Thoracic Trauma
Introduction
Among trauma victims, thoracic trauma is the second most common cause of death in the field and the third most common cause of death in patients who make it to the hospital. In civilian trauma, the most common mechanism is blunt trauma and the most frequently encountered injuries (i.e., hemothorax and pneumothorax) are seen in 20% of patients. The majority require only tube thoracostomy for definitive treatment; however, up to 15% of these patients will require operative intervention, whether immediate (resuscitative), urgent, or delayed. Overall mortality in patients with thoracic trauma is about 10%, with a low Glasgow Coma Scale score exerting the strongest influence on mortality. Additional predictors of poor outcomes include increasing age, mechanism of blunt injury, increasing number of ribs fractured, and concomitant long bone extremity fractures.
Indications for Resuscitative Thoracotomy
Patients who arrive without signs of life may benefit from emergency department thoracotomy (EDT) or “resuscitative” thoracotomy depending on the amount of time elapsed since the loss of vital signs (<15 minutes for penetrating trauma; <10 minutes for blunt trauma). The goal of this procedure is to restore spontaneous circulation by rapid repair of intrathoracic injuries or release of pericardial tamponade, and occlusion of the descending thoracic aorta to divert perfusion to the brain and heart. Internal cardiac massage with injection of intracardiac medications may also be indicated. In cases of bronchovenous air embolism, clamping of the pulmonary hilum prevents further propagation. Because resuscitative thoracotomy should always proceed to the operating room for definitive repair, it should not be attempted without a surgeon immediately available. Rhee and colleagues reported that the overall survival rate after EDT is 7.4%, with a survival rate of 15% among all patients with penetrating injuries and 35% among patients with penetrating cardiac injuries. On the contrary, survival rates are very low (2%) after blunt traumatic arrest.
It is important to perform an initial cardiac ultrasound, because EDT can be considered futile if asystole is the presenting rhythm and pericardial tamponade is absent. In patients with appropriate indications, aggressive efforts are justified, because functional long-term outcomes in survivors are excellent. More than half of survivors are discharged home from the hospital, and more than 75% have normal cognition, are ambulatory, and have no evidence of posttraumatic stress disorder. However, inappropriate EDT results in no survival benefit, wasted use of resources, and exposure of health care workers to needle-stick and bone fragment injuries.
Indications for Urgent Thoracotomy
Commonly accepted indications for urgent thoracotomy include any of the following: initial chest drain output of 1500 mL; greater than 1500 mL output in the first 24 hours after injury; more than 200 mL bloody output per hour for 3 consecutive hours; massive air leak (present during all phases of respiration, associated with inability to reexpand the lung or affecting ventilation secondary to loss of tidal volume); or hypotension. Rarely, a patient has sudden cardiovascular collapse or new neurologic symptoms, typically after the patient is placed on positive pressure ventilation, that are consistent with air embolism and require urgent thoracotomy.
Indications for Delayed Thoracotomy
Delayed thoracotomy is performed several days after injury in stable patients, usually for retained hemothorax, trapped lung, persistent air leak, or rib fixation (discussed later).
Thoracic Cage Injuries
Rib Fractures
Rib fractures are seen in 10% of hospitalized trauma patients. The morbidity of rib fracture pain is underappreciated; one third of patients require hospitalization for pain control, and pneumonia develops in another third of these patients. The true incidence of rib fractures is likely higher, in that, a supine chest radiograph is, at best, about 50% sensitive for detecting rib fractures. Chest computed tomography (CT) ( eFig. 76-1 ) is much more sensitive and can also evaluate for other occult injuries. In general, ribs tend to fracture at their weakest structural point (the posterior angle) or directly at the point of impact. They are rarely the sole injury, and 50% of patients with rib fractures have other significant injuries. In a patient with lower rib injury (9th through 12th), abdominal solid organ injury (spleen or liver) should be ruled out.
Pressley and colleagues developed a simple scoring system based on initial clinical findings (age, number of fractured ribs, severity of pulmonary contusion [PC], unilateral vs. bilateral rib fractures), to predict the likelihood of requiring mechanical intubation or intensive care unit (ICU) admission ( Table 76-1 ). The authors found that a patient with a score of 7 or 8 had a high probability of death, need for ICU admission, and a need for mechanical ventilation. Similarly, a score of greater than 5 predicted the need for a longer hospital stay and prolonged mechanical ventilation. Other studies found that an increasing number of rib fractures correlates to increasing ICU length of stay (LOS), hospital LOS, and mortality.
| Age (years) | Number of Rib Fractures |
|---|---|
| <45 = 1 point | <3 = 1 point |
| 45-65 = 2 points | 3-5 = 2 points |
| >65 = 3 points | >5 = 3 points |
| Score : _____ | Score : ______ |
| Pulmonary Contusion | Bilateral Rib Fractures |
| None = 0 points | No = 0 points |
| Mild = 1 point | Yes = 2 points |
| Severe = 2 points | Score : ______ |
| Bilateral = 3 points | |
| Score : _____ | |
| Total Score: _____ | |
Treatment for the vast majority of patients with rib fractures is supportive care, consisting of aggressive pain control and pulmonary rehabilitation. Deep breathing facilitates clearance of secretions to reduce the incidence of pneumonia. Analgesic options include oral analgesics (including opioids), intermittent intravenous analgesics, patient-controlled analgesia, thoracic epidural analgesia (TEA), intrapleural blocks, intercostal blocks, and thoracic paravertebral blocks. Non-opioid analgesic options include acetaminophen, nonsteroidal anti-inflammatory drugs (including ketorolac), anticonvulsants such as gabapentin, and topical lidocaine patches. Numerous small, single-center trials have reported that pain control with TEA, compared with intravenous opioids, results in superior outcomes as measured by less pain, improved pulmonary function, fewer ventilator days, fewer infections, fewer pulmonary complications, shorter ICU LOS, and shorter hospital LOS. Furthermore, TEA may also attenuate the postinjury inflammatory response. At present, TEA is the preferred analgesic modality for pain secondary to blunt thoracic injury. However, TEA is underused, with studies reporting less than 30% TEA use in eligible patients. Epidural analgesia is not without risk, though, and side effects include hypotension, epidural hematoma, urinary retention, and epidural abscess.
Interestingly, a recent pooled meta-analysis did not find any benefit of TEA for duration of mechanical ventilation, ICU LOS, hospital LOS, or mortality. For epidural analgesia to be of benefit, appropriate patient selection is paramount. Although absolute number of fractured ribs is predictive of the need for TEA, it is important to examine the patient carefully to determine whether pain is severe enough to cause respiratory embarrassment. In fact, one review reported that epidural analgesia was associated with increased complications and prolonged LOS in older patients.
Another underused analgesic option is the thoracic paravertebral block (TPVB) or paravertebral catheterization, which is easy to learn, has a greater than 90% success rate, and has a low incidence of side effects. When successful, TPVB has been shown to improve pain scores and pulmonary function tests. TPVB provides unilateral pain relief and thus is usually used only for patients with unilateral rib fractures; however, it is also useful when there are contraindications to TEA, such as coagulopathy, spinal fractures, or altered mental status.
In addition to pain control, a formalized multidisciplinary pathway that includes aggressive respiratory therapy and nutritional support has been shown to decrease ventilator days, LOS, infectious morbidity, and mortality among patients older than age 45 years with more than four fractured ribs.
Older patients with rib fractures are at especially high risk for complications: 15% require intubation, and pneumonia develops in up to 31% ( Fig. 76-1 ). In patients older than age 45 years, morbidity sharply increases when more than four ribs are fractured. Bulger and coworkers reported that patients older than 65 years with fractured ribs have double the morbidity and mortality rates compared with younger patients with similar injuries. Bedside vital capacity measurement can predict hospital LOS and identify those patients requiring discharge to an extended care facility.
Nonunion results in a small percentage of rib fractures and can manifest as chronic pain and discomfort. Typically, follow-up demonstrates that a significant proportion of patients have chronic persistent pain and impairment of both work and personal life. In one study, patients with rib fractures had more disabilities at 30 days after injury than did patients with chronic medical illness. These patients with rib injuries missed an average of 70 days of work. Two months after injury, more than 75% of patients with rib fractures reported some form of disability. Interestingly, the single most important predictive factor for long-term disability after rib injury was the initial intensity of pain. The total number of rib fractures and injury on both sides was not predictive.
Flail Chest.
A flail chest, also known as “stoved-in” or “crushed” chest, is the most severe form of blunt thoracic injury (see Chapter 98 ). The mortality associated with flail chest is up to 40%. Radiographically, it is defined as three or more consecutive ribs fractured in at least two locations. Clinically, a flail chest manifests as paradoxical incursion (rather than excursion) of the “floating segment” of chest wall during inspiration ( 
). Due to the significant energy transfer required to produce this injury, flail chest is almost universally accompanied by PC.
The management of flail chest has evolved over the past half century. Previously, it was believed that the paradoxical chest wall movement was the cause of the respiratory failure and hypoxia. Now, it is understood that the respiratory impairment is due to the underlying pulmonary parenchymal injury. Historically, efforts were focused on correcting the paradoxical motion through external stabilization (“sandbagging”), and later, “internal pneumatic stabilization” (i.e., positive pressure ventilation). Hence, in the mid-twentieth century, the predominant treatment method for all patients with flail chest was mechanical ventilation. Starting in the mid 1970s, some physicians found that these patients could be adequately managed without ventilatory support. It was at this time that it was recognized that the underlying PC rather than the chest wall instability was the driving factor in outcome. Currently, less than half of patients with flail chest require mechanical ventilation. Abnormal gas exchange, not chest wall movement, should drive the decision to mechanically ventilate a patient with flail chest.
In the modern management of flail chest, optimal pain control is paramount. According to the Eastern Association for the Surgery of Trauma practice management guidelines, TEA is the preferred pain treatment modality in the treatment of flail chest. When an epidural catheter is contraindicated, TPVB may be considered. If mild to moderate respiratory compromise is present, a trial of noninvasive ventilation in conjunction with TEA may be considered before proceeding to endotracheal intubation. However, in the absence of respiratory embarrassment, mechanical ventilation to treat paradoxical chest wall motion is not recommended.
Surgical fixation of the “floating” chest wall segment has been practiced for decades in Europe and Asia, but is underused in the United States, likely due to a combination of relative unfamiliarity with the procedure itself and unfamiliarity with the evidence supporting the procedure. In a survey of trauma surgeons, orthopedic surgeons, and thoracic surgeons, only 26% had ever performed or assisted on the procedure and most were unaware of the published randomized trials supporting its use. European and Asian studies report clinical benefit, yet the quality of evidence is poor and consists mainly of small, observational single-center studies. To date, three randomized controlled studies and a meta-analysis evaluating surgical fixation in patients with flail injury have been published supporting rib fracture fixation. Tanaka and colleagues reported that patients who underwent internal fixation of their fractured ribs benefited by less mechanical ventilation, lower incidence of pneumonia, shorter ICU LOS, improved pulmonary function, and quicker return to employment. Granetzny and coworkers also reported decreased need for mechanical ventilation, shorter ICU LOS, and lower incidence of pneumonia in patients randomized to operation. Most recently, Marasco and associates demonstrated decreased ICU LOS and decreased need for tracheostomy in patients randomized to operative repair of flail chest, with no difference in duration of invasive mechanical ventilation. The optimal time for operative intervention is currently unknown and no trial has compared surgical fixation with modern nonoperative management with TEA and chest physiotherapy. An economic analysis based on reported incidences of complications and outcomes concluded that, despite the additional cost of surgery, rib fixation for flail chest remained cost-effective compared with nonoperative management.
Numerous techniques are described for rib fracture fixation, including the use of wire suture, staples, metal or absorbable plates, and screws. A case-control study reported by de Moya and colleagues concluded that rib fracture fixation significantly decreased the need for analgesia. Infection of rib fixation hardware is uncommon, reported to be approximately 2%.
Rescue therapies such as single lung ventilation and high-frequency oscillatory ventilation may be considered when traditional mechanical ventilation fails to improve oxygenation. However, there is no evidence to support routine use of these treatment modalities.
The long-term outcome of flail chest managed nonoperatively is marked by disability, with 70% of patients reporting dyspnea and more than 50% reporting chronic chest wall pain. Less than half of patients are able to return to work.
Sternal Fractures.
The most common cause of sternal fracture is motor vehicle crash ( eFig. 76-2 ). The presence of sternal fractures has traditionally been considered a marker of injury severity, especially in previous decades when seatbelt use was not as widespread. As such, some advocate for hospital admission and close monitoring to rule out other serious injuries, such as blunt cardiac injury. Others report that sternal fracture, ipso facto, is not a significant cause of morbidity or mortality, and many believe that morbidity is mainly attributable to other associated injuries. With increasing seatbelt use, the incidence of sternal fractures has increased while mortality has decreased.
In the initial workup of a patient with sternal fracture, it is important to rule out a blunt cardiac injury with a 12-lead electrocardiogram and serum troponin level. Arrhythmias, ST changes, heart block, signs of ischemia, and elevated troponin levels are considered abnormal screening tests and should be followed by a confirmatory echocardiogram; normal findings on electrocardiogram and initial troponin level essentially exclude the diagnosis of blunt cardiac injury. Nuclear medicine studies are not useful in the diagnostic workup of blunt cardiac injury. As in rib fractures, adequate pain control is paramount in the treatment of sternal fractures. Sternal fracture fixation is rarely indicated.
Clavicle Fractures.
The clavicle is an S-shaped bone that acts as a strut between the shoulder and the axial skeleton. It protects the apex of the lung, brachial plexus, and subclavian vessels. Clavicle fractures follow direct impact to the extremity or chest, and account for 44% of all fractures to the shoulder girdle.
Fractures of the middle third are the most common, accounting for 69% to 81% of all clavicular fractures. Diagnosis of clavicle fractures requires some type of imaging, usually radiography, although many are only visualized on CT. Special radiographic views can be ordered to improve the evaluation for subtle fractures, fracture displacement, or sternoclavicular joint dislocations ( eFigs. 76-3 and 76-4 ). These views include axillary views, 45-degree cephalic tilt, or a serendipity view, which is a 40-degree cephalic tilt.
Treatment generally is nonoperative, with a sling or figure-of-eight brace, and 2 to 6 weeks of immobilization, as well as avoidance of heavy lifting and contact sports for 4 to 6 months. Operative treatment is indicated for all open fractures, skin tenting that will result in an open fracture, and any neurovascular compromise. A recent Cochrane meta-analysis of eight trials involving 555 patients examined the difference between conservative treatment and surgical fixation among patients with fractures of the middle third of the clavicle. Unfortunately, due to heterogeneity between the studies and overall high risk of bias, no strong conclusions can be made and the authors concluded that the decision to operate must be made on a case-by-case basis.
Lung Parenchyma Injuries
Pulmonary Contusion
PC is a common injury, with an incidence of 30% to 75% in patients suffering blunt thoracic injury and up to 17% of all trauma admissions. PC is most common after a blunt mechanism of injury, but can also manifest adjacent to a missile tract through lung parenchyma. At the microscopic level, the contused lung displays edema, alveolar and intraparenchymal hemorrhage, and atelectasis, which results in intrapulmonary shunting, ventilation-perfusion mismatch, and decreased lung compliance. This manifests as hypoxemia, hypercarbia, and increased work of breathing. The full effects of PC may not be obvious immediately; however, clinically significant PC becomes apparent within 24 hours. The natural history of PC is progressive dysfunction over the first few days and resolution within a week.
Severe PC can produce systemic effects. Animal studies demonstrate that after unilateral contusion, there is capillary leak in both ipsilateral and contralateral sides. Both lungs develop increased edema and accumulation of inflammatory cells. Inflammatory cytokines are increased both locally and systemically, and there is evidence of global immune dysfunction. Additionally, PC primes the immune system for an exaggerated response to a subsequent second hit, such as infection ( Fig. 76-2 ). Trauma patients with PC have twice the rate of ventilator-associated pneumonia as those without PC. At 6 years after injury, more than half of patients with PC have evidence of lung fibrosis on CT scan, and long-term lung function can be compromised.
Because not all PCs are clinically significant, several authors have attempted to identify factors predictive of outcome. De Moya and associates developed a simple scoring system, combining the initial CT findings, Glasgow Coma Score, and number of fractured ribs, to predict the need for mechanical ventilation. Interestingly, in this study, less than one third of all PCs were evident on the initial chest radiograph ( eFig. 76-5 ). Other authors have questioned the significance of “occult” PC (i.e., apparent only on CT, eFig. 76-6 ). In a prospective study of 255 patients with PC, Deunk and colleagues reported that patients with occult PC fared no worse than those without PC, while those with PC seen on both chest radiograph and chest CT scan had significantly worse outcomes. Others have attempted to correlate PC size (as a percentage of total lung volume) with outcomes. Studies have demonstrated that patients with PC volume greater than 20% of total lung volume are at increased risk for requiring mechanical ventilation, developing pneumonia, and developing acute respiratory distress syndrome (ARDS).
At this time there is no well-supported intervention to treat PC, and management consists mainly of supportive care and avoidance of iatrogenic injury. Steroids are not recommended and prophylactic antibiotics are strongly discouraged. Four decades ago, Trinkle and coworkers recognized that crystalloid administration increased PC size while diuresis decreased PC size. Pharmacologic therapies being investigated include arginine vasopressin and dexmedetomidine. A recent animal study reported that dexmedetomidine infusion in a PC model improved hemodynamic parameters, decreased inflammatory infiltration, limited the extent of lung damage, and abrogated pulmonary edema. In patients with early, severe hypoxemia (arterial P o 2 /F io 2 <200), a trial of noninvasive ventilation may be attempted in order to decrease the need for intubation; however, the development of pneumothorax must be carefully monitored. In animal studies, the application of positive end-expiratory pressure has been shown to decrease the size of PCs. Small clinical studies have reported that recruitment maneuvers are successful in improving aeration (“open lung” strategy). In patients with PC, the use of airway pressure release ventilation has been reported to decrease the incidence of ventilator-associated pneumonia; however, experience and evidence is limited. For severe, unilateral PC, lung isolation ventilation may be considered. The use of rescue therapies such as high-frequency oscillatory ventilation, surfactant administration, prone positioning, and extracorporeal membrane oxygenation for the treatment of PC is poorly studied and considered experimental at this time.
In the management of PC, the patient should be resuscitated to maintain signs of adequate tissue perfusion. Once this has been achieved, however, meticulous attention should be paid to the avoidance of excessive fluid administration, to the point of using a pulmonary artery catheter if necessary to help guide diuretic therapy. Aggressive pulmonary toilet and adequate analgesia are paramount in preventing pneumonia.
Pulmonary Laceration
Pulmonary laceration reflects tearing of the pulmonary parenchyma that disrupts the alveolar walls. Pulmonary lacerations result from several mechanisms, such as alternate compression and decompression of the chest wall or from a sudden, rapid increase in intrathoracic pressure with a closed glottis leading to high intra-alveolar pressure that produces shearing of pulmonary parenchymal tissue. Alternatively, pulmonary laceration may also result from direct puncture of lung tissue by a fractured rib, missile, or stab wound, or from shearing of lung tissue fixed by previously formed pleural adhesions. The disrupted pulmonary tissue fills with blood and/or air and manifests on thoracic imaging as one or more pulmonary parenchymal cavities ( eFig. 76-7 ), appearing as gas-fluid levels, frequently with surrounding pulmonary consolidation and ground-glass opacity related to hemorrhage and atelectasis.
While only a small fraction of patients with thoracic injury ultimately require urgent thoracotomy, of those that proceed to surgery, about one third may require lung resection, usually to remove severely injured lung tissue, to control hemorrhage, or to remove irreparable proximal bronchus injuries. The extent of resection can range from simple, nonanatomic “wedge” resection, to formal anatomic lobectomy, to the extremely morbid pneumonectomy. For the majority of penetrating injuries, lung-sparing techniques such as simple suture or “tractotomy” are sufficient ( Fig. 76-3 ). Not surprisingly, there is a stepwise increase in mortality with increasing extent of lung resection for trauma: wedge resection (19%), lobectomy (27%), and pneumonectomy (53%).
While most healthy patients can tolerate a wedge resection or lobectomy, pneumonectomy imposes a tremendous physiologic burden and patients may succumb to right ventricular failure. Using a porcine model, Cryer and colleagues demonstrated that pulmonary vascular resistance increases up to 500% within 4 hours after pneumonectomy. Postoperative complications in survivors are common, and include pneumonia, bronchopleural fistula, empyema, or erosion into the pulmonary artery.
Bronchopleural Fistula
A bronchopleural fistula is a direct connection between the bronchus and atmosphere by way of the pleural space and tube thoracostomy. While most heal spontaneously, a persistent bronchopleural fistula may seriously impair the ability to ventilate a patient secondary to loss of tidal volume through the fistula. It is believed that air flowing through the fistula impedes healing and therefore medical therapies are aimed at reducing bronchopleural fistula flow.
Nonsurgical treatments include minimizing airway pressures (via lower tidal volumes, positive end-expiratory pressure, and inspiratory time), application of positive intrapleural pressure through the chest tube, isolated contralateral lung ventilation, dependent positioning of the side with the bronchopleural fistula, and use of high-frequency oscillatory ventilation. Bronchoscopy is useful to identify the injury directly (proximal) or the affected bronchial segment (distal). A balloon is used to occlude the segments sequentially until a reduction in air leak is noted. Once the affected bronchus is identified, agents such as silver nitrate, cyanoacrylate-based agents, gelatin, fibrin, and even fishing weights have been applied to occlude the bronchial segment and close the fistula. Bronchopleural fistula refractory to medical therapy for more than 7 days may require pleurodesis or operative intervention.
Miscellaneous
Pneumothorax.
Pneumothorax, air within the pleural cavity, is present in 40% and 20% of blunt and penetrating thoracic trauma patients, respectively (see Chapter 81 ). The clinical consequence of pneumothorax ranges from asymptomatic to the most life-threatening manifestation, “tension” pneumothorax, whereby progressive air trapping in the thorax results in a contralateral mediastinal shift, compression or kinking of the superior vena cava/inferior vena cava, and a precipitous drop in preload and cardiac output ( Fig. 76-4 ). Clinically, this form of “obstructive” shock manifests as hypotension and hypoxia, and should be treated with immediate thoracic decompression, either with needle thoracostomy or tube thoracostomy. Current recommendations from the Advanced Trauma Life Support program are to insert the needle in the mid-clavicular line at the second intercostal space; however, some authors have questioned the efficacy of inserting the needle in this location and recommend instead inserting the needle in the mid-axillary line at the fifth intercostal space because of decreased chest wall thickness in this location. Additionally, lateral needle insertion avoids potentially life-threatening hemorrhagic complications such as laceration of the internal mammary artery, subclavian artery, or pulmonary artery. Others recommend inserting a chest tube at the outset instead of a needle due to frequent failure rates associated with the latter.
