Traumatic intracranial injury

CT

First and second year — the floor first, then every step

Read the head CT in a fixed order — extra-axial blood, brain, ventricles and cisterns, bone — and measure the three numbers the neurosurgeon acts on: thickness, midline shift and cistern status.

Orient first

  • Acute blood is hyperdense (roughly 50–70 HU) only for the first days; an anaemic patient, a coagulopathic patient or a hyperacute bleed can be isodense. Swirling mixed density inside a haematoma means active bleeding into it.
  • Extradural blood is biconvex and stops at sutures (dura is bound there) but can cross the midline and the tentorium; subdural blood is crescentic, crosses sutures, and stops at the falx and tentorium.
  • Mass effect kills before the blood does. The basal cisterns (ambient, quadrigeminal, suprasellar) are the pressure gauge: open, compressed or effaced.

Acquire the study

  • Non-contrast CT head: thin axial reconstructions with SAGITTAL and CORONAL reformats — a thin subdural along the tentorium or the vertex is seen on coronals and missed on axials.
  • Three windows every time: brain (~W80/L40), a narrow SUBDURAL window (~W150–200/L50–80), and BONE on the thin-slice kernel.
  • Include the craniocervical junction; add CT cervical spine per the trauma criteria in use, and CTA when the skull-base fracture pattern or mechanism meets BCVI screening criteria.
  • MRI is not the acute study; it is for unexplained deficit with a near-normal CT (diffuse axonal injury: SWI and DWI), or for the brainstem.

The manoeuvre

  • Extra-axial first, on the SUBDURAL window: convexities, along the falx and tentorium, the middle cranial fossa floor and the posterior fossa. Measure the maximum THICKNESS of any collection perpendicular to the inner table, in mm.
  • Measure MIDLINE SHIFT in mm at the level of the septum pellucidum (foramen of Monro), from the ideal midline drawn between the anterior and posterior falx attachments.
  • Grade the BASAL CISTERNS: open, compressed, or absent — and check the fourth ventricle and the ambient cisterns for posterior fossa mass effect.
  • Parenchyma: contusions at the classic coup and contrecoup sites (inferior frontal lobes, anterior temporal poles), and small haemorrhages at the grey–white junction, corpus callosum and dorsolateral brainstem (diffuse axonal injury).
  • Subarachnoid blood in sulci and the sylvian fissures, and intraventricular blood in the occipital horns (dependent layering).
  • Herniation: subfalcine (cingulate under the falx), uncal (temporal horn dilatation on the opposite side, effaced suprasellar cistern), tonsillar, and upward.
  • Bone window: the fracture line, its depression in mm relative to the inner table, whether it crosses a VENOUS SINUS or the middle meningeal groove, and pneumocephalus.

What confirms it

  • Extradural haematoma: biconvex, bounded by sutures, usually under a fracture across the middle meningeal groove or a venous sinus.
  • Subdural haematoma: crescentic, crosses sutures, limited by the dural reflections; acute is hyperdense, subacute can be isodense and is found by the displaced grey–white junction and effaced sulci.
  • Traumatic SAH is convexity-predominant beside contusions; blood filling the basal cisterns with little else should raise an aneurysmal bleed that caused the fall.

What licenses you to exclude it

  • A normal CT does not exclude diffuse axonal injury — if the deficit is out of proportion to the CT, say so and recommend MRI with SWI and DWI.
  • A normal early CT does not exclude a delayed bleed or blossoming contusion in an anticoagulated patient; state the time from injury and the anticoagulation status if known.

The classic misread

  • Reading only the brain window: a thin acute subdural against the inner table is invisible without the subdural window.
  • Calling a hyperdense tentorium or falx normal on an axial slice without checking the coronal reformat.
  • Missing an isodense subacute subdural — look for sulci that do not reach the inner table and a buckled grey–white junction.
  • Measuring midline shift at the pineal or the third ventricle and quoting it as the septum pellucidum shift.

Reporting the injury

Classification to use

  • Name each lesion type (EDH, SDH, tSAH, IVH, contusion, DAI) — management follows the lesion, not a composite score.
  • Marshall CT classification (diffuse injury I–IV, evacuated / non-evacuated mass lesion) and the Rotterdam CT score (cisterns, shift, EDH, IVH/tSAH) for prognosis and research; state which one you use.

Measurements — and how to take them

  • Haematoma thickness: maximum, perpendicular to the inner table, in mm on the slice where it is thickest.
  • Haematoma volume (focal lesions): ABC/2 — the three orthogonal maximum diameters in cm, multiplied, halved, in mL.
  • Midline shift: perpendicular distance of the septum pellucidum from the line joining the anterior and posterior falx attachments, in mm.
  • Depressed fracture: depth of the most depressed fragment below the inner table of the adjacent vault, in mm.

What to report

  • Each haematoma: type, side, location, maximum thickness in mm (and volume if a lesion is focal), and whether it is mixed-density (active bleeding).
  • Midline shift in mm at the septum pellucidum; state of the basal cisterns; any herniation by type.
  • Hydrocephalus, including trapping of a temporal horn.
  • Skull fractures: depressed (by how many mm), compound (overlying scalp laceration/gas), involving a venous sinus or the skull base.

How to report it

  • CT: "Acute right frontoparietal subdural haematoma, maximal thickness 12 mm, with 8 mm leftward midline shift at the septum pellucidum and effacement of the right ambient cistern."
  • CT: "Mixed-density content within the collection is in keeping with active haemorrhage."
  • MRI: "Multiple foci of susceptibility at the grey–white junction of both frontal lobes and in the splenium of the corpus callosum, in keeping with diffuse axonal injury."

What not to report

  • Do not describe a haematoma as "small" or "trace" without a number; the thickness is what is acted on.
  • Do not attribute a basal-cistern-predominant SAH to trauma without saying an aneurysmal cause has not been excluded.
  • Do not quote a Marshall or Rotterdam grade instead of the measurements — give both, or the measurements alone.

Associated injuries to look for

  • Cervical spine injury — image the cervical spine by the rule in use in every significant head injury.
  • Blunt cerebrovascular injury — a skull-base fracture through the carotid canal or a petrous fracture meets screening criteria for CTA.
  • Venous sinus thrombosis or laceration under a fracture crossing a sinus — CT venography.
  • Orbital and facial fractures; CSF leak with pneumocephalus.

What changes management

  • Brain Trauma Foundation surgical guidance (2006) is commonly applied: EDH > 30 mL, or acute SDH > 10 mm thick or with > 5 mm midline shift, is an indication for evacuation regardless of GCS — verify against the local neurosurgical protocol.
  • Effaced basal cisterns, new hydrocephalus, or a posterior fossa mass lesion with fourth-ventricle effacement: urgent neurosurgical communication.
  • Depressed fracture deeper than the thickness of the vault, compound, or involving a sinus.
  • Any expanding lesion on a repeat CT — state the interval change in mm.

Reference values

Each value carries the caveat that keeps it from being misused. Normal limits and diagnostic criteria are kept apart on purpose: a disease cut-off read as a normal range is the more dangerous mistake.

Diagnostic criteria

  • Extradural haematoma · EDH volume at the surgical-evacuation threshold

    an extradural haematoma over 30 cm³ meets the conventional evacuation criterion regardless of GCS

    Volume is estimated by the ABC/2 method (see the ICH entry for the technique). Temporal location, lucid-interval history and a swirl sign (active bleeding) all argue for urgency below the volume threshold. Versioned criterion — verify against the current edition before clinical use.

    CT

  • Subdural haematoma · Acute SDH thickness and shift at the surgical-evacuation thresholds

    an acute subdural over 10 mm thick, or with midline shift over 5 mm, meets the conventional evacuation criteria regardless of GCS

    These are DECISION thresholds for the neurosurgical conversation, not automatic verdicts — GCS trend, pupillary signs and comorbidity sit alongside them. Measure the maximal thickness perpendicular to the inner table, and the shift at the septum pellucidum. Versioned criterion — verify against the current edition before clinical use.

    CT

  • Extra-axial collections (density) · Expected density time-course — how to report it

    a named qualitative course: acute blood is hyperdense, becomes isodense over days to a couple of weeks, then hypodense when chronic — exact timing varies with haemoglobin and CSF admixture

    Anaemia and hyperacute unclotted blood are less dense; a subdural of mixed density may be acute-on-chronic or active bleeding, not a precise clock. Date cautiously.

    CT

  • Midline structures · Midline shift — how to report it

    none; any shift is measured and stated in mm

    Measure at the level of the foramen of Monro, from the septum pellucidum to a line between the anterior and posterior falcine attachments.

    ⚠️ THE LEVEL IS PART OF THE MEASUREMENT — a shift measured at a different level is not comparable with the prior study, and serial comparison is the reason the number exists. State the level.

    CT · MRI

  • Intracerebral haemorrhage · Haematoma volume by ABC/2

    volume in cm³ approximated as A × B × C ÷ 2; 30 cm³ is a widely used severity and decision inflection

    A = largest axial diameter of the haematoma; B = the diameter perpendicular to A on the same slice; C = the craniocaudal extent (number of slices × slice thickness). All in centimetres.

    ABC/2 systematically over-estimates irregular and anticoagulation-related haematomas — say "approximately". Location changes the meaning of any volume: 10 cm³ in the posterior fossa is a different emergency from 10 cm³ in the frontal lobe.

    CT

See it on real cases

Direct links to Radiopaedia — the reference article and worked cases with their images. Each opens on Radiopaedia.

Key papers

Reviews and guidelines from RSNA, ESR and related journals. Each opens at its DOI.

  1. Traumatic Brain Injury: Imaging Patterns and Complications ↗Schweitzer AD, Niogi SN, Whitlow CT, et al. · RadioGraphics 2019RSNA · PubMed
  2. MDCT imaging of traumatic brain injury ↗Lolli V, Pezzullo M, Delpierre I, et al. · British Journal of Radiology 2016BIR · PubMed
  3. Imaging evidence and recommendations for traumatic brain injury: conventional neuroimaging techniques ↗Wintermark M, Sanelli PC, Anzai Y, et al. · Journal of the American College of Radiology 2015ACR · PubMed
  4. Imaging evidence and recommendations for traumatic brain injury: advanced neuro- and neurovascular imaging techniques ↗Wintermark M, Sanelli PC, Anzai Y, et al. · AJNR 2015ASNR · PubMed
  5. MR Imaging Applications in Mild Traumatic Brain Injury: An Imaging Update ↗Wu X, Kirov II, Gonen O, et al. · Radiology 2016RSNA · PubMed

More searches

More in Trauma