imaging modalities in acute stroke: time is...

Jeremiah ScharfGillian Lieberman, MD

Imaging Modalities in Acute Stroke:

Jeremiah Scharf, Harvard Medical School, MS IVGillian Lieberman, MD

Beth Israel-Deaconess Medical CenterDepartment of Radiology

April 2001

Time is Brain

2

Jeremiah ScharfGillian Lieberman, MD

Stroke - Definition and StatisticsAcute, vascular injury to CNS

<24 hrs = TIA>24 hrs = stroke (CVA)

Affects 600,000 people/ yr(that is 1 stroke per minute!)

Is #3 cause of mortality in adultsIs #1 cause of disabilityhttp://www.swmed.edu/stars/resources/stroke.html

3

Jeremiah ScharfGillian Lieberman, MD

Types of Stroke

Hemorrhagic (20%)usually hypertensive hemorrhage

Ischemic (80%)Thrombotic (40%)

intracerebral atherosclerosis

Embolic (60%)Cardiac embolus (thrombus, tumor, septic embolus)artery-to-artery (mainly carotid thrombus)Paradoxical embolus (thrombus, fat, air)

4

Jeremiah ScharfGillian Lieberman, MD

Risk Factors for Stroke

Atherosclerosis risk factorsFamily history of CVA, TIA, or MIHypertensionSmokingDiabetesHypercholesterolemia

Previous CVA, TIA, or MIAtrial fibrillation

5

Jeremiah ScharfGillian Lieberman, MD

Our Patient - BF

86 yo F w/ Hx of HTN, CAD s/p MI,and high cholesterol

presented to PCP for routine visitFelt “funny” -> began seizingIn ED, unresponsive, L. sided hemiplegia

eyes deviated to the right

6

Jeremiah ScharfGillian Lieberman, MD

Differential Diagnosis

Many CNS diseases can mimic ischemic stroke

HemorrhageMass lesion (tumor, abscess, AVM)Seizure (Todd’s paralysis)Hemiplegic migraineMS flareVenous infarct

7

Jeremiah ScharfGillian Lieberman, MD

Goals of Imaging in Acute Stroke

1. Rule in or out other disease processes2. Define location, extent and age of infarct3. Do so as rapidly as possible

TIME IS BRAIN

8

Jeremiah ScharfGillian Lieberman, MD

Cerebrovascular Anatomy

MGH Handbook of Neurology

PosteriorCirculation

AnteriorCirculation

9

Jeremiah ScharfGillian Lieberman, MD

Anatomy of the Anterior Circulation

High Yield Neuroscience

Internal carotid artery

Anterior cerebral artery

Middle cerebral artery

Middle cerebral artery

Internal carotid artery

Anterior cerebral artery

MGH Handbook of Neurology

10

Jeremiah ScharfGillian Lieberman, MD

Vascular territories in the brain

MGH Handbook of Neurology

11

Jeremiah ScharfGillian Lieberman, MD

Imaging Modalities in Acute Stroke

CT without contrastConventional MRIDiffusion-Weighted and Perfusion MRIMRAUltrasound

12

Jeremiah ScharfGillian Lieberman, MD

CT Imaging in Acute Stroke - 1Initial test of choiceBest modality for detecting hemorrhageIdentifies mass lesions

(tumor, abscess, AVM)Fast and readily available

= Crucial for stroke triage

BIDMCPatient #2 - LL

(rule in/out other diseases)

13

Jeremiah ScharfGillian Lieberman, MD

CT Imaging in Acute Stroke - 2

BIDMC

HOWEVER,CT is poor at detecting acute

infarcts Only 40% sensitivity <24 h

Film Findings for our patient, BF:Normal Initial Head CT

Patient #1 – BF; 1-2 hrs post stroke

2 hours post stroke

Our Patient

14

Jeremiah ScharfGillian Lieberman, MD

Our patient BF: CT#2 – 8 hours laterAs time passes, classic signs of stroke appear:

Patient #1 – BF; 8 hrs post strokeBIDMC

Loss of gray-white matterdifferentiation

Sulcal effacementNormal

sulci

Normal G/W diff.

15

Jeremiah ScharfGillian Lieberman, MD

Our patient BF CT#3 - 2 days later

Patient #1 – BF; 48 hrs post strokeBIDMC

Complete loss ofgray-white matter

differentiation

Sulcal effacement

? hemorrhagictransformation

16

Jeremiah ScharfGillian Lieberman, MD

Therefore, other imaging modalities are used to detect

strokes < 6 hours!

17

Jeremiah ScharfGillian Lieberman, MD

CT without contrastConventional MRIDiffusion-Weighted and Perfusion MRIMRAUltrasound

Imaging Modalities in Acute Stroke

18

Jeremiah ScharfGillian Lieberman, MD

Conventional MR Imaging in StrokeSlight incr. detection rate over CT in early stroke

T2 hyperintensity visibleat 12-24 hrs (80% +)

represents edema

May see absent flow voids = arterial occlusion

T1 imaging basics T2 imaging basicsBIDMC BIDMC

CSF is darkSoft tissue is bright

Good for mass lesions

CSF is brightSoft tissue is dark

Good for edema (bright)

T1 T2Our Patient

19

Jeremiah ScharfGillian Lieberman, MD

Conventional MR Imaging in Stroke

Our patient, BF, underwent an MRI study immediately following her initial CT, 2 hours after her stroke

T2 imageBIDMC

Patient BF; 2 hrs post stroke

Normal Initial MRI:? Absent R. MCA Flow Void

(suggestive of MCA occlusion)

Film Findings:

20

Jeremiah ScharfGillian Lieberman, MD

Our patient BF - MRI #2 :30 hours later

BIDMC

At 30 hrs., classic MR signs of infarct are present

Patient BF; 30 hrs post stroke

T2 hyperintensity in temporal lobe,MCA distribution

Film Findings:

T2 image

21

Jeremiah ScharfGillian Lieberman, MD

Conclusions - Conventional MR Imaging in Acute Stroke

Conventional MRI can detect acute infarcts slightly earlier than CT

Nonetheless, additional techniques are still needed for early stroke detection

22

Jeremiah ScharfGillian Lieberman, MD

Imaging Modalities in Acute Stroke

CT without contrastConventional MRIDiffusion-Weighted and Perfusion MRIMRAUltrasound

23

Jeremiah ScharfGillian Lieberman, MD

Diffusion Weighted MRI Imaging (DWI)

Osmotic pump failure is 1st event in ischemia

Fluid shift extracellular->intracellularWater in cells now can’t diffuse!

Detected as decreased diffusion coefficient (ADC)

Increased restriction of diffusion (DWI)

Detects change within 30 minutesof onset of stroke

Beats T2 signal by 3-6 hours !!!Schaefer et al. Radiology 217:331-345, 2000

24

Jeremiah ScharfGillian Lieberman, MD

Our Patient had a DWI MRI immediately following the initial routine MRI

25

Jeremiah ScharfGillian Lieberman, MD

In our patient, DWI sequences were performed during her initial MRI

A faint increase in DWI signal was observed in the temporal and insular cortex.Indicated early ischemia in

MCA territory

Led to treatment with IV thrombolytic therapy (t-PA)

BIDMC

Patient BF; 2 hrs post stroke

DWI in BF at 2 hours post-stroke

26

Jeremiah ScharfGillian Lieberman, MD

That’s good, but could we predict how bad her stroke might get?

27

Jeremiah ScharfGillian Lieberman, MD

Imaging the Penumbra: the Holy Grail of Stroke Diagnostics

DWI is thought to show the area currently infarcting.

But is there an “area at risk”where blood flow is reducedbut cells haven’t died yet?

Imaging this region (the penumbra)= goal of MR perfusion imaging

Uses gadolinium for contrastChanges magnetic properties of perfused tissue vs. non-perfusedMeasures decreased flow in penumbra!

-increased mean-transit-time(MTT) of blood flow to penumbra

http://www.swmed.edu/stars/resources/stroke.html

28

Jeremiah ScharfGillian Lieberman, MD

Correlation of Perfusion Imaging with Infarct Progression

Baird AE and Warach S. J. Cereb. Blood Flow Metab. 18(6): 583-609, 1998.

= area of low/ slow blood flow

= area at risk forstroke extension

DWI

Late (29h):=larger area

of injury (correlates w/

2h MTT)

MTTEarly (2h):=small area

of injury

29

Jeremiah ScharfGillian Lieberman, MD

Progression of Infarct in BF: DWI at 30 hrs post-stroke

BIDMC

serial,axial sections

demonstratingextent of infarct

at 30 hours

normal DWInormal brain

30

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 2

BIDMC

normal DWInormal brain

31

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 3

BIDMC

increased DWIinjured brain

32

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 4

BIDMC

increased DWIinjured brain

33

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 5

BIDMC

increased DWIinjured brain

34

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 6

BIDMC

increased DWIinjured brain

35

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 7

BIDMC

increased DWIinjured brain

36

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 8

BIDMC

increased DWIinjured brain

37

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 9

BIDMC

increased DWIinjured brain

38

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 10

BIDMC

increased DWIinjured brain

39

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 11

BIDMC

increased DWIinjured brain

40

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 12

BIDMC

increased DWIinjured brain

41

Jeremiah ScharfGillian Lieberman, MD

DWI in 3D - 13

BIDMC

increased DWIinjured brain

42

Jeremiah ScharfGillian Lieberman, MD

Summary – Extent of infarct

BIDMCMGH Handbook of Neurology

our patient

R. PCAterritory

(complete)

R. MCAterritory

43

Jeremiah ScharfGillian Lieberman, MD

Imaging Modalities in Acute Stroke

CT without contrastConventional MRIDiffusion-Weighted and Perfusion MRIMRAUltrasound

44

Jeremiah ScharfGillian Lieberman, MD

MRA/Angiography of occluded MCA

Bahn et al. JMRI 6:833-845, 1996

BIDMC

BIDMC

Patient BF; R. MCA occlusion

Patient BF; R. MCA occlusion

L. MCA occlusion -literature

our patient – R. MCA

our patient – R. MCA

companion patient –L. MCA

45

Jeremiah ScharfGillian Lieberman, MD

Imaging Modalities in Acute Stroke

CT without contrastConventional MRIDiffusion-Weighted and Perfusion MRIMRAUltrasound

46

Jeremiah ScharfGillian Lieberman, MD

Ultrasound in Stroke

Has a primary role in working up cause of strokeEchocardiography

TEE for LA/LV thrombusBubble echo study for PFO (paradoxical embolus)

Carotid UltrasoundEvaluates patency of carotids and degree of stenosis

Transcranial Doppler UltrasoundEvaluates patency of intracranial arteries (MCA)

47

Jeremiah ScharfGillian Lieberman, MD

Summary I – Patient Course

BF was given t-PA within 3 hours of onset30% increase in recovery over controls

(31% t-PA vs. 20% placebo = minimal/ no disability @ 3 mos.)6-fold increased risk of bleeding (6% vs. 1%?)

Symptoms of stroke did not improve considerablyIntubated in ICU for 1 weekTransferred to floor with residual weaknessDischarged to rehabilitation facility after 10 days

48

Jeremiah ScharfGillian Lieberman, MD

Summary II – Timing of stroke detection in our patient

CT without contrast

Conventional MRI

DWI MRI

MRA

Imaging Modality Time of post-stroke imaging2h 8h 30h

- + +

ND +

ND ++

+

-

ND+

+ = evidence of acute stroke; - = no evidence of acute stroke; ND = not determined

49

Jeremiah ScharfGillian Lieberman, MD

Summary III – Imaging in Stroke

Acute Stroke (<6 hours)CT (without contrast!!)

Excellent for ruling out hemorrhage, other diseasesPoor in defining early infarcts

Conventional MRI (T1,T2)Bad for hemorrhage, fair for early infarcts

Diffusion-Weighted and Perfusion MRIExcellent for defining early infarcts (1-2 hrs)

and for estimating areas at risk

50

Jeremiah ScharfGillian Lieberman, MD

Summary IV - Imaging in Stroke

Sub-acute Stroke (>6 hours)CT (without contrast)

Excellent in defining late infarcts (>24 h)sulcal effacement, loss of gray-white differentiation

Conventional MRI (T1,T2)Better than CT at 6-24h; Same as CT in infarcts > 24h

T2 hyperintensity is most indicative of injury

UltrasoundCritical for workup of origin of stroke (TEE, TCD,

Carotid Doppler)

51

Jeremiah ScharfGillian Lieberman, MD

Summary V - Imaging in Stroke

In the near future, there will hopefully be effective treatments for acute stroke.

Patients will need to come to the ER at first sign of “brain attack”.

Patients will need to be imaged by multiple modalities rapidly.

Remember ...

TIME IS BRAIN

52

Jeremiah ScharfGillian Lieberman, MD

ReferencesAHA Website: http://www.americanheart.org/statistics/stroke.htmlBahn MM, Oser AB, Cross DT. CT and MRI of Stroke. JMRI 6:833-845, 1996.Baird AE and Warach S. Magnetic Resonance Imaging in Acute Stroke. J. Cereb. Blood Flow Metab.

18(6): 583-609, 1998.Beauchamp NJ, Barker PB, Yang PY, vanZijl PCM. Imaging of Acute Cerebral Ischemia. Radiology

212:307-324, 1999.Culebras A et al. AHA Scientific Statement: Practice Guidelines for the Use of Imaging in Transient

Ischemic Attacks and Acute Stroke. Stroke. 28:1480-1497, 1997.Flaherty AW. MGH Handbook of Neurology. Lippincott Williams and Wilkins 2000.Fix J. High-Yield Anatomy. Lippincott Williams & Wilkins. Philadelphia. 2000.Lev MH, Farkas J, Gemmete JJ, Hossain ST, Hunter GJ, Kroshetz WJ, Gonzalez RG. Acute Stroke:

Improved Nonenhanced CT Detection – Benefits of Soft-Copy Interpretation by Using Variable Window Width and Center Level Settings. Radiology 213: 150-155, 1999.

Petrella JR and Provenzale JM. MR Perfusion Imaging of the Brain. AJR 175:207-219, 2000.Schaefer PW, Grant PE, Gonzalez RG. Diffusion Weighted MR Imaging of the Brain. Radiology

217:331-345, 2000Simon RP, Aminoff MJ, Greenberg DA. Clinical Neurology. Appleton & Lange: Connecticut, 1999.UT Southwestern STARS Website: http://www.swmed.edu/stars/resources/stroke.html

53

Jeremiah ScharfGillian Lieberman, MD

Acknowledgments

Andru Bageac and Daniel Saurborn- for help with case identification

Beverlee Turner-for help with Power point and PACS

Gillian Lieberman-for her enthusiasm in teaching medical students

My classmates-for enduring this talk