Thoracic Trauma: Principles and Practice of Emergency Stabilization, Evaluation, and Management

Part II: Tracheobronchial Injury, Pneumothorax, and Systemic Complications

Authors: Elizabeth B. Jones, MD, Assistant Professor, Department of Emergency Medicine, University of Texas-Houston Medical Center; Kimberly Chambers, MD, Assistant Professor, Department of Emergency Medicine, University of Texas-Houston Medical Center; Luis Haro, MD, Assistant Professor, Department of Emergency Medicine, Mayo Clinic, Rochester, MN.

Peer Reviewers: Larry B. Mellick, MD, MS, FAAP, FACEP, Chair and Professor, Department of Emergency Medicine, Section Chief, Pediatric Emergency Medicine, Medical College of Georgia, Augusta; Steven Winograd, MD, FACEP, Attending Physician, Department of Emergency Medicine, Allegan General Hospital, Allegan, MI; Southwestern Michigan Emergency Services, PC.

Blunt and penetrating chest trauma leads to significant morbidity and mortality throughout the world. Many of the injuries that result from chest trauma were first identified decades ago, but the evolution of medical care in the United States has led to changes in the way these injuries are best diagnosed and treated. This paper, the second in a three-part series, focuses specifically on the non-cardiovascular injuries resulting from thoracic trauma and a review of the latest information about them in the medical literature. Included in the article are the following injuries: esophageal perforation, diaphragmatic rupture and laceration, tracheobronchial injuries, pulmonary contusion, flail chest, pneumothorax, hemothorax, air embolism, traumatic asphyxia, blast injuries, thoracic vertebral fractures, and other minor fractures.

The initial approach to the patient should be performed using Advanced Trauma Life Support (ATLS) guidelines. Management of the patient’s airway, breathing, and circulation is critical. In chest trauma, the primary survey should emphasize identification and treatment of tension pneumothorax and hemothorax. Eighty-five percent of penetrating chest wounds can be treated by tube thoracostomy and supportive measures. Wounds that enter or exit inferior to the nipple or the tip of the scapula may be associated with intra-abdominal injuries such as a perforated diaphragm or a liver or spleen laceration.

In chest trauma, management of circulation generally should follow ATLS protocols, except in one controversial area. In penetrating chest trauma, a landmark study in 1994 showed statistically significant benefit from delaying fluid resuscitation until bleeding was controlled via operative management.1 The decision to delay fluid resuscitation in penetrating chest trauma depends on protocols agreed upon by the departments of emergency medicine and surgery at individual institutions. (See Figure 1 by clicking here.)

After the initial stabilization of the patient, x-ray and other diagnostic studies can be performed. In major blunt trauma, a lateral cervical spine, chest x-ray, and pelvic x-ray routinely are performed. In penetrating thoracic trauma, a chest x-ray may be all that is needed.— The Editor

Esophageal Injuries

Injuries can result from blunt or penetrating trauma, barotrauma (from blast injury), or ingestion of caustic material. Penetrating trauma leads to immediate perforation in most cases, but high velocity missiles (such as those from military weapons) can lead to late necrosis and delayed perforation.2

The symptoms of esophageal injury are chest, throat, or neck pain; tenderness; dysphagia; and odynophagia. The signs include fever, choking, subcutaneous emphysema, redness, swelling, leukocytosis with left shift, and shock. Known injuries to the trachea or to the third and fourth thoracic vertebrae suggests possible esophageal injury, as these structures are quite close to the esophagus.

On physical exam, palpate the neck and supraclavicular areas for crepitus. Neck and chest x-ray may reveal air in the deep cervical tissue, air in the mediastinum, or bilateral pleural effusions. To diagnose esophageal injury, obtain an esophagram using Gastrografin swallow. If the suspected injury is not identified, this study should be followed by a barium swallow. Using this strategy, 62% of esophageal injuries will be identified.2 Endoscopy will demonstrate the injury in 50-100% of cases, but it is riskier than esophagography in this setting.2,3 Endoscopy actually may enlarge the perforation. Surgical exploration is a definitive diagnostic maneuver.

Surgical repair of the esophagus is the treatment of this injury. If the patient is a very poor surgical candidate and there is a contained leak, medical management can be attempted. The patient is treated with intravenous antibiotics, nasogastric suctioning, parenteral alimentation, and hemodynamic monitoring. Mortality is so high with medical management, however, that most patients should receive immediate surgical intervention. The decision to delay surgical management should be left to the consulting surgeon.

Prompt diagnosis of an esophageal tear is essential, as any delay in repair of the injury inevitably leads to higher mortality. If surgery is delayed 24 hours, mortality may be 50% or greater.12 Mediastinitis is the main complication responsible for this increased mortality.

Diaphragmatic Injuries

Blunt trauma causes 75% of diaphragmatic injuries, and penetrating trauma 25%. The injury is found in 0.8-1.6% of blunt trauma victims.4 Injuries to the diaphragm from blunt trauma tend to be larger than those due to penetrating trauma. Approximately 70% will be right-sided injuries, and 1.5% will be bilateral. A recent, prospective study of penetrating left thoracoabdominal injuries in which all patients had exploratory laparotomies found a 42% incidence of diaphragmatic injuries.5

Which signs and symptoms present diaphragmatic rupture depend on whether the bowel has herniated into the chest cavity. If there is no herniation through the rupture site, patients often are asymptomatic or experience mild pain at the time of injury. In one study, of the 45 patients with penetrating diaphragmatic injuries, 31% had no abdominal tenderness.5

If herniation has occurred, compression of the lung by herniated abdominal contents may cause dyspnea and tachypnea. If the bowel becomes strangulated, the patient will have signs of bowel obstruction (distended abdomen, colicky pain, hyperactive bowel sounds) and intestinal ischemia progressing to infarct (melena, shock). If the stomach has become incarcerated, a closed loop will be present that may cause retching without emesis and a flat upper abdomen.

If herniation does not occur at the time of injury, it may eventually develop—even if the initial tear in the diaphragm is small. In fact, the negative pressure in the pleural space will encourage herniation. Eighty-five percent of patients who go on to develop bowel strangulation from diaphragmatic injury will do so within three years.6 Some patients will present with incarceration even decades post injury.

In the interval before strangulation, the patients may have symptoms of intermittent incarceration of the bowel. Sudden, sharp pain can occur in the epigastrium, substernum, left upper quadrant, or left chest and may radiate to the shoulder. Pain may be initiated by eating and be relieved by vomiting or by standing or sitting up. Other symptoms are nausea, vomiting, dysphagia, constipation, melena, hiccups, or difficulty belching.4,7

Bowel loops or a nasogastric tube in the chest on chest x-ray are classic signs. Other x-ray findings are an irregular outline of the diaphragm, elevated hemidiaphragm, or atelectasis of the lower lobe.8 Diagnostic peritoneal lavage is helpful if positive, but has a false negative rate of up to 34%. Computed tomography (CT) scan will diagnose the injury initially if there is herniation of bowel contents into the chest. If there is rupture without herniation, CT scan is inconsistently reliable for diagnosis.7

In one study, 50% of patients with a penetrating diaphragmatic injury from left chest/abdominal wounds had a chest x-ray that demonstrated a chest hemopneumothorax, while 40% had a normal chest x-ray. Twenty-four percent of patients had no tenderness and negative radiographic findings such that the injury would not have been found other than by laparoscopy.5

Treatment includes surgical repair of the diaphragm. If initially repaired, morbidity and mortality are related to other injuries. If strangulation occurs, sequelae vary with the length of time of strangulation.

Tracheobronchial Injuries

An injury to the trachea or lower air passages can be subtle and may not be recognized. An injury of this type occurs in 1-2% of blunt chest trauma patients and 2-9% of those with penetrating trauma. Of all tracheal injuries, 75% affect the cervical trachea and 25% the thoracic trachea. These injuries may be caused by penetrating trauma or by the shearing forces of blunt trauma.

Symptoms include dyspnea, dysphonia, hoarseness, and hemoptysis. Signs are subcutaneous emphysema, hypoxia, air escaping the wound with respiration, and an unrelenting air leak from a chest tube that was placed for pneumothorax.

A chest x-ray is the study with which to begin investigation. A tracheal or bronchial injury should be suspected if there is pneumomediastinum on chest x-ray. Also, deep cervical emphysema seen on cervical spine films should raise suspicion of this injury. (See Figures 2 and 3: Click here.)

If the injury still is suspected and the patient is not suffering from significant respiratory compromise, a tracheogram may be performed to attempt to identify the injury. An alternative to the tracheogram (or as an adjunct if it fails to identify a suspected injury) is fiberoptic bronchoscopy. Not all injuries will be identified by tracheogram or bronchoscopy and surgical exploration may be necessary when the index of suspicion for this injury is high.

Respiratory distress and hypoxia clearly will prompt intubation, but it is important to realize that if there is tracheal disruption, extratracheal placement of the endotracheal tube can occur. There are case reports of endotracheal tube placement in the mediastinum causing a large air leak from the chest tube and not providing any oxygenation or ventilation. Whenever possible, intubation should be performed during flexible bronchoscopy, preferably in the controlled environment of the operating suite. Repair of these injuries is ultimately surgical.

If this injury is missed, there is a risk of tracheobronchial stricture, bronchopleural fistula, and death. Mortality in children is reported to be as high as 30%, and one-half of those who die will do so within an hour of the event.9 Seventy-five percent of patients with tracheobronchial injury die prior to presentation to the emergency department, and 15-25% die after reaching the hospital.10

Pulmonary Contusion

Pulmonary contusion is an important clinical entity and is the most common injury following major chest trauma. Pulmonary contusion in adults most commonly is seen in combination with other injuries such as rib fractures or flail chest, but it can be found without other injuries. In children, in whom the chest wall is quite pliable, there are often no rib fractures or outward signs of trauma.

Pulmonary contusion can arise from blunt or penetrating trauma or from a blast injury. It ranges from a small area of damage with local blood extravasation and lung tissue edema to extensive hemorrhage with alveolar rupture and diffuse lung edema. Lung compliance, which is normally 0.10-0.27 L/cm H20, decreases to 0.03-0.10 L/cm H2O on the second and third days after injury.11

Recently, it has been proposed that lacerations of the lung are responsible for alveolar hemorrhage and atelectasis. Wagner noted that 95% of patients studied with pulmonary contusion had at least one laceration that could be identified on CT scan.12 Normally, chest x-ray reveals patchy infiltrates but is not sensitive enough to pick up small lacerations that can be identified in the same patients on CT scan. The lacerations may be caused by lung rupture from compressive forces, shearing tears, or missile penetration.

Patients often present with symptoms of dyspnea and tachypnea. They may have hemoptysis, cyanosis, and hypotension. Auscultation of the chest may reveal fine rales or be normal. Patients should be maintained on continuous pulse oximetry as hypoxia may develop and worsen over time.

Chest x-ray is the screening exam of choice. On initial presentation, the chest x-ray may or may not show the contusion. The alveolar edema and other macroscopic changes develop over time, and the contusion can be seen on repeat chest x-ray 4-6 hours after the injury. CT scan is more sensitive, and pulmonary contusions often are identified on the initial CT scan.

Generally, patients are hospitalized for observation. They should have chest physical therapy, incentive spirometry, nasotracheal suctioning, analgesia, supplemental oxygen as needed, and repeat evaluations. Maximal hemorrhage usually has occurred by six hours; therefore, a repeat chest x-ray and clinical assessment at 8-10 hours post-injury will predict the patient’s disposition (discharge at the end of the observation period or full admission to the hospital.) In patients who remain hospitalized, repeat chest x-rays beyond 48 hours that reveal progression of the pulmonary contusion should prompt suspicion of other pathology such as aspiration or systemic inflammatory response syndrome (SIRS).

Richardson performed a selective management trial with 427 patients who had suffered major blunt chest trauma and who were diagnosed with flail chest (95 patients), pulmonary contusion (135 patients), pneumothorax, hemothorax, and/or multiple rib fractures. Patients were intubated for hypoxia, respiratory distress, pCO2 greater than 55 mmHg, airway compromise, and general anesthesia, or if intubation was part of the standard of care for other injuries. Patients were not intubated based on x-rays or paradoxical chest wall movement. Using these criteria, only 10 of 328 patients who were not initially intubated had to be intubated later in their hospital course.13

Allen provided guidelines for pediatric pulmonary contusion and reported that an alveolar to arterial gradient of 15 mmHg or higher or a Pa02/FI02 ratio of less than 250 on initial blood gas indicates respiratory insufficiency and the need for intubation. Frequent serial exams are critical in children who have pulmonary contusion as the injury will worsen initially. One-half of the children with pulmonary contusion will have respiratory failure over the first few hours.14

Those patients who do not die in the hospital will have normal p02 and lung compliance measurements at three weeks post-injury. Mortality rates for pulmonary contusion vary greatly, as many of the series reported included patients with severe associated injuries. In children, Allen reports that mortality is 10-20%. Richardson reported a 6.4% mortality in adult patients with pulmonary contusion who had to be intubated. However, only 1.4% of the patients died of respiratory causes.13

Flail Chest

Flail chest develops when 3-4 consecutive ribs are broken in two or more places. Functionally, a segment of the chest wall moves paradoxically during respiration and thus impairs ventilation. Patients will present with pain, dyspnea, and paradoxical motion of the chest wall. Respiratory distress may be present. Consecutive rib fractures may be diagnosed by chest x-ray, but a flail segment causing dysfunctional respiration is identified by direct visualization of chest wall motion.

Prior to 1956, patients were treated with stabilization of the chest wall. After that time, patients were managed with mechanical ventilation until 1975 when Trinkle suggested that not all patients with flail chest needed to be intubated.15 However, isolated flail chest requires mechanical ventilation twice as often as isolated pulmonary contusion.16

The selective-management strategy described earlier in the pulmonary contusion section also has been applied to patients with flail chest with the success noted above.13 Surgical intervention to stabilize the chest wall with hardware is clearly beyond the scope of the emergency department intervention but is undertaken at some centers and allows patients to be extubated sooner according to some studies.20,27

In a series of 144 trauma victims, patients with flail chest had a mortality of 16% (pulmonary contusion without flail chest in this series also had a mortality of 16%), while those with both flail chest and pulmonary contusion had a mortality rate of 42%. A recent retrospective study attempted to identify factors that affect outcome in 57 patients with flail chest. Factors that correlated with the need for mechanical ventilation were an injury severity score (ISS) greater than 22, blood transfusions in the first 24 hours, moderate to severe associated injuries (defined as fractures, head injury, or truncal organ injuries requiring operation), and shock on admission. Patients with an ISS of greater than 30, moderate or severe associated injuries, or the need for transfusion had adverse outcomes, including prolonged ventilatory assistance (> 2 weeks) and death from pneumonia and sepsis.17 (See Table 1.)

Table 1. Differential Diagnosis and Radiographic Findings
in Chest Trauma
Injury Symptoms / Signs X-ray Findings

Chest pain, throat pain, fever, crepitus Deep cervical emphysema, pneumomediastinum, pleural effusions
Dyspnea, dysphonia, hoarseness, hemoptysis crepitus, persistent air leak from chest tube Deep cervical emphysema, pneumomediastinum,  persistent pneumothorax after chest tube
Asymptomatic, chest or abdominal pain, vomiting/signs of obstruction Bowel gas or nasogastric tube in the chest, DPL fluid draining from chest tube, elevated or irregular contour of the diaphragm
Pulmonary contusion
Dyspnea, tachypnea, hemoptysis, hypoxia, rales on ausculation None or patchy infiltrates on initial film; infiltrates on repeat chest film
Flail chest
Paradoxical chest wall motion, chest pain, dyspnea, tachypnea, hypoxia 3-4 contiguous rib fractures in two or more places
Diminished breath sounds on affected side, tympanitic to percussion, dyspnea, tachypnea, hypoxia Pneumothorax
Tension pneumothorax
Signs of pneumothorax + jugular venous distension and/or hypotension, tracheal deviation away from the affected side Do not wait for a chest x-ray to treat this injury
Diminished breath sounds on affected side, dullness to percussion, dyspnea, tachypnea, hypotension Hemothorax
Traumatic asphyxia
Cervicofacial cyanosis, facial petechiae, subconjunctival hemorrhage, periorbital edema, disorientation None


Pneumothorax is the accumulation of air in the pleural space. It can be caused by blunt or penetrating trauma. There are different types of pneumothorax. In a simple pneumothorax, air enters the pleural space through a tear in the parietal pleura that acts as a one-way valve allowing air to enter, but not completely exit, the pleural space. As more air accumulates, the ipsilateral lung will collapse. If air continues to enter the pleural space, despite complete lung collapse, the non-compressible air will begin to compress mediastinal structures, causing a tension pneumothorax. As structures in the mediastinum are shifted away from the growing tension pneumothorax, there will be compression of the vascular beds as intrathoracic pressure exceeds venous pressure. When this occurs, signs and symptoms develop and rapid treatment must be performed to prevent cardiovascular collapse. In open pneumothorax, a defect in the chest wall allows air to be drawn into the pleural space during respiration. If the hole in the chest wall is two-thirds the size of the trachea, air will enter through the chest wall instead of the trachea with inspiration. In some cases, the defect in the chest wall will form a one-way valve and a tension pneumothorax will develop.

Dyspnea and tachypnea will be frequent presenting features. Crepitus of the soft tissue of the chest wall can be palpated in some cases. Patients may or may not exhibit hypoxia on pulse oximetry or blood gases depending on the size of the pneumothorax. Auscultation will reveal diminished breath sounds on the side of the pneumothorax and hyper-resonance to percussion. Distended neck veins and hypotension, along with diminished breath sounds, are indicative of a tension pneumothorax, which should be treated prior to any other evaluation or intervention.

Tension pneumothorax is identified by clinical exam. With treatment, neck vein distention and hypotension should resolve. If they do not resolve, hemopericardium with resulting cardiac tamponade should be considered.

In a prospective study, Chen sought to determine the accuracy of auscultation in detecting hemopneumothorax. In 148 patients who had isolated hemothoraces or pneumothoraces, or combined hemopneumothoraces, ausculation had a sensitivity of 84%, a specificity of 97%, an accuracy of 89% and a positive predictive value of 97% for the detection of the injuries. Chest x-ray was more accurate than auscultation. The authors recommend that therapy be instituted if auscultation predicts an injury, but if no injury is detected, that chest x-ray be performed to rule out an injury missed by auscultation.18

Remember that in an intubated patient with decreased breath sounds on the left side, the cause may be right mainstem intubation rather than a hemo-/pneumothorax. If other signs of these injuries do not exist, a chest x-ray can be performed prior to therapy to confirm the diagnosis and the location of the distal end of the endotracheal tube. An upright expiratory chest x-ray, although not practical in critically injured patients, is the best technique to confirm a pneumothorax. With the patient upright, the air in the pleural space will rise to the apical area and exhalation will make the pneumothorax a larger proportion of the total intrathoracic volume. As a result, the upright expiratory view makes the pneumothorax easier to identify.

The majority of pneumothoraces will be treated with a chest tube alone. Only 10-20% will require a thoracotomy. The chest tube is placed for evacuation of the air in the pleural space. In the case of a tension pneumothorax, a large bore needle should be introduced into the second intercostal space in the midclavicular line to decompress the pleural space while preparations to perform chest tube thoracostomy are made. Needle decompression will convert a tension pneumothorax into an open pneumothorax.

In the case of penetrating trauma to the chest wall, the wound site should not be used as the track for the chest tube insertion. The wound site may lead directly into the lung parenchyma or other undesirable location for a chest tube. Chest tube should generally be placed at the fifth intercostal space, anterior to the axillary line.

There is controversy regarding whether or not prophylactic antibiotics to prevent infectious complications should be instituted when tube thoracostomy is being performed. The reported incidence of empyema ranges from 1.6%-26% after tube thoracostomy for traumatic pneumo- or hemothorax.19

Nonrandomized, retrospective studies published in recent years offer guidelines for decreasing infections, but do not recommend antibiotics. Mandal recommends antibiotics for emergent or urgent thoracotomy, destruction of the chest wall from shotgun blasts, lung contusion with hemoptysis, exploratory laparotomy for abdominal trauma, or open long bone fracture. Using this protocol in 5474 patients, only 1.6% developed empyema following their protocol for antibiotic therapy.20 Additionally, a retrospective review identified retained hemothorax, pulmonary contusion, and multiple chest tubes as risk factors for empyema. Administration of antibiotics did not decrease the incidence of empyema in the series.19

A review article that included data from seven randomized trials comparing the use of antibiotics to placebo showed that the incidence of empyema/infection in the study groups was 4.5%, whereas the control groups had a 20.9% rate of infectious complications.21

A double-blind, randomized clinical trial of 139 patients published in 1998 found that there were no infectious complications (empyema or pneumonia with effusion) in the treatment group. The rate of infectious complication in the placebo group was 5.8%. The treatment group received 1 gram of cefazolin intravenously every eight hours starting just prior to chest tube insertion. The control group received placebo on the same schedule.11

In the above study, no infectious complications occurred in patients with hemothoraces of more than 500 cc. This is interesting because several studies identify retained hemothorax as an important nidus for infection which can result in empyema development,20-22 and a number of centers are performing video-aided drainage of retained or recurrent hemorrhage to prevent infection.

All the publications reviewed identify Staphylococcus aureus as the most common pathogen. Gram-negative bacteria, anaerobes, and mixed organisms also can lead to infection complications.

The majority of pneumothoraces, if recognized and treated with a chest tube, need no further therapy. The average hospital stay is 5-6 days.23 Complications from infection, such as empyema, require drainage and often decortication and increase morbidity and mortality and hospital stay. Persistent air leaks suggest a tracheal or bronchial injury.

The Aerospace Medicine Association recommends that patients wait 2-3 weeks after radiographic resolution of pneumothorax before traveling by air. A prospective study of 12 patients found, by chest x-ray, that those who traveled by air at least 14 days after pneumothorax had resolved, and that they had no symptoms during air travel. Of the two patients who flew prior to the 14 days, one experienced symptoms of recurrent pneumothorax.24


Blunt or penetrating trauma can cause a hemothorax by laceration of the lung, intercostal vessels, mammary arteries, or great vessels. The incidence is between 15% and 30% in chest trauma. Most hemothoraces are caused by lung laceration and are self-limited since there is a high content of thromboplastin in the lung and the pulmonary venous pressure is low.

Dyspnea, tachypnea, hypovolemia, decreased breath sounds, and dullness to percussion on the affected side may be present with this injury.

A chest x-ray is the screening exam of choice. A chest x-ray should not, however, be used to predict the volume of blood in the chest cavity. A recent study showed that experienced surgeons and radiologists were unable to accurately predict traumatic hemothorax volume using chest x-ray. The study also showed that CT scan can accurately predict the volume present as confirmed by drainage of the hemothorax.25

Hemothorax is managed with a #38 French chest tube placed at the nipple level in the anterior axillary line. If the hemothorax is massive (> than 1500 mL), the chest tube drainage should be autotransfused back into the patient and early thoracotomy should be strongly considered. Continuing output from the chest tube of greater than 200 mL of blood per hour for the following 2-4 hours increases the likelihood that the patient will need a thoracotomy by the surgical team to achieve hemostasis. The decision to operate will be made by the surgical team based on the patient’s hemodynamic status. Only between 5% and 15% of hemothoraces will require thoracotomy.

The mortality varies depending on the etiology of the bleeding and concomitant injuries. Mortality rate for hemothorax in penetrating trauma is roughly 20%. The rate is almost double that if the hemothorax is from blunt trauma.12

Air Embolism

Air embolism occurs when air or gas enters the pulmonary venous system, and travels to the heart and on to the coronary arteries and/or cerebral vessels. Factors in thoracic trauma that predispose patients to this are low pulmonary venous pressure (as in hypovolemia) and increased airway pressure (as in tension pneumothorax or positive pressure ventilation). As little as two milliliters of air in the cerebral circulation can be fatal, and half a milliliter in a coronary artery can lead to cardiac ischemia and secondary cardiac arrest.

The prevalence of air embolism after major thoracic trauma is estimated to be between 4% and 14%. It occurs twice as often in penetrating trauma as it does in blunt trauma.26,27

Cardiac arrest or change in mental status directly after intubation is a typical presentation. The differential includes tension pneumothorax, pericardial tamponade, hypovolemia, or cardiac arrhythmia. Air in the retinal vessels can be seen in some cases. Precordial doppler can show air in the right atria. Transesophageal echocardiography (TEE) can identify venous air emboli as well as systemic air embolism. CT scan will identify intracerebral air. (See Figure 4: Click here.)

If the patient arrests directly after intubation from air embolism, mortality is 100%. If TEE identifies air in the right heart or if air embolism is suspected, the patient should be placed in the left lateral decubitus position. If the source of embolism is identified, unilateral lung ventilation or surgical clamping of the ipsilateral hilum may be performed. Cerebral air embolism is treated with hyperbaric oxygen therapy.

Clearly, preventing air embolism is preferable to treatment. Saada used high-frequency, low-volume ventilation in three patients with systemic air embolism and, using TEE, documented a cessation of embolization after the switch to this ventilation method.21 Investigating changes in ventilation in the acute phase of resuscitation of major thoracic trauma is an area in which further study is needed.

Traumatic Asphyxia

Traumatic asphyxia is a rare syndrome that is produced by a sudden compressive force which crushes the chest. It is most often caused by a motor vehicle running over the patient’s chest but industrial accidents and falls also have been associated with this injury pattern.

Inspiration and a closed glottis, when combined with tensing of the chest and abdominal muscles during the crush injury, lead to increased intrathoracic pressure and compression of the superior vena cava causing cyanosis. Increased pressure in superficial capillaries leads to the development of petechiae.

The hallmark of this injury is cervicofacial cyanosis and/or facial petechiae. The cyanosis and petechiae may extend to the upper chest, but rarely reaches the elbows. Patients often are agitated, restless, and disoriented. Subconjunctival hemorrhage, periorbital edema, and ecchymosis also are seen.

Following traumatic asphyxia, look for associated injuries such as pulmonary contusion, flail chest, ruptured diaphragm, head injuries, and particularly in children, liver and splenic lacerations. In several recent cases in the literature, cardiovascular injuries were not reported, but patients who died on the scene or presented dead on arrival to the emergency department were not included and it may be that cardiovascular injuries were immediately catastrophic.11

Treatment includes airway management and admission for monitoring and treatment of associated injuries. Patients with traumatic asphyxia and no associated injuries who present to the emergency department should be expected to have a full recovery. The neurologic symptoms generally resolve in 24-48 hours. The petechiae will resolve in 1-3 weeks. The prognosis for the patient is determined by the concomitant injuries.

Blast Injuries

Explosions create blast energy that acts on the body as a blunt force. Distance from the blast determines how much energy will be transferred to the lung through the chest wall. Energy from the blast disperses at places where there are tissues of different densities. Tissues that contain plasma or hollow organs that contain gas are particularly susceptible to injury.2,28

Hemoptysis, dyspnea, rales and rhonchi on auscultation, chest pain, hypoxemia, and cyanosis are seen with blast injury to the lungs.29 There often is no evidence of trauma to the chest wall such as soft-tissue contusion or rib fracture. The clinician must look carefully for signs of respiratory distress and hypoxia.

Patients also may present with abdominal pain and symptoms of intestinal perforation or neurologic symptoms such as paralysis. Bradycardia and profound shock also can occur. Randomized animal trials suggest the bradycardia and hypotension are vagally mediated and, thus, that vagolytic therapy may be helpful.30

The tympanic membranes are more susceptible to barotrauma and blast injury than are the lungs or gastrointestinal tract. If the patient does not have perforation of either tympanic membrane, significant injury to other air-containing structures is unlikely.2

Chest x-ray will be abnormal on patient presentation if signs and symptoms of injury are present. Possible x-ray findings include pulmonary edema, pneumothorax, pneumomediastinum, hemothorax, and pulmonary contusion.

Treatment of pulmonary compromise after this injury is supportive and the same principles apply as are mentioned in the review of the underlying pathology. The extent of injury to the lung is directly proportional to the proximity to the blast. Morbidity and mortality depend on the extent of lung damage and on the presence of associated injuries.

Thoracic Vertebral Fractures

Injuries of the thoracic vertebrae occur with compressive, rotational, or shearing forces, as well as from flexion, extension, and distraction. Fractures of the thoracolumbar spine occur in roughly 9% of patients with multiple trauma, and approximately 35% of these will have a neurologic deficit.31

The majority of patients with thoracic spine injury will have back pain and point tenderness over the thoracic spinous processes, but not all patients will have clinical evidence to support the need for radiographs. A retrospective study of 181 trauma patients with major thoracic and lumbar fractures found that 13 of the patients (who all had a Glasgow Coma Scale (GCS) score of 15) had no back pain or tenderness to prompt diagnostic studies. Of these patients, the majority had other fractures and several had closed head injuries. Intoxication, if present, was not noted. The mechanisms were: five falls, four motor vehicle collisions, two motorcycle collisions, and two pedestrians hit by motor vehicles.32

Another retrospective review of 145 cases of thoracic and lumbar spinal fractures revealed that 81% had back pain or tenderness. The presence of the clinical clues were significantly higher in those without an altered sensorium or other major injury. Conversely, if the patients had clear sensorium, no distracting injuries, and no clinical evidence of injury (back pain or tenderness), there were no thoracolumbar fractures.33

A reasonable approach to follow when looking for thoracic vertebral fractures seems to mirror the standard approach to cervical spine evaluation. Clearly, if there is pain or tenderness, films should be obtained. Additionally, if the patient has altered sensorium from any cause, has a painful injury elsewhere, or has any neurologic signs or symptoms suggesting injury, thoracic spine radiographs should be ordered.

Several papers report that the incidence of thoracolumbar fracture is 7-10% in patients with cervical fractures. Therefore, if a cervical fracture is found, a survey of the rest of vertebral column seems appropriate.31-33

Despite the close proximity of the aorta to the thoracic vertebrae, a 1997 study revealed that the incidence of traumatic aortic injuries is not increased with thoracic vertebral fracture.34

During evaluation for thoracic spine injury, physicians should keep the patient’s spine immobilized. If an injury is found, the physician should continue immobilization and consult the spine service. The spine service will determine whether surgical intervention or conservative management with a brace is warranted. If there is a neurologic deficit, start high-dose steroid therapy immediately—even if the spine has not yet been imaged.

Neurologic impairment will occur in up to 50% of patients with bony thoracic spine injury. Only 11% of patients with complete spinal cord injuries regain useful motor function.35 Patients without deficit are, in some cases, being treated only with orthoses and physical therapy with good success.36

Other Fractures

Clavicle fractures are common. Standard therapy has been analgesia and figure-eight brace for treatment. The figure-eight brace is cumbersome for patients to put on and often can be uncomfortable since one of the straps may lie directly against the fracture site. As a result, there has been a noncompliance problem among patients given the brace. This has led many clinicians to give patients an arm sling rather than the figure-eight brace. There are no prospective studies comparing the outcomes of patients using the brace vs. the sling. One advantage of the figure-eight brace is that it allows mobility at the shoulder and elbow joints, and patients who are placed in arm slings should be advised via verbal and written instructions to perform range of motion exercises at each of these joints.

The overall prognosis for clavicle fractures is good. A 17-year follow-up of 225 patients with clavicle fractures revealed that 185 had no symptoms, 39 still had some mild pain, and one complained of severe pain. Two patients required surgery for neuropathy. Radiographically, 125 had normal healing, 53 had malunion with persistent displacement, and seven had nonunion. Nonunion more commonly was found in the group of patients whose original fractures had been displaced.37

Scapular fractures were seen in 2.9% of blunt thoracic trauma at one high-volume center.16 High energy is required to cause scapular fractures. More than one-half of the patients with scapular fracture will have pulmonary contusions, and 11% will have an arterial injury (subclavian, axillary, or brachial).16

High energy transfer to the sternum causes sternal fractures, and these fractures traditionally were thought to be associated with a higher incidence of mediastinal injury, especially cardiac contusion. Chiu performed a prospective study of 33 patients (all with GCS score of 15) with sternal fractures and no other life-threatening injuries identified. These blunt trauma patients presented with chest pain that was often pleuritic in nature. The diagnosis of sternal fracture was made by chest x-ray. Patients were evaluated with electrocardiogram (ECG), echocardiogram, and serial cardiac enzymes and were continuously monitored. None of the patients developed ECG changes, cardiac wall motion abnormalities, elevated cardiac enzymes, or arrhythmias. Based on these data, the authors recommend an ECG at presentation, six hours of cardiac monitoring, and repeat ECG at six hours. If there are neither ECG changes nor arrhythmias and no other symptoms develop, the patient may be discharged with analgesics and instructed to follow up.38

In the above study, the rate of aortic rupture in patients without sternal fracture was 3.6%, while in those with fracture it was 2%. If there is widening of the mediastinum suggesting a mediastinal hematoma, 12-18% will have a rupture of the aorta. All mediastinal widening warrants evaluation.39


1. Bickell WH, Wall MJ Jr., Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med 1994;331:1,105-1,109.

2. Bastos RBN, Graeber, GM. Esophageal injuries. Chest Surg Clin N Am 1997;7:357-371.

3. Attar S, Hankins JR, Suter CM, et al: Esophageal perforation: A therapeutic challenge. Ann Thorac Surg 1991;50:45-51.

4. Mansour KA. Trauma to the diaphragm. Chest Surg Clin N Am 1997;7:373-383.

5. Murray JA, Demetriades D, Arsenio JA, et al. Penetrating left thoracoabdominal trauma: The incidence and clinical presentation of diaphragm injuries. J Trauma 1997;43:624-626.

6. Shah R, Sabanathan S, Mearns AJ, et al. Traumatic rupture of the diaphragm. Ann Thorac Surg 1995;60:1,444-1,449.

7. Reber PU, Schmied B, Seiler CA, et al. Missed diaphragmatic injuries and their long-term sequelae. J Trauma 1998;44: 183-188.

8. Singh N, Narasimhan KL, Rao KL, et al. Bronchial disruption after blunt trauma chest. J Trauma 1999;46:962-964.

9. Grant WJ, Meyers RL, Jaffe RL, et al. Tracheobronchial injuries after blunt chest trauma in children—Hidden pathology. J Ped Surg 1998;33:1,707-1,711.

10. Lee RB. Traumatic injury of the cervicothoracic trachea and major bronchi. Chest Surg Clin N Am 1997;7:285-303.

11. Garzon AA, Seltzer B, Karlson KE. Physiopathology of crushed chest injuries. Ann Surg 1968;168:128-136.

12. Wilson RF. Thoracic Trauma. In: Tintinalli JE et al, eds. Emergency Medicine—A Comprehensive Study Guide. 4th ed. New York: McGraw Hill; 1996:1,156-1,182.

13. Richardson JD, Adams L, Flint LM. Selective management of flail chest and pulmonary contusion. Ann Surg 1982;196(4):481-487.

14. Allen GS, Cox CS Jr. Pulmonary contusion in children: Diagnosis and management. South Med J 1998; 91:1,099-1,106.

15. Trinkle JK, Richardson JD, Franz JL, et al. Management of flail chest without mechanical ventilation. Ann Thorac Surg 1975;19: 355-363.

16. Mayberry JC, Trunkey DD. The fractured rib in chest wall trauma. Chest Surg Clin N Am 1997;7:239-260.

17. Freedland M, Wilson RF, Bender JS, et al. The management of flail chest injury: Factors affecting outcome. J Trauma 1990;30: 1,460-1,468.

18. Chen SC, Chang KJ, Hsu CY. Accuracy of auscultation in the detection of haemopneumothorax. Eur J Surg 1998;164:643-645.

19. Aguilar MM, Battistella FD, Owings JT, et al. Posttraumatic empyema. Risk factor analysis. Arch Surg 1997;132:647-651.

20. Mandal AK, Thadepalli H, Mandal AK, et al. Posttraumatic empyema thoracis: A 24-year experience at a major trauma center. J Trauma 1997;43:764-771.

21. Richardson JD, Carrillo E. Thoracic infection after trauma. Chest Surg Clin N Am 1997;7:401-426.

22. Eddy AC, Luna GK, Compass M. Empyema thoracis in patients undergoing emergent closed tube thoracostomy for thoracic trauma. Am J Surg 1989;157:494-497.

23. Gonzalez RP, Holevar MR. Role of prophylactic antibiotics for tube thoracostomy in chest trauma. Am Surg 1998;64:617-620.

24. Cheatham ML, Safcsak K. Air travel following traumatic pneumothorax: when is it safe? Am Surg 1999;65:1,160-1,164.

25. Velmahos GC, Demetriades D, Chan L, et al. Predicting the need for thorascopic evacuation of residual traumatic hemothorax: Chest radiograph is insufficient. J Trauma 1999;46:65-70.

26. Stene JK, Grande CM, Bernhard WN, et al. Perioperative anesthetic management of the trauma patient: Thoracoabdominal and orthopaedic injuries. Trauma Anesthesia 1991;219.

27. Trunkey D. Initial treatment of patients with extensive trauma. N Engl J Med 1991;324:1,259-1,263.

28. Boyd AD, Glassman LR. Trauma to the Lung. Chest Surg Clin N Am 1997;7:263-284.

29. Huller T, Bazini Y. Blast injuries of the chest and abdomen. Arch Surg 1970;100:24-30.

30. Irwin RJ, Lerner MR, Bealer JF, et al. Shock after blast wave injury is caused by a vagally mediated reflex. J Trauma 1999;47: 105-110.

31. Meek S. Fractures of the thoracolumbar spine in major trauma patients. BMJ 1998;317:1,442-1,443.

32. Anderson S, Biros MH, Reardon RF. Delayed diagnosis of thoracolumbar fractures in multiple-trauma patients. Acad Emerg Med 1996;3:832-839.

33. Meldon SW, Moettus LN. Thoracolumbar spine fractures: Clinical presentation and the effect of altered sensorium and major injury. J Trauma 1995;39:1,110-1,114.

34. Lee J, Harris JH Jr., Duke JH Jr., et al. Noncorrelation between thoracic skeletal injuries and acute traumatic aortic tear. J Trauma 1997;43:400-404.

35. McCullen G, Vaccaro AR, Garfin SR. Thoracic and lumbar trauma. Rationale for selecting the appropriate fusion technique. Orthop Clin North Am 1998;29:813-822.

36. Melchiorre PJ. Acute hospitalization and discharge outcome of neurologically intact trauma patients sustaining thoracolumbar vertebral fractures and managed conservatively with thoracolumbosacral orthoses and physical therapy. Arch Phys Med Rehabil 1999;80: 221-224.

37. Nordqvist A, Petersson CJ, Redlund-Johnell I. Mid-Clavicle fracture in adults: End result study after conservative treatment. J Orthop Trauma 1998;12:572-576.

38. Chiu WC, D’Amelio LF, Hammond JS. Sternal fractures in blunt chest trauma: A practical algorithm for management. Am J Emerg Med 1997;15:252-255.

39. Crestanello JA, Samuels LE, Kaufman MS, et al. Sternal fracture with mediastinal hematoma: Delayed cardiopulmonary sequelae. J Trauma 1999;47:161-164