contusion.2 Furthermore, fracture of the lower ribs (ribs 10-12) should heighten suspicion for intra-abdominal injuries involving the liver and spleen.
A flail chest results when fractures occur in two or more ribs, or a fracture involves multiple segments of a rib. This results in a lack of continuity between the fractured segments and the rest of the thoracic cage, resulting in the disruption of normal chest wall movement. The morbidity from flail chest is not from the paradoxical chest wall movement produced, but rather from the underlying injury to the lung parenchyma (pulmonary contusion), and persistent splinting that results from severe pain. The pulmonary contusion leads to ventilation perfusion mismatch, leading to hypoxia. (See Figure 1; see more on pulmonary contusion in the section dedicated to this entity.)
Elderly and Pediatric Patients. The presence of rib fractures in older individuals carries significant morbidity and mortality. An elderly patient with a rib fracture resulting from blunt chest trauma has twice the morbidity and mortality of younger patients.4,5 For each additional rib fracture, mortality increases by 19%.4 Similarly, pediatric patients who present with rib fractures should alert the practitioner to possible child abuse. Although a rib fracture could occur in a pediatric patient who sustains blunt chest trauma, it is very rare. Its presence portends the possibility of concomitant severe injuries. Thus, geriatric and pediatric patients with multiple rib fractures warrant extreme vigilance.
The initial evaluation of a patient with a rib fracture involves palpation for localized pain, crepitus, subcutaneous emphysema, and deformities. In simple rib fractures, chest radiography alone will suffice. The purpose of the study is not to visualize the rib per se, but to evaluate for possible coexisting complications such as pneumothorax. Patients who sustain rib fractures from significant trauma to the thoracic cage, in addition to chest radiography, require computed tomography (CT) of the chest to qualify the extent of intrathoracic organ involvement (e.g., hemopneumothorax or aortic, bronchial, esophageal, cardiac, or diaphragmatic injuries).
Initial therapeutic intervention for an isolated rib fracture focuses on adequate ventilation and analgesia for pain control. Many therapeutic regimens are available for pain management. These modalities include opioid analgesics, transcutaneous electric nerve stimulation, non-steroidal anti-inflammatory drugs, and regional nerve blocks (intercostals, epidural, interpleural, and thoracic paravertebral).6,7
In patients with simple flail chest and minimal or no lung contusion, analgesia or intercostal nerve blocks with chest physiotherapy may be adequate. In patients with flail chest and moderate to severe pulmonary contusion, evidence of hypoxia, and signs of shock, mechanical ventilation should be initiated. Mechanical ventilation should be considered in individuals with flail chest and shock, three or more associated injuries, severe head injury, underlying pulmonary disease, fracture of eight or more ribs, or age older than 65 years.1 Bergeron et al showed that after adjusting for severity, co-morbidity, multiple rib fractures, and age, patients older than 65 years had five times the odds of dying when compared to those younger than 65 years.8 In patients with flail chest and two or more injuries, early mechanical ventilatory support has been shown to reduce the mortality from 69% to 7%.1
The disposition of a patient with a rib fracture resulting from blunt chest trauma should be individualized. Patients with isolated rib fractures may be discharged home safely with good analgesia and an incentive spirometer. In patients with multiple rib fractures, admission often is recommended for pain control and to minimize the potential associated morbidities. All geriatric patients with multiple rib fractures require admission for management.4,5,8
Sternal Fracture. Sternal fractures commonly result from a direct blow to the chest wall during motor vehicular accidents or from falls, and account for approximately 8% of thoracic injuries.9 The majority of these injuries result from direct contact with the steering wheel or from seatbelt compression. The presence of a sternal injury should alert practitioners to the possibility of an underlying cardiac or pulmonary injury. However, in the absence of concomitant injuries and other comorbidities, the incidence of morbidity is only about 5.4%,9 and mortality is approximately 0.7-0.8%.1,9 Previously, fractures of the sternum were believed to be harbingers of underlying cardiac injury; however, multiple recent studies9-12 have proven otherwise. It is prudent to remember anecdotal cases of ventricular wall rupture associated with a sternal fracture.13 (See Figure 2.)
The initial diagnostic modalities for sternal fracture are posterior-anterior and lateral chest radiographs. Although some studies have reported better diagnostic yield with sternal ultrasound and bone scintigraphy,14,15 chest radiography still remains the initial diagnostic aid.
The extent of the evaluation and management should be guided by the severity of the injury and co-morbid factors. Most studies recommend a baseline electrocardiogram (ECG) and creatinine kinase (CK-MB),1,11,12 with a 2-D echocardiogram reserved for those patients whose clinical presentation warrants the diagnostic test. Although there are significant variations in institutional management of sternal fractures, the disposition of these patients should be individualized. Most patients with an isolated sternal fracture and no significant underlying medical problems may be discharged safely with appropriate analgesia.10-12
Thoracic Spine Injury. The thoracic spine supports the posterior segment of the thoracic cage. The first 10 vertebrae are fixed owing to their articulation with the thoracic cage.1 Thoracic spine fractures often result from a fall from a height, a motor vehicle collision (especially with ejection), or a motorcycle accident.16 Motorcycle accidents are the most common mechanisms because of the forced hyperflexion of the thoracic spine.17,18 Fractures of the thoracic spine usually result from an excessive force, and an anatomically narrow thoracic spinal canal leads to a high incidence of associated neurologic complications.1,2 Thoracic spinal fractures include anterior wedge compression, burst, chance, and fracture dislocation injuries. (See Figure 3.)
The initial evaluation of a patient with a possible spinal injury involves palpation of the spine for step off or subluxation, deformity, or midline tenderness. In most minor blunt chest trauma patients, an ordinary thoracic spine radiograph is adequate for evaluation. However, in severe multi-trauma patients with decreased levels of consciousness, severe alcohol intoxication, or respiratory distress requiring mechanical ventilation, it may be difficult to adequately assess the thoracic spine with routine radiographs. In these patients, CT of the spine has been shown to have better sensitivity, specificity, and negative and positive predictive values when compared to routine thoracic spine radiography.19-21 Hence, in severe multi-trauma patients, CT scan of the thoracic spine should be the diagnostic study of choice.
The initial therapeutic management of a potential thoracic spinal injury associated with chest trauma involves spinal immobilization until radiographic clearance. However, because of the high incidence of spinal cord involvement associated with thoracic spine injury, the early utility of steroids is very controversial. While some studies support the utilization of cortico-steroids,22,23 others do not consider it as a mandated standard of care, but rather as a treatment option.24,25 Currently, the National Acute Spinal Cord Injury Studies (NASCIS) advocates the early use of steroids.
Pneumothorax. Pneumothorax results when air enters the potential space between the parietal and visceral pleura. This injury can be caused by either penetrating or blunt trauma.2 The most common cause of pneumothorax in blunt trauma is a lung laceration with air leak.2 It is a common complication of chest trauma, and a recognizable cause of preventable death.26,27 About 6.7% occur without rib fractures, while the incidence increases with number of rib fractures.28
For the normal functioning of the pulmonary apparatus, the thorax is completely filled by the lungs, which are held to the chest wall by surface tension between the pleural surfaces. The presence of air in the pleural space leads to eventual collapse of the lungs. This collapse results in a ventilation/perfusion (V/Q) mismatch, because the blood perfusing the non-ventilated area is not oxygenated.2 This ultimately results in respiratory distress and hypoxia. Early in the course of a pneumothorax or a small pneumothorax, the patient may be asymptomatic. However, as the intrathoracic pressure increases because of air in the pleural space, a simple asymptomatic pneumothorax may become a life-threatening tension pneumothorax.
The diagnosis of pneumothorax often is accomplished by clinical examination and chest radiography. However, in a severely traumatized patient, supine chest radiography may miss a pneumothorax.29 Adjunctive diagnostic aids, such as ultrasonography and CT scan, have been shown to have both high sensitivity and specificity in diagnosing pneumothorax.29,30
The management of simple pneumothorax is expectant. However, when mechanical ventilation is indicated, prophylactic chest tube thoracostomy should be performed to prevent tension pneumothorax. In acutely dyspneic patients with hemodynamic instability after chest trauma, rapid needle decompression followed by tube thoracostomy is the standard of care.
Hemothorax. Hemothorax commonly results from lacerations to the lungs, intercostal vessels, or the internal mammary artery due to either blunt or penetrating trauma.1,2 Patients with large hemothoraces often are dyspneic, with some degree of respiratory distress because of restricted ventilation. (See Figure 4.)
The diagnosis of hemothorax should be made promptly. Although the majority of these injuries are diagnosed during the initial phase of assessment and management of the patient, there have been reported cases of delayed manifestation up to eight hours after initial presentation.31 The initial diagnostic method is the chest radiograph, and a hemothorax that is large enough to be apparent on chest radiograph should be evacuated. However, in a supine trauma patient, up to 1000 mL of blood could be missed on a radiograph.1 Studies showing the utility of ultrasound have produced mixed results.32,33 The use of CT scans has been shown to be the most sensitive and specific for diagnosing and accurately assessing hemothoraces in chest trauma patients.34-36 CT adequately quantifies the extent of the injury complex and the underlying complications.
The initial management of a hemothorax involves the insertion of a large-caliber chest tube for drainage, or open thoracotomy. In the majority of patients with hemothoraces, tube thoracostomy alone is adequate and is effective in more than 80% of the cases.37 In the presence of persistent bleeding, videothoracoscopic evaluation and treatment have been shown to be as effective as open throracotomy, and minimize the complications that accompany thoracotomy.37 The presence of more than 1500 mL of blood in the initial chest tube drainage, drainage of more than 200 mL an hour for 2-4 hours, or ongoing transfusion requirements mandate surgical exploration with open thoracotomy.2 (See Figure 5.)
Pulmonary Contusion. Contusion to the lung parenchyma is a significant cause of morbidity and mortality resulting from thoracic trauma. It is the most common cause of lethal chest injury.2 Usually a result of direct force to the chest wall, the injury results from a coup and countrecoup effect. The transmitted force produces direct damage to the lung parenchyma and associated hemorrhage and edema. In the majority of cases, respiratory failure develops over days rather than instantaneously.2
The pathophysiology of pulmonary contusion is poorly understood, and only minimal advances have been made in the past 20 years.38 The primary injury to the lung results in increased blood flow to the uninjured lung and parenchyma due to reflex decrease in pulmonary vascular resistance. This results in the extravasation of fluid into the alveoli and the interstitial spaces. With aggressive resuscitation, the level of the edema and blood in the lung parenchyma and interstitial spaces increases, thus producing a V/Q mismatch that is manifested as worsening hypoxia, hypercarbia, and acidosis.1
The diagnosis of pulmonary contusion is dependent on the extent of the lung parenchyma involved. Initial radiographic manifestations may be minimal or significant, depending on the individual patient. Even so, the chest radiograph should be the initial diagnostic aid, and for the majority of patients will show diffuse opacification of the involved lung parenchyma. (See Figure 6.) CT scan of the chest may be used in patients to adequately identify the exact lung segment involved, quantify the contusion volume, and act as a prognosticator of morbidity and mortality.39,40
The management of pulmonary contusion depends on the severity, associated injuries, and co-morbid conditions of the patient. The physiologic parameters that determine the severity of a pulmonary contusion in a chest trauma patient include: oxygen saturation less than 90% that is not responding to routine oxygen supplementation (nasal canula, bag-valve-mask, or facemask); PaO2 less than 65 mmHg; persistent tachypnea or FiO2/PaO2 less than 300.1,2,41
The primary management goal is to maintain adequate ventilation. The severity of the injury determines the modality of mechanical ventilation utilized. The need for intubation should be individualized, since the majority may be managed non-invasively.42 In a patient with a contused lung, the plateau airway pressure is increased, while static compliance and ETCO2 is decreased.43 Independent lung ventilation (ILV) now is a commonly used modality for managing pulmonary contusion, because it is effective in reducing V/Q mismatch, improving oxygenation with a setting of tidal volume (TV) and positive end-expiratory pressures (PEEP) to keep pulmonary plateau pressure below safe thresholds for barotrauma.42,43 Presently, the use of lung protective ventilatory strategy with low TV and high PEEP has become a standard practice.43,44 In most severe cases, placing the patient in a decubitus position (involved lung dependent position), use of nitrous oxide (because of its less dense characteristics), utility of pressure-control ventilation with paralysis, and the use of high-frequency oscillator have been shown to improve survival.
Blunt Myocardial Injury. Blunt myocardial injury (BMI) commonly is caused by high-speed motor vehicle collisions, although injuries resulting from low-velocity collisions have been reported.1 Due to different criteria used in diagnosing BMI, the incidence is difficult to determine. However, the prevalence ranges from 3-56%, depending on the study. The mechanisms of this injury include direct blows, athletic trauma, industrial crush, blasts, rapid deceleration, or falls from a height.1
A direct blow to the precordium (such as the chest wall striking the steering wheel in a motor vehicle collision, or the handlebar or a guardrail in a motorcycle collision, or being struck by a high-velocity object such as a baseball) creates a force that compresses the myocardium against the spine. (See Figure 7.) Because of the relatively free anterior-posterior movement of the heart, the momentum generated from a rapid deceleration accident maintains the heart in a uniform, straight-line motion, resulting in a direct strike against the internal sternum.1,2 BMI includes injuries such as cardiac contusion, cardiac concussion (commotio cordis), cardiac chamber rupture, or valvular disruption.
The manifestations of BMI originate from a direct injury to the myocardium that results in sub-endocardial hemorrhage, in turn leading to the formation of localized edema and the mobilization of the inflammatory response system. The resulting inflammatory changes cause a redistribution of coronary flow that may manifest as ischemic type chest pain.1 The severity of the presentation is contingent upon the underlying coronary artery disease, since transient coronary redistribution can produce a total or near-total occlusion of coronary vessels that already are stenosed. The worsening stenosis may lead to signs of acute coronary syndrome or acute myocardial infarction (AMI).
The clinical manifestations of this injury, in addition to ischemic or ischemic-like chest pain, include rhythm and conduction disturbances (e.g., multiple premature ventricular contractions and premature atrial contractions, atrial fibrillation, ventricular tachycardia, etc.), tachycardia out of proportion to volume loss or pain, right bundle-branch block (RBBB), and heart failure.1,2 Also, the clinician should be suspicious of BMI in the presence of increased central venous pressure (CVP) and the absence of obvious cause.
Presently, there is no gold standard for diagnosing cardiac contusion. The true diagnosis only can be ascertained through direct visualization of the injured myocardium. Currently utilized diagnostic aids include:
The management of BMI depends on the clinical manifestations. Although the majority of the cases are benign, some may present with AMI,53,54 persistent hypotension, or cardiogenic shock that warrants acute intervention. Recent literature proposes the use of ECG and TnI as the initial screening tools. If the initial ECG and TnI (four hours after the injury) are normal in the absence of concurrent injuries, the patient safely may be discharged from the ED.52 However, if the ECG or the TnI is abnormal, further workup is indicated. The patient should be admitted to a monitored unit with serial cardiac enzymes and a TEE to adequately evaluate the cardiac apparatus. In the absence of abnormalities on TEE and the normalization of cTnI, the patient should be reassessed, and the clinical parameters should determine the disposition.
Commotio Cordis. Commotio cordis, or cardiac concussion, is the most common cause of traumatic death in an athlete.55-57 It causes sudden or near cardiac death in the absence of structural abnormalities, and results from an object directly striking the chest wall at a phase of ventricular depolarization.56 Ventricular fibrillation is the most commonly induced arrhythmia. The survival rate is very low when the condition is not recognized. With the popularity of automatic electric defibrillators in public places, the survival from this phenomenon could be improved, since the ventricular fibrillation may respond to rapid defibrillation.
Esophageal Injury. Traumatic esophageal injury is very uncommon.2,57 When it occurs, injury to the esophagus most commonly results from penetrating trauma; however, it also can result from blunt injury to the lower thoracic cavity. Upper esophageal injury may accompany lower cervical or upper thoracic spine injuries, while distal injuries are rarely caused by blunt chest trauma.57,58 The mechanism of this injury is the forceful expulsion of gastric contents into the esophagus caused by a severe blow to the lower chest wall or upper abdomen. The resulting increase in intragastric pressure being transmitted to the esophagus results in a linear tear of the lower esophagus. This tear leads to the leakage of gastric contents into the mediastinum.2
The clinical presentation of the disease is similar to that of an esophageal tear secondary to persistent, profuse retching. In a severely traumatized patient, the diagnosis initially may be missed, and delayed diagnosis may lead to increased morbidity and mortality.
Esophageal rupture should be suspected in a patient with blunt trauma and one of the following features:
The diagnosis of esophageal perforation is very challenging. If suspected, a non-ionic contrast esophagogram should be performed.1 However, this study carries a high false-negative rate, and in patients with a high degree of suspicion for this injury, flexible esophagoscopy in conjunction with esophagogram may increase diagnostic yield.
The management of esophageal perforation is very controversial and is guided by the location of the perforation (hypopharynx, cervical, thoracic, or distal). While some authors propose surgical management for all perforations,59-63 some propose conservative management based on Cameron’s criteria (i.e., minimal signs of sepsis, disruption contained within the mediastinum, drainage of cavity back into the esophagus, and minimal symptoms).64 According to Rios et al, "Conservative treatment should be considered for patients meeting Cameron’s criteria, since their evolution is favorable with low morbidity and mortality."64
Despite technical and nutritional advances, the mortality rate for esophageal injuries ranges from 5-25% for those treated within 12 hours, to 25-66% for those treated after 24 hours.1 Regardless of the management approach, esophageal perforation is a life-threatening condition that requires early diagnosis and management to minimize the morbidity and mortality.
Diaphragmatic Injury. Blunt injury to the diaphragm is uncommon, occurring in about 0.8-8% of hospitalized chest trauma patients.65 The incidence of the laterality of this injury varies depending on the study. However, most studies report the incidence of left-side injury to be between 60-90%.1,66-68 This probably is because of the protection provided by the liver on the right and the possible left posterior lateral weakness of the diaphragm.1,2 Because of the significant contribution of the diaphragm to normal ventilation, injury to this structure may lead to significant respiratory compromise.
This injury often initially is missed unless the defect is large enough to cause acute herniation of the abdominal viscera into the thoracic cavity. In this instance, the chest radiograph will show the gastrium in the thoracic cavity or a coiled nasogastric tube in the thoracic cavity. Smaller defects may cause a delay in diagnosis until the abdominal viscera slowly migrates into the thoracic cavity, causing compression of the adjacent lung leading to either bowel strangulation or tension pneumothorax.1
Diaphragmatic injuries create a diagnostic dilemma. The chest radiograph is diagnostic in most cases of large ruptures; however, in the small defects, the injury often is missed or the chest radiograph is misinterpreted as showing an elevated diaphragm, acute gastric dilatation, a loculated pneumohemothorax, or a sub-pulmonary hematoma.2 The general consensus for the modality of choice for diagnosis is the helical CT; multiple studies have shown helical CT scan with sagittal and coronal reconstruction to be very sensitive in diagnosing diaphragmatic injury.65,69,70 In one study, sensitivity approaches 84%, with specificity of 77%.70 The most accurate modality for diagnosis is MRI. MRI is capable of directly acquiring both coronal and sagittal images, and allows the evaluation of the entire diaphragm, and shows the exact site and size of rupture in all cases as reported in one study.69 However, its usefulness is limited in the acutely traumatized patient. Its utility is beneficial in otherwise stable patients with blunt chest trauma with a high index of suspicion for isolated diaphragmatic injury. With technological advances in the field of thoracic surgery, videothora-scopic evaluation of the diaphragm in the hand of an experienced surgeon is emerging as a diagnostic modality.71,72
The management of diaphragmatic injury mostly is surgical. Most trauma patients with suspected diaphragmatic injury undergoing exploratory laparotomy for any intra-abdominal injury should have the diaphragm evaluated for possible defect. However, with the advances in laparoscopic techniques, thoracolaparoscopic repair is becoming the modality of choice for repair of an isolated diaphragmatic injury.71 Isolated diaphragmatic injury in the absence of other surgical injuries carries low mortality, and Bergeron et al showed that operative repair can be deferred without appreciable increase in mortality if no other indications mandate surgery.72
Blunt Aortic Injury. Blunt traumatic thoracic aortic injury (BAI) is a rare, but very lethal, condition. It often occurs as a result of a decelerating injury from high-speed motor vehicle collision (low-speed in older population) or a fall from a height.1 The incidence of BAI is about 6.8 per 100,000 motor vehicle occupants, with a steady increase with increasing age.73 Approximately 80-90% of the patients die at the scene,1,73 and up to 50% of the remaining patients die within 24 hours if not promptly diagnosed and treated.1 Of the survivors who make it to the hospital, the ultimate survival rate is lower for patients who are older than 60 years.73
Three mechanical factors contribute to aortic rupture. These factors include shearing stress, bending stress, and torsion stress.1 As deceleration occurs, a gradient is created between the mobile aortic arch and the immobile descending aorta. This gradient places the aortic isthmus under tension, and the resultant shearing stress can lead to rupture or tear opposite the fixation site.1 Bending stress results from the hyperflexion of the aortic arch produced by the downward traction exerted by the heart, and torsion stress results from anterior posterior compression of the chest, with the heart displaced to the left combined with an intravascular pressure wave transmitted to the aorta.1 When combined, these three forces produce maximum stress to the inner surface of the aorta at the ligamentum arteriosum, which is the point of greatest fixation.1
The clinical presentation of BAI is vague, since specific signs and symptoms often are absent.2 Most patients with free rupture, as elucidated earlier, die at the scene. However, patients with contained rupture who make it to the hospital may exhibit transient hypotension, dysphagia, hoarseness, or acute dyspnea.1,74 Although clinical factors (e.g., multiple rib fractures, flail chest, pulse deficits, or hoarseness without laryngeal injury) may be suggestive of BAI, about one-third of patients with this injury have no external signs of thoracic injury on initial physical examination.1
The diagnosis of BAI requires a high index of suspicion. A chest radiograph remains the most appropriate initial screening modality, with a negative predictive value approaching 95%.75 The supine chest x-ray may not show the classic findings; however, the presence of a widened mediastinum mandates further investigation. On chest x-ray, the presence of a widened mediastinum and a hemothorax in a patient with transient hypotension should increase the suspicion of aortic injury.74 (See Table 1.)
The radiographic algorithm utilized after the initial chest x-ray should be dictated by the patient’s clinical condition. Although aortography still is the recognized gold standard, fast spiral helical CT scan has emerged as a diagnostic study that potentially will supplant aortography.76 Multiple studies77-80 have shown spiral CT angiography to have a sensitivity of 96-100%, specificity of 98-100%, and accuracy of 99-100%. In the hemodynamically stable patient, contrast-enhanced helical CT has a critical role in the exclusion of BAI and prevents unnecessary thoracic aortography.80
The role of TEE in the evaluation of BAI has been well documented.75,77 Since the specificity and sensitivity of TEE are similar to those of helical CT, the indication for TEE is for the hemodynamically unstable patient with suspected BAI.75,77 Although TEE and CT have similar diagnostic accuracy, TEE allows for the diagnosis of associated cardiac injuries, and is more sensitive than CT for the identification of intimal or medial lesions of the thoracic aorta.77 (See Figure 8.)
While aortography still is the recognized gold standard for diagnosing BAI, in the case of equivocal aortographic findings, intravascular ultrasound (IVUS), with sensitivity approaching 92%, and specificity of 100%, could be used as an adjunctive diagnostic aid.81 The role of MRI in the evaluation of trauma patients remains indeterminate. Despite sensitivity and specificity approaching 100% for aortic injury, MRI’s utility in trauma patients is not feasible logistically because of the requirements of a metal-free environment and the long period of time that the patient must lie in isolation in a quiet room.1
The management of BAI typically is surgical following the initial resuscitation using the American College of Surgeons Advanced Trauma Life Support protocol. Emergent surgical intervention is the accepted standard of care. However, in some selected cases, such as patients deemed to be at high operative risk because of associated injuries or pre-existing medical conditions, or in stable patients for whom conditions for surgery are not ideal, delayed surgical intervention may be warranted.1 The surgical approach utilized is institutionally dependent.
Tracheobronchial Injury. Injury to the tracheobronchial area occurs very rarely and often is associated with other injuries.84 Although this injury potentially is fatal, it often is overlooked on initial assessment.2 The reported mortality for tracheobronchial injury (TBI) has fallen from 36% in the 1950s to less than 9% in the 1970s.85 The mechanism of this injury results from the effect of rapid deceleration on a relatively mobile bronchial structure and its fixed proximal segments.1 The majority of these injuries occur within 2 cm of the carina, or at the origin of the lobar bronchi.1,86
The common clinical presentations of TBI are signs of respiratory distress (dyspnea, stridor, or hemoptysis), subcutaneous emphysema, and sternal tenderness.1,87,88 The presence of pneumomediastinum, pneumothorax, widened mediastinum, or deep cervical emphysema on chest radiograph may suggest TBI.1,2,84,88
The diagnosis of TBI requires a high index of suspicion. The morbidity and mortality increases if not diagnosed early.87 The initial screening modality is the chest radiograph, which often demonstrates the signs suggestive of TBI (as described in the paragraph above). According to one study, the CT scan of the chest demonstrated similar findings to the chest radiograph, but failed to confirm the diagnosis.89 The presence of the "fallen lung" sign (a collapsed lung in a dependent position hanging on the hilum by its vascular attachment) on CT scan is highly suggestive of TBI.84 Tracheobronchoscopy is the definitive diagnostic modality of choice.1,2,84,88 On bronchoscopy, the injury pattern visualized is a transverse tear in the main bronchus involved or a disruption at the origin of an upper lung bronchus, while the trachea shows a vertical tear in the membranous portion near its attachment to the tracheal cartilages.1
Since most patients with TBI present with pneumothorax or tension pneumothorax, initial needle decompression with subsequent thoracostomy tube placement is required. The presence of persistent air leak with proper chest tube placement and drainage is highly suggestive of TBI until proven otherwise.1,2,89
The presence of persistent hypoxia despite intubation and chest tube placement mandates the use of temporizing opposite main stem bronchus intubation to provide adequate oxygenation,2 and also minimizes the effect of the ventilation and perfusion mismatch. In some instances, blind endotracheal intubation may be difficult, owing to anatomic distortion from pharyngeal injuries, paratracheal hematoma, or the TBI itself. For these patients, immediate operative intervention is required.2 However, in the more stable patients, acute surgical intervention could be delayed until inflammation and edema resolve.2,90 The definitive surgical treatment involve the reestablishment of the anatomic continuity of the tracheobronchial tree if the lesion affects more than one-third of the circumference.86
Finally, independent of mechanism and anatomic location of injury, delay in diagnosis is the single most important factor influencing outcome.91
Penetrating Chest Injury. Penetrating chest injuries (PCIs) are more common in urban medical centers. Most of these injuries involve firearms and knives. Injuries involving the cardiac, vascular, and transmediastinal structures are the most lethal, with prehospital mortality approaching 86% for cardiac injuries, and 92% for extrapericardial vasculatures.92 Of patients who make it to the hospital alive, only about 5-15% require thoracotomy.1 Of those who survive to the hospital, the mortality for those with cardiac injury is 21.9%, and 1.5% for those without cardiac injury.93 Survival rates from stab wounds generally are much higher than those from gunshot wounds.1,93
The injury pattern in PCI may include extensive lung laceration (See Figure 9), a sucking chest wound, or mediastinal traversing injuries. A sucking chest wound acts as a one-way valve that allows air to enter the pleural cavity during inspiration and none to leave during expiration. This eventually leads to an expanding or tension pneumothorax.1 In the prehospital setting, it is imperative that the wound be covered with occlusive dressings on only three sides. This allows air to escape the pleural cavity during expiration and, thus, prevents development of a tension pneumothorax. On arrival to the ED, the wound should be examined and covered completely with occlusive dressing, with simultaneous insertion of a chest tube at a site other than the initial injury location.1,2
Wounds that traverse the mediastinum may involve the great vessels, heart, tracheobronchial tree, or the esophagus.2 The overall mortality for these injuries approaches 20%.2 The evaluation of these injuries in hemodynamically stable patients can be performed non-operatively.94 Trauma ultrasound may be used for assessing pericardial tamponade, spiral CT for evaluating transmediastinal injuries, and organ-specific studies (e.g., esophagogram, aortography, bronchoscopy, thoracolaparoscopic evacuation of retained hemothorax, or repair of diaphragmatic injury) are minimally invasive management techniques for stable PCI patients.94
In hemodynamically unstable patients, there should be a high index of suspicion for exsanguinating thoracic hemorrhage, tension pneumothorax, or pericardial tamponade.1,2 In this situation, a bilateral chest tube thoracostomy is warranted to decompress possible hemopneumothorax and document volume of blood in chest tube drainage.2 The performance of ED thoracotomy is mainly for evacuation of pericardial blood, direct control of exsanguinating hemorrhage, open cardiac massage, or cross-clamping the descending aorta to slow blood loss below the diaphragm and increase perfusion to the brain and heart.1,2 With a bleak survival report for emergent ED thoracotomy, each facility should develop a uniform guideline for performance of this procedure. (See Trauma Reports 2003;4:1-12 for a thorough discussion of ED thoracotomy.) Recent studies95,96 show that the presence of signs of life on arrival to the hospital, in addition to the mechanism of injury and location of major injury, should be the determinants of the indications for emergent thoracotomy.
Thoracic cavity trauma carries significant morbidity and mortality because of the vital structures it involves. With technologic advances, most of these injuries now can be evaluated with minimally invasive diagnostic aids. The evaluation and management of injuries involving this cavity should be individualized, with special consideration for the pediatric and geriatric population.
1. Tintinalli JE, Gabor D, Saphcynski JS, eds. Emergency Medicine: A Comprehensive Study Guide. 5th Ed. New York: McGraw-Hill Professional; 2000:1675-1699.
2. American College of Surgeons Committee on Trauma. Advanced Trauma Life Support for Doctors (ATLS). 6th Ed. Chicago: American College of Surgeons; 1997.
3. Sirmali M, Turu H, Topcu S, et al. A comprehensive analysis of traumatic rib fractures: Morbidity, mortality, and management. Eur J Cardiothorac Surg 2003;24:133-138.
4. Bulge EM, Arenson MA, Mock CN, et al. Rib fractures in the elderly. J Trauma 2000;48:1040-1046.
5. Holcomb JB, McMullin NR, KozarAA, et al. Morbidity from rib fractures increase after age 45. J Am Coll Surg 2003;196:549-555.
6. Govindarajan R, Batalova T, Michael R, et al. Epidural buprenorphine in management of pain in multiple rib fractures. Acta Anaesthesiol Scand 2002;46:660-665.
7. Karmaker MK, Ho AM. Acute pain management of patients with multiple fractured ribs. J Trauma 2003;54:615-625.
8. Bergeron E, Lavoie A, Clas D, et al. Elderly trauma patients with rib fractures are at greater risk of death and pneumonia. J Trauma Inj, Infect, Crit Care 2003;54:478-485.
9. Potaris K, Gakidis J, Mihos P, et al. Management of sternal factures: 239 cases. Asian Cardiovasc Thorac Ann 2002;10:145-149.
10. Velissaris T, Tang AT, Patel A, et al. Traumatic sternal fractures: Outcome following admission to thoracic surgical unit. Injury 2003;34: 924-927.
11. Bar I, Friedman T, Rudis E, et al. Isolated sternal fractures: A benign condition. Isr Med Assoc J 2003;5:105-106.
12. Sadaba JR, Oswal D, Munsch CM, et al. Management of isolated sternal fractures: Determining the risk for blunt cardiac injury. Ann R Coll Surg Engl 2000;83:144-145.
13. Yilmaz EN, van Heek NT, van der Spoel JI, et al. Myocardial contusion as a result of isolated sternal fracture: A fact or a myth? Eur J Emerg Med 1999;6:293-295.
14. Erhan Y, Solak I, Kocabs S, et al. The evaluation of diagnostic accordance between plain radiography and bone scintigraphy for assessment of sternal and rib fractures in early periods of blunt trauma. Ulus Travma Derg 2001;7:242-245.
15. Engin G, Yekeler E, Guloglu R, et al. Ultrasound versus conventional radiography in the diagnosis of sternal fractures. Acta Radiol 2000;41: 296-299.
16. Hsu JM, Joseph T, Ellis AM, et al. Thoracolumbar fracture in blunt trauma patients: Guidelines for diagnosing and imaging. Injury 2003;34: 426-433.
17. Robertson A, Giannoudis PV, Branfoot T, et al. Spinal injuries in motorcycle crashes: Patterns and outcomes. J Trauma 2002;53:15-18.
18. Robertson A, Branfoot T, Barlow IF, et al. Spinal injury patterns resulting from car and motorcycle accidents. Spine 2002;27:2825-2830.
19. Sheridan R, Peralta R, Rhea J, et al. Reformatted visceral protocol helical computer topography scan allows conventioan radiograph of the thoracic and lumbar spine to be eliminated in the evaluation of blunt trauma patients. J Trauma 2003;55:665-669.
20. Hauser CJ, Visvikis G, Hinich SC, et al. Prospective validation of computer topography screening of the thoracolumbar spine in trauma. J Trauma 2003;55:228-234.
21. Wintermark M, Mouhsine E, Theumann N, et al. Thoracic and lumbar spine fracture in patient who has sustained a severe trauma: Depiction with multi-detector row CT. Radiology 2003;227:681-689.
22. Peter Vellman W, Hawkes AP, Lammertse DP, et al. Administration of corticosteroids for acute spinal cord injury: The current practice of trauma medical directors and emergency medical system physician advisors. Spine 2003;28:941-847.
23. Molloy S, Middleton F, Casey AT. Failure to administer methylprednisolone for acute spinal cord injury—A prospective audit of 100 patients from a regional spine unit. Injury 2002;33:575-578.
24. Hurlbert RJ. Methylprednisolone for acute spinal cord injury: An inappropriate standard of care. J Neurosurg 2000;93:1-7.
25. Hugenholtz H, Cass DE, Dvorak MF, et al. High-dose methylprednisone for acute closed spinal cord injury—Only a treatment option. Can J Neurol Sci 2002;29:227-235.
26. Di Bartolomeo S, Sanson G, Nardi G, et al. A population-based study of pneumothorax in a severely traumatized patient. J Trauma 2001;51: 677-682.
27. Tekinbas C, Eroglu A, Kurkcuoglu IC, et al. Chest trauma: Analysis of 592 cases. Ulus Travma Derg 2003;9:275-280.
28. Liman ST, Kuzucu A, Tastepe AI, et al. Chest injury due to trauma. Eur J Cardiothorac Surg 2003;23:374-378.
29. Rowan KR, Kirkpatrick AW, Liu D, et al. Traumatic pneumothorax detection with thoracic ultrasound: Correlation with chest radiograph and computer topography—Initial experience. Radiology 2002;225: 210-214.
30. Kirkpatrick AW, Ng AK, Dulchavsky SA, et al. Sonographic diagnosis of pneumothorax inapparent on plain radiograph: Confirmation by CT. J Trauma 2001;50:750-752.
31. Bundy W, Tilton DM. Delayed hemothorax after blunt trauma without rib fractures. Mil Med 2003;168:501-502.
32. Abboud PA, Kendall J. Emergency department ultrasound for hemothorax after blunt thoracic injury. J Emerg Med 2003;25):181-184.
33. Dulchavsky SA, Schwarz KL, Kirkpatrick AW, et al. Prospective evaluation of thoracic ultrasound in the detection of pneumothorax. J Trauma 2001;50:201-205.
34. Lomoschitz FM, Eisenhuber E, Linnau KF, et al. Imaging of chest trauma: Radiologic patterns of injury and diagnostic algorithm. Eur J Radiol 2003;48:61-70.
35. Gavelli G, Canini R, Beraccini P, et al. Traumatic injuries: Imaging of thoracic injuries. Eur Radiol 2002;12:1273-1294.
36. De Wever W, Bogaert J, Verschakelen J. Radiology of lung trauma. JBR-BTR 2000;83:167-173.
37. Ambrogi MC, Lucchi M, Din P. Videothoracoscopy for evaluation and treatment of hemothorax. J Cardiovasc Surg (Torino) 2002;43:109-112.
38. Miller PR, Croce MA, Bee TK, et al. ARDS after pulmonary contusion: Adequate measurement of contusion volume identifies high-risk patients. J Trauma 2001;51:223-230.
39. Christin F, Meyer N, Launoy A, et al. Lung contusion: Relevance of initial injured pulmonary volume measurement by CT. Ann Fr Anesth Reanim 2003;22:408-413.
40. Mizushima Y, Hiraide A, Shimazu T, et al. changes in contused lung volume and oxygenation in patients with pulmonary parenchymal injury after blunt chest trauma. Am J Emerg Med 2000;18:385-389.
41. Tyburski JG, Collinge JD, Wilson RF, et al. Pulmonary contusion: Quantifying the lesion on chest x-ray films and the factors affecting prognosis. J Trauma 1999;46:833-838.
42. Vidhani K, Kause J, Parr M. Should we follow ATLS guidelines for the management of traumatic pulmonary contusion: The role of non invasive ventilatory support. Resuscitation 2002;52:265-268.
43. Cinnella G, Dambrosio M, Brienza N, et al. Independent lung ventilation in patients with unilateral pulmonary contusion: Monitoring with compliance and ETCO2. Intensive Care Med 2001;27:1860-1867.
44. Offner PJ, Moore EE. Lung injury severity scoring in the era of lung protective mechanical ventilation: The PaO2/FiO2 ratio. J Trauma 2003;55:285-289.
45. Velmahos GC, Karaiskakis M, Salim A, et al. Normal electrocardiography and serum troponin I levels preclude the presence of clinically significant blunt cardiac injury. J Trauma 2003;54:45-51.
46. Mori F, Zuppiroli A, Ognibene A, et al. Cardiac contusion in blunt chest trauma: A combined study of tranesophageal echocardiography and cardiac troponin I determination. Ital Heart J 2001;2:222-227.
47. Chirillo F, Totis O, Cavarzerani A, et al. Usefulness of transthoracic and transesophageal echocardiography in recognition and management of cardiovascular injuries after blunt chest trauma. Heart 1996;75:301-306.
48. Descat E, Montaudon M, Latrabe V, et al. MR imaging of myocardial haematoma after blunt chest injury. Eur Radiol 2002;12:S174-176.
49. Bertinchant JP, Polge A, Mohty D, et al. Evaluation of incidence, clinical significance, and prognostic value of circulating cardiac troponin I and T elevation in hemodynamically stable patients with suspected myocardial contusion after blunt chest trauma. J Trauma 2000;48: 924-931.
50. Salim A, Velhamos GC, Jindal A, et al. Clinically significant blunt cardiac trauma: Role of serum troponin level combined with electrocardiographic findings. J Trauma 2001;50:237-243.
51. Sybrandy KC, Cramer MJ, Burgersdijk C. Diagnosing cardiac contusion: Old wisdom and new insights. Heart 2003;89:485-489.
52. Collins JN, Cole FJ, Weireter LJ, et al. The usefulness of serum troponin levels in evaluating cardiac injury. Am Surg 2001;67:821-825.
53. Tsoukas A, Andreades A, Zacharogrannis C, et al. Myocardial contusion presented as acute myocardial infarction after chest trauma. Echocardiography 2001;18:167-170.
54. Vasudevan AR, Kabinoff GS, Kelte TN, et al. Blunt cardiac trauma producing acute myocardial infarction in a rugby player. Lancet 2003; 362:370.
55. Wang JN, Tsai YC, Chen SL, et al. Dangerous impact—Commotio cordis. Cardiology 2000;93:124-126.
56. Perron AD, Brady WJ, Erling BF. Commotio cordis: An under-appreciated cause of sudden death in young patient: Assessment and management in the emergency department. Am J Emerg Med 2001;19:406-409.
57. Lateef F. Commotio cordis: An under appreciated cause of sudden death in athletes. Sports Med 2000;30:301-308.
58. Nerot C, Jeanneret B, Lardenois T, et al. Esophageal perforation after fracture of the cervical spine: Case report and review of literature. J Spinal Disord Tech 2002;15:513-518.
59. Cheynel N, Arral E, Peschaud F, et al. Perforation and rupture of the esophagus: Treatment and prognosis. Ann Chir 2003;128:163-166.
60. Port JL, Kent MS, Korst Rj, et al. Thoracic esophageal perforation: A decade of experience. Ann Thorac Surg 2003;75:1071-1074.
61. Uchikov AP, Safev GP, Nedev PI, et al. Perforation of the esophagus. Folia Med (Plovdiv) 2003;45:9-11.
62. Zumbro GL, Anstadt MP, Mawulawde K, et al. Surgical management of esophageal perforation: Role of esophageal conservation in delayed perforation. Am Surg 2002;68:36-40.
63. Dosios T, Safiolea M, Xipolitas N. Surgical treatment of esophageal perforation. Hepatogastroenterology 2003;50:1037-1040.
64. Rios ZA, Martinez de Haro LF, Ortiza EMA, et al. Esophageal perforation: Presentation of 23 cases. Gastroenterol Hepatol 2000;23:379-383.
65. Iochum S, Ludig T, Walter F, et al. Imaging of diaphragmatic injury: A diagnostic challenge. Radiographics 2002;22:S103-116.
66. Simpson J, Lobo DN, Shah AB, et al. Traumatic diaphragmatic rupture: Associated injuries and outcome. Ann R Coll Surg Engl 2000;82:97-100.
67. Mihos P, Potaris K, Gakidis J, et al. Traumatic rupture of the diaphragm: Experience with 65 cases. Injury 2003;34:169-172.
68. Patselas TN, Gakkagger EG. The diagnostic dilemma of diaphragm injury. Am Surg 2002;68:633-639.
69. Barbiera F, Nicastro N, Finazzo M, et al. The role of MRI in traumatic rupture of the diaphragm. Our experience in 3 cases and review of literature. Radiol Med (Torino) 2003;105:188-194.
70. Larici AR, Gotway MB, Litt HI, et al. Helical CT with sagittal and coronal reconstruction: Accuracy for detection of diaphragmatic injury. AJR Am J Roentgenol 2002;179:451-457.
71. Lamanto D, Poon PL, So JB, et al. Thoracolaparoscopic repair of traumatic diaphragmatic rupture. Surg Endosc 2001;15:323.
72. Bergeron E, Clas D, Ratte S, et al. Impact of deferred treatment of blunt diaphragmatic rupture: A 15-year experience in 6 trauma centers in Quebec. J Trauma 2002;52:633-640.
73. McGwin G Jr, Reiff DA, Moran SG, et al. Incidence and characteristics of motor vehicle collision-related blunt thoracic aortic injury according to age. J Trauma 2002;52:859-865.
74. Simon BJ, Leslie C. Factors predicting early in-hospital death in blunt thoracic aortic injury. J Trauma 2001;51:906-910.
75. Wintermark M, Wicky S, Schnyder P. Imaging of acute traumatic injury of the thoracic aorta. Eur Radiol 2002;12:431-442.
76. Britt LD, Campbell MM. Blunt thoracic aortic injury: Old problem and new technology. J Inj Term Eff Med Implants 2003;13:419-427.
77. Vignon P, Boncoeur MP, Francois B, et al. comparison of multilane transthoracic echocardiography and contrast enhanced helical CT in the diagnosis of blunt traumatic cardiovascular injury. Anesthesiology 2001;94:615-622.
78. Parker MS, Matheson TL, Rao AV, et al. Making the transition: The role of helical CT in the evaluation of potentially acute traumatic aortic injury. AJR Am J Roentgenol 2001;176:1267-1272.
79. Dyer DS, Moore EE, Ilke DN, et al. Thoracic aortic injury: How predictive is mechanism and is chest CT a reliable screening tool? A prospective study of 1561 patients. J Trauma 2000;48:673-682.
80. Scaglione M, Pinto A, Pinto F, et al. Role of contrast enhanced helical CT in evaluation of acute traumatic aortic injury after blunt chest trauma. Eur Radiol 2001;11:2444-2448.
81. Patel NH, Hahn D, Comess KA. Blunt chest trauma victims: Role of intravascular ultrasound and transesophageal echocardiography in cases of abnormal thoracic aortogram. J Trauma 2003;55:330-337.
82. Taylor PR, Bell RE, Reidy SF. Aortic transection due to blunt trauma: Evolving management using endovascular techniques. Int J Clin Pract 2003;57:652.
83. Fujikawa T, Yukioka T, Ishimaru S, et al. Endovascular stent grafting for treatment of blunt thoracic aortic injury. J Trauma 2001;50:223-229.
84. Wintermark M, Schnyder P, Wicky S. Blunt traumatic rupture of mainstem bronchus: Spiral CT-scan demonstration of the "fallen lung" sign. Eur Radiol 2001;11:409-411.
85. Kiser AC, O’Brien SM, Detterbeck FC. Blunt tracheobronchial injuries: Treatment and outcomes. Ann Thorac Surg 2001;71:2059-2065.
87. Balci AE, Eren N, Eren S, et al. Surgical treatment of post traumatic tracheobronchial injuries: 14 years experience. Eur J Cardiothorac Surg 2002;22:984-989.
88. Chen D, Han L, Hu Y, et al. Diagnosis and treatment of bronchial rupture from blunt thoracic trauma. Chin Med J (Engl) 2001;114:540-541.
89. Kunisch-Hoppe M, Hoppe M, Rauber K, et al. Tracheal rupture caused by blunt chest trauma: Radiological and clinical features. Eur Radiol 2000;10:480-483.
90. Triaridis S, Konstantinidis I, Soussios G, et al. A diagnostic challenge: Investigating suspected tracheobronchial and pharyngoesophageal tear. Are there gold standards? Acta Otorhinolaryngol Belg 2002;56: 331-334.
91. Cassada DC, Munyikwa MP, Moniz MP, et al. Acute injuries of the trachea and major bronchi: importance of early diagnosis. Ann Thorac Surg 2002;69:1563-1567.
92. von Oppel UO, Bautz P, De Groot M. Penetrating thoracic injuries: What we have learnt. Thorac Cardiovasc Surg 2000;48:55-61.
93. Mandal AK, Sanusi M. Penetrating chest wound: 24 year experience. World J Surg 200;25:1145-1149.
94. Demetriades D, Velmahos GC. Penetrating injuries of the chest: Indications for operation. Scan J Surg 2002;91:41-45.
95. Rhee PM, Acosta J, Bridgeman A, et al. Survival after emergency department thoracotomy: Review of published data for 25 years. J Am Coll Surg 2000;190:288-298.
96. Alanezi K, Milencoff GS, Baillie FG, et al. Outcome of major cardiac injuries at a Canadian trauma center. BMC Surg 2002;10:4.