Diffusion and Perfusion MR Imaging

1
Diffusion and Perfusion MR Imaging
2
MRI in Stroke

• Increased sensitivity to stroke relative to CT
• Equivalent sensitivity for detection of
  intraparenchymal hemorrhage
• Conventional MRI may not demonstrate infarct for
  6 hours

3
Physiology of Stroke I

• Brain has little capacity to store energy
• Available energy will maintain brain viability
  for only 3-4 minutes
• Reduction of flow to below 15-20 ml / 100g of
  brain tissue / minute will quickly produce
  ischemia and result in infarct if flow is not
  rapidly restored

4
Physiology of Stroke II

• Loss of metabolic substrate leads to failure of
  the Na / K pumps
• This leads to a net movement of water from the
  extracellular to the intracellular space
• This produces cell swelling
• This process results in typical cytotoxic edema
  seen in infarction

5
Cell Swelling With Acute Infarct
6
Physiology of Stroke IIIThe Penumbra

• It is common for acute stroke to progress
• Hypotension, edema, loss of collateral vessels
  and further thrombosis of the supplying vessel
  may all result in infarction in the penumbra
• Restoration of flow to the penumbra is the goal
  of thrombolytic therapy

7
Basic Principles of Diffusion
L2 2 ?t D Where L the translational
displacement t time D the diffusion
coefficient
8
Diffusion Coefficient
In MR, an estimate of the tissue diffusion
coefficient is called the apparent diffusion
coefficient
L2 2 ?t D
9
Stejskal-Tanner
10
Variation of b Values
11
ADC Map
ln (Sb/So) -b (ADC)
12
Utility of DWI - Early detection of Acute Infarct
DWI
FLAIR
13
Utility of DWI - Acute vs. Chronic Infarct
FLAIR
T2
ADC Map
DWI
14
Diffusion Changes With Time in Cerebral
Infarction
Weeks 1 2 3 4

T2
DWI
ADC
15
Time Course of ADC and T2
16
Perfusion of the Brain

• Delivery of nutrients
• Cerebral blood flow (CBF) normally 60 ml/100g
  brain tissue/min
• Cerebral blood volume (CBV) is the fraction of
  tissue volume occupied by blood vessels normally
  4
• Mean Transit Time (MTT) is the time for blood to
  travel through the tissue volume normally about 4
  seconds

17
Perfusion Principles

• MTT CBV/CBF
• Note similarity
• R V/I

18
Measuring CBF

• Labeled microspheres - Gold Standard
• Embolize in capillaries
• Diffusible Tracer
• Freely moves into brain with a partition
  coefficient ? (if completely diffusible, ? 1)
• NO2
• 133Xe
• 15O H2O in PET
• Intravascular tracer
• Gadolinium

19
Ideal Perfusion

• --- arterial concentration
• flow
• ? tissue distribution constant
• CBV area under curve

20
MR Perfusion Imaging

• Rapid, repeated scanning (Echo planar) at a
  limited number of slices
• Gradient echo technique (T2-weighted)
• Gadolinium bolus
• Decrease in signal as Gd passes through secondary
  to large magnetic moment and susceptibility effect

21
MR Perfusion Principle
22
Perfusion Data

• In PET the arterial input can be obtained (radial
  artery blood sample) and flow can be calculated
• With MR perfusion imaging the flow cannot be
  directly calculated and the CBV is only and
  estimate and is thus termed the rCBV

23
Perfusion Data

• From the perfusion curve that is generated, the
  following information can be obtained
• rCBV can be obtained from integration of the area
  under the curve
• Although flow cannot be directly measured, the
  time to peak (TTP) is proportional to flow
• The peak height is proportional to both the bolus
  delivery and the rCBV

24
Perfusion Maps
25
Perfusion in Stroke

• May detect ischemic areas before diffusion
  changes
• Identify brain at risk adjacent to infarction
• Allow selection of patients outside the window
  for thrombolysis by identifying salvageable brain
• Identify patients with matched perfusion
  diffusion defects who likely will not benefit
  from thrombolysis

26
Perfusion Findings in Acute Stroke

• rCBV abnormality is usually slightly larger than
  diffusion abnormality
• TTP abnormality is bigger than rCBV
• Eventual size of infarct is usually between rCBV
  and TTP abnormality

27
Unmatched vs. Matched Diffusion Perfusion Defects
28
(No Transcript)
29
Other Applications

• Ischemia without infarction
• Toxemia
• Vasospasm
• Cerebral reserve
• Migraine
• Sickle Cell Disease

30
Tumor Imaging

• Perfusion imaging assesses the microvasculature
• Malignant tumors have greater microvascular
  supply
• Uses
• Direct biopsy
• Monitor for recurrence

31
High Grade Glioma
FDG-PET Gad TTP rCBV
32
Radiation Necrosis
33
Recurrent Glioma and Radiation Necrosis
Gad TTP rCBV
34
Functional MRI of Disease States
Functional MRI of Disease States
Mapping of lateralization Activation center
location in altered brain Presence and impact
of disease (stroke) Therapy and relocalization
Mapping of lateralization Activation center
location in altered brain Presence and impact
of disease (stroke) Therapy and relocalization
35
What is Meant by Functional MRI?
What is Meant by Functional MRI?
Flow-sensitive MRI - TOF-MRA, PC-MRI Diffusion-se
nsitive MRI Metabolic mapping - MRS Perfusion
mapping with contrast agents Perfusion "Inflow"
MRI T2-sensitive BOLD MRI
Flow-sensitive MRI - TOF-MRA, PC-MRI Diffusion-se
nsitive MRI Metabolic mapping - MRS Perfusion
mapping with contrast agents Perfusion "Inflow"
MRI T2-sensitive BOLD MRI
"activation mapping"
"activation mapping"
36
Functional MRI of Disease States
Functional MRI of Disease States
Mapping of lateralization Handedness...
Location of foci... Seizures
Epilepsy Tumor impact Stroke
impact
Mapping of lateralization Handedness...
Location of foci... Seizures
Epilepsy Tumor impact Stroke
impact
37
Central sulcus localization with motor task
X
X
38
Functional MRI of Disease States
Functional MRI of Disease States
Mapping of localization of Sensori-motor centers
Tumors / Edema / Mass effects
AVM's Stroke Trauma
Mapping of localization of Sensori-motor centers
Tumors / Edema / Mass effects
AVM's Stroke Trauma By finger
-to-lip control / sensation activations...
39
Topics studied using fMRI Language
lateralization in children w/ partial seizures,
NIH. Long-term motor cortex plasticity,
NIH, Inst. Neurology (London). Motor Recovery
following Cortical Stroke, University of
Essen. Cerebral Oxygenation in Carotid Occlusive
Disease, Biomed. NMR Forschung's
GmbH. Primary/supplementary Motor Activation in
Hemiplegia, University of Edinburgh. Cortical
Processing during Sign Language in the Deaf,
NIH, UWS, UCSD, Inst. Neurology (London). Word
Production in Schizophrenics,
Harvard. Photic stimulation in Schizophrenics,
Harvard. Pre-surgical Mapping of activation
Gliomas and AVM's, HUP.
SMR, San Francisco, 1994 Language
lateralization in children w/ partial seizures,
NIH. Long-term motor cortex plasticity,
NIH, Inst. Neurology (London). Motor Recovery
following Cortical Stroke, University of
Essen. Cerebral Oxygenation in Carotid Occulusive
Disease, Biomed. NMR Forschung's
GmbH. Primary/supplementary motor activation in
hemiplegia, University of Edinburgh. Cortical
Processing during Sign Language in the Deaf,
NIH, UWS, UCSD, Inst. Neurology (London). Word
Production in Schizophrenics,
Harvard. Photic stimulation in Schizophrenics,
Harvard. Pre-surgical Mapping of activation
Gliomas and AVM's, HUP.
40
More Topics studied Pre-surgical Mapping of
Activation, MCW.. Olfactory cortical
dysfunction, NIH. Regional Brain Function
and AVM Location, HUP. Longitudinal fMRI and
Progressive Neuro Disorders, Berlin. Hand
Motor Activation in Congenital Mirror Movements,
Munich. Somatosensory Localization and
Cerebral Tumors, MCW.
RSNA, CHicago, 1994 Pre-surgical Mapping of
Activation, MCW.. Olfactory cortical
dysfunction, NIH. Regional Brain Function
and AVM Location, HUP. Longitudinal fMRI and
Progressive Neuro Disorders, Berlin. Hand
Motor Activation in Congenital Mirror Movements,
Munich. Somatosensory Localization and
Cerebral Tumors, MCW.
41
fMRI Is A Question of Hemodynamics
T2 vs. Proton Inflow
BOLD fMRI relates to bulk T2 within a slice.
Alterations in T2 can be caused by ? CBV ?
CBF ? Oxygen usage/ extraction ? Metabolism
arterial
Inflow fMRI depends on delivery of
arterial protons into an imaged slice - induced
altered magnetization by ? CBF
Imaged slice
venous
42
MR Task Activation of the Human Visual Cortex
GE Signa SPGR TR 75 TE 60 2 NEX 20 sec 128X256 40
flip 5 mm 24 FOV surface coil LED excitation 10
Hz
Placement of Oblique Slice
Tsukamoto Moseley de Crespigny et al. YMS- UCSF
43
MR Task Activation of the Human Visual Cortex
Mask (2 NEX)
Effect of activation
Tsukamoto Moseley de Crespigny et al. YMS- UCSF
44
MR Task Activation of the Human Visual Cortex
Visual cortex
Change
Brain
3.5 minutes
Tsukamoto Moseley de Crespigny et al. YMS- UCSF
Image Number (x 10 seconds)
45
Task Activation of Visual Cortex by MRI
GE 1.5T, Spiral MRI Oblique thro V1 with surface
coil
T1
T2
Spiral
Glover, Lee, et al. Stanford
46
Task Activation of Visual Cortex by MRI
8 images of a train of 252 over 6 minutes
Raw images, 188x188, 1.5 second scans, 20 spirals
Glover, Lee, et al. Stanford
47
Task Activation of Visual Cortex by MRI
"Paradigm"

"Response"
Image Number 1- 252 6 minutes
48
Approximate stimulus as a sinusoid. Calculate r.
MR signal
R0.8
Correlation coefficient, R
49
Task Activation of Visual Cortex by MRI
GE 1.5T, Spiral MRI Oblique thro V1 with surface
coil
Spiral
T1
Correlation function
Glover, Lee, et al. Stanford
50
Optical Imaging
Diffuse Optical Tomography - DOT Informs about
concentrations of Hb in arterial, capillary,
venial compartments. Sensitive to total optical
absorption by HbO and Hb at two wavelengths. --gt
2 measurements, 2 unknowns. Animal model or
neurosurgical microscopy. Temporal resolution 0.1
mm possible.
51
(No Transcript)
52
Optical Imaging
Bonhoeffer Grinvald, Brain Mapping The
Methods, 1996, Eds Toga Mazziota.
53
(No Transcript)