Blunt Chest Trauma: Who Needs Computed Tomography Imaging?

Authors: Erica Simon, DO, MHA (@E_M_Simon – EM Chief Resident at SAUSHEC, USAF) and Daniel Sessions, MD (@Danosaurus – EM Associate Program Director, SAUSHEC, USAF / Medical Toxicologist, South Texas Poison Center) // Edited by: Alex Koyfman, MD (@EMHighAK – EM Attending Physician, UT Southwestern Medical Center / Parkland Memorial Hospital) and Manpreet Singh, MD (@MPrizzleER – Assistant Professor of Emergency Medicine / Department of Emergency Medicine – Harbor-UCLA Medical Center)

Case

A 37-year-old male presents following an MVC in which he was the restrained driver of compact vehicle.  Per bystander, the patient’s car was struck on the passenger side by an SUV traveling at speeds > 60 mph.  Upon EMS arrival the man was noted as GCS 14 (confused) – unable to recall the events surrounding the accident.  Initial VS: BP 148/88, HR 102, RR 24, SpO2 89% on room air.

As the patient enters the trauma bay you ask yourself: do I need to order a CT chest to rule out blunt injury?

Although screening CXRs and trauma pan-scans are common-place, let’s take a minute to review current literature to aid in our decision-making regarding the use of imaging in the evaluation of blunt thoracic injury.

Epidemiology

In the U.S., blunt chest trauma is associated with a mortality rate as high as 60%.¹  Occurring most frequently in the setting of MVCs (63-78%²), falls, blows from blunt objects, and blast injuries, blunt chest injury represents the third leading cause of death in polytrauma patients.³

Rib fractures are the most commonly identified injury in individuals having sustained blunt chest trauma (present in up to 67% of patients have undergone radiologic imaging).^4-6  Of persons presenting to trauma centers, approximately 30-50% are diagnosed with pulmonary contusions (mortality 10-25%⁷), pneumothoraces, or hemothoraces following blunt thoracic injury.⁸  Sternal fractures reportedly occur in 8% of blunt chest trauma patients, and are associated with cardiac contusions in 20-40% of cases.^2,5 Tracheobronchial injuries are rare, identified in < 1% of individuals suffering blunt thoracic trauma.⁹

Morley, et al.⁵ offer a list of differential diagnoses to be considered in the setting of blunt chest injury:

Imaging the Chest

After initial stabilization, radiologic imaging plays an important role in the evaluation of the trauma patient.  Let’s take a moment to discuss imaging modalities frequently utilized in the evaluation of a patient with blunt chest injury.

CXR versus Chest CT scan

While CXRs are useful for the expeditious detection of serious life-threatening conditions including tension pneumothorax/hemothorax, widened mediastinum, mediastinal hematoma, flail chest, and malpositioned lines/tubes,^10-13 numerous studies have demonstrated the superior sensitivity of CT in the detection of vascular and non-vascular thoracic injuries.  As examples:

Chardoli, et al., 2013:¹⁴ Cross-sectional study of hemodynamically stable trauma patients ≥ age 16 years (SPB > 90 mmHg, HR 60-100 bpm, RR 10-35/min) with blunt chest injury presenting to Sina Hospital, Tehran from March 2011- March 2012:

• 200 Patients: 168 male, 32 female – all undergoing CXR and chest CT without contrast:
  • CT detected 11 hemothoraces, 11 pulmonary contusions, 8 pneumothoraces, 2 foreign bodies, 2 cases of subcutaneous emphysema, and 1 sternal fracture missed by CXR.

Traub, et al., 2007:¹⁵ Retrospective review of the health records of 141 patients (106 male; mean age 47.2 years; mean Injury Severity Score (ISS) 24) having sustained blunt chest trauma (MVC: n=12; fall: n=33; pedestrian injury: n=24) and having undergone both CXR and CT evaluation:

• 98% of patients (n = 82) had pathologic findings on CT scan vs. 58% (n = 82) on CXR
• Additional investigations/interventions performed on 19% of patients (n=27) following CT:
  • 15 patients required chest tubes, 4 required aortic angiography, 4 transesophageal echocardiography, 2 endoluminal stent insertion for aortic injuries, and 2 patients required thoracotomies

Trupka, et al., 1997:¹⁶ Prospective study comparing CXRs and thoracic CTs of 103 consecutive patients with clinical or radiological signs of isolated blunt chest injury (Average ISS: 30; Average Abbreviated Injury Scale (AIS): 3):

• Thoracic CT detected major chest trauma complications previously unidentified by CXR in 27 patients:
  • 33 pulmonary contusions, 27 pneumothoraces, 21 hemothoraces, 7 residual pneumothoraces following chest tube placement, 5 displaced chest tubes, 2 diaphragmatic ruptures, and 1 myocardial rupture

Note: The following offers additional information regarding the assignment of ISS and AIS: Brohi, K. Trauma Scoring. Trauma.Org. Available from: http://www.trauma.org/archive/scores/iss.html

Let’s compare CXR and Chest CT in terms of specific differential diagnoses:

Tracheobronchial Injury:

• Bronchial tears commonly occur on the right side of the thorax, within 2.5 cm of the carina.¹⁷
• CXR: non-specific: may see pneumothorax, pneumomediastinum, subcutaneous emphysema.¹²
• CT is the imaging modality of choice.

Esophageal Injury:

• Rare in the setting of blunt chest trauma (<1% of patients¹⁸).
  • CXR findings suggestive of esophageal tear: pneumomediastinum, left pneumothorax, left pleural effusion, subcutaneous emphysema, left lower lobe atelectasis.¹⁸
  • CT with water-soluble contrast (esophagram) = definitive diagnosis
    • CT findings: CXR findings + mediastinal fluid and extraluminal enteric contrast.¹⁸

Thoracic Aortic Injury:

• CXR utilized primarily to exclude the diagnosis (98% negative predictive value¹⁹).
  • Findings concerning for thoracic aortic injury: mediastinal widening, abnormal contour or indistinctness of the aortic knob, apical pleural cap, rightward deviation of the nasogastric tube within the esophagus, rightward deviation of the trachea, downward displacement of the left mainstem bronchus, thickening of the right paratracheal stripe.^12,20

Pulmonary Parenchymal Injury:

• Pulmonary contusion:
  • Appears on CXR and CT as non-segmental consolidation and ground-glass opacification, often adjacent to the area of trauma.^12,21
  • CT demonstrates improved sensitivity as compared to CXR:
    • Allows for detection of pathology immediately following injury, whereas radiographic evidence of pulmonary contusion may not develop until 6 hours following the initial trauma.^12,21
    • Pulmonary contusion often obscured by concomitant hemothorax/pneumothorax on CXR.¹²

• Pulmonary laceration:
  • Appears on CXR as the formation of a cavity filled with blood or air.¹²
  • Characterized by CT as air collections within an area of consolidation.¹²
  • CT demonstrates improved sensitivity as compared to CXR:
    • Study by Wagner et al.:²² Retrospective review of 85 patients with pulmonary lacerations having received CXR and CT: 99% of lacerations present on CT, 5% seen radiographically.

Pneumothorax, Pneumomediastinum, and Hemothorax

• Pneumothorax occurs in 30-40% of patients following blunt chest injury.²²
  • CXR: Pleural air rises to the most non-dependent portion of the thorax (supine patient = anterior, caudal aspect of the pleural space).¹²
    • Radiographic signs: deep, lucent costophrenic sulcus (“deep sulcus sign”), a relative increase in the lucency of the affected lung base, and the “double diaphragm sign” (air outlining the dome and insertion of the diaphragm).¹²
  • CT is the gold standard for diagnosis.¹²

• Pneumomediastinum is seen on CXR and CT as free air collections surrounding mediastinal structures and dissecting along mediastinal fat.¹³
  • A continuous diaphragm sign may be seen if air is present between the pericardium and diaphragm.¹²
  • CT demonstrates increased sensitivity for the detection of pneumomediastinum and allows for the differentiation between pneumothorax and pneumomediastinum (presence of septa within the pneumomediastinum delineated in the lung window).¹³

• Hemothorax occurs in 50% of blunt chest trauma cases.¹⁰
  • Appearance varies according to the volume of blood and coagulation state of pleural fluid blood products: may be seen on supine CXR as diffuse infiltrate.
  • CT is highly sensitive for the detection of even small hemothoraces: allows for the measurement of Hounsfield units and subsequent differentiation between pleural fluid and mediastinal vessels.¹⁰

Thoracic Skeletal Injuries:

• Fractures of the thoracic spine occur in 16-30% of all trauma patients, and up to 62% are associated with neurologic deficits.²³
  • Chest CT is the imaging modality of choice (acute pathology requires subsequent dedicated thoracic CT).
• Sternal fractures and sternoclavicular dislocations are often missed on supine CXR. CT frequently identifies the diagnosis, frequently with concomitant retrosternal hematoma.¹²
• Scapular fractures rarely occur in the setting of blunt trauma (3% of cases) and are readily identified upon initial CXR.¹⁰
• As above, rib fractures are the most common finding in patients sustaining blunt thoracic injury. Supine CXR demonstrates poor sensitivity in the identification of rib fractures (particularly posterior rib fractures).¹²

Diaphragmatic Injury

• Occurs in 1-8% of patients sustaining blunt trauma to the chest and abdomen.²⁴
• CXR may demonstrate herniation of hollow viscus into the thorax with abnormal course of the nasogastric tube, elevated hemidiaphragm, or loss of contour of the hemidiaphragm.²⁴
• Axial CT demonstrates a sensitivity of 70-90% in the identification of traumatic diaphragmatic injuries.²⁵
  • CT findings consistent with diaphragmatic injury include: discontinuity of the diaphragm, herniation of abdominal contents into the thorax, and constriction of bowel at the herniation site (“collar sign”).¹²
  • CT may miss diaphragmatic injury when intraabdominal blood or hemothorax obscures the diaphragm.¹²

CXR vs. CT – How Do We Decide?

To date, there is no clinical decision rule regarding indications for chest radiography in trauma patients.¹¹

What does the literature tell us?  Studies regarding clinical variables for selective chest radiography in blunt trauma patients are few and far between.  Of note:

Rodriguez, et al., 2006:²⁶ Prospective study of adult (age > 15 years) blunt trauma patients who received chest radiographs at two urban trauma centers:

• 492 patients enrolled:
  • Palpation tenderness and chest pain had the highest sensitivity for identifying significant acute intrathoracic injury (90%). Hypoxia had the highest specificity for identifying significant acute intrathoracic injury (97%).
  • Palpation tenderness + hypoxia: identified all significant acute intrathoracic injuries (sensitivity 100% (95% CI: 91-100%); specificity 50% (95% CI: 45-54%)).
    • Significant acute intrathoracic injury defined as: pneumothorax, hemothorax, aortic injury, 2 or more rib fractures, sternal fracture or pulmonary contusion by blinded radiologists’ chest radiograph interpretation.

In the setting of blunt thoracic injury, Advanced Traumatic Life Support (ATLS) guidelines, developed by the American College of Surgeons Committee on Trauma (ACS-COT), require only that an anteroposterior (AP) CXR be obtained following an applicable procedure to document the position of tubes and lines, and to evaluate for iatrogenic pneumothorax, hemothorax, and mediastinal abnormalities.^10-13

Today the use of CXR as an initial screening examination in the trauma patient is a common practice.

What about CT evaluation? 

The ATLS manual makes little mention of CT evaluation, save a comment noting CT chest as an accurate screening method for traumatic aortic injury.¹¹  What can we learn from trauma literature?  The majority of texts recommend CT evaluation in the setting of severe mechanism of injury (MOI), and for patients of special populations. 

To better understand this terminology, let’s review ACS-COT classifications, published in the Centers for Disease Control and Prevention’s Guidelines for Field Triage of Injured Patients:²⁶

Note: Additional trauma literature notes the following as mechanisms predictive of significant thoracic injury: assault with depressed level of consciousness without any other evidence of trauma.²⁷

What if we’re concerned for a major injury following a non-severe blunt thoracic injury?

Previously experts recommended the use of a thorough history and physical examination, combined with clinical gestalt, to direct imaging evaluation.  As a review:

Tracheobronchial Injury:¹²

• CXR demonstrating persistent pneumothorax following chest tube placement, collapse of the lung away from the hilum (“fallen lung sign”), overdistention/herniation of the endotracheal balloon, or clinical concern for tracheobronchial injury => CT (+ bronchoscopy).

Esophageal Injury:¹²

• Obtain a CT with water-soluble contrast (esophagram).

Thoracic Aortic Injury:¹²

• All patients receive a screening CXR.
• Abnormal CXR concerning for aortic injury or unstable patient suspected of having a traumatic aortic injury => aortography.
• Stable patient suspected of having a traumatic aortic injury => contrasted helical CT.

Pulmonary Parenchymal Injury:¹²

• Pulmonary contusion or pulmonary laceration/hematoma
  • Consider CXR or repeat CXR (taking into account time since initial injury).
  • Consider CT: delineates alternative etiologies (aspiration, atelectasis, cardiogenic/non-cardiogenic pulmonary edema).

Pneumothorax, Pneumomediastinum, or Hemothorax:¹⁰

• Consider CT if CXR unremarkable and patient symptomatic (e.g. tachypneic, hypoxic, hypotense): CT demonstrates high sensitivity, is capable of differentiating between a pneumothorax and pneumomediastinum, and detects small hemothoraces.

Thoracic Skeletal Injuries:¹²

• Obtain a CT chest.
• If sternoclavicular dislocation is identified => CT angiography as appropriate.
• If a thoracic spinal fracture is identified on chest CT => obtain a dedicated thoracic spine CT at the level in question with sagittal and coronal reconstructions (allows for assessment of the type of fracture and fracture stability).¹²

Diaphragmatic Injury:¹²

• If CXR fails to demonstrate herniation of bowel contents and suspicion remains high => Chest CT.
  • Recall that chest CT demonstrates poor sensitivity => surgical consult for consideration for exploratory laparotomy.

Can we trust our clinical gestalt?

Given the risks associated with non-selective CT scanning (increased attributable risk for cancer mortality and risk of contrast nephropathy), in 2011 Smith, et al. set out to assess the emergency physician’s (EPs) skill at predicting injury in patients presenting following blunt chest trauma:²⁸

• 18 board-certified/board-eligible EPs with a median of 11 years of experience completed pre-test probability worksheets for 329 patients regarding the likelihood of CT demonstrated injury based upon history, physical examination, CXR, and FAST (if performed):

• Conclusions: EPs are skilled at identifying patients at very low risk for thoracic injuries following blunt trauma, however, even in the patients thought to be at low risk, numerous serious injuries would have been missed if a CT had not been performed.²⁸

Aside: This study is referenced as an example and is limited in that data points were obtained from pre-test probability worksheets completed by one attending group of physicians.

Is there a manner in which to augment our clinical-decision making?

Enter the NEXUS Chest CT decision instruments:

• Rodriguez, et al., 2015:²⁹ Prospective observational study of blunt trauma patients ( > 14 yrs of age) presenting to 8 US, urban level on trauma centers within 6 hours of initial injury:
  • Derivation phase: 6,002 patients => physicians recorded the presence or absence of 14 clinical criteria prior to viewing chest imaging results.
    • CT radiology readings were utilized to classify injury outcomes (major vs. minor) based upon an expert-panel (EPs and trauma surgeons) derived classification scheme (see below).
    • Recursive partitioning (statistical method which creates a decision tree to classify members of a population into sub-populations based upon dichotomous independent variables) was utilized to create two decision instruments.
  • Validation phase: 5,475 patients => prospectively tested the performance of the decision instruments.

Decision Instruments

Patient are considered low risk and unlikely to benefit from chest CT if they have none of the following:

 NEXUS Chest CT-All (identifies injuries of major and minor clinical significance – as above):²⁸

• Abnormal CXR (any thoracic injury including clavicular fracture or a widened mediastinum)
• Rapid deceleration mechanism (fall from a height > 20 ft or MVC at speeds > 40 mph with sudden deceleration)
• Distracting injury
• Chest wall tenderness (boundaries defined as the upper and lower costal margins circumferentially; isolated clavicular tenderness does not qualify)
• Sternal tenderness
• Thoracic spine tenderness
• Scapular tenderness

NEXUS Chest CT-Major (identifies injuries of major clinical significance – as above):²⁸

• Abnormal CXR (any thoracic injury including clavicle fracture or a widened mediastinum)
• Distracting injury
• Chest wall tenderness (boundaries defined as the upper and lower costal margins circumferentially; isolated clavicular tenderness does not qualify)
• Sternal tenderness
• Thoracic spine tenderness
• Scapular tenderness

Bottom line

Both NEXUS Chest CT-All and CT-Major have > 99% sensitivity in detecting major thoracic injuries.  NEXUS Chest CT-All has > 95% sensitivity in detecting major and minor thoracic injuries following blunt trauma (in patients presenting within 6 hours of injury).

If you’ve read to this point and are wondering, what about ultrasound?  We’ve got you covered:

Bedside ultrasound is a useful adjunct in the evaluation of patients presenting following blunt chest injury as it allows for the rapid detection of pneumothoraces, hemothoraces, and pericardial effusions.  As compared to CXR, US demonstrates greater sensitivity for the detection of pneumothoraces (sensitivity 86%³⁰) and is more accurate in the diagnosis of hemothoraces,⁵  however, CT remains the gold standard.⁵

To summarize our discussion:

Note: This varies from the widely accepted trauma surgery approach, which would advocate for a screening CXR in the evaluation of a patient who sustains blunt chest injury.

Back to our case

Given the patient’s severe MOI, a CT is warranted.  Further detail regarding the secondary survey would be required to utilize the NEXUS Chest CT decision instruments.

Key Pearls

• The differential diagnosis of blunt chest injury is broad:
  • Understand the limitations of imaging modalities in this setting => even CT is not sensitive enough to exclude diaphragmatic injury.
• Recognize that there are no criteria that direct the use of CXR in trauma patients.
  • The practice of performing a screening CXR in patients sustaining blunt trauma to the chest has become widely accepted, but is not always necessary.
    • Perform in patients in whom aortic injury is a concern.
    • Consider (with or without US) in patients who are hemodynamically unstable, or those who are persistently symptomatic (hypoxia despite O2 therapy, etc.).
  • Utilize the H&P to develop your clinical gestalt – understand what constitutes a severe MOI and be aware of populations at risk for significant injury following blunt chest injury (the elderly, the anticoagulated, etc.)
  • Utilize the NEXUS Chest CT decision instruments to augment, but not substitute for, your clinical decision-making.

References / Further Reading

1. Clark G, Schecter W, Trunkey D. Variables affecting outcome in blunt chest trauma: flail chest vs. pulmonary contusion. J Trauma. 1988; 28:298-304.
2. Mayberry J. Imaging in thoracic trauma: the trauma surgeon’s perspective. J Thorac Imaging. 2000: 15:76-86.
3. American College of Surgeons Committee on Trauma Leadership, National Trauma Data Bank (NTDB) annual report 2007. In: Clear D, Fantus R, eds. Chicago, IL: American College of Surgeons. 2007:1-64.
4. Newman R, Jones I. A prospective study of 413 consecutive car occupants with chest injuries. J Trauma. 1984; 24(2):129-135.
5. Morley E, Johnson W, Leibner E, Shahid J. Emergency department evaluation and management of blunt chest and lung trauma (Trauma CME). Emerg Med Pract. 2016; 18(6):1-20.
6. Palvanen M, Kannus P, Niemi S, et al. Epidemiology of minimal trauma rib fractures in the elderly. Calcif Tissue Int. 1998; 62(3):274-277.
7. Cohn S. Pulmonary contusion: review of the clinical entity. J Trauma. 1997; 42(5):973-979.
8. Sangster G, Bonzalez-Beicos A, Carbo A, et al. Blunt traumatic injuries of the lung parenchyma, pleura, thoracic wall, and intrathoracic airways: multidetector computer tomography imaging finding. Emerg Radiol. 2007; 14(5):297-310.
9. Wintermark M, Schnyder P, Wicky S. Blunt traumatic rupture of a mainstem bronchus CT demonstration of the “fallen lung” sign. Eur Radiol. 2001; 11(3):409-411.
10. Oikonomou A, Prassopoulous P. CT imaging of blunt chest trauma. Insights Imaging. 2011; 2(3):281-295.
11. Kool D, Blickman J. Advanced Trauma Life Support. ABCDE from a radiological point of view. Emerg Radiol. 2007; 14(3):135-141.
12. Zinck S, Primack S. Radiographic and CT findings in blunt chest trauma. J Thorac Imaging. 2000; 15(2): 87-96.
13. Palas J, Matos A, Mascarenhas V, Heredia V, Ramalho M. Multidetector computer tomography: evaluation of blunt chest trauma in adults. Radiol Res Pract. 2014. doi: 10.115/2014/864369 [epub ahead of print].
14. Chardoli M, Hasan-Ghaliaee T, Akbari H, Rahimi-Movaghar V. Accuracy of chest radiography versus chest computed tomography in hemodynamically stable patients with blunt chest trauma. Chin J Traumatol. 2013; 16(6):351-354.
15. Traub M, Stevenson S, McEvoy S, Briggs G, Lo S, et al. The use of chest computed tomography versus chest X-ray in patients with major blunt trauma. Injury. 2007; 38(1):43-47.
16. Trupka A, Waydhas C, Hallifeldt K, Nast-Kolb D, Pfeifer K, et al. Value of thoracic computed tomography in the first assessment of severely injured patients with blunt chest trauma: results of a prospective study. J Trauma. 1997; 43(3): 405-411.
17. Mirvis S, Templeton P. Imaging in acute thoracic trauma. Semin Roentgenol. 1992; 27:184-210.
18. Maroney M, Mirvis S, Shanmuganathan K. Esophageal occlusion caused by thoracic spine fracture or dislocation: CT diagnosis. AJR. 1996; 167:714-715.
19. Mirvis S, Bidwell J, Buddemeyer E, et al. Value of chest radiography in excluding traumatic aortic rupture. Radiology. 1987; 163:487-493.
20. Groskin S. Selected topics in chest trauma. Semin Ultrasound CT MRI. 1996; 17:119-141.
21. Ganie F, Lone H, Lone G, Wani M, Singh S, et al. Lung contusion: A clinic-pathological entity with unpredictable clinical course. Bull Emerg Trauma. 2013; 1(1):7-16.
22. Wagner R, Crawford W, Schimpf P. Classification of parenchymal injuries of the lung. Radiology. 1988; 167:77-82.
23. Pal J, Mulder D, Brown R, et al. Assessing multiple trauma: Is the cervical spine enough? J Trauma. 1988; 28:1282-1284.
24. Boulanger B, Milzman D, Rosati C, Rodriguez A. A comparison of right and left blunt traumatic diaphragmatic rupture. J Trauma. 1993; 35:255-260.
25. Murray J, Baoili E, Gruden J, et al. Acute rupture of the diaphragm due to blunt trauma: diagnostic sensitivity and specificity of CT. AJR. 1996; 166:1035-1039.
26. Rodriguez R, Hendey G, Marek G, Dery R, Bjoring A. A pilot study to derive clinical variables for selective chest radiography in blunt trauma patients. Ann Emerg Med. 20016; 47(5):415-418.
27. Frieden T, Jaffe H, Stephens J, Thacker S, Zaza S, et al. Center for Disease Control and Prevention: Guidelines for Field Triage of Injured Patients. 2011. Available from: https://www.facs.org/~/media/files/quality%20programs/trauma/vrc%20resources/6_guidelines%20field%20triage%202011.ashx
28. Salim A, Sangthong B, Martin M, et al. Whole body imaging in blunt multisystem trauma patients without obvious signs of injury: results of a prospective study. Arch Surg. 2006; 141(5):468-473.
29. Smith C, Barrett T, Berger C, Zhou C, Thurman R, et al. Prediction of blunt traumatic injury in high-acuity patients: bedside examination vs computed tomography. Am J Emeg Med. 2011; 29(1): 1-10.
30. Rodriguez R, Langdorf M, Nishijima D, Bauman B, Hendey G, et al. Derivation and validation of two decision instruments for selective chest CT in blunt trauma: a multicenter prospective observational study (NEXUS Chest CT). PLoS Med. 2015; 12(10):e1001883.
31. Wilderson R, Stone M. Sensitivity of bedside ultrasound and supine anteroposterior chest radiographs for the detection of pneumothorax after blunt trauma. Acad Emerg Med. 2010; 17(1):11-17.