Neuroradiology Lecture Aug2007

Diagnostic Imaging Methods in Central Nervous System Disorders P. Danilo J. Lagamayo, MD

Headache: Primary: Migraine Cluster Tension Secondary: Increased Intracranial Pressure: Neoplasms Abscess Granulomas Meningeal Irritations: Meningitis Subarachnoid Hemorrhage Vascular Disorders: Stroke Malformations Arteritis Head Trauma: Concussion Hematoma Other Cranio-Facial Pains: Trigeminal Neuralgia

Incidence of Primary Brain Tumors: - 6 persons / 100,000 population / year - about 1 in 12 primary brain tumors occur in children under 15 years old.

Clinical Presentation of Brain Tumors: Focal neurologic deficit Increase intracranial pressure: Headache that is more severe in AM - Nausea / Vomiting - Diplopia - Papilledema - Ontundation & Lethargy (ominous) Focal neurologic signs and symptoms: - seizure, seen in about ½ of

Clinical Presentation of Brain Tumors: (II) Non-localizing findings:

- fatigue - malaise - impotense -

glactorrhea growth failure - macrocephaly in young children

Pathological Classification of Intracranial Tumors: Neuroepithelial: Astrocytes Astrocytoma Oligodendrocytes - Oligodendroglioma Ependymal cells & Choroid PlexusEpendymoma Choroid Plexus Papilloma Neurons Gangliomas, Gangliocytomas, Neuroblastomas Pineal cells Pineocytomas Pineoblastomas Poorly differentiated

Pathological Classification of Intracranial Tumors: (cont.) Meninges - Meningioma Nerve sheet cells Neuroma - Neurofibroma Blood vessels - Hemangioblastomas Germ cells - Germinoma - Teratoma Tumors of maldevelopmental origin Craniopharyngioma - Epidermoid/Dermoid cyst

Pathological Classification of Intracranial Tumors: Anterior pituitary gland - Pituitary adenoma - Adenocarcinoma Local extension - Chordoma from adjacent - Glomus jugulare tumors - Chondroma - Chondrosarcoma - Cylindroma

Incidence of Tumors: Glioblastoma - - - - - - - - 55% Astrocytoma - - - - - - - - 20.5% Ependymoma - - - - - - - 6% Medulloblastoma - - - - - 6% Oligodendroglioma _ - - 5% Choroid Plexus Papilloma 2% Other less common entities: Neuronal tumors – Gangliocytomas, ganglioglioma Embryonal – PNET Pineal Region – germ cell

Primary Imaging Methods for Diagnosis of CNS tumors: - MRI - CT scan - Angiography

Advantages of CT Scan -

Wide availability; Can accommodate life support systems; Fast imaging methods Can show bone structures and their pathologic changes like fractures; - Cheap.

Disadvantages of CT Scan • Cannot demonstrate soft tissue detail of the sella turcica, the brain stem and the cerebellum; • Not very sensitive to white matter lesions; • Cannot differentiate encephalomacic lesions of hemorrhage from infarcts; • Uses ionizing radiation and cannot be used on pregnant patients.

Advantages of MRI • Unaffected by the thick bone encasement of the calvarium in the posterior fossa and the sellar turcica; • Accurate determination of the age and evidence of hemorrhage; • More sensitive for detection of white matter lesion; • Ability to perform multiplanar imaging; • Does not use radiation and can be safely used on pregnant patients.

Disadvantages of MRI • Longer time needed to complete an examination; • Needs patient cooperation, e.g. patient must not move during an imaging sequence that can take anywhere from 1 minute to as long as 7 minute; • Any metal implement is not allowed into the MR room.

Disavantages of MRI • Cannot image patients with: – Pacemakers – Neurostimulators – Newly applied vascular clips – Vacular clips with ferromagnetic materials (steel or iron) – Metal foreign bodies in the orbit – Claustrophobic patients – Patients who need extensive life support

Causes of Low Signal Intensity in Tumors in T2WI: Paramagnetic effects - Iron with dystrophic Ca or necrosis Ferritin/hemosiderin from prior bleed - Deoxy hgb in acute bleed - Intracellular met hgb in early subacute bleed - Melanin (or other free radicals)

Causes of Low Signal Intensity in Tumors in T2WI (cont.): Low spin density - Calcifications - Scant cytoplasm (high nucleus:cytoplasm ratio) Dense Cellularity Fibrocollagenous stroma Macromolecule content Very high (nonparamagnetic) protein content

-

Causes of High Signal Intensity in Tumors in T1WI: Paramagnetic effects from hemorrhage - Subacute – chronic blood (met hgb) Paramagnetic materials w/out hemorrhage - Melanin - Naturally occuring ions associated with necrosis of calcification: Manganese, Iron, Copper Non-paramagnetic effects - Very high (non-paramagnetic) proteins - Fat -

Requirements for contrast enhancement: absence of blood-brain barrier - adequate delivery of contrast material - extracapillary interstitial space to accommodate contrast - appropriate contrast dosage - spatial resolution - imaging parameters to allow contrast detection - time for -

Mechanism for contrast enhancement in CNS tumors: Formation of capillaries with deficient blood-brain barrier rather than the destruction of blood-barrier is presumed as the mechanism for tumor enahcement. The capillaries of metastatic tumors to the brain has no blood-brain barrier since these tumors come from elsewhere and not from the brain.

Type of enhancement: immediate or delayed - evanescent or persistent - dense and homogenous minimal or irregular -

Note: Lack of tumor enhancement do not signify lack of tumor.

Effects of Tumor Necrosis on Signal Intensity: Short relaxation times Hemorrhage Liberation of cellular iron Release of free radicals Proteinaceous debris Prolong relaxation increased times water

Cystic change with

Frequently Cystic Tumors Colloid cyst Craniopharyngioma Desmoplastic infantile ganglioma Dermoid Ependymoma (supratentorial and spinal) Epidermoid Ganglion cell tumors Glioblastoma (cystic necrosis) Hemangioblastoma Pilocytic astrocytoma Pleomorphic xanthoastrocytoma Rathke cleft cyst

Magnetic Resonance Criteria for Cystic Lesions

Morphology

Signal Intensity

Sharply demarcated, round smooth Isointense to cerebrospinal fluid on all spin echo images (tumor cysts can be hyperintense due to ↓ T1)

Fluid-debris levels (bleed into necrotic or cystic regions)

Intracellular blood-cyst fluid Intracellular blood-extracellular blood

Motion of intralesional fluid

Lesion emanates ghost images along phase-encoding axis Intralesional signal loss (especially on steady-state sequences)

Hemorrhagic Tumors Primary brain tumors Glioblastoma/anaplastic asctrocytoma Anaplastic oligodendroglioma/oligodendroglioma Ependymoma Teratoma Metastatic disease Melanoma Renal cell carcinoma Choriocarcinoma Lung carcinoma Breast carcinoma Thyroid carcinoma

Intratumoral Hemorrhage vs. Benign Intracranial Hematomas Intratumoral hemorrhage: - Markedly heterogenous, related to: Mixed stages of blood Debris-fluid (intracellular-extracellular blood) levels Edema + tumor + necrosis with blood - Identification of nonhemorrhagic tumor component - Delayed evolution of blood breakdown products - Absent, diminished, or irregular ferritin/hemosiderin -Persistent surrounding high intensity on long TR images (i.e., tumor/edema) and mass effect, even in late stages

Intratumoral Hemorrhage vs. Benign Intracranial Hematomas Benign hemorrhage: - Shows expected signal intensities of acute, subacute or chronic blood, depending on stage of hematoma - No abnormal nonhemorrhagic mass - Follows expected orderly progression - Regular complete ferritin/hemosiderin rim - Complete resolution of edema and mass effect in chronic stages

Intratumoral Melanin vs. Hemorrhage Signal intensity (relative to gray matter) T1-weighted T2-weighted Image Image Amelanotic tumor Melanotic tumor Early subacute blood (intracellular methemoglobin) Late subacute blood (extracellular methemoglobin)

↓ ↑↑

sl. ↑ = or sl. ↓

↑↑

↓↓

↑↑

↑↑

ssification of Astrocyctic Brain Tum Diffuse (infiltrative) Astrocytoma Anaplastic astrocytoma Glioblastoma multiforme

Localized (circumscribed) Pilocytic astrocytoma Pleomorphic xanthoastrocytoma Subependymal giant cell astrocytoma

Diffuse Astrocytic Brain NeoplasmsAnaplastic astrocytoma Astrocytoma Typical site(s) of origin

Cerebral hemisphere (adult) Brainstem (child) Cerebellum (young adult) Signal intensity Homogeneous; high characteristics(on intensity T2-weighted image) Not seen Vascular flow voids Variable; irregular Contrast enhancement 7-8 yr Prognosis (median survival, if available)

Cerebral hemisphere (adult) Brainstem (child) Some heterogeneity

Unusual Common; irregular 2-3 yr

Diffuse Astrocytic Brain Neoplasms Glioblastoma

Gliomatosis

Typical site(s) of origin

Cerebral hemisphere (adult)

Cerebral hemisphere (young or middle-aged adult)

Signal intensity characteristics(on T2-weighted image) Vascular flow voids Contrast enhancement Prognosis (median survival, if available)

Markedly heterogeneous; hemorrhage and necrosis common Common Common; irregular

Ill-defined; high intensity

Rare Uncommon

12 mo Estimated as months

atric Supratentorial Hemispheric Neopla

Signal intensity characteristics (on T2-weighted image) Contrast enhancement Hemorrhage Calcification Prognosis

Juvenile pilocytic astrocytoma

Ganglioglioma

Sharply demarcated; commonly cystic

Sharply demarcated; commonly cystic

Common; dense

Common; irregular

Rare Uncommon Excellent

Rare Common Excellent

atric Supratentorial Hemispheric Neopla

Signal intensity characteristics (on T2-weighted image) Contrast enhancement Hemorrhage Calcification Prognosis

Pleomorphic xanthoastrocytoma

Embryonal tumor (e.g., cerebral neuroblastoma)

Sharply demarcated with subjacent cyst

Markedly heterogeneous

Common in solid portion Rare Uncommon Variable

Common; irregular Common Common Poor

iatric Supratentorial Hemispheric Neopla DNT Signal intensity characteristics (on T2-weighted image) Contrast enhancement Hemorrhage Calcification Prognosis

Sharply demarcated heterogeneous Unknown Rare Common Excellent

ntraventricular Masses Tumor type

Typical location

Central neurocytoma Ependymoma Subependymoma Oligodendroglioma Pilocytic astrocytoma Meningioma Choroid plexus tumor Epidermoid Subependymal giant cell astrocytoma Colloid cyst Arachnoid cyst

Lateral (attached to septum pellucidum) Fourth, lateral Lateral, fourth Lateral Lateral, third, or fourth Lateral (atrium) Lateral (atrium) or third in children, fourth in adults Any ventricle Lateral Third Any ventricle

ntraventricular Masses Tumor type

Intensity characteristics Contrast enhancement on T2-weighted images

Central neurocytoma Ependymoma Subependymoma Oligodendroglioma Pilocytic astrocytoma Meningioma Choroid plexus tumor Epidermoid Subependymal giant cell astrocytoma Colloid cyst

Isointense to gray matter Heterogeneous Hyperintense to gray matter Heterogeneous Hyperintense to gray matter Isointense to gray matter Heterogeneous Slightly hyperintense to CSF Hyperintense to gray matter Hyperintense to gray matter

Arachnoid cyst

Usually dense Heterogeneous None Variable; irregular Dense Dense None Generally enhance Limited enhancement at periphery None Isointense to CSF

Posterior Fossa Tumors in Juvenile pilocytic Medulloblastoma Childhood astrocytoma Signal intensity characteristics (on T2-WI) Contrast enhancement Calcification Hemorrhage Tendency to seed CSF pathways Prognosis (estimated survival)

Sharply demarcated; commonly cystic

Homogeneous; low to moderate intensity

Common in solid portion (mural nodule) Uncommon Rare Extremely low

Common; dense Uncommon Uncommon High

>90% 10-yr survival

50% 5-yr survival

Posterior Fossa Tumors in Childhood Signal intensity characteristics (on T2WI) Contrast enhancement Calcification Hemorrhage Tendency to seed CSF pathways Prognosis (estimated survival)

Ependymoma

Diffuse pontine glioma

Markedly heterogeneous

Ill-defined; high intensity

Common; irregular

Variable

Common Common Low to moderate

Rare Common Low

65-70% 5-yr survival

<1-2% 5-yr survival

ineal Region Tumors Age; sex predilection Pineal vs. parapineal Signal intensity (heterogeneous vs. homogeneous) Hemorrhage Calcification Brain edema or invasion Tendency to metastasize Enchancement Prognosis

Germinoma

Teratoma

Pineoblastoma

Child; male

Child; male

Child; none

Pineal

Pineal

Pineal

Homogeneous (but often hemorrhagic)

Strikingly heterogeneous

Homogeneous (unless hemorrhagic)

Common Rare Common

Typical Typical Variable

Common Common Common

Yes

Variable

Yes

Dense Excellent

Variable Variable

Dense Poor

neal Region Tumors

Age; sex predilection Pineal vs. parapineal Signal intensity (heterogeneous vs. homogeneous) Hemorrhage Calcification Brain edema or invasion Tendency to metastasize Enchancement Prognosis

Pineocytoma

Glioma

Meningioma

Adult; none

Child; none

Adult; none

Pineal

Parapineal (usually) Homogeneous (usually)

Parapineal (usually) Homogeneous

Rare Common Occasional

No

Rare Common Primarily midbrain Variable

Dense Variable

Variable Variable

Dense Excellent

Variable

Common Common Uncommon

No

Magnetic Resonance Findings in Extraaxial Mass Lesions Suggestive Peripheral, broadly based along calvarium Overlying bone changes Enhancement of adjacent meninges Displacement of brain from skull

Definitive Cerebrospinal fluid cleft between brain and lesion Vessels interposed between brain and lesion Cortex between mass and (edematous) white matter Dura (meninges) between (epidural) mass and brain

Stroke: neurologic function secondary to parenchymal ischemia or

a new, often acute, loss of

Main Etiologies for Symptomatology of Stroke: 1. Cerebral Infarction 2. Intraparenchymal Hemorrhage 3. Subarachnoid Hemorrhage

Role of Imaging in Stroke: 1. Rule out hemorrhage 2. Rule other causes of stroke syndrome 3. Help assess etiology in known ischemic infarction

The Normal Brain: To sustain the normal brain, a normal mean regional cerebral blood flow (rCBF) must be maintained at about

54 (± 12 ml) / 100 g / min

The Normal Brain: The threshold for cerebral ischemia is approximately at:

23 ml / 100 g /min.

Autoregulation plays a very important role in maintaining intracerebral blood flow. This mechanism can be temporarily lost in ischemia leaving the control of blood flow to peripheral flow volumes.

Ischemic Strokes: 1. Large Artery or Atherosclerotic Infarction 2. Cardioembolic Infarction 3. Small Vessel Infarction 4. Venous Infarction

The Abnormal Brain: Between cerebral blood flow rate of:

15 & 20 ml / 100 gm / min., ischemic brain injury begins w/ loss of neurologic function, noted as flattening of the electroencephalogram

The Abnormal Brain: Blood flow values below:

10 ml / 100 gm / min., may lead to infarction within a few minutes.

The Ischemic Brain: There are two ischemic changes thresholds, one occurring at blood flow range of 15-20 ml / 100 gm / min., resulting to loss of electrical function and another one at 10ml / 100 gm / min. , resulting to loss of cell polarizaton.

PENUMBRA Heterogeneity in brain injury has been documented in an infarcted zone. Blood flow to an infarcted zone is said to have: A. a central region or core of very low flow that results in rapid cell demise and B. a peripheral penumbra where decline in flow is more moderate and cell death is not immediate.

PENUMBRA The penumbra is thought to represent salvageable tissues that may go on to infarction. If blood flow is normalized at an adequate time, the brain cells will normalize.

Imaging in stroke: Most commonly used imaging method non-contrast CT scan but MRI is fast catching-up.

CT scan is commonly used in stroke due to: - Widespread and ready availability; - Ease of hemorrhage detection; - Compatibility with monitoring equipment; - Rapid scanning techniques for unstable patients.

Emergent evaluation in Acute Stroke: Goals: - Confirm cause of deficit is stroke related. - Assess possible reversibility of the lesion. - Determine most likely etiology. - Predict likelihood of immediate complications. - Begin appropriate treatment.

Emergent Evaluation in Acute Stroke: Opportunities for Intervention: - Before any clinical symptoms. - After transient ischemic attack or minor stroke. - During acute ischemic stroke. - Before a recurrence.

Imaging Signs of Hyperacute Infraction: 1. Hyperdense LMCA sign 2. Loss of gray-white matter differentiation 3. Sulcal effacement. 4. Loss of insular ribbon. 5. Obscurred lentiform nucleus.

Lacunar Infarction: - Not larger than 1.5 cm - Deep gray matter - Brain stem - Deep hemispheric

Cardioembolic Infarction - Relative stasis resulting to mural thrombus, ex.: M.I., atrial fib., ventricular aneurysm - Valvular heart disease resulting to vegetation or from prosthesis - Cardiac tumors - Congenital HD, ex.: right to left shunt

Watershed Infarction: - Boundary zone infarct - Internal carotid stenosis or occlusion - Systemic hypotension - Embolic events

Hemorrhagic Infarction: - Hemorrhagic transformation results to petechial hemorrhage or frank hematoma Anticoagulant therapy - Thrombolytic agents - More common in cardioembolic strokes - Larger cardioembolic strokes are more likely to bleed

Temporal Evolution of Infarction on CT Scan: 0 – 4 hrs.

Normal to subtle hypodensity ± sulcal effacement

1 – 7 days

Mass effect peaks at 3 – 4 days

1 – 8 weeks

Contrast enhancement

Days to months/ yrs

Hypodensity

Acute to Subacute Infarction Changes: 1. Vasogenic Edema that later on wanes 2. Enhancement (Luxury perfusion) 3. Petechial hemorrhage

Hypertensive Hemorrhage In hypertensives, hyalinization within the walls of small cerebral vessels results in

microaneurysms that are less than 1.0 mm in size,

(Charcot & Bouchard), that tend to arise from perforating vessels that will later on bleed.

Some of the Causes of ICH: Hypertension Amyloid Vasculopathy Aneurysm A-V malformation Neoplasm Coagulation disorders, e.g. hemophilia Aticoagulants Vasculitis Drug abuse e.g. cocaine Trauma Idiopathic

Hypertension accounts for 40-50% of deaths from non-traumatic hemorrhage in an autopsy series. In young (less than 40 y/o) normotensive patients, cause remains unknown but cryptic AVM is a suspect.

Why is there a need to measure hemorrhage size? Volume of the hemorrhage is a strong indicator of the 30 day survival of the patient.

Methods of measuring ICH Volume: A. Direct volume measurement in the CT Scan system or in a work station; B. Planimetry C. Application of the formula for the sphere:

Volume = 4/3 π (r)3

D. ABC/2 method

Among different methods of volume measurements, the direct volume measurement in the CT scanner is the most accurate but this would depend on the cooperation of the facility operators. Once the patient data is deleted from the memory file of the system, the direct volume measurement can no longer be applied on the data in the hard copy (film).

In older model CT Scan where volume measurement is not available, an alternative method is possible by using the area of the hemorrhage: Volume in cubic cm = Area x slice thickness (millimeters) 1000

ABC/2 Method: Kothari, et. al., has developed a simple bedside method of ICH volume determination with the following formula:

ICH volume = A x B x C 2

ABC/2 Method (continued): Step 1:

The largest dimension of the hemorrhage is determined in the series of CT slices, then the largest diameter of the hematoma is measured and labeled - A;

Step 2:

On the same slice, the largest diameter of hemorrhage 90o to A is determined and labeled – B.

ABC/2 Method (continued): Step 3:

“C” or the cephalocaudal dimention of the hemorrhage is determined by comparing the rest of the CT slices to the largest hemorrhage on the scan. If the hemorrhage area is 75 % of the largest hemorrhage area = one (1) slice for determining C;

ABC/2 Method (continued): Step 3: If the area was 25 to 75% of the slice where the hemorrhage was largest, the slice is considered as one-half a hemorrhage slice; If the area was less than 25 % of the largest hemorrhage, this is not considered as a hemorrhage slice.

When the CT slice thickness is smaller than the table movement, as will be commonly encountered in CT slices of the posterior fossa, there will necessarily be the presence of inter-slice gaps. To remedy this, use the table movement measurement for thickness of the slice instead of the actual slice thickness to calculate for volume.

1

2

3

4

(2)

A B

“1” slice

ABC/2 Method: (A x B x C ) ÷ 2 = Volume in cc A = 4.0 cm B = 2.6 cm C = 2.5 cm (4.0 x 2.6 x 2.5) ÷ 2 = 13 cc Actual computation directly done in the CT scan = 13.3 cc

Reliability & Reproducibility of the ABC/2 Method of Measuring Intraparenchymal Hemorrhage Volume Reader

No.

Intraclass Correlatio n

Difference From Planimetric,* cm3

P†

Mean Time per Measurement,‡ s

1 (Neurosurgery faculty)

20

.99

-2.0 ± 1.2

.11

35

2 (Neurosurgery resident)

20

.99

0.6 ± 3.0

.85

40

3 (Emergency physician)

20

.99

0.8 ± 1.3

.55

33

4 (Nurse) 20 Interrater reliability (readers 1-4): Intrarater reliability (reader 3):

.99

-2.5 ± 1.5 .07 31 Interclass correlation = .99 Interclass correlation = .99 (P=.19)

* Mean±SE difference from planimetric measurement. † Difference from planimetric measurement. ‡ Mean time to determine hemorrhage volume per CT scan with the ABC/2 technique

Mean Hemorrhage Volumes Hemorrhage Volume, cm3

Location

No.

Planimetric

ABC/2

R2

Deep Lobar Brain Stem

83 21 8

23.0 ± 2.7 44.6 ± 8.4 13.6 ± 7.2

23.5 ± 2.9 49.9 ± 9.9 12.3 ± 6.3

.94 .96 .99

Cerebellar Total

6 118

19.6 ± 4.3 26.0 ± 2.6

24.4 ± 5.9 27.5 ± 2.9

.78

Hemorrhage volumes are mean ± SE.

.96

Temporal Evolution of ICH Biochemical Form OxyHg in RBCs DeoxyHg in RBCs MetHg in RBCs Extracellular MetHg Ferritin and Hemosiderin

Clinical Stage Approximate Time of Appearance Immediately to first Hyperacute several hours Acute

Hours to days

Early subacute First several days Subacute to Days to months chronic Remote

Days to indefinitely

Temporal Evolution of ICH Biochemical Form

Intensity on T1WI

Intensity on T2WI

OxyHg in RBCs

≈



DeoxyHg in RBCs MetHg in RBCs Extracellular metHg Ferritin and hemosiderin

≈, 











≈, 



Acute Infarction findings in MRI: 1. Lesion in arterial distribution 2. High intensity in Proton density or in T2 FLAIR 3. Gyral swelling / sulcal effacement 4. Absent arterial flow void 5. Subcortical white matter hypointensity 6. Intravascular contrast enhancement

Diffusion weighted imaging: -Signal attenuation is noted in areas of free diffusion Signal intensity is increased in areas of restricted diffusion with decrease in apparent diffusion coefficient in brain tissue - Decrease in diffusion of water in early ischemia is due to shift of water from extracellular to intracellular