Investigation

Who to image

Age: younger

ICH location: Post fossa > Lobar >> Deep

Evidence of small vessel disease

Non contrast CT features suggestive

Enlarged vessels around haematoma or draining from it
Calcification around haematoma
ICH location adjacent to Circle of willis
Hyperdense veins for venous sinus thrombosis
Finger like projection of haematoma

Amyloid angiopathy in elderly
Pt on anticoagulants
Venous haemorrhage

CT

Mass effect: clot can dissect brain tissue when enlarging so it can appear significantly large without cause it much mass effect.

Clot volume:

Prognostic
Calculated: ellipsoid method: diameter in three planes/ 2=(AP X LAT X HT)/2

Progression 1st 2 wks very little change

Clot size dec. 0.75mm/day
Density dec. 2 CT units/ day

Pitfalls

Calcification
Teratoma
High attenuation of blood and changes through time. The dotted lines are normal brain

MRI

Not good for ICH as it doesn’t show blood well for first few days

However useful for cerebral amyloid angiopathy

Appearance complicated by age of clot

Variation of MRI appearance of ICH with time since hemorrhage

Stage	Age	Condition of hemoglobin	T1WI	T2WI
Hyper-acute	< 24 hrs	oxy-Hgb (intracellular)	iso	sl. ↑
Acute	1-3 d	deoxy-Hgb (intracellular)	sl. ↓	very ↓
Subacute • Early • Late	> 3d > 7d	met-Hgb (intracellular) met-Hgb (extracellular†)	very ↑ very ↑	very ↓ very ↑
Chronic • Center • Rim	> 14d	hemichromes‡ (extracellular) hemosiderin (intracellular)	iso sl. ↓	sl. ↑ very ↓

oxy-Hgb = oxyhemoglobin, deoxy-Hgb = deoxyhemoglobin, met-Hgb = methemoglobin, iso = iso-intense to brain, ↓ = hypo-intense, ↑ = hyperintense, sl = slightly
† when the RBCs lyse, the Hgb becomes extracellular
‡ diamagnetic (non-paramagnetic) heme derivatives

Mnemonic graph describing the progression of MRI signal intensity changes of intracranial hemorrhage

Evolution of intracranial hemorrhage appearance on MRI

Stage	Age	Hemoglobin	T1W SI	T2W SI	Point on the graph
Hyperacute	<24h	Oxyhemoglobin	Isointense	Hyperintense	1
Acute	1–3d	Deoxyhemoglobin	Isointense to hypointense	Hypointense	2
Early subacute	>3d	Intracellular methemoglobin	Hyperintense	Hypointense	3
Late subacute	>7d	Extracellular methemoglobin	Hyperintense	Hyperintense	4
Chronic	>14d	Hemosiderin	Hypointense	Hypointense	5

T1 great Washington bridge gray white black
T2 black white black oreo

Hyperacute stage

Oxy-haemoglobin in cells (Fe2+): no free electrons —> diamagnetism (weak effect from Hb all effect on MRI due to surrounding tissue)

T1: iso: due to presence of other protein within haematoma
T2: hyper: due to edema —> slowing fluid rotation to larmor fq —> shorter T2

Acute stage

Deoxyhaemoglobin in cells (Fe2+): has 4 free electrons —> paramagnetism

T1: iso: Fe held tightly in Hb —> H2O cant get close to it
T2: hypo: Hb still held in cell and in a macro molecule —> Fe cannot move —> local static magnetic effect increased causing dephasing of nuclei precession —> very short T2 —> so short that it is not detected by current MRI scans

Early sub-acute stage

Methemoglobin in cell (Fe3+): has 5 free electrons —> paramagnetism. When Hb is a oxygen poor state methaemoglobin reductase cannot function to keep the Fe in Hb in a reduced Fe2

+ state therefore Fe2+—>Fe3+. This is shown as red meat (Fe2+) turning to brown meat (Fe3+). When Fe3+ forms the core of the Hb, Fe3+ can no longer accept O2 & deoxyHb now becomes metHb. MetHb has a different conformational S(x) than deoxyHb, where it can accept water to bind very closely to it (see image) forming aqua-methHb. This close proximity of water allows for the T1 image

T1: hyper: Fe still in cell but in MetHb allows for water to come into contact with it—> electromagnetic energy transfer from water to the Fe3+ on the MetHb—> shortening the water’s T1
T2: hypo: Hb still held in cell and in a macro molecule —> Fe cannot move —> local static magnetic effect increased causing dephasing of nuclei precession —> very short T2 —> so short that it is not detected by current MRI scans

Late sub-acute stage

Methemoglobin out of cell (Fe3+): has 5 free electrons —> paramagnetism

T1: hyper: RBC lysis —> MetHb exited cells with H2O still attached to it (HydroMetHb) —> electromagnetic energy transfer from the water to the Fe3+—> shortening the water’s T1
T2: hyper: RBC lysis —> MetHb with its paramagnetism is now not confined in a cell and is flowing within the haematoma—> the previously resonance dephasing is gone —> the T2 is now much longer —> bright on T2

Chronic

MetHb phagocytosed by macrophages and haematoma are all broken down. Hemichromes form that has no paramagnetism. Within the core of the haematoma, its more fluid like. The periphery of the haematoma you can get deposition of ferritin and haemosiderin.

CSDH: center will remain bright for several months on T1-weighted images.

Due to residual methemoglobin with possible contributions from soluble ferrous salts.

Ferritin: a protein shell (aka apoferritin) with 100-1000s of Fe atoms inside. The protein shell prevents water from coming into contact

Haemosiderin:

Conglomerates of clumped ferritin particles, denatured proteins, and lipids.
The iron within hemosiderin is insoluble, but is in equilibrium with the soluble ferritin pool.
Since more Iron —> greater super paramagnetic effect

Hemichromes:

Are oxidative denaturization of MetHb —> The histidine a.a. Originates from the globin chain. —> This binding unravels the globin chain which the affects the alpha and beta subunit of the Hb —> promoting Hb breakdown into a polypeptide chain —> releasing Fe —> Free Fe phagocytosed by macrophages —> forming ferritin
Hemichromes can deposit in RBC to form Heinz bodies (associated with G6P{D deficiency)

Hypo: Center of the haematoma is fluid like —> any hemichromes and protein in the center has no paramagnetism and is tumbling so fast like proteins in CSF —> cannot steal electromagnetic energy away from water —> Long T1 —> dark
Hypo: Periphery of the haematoma has deposition of haemosiderin and ferritin —> although these are superparamagnetism materials (10000s of free electrons) that are stuck to the wall, they are SEPARATED from the surrounding water —> no T1 shortening occurs

Hyper: Center of haematoma is like fluid —> high hemichromes (lack of paramagnetism) and protein molecular motion —> no static electromagnetic field to dephase protons on Water —> the T2 time is not extremely short and therefore detectable on MRI.
Hypo: Immobile and superparamagnetic haemosiderin and ferritin can cause variations in magnetic field in a large surrounding area —> these causes phase differences —> shortening t2 time so that it is NOT detectable on MRI

SWI help differentiate between

Blood and calcification

Haemosiderin is paramagnetic

Hypointense signal

Calcification is not paramagentic

Hyperintense signal at the calcification regions.

To look for other small previous sub clinical bleeds outside of the current lobar ICH.

CTV for venous thrombosis

Cerebral angiography

Indication

Recommended in all patients except

Patients >45 yrs and,
Pre-existing HTN and,
Deep (thalamus/putamen/post fossa)

0% yield and low yield in (Zhu 1997) patients with isolated deep ICH (Laissy 1991)
Patients >45 yrs + HTN + lobar: 10 % yield (zhu 1997), AVM:Aneurysm is 4:1 (normally 1:6)
Patients with intraventricular haemorrhage (no parenchyma haematoma) yield =65% mainly due to AVM

To look for AVM or aneurysm

Angiography is not good as early study

CTA is first line

CTA is first line

95% sensitivity
99% specificity

Can miss small <1 cm AVMs
Negative CTA

Depends on clinical details
MRA adds little
Intra-arterial angiography next line
DSA timing is still up for debate

Cannot rule out AVM or tumour early

Lesion might be obliterated by ICH
Delayed CT scan at: 2/3 months then another at every 6 months for one year
Can try delayed Angiogram after few weeks
MRI/MRA 90% sensitivity for detecting s(x) abnormalities eg. aneurysm >3mm, so a negative study cannot completely exclude a s(x) abnormality