Emergency and Trauma · Head trauma · 1.0

Rotterdam CT score for traumatic brain injury

A six-point admission CT score for predicting six-month mortality in moderate to severe traumatic brain injury, developed as a successor to the Marshall classification.

Purpose

To grade admission head CT findings in moderate to severe traumatic brain injury into an additive prognostic score for six-month outcome, improving on the Marshall classification by scoring individual CT predictors separately rather than assigning a single anatomical category.

Component	Finding	Points
Basal cisterns	Normal	0
Basal cisterns	Compressed	1
Basal cisterns	Absent	2
Midline shift	No shift or shift 5 mm or less	0
Midline shift	Shift greater than 5 mm	1
Epidural mass lesion	Present	0
Epidural mass lesion	Absent	1
Intraventricular blood or traumatic subarachnoid haemorrhage	Absent	0
Intraventricular blood or traumatic subarachnoid haemorrhage	Present	1
Baseline constant	Added to every study	+1
Total score	Sum of all components	1 to 6

How to use it

Four CT variables are scored independently and summed, then a constant of 1 is added so that the total ranges from 1 to 6 and aligns numerically with the six grades of the Marshall classification.
Basal cistern status and midline shift are surrogates of mass effect and raised intracranial pressure, while intraventricular or traumatic subarachnoid haemorrhage marks diffuse vascular and shear injury.
The epidural component is scored inversely: absence of an epidural mass lesion attracts a point, because an epidural haematoma is an evacuable extra-axial lesion associated with better outcome than equivalent diffuse or parenchymal injury.
Six-month mortality rises progressively with increasing score, with the steepest gradient at the upper end of the range; the score is prognostic at group level and is not a surgical decision rule for an individual patient.
The score is applied to the admission CT; repeat imaging after deterioration or decompression frequently changes the components and does not carry the original prognostic estimate.

Common mistake

Scoring a point for the presence of an epidural haematoma. The point is awarded for its absence. A second frequent error is awarding separate points for intraventricular blood and traumatic subarachnoid haemorrhage, which together contribute a maximum of one point, and forgetting the mandatory baseline constant of 1, which makes a minimum score of 0 impossible.

Exam pearl

Cisterns 0-2, midline shift over 5 mm 1, ABSENT epidural lesion 1, IVH or traumatic SAH 1, plus 1 constant - total 1 to 6. Do not confuse with Marshall (six non-additive anatomical categories) or Helsinki (lesion type and volume based).

Viva questions

What is the Rotterdam CT score and what does it predict?: It is an additive admission head CT score, ranging from 1 to 6, derived by Maas and colleagues in 2005 to predict six-month mortality and outcome in moderate to severe traumatic brain injury.
Name the four CT components scored.: Basal cistern status, midline shift, presence or absence of an epidural mass lesion, and intraventricular blood or traumatic subarachnoid haemorrhage.
How are the basal cisterns and midline shift scored?: Basal cisterns score 0 if normal, 1 if compressed and 2 if absent. Midline shift scores 0 if absent or 5 mm or less, and 1 if greater than 5 mm.
Why does absence of an epidural haematoma score a point?: Because an epidural haematoma is an evacuable extra-axial lesion carrying a comparatively better prognosis, so its absence implies the mass effect arises from diffuse or parenchymal injury, which predicts worse outcome.
Why is a constant of 1 added?: It shifts the range from 0 to 5 up to 1 to 6, so the score maps numerically onto the six grades of the Marshall classification and no patient scores zero.
How does the Rotterdam score differ from the Marshall classification?: Marshall assigns a single anatomical category based on diffuse injury grade and evacuated versus non-evacuated mass lesions, whereas Rotterdam scores individual CT predictors independently and sums them, handling mass lesions and subarachnoid or intraventricular blood better.
Which CT should be scored?: The admission scan. Repeat imaging after clinical deterioration or after decompressive surgery alters the components and does not carry the original prognostic estimate.
How should the score be used clinically?: As a group-level prognostic descriptor for outcome prediction, research stratification and audit. It is not a triage rule or an indication for surgery in an individual patient, and management decisions rest on clinical status and intracranial pressure.