CT head (protocol)

CT head (sometimes termed CT brain), refers to a computed tomography examination of the brain and surrounding cranial structures. It is most commonly performed as a noncontrast study, but the addition of a contrast-enhanced phase is performed for some indications.

This article covers non-contrast and delayed post-contrast imaging. Other specific types of contrast-enhanced CT for cerebrovascular evaluation are discussed separately: CT venography, CT angiography, and CT perfusion.

Indications

The following are common indications for which noncontrast head CT (CT head without intravenous contrast) is usually appropriate ^1,2:

• altered mental status in specific scenarios:
  • known intracranial hemorrhage, mass, infection, or infarct
  • coagulopathy or anticoagulation
  • suspected central nervous system infection
  • suspected elevated intracranial pressure
  • hypertensive emergency
  • acute delirium
  • unknown cause
• cerebrovascular disease in specific scenarios:
  • intracranial hemorrhage suspected or proven
  • dural venous sinus thrombosis suspected
  • ischemic stroke suspected due to focal neurological deficit (new, fixed, or worsening)
  • transient ischemic attack, initial screening
• dementia, initial imaging
• head trauma in specific scenarios:
  • acute head trauma, initial imaging:
    • mild (GCS 13-15) but imaging indicated by clinical decision rule
    • moderate or severe (GCS <13)
    • penetrating trauma
  • acute head trauma, short-term follow-up imaging:
    • positive findings on initial imaging
    • new or progressive neurologic deficits
  • subacute or chronic head trauma with unexplained cognitive or neurologic deficits
  • recent head trauma with suspected cerebrospinal fluid (CSF) leak, initial imaging
  • recent head trauma with acute ataxia
  • pediatric abusive head trauma suspected due to the presence of neurologic signs or symptoms, apnea, complex skull fracture, other fractures, or other injuries highly suspicious for child abuse
• headache in specific scenarios:
  • sudden and severe headache (worst headache of life, thunderclap headache)
  • new headache with papilledema
  • new or worsening headache in the following scenarios:
    • subacute head trauma
    • inciting activity/event such as sexual activity, exertion, or position
    • neurologic deficit
    • known or suspected cancer
    • immunocompromise
    • pregnancy
    • at least 50 years old
• seizures in specific scenarios:
  • epilepsy disorder with change in clinical symptoms or seizure pattern
  • new seizure, initial imaging

Additional indications for noncontrast head CT include the following ²:

• surgery-related indications
  • surgical guidance or preoperative planning
  • postoperative evaluation after intracranial surgery
  • evaluation for CSF shunt malfunction
• skull lesions (such as craniosynostosis, fibrous dysplasia, Paget disease, tumors)
• detection or evaluation of calcification

The administration of intravenous contrast media may improve the sensitivity for detecting brain neoplasms or infections. CT head without and with contrast can be performed for these indications if MRI, which is generally superior for these diagnoses, is contraindicated or unavailable.

Purpose

The purpose of noncontrast head CT is to evaluate for neurosurgical emergencies with high sensitivity, including acute intracranial hemorrhage, mass effect, territorial infarct, or hydrocephalus. Due to its widespread availability, CT is more often performed than MRI in the acute setting. In addition, CT is superior to MRI for evaluating osseous structures, such as for calvarial or skull base fractures or craniosynostosis.

Contrast-enhanced CT allows the identification of abnormal contrast enhancement, such as in brain metastases, some primary brain tumors, and brain abscesses. 

Technique

• patient position
  • supine with their arms by their side
• scout
  • C2 to vertex
• scan extent
  • C2 to vertex
  scan direction
  • caudocranial 
• scan delay
  • minimal scan delay 
• ​respiration phase
  • ​suspended

Practical points

The technique for performing a CT of the head depends on the scanner available and fall into two broad camps: 

• step-and-shoot (sequential)
• volumetric acquisition (helical) 

Historically, only axial planes were obtained. More recently it has become standard practice to obtain volumetric/helical scans, with subsequent multiplanar reconstructions.

In addition to various planes, the images can also be reconstructed using different algorithms (e.g. bone algorithm or soft-tissue algorithm) and viewed with different windows (e.g. brain window, subdural window, or bone window) to emphasize various tissue characteristics.

Step-and-shoot

Step-and-shoot scanning was the first described technique but has largely been superseded in more modern scanners in favour of helical scanning and volumetric datasets (see below)

In step-and-shoot scanning, the CT tube/detector perform a complete resolution around a stationary patient (shoot) generating a single axial image. The table/gantry then advances a specific distance (step) and the process is repeated to acquire the next axial image.  Traditionally, these were 10 mm slices through the cerebrum and 5 mm slices through the base-of-skull and posterior fossa. 

Initially, scanners were fixed such that the scanning plane was at right angle to the floor. The initial axial plane described for CT brain was the orbitomeatal line. Relatively soon, however, the ability to tilt the scan plane became possible and the standard plane was then shifted to one parallel to the orbital roof. This had the advantage of avoiding the lens (at least in some patients) and reduce the artefact from dental fillings which would be projected below the posterior fossa. 

Volumetric acquisition

More recently, CT scanners have become able to obtain a complete volume of scan data by continuously scanning as the gantry is moved. This generates a helical scan path through the patient (thus helical scanning). The advantage of this technique is that it generates a complete 3D volume of data which in turn allows the creation of multi-planar reconstruction (MPR) with thick or thin slices using different algorithms.

The axial plane can then be chosen to match any desired plane, regardless of the position the patient's head was in when scanned. Increasingly, the standard axial plane is being set to match that of MRI scans often parallel to thetuberculum sellae-occipital protuberance line (which is close to parallel to the AC-PC line). Coronal and sagittal reconstructions are then usually at right angles to this. 

The ability to create MPRs quickly and easily does result in a significant increase in the number of images to be reviewed and the amount of space required to save them to disc. 

It is not inconceivable to see a CT brain resulting in 3-plane 4 mm soft and 3-plane 1 mm bone reconstructions being sent to PACS with a 3D reformat and even the 0.6 mm overlapping data (to allow reimport into the volume rendering system for future use). We have moved from the traditional 30 images for a CT brain to a mammoth set of data with 1000+ images.