Radiology in major trauma

Initial trauma series

• Supine CXR and pelvic X-ray
  • CXR should include both clavicles, ribs, lung, mediastinum and diaphragm
• If cervical spine injury suspected, CT is first-line from occiput to T4/5 with axial images + coronal and sagittal reconstructions
  • Lateral C-spine X-ray poorly visualises the atlanto-occipital junction and cervicothoracic junction

Hazards of radiation

• Lead gowns and thyroid shields have been shown to protect against ionising radiation within recommended occupational limits
• Minimise the number of X-rays taken in trauma bay
• Radiation exposure decreases inversely with the square of the distance from the source so move away if possible but DO NOT stop resuscitation (hence everyone should have gowns on)
• A single CT scan gives a tissue dose (effective dose) of 10-30mSv
• 50-200mSv has been shown to cause an increase in cancer risk among atomic bomb survivors and risk is higher for younger people
• US FDA states that a single CT scan with effective dose of 10mSv carries 1/2000 lifetime risk of fatal cancer

Radiation dose estimates

Panscan or selective?

• REACT-2
  • Included either unstable OR suspected severe injuries OR severe mechanism of injury (fall >3m, ejection, death of occupant, severely injured other patient, wedged/trapped chest or abdomen)
  • Found reduced time to diagnosis and reduced ED LOS
  • No difference in 24 hour mortality (8% vs. 6%), 30-day mortality
  • Very high mortality in this patient cohort overall
  • Although, 46% of patients in selective imaging group ended up with panscan
  • The selective imaging group only had 1mSv less radiation overall!
• Increases radiation exposure (20mSv), incidental findings and logistically resource-intensive (although balanced against ED LOS reduction)
• Mortality benefits shown in non-randomised retrospective studies with significant confounders
• Overall recommendation is for panscan if abnormal mental status, unstable or suspicion of critical injury

Specific regional radiology

• Head
  • Head trauma accounts for 50-75% of mortality associated with major trauma
  • Indications in significant head injury
    • GCS <9 after resuscitation
    • Neurological deterioration of 2 or more points
    • GCS 9-13 that persists longer than 2 hours
    • Persistent headache or vomiting
    • Focal neurological signs
    • Suspected or known skull fracture
    • Penetrating injury to cranium known or suspected
    • Age >50 with suspicious mechanism of injury
    • Any head injury in a patient on anticoagulation therapy

Specific regional radiology

• Canadian CT head rule for patients with minor head injury  as defined by witnessed LOC, definite amnesia or witnessed disorientation in a patient with GCS 13-15
  • High-risk for neurosurgical intervention (100% sensitive)
    • GCS <15 at 2 hours post-injury
    • Suspected open or depressed skull fracture
    • Any sign of basal skull fracture
    • Vomiting 2 or more episodes
    • Age 65 or older
  • Medium risk for brain injury on CT (97.2% sensitive)
    • Amnesia before impact >30 minutes
    • Dangerous mechanism (pedestrian struck by car, occupant ejected, fall >3 feet or five stairs)

Specific regional radiology

• Classification of intracranial haemorrhage
  • Epidural:
    • Typically middle meningeal artery disruption in setting of skull fracture
    • Ovoid, does not cross sutures but may cross the midline
  • Subdural haematoma:
    • Usually venous bleeding, crescentic, may be larger, may cross sutures but do not cross the midline
  • Subarachnoid: 
    • Small subarachnoid vessel disruption or direct extension from intraparenchymal bleed. 
    • May visualise haemorrhage in sulci of cerebral convexities or within subarachnoid cisterns
  • Intraventricular: 
    • May be caused by tearing of subependymal veins on surface of ventricles or direct extension of intraparenchymal bleed
    • Blood products tend to layer, particularly within occipital horns of lateral ventricles when supine

Specific regional radiology
 – Epidural haematoma

Specific regional radiology
 – Subdural haematoma

Specific regional radiology
 – Subarachnoid haemorrhage

Specific regional radiology

• Cerebral contusions
  • Focal parenchymal bleeding
  • May occur within superficial grey matter/subcortical white matter due to direct trauma
  • If in deeper white matter, suggests intraparenchymal vessel disruption
  • Commonly increase in size and number over 24 hours with associated oedema that may show mass effect
• Patients on anticoagulation or antiplatelet therapy
  • Bleeding may progress over 24 hours

Specific regional radiology

• Diffuse axonal injury
  • Non-contrast CT may show tiny foci of petechial haemorrhage at grey-white matter interface, within corpus callosum or within brainstem
• Acute established infarction
  • May be visualised on non-contrast CT in trauma and may indicate acute vascular injury or mass effect from cerebral oedema
• Basal skull fractures
  • Opacification of mastoid air cells, fluid in middle ear cavity and pneumocephalus
  • Longitudinal temporal bone fractures signify blow to side of head and constitute 70-90% of temporal bone fracture
  • Transverse fractures indicate blow to front or back of head

Specific regional radiology

• Blunt cerebrovascular injury (BCVI)
  • 0.1% of all trauma patients suffer blunt injury to the carotid or vertebral arteries
  • Many diagnosed late after neurological deterioration occurs with morbitidy in 80% and mortality of 40%
  • When asymptomatic patients are screened, incidence rises to 1% (10x) and up to 2.7% for patients with ISS >15
  • The Denver modification of screening criteria can be utilised (see next slide)

Specific regional radiology – Denver modification of screening tool for BCVI

• Risk factors for BCVI
  • High energy transfer mechanism with:
    • LeFort 2 or 3 fractures
    • Cervical spine fracture patterns
      • Subluxation, fractures extending into foramen transversarium, fractures of C1-3
    • Basilar skull fractures with carotid canal involvement
    • DAI with GCS <7
    • Near hanging with anoxic brain injury
• Symptoms/signs of BCVI
  • Arterial haemorrhage
  • Carotid bruit
  • Expanding cervical haematoma
  • Focal neurological deficit
  • Neurological findings incongruous with CT scan findings
  • Ischaemic stroke on secondary CT scan

Specific regional radiology – Grading scale for BCVI

• Grade 1: Intimal irregularity with <25% narrowing
• Grade 2: Dissection or intramural haematoma with >24$ narrowing
• Grade 3: Pseudoaneurysm
• Grade 4: Occlusion of lumen
• Grade 5: Transection with extravasation

Specific regional radiology – Facial injury

• Most common injury to midface is the blowout fracture of the orbit
  • Fracture of orbital floor or medial wall of orbit in the region of thin lamina papyracea
• Mandibular fractures
  • OPG sensitive but CT optimal (esp. for mandibular neck, condyle and TMJ + dislocation)
• Zygomatic fractures
  • Tripod fractures (zygomaticomaxillary) separate the malar eminence of the zygoma from frontal, temporal and maxillary connections (floating)
• LeFort fractures
  • Direct trauma to midface
  • LeFort I: Maxilla at level of nasal floor allowing floating palate
  • LeFort II: Passes through nasal bones as well as medial, inferior and lateral walls of the maxillary antrum
  • LeFort III: Floating face/craniofacial dysjunction involving the nasal bones, medial and lateral orbital walls and zygomatic arch

Specific regional radiology – Zygmomatic tripod fracture

Specific regional radiology – LeFort

Specific regional radiology – Facial injuries

• Frontal sinus fractures
  • Often compound with risk of intracranial infection
  • May be associated intracranial haematoma or cerebral contusion
  • CT can evaluate involvement of posterior sinus wall
  • Often require surgical exploration

Specific regional radiology – Cervical spine

• 2-6% incidence in blunt trauma
• If posterior midline neck tenderness, altered mental status or neurological deficit
  • 9.5% incidence
• 5-10% of those with traumatic brain injury suffer cervical spine injury
• Plain X-ray misses 12-16% of injuries
• MDCT at Alfred with 1mm cuts from C0-3 and 3mm cuts from C4-T4/5 missed no injuries and found that passive flexion/extension plain films added nothing
• The role of routine clearance after normal MDCT without focal neurology in unconscious patients is not clear but most authors regard this as enough (depends on clinical suspicion based on mechanism)
  • If pressure areas arise from rigid collar (31% incidence in unconscious patients) median LOS increases from 10 to 23 days

Specific regional radiology
 – Cervical spine

• NEXUS
  • 99.6% sensitive and 12.9% specific
  • Also found C-spine X-rays missed 1/3 of secondary spinal injuries (up to 25% of these were non-contiguous with the original injury) and confirmed the benefits of MDCT
• Canadian C-spine rule
  • Well-validated in prospective cohort study
  • 100% sensitivity and 42.5% specificity with good inter-observer agreement
• Should apply both with caution in elderly, spinal disease and the very young (<2yo)
  • At high risk of spinal injury with minor mechanism

Specific regional radiology
 – Cervical spine

• Plain film
  • Need adequate views (C0-T1), alignment and inspect bones, cartilage and soft tissues
    • 4 lordotic lines: Anterior and posterior vertebral lines, spinolaminar line and spinous process line
    • In adults, up to 1mm of anterior subluxation (and up to 3mm in children) is allowed
      • Pseudosubluxation is most common up to 8yo but can be seen up to 18. Most common at C2/3 (and C3/4 and C4/5 less commonly) with key radiological feature being preservation of the spinolaminar line in flexion/extension views
    • AP diameter of spinal canal should be measured
      • At C2 should be >14mm
      • At C7 should be >12mm
    • Should look at vertebral body, pedicles, facet joints, laminae, spinous processes and interspinous distances for all vertebrae systematically
    • Interspinous spaces should all be equal
    • Prevertebral soft tissues
      • >7mm at C2 and >22mm at C6 in adults indicates prevertebral haematoma
      • Upper prevertebral space larger in children due to lymphoid tissue

Specific regional radiology
 – Cervical spine

Specific regional radiology
 – Cervical spine

• Atlanto-occipital dislocation – Usually fatal
• Atlanto-occipital subluxation – Usually survive and subtle
  • Diagnosis based on abnormal basion-axial interval and/or basion-dental interval (RULE OF 12’s)
  • Usually total =<12mm
  • Space between odontoid and anterior arch of C1 should be <3mm in adults (<5mm in children)
    • If larger, suggests traverse atlantal ligament disruption
    • Can be ‘normally’ enlarged in RA or Down’s syndrome
• Odontoid view
  • Lateral masses should be inspected for alignment (abnormal in Jefferson fracture, odontoid fracture and rotatory subluxation of C1 on C2
• AP view
  • Malalignment of spinous processes may suggest unilateral facet dislocation
  • Look at all vertebral bodies for height

Specific regional radiology
 – Rule of 12’s

Specific regional radiology
 – Thoracolumbar spine

• Second most commonly injured area of the spine is the thoracolumbar junction (T11-L2)
  • Cord injuries at the junction comprise 20% of all spinal cord injuries
  • Due to rigidly fixed thoracic spine and more mobile lumbar spine with narrow spinal canal (as compared to cervical and lumbar spine)
• Injuries to T1-T10 comprise 16% of cord injuries and lumbosacral injuries comprise 4% of spinal neurological injuries
• T1-4 can be difficult to visualise on plain film and it is routine to include T4-5 in cervical spine CT and to perform bony reconstructions of the thoracolumbar spine when CT chest/abdo/pelvis performed

Specific regional radiology
 – Thoracolumbar spine

• Three column theory (Dennis)
  • Anterior column: Anterior longitudinal limit and anterior half of vertebral body
  • Middle column: Posterior half of vertebral body, posterior LL
  • Posterior column: All bony and ligamentous structures posterior to PLL
• Anterior column only = Stable
• Anterior + middle column = Unstable
• All three columns = Unstable

Specific regional radiology
 – Thoracolumbar spine

Specific regional radiology
 – Thoracolumbar spine

• Stable fractures therefore include:
  • Transverse process fractures
  • Spinous process fractures
  • Pars interarticularis fractures
  • Wedge compression fractures involving only the anterior 2/3 of the vertebral body (although this is technically anterior and middle column involvement)
• Unstable fractures include:
  • Compression fractures with middle and/or posterior column disruption
  • Chance fracture
  • Burst fracture
  • Flexion/distraction injuries

Specific regional radiology
 – Chance fracture

Specific regional radiology
 – Burst fracture

Specific regional radiology
 – Thoracolumbar spine

• If fused thoracic spine (e.g. ankylosing spondylitis, diffuse interstitial spinal hypertrophy (DISH) and advanced degenerative disc disease) can result in ‘carrot stick’ fracture
  • Typically through disco-vertebral junction
  • Typically due to hyperextension and usually involve all three columns = unstable
  • Cord damage is common
• If CT of the torso IS NOT performed, plain X-rays of the T/L spine should be taken with limited CT of areas difficult to visualise or where there is clinical suspicion
• Chance fracture
  • Flexion/distraction injury with lap belt of axis of rotation
  • Failure of spinal column in posterior elements
  • Fractured directed in horizontal plane through entire bony column = three columns and inherently unstable
  • May also have associated fractures to duodenum or pancreas

Chest trauma

• Chest CT more sensitive for:
  • Lung contusions, pneumothoraces, mediastinal haematoma, fractured ribs/scapula/sternum/vertebrae
• Chest CT more likely to provide additional information in:
  • Chest wall tenderness
  • Reduced AE
  • Abnormal respiratory rate
• CXR can be done rapidly and allows diagnosis of life-threatening injuries such as pneumothorax/haemothorax or malpositioned ETT or ICC
  • Haemothorax requires 800-1000mL to be visualised on supine X-ray (as opposed to 200-300mL in erect X-ray)
• CXR misses 50% of anterior and lateral rib fractures so directed mainly at complications such as PTX, haemothorax and lung contusion
• Fractures of clavicles, scapulae, glenohumeral joints and first 2 ribs suggests high-energy blunt trauma and high risk of underlying visceral/vascular injuries

Chest trauma

• 2% incidence of sternal fractures in MVA
  • 1.5% of these had arrhythmias due to cardiac contusion requiring treatment
  • Mortality rate of <1%
  • Those most at risk were >65yo and/or on digoxin therapy
• Subcutaneous emphysema
  • May be due to injury to lung, tracheobronchial tree, larynx, pharynx and/or oesophagus
  • Prompts careful examination for pneumothorax and/or pneumomediastinum
• Penetrating chest trauma
  • Pneumothorax may be delayed so warrants check CXR at 6 and 12 hours
• Lung contusion
  • Usually evident within 6 hours on CXR as opposed to aspiration and pulmonary infarction which are delayed up to 12-24 hours

Chest trauma

• CXR signs of tracheal laceration
  • Subcutaneous emphysema/pneumomediastinum/pneumothorax
  • Deviation of ETT tip to the right relative to tracheal lumen
  • Distension of ETT balloon
  • Migration of ETT balloon distally towards the tube tip (normally 2.5cm proximal to tip)
• Signs of diaphragm injury on CXR
  • Elevated hemidiaphragm
  • Bowel in chest
  • NG tube in chest
  • Collapse of lower lung fields
  • Displacement of mediastinum away from injury

Chest trauma

• Thoracic aortic injury
  • Mediastinal width (measured at top of aortic knob) >8-8.5cm in supine film (>6cm in erect film) is suggestive of mediastinal haematoma
  • Sensitivity 90% and specificity 10%
  • 7% of patients with aortic rupture have normal CXR
  • CXR signs
    • Widened mediastinum as above
    • Deviation of oesophagus/NG/trachea to the right of T4 spinous process
    • Obliteration of aortic knob
    • Increased right paratracheal stripe >4mm
    • Increased left paravertebral stripe >5mm
    • Opacification of aortopulmonary window
    • Depression of left main bronchus > 40 degrees from horizontal
    • Left apical cap

Right paratracheal stripe

Left paravertebral stripe

Chest trauma

• Oesophageal injury
  • Pneumomediastinum
  • Subcutaneous emphysema
  • Pneumothorax
  • Left pleural effusion
  • Widened mediastinum
• If suspected mediastinal major vascular injury, CT angiogram is preferred screening investigation (although DSA is gold standard)
  • 83-100% sensitive for aortic injury and NPV 99-100%

Chest trauma

• TOE for mediastinal investigation
  • Advantages include use at bedside, minimally invasive and low rate of complications
    • Can demonstrate myocardial, pericardial and valvular injuries not seen on MDCT
  • Disadvantages include need for sedation/intubation, limited information about distal ascending aorta, aortic arch and arch vessels
• Oral contrast studies
  • Useful in Ix of oesophageal and diaphragmatic injuries
  • Gastrograffin preferred for suspected oesophageal injury as less risk of mediastinitus than barium
    • Does however result in severe pneumonitis if aspirated so CANNOT be risk of this. If there is a risk of aspiration, IV contrast should be used orally
  • If unconscious, NG tube is placed at start of oesophagus and contrast provided
  • Flexible or rigid oesophagoscopy may be preferred

Abdominopelvic trauma

• Oral contrast for CT is of limited use in this setting given delayed gastric emptying and most contrast remains in stomach
• Injuries missed on CT include diaphragmatic injuries, pancrease, bladder and upper intestinal perforation
• CT gives some information on renal function and with contrast can provide information on extravasation

Abdominopelvic trauma
 – Spleen injury scale

Abdominopelvic trauma
 – FAST

• DLP has low specificity with non-therapeutic laparotomy rate of up to 30%
• FAST diagnostic modality of choice in unstable blunt trauma patient
• Limitations
  • Requires training
  • Cannot differentiate between blood, urine and ascites
  • Poor-quality images in obesity, subcutaneous emphysema and dilated bowel loops
• Done in resus room, non-invasive, no radiation exposure, should take 2-5 minutes and is repeatable
• Sensitivity 80-100%; specificity 88-100% for intraperitoneal free fluid
• Some evidence for benefit in penetrating trauma to diagnose pericardial or intraperitoneal fluid

Pelvic trauma

• Most anterior fractures seen on plain film
• Up to 30% of posterior fractures involving sacrum and sacroiliac joints will not be seen on plain film
• Acetabular fractures also often missed
• The greater the AP disruption of the pelvic ring, the greater the pelvic cavity volume and risk of haemorrhage shock and visceral damage
• See next slide for Young and Resnick classification scheme
  • Does not take into consideration isolated fractures outside the pelvic ring or acetabular fractures
• If pelvic ring is fractured anteriorly AND posteriorly, stability is usually lost, with disruption of the posterior ligament complex and widening of the sacroiliac joints on AP view

Young and Burgess classification

• AP compression
  • Type 1: Disruption of symphysis pubis with <2.5cm diastasis. No posterior pelvic injury.
  • Type 2: >2.5cm SP diastasis with tearing of anterior sacroiliac, sacrospinous and sacrotuberous ligaments
  • Type 3: Complete disruption of SP and posterior ligament complexes with hemipelvic displacement
• Lateral compression
  • Type 1: Posterior compression of SI joint without ligament disruption; oblique pubic ramus fracture
  • Type 2: Rupture of posterior sacroiliac ligament; pivotal internal rotation of hemipelvis on the anterior SI joint with crush injury of sacrum and oblique pubis ramus fracture
  • Type 3: As above with evidence of AP compression injury to contralateral hemipelvis
• Vertical shear
  • Complete ligament or bony disruption of a hemipelvis associated with hemipelvis displacement

Young and Burgess classification

CT scan of the pelvis

• If pelvic fracture seen and unstable, crucial to determine if intra-abdominal bleeding exists
  • If FAST excludes this, then patient should go to angiography and embolisation of pelvic arterial bleeding OR to operating theatre for X-fix if pelvic ring is widely disrupted
• If stable, CT is indicated to demonstrate posterior fractures and sacroiliac joint diastasis
• 3D images are particularly useful for assessment of acetabular and pubic bone fractures
• Contrast-enhanced MDCT can rapidly identifying ongoing arterial bleeding (blush/extravasation) to help direct embolisation techniques

Pelvic angiography

• If posterior aspect of pelvic ring is disrupted, considerable arterial and/or venous haemorrhage can occur
• Open-book (AP compression fractures) are more likely to have venous than arterial bleeding and compression by pelvic binder + X-fix should help to minimise blood loss
• Age >65, absence of long-bone fractures and haemodynamic instability necessitating urgent angiography all predict likelihood of arterial bleeding
• Angiographic embolisation has a success rate of >90% in controlling bleedings
  • Risk of impotence does exist but is rare

Contrast studies

• Urethrogram/cystogram to assess for urethral injuries and bladder rupture when suspected

Extremities

• 2-6% of blunt trauma patients have missed injuries
• More likely if closed head injury or intoxicated
• Commonly missed
  • Perilunate/lunate dislocations at wrist
  • Posterior shoulder dislocations
  • Knee/hip dislocations
  • Medial or lateral epicondylar fractures
  • Supracondylar fractures of elbow
  • Scaphoid and triquetrum fractures
  • Metacarpal and phalangeal fractures
  • Gamekeeper’s thumb
  • Tibial plateau/calcaneus fractures unless specifically sought
  • Lumbar spine fractures ( esp. if calcaneus fracture evident)

Extremities

• CT scans of major joints and calcaneus where fracture is suspected is often helpful for operative planning and confirmation
• MRI investigation of choice for knee ligamentous/meniscal injuries
• Angiography
  • Required if suspected vascular compromise to limb
  • Brachial plexus injuries need to be screened for in shoulder joint trauma

Last Updated on October 9, 2020 by Andrew Crofton