Imaging of Head Trauma

Introduction

Traumatic brain injury (TBI) is a major cause of mortality and morbidity. In England and Wales, approximately 1.4 million patients per year attend hospital following head injury and it is the most common cause of death under the age of 40 years. In the following chapter, the imaging techniques best placed for the initial assessment of these patients and the complications of trauma are introduced. Imaging manifestations of the sequelae of trauma are then individually discussed.

Modality and Technique

The aim of imaging at initial presentation of an acute trauma patient is to depict the primary injuries and their extent and to determine if the patient is a potential candidate for neurosurgical intervention. For example, this includes the identification of active bleeding, intra/extra-axial collections, hydrocephalus, brain herniation and depressed skull fractures. Additional imaging for secondary injuries may also be required during both the acute/subacute setting for complications that include the development of brain oedema, ischaemia/infarction and delayed haemorrhage.

Computed tomography (CT) remains the first-line imaging modality in the acute trauma setting given its availability, scanning speed and efficiency and is usually performed without contrast medium being administered. This is of benefit in polytrauma patients where the brain, whole spine and body can be imaged within a short time interval. Furthermore, it provides imaging with excellent spatial resolution and slice thickness imaging of 0.75 to 1 mm that enables the axial imaging to be reformatted into sagittal and coronal imaging planes.

CT head imaging first involves the acquisition of a scout radiograph to plan the subsequent imaging. The scout radiograph is akin to a skull plain radiograph and can provide useful information with regard to linear calvarial fractures that may be harder to identify in the axial plane. It is of use to identify foreign bodies and the scout radiograph may include the mandible and craniocervical junction/upper cervical spine that can prompt additional facial bone/cervical spine imaging if there are any abnormalities identified ( Fig. 54.1 ). The CT head imaging is performed with the patient in the supine position from the skull base to the vertex and a helical acquisition is obtained. From the initial source imaging, a number of reconstructions can be generated, including brain imaging varying from 1 and 5 mm slice thickness that can be used to generate a combination of axial, coronal and sagittal images. Furthermore, a dedicated bone algorithm is typically used to depict the bony anatomy of the calvarium, skull base and facial bones.

On viewing the CT imaging, knowledge of how to optimise the imaging and contrast is essential. To assess the brain parenchyma, thicker slice images (typically approximately 5 mm) when viewed on a narrow-tissue window setting (e.g. window level 40 Hounsfield units [40 HU], window width 80 HU) enables accurate assessment of the brain parenchyma with reduced levels of noise. To assess for acute haemorrhage (parenchymal contusions, subarachnoid/intraventricular haemorrhage, subdural and extradural haematomas) a wider window setting is used (for example, window level 70 HU, window width 150 to 200 HU). This accentuates the contrast between blood and bone and is of benefit when assessing for small extra-axial haemorrhages that may be occult on thicker-slice imaging and with standard brain imaging window settings ( Fig. 54.2 ). Reviewing the coronal reformatted imaging aids identification of haemorrhage in relation to the vertex, falx cerebri and, particularly, the tentorium cerebelli and posterior fossa (see Fig. 54.2 ). The coronal imaging is also of benefit in assessment of trans-compartmental shift (e.g. subfalcine, uncal and tonsillar herniation syndromes) whilst the sagittal imaging is of particular use to see the pituitary fossa, central skull base/clivus and craniocervical junction. A major limitation of CT imaging is in the assessment of the subfrontal, infratemporal and posterior fossa regions given the artefact generated at the interface between the brain parenchyma and skull base. Unfortunately, these are also common sites of traumatic parenchyma injury and extra-axial haemorrhage that require a more meticulous assessment of thinner-slice imaging and in multiple planes.

CT imaging provides exquisite detail in terms of bony anatomy and this enables assessment in multiple planes. This can be combined with three-dimensional (3D) surface display to aid in the identification of fractures, particularly non-displaced linear fractures and sutural diastasis ( Fig. 54.3 ). CT head imaging will typically include the calvarium, skull base, orbits, temporal bones, craniocervical junction and, potentially, the facial bones including the mandible, depending on the extent of coverage. Assessment of CT imaging on dedicated bone windows also enables accurate assessment of foreign bodies, particularly in the context of penetrating trauma.

CT imaging assessment of the acute trauma patient can also be combined with vascular imaging of the intra- and extra-cranial vessels. This involves the administration of contrast preferably with a pump injector via a large-bore cannula (e.g. 18G cannula). Image acquisition is from the aortic arch to the vertex, which is typically triggered automatically when the density of contrast in a region of interest over the aortic arch reaches a certain threshold (typically 100 HU). A combination of both thin (0.75 mm and 3 mm) axial imaging is acquired from which a combination of axial, coronal and sagittal maximum-intensity projections can be produced to aid depiction of the intra- and extra-cranial vessels. CT angiographic imaging provides information tailored to the arterial vascular tree; however, depending on timing of acquisition and cardiovascular dynamics, a degree of venous opacification can occur that enables assessment of the dural venous sinuses and cortical veins. Nevertheless, if there is a high index of suspicion of a venous injury, a CT venogram can also be acquired with the same bolus of contrast, with an imaging delay of approximately 45 seconds. As per the arterial CT angiogram, maximum-intensity projections in multiple planes can be generated. CT angiography has the advantage of accessibility and speed as well as providing excellent spatial resolution. It is less prone to flow artefacts when compared with magnetic resonance imaging (MRI), particularly around the skull base that is a common site of vascular injury.

Indications for CT vascular imaging are guided by the trauma team, taking into account the mechanism of injury and the context of other injuries in a polytrauma patient. Typical indications for vascular imaging include penetrating head and neck trauma, central skull base (e.g. clival) and craniocervical junction fractures, cervical spine facet dislocation and fractures involving the foramina transversaria, the carotid canal and jugular foramina/dural venous sinuses. In addition, vascular imaging can be considered in the absence of skull base/craniocervical fractures if the patient develops focal neurological signs that suggest a potential stroke/parenchymal infarction.

MRI is more sensitive than CT in the detection of all stages of haemorrhage. The combination of routine brain MRI sequences (particularly the fluid-attenuated inversion recovery [FLAIR] sequences) and blood-sensitive sequences (either gradient-recalled echo [GRE] or susceptibility weighted sequences [SWI]) can accurately delineate the extent of intra-cranial haemorrhage and traumatic parenchymal injuries ( Fig. 54.4 ) (see sections ‘traumatic intra-cranial haemorrhage’ and ‘traumatic injury to the brain’ for further detail). Unlike CT, it is not prone to beam hardening artefact and enables more accurate assessment of the subfrontal, infratemporal and posterior fossa regions. Furthermore, it is more sensitive to parenchymal injuries, including shear-type injuries and haemorrhagic/non-haemorrhagic axonal injury. MRI can also be used to assess the intra- and extra-cranial vasculature with either a time-of-flight (TOF) or contrast-enhanced MR angiography. The angiographic imaging can also be combined with fat-suppressed sequences to enhance the detection of vascular dissections and injury of the surrounding soft tissues.

The main limitations of MRI in the acute trauma setting are the practical implications of accessibility and patient monitoring whilst in the scanner, particularly in the context of polytrauma. MRI sequences are more time consuming and require the patient to lie still whilst the imaging is acquired.

Mechanism of Injury

Traumatic head injuries can be classified as open or closed head injuries. An open head injury is used to describe an injury whereby the intra-cranial contents are exposed or herniate through a scalp/calvarial defect with direct communication between the intra- and extra-cranial contents ( Fig. 54.5 ). As such there is an increased risk of pneumocephalus and of subsequent cerebrospinal fluid (CSF) leaks and infection (cerebritis, meningitis, subdural empyema/extradural collections). A closed injury does not result in communication between the intra- and extra-cranial contents.

The mechanism of trauma can be broadly divided into blunt/non-missile and penetrating/missile trauma. Blunt trauma is more common and is typically the result of road traffic accidents (either as a driver or pedestrian) and falls, the latter being more common in the elderly (aged >75 years) or in the young (aged <4 years). Blunt trauma may result in a direct blow to the head and injury to the adjacent calvarium and brain parenchyma (coup site). Additional acceleration/deceleration, tensile and rotational forces may also result in parenchymal injuries opposite to the coup site and result in contre-coup injury ( Fig. 54.6 ) (see section on ‘ contusional brain injury ’ for further detail). Significant brain parenchymal injury can also occur with indirect trauma in the absence of a direct blow to the head. This is most often observed in the setting of high-energy trauma such as a road traffic accident, whereby the brain parenchyma can be exposed to significant acceleration, deceleration and rotational forces creating shear-type injuries (see section on ‘ diffuse axonal injury (DAI) ’). In this setting, there may be no extra-calvarial soft tissue swelling or calvarial/skull base fracture.

Penetrating/missile trauma is used in the context of ballistics injuries and knife/stab injuries as well as penetrating bone fragments. In this setting, a close working relationship with the acute trauma team is required to determine the exact mechanism of injury and relevant entry and exit wounds. CT imaging is the modality of choice given its ability to demonstrate radio-opaque foreign bodies and more accurately delineate the injury pathway in penetrating trauma.

Primary Injury

Sequelae of head trauma can be categorised into those that occur at the time of the initial trauma (primary injury) or those that occur later on (secondary injury). In the following section, primary injuries are considered.

Calvarial/Skull Fractures

Identification of a calvarial fracture is aided by the assessment of the scalp soft tissues in the first instance, which may indicate sites of direct injury (see Fig. 54.6 ). Scalp lacerations are evident clinically and can be correlated with a discontinuity of the cutaneous/subcutaneous soft tissues and may be associated with foreign bodies. The presence of facial/preseptal lacerations should prompt imaging of the facial bones if not already performed. Extracalvarial haematomas are common and often indicate the site of a coup injury/direct impact that necessitates assessment of the underlying calvarium and the contre-coup site within the intra-cranial compartment. Subgaleal haematomas refers to haemorrhage that collects beneath the aponeurosis of the occipitofrontalis muscle and is external to the periosteum and thus not confined to sutures ( Fig. 54.7 ). Cephalohaematomas are associated with instrumented deliveries and, as opposed to subgaleal haematomas, haemorrhage occurs between the skull and periosteum.

There are many different types of calvarial fractures that can be broadly divided into simple and comminuted fractures. Simple calvarial fractures include linear fractures of the inner/outer tables of the skull as well as diastatic fractures that may involve cranial sutures or synchondroses (see Fig. 54.3 ). Children are at higher risk of diastatic fractures typically involving the spheno-occipital, petro-occipital and occipitomastoid sutures. Comminuted fractures describe fractures with more than two fragments and may be associated with elevation (typically a sharp penetrating injury like a machete) or depression secondary to a high-energy direct blow or penetrating injury (see Fig. 54.5 ). Depressed fractures are associated with higher rates of underlying traumatic brain parenchymal injury ( Fig. 54.8 ). Open calvarial fractures are synonymous with open head injuries whereby the intra- and extra-cranial contents communicate and there is a higher risk of subsequent infection and CSF leaks. Assessing CT head imaging on lung windows is the best method to identify pneumocephalus and, given the location, this may imply that the fracture has involved the sinuses and temporal bones or has resulted in an open fracture of the calvarium (see Fig. 54.8 ).

A leptomeningeal cyst or growing skull fracture is a separate entity and typically occurs in the paediatric population in children aged less than 3 years. It is considered to be secondary to a skull fracture that results in a dural tear that allows the leptomeninges and/or brain parenchyma to herniate through the dura. CSF pulsation results in an enlarging cleft and bony defect, eroding the fracture margins, and results in non-union of the fracture site.