Estimating the volume of intracranial hematoma on CT

The commonly used formula for estimating the volume of intracranial bleed on CT is

ABC/2 

where

A is the maximal diameter of the hematoma by CT
B is the diameter 90° to A, and
C is the approximate number of CT slices with hemorrhage multiplied by the slice thickness

For a number of years, I could not really grasp why the formula is as such. I thought that the formula is derived from the formula of volume for a cylinder. The part of (C) in the formula I could understand as it involves the height of the cylinder. But what I could not grasp is that, if this formula is derived from the volume of a cylinder, the formula should entail the use of "pi" (π) which is taken to be 3.14 (up to 2 decimal points).  This is because the surface area of a circle is π * r *r; and therefore, the formula for volume of cylinder is π * r *r * h where h is the height of the cylinder.

Until I found this article Kothari et al (1996) which is helpful in explaining how this formula of ABC/2 comes about.

This formula is actually derived from the volume of an ellipsoid object and NOT of cylinder.

Figure 1

Image copyrighted to JoshDif licensed under under the Creative Commons Attribution-Share Alike 3.0 Unported license. Original site: Wikipedia

The volume for an ellipsoid object is

4/3 * π * a *b * c (a, b, and c as of Figure 1 above)

Applying this formula to the hematoma object on CT,
a and b are actually half of the two diameters mentioned on 2 dimension plane; and c is half of the height of the hematoma.

Therefore, the volume of the hematoma is

4/3 * π * (A/2) * (B/2) * (C/2)

where
A is the maximal diameter of the hematoma by CT
B is the diameter 90° to A, and
C is the approximate number of CT slices with hemorrhage multiplied by the slice thickness

assuming π ~ 3, therefore,

the volume of the hematoma is therefore,
4/3 * 3 * (A/2) * (B/2) * (C/2) = ABC/2

In an article by Freeman et al (2008), it is found that the ABC/2 method accurately estimates smaller ellipsoid hematoma volumes but inaccurately measures larger, irregularly shaped hematoma, or multicompartment hemorrhage such as intraventricular hemotoma and subdural hematoma. This article by Freeman et al (2008) has great pictures showing the ellipsoid shape of intracranial hematoma.

The importance of estimating the volume is that if the volume is large (~20 - 30 cc), it may be one of the indications for neurosurgical intervention, depending on the local neurosurgical management protocol of the center.

References:
1. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, Khoury J. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996 Aug;27(8):1304-5. Click here.

2. Freeman WD, Barrett KM, Bestic JM, Meschia JF, Broderick DF, Brott TG. Computer-assisted volumetric analysis compared with ABC/2 method for assessing warfarin-related intracranial hemorrhage volumes. Neurocrit Care. 2008;9(3):307-12. Click here.

Grey-Turner Sign

Grey-Turner’s sign simply refers to the bluish discoloration of the flanks. The interesting thing about this sign is that whenever this sign is found, medical students are alerted to the fact that there is a possible underlying retroperitoneal bleeding going on.  Obviously, this sign could also indicate a possible intraperitoneal bleed besides the possibility of retroperitoneal bleed. The source of bleed could be traumatic or non-traumatic, as in hemorrhagic pancreatitis.

However, when British surgeon George Grey Turner (1877-1951) first described it in 1920, in the British Journal of Surgery, it was described as a sign of hemorrhagic pancreatitis.

Not many know about the pathophysiologic basis of this sign, however.  It is actually due to the action on the abdominal wall and skin of leaking extravasated pancreatic juice from the hemorrhagic or necrotizing pancreatitis tracked subcutaneously. It could also be the blood collection tracked subcutaneously from retroperitoneal organs in the flank region.

In this picture, CT scan was done; showing no evidence of retroperitoneal or intraperitoneal bleed. This patient had an intramuscular bleeding resulting in the bruise. Does this considered as a Grey Turner sign? But how do we know conclusively whether retroperitoneal bleed has actually occurred  without performing a CT scan? Isn't Grey-Turner sign a clinical sign?

Grey-Turner sign can be accompanied by another sign, the Cullen sign; also of similar pathophysiologic basis but at a different location. Grey-Turner refers to bruising at the flank; Cullen sign refers to bruising at the periumbilical region.

Grey-Turner vs Cullen: How to remember which is which?
The way I remember which is Grey-Turner and which is Cullen:

C = Cullen = Central abdomen (periumbilical)

Grey-Turner is the other one, then ("periphery", flank)

Can We Avoid Head CT scan in Some Pediatric Patients With Head Trauma?

In a multicenter trial published in Lancet recently, Kuppermann N et al. enrolled 42,412 children (age, ≤18 years) with mild head trauma (defined as Glasgow Coma Scale score >13) to derive and validate decision rules for two separate age groups:
- those below 2 years and
- those aged 2 years and above.

In that paper, the authors highlighted the need to identify pediatric patients with very low risk of clinically important brain injuries who might not need a CT scan after all:

1. 40 - 60% of those with traumatic brain injuries seen on CT scan are from this group of patients with minor head injuries or those with GCS 14 and 15.

2. But the converse is not true - less than 10% of those with minor head injuries show traumatic brain injuries

3. Furthermore, even if there are head trauma identified, injuries needing neurosurgery are very uncommon in children with GCS scores of 14 - 15

4. The risk of radiation exposure. Ionising radiation from CT scans can cause lethal malignancies. The estimated rate of lethal malignancies from CT is between 1 in 1000 to 1 in 5000 pediatric cranial CT.

But predictive models in such cases are not new. There have been predictive models to identify low risk patients where Head CT may probably be avoided. However, in that paper, it is stated that the problems with previous predictive models:

- are limited by small sample size
- lack of validation
- no independent assessment of preverbal children (less than 2 years old)

From that study, the authors identified that:

The Decision To Avoid CT in Children with Head Trauma could be made:

A. In Children less than 2 years old if
* Normal mental status
* No scalp hematoma except frontal hematoma
* LOC<5 seconds
* Non severe mechanisms of injuries (see below for the list)
* No palpable scalp fracture
* Normal behavior

B. In Children 2 years and above if
* Normal mental status
* No LOC
* No vomiting
* Non severe mechanisms of injuries (see below for the list)
* No signs of base of skull fracture
* No severe headache

Non severe mechanisms of injuries:
# death of a passenger in the accident
# ejection of patient from the vehicle
# rollover
# pedestrian or bicyclist without helmet struck by the vehicle
# fall more than 1.5 m for children above 2 years old and more than 0.9 m for children less than 2 years
# head struck by high impact object

In the validation group of 2216 children younger than 2 years, the rule had 100% sensitivity and negative predictive value. In the validation group of 6411 children 2 years and older, the rule had 96.8% sensitivity and 99.5% negative predictive value.

Using the list of features identified, the suggested algorithm for mild head trauma in that paper:
A. In Children less than 2 years old:
Step 1:
Is the patient with altered mental status OR a palpable skull fracture: If yes - CT;
If no, then proceed to Step 2

Step 2:
Is the patient has occipital or parietal or temporal scalp hematoma OR LOC 5 or more seconds OR severe mechanism of injury OR "not acting normally" as per parent:
if yes: (use clinical judgement with the following in mind)
Observation vs CT on the basis of other clinical features include
- physician experience
- multiple versus isolated findings
- worsening symptoms of signs after emergency department observation
- age of less than 3 months
- parental preference


if not
- then CT SCAN IS NOT RECOMMENDED

B. In Children 2 years or older

Step 1:
Is the patient with altered mental status OR other signs of basilar skull #: if yes - CT

If no, then proceed to Step 2:

Step 2:
Is the patient with history of LOC OR vomiting OR severe mechanism of injury OR severe headache:

If yes, again Observation vs CT on the basis of other clinical factors including:
- physician experience
- multiple versus isolated findings (see note below)
- worsening symptoms or signs after emergency department observation
- parental preference

If no, then CT SCAN IS NOT RECOMMENDED.

Note: Patients with certain isolated findings (i.e. without other findings suggestive of traumatic brain injury) such as
- isolated LOC
- isolated headache
- isolated vomiting
- certain types of isolated scalp hematomas in infants older than 3 months
have risk of clinical important TBI (traumatic brain injuries) of less than 1%.

In general, if risk of clinically important TBI is exceedingly low, lower than risk of CT induced malignancies, then CT scans are not indicated.

In short, while using this rule may identify pediatric patients at very low risk of having clinically important TBI when ALL of the CRITERIA are fulfilled; the converse is not true. Doctors still have to use their own clinical judgment to see which patients they would order a CT scan. Nevertheless, as always, extra caution is still advisable in children younger than 3 months, in whom clinical evaluation may be less reliable.


Reference:
Kuppermann N et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: A prospective cohort study. Lancet 2009 Sep 15; [e-pub ahead of print]. (http://dx.doi.org/10.1016/S0140-6736(09)61558-0)