Visit https://ebooknice.

com to download the full version and

explore more ebooks

(Ebook) Magnetic Resonance Imaging in Stroke by

Stephen Davis, Marc Fisher, Steven Warach ISBN
9781588293978, 9781597450102, 1588293971, 1597450103

_____ Click the link below to download _____

https://ebooknice.com/product/magnetic-resonance-
imaging-in-stroke-1099072

Explore and download more ebooks at ebooknice.com

Here are some recommended products that might interest you.
You can download now and explore!

(Ebook) Biota Grow 2C gather 2C cook by Loucas, Jason; Viles, James

ISBN 9781459699816, 9781743365571, 9781925268492, 1459699815,
1743365578, 1925268497

https://ebooknice.com/product/biota-grow-2c-gather-2c-cook-6661374

ebooknice.com

(Ebook) Magnetic resonance imaging in stroke by Stephen M Davis; Marc

Fisher; Steven Warach ISBN 9780511063770, 9780511072239,
9780511119804, 0511063776, 0511072236, 0511119801

https://ebooknice.com/product/magnetic-resonance-imaging-in-
stroke-4062662

ebooknice.com

(Ebook) Matematik 5000+ Kurs 2c Lärobok by Lena Alfredsson, Hans

Heikne, Sanna Bodemyr ISBN 9789127456600, 9127456609

https://ebooknice.com/product/matematik-5000-kurs-2c-larobok-23848312

ebooknice.com

(Ebook) SAT II Success MATH 1C and 2C 2002 (Peterson's SAT II Success)

by Peterson's ISBN 9780768906677, 0768906679

https://ebooknice.com/product/sat-ii-success-
math-1c-and-2c-2002-peterson-s-sat-ii-success-1722018

ebooknice.com
(Ebook) Master SAT II Math 1c and 2c 4th ed (Arco Master the SAT
Subject Test: Math Levels 1 & 2) by Arco ISBN 9780768923049,
0768923042

https://ebooknice.com/product/master-sat-ii-math-1c-and-2c-4th-ed-
arco-master-the-sat-subject-test-math-levels-1-2-2326094

ebooknice.com

(Ebook) Cambridge IGCSE and O Level History Workbook 2C - Depth Study:

the United States, 1919-41 2nd Edition by Benjamin Harrison ISBN
9781398375147, 9781398375048, 1398375144, 1398375047

https://ebooknice.com/product/cambridge-igcse-and-o-level-history-
workbook-2c-depth-study-the-united-states-1919-41-2nd-edition-53538044

ebooknice.com

(Ebook) Advanced Image Processing in Magnetic Resonance Imaging by

Luigi Landini, Vincenzo Positano, Maria Santarelli ISBN 9780824725426,
0824725425

https://ebooknice.com/product/advanced-image-processing-in-magnetic-
resonance-imaging-1098214

ebooknice.com

(Ebook) Hyperpolarized Carbon-13 Magnetic Resonance Imaging and

Spectroscopy by Larson, Peder; ISBN 9780128222690, 0128222697

https://ebooknice.com/product/hyperpolarized-carbon-13-magnetic-
resonance-imaging-and-spectroscopy-37036818

ebooknice.com

(Ebook) Cardiovascular Magnetic Resonance Imaging by Raymond Y. Kwong,

P. Libby ISBN 9781588296733, 9781597453066, 1588296733, 1597453064

https://ebooknice.com/product/cardiovascular-magnetic-resonance-
imaging-1385580

ebooknice.com
Magnetic Resonance Imaging in Stroke

Magnetic resonance imaging (MRI) provides non-invasive

information about the brain’s blood flow, water movement
and biochemical abnormalities following stroke, and
advances in MRI are transforming the investigation and
treatment of cerebrovascular disease. Echoplanar techniques
with diffusion- and perfusion-weighted imaging, together
with developments in magnetic resonance spectroscopy and
angiography, are replacing CT scanning as the diagnostic
modality of choice. In this profusely illustrated book, world
leaders in these technologies review the scientific basis and
clinical applications of MRI in stroke. It will appeal to a broad
readership including stroke physicians, neurologists,
neurosurgeons, rehabilitation specialists, and others with a
clinical or research interest in cerebrovascular disease.

Stephen Davis is Professor of Neurology at the University of

Melbourne. He heads the Stroke Research Group and is
Co-Director of the Brain Imaging Laboratory at the Royal
Melbourne Hospital, where he is Director of Neurology.

Marc Fisher is Professor of Neurology at the University of

Massachusetts, and a leading authority on the use of MRI in
the evaluation of stroke therapies.

Steven Warach is Chief of the Section on Stroke Diagnostics

and Therapeutics in the Stroke Branch at NINDS, National
Institutes of Health, Bethesda, Maryland. He pioneered the
use of diffusion and perfusion MRI in the evaluation of stroke
and in clinical trials.
Magnetic Resonance
Imaging in Stroke

Edited by

Stephen Davis
University of Australia, Melbourne

Marc Fisher
University of Massachusetts Memorial Medical Care, USA

Steven Warach
National Institutes of Health, Bethesda, MD, USA
  
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press

The Edinburgh Building, Cambridge  , United Kingdom
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521806831

© Cambridge University Press 2003

This book is in copyright. Subject to statutory exception and to the provision of

relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.

First published in print format 2003

-
isbn-13 978-0-511-06377-0 eBook (NetLibrary)
-
isbn-10 0-511-06377-6 eBook (NetLibrary)

-
isbn-13 978-0-521-80683-1 hardback
-
isbn-10 0-521-80683-6 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of

s for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.

Every effort has been made in preparing this book to provide accurate and up-to-date
information which is in accord with accepted standards and practice at the time of
publication. Nevertheless, the authors, editors and publishers can make no warranties
that the information contained herein is totally free from error, not least because
clinical standards are constantly changing through research and regulation. The authors,
editors and publisher therefore disclaim all liability for direct or consequential
damages resulting from the use of material contained in this book. Readers are strongly
advised to pay careful attention to information provided by the manufacturer of any
drugs or equipment that they plan to use.
Contents

List of contributors page vii

Preface xiii

1 The importance of specific diagnosis in 1

stroke patient management
John N. Fink and Louis R. Caplan

2 Limitations of current brain imaging

modalities in stroke 15
P. Alan Barber and Stephen M. Davis

3 Clinical efficacy of CT in acute cerebral

ischemia 31
Rüdiger von Kummer

4 Computerized tomographic-based 47
evaluation of cerebral blood flow
Lawrence R. Wechsler, Steven Goldstein
and Howard Yonas

5 Technical introduction to MRI 55

Rohit Sood and Michael Moseley

6 Clinical use of standard MRI 69

Brian M. Tress

7 MR angiography of the head and neck: 85

basic principles and clinical applications
Robert R. Edelman and Joel Meyer

8 Stroke MRI in intracranial hemorrhage 103

Peter D. Schellinger, Olav Jansen and
Werner Hacke

9 Using diffusion–perfusion MRI in animal 113

models for drug development
Marc Fisher

v
vi Contents

10 Localization of stroke syndromes using 121 16 New MR techniques to select patients for 207
diffusion-weighted MR imaging (DWI) thrombolysis in acute stroke
Max Wintermark, Marc Reichhart, Vincent N. Thijs and Gregory W. Albers
Reto Meuli and Julien Bogousslavsky
17 MRI as a tool in stroke drug development 223
11 MRI in transient ischemic attacks: clinical 135 Steven Warach
utility and insights into pathophysiology
18 Magnetic resonance spectroscopy in stroke 233
Jeffrey L. Saver and Chelsea Kidwell
Dawn E. Saunders and Martin M. Brown
12 Perfusion-weighted MRI in stroke 147
19 Functional MRI and stroke 251
William A. Copen and A. Gregory Sorensen
Amy Brodtmann, Leeanne Carey and
13 Perfusion imaging with arterial spin 161 David G. Darby
labelling
David C. Alsop and John A. Detre Index 263
Colour figures between pp. 120 and 121.
14 Clinical role of echoplanar MRI in stroke 175
Stephen Davis and Mark Parsons

15 The ischemic penumbra: the evolution of

a concept 191
Geoffrey A. Donnan, Peter M. Wright,
Romesh Markus, Thanh Phan and
David C. Reutens
Contributors

Stephen M. Davis,
Department of Neurology,
Royal Melbourne Hospital,
Parkville,
Victoria 3050,
Australia

Marc Fisher,
Department of Neurology,
UMACS,
Memorial Health Care,
119 Belmont Street,
Worcester,
MA 01605,
USA

Steven Warach,
National Institutes of Health,
NINDS,
10 Center Drive,
MSC 1063, Room B1D733,
Bethesda,
MD 29892-1063,
USA

Gregory W. Albers,
Stanford Stroke Center,
Stanford University Medical Center,
Palo Alto,
CA 94394,
USA

vii
viii List of contributors

David C. Alsop, Leeanne Carey,

Department of Radiology, Department of Neurology,
Beth Israel Deaconess Medical Center and Harvard Royal Melbourne Hospital,
Medical School, Parkville,
USA Victoria,
Australia
P.A. Barber,
Department of Neurology, William A. Copen,
Royal Melbourne Hospital, Department of Radiology,
University of Melbourne, Massachusetts General Hospital,
Parkville, PO Box 9657,
Victoria 3050, 55 Fruit Street,
Australia Boston,
MA 02114,
Julien Bogousslavsky, USA
Department of Neurology,
University Hospital (CHUV) BH07, David G. Darby,
1011 Lausanne, Department of Neurology,
Switzerland Royal Melbourne Hospital,
Parkville,
Amy Brodtmann, Victoria 3050,
Department of Neurology, Australia
Royal Melbourne Hospital,
Parkville, Stephen M. Davis,
Victoria 3050, Department of Neurology,
Australia Royal Melbourne Hospital,
University of Melbourne,
Martin M. Brown, Parkville,
Stroke Medicine, Victoria 3050,
Institute of Neurology, Australia
University College London,
The National Hospital for Neurology and John A. Detre,
Neurosurgery, Departments of Neurology and Radiology,
Queen Square, University of Pennsylvania Medical Center,
London, 3 W Gates,
UK 3400 Spruce Street,
Philadelphia,
Louis R. Caplan, PA 19104–4283,
Department of Neurology, USA
Beth Israel Deaconess Medical Center,
Boston, Geoffrey A. Donnan,
MA, National Stroke Research Institute,
USA Heidelberg,
Victoria,
Australia
 List of contributors ix

Robert R. Edelman, Rüdiger von Kummer,

Department of Radiology, Room 5106 Department of Neuroradiology,
Evanston Hospital, University of Technology,
2650 Ridge Avenue, Fetscherstr. 74,
Evanston, Dresden,
IL 60201, Saxonia D-01307,
USA Germany

John N. Fink, Romesh Markus,

Department of Neurology, National Stroke Research Institute,
Christchurch School of Medicine, Heidelberg,
New Zealand Victoria,
Australia
Marc Fisher,
Department of Neurology, Reto Meuli,
UMACS, Department of Diagnostic and Interventional
Memorial Health Care, Radiology,
119 Belmont Street, University Hospital (CHUV) BH07,
Worcester, 1011 Lausanne,
MA 01605, Switzerland
USA
Joel Meyer,
S. Goldstein, Department of Radiology,
University of Pittsburgh Health System, Evanston Northwestern Healthcare,
Stroke Institute, Northwestern University School of Medicine,
Departments of Neurology and Neurosurgery, 2650 Ridge Avenue,
200 Lothrop Street, Evanston,
PA 15213, IL 60201
USA USA

Werner Hacke, Michael Moseley,

Department of Neurology, Department of Radiology,
University of Heidelberg, 1201 Welch Road,
Germany Stanford University,
CA 94305-5488,
Olav Jansen, USA
Department of Neuroradiology,
University of Kiel, Mark Parsons,
Germany Department of Neurology,
Royal Melbourne Hospital,
Chelsea Kidwell, University of Melbourne,
UCLA Stroke Center, Parkville,
710 Westwood Plaza, Victoria 3050,
Los Angeles, Australia
CA 90095,
USA
x List of contributors

Thanh Phan, A. Gregory Sorensen,

National Stroke Research Institute, Massachusetts General Hospital,
Heidelberg, NMR Center,
Victoria, 149 13th Street,
Australia Charlestown,
MA 02129,
Marc Reichhart, USA
Department of Neurology,
University Hospital (CHUV) BH07, Vincent N. Thijs,
1011 Lausanne, Department of Neurology,
Switzerland U2 Gasthuisberg,
Herestraat 49,
David C. Reutens, 3000 Leuven,
National Stroke Research Institute, Belgium
Heidelberg,
Victoria, Brian M. Tress,
Australia The University of Melbourne Department of
Radiology,
Dawn E. Saunders, c/o Post Office,
Department of Neuroradiology, Parkville,
The National Hospital of Neurology and Victoria 3050,
Neurosurgery, Australia
Queen Square,
London, Steven Warach
UK National Institutes of Health,
NINDS,
Jeffrey L. Saver, 10 Center Drive,
UCLA Stroke Center, MSC 1063, Room B1D 733
710 Westwood Plaza, Bethesda
Los Angeles, MD 29892–1063,
CA 90095, USA
USA
Lawrence R. Wechler,
Peter D. Schellinger, University of Pittsburgh Health System,
Department of Neurology, Stroke Institute,
University of Heidelberg, Departments of Neurology and Neurosurgery,
Im Neuenheimer Feld 400, 200 Lothrop Street,
D69120 Heidelberg, PA 15213,
Germany USA

Rohit Sood, Max Wintermark,

Department of Radiology, Department of Diagnostic and Interventional
Stanford University, Radiology,
CA 94305-5488, University Hospital (CHUV) BH07,
USA 1011 Lausanne,
Switzerland
 List of contributors xi

Peter M. Wright, H. Yonas,

National Stroke Research Institute, University of Pittsburgh Health System,
Heidelberg, Stroke Institute,
Victoria, Departments of Neurology and Neurosurgery,
Australia 200 Lothrop Street,
PA 15213,
USA
Preface

Stroke is a leading cause of death in Western coun-

tries, with a mortality rate higher than most forms
of cancer and now the commonest cause of long-
term adult disability. Stroke diagnosis and manage-
ment were revolutionized by the widespread
introduction of computed tomographic (CT) scan-
ning in the 1970s. CT scanning sensitively excludes
cerebral hemorrhage, but early ischemic changes
can be subtle. In the first few hours after stroke
onset, when acute therapies such as thrombolysis
are being considered, CT is often normal, although
acute ischemic changes have become better recog-
nized in recent years. Conventional magnetic reso-
nance imaging (MRI) became widely available in
most countries a decade after the advent of CT
scanning, but has had a limited role in stroke diag-
nosis and management. Although MRI provides far
better imaging of posterior fossa structures and
facilitated non-invasive angiography (MRA), its
sensitivity in acute stroke is not much better than
CT. Other functional imaging techniques such as
single photon emission computed tomography
(SPECT) and positron emission tomography (PET)
have been valuable research tools, but have not
been of routine clinical use in the management of
stroke.
Since the 1990s, the increasingly widespread
availability of echoplanar MRI technology facili-
tated the introduction of diffusion-weighted
imaging (DWI), perfusion imaging (PWI) and mag-
netic resonance spectroscopy (MRS). Diffusion-
weighted imaging allows the hyperacute evaluation
of the ischemic core within minutes of stroke onset
and the distinction between acute and chronic

xiii
xiv Preface

ischemic lesions. It represents an extraordinary being tested in randomized trials. A series of

advance in stroke imaging, specifically in the region chapters details the diagnostic advances facili-
of ischemic tissue that is usually destined for infarc- tated by MRA, DWI, PWI and MRS. Following a
tion. PWI provides a measure of the hypoperfused review of the pathophysiology and clinical impor-
tissue at risk, particularly in the ischemic penum- tance of the ischemic penumbra, our contributors
bra, where acute therapies are targeted. Currently, illustrate the role of MRI in drug development and
PWI is dependent on contrast injection, but arterial selection of acute therapies. Recent studies
spin labelling may well supersede this technique. provide insights into the use of MRI in individual-
These new MRI methods also permit topographic ization of the time window, providing a ‘tissue
analysis of acute infarcts and some insights into clock’ for therapeutic interventions such as
stroke pathophysiology and prognosis. Concurrent thrombolysis. Currently, MR-based studies are
MRA allows analysis of acute arterial occlusion and testing the hypothesis that perfusion–diffusion
monitors recanalization. Magnetic resonance spec- mismatch, the postulated MR signature of the
troscopy provides insights into metabolically ischemic penumbra, can suggest the benefit of
deranged cerebral tissues and provides information thrombolysis beyond the clinically established 3-
that is complementary to DWI and PWI. These new hour time window. Finally, functional brain
techniques are transforming the diagnosis and imaging using brain activation studies and MRI
management of acute stroke. We believe that CT is are leading to a better understanding of brain pro-
likely to be widely replaced by these new MR tech- cessing and brain recovery after stroke.
niques within the next few years. This has already In this book, we have targeted neurologists, other
occurred in many expert stroke centres. stroke physicians, neuroradiologists and other clin-
In this book we have aimed to provide a com- icians involved in stroke diagnosis, imaging and
prehensive and up-to-date summary of the dra- management. We have aimed to encapsulate the
matic developments that have occurred in this development, current and emerging clinical role of
field in the last few years and have also tried to MRI in stroke. We are grateful for the contributions
predict likely advances. The scope of the text of our chapter authors, all leaders in the field of
includes background on the importance of MRI and stroke. A few years ago, experts debated
precise stroke diagnosis, the current uses of CT whether MRI, in acute stroke diagnosis, was ready
including perfusion imaging and an introduction for ‘prime time’. After reading this book, we suspect
to standard and echoplanar MRI techniques. you will agree that it is.
Recent advances in MRI permit exclusion of
intracerebral hemorrhage and this is currently Stephen Davis, Marc Fisher and Steven Warach, 2002
 1
The importance of specific diagnosis in stroke patient
management

John N. Fink1 and Louis R. Caplan2

1
Department of Neurology, Christchurch School of Medicine, New Zealand
2
Department of Neurology Beth Israel Deaconess Medical Center, Boston, MA, USA

Introduction: A stroke is not a ‘stroke’ cians should manage patients according to

methods that have been tested by well-designed
Stroke cannot be considered a diagnosis in itself. randomized controlled trials. Unfortunately, few
Stroke refers to any damage to the brain or spinal therapies for patients with stroke have been tested
cord caused by a vascular abnormality, the term gen- with randomized trials, and even fewer have been
erally being reserved for when symptoms begin thoroughly investigated for patients with specific
abruptly. Stroke is anything but a homogeneous stroke subtypes.
entity, encompassing disorders as different as Randomized trials have limitations, including the
rupture of a large blood vessel that causes flooding issue of numbers v. specificity, or ‘lumping’ v. ‘split-
of the subarachnoid space with blood, the occlusion ting’. To provide statistically valid results, random-
of a tiny artery supplying a small but strategic brain ized trials must contain large numbers of patients
site and thrombosis of a venous conduit obstructing with enough end points to analyse within a rela-
outflow of blood from the brain. Each stroke subtype tively short period; therefore ‘lumping’ must pre-
carries with it different implications for acute treat- dominate over ‘splitting’. But, if the results are to be
ment, prognosis and secondary prevention. Each useful for clinical practice, the data must be specifi-
stroke patient has additional variables that influence cally applicable to individual patients. Too often,
management, including the time from onset to pres- there are significant obstacles to doing this.
entation, the severity of the lesion, and associated Investigators have continued to design trials as if
comorbidities as well as social and psychological they expect a single treatment to be effective for all
factors. The availability of non-invasive imaging ischemic stroke patients, resulting in inevitable dis-
techniques has revolutionized the diagnostic appointment. Even when treatments have been
process, enabling a much greater understanding of found effective, there is still a great deal of room for
the relevant pathophysiological processes active in improvement. For example, aspirin has been
the individual patient. This chapter provides an proven to be effective for early secondary preven-
overview of how the specific diagnostic information tion of ‘stroke’ generally, but only prevents 25% of
available from non-invasive investigations can be recurrent strokes within 14 days.1,2 Cost contain-
applied to the management of individual patients. ment and the need to involve a large number of
centres with varying expertise and resources in
trials results in a minimum of patient investigation.
‘Lumping’ vs. ‘splitting’ As a result, accurate subgroup comparisons in trials
become impossible, even when these are reported
The goal of every clinician is to provide the best in a post hoc analysis. Patients who are too ill, old,
care for his or her patients. Where possible, physi- young, or of child-bearing age are often excluded

1
2 John N. Fink and Louis R. Caplan

from trials. Those unable to give informed consent Table 1.1. Stroke classification
or who have too complex, or multiple, illnesses are
also frequently left out. The type of patients that are (a) Clinical stroke classification systems
excluded from these trials are those that doctors are ‘Traditional’
Transient ischemic attack (TIA)
called on to care for every day.
Minor stroke
The term ‘evidence-based’ must be used cau-
Reversible ischemic neurologic deficit (RIND)
tiously when applied to a particular circumstance if
Stroke in progress
that circumstance has not been specifically studied. Completed stroke
Information from trials must be weighed according (a) Oxfordshire Community Stroke Project6
to the context of specific treatment decisions for Total anterior cerebral infarction syndrome (TACI)
individual patients. George Thibault said it well:3 Partial anterior cerebral infarction syndrome (PACI)
Lacunar infarction syndrome (LACI)
We then need to decide which approach in our large thera-
Posterior cerebral infarction syndrome (POCI)
peutic armamentarium will be most appropriate in a particu-
(b) Etiologic classification systems
lar patient, with a particular stage of diseases and particular
TOAST7
coexisting conditions, and at a particular age. Even when ran-
Large artery
domized clinical trials have been performed (which is true
Cardioembolism
for only a small minority of clinical problems), they will often
Small vessel
not answer this question specifically for the patient sitting in
Other determined etiology
front of us in the office or lying in the hospital bed.
Undetermined etiology
The complexity of managing stroke patients is (a) Baltimore-Washington8
increasing. Improvements in diagnostic accuracy Atherosclerotic vasculopathy
have raised new questions about the correct appli- Non-atherosclerotic vasculopathy
cation of existing treatments. There have been Vasculopathy of uncertain cause (lacunar infarct)
Cardiac/transcardiac embolism
many new developments in stroke therapeutics,
Hematological/other
including intravenous and intra-arterial thromboly-
Migrainous stroke
sis, catheter-based interventions such as angio-
Oral contraceptive or exogenous estrogen use
plasty and stenting for both extracranial and Other drug related
intracranial stenoses, the development of new anti- Indeterminate
platelet agents with potentially complimentary
mechanisms of action, and hypothermic treatment,
to name a few. The exact place for all of these thera- and how we diagnose stroke. Early stroke classifica-
pies is not established, yet it is extremely likely that tions relied on clinical information. Terms such as
many of the new treatments that are currently ‘transient ischemic attack (TIA)’, ‘minor stroke’,
‘unproven’ will be able to deliver improved out- ‘reversible ischemic neurologic deficit (RIND)’,
comes for carefully selected patients. Ignoring ‘stroke in progress’ and ‘completed stroke’ were
these new diagnostic and therapeutic develop- used to distinguish stroke subtypes.4 These simplis-
ments is not an option, although a conservative tic distinctions now have little clinical usefulness.
approach must be taken when potentially hazard- Even the term ‘TIA’ is becoming obsolete as smaller
ous therapies have not been rigorously tested. A infarctions have become detectable with magnetic
specific diagnosis is required to optimize treatment resonance imaging (MRI).5
selection. Subsequent classifications have increasingly
focused on stroke etiology, because of its impor-
tance in determining treatment strategies for sec-
Advances in imaging and stroke diagnosis ondary prevention of stroke (Table 1.1). This has
required an increasing emphasis on the results of
Advances in imaging have led to dramatic changes imaging investigations, rather than clinical features.
in our understanding of stroke pathophysiology The authors of the Trial of Org 10172 in Acute
 The importance of specific diagnosis in stroke patient management 3

Stroke Treatment (TOAST) classified strokes as lar system. Brain MRI examinations for stroke
being due to large artery atherosclerosis, cardioem- should routinely include magnetic resonance
bolism, small vessel occlusion, other determined angiography (MRA) of the intracranial vasculature.
etiology or undetermined etiology.7 This system of Magnetic resonance venography (MRV) and MRA
stroke classification represents an important of the cervical carotid and vertebral arteries can
advance, but still has shortcomings that limit its easily be performed at the same sitting as brain
application to the diagnosis and management of imaging, without the need for contrast. Assessment
individual patients. One major limitation is the of the aortic arch and proximal vessels is possible
oversimplified ‘large artery’ classification. This cat- with gadolinium-enhanced MRA. Diffusion-
egory ‘lumps’ embolic strokes from sources in the weighted MR imaging (DWI) and perfusion imaging
aorta, large vessel origins in the thorax, cervical (PI) enable determination in real time of the pres-
arterial lesions, and intracranial arterial stenoses ence and severity of an ischemic deficit and the
with strokes due to thrombotic occlusion of cervical response of the brain to the insult. These new tech-
or intracranial vessels of either anterior or posterior niques enable the concept of stroke diagnosis to go
circulations. beyond that of simple stroke etiology to establish a
Stroke subtype classifications used today, such as comprehensive and dynamic model of stroke
the TOAST system, reflect the type of stroke pathophysiology for individual patients.
imaging techniques that were generally available a
decade ago, namely non-contrast computed
tomography (CT) head scan and ultrasound exam- Initial stroke diagnosis
inations of the cervical carotid arteries and of the
heart. A diagnostic strategy that continues to rely
Stroke or stroke-mimic
solely on these modalities will not achieve a more
accurate diagnosis. Not only are important parts of The initial diagnostic step should be to determine if
the vascular system overlooked entirely by such an the event is due to stroke or a non-vascular stroke
approach, but the accuracy of even these simple mimic. Clinical information remains very impor-
classifications is often poor.9 A lacunar stroke tant in distinguishing disorders such as migraine,
cannot be reliably diagnosed on the basis of clinical seizure, and factitious and psychogenic disorders
and acute CT findings.10 Some patients with a from stroke. Sometimes the diagnosis is relatively
lacunar syndrome have multiple acute lesions on clear, but when this is not the case, imaging results
diffusion-weighted MRI, consistent with an are critical. A typical appearance on a CT scan will
embolic etiology.11 Moreover, diagnosis using the often confirm the diagnosis of stroke; however,
traditional approach is not made in real time, but false-negative CT findings are common in the acute
retrospectively. A subacute CT scan is required if phase, particularly if image quality is poor, readers
the diagnosis of lacunar infarction is to be con- are inexperienced or if the patient presents with
firmed and a cortical lesion excluded. Ultrasound lacunar or brainstem stroke.10,12–14 Diffusion-
tests may be obtained days after the initial presen- weighted MRI is extremely sensitive to acute brain
tation. This is a critical limitation that prevents a ischemia and false-negatives are very rare, with the
specific diagnosis prior to consideration of acute exception of small brainstem lacunes.12,15 DWI is
stroke therapies that can only be overcome if other therefore the diagnostic modality of choice when
protocols for acute imaging and assessment are the diagnosis of stroke is uncertain and positive evi-
used.9 dence of a stroke is required. The importance of an
Newer imaging techniques that allow rapid, non- accurate diagnosis even at this level should not be
invasive assessment of a much greater extent of the underestimated; as many as 20% of initial stroke
vascular system are now widely available. MRI, as diagnoses are erroneous,16 and some patients with
this book demonstrates, is an extremely powerful stroke mimic have been treated with thrombolysis
technique for imaging the brain and cerebrovascu- as a result.17
4 John N. Fink and Louis R. Caplan

management decisions are made. MRI is very sensi-

Arterial occlusion, arterial rupture, or venous
tive for the initial detection of AVM and other vas-
thrombosis
cular abnormalities in the brain, including
The next level of stroke diagnosis is primarily to dis- cavernous angiomata, which are often undetect-
tinguish hemorrhagic from ischemic stroke. able with DSA. MRA has a role in follow-up of any
However, conceptualizing the mechanism and its untreated lesions and screening of high-risk
vascular pathology ensures that stroke due to families.25–27
venous thrombosis is not overlooked. The majority
of the remainder of the chapter considers diagnosis
of ischemic stroke; cerebral venous thrombosis and Specific diagnosis and management of
hemorrhagic stroke are considered briefly below. ischemic stroke

Cerebral venous thrombosis

Acute stroke
Although cerebral venous thrombosis (CVT) is rare
in comparison with other stroke types, it is treat- The NINDS trial28 established the effectiveness of
able and the diagnosis is frequently missed on CT intravenous tissue plasminogen activator (tPA) for
scan. MRI is very sensitive in the detection of CVT;18 acute ischemic ‘stroke’ within 3 hours of symptom
however, the diagnosis can be overlooked if it is not onset. While this is a major advance in stroke
considered in the differential diagnosis, or if sus- treatment, the advancement must not stop there.
ceptibility-weighted (T2*) imaging or MR venogra- Thrombolysis according to the NINDS protocol
phy is not specifically requested. Patients adds one favourable outcome for every 13 patients
presenting with what appear to be lobar hemor- treated, while causing harm to one in 17.28 The
rhages on CT (young patients with temporal lobe NINDS trial and other negative multi-centre trials
hemorrhage especially) are particularly at risk of of intravenous thrombolysis29–33 relied on a CT
being misdiagnosed and mismanaged before the scan and a clock to characterize their patients
correct diagnosis is made.19 before treatment decisions were made. This was
appropriate at the time as other rapid methods of
Hemorrhagic Stroke more detailed assessment were not generally avail-
CT scanning has generally been considered the able, but inevitably resulted in some patients
investigation of choice for identification of intracra- being exposed to the risk of treatment without
nial blood; however MRI protocols including T2* hope of benefit, such as patients whose vessels
imaging are now able to reliably detect acute cere- have spontaneously recanalized,34 or those with
bral hemorrhage, and are far superior to CT in the little salvageable brain tissue within the hypoper-
detection of subacute and chronic hemorrhage.20–23 fused region. At the same time, some patients who
The sensitivity of T2* and FLAIR MRI for the detec- might benefit beyond 3 hours were denied treat-
tion of acute subarachnoid hemorrhage is compar- ment on the basis of time alone, not individual
able with CT.24 pathophysiological features.35 The PROACT trials
Clinicians are already familiar with the need to have subsequently demonstrated the potential of
make a specific diagnosis of the cause of hemor- intra-arterial thrombolytic agents.36,37 Occlusions
rhage, when it is detected. The development of of the internal carotid artery and proximal middle
catheter-based interventions for treatment of aneu- cerebral artery do not respond as well as more
rysms and arterio-venous malformations (AVM) distal occlusions to intravenous thrombolysis, and
has meant that an even more detailed characteriza- may be better treated via the intra-arterial
tion of the size, morphology and anatomic location route.37–39 Determining the appropriate applica-
of these lesions is required to determine the appro- tions for these potentially hazardous therapies
priate therapeutic approach. Digital subtraction requires a specific diagnosis.
angiography (DSA) is generally required before final More advanced, rapid, non-invasive imaging
 The importance of specific diagnosis in stroke patient management 5

techniques for assessment of acute ischemia are bility images, and additional CT scanning is not
increasingly available. Multimodal stroke MRI required before administering acute treatments.20–22
protocols that include diffusion-weighted imaging, Patients with multifocal small chronic hemorrhagic
perfusion imaging, MRA and susceptibility- lesions due to presumed amyloid angiopathy can
weighted imaging can be performed rapidly, also be identified, who may be at increased risk of
exclude brain hemorrhage, define areas of hypoper- hemorrhage if thrombolytic agents are given.23,43
fusion and tissue damage and identify occluded The scanning time of an acute imaging protocol is
arteries, enabling decisions about thrombolysis to less than 15 minutes. In the last 4 years at our insti-
be made according to individual pathophysiologi- tution, we have performed perfusion studies in over
cal criteria.34,35 DWI and MRA can enable an 300 acute stroke patients and have treated 29 acute
unequivocal diagnosis of acute stroke to allow stroke patients with t-PA on the basis of MRI results
stroke patients who might be excluded from throm- alone.
bolysis on CT-based criteria to be treated, such as
patients presenting with seizure at stroke onset,
Stroke etiology and secondary prevention
hypoglycemia or hyperglycemia. Definition of the
‘ischemic penumbra’ with diffusion and perfusion A detailed diagnosis of stroke etiology is required to
MRI may allow expansion of the therapeutic plan management strategies for secondary stroke
window beyond the current 3-hour guideline for prevention. This requires identification of the loca-
selected patients.35 Parameters are being estab- tion and nature of the vascular lesions responsible,
lished to identify those with an unacceptably high identification of systemic stroke risk factors and
risk of hemorrhage due to the severity of the consideration of the likely pathophysiological
ischemic damage present at the infarct core.40 mechanism of stroke. The elements of specific diag-
Continued refinements in MR perfusion imaging nosis of ischemic stroke are summarized in Table
techniques promise to allow more accurate predic- 1.2.
tions of the volume of brain tissue that is at risk of
infarction if reperfusion does not occur, based on Diagnosis of vascular lesions
perfusion thresholds.41 All levels of the vascular supply to the brain should
In addition to enabling more specific application be considered when determining stroke etiology,
of thrombolytic therapies, physicians can use that is: the heart, aorta, proximal carotid or verte-
detailed knowledge of their patients’ pathophysio- bral arteries in the thoracic cavity, cervical carotid
logy to select candidates for other acute stroke ther- and vertebral arteries and intracranial vessels. Not
apies. In particular, patients who are not candidates only must the anatomical location of vascular
for t-PA but who have a persistent vascular occlu- lesions be determined, but knowledge of the nature
sion and a significant volume of brain at risk of and severity of lesions is required, also. MRA can
infarction due to tenuous collateral supply may provide a comprehensive assessment of the vascu-
benefit from hypertensive therapy to improve col- lar tree to determine the location and severity of
lateral circulation.42 vascular lesions. MRI with MRA is the non-invasive
investigation of choice for the diagnosis and follow-
Practical application of acute stroke MRI up of carotid and vertebral artery dissection.44
Multimodal stroke MRI has been in use for several Duplex ultrasound remains more established than
years in institutions in many countries, including MRA for assessment of cervical internal carotid
our own hospital. The hardware and software artery lesions, but promising results are being
required are increasingly available. Stroke fellows shown with contrast-enhanced MRA,45 and verte-
can be trained to perform the studies enabling 24- bral artery assessment is superior with MRA. MRA
hour coverage independent of technician rosters. has great promise in the evaluation of aortic
Acute hemorrhagic stroke can be accurately iden- lesions; 46 it is possible that in the future, MRI of the
tified using an MRI protocol that includes suscepti- heart and great vessels will reduce the need for the
6 John N. Fink and Louis R. Caplan

Table 1.2. Approach to ischemic stroke diagnosis more invasive procedure of transesophageal echo-
cardiography. Transcranial Doppler
1. Initial ischemic stroke diagnosis: Ultrasonography is a useful method of assessing
(a) stroke vs. non-vascular stroke mimic the major intracranial vessels but MRA or CTA offer
(b) ischemic stroke vs. hemorrhagic stroke vs. venous
the convenience of being performed at the same
thrombosis
time as brain imaging. Digital subtraction angiogra-
2. Acute stroke pathophysiology
phy is still required when intravascular interven-
(a) severity and extent of ischemic brain injury
(b) persistence and severity of cerebral hypoperfusion tions are contemplated, on occasion to distinguish
(c) identification of vascular occlusive lesion between critical stenosis and occlusion of the inter-
3. Stroke etiology: vascular lesion nal carotid artery, and to confirm the diagnosis of
(a) location of vascular lesion(s) certain non-atherosclerotic vasculopathies, such as
e.g. cardiac, aorta, vascular origins, cervical fibromuscular dysplasia, inflammatory and infec-
vessels, intracranial vessels tious arteritides, drug abuse-associated vasculopa-
(b) nature of vascular lesion(s) thy, and radiation-induced stenosis.
e.g. cardiac: thrombus, AF, valvular, PFO,
akinesis, endocarditis, other Specific vascular diagnosis and management
vascular: atherosclerosis – severity, ulceration,
other high risk features
Cardiac-origin embolism
other lesions – dissection,
A full discussion of the diagnosis and management
vasospasm, fibromuscular
dysplasia, arteritis, drug- of cardiac-origin embolism is beyond the scope of
associated vasculopathy this chapter and is available elsewhere.47 Secondary
4. Systemic stroke risk factors prevention strategies can include anticoagulants,
(a) traditional risk factor identification: hypertension, antiplatelet agents or their combination, antibio-
smoking, diabetes, hyperlipidemia. tics, antiarrhythmics and cardioversion, pace-
(b) thrombophilia maker, surgery, or catheter-based interventions.
acquired: antiphospholipid syndrome, Therapeutic decisions depend on a specific diagno-
polycythemia, thrombocytosis, sis of the structural lesions involved and the likely
hyperfibrinogenemia, other composition of the embolic particle itself.47,48
inherited: protein C, S, ATIII deficiency,
prothrombin mutation
Lesions of the aorta and great vessels
(c) other, e.g. hyperhomocysteinemia
That the aorta is an important source of brain
5. Stroke mechanism
(a) embolic stroke embolism is now well established.49,50 The embolic
(b) in situ thrombosis risk is greatest for thick, complex and mobile
(c) lacunar infarction plaques.51,52 Gadolinium-enhanced MRA can estab-
(d) hemodynamic / ‘watershed’ stroke lish this diagnosis quickly and accurately.46 The best
(e) vasospasm treatment to prevent embolism from aortic lesions
6. Stroke severity is not yet known. Cases have been reported where
(a) clinical features, e.g. NIH Stroke Scale Score aortic thrombotic masses have disappeared after
(b) lesion volume / location anticoagulant therapy.53,54 Intravenous thrombo-
7. Patient factors lytic treatment 55 and surgical removal of protrud-
(a) premorbid functioning, age
ing atheromas56 have also been reported to be
(b) comorbidities
successful in treating patients with aortic athero-
(c) psychological, social and economic factors
mas.
Atheromatous disease of the origins of the verte-
bral arteries is a common, yet often overlooked
source of posterior circulation TIA and stroke.57
Antiplatelet agents or anticoagulants are generally
 The importance of specific diagnosis in stroke patient management 7

the first line of treatment, but angioplasty and eficial for carefully selected patients with poor
stenting of such lesions may sometimes be appro- untreated prognosis refractory to medical therapy.77
priate.58
Pathophysiological stroke diagnosis
Cervical vascular lesions Vascular imaging studies define the structural
Carotid endarterectomy is well-established for the lesions important in stroke etiology, but may not
treatment of symptomatic severe (70–99%) internal show whether the stroke was due to thrombotic,
carotid artery (ICA) stenosis.59,60 The benefit of embolic or hemodynamic mechanism, and do not
endarterectomy for symptomatic moderate inform about the nature of the embolic material
(50–69%) stenosis is more modest and decisions itself. Some stroke subtypes, such as migrainous
about treatment must take individual and surgeon stroke, may not be associated with a structural vas-
characteristics into account.60,61 The benefit–risk cular lesion. Clinical information must be com-
ratio for carotid endarterectomy for unselected bined with imaging data to achieve a specific
patients with asymptomatic ICA lesions is even diagnosis and tailor management for the individual
lower62 and treatment decisions must be individu- patient.
alized.63,64 A significant increase in severity of sten-
osis increases stroke risk and favours surgery.65 Thrombosis and embolism
Identification of individual patients with higher Our understanding of stroke pathophysiology has
stroke risk who would benefit most from surgical changed dramatically during recent years, empha-
treatment may be possible using TCD microembo- sizing the importance of embolism in stroke patho-
lus detection,66 or possibly platelet scintigraphy67 or genesis.47,48 The majority of non-lacunar ischemic
indicators of cerebral perfusion or vascular strokes are likely to be embolic in origin. Secondary
‘reserve’68,69 including MRI perfusion techniques;70 prevention strategy depends on identification of
however, more studies are still required.71 the donor source, risk factors, and consideration of
Certain cervical carotid artery lesions may be the likely nature of the embolic particle itself.47,48
better treated with intravascular interventions than Thrombosis is likely when complete ICA occlu-
traditional endarterectomy. Careful patient selec- sion is found, although even then embolism from
tion is required; indications might include high cer- the distal ICA thrombus may be the final stroke
vical lesions with difficult surgical access, mechanism, and embolism from the heart may
radiation-induced stenosis, postsurgical restenosis, have caused the ICA occlusion.78–80 Thrombosis
fibromuscular dysplasia and patients with high sur- may be an important mechanism when intracranial
gical risk due to severe medical comorbidity. A ran- vascular stenoses are present. It may be difficult to
domized controlled trial of carotid stenting and know in some cases whether intracranial stenoses
endarterectomy is planned.72 detected in the subacute period represent chronic
Dissection of the internal carotid artery generally lesions or partial recanalization of an embolus.
does not require surgical intervention, even when Repeat imaging at a chronic time point with MRA,
aneurysms are associated;73 however, patients may CTA or TCD may be required.
benefit from a period of anticoagulation.
Hemodynamic stroke
Intracranial stenoses The importance of hemodynamic factors as sole
The identification of intracranial stenoses can have mechanism in stroke etiology has also been over-
important prognostic implications.74,75 Whether emphasized in the past. Many strokes that may pre-
anticoagulation is more appropriate treatment than viously have been considered ‘hemodynamic’,
aspirin for patients with intracranial disease is cur- particularly ‘posterior borderzone’ infarctions are
rently the subject of a multicentre randomized con- likely to be caused by embolism.81–84 However,
trolled trial.76 Intracranial angioplasty and stenting, impaired regional blood flow due to severe vascular
in the hands of experienced operators, may be ben- stenosis or occlusion is likely to contribute to the
8 John N. Fink and Louis R. Caplan

pathogenesis of embolic stroke. A small embolic patient with apparently ‘well-controlled’ blood-
vascular occlusion is more likely to result in infarc- pressure should prompt consideration of intensifi-
tion when insufficient collateral circulation is cation of treatment. Ambulatory blood pressure
present, and low flow may impair clearance of recording may be very helpful to optimize the man-
emboli, or ‘washout’.85 Possibly the most reliable agement in individual patients. The role of inten-
marker for hemodynamic infarction is the topo- sive lipid-lowering therapy in secondary stroke
graphic pattern of infarction seen on acute multi- prevention is currently the subject of a large ran-
modal MRI, including DWI, MRA and perfusion domized trial.
imaging, along with the appropriate clinical setting. Young patients and those without major vascu-
Multiple small acute lesions are seen in widespread lar risk factors for stroke should also be tested for
distribution within the internal borderzone region, hereditary and acquired thrombophilic states
in the absence of a vascular occlusion.86 Evidence (Table 1.2), the discovery of which can lead to
of a hemodynamic cause for stroke warrants con- modification of treatments prescribed, such as the
sideration of reduction of antihypertensive medica- use of higher intensity anticoagulation for patients
tions or other measures to raise blood pressure, as with the antiphospholipid antibody syndrome,94
well as consideration of revascularization proce- or the introduction of additional treatments such
dures. Magnetic resonance perfusion and other as venesection for polycythemia, folate supple-
methods of brain perfusion imaging such as SPECT, mentation for hyperhomocysteinemia and use of
or TCD assessment of ‘vascular reserve’ may agents such as eicosapentanoic acid (fish-oil) to
provide helpful information in the management of reduce fibrinogen levels in hyperfibrinogene-
these patients in the future.70,71 mia.95,96

Migraine and vasoconstriction Multiple possible causes of stroke

The diagnosis of migrainous stroke remains pri- Some individuals presenting with stroke may have
marily clinical. Infarction can be caused by pro- more than one potential cause identified. The
longed intense vasoconstriction87,88 either directly major risk factors that predispose to atherosclero-
as a result of impeded blood flow or due to secon- sis, such as hypertension, cigarette smoking, dia-
dary thrombosis.88,89 Treatment for secondary pre- betes and hypertension promote plaque formation
vention of migrainous infarction should include a and occlusive disease in the coronary arteries, aorta
migraine prophylactic agent as well as antiplatelet and peripheral vasculature, as well as the cranio-
therapy. We have most often used verapamil in this cervical arteries.97 Hypertension also predisposes
setting. the penetrating arteries of the brain to lipohyalino-
sis and atheromatous branch disease. The true fre-
Systemic stroke risk factors quency of multiple potential causes for stroke in
Modifiable systemic stroke risk factors must be the stroke population is not known. Variation in
incorporated into the patient’s diagnosis. reported frequencies in the literature have been
Management must be individualized. Risk factors due to definitions used, such as the degree of
are not simply present or absent, but there is a con- carotid artery stenosis required before it is consid-
tinuum of increasing stroke risk with higher blood ered an etiological candidate, and how thoroughly
pressure and cholesterol levels.90–93 The actual the patients in each series were investigated.
stroke risk may differ between individuals with the Improvements in diagnostic techniques will inevi-
same blood pressure recordings, depending on how tably result in more such patients being identified.
accurately recordings in the office reflect true daily Understanding of the activity of all of these pro-
levels, the duration of hypertension and the pres- cesses in the individual patient is important. Data
ence of additional risk factors. Evidence of signifi- from the Lausanne Stroke Registry indicated that 46
cant end-organ damage such as extensive cerebral (38%) out of 121 recurrent strokes had a different
white-matter disease on CT or MRI scanning in a etiology than the initial index stroke.98 In asympto-
 The importance of specific diagnosis in stroke patient management 9

matic carotid stenosis, 40% of strokes observed medical and surgical treatments as well as younger
ipsilateral to severe carotid lesions were attribut- patients, nor do they rehabilitate as well from the
able to cardioembolic or lacunar etiologies.99 In effects of a stroke. Secondary prevention studies
addition, the coexistence of coronary artery disease have demonstrated that the absolute benefit of
in patients presenting with stroke should not be treatments such as antihypertensive medication
overlooked. Patients who survive ischemic stroke and carotid endarterectomy may be greater for
face a similar risk of death from future myocardial elderly patients.60,107 Socioeconomic and psycho-
infarction to that from recurrent ischemic stroke.100 logical factors may influence treatment decisions
for some patients and their families.

Stroke severity

Stroke severity is an important diagnostic consider- Specific diagnosis and stroke patient
ation in determining stroke prognosis, which in management
turn influences management decisions. Clinical
features, which can be quantified using clinical The implications of the enormous heterogeneity of
scales such as the NIH Stroke Scale generally stroke and stroke patients for patient management
provide the most important prognostic informa- should be obvious. Patients should be regarded as
tion.101 Early ischemic lesion volume detected with individuals and modern non-invasive imaging tech-
DWI is also an independent predictor of stroke niques should be used to obtain a specific diagnosis
outcome.102 Imaging studies can be particularly of stroke pathophysiology for each, in order to
important for prognosis in specific cases. The use of ensure optimal management. Acute stroke therapy
diffusion and perfusion MR imaging techniques should be offered when possible, preferably on the
and MRA in determining the prognosis of patients basis of pathophysiological, rather than arbitrary,
presenting with acute stroke has already been dis- criteria. All patients deserve assessment of potential
cussed. Detection of a large infarction involving the risk factors, such as hypertension, diabetes,
entire middle cerebral artery territory in a younger smoking, lifestyle, etc, and appropriate modifica-
stroke patient is associated with a high risk of tions should be instituted. The mechanism of stroke
‘malignant’ cerebral edema.103 Large infarctions or must be considered. Patients with atherosclerosis
hemorrhages in the posterior fossa may also be who have had evidence of cerebral ischemia should
associated with the development of raised intracra- have an evaluation of their heart, coronary arteries,
nial pressure.104 Recognition of these patterns aorta and extracranial and intracranial arteries.
allows early discussion of treatment options that When atherosclerosis is not the cause, a careful
may include hematoma excision, hypothermia105 search for a specific alternative diagnosis must be
and hemicraniectomy.106 made. Therapeutic strategies should then be insti-
tuted for each of the potential risks found and clini-
cians should carefully weigh the risk–benefit ratio of
Patient variables
each strategy based on the totality of their knowl-
Individualized stroke management requires consid- edge of that individual patient. Some treatments
eration of the whole individual. Even once a spe- such as antiplatelet medications or standard antico-
cific pathophysiological diagnosis of stroke is agulants might be effective against more than one of
achieved, other variables peculiar to that individual the lesions found, while other treatments such as
patient must also be considered before planning carotid endarterectomy or intracranial angioplasty
management. Pre-existent or coexistent illness may are effective only for the lesions treated. Some treat-
limit or affect treatment. The patient’s premorbid ments that should benefit one lesion (e.g. coronary
function is also an important consideration. Age is artery bypass grafting) might pose a risk for patients
never an absolute contraindication to stroke with other lesions such as severe extracranial and
therapy, however elderly patients do not tolerate intracranial occlusive disease.
10 John N. Fink and Louis R. Caplan

Treatment decisions should be based on infor- 11 Ay H, Oliveira-Filho J, Buonanno FS et al. Diffusion-

mation gained about the individual patient, not on weighted imaging identifies a subset of lacunar
assumptions or arbitrary criteria. The availability of infarction associated with embolic source. Stroke
powerful non-invasive imaging technology makes 1999; 30: 2644–50.
thorough evaluation of each patient a realistic 12 Lansberg MG, Albers GW, Beaulieu C, Marks MP.
Comparison of diffusion-weighted MRI and CT in
expectation. The increasing use of these techniques
acute stroke. Neurology 2000; 54: 1557–1561.
in randomized controlled trials will provide
13 Schriger DL, Kalafut M, Starkman S, Krueger M,
answers to important unresolved questions in
Saver JL. Cranial computed tomography interpreta-
stroke therapeutics and continue to improve out- tion in acute stroke: physician accuracy in determin-
comes for our patients. ing eligibility for thrombolytic therapy. J Am Med
Assoc 1998; 279: 1293–1297.
14 Beauchamp NJ, Barker PB, Wang PY, vanZijl PC.
Imaging of acute cerebral ischemia. Radiology 1999;
REFERENCES 212: 307–324.
15 Ay H, Buonanno FS, Rordorf G et al. Normal
1 International Stroke Trial Collaborative Group. The diffusion-weighted MRI during stroke-like deficits.
International Stroke Trial (IST): a randomised trial of Neurology 1999; 52: 1784–1792.
aspirin, subcutaneous heparin, both, or neither 16 Libman RB, Wirkowski E, Alvir J, Rao TH. Conditions
among 19435 patients with acute ischaemic stroke. that mimic stroke in the emergency department:
Lancet 1997; 349: 1569–1581. implications for acute stroke treatment trials. Arch
2 CAST (Chinese Acute Stroke Trial) Collaborative Neurol 1995; 52: 1119–1122.
Group. CAST: randomized placebo-controlled trial of 17 Katzan IL, Furlan AJ, Lloyd LE et al. Use of tissue-
early aspirin use in 20 000 patients with acute type plasminogen activator for acute ischemic
ischaemic stroke. Lancet 1997; 349: 1641–1649. stroke: the Cleveland area experience. J Am Med
3 Thibault GE. Clinical problem solving: too old for Assoc 2000; 283: 1151–1158.
what? N Engl J Med 1993; 328: 946–950. 18 Dormont D, Anxionnat R, Evrard S, Louaille C, Chiras
4 Caplan LR. Are terms such as completed stroke or RIND J, Marsault C. MRI in cerebral venous thrombosis.
of continued usefulness? Stroke 1983; 14: 431–433. Am J Neuroradiol 1994; 21: 81–99.
5 Kidwell CS, Alger JR, Di Salle F et al. Diffusion MRI in 19 Fink JN, McAuley DL. Cerebral venous sinus throm-
patients with transient ischemic attacks. Stroke 1999; bosis – a diagnostic challenge. Intern Med J 2001; 31:
30: 1174–1180. 384–90.
6 Bamford J, Sandercock P, Dennis M, Burn J, Warlow 20 Linfante I, Llinas RH, Caplan LR, Warach S. MRI fea-
C. Classification and natural history of clinically tures of intracerebral hemorrhage within 2 hours
identifiable subtypes of cerebral infarction. Lancet from symptom onset. Stroke 1999; 30: 2263–2267.
1991; 337: 1521–1526. 21 Patel MR, Edelman RR, Warach S. Detection of
7 Adams HP, Bendixen BH, Kappelle LJ et al. hyperacute primary intraparenchymal hemorrhage
Classification of subtype of acute ischemic stroke. by magnetic resonance imaging. Stroke 1996; 27:
Definitions for use in a multicenter clinical trial. 2321–2324.
TOAST. Trial of Org 10172 in Acute Ischemic Stroke. 22 Schellinger PD, Jansen O, Fiebach JB, Hacke W, Sartor
Stroke 1993; 24: 35–41. K. A standardized MRI stroke protocol: comparison
8 Johnson CJ, Kittner SJ, McCarter RJ et al. Interrater with CT in hyperacute intracerebral hemorrhage.
reliability of an etiologic classification of ischemic Stroke 1999; 30: 765–768.
stroke. Stroke 1995; 26: 46–51. 23 Roob G, Schmidt R, Kapeller P, Lechner A, Hartung
9 Lee LJ, Kidwell CS, Alger J, Starkman S, Saver JL. HP, Fazekas F. MRI evidence of past cerebral micro-
Impact on stroke subtype diagnosis of early diffusion- bleeds in a healthy elderly population. Neurology
weighted magnetic resonance imaging and magnetic 1999; 52: 991–994.
resonance angiography. Stroke 2000; 31: 1081–1089. 24 Mitchell P, Wilkinson ID, Hoggard N et al. Detection
10 Toni D, Iweins F, von Kummer R et al. Identification of subarachnoid haemorrhage with magnetic reso-
of lacunar infarcts before thrombolysis in the ECASS nance imaging. J Neurol Neurosurg Psychiatry 2001;
I study. Neurology 2000; 54: 684–688. 70: 205–211.
 The importance of specific diagnosis in stroke patient management 11

25 Ross JS, Masaryk TJ, Modic MT, Ruggieri PM, Haacke delivery in acute middle cerebral artery stroke. Stroke
EM, Selman WR. Intracranial aneurysms: evaluation 1998; 29: 4–11.
by MR angiography. Am J Neuroradiol 1990; 11: 37 Furlan A, Higashida R, Wechsler L et al. for the
449–455. PROACT Investigators. Intra-arterial prourokinase
26 Ronkainen A, Hernesniemi J, Puranen M et al. for acute ischemic stroke. The PROACT II study: a
Familial intracranial aneurysms. Lancet 1997; 349: randomized controlled trial. J Am Med Assoc 1999;
380–384. 282: 2003–2011.
27 Crawley F, Clifton A, Brown MM. Should we screen 38 Pessin MS, del Zoppo GJ, Furlan AJ. Thrombolytic
for familial intracranial aneurysm? Stroke 1999; 30: treatment in acute stroke: review and update of
312–316. selective topics. In: Moskowitz MA, Caplan LR, eds.
28 The National Institute of Neurological Disorders and Cerebrovascular Diseases: Nineteenth Princeton
Stroke rt-PA Stroke Study Group. Tissue plasminogen Stroke Conference. Boston: Butterworth-Heinemann,
activator for acute ischemic stroke. N Engl J Med 1995: 409–418.
1995; 333: 1581–1587. 39 Linfante I, Llinas RH, Chaves C, Caplan LR, Schlaug
29 The Multicentre Acute Stroke Trial – Italy (MAST-I) G. Reperfusion rates and clinical outcome of MCA
Group. Randomised controlled trial of streptokinase, versus ICA occlusion: MRI/CT before and after t-PA
aspirin, and combination of both in treatment of within 3 hours of symptom onset [Abstract].
acute ischemic stroke. Lancet 1995; 346: 1509–1514. American Academy of Neurology, 53rd Annual
30 Clark WM, Wissman S, Albers GW, Jhamandas JH, Meeting, 2001.
Madden KP, Hamilton S. Recombinant tissue-type 40 Kidwell CS, Saver JL, Mattiello J et al. A diffusion-
plasminogen activator (Alteplase) for ischemic stroke perfusion MRI signature predicting hemorrhagic
3 to 5 hours after symptom onset. The ATLANTIS transformation following intra-arterial thrombolysis
Study: a randomized controlled trial. J Am Med Assoc [Abstract]. Stroke 32: 318.
1999; 282: 2019–2026. 41 Schlaug G, Benfield A, Baird AE et al. The ischemic
31 Hacke W, Kaste M, Fieschi C et al. for the ECASS penumbra: operationally defined by diffusion and
Study Group. Intravenous thrombolysis with recom- perfusion MRI. Neurology 1999; 53: 1528–1537.
binant tissue plasminogen activator for acute hemi- 42 Hillis AE, Barker PB, Beauchamp NJ, Winters BD,
spheric stroke. The European Cooperative Acute Mirski M, Wityk RJ. Restoring blood pressure reper-
Stroke Study (ECASS). J Am Med Assoc 1995; 274: fused Wernicke’s area and improved language.
1017–1025. Neurology 2001; 56: 670–672.
32 Hacke W, Kaste M, Fieschi C et al. for the Second 43 Wijdicks EF, Jack CR Jr. Intracerebral hemorrhage
European-Australasian Acute Stroke Study after fibrinolytic therapy for acute myocardial infarc-
Investigators. Randomised double-blind placebo- tion. Stroke 1993; 24: 554–557.
controlled trial of thrombolytic therapy with intrave- 44 Kasner SE, Hankins LL, Bratina P, Morganstern LB.
nous alteplase in acute ischaemic stroke (ECASS II). Magnetic resonance angiography demonstrates vas-
Lancet 1998; 352: 1245–1251. cular healing of carotid and vertebral artery dissec-
33 The Multicentre Acute Stroke Trial – Europe Study tions. Stroke 1997; 28: 1993–1997.
Group. Thrombolytic therapy with streptokinase in 45 Nederkoorn PJ, Kappelle LJ, van der Graaf Y, Hunink
acute ischemic stroke. N Engl J Med 1996; 335: MG, Eikelboom BC, Mali WP. Duplex ultrasound, MR
145–150. angiography and conventional angiography in
34 Caplan LR, Mohr JP, Kistler JP, Koroshetz W. Should carotid artery disease: interpretation of separate test
thrombolytic therapy be the first-line therapy for results and their combinations [Abstract]. Stroke
acute ischemic stroke? Thrombolysis: not a panacea 2001; 32: 335.
for ischemic stroke. N Engl J Med 1997; 337: 46 Fayad ZA, Nahar T, Fallon JT et al. In vivo magnetic
1309–1313. resonance evaluation of atherosclerotic plaques in
35 Albers GW. Expanding the window for thrombolytic the human thoracic aorta: a comparison with trans-
therapy in acute stroke: the potential role of acute esophageal echocardiography. Circulation 2000; 101:
MRI for patient selection. Stroke 1999; 30: 2230–2237. 2503–2509.
36 del Zoppo GJ, Higashida RT, Furlan AJ, Pessin MS, 47 Caplan LR. Brain embolism. In: Caplan LR, Hurst JW,
Rowley HA, Gent M. PROACT: a phase II randomized Chimowitz MI, eds. Clinical Neurocardiology. New
trial of recombinant pro-urokinase by direct arterial York: Marcel Dekker; 1999: 35–185.
12 John N. Fink and Louis R. Caplan

48 Caplan LR. Brain embolism, revisited. Neurology North American Symptomatic Carotid
1993; 43: 1281–1287. Endarterectomy Trial Collaborators. Benefit of
49 Amarenco P, Duyckaerts C, Tzourio C, Hennin D, carotid endarterectomy in patients with sympto-
Bousser MG, Hauw JJ. The prevalence of ulcerated matic moderate or severe stenosis. North American
plaques in the aortic arch in patients with stroke. N Symptomatic Carotid Endarterectomy Trial
Engl J Med 1992; 326: 221–225. Collaborators. N Engl J Med 1998; 339: 1415–1425.
50 Amarenco P, Cohen A, Baudrimont M, Bousser M-G. 62 Executive Committee for the Asymptomatic Carotid
Transesophageal echocardiographic detection of Atherosclerosis Study (ACAS). Endarterectomy for
aortic arch disease in patients with cerebral infarc- asymptomatic carotid artery stenosis. J Am Med
tion. Stroke 1992; 23: 1005–1009. Assoc 1995; 273: 1421–1428.
51 The French Study of Aortic Plaques in Stroke Group. 63 Caplan LR. A 79 year-old musician with asympto-
Atherosclerotic disease of the aortic arch as a risk matic carotid artery disease. J Am Med Assoc 1995;
factor for recurrent ischemic stroke. N Engl J Med 274: 1383–1389.
1996; 334: 1216–1221. 64 Caplan LR. Caplan’s Stroke: a Clinical Approach. 3rd
52 Mitusch R, Doherty C, Wucherpfennig H et al. edn. Boston: Butterworth-Heinemann; 2000:
Vascular events during follow-up in patients with 175–176.
aortic arch atherosclerosis. Stroke 1997; 28: 36–39. 65 Chambers BR, Norris JW. Outcome in patients with
53 Blackshear JL, Jahangir A, Oldenberg WA, Safford RE. asymptomatic neck bruits. N Engl J Med 1986; 315:
Digital embolization from plaque-related thrombus 860–865.
in the thoracic aorta: identification with transesoph- 66 Seibler M, Nachtman A, Sitzer M et al. Cerebral
ageal echocardiography and resolution with warfarin microembolism and the risk of ischemia in asympto-
therapy. Mayo Clin Proc 1993; 68: 268–272. matic high-grade internal carotid artery stenosis.
54 Freedberg RS, Tunick PA, Culliford AT, Tatelbaum RJ, Stroke 1995; 26: 2184–2186.
Kronzon I. Disappearance of a large intraaortic mass 67 Manca G, Parenti G, Bellina R et al. 111In platelet scin-
in a patient with prior systemic embolization. Am tigraphy for the noninvasive detection of carotid
Heart J 1993; 125: 1445–1447. plaque thrombosis. Stroke 2001; 32: 719.
55 Hausmann D, Gulba D, Bargheer K, Niedermeyer J, 68 Markus H, Cullinane M. Severely impaired cerebro-
Comess KA, Daniel WG. Successful thrombolysis of vascular reactivity predicts stroke and TIA risk in
an aortic-arch thrombus in a patient after mesen- patients with carotid artery stenosis and occlusion.
teric embolism. N Engl J Med 1992; 327: 500–501. Brain 2001; 124: 457–467.
56 Belden JR, Caplan LR, Bojar RM, Payne DD, 69 Gur AY, Bova I, Bornstein NM. Is impaired cerebral
Blachman P. Treatment of multiple cerebral emboli vasomotor reactivity a predictive factor of stroke
from an ulcerated, thrombogenic ascending aorta asymptomatic patients? Stroke 1996; 27: 2188–2190.
with aortectomy and graft replacement. Neurology 70 Kim JH, Lee SJ, Shin T et al. Correlative assessment of
1997; 49: 621–622. hemodynamic parameters obtained with T2*-
57 Caplan LR. Posterior Circulation Disease. Clinical weighted perfusion MR imaging and SPECT in symp-
Findings, Diagnosis and Management. Boston: tomatic carotid artery occlusion. Am J Neuroradiol
Blackwell, 1996. 2000; 21: 1450–1456.
58 Spelle PM, Martin JB, Weill A et al. Percutaneous 71 Derdeyn CP, Grubb RL Jr, Powers WJ. Cerebral hemo-
transluminal angioplasty and stenting of the proxi- dynamic impairment: methods of measurement and
mal vertebral artery from symptomatic stenosis. Am association with stroke risk. Neurology 1999; 53:
J Neuroradiol 2000; 21: 727–731. 251–259.
59 NASCET Collaborators. Beneficial effect of carotid 72 Hobson RW. Carotid angioplasty-stent: clinical expe-
endarterectomy in symptomatic patients with high- rience and role for clinical trials. J Vasc Surg 2001; 33
grade carotid stenosis. N Engl J Med 1991; 325: 445–453. (Suppl): S117–123.
60 European Carotid Surgery Trialists Collaborative 73 Guillon B, Brunereau L, Biousse V, Djouhri H, Levy C,
Group. Randomised trial of endarterectomy for Bousser MG. Long-term follow-up of aneurysms
recently symptomatic carotid stenosis: final results developed during extracranial internal carotid artery
of the MRC European Carotid Surgery Trial (ECST). dissection. Neurology 1999; 53: 117–122.
Lancet 1998; 351: 1379–1387. 74 The Warfarin-Aspirin Symptomatic Intracranial
61 Barnett HJM, Taylor DW, Eliasziw M et al. for the Disease (WASID) Study Group. Prognosis of patients
 The importance of specific diagnosis in stroke patient management 13

with symptomatic vertebral or basilar artery stenosis. Clinical Findings, Diagnosis, and Management.
Stroke 1998; 29: 1389–1392. Boston: Blackwell; 1996: 544–568.
75 Bogousslavsky J, Barnett HJM, Fox AJ, Hachinski VC, 89 Caplan LR. Migraine and vertebrobasilar ischemia.
Taylor W, EC/IC Bypass Study Group. Atherosclerotic Neurology 1991; 41: 55–61.
disease of the middle cerebral artery. Stroke 1986; 17: 90 Voko Z, Bots ML, Hofman A, Koudstaal PJ, Witteman
1112–1120. JC, Breteler MM. J-shaped relation between blood
76 Benesch CG, Chimowitz MI. Best treatment for intra- pressure and stroke in treated hypertensives.
cranial arterial stenosis? 50 years of uncertainty. Hypertension 1999; 34: 1181–1185.
Neurology 2000; 55: 465–466. 91 Hansson L, Zanchetti A, Carruthers SG et al. for the
77 Gomez CR, Misra VK, Liu MW et al. Elective stenting HOT Study Group. Effects of intensive blood-pres-
of symptomatic basilar artery stenosis. Stroke 2000; sure lowering and low-dose aspirin in patients with
31: 95–99. hypertension: principal results of the Hypertension
78 Jorgenson L, Tovrik A. Ischemic cerebrovascular dis- Optimal Treatment (HOT) randomised trial. Lancet
eases in an autopsy series. Part 2: prevalence, loca- 1998; 351: 1755–62.
tion, pathogenesis and clinical course of cerebral 92 Tell GS, Crouse JR, Furberg CD. Relation between
infarcts. J Neurol Sci 1969; 9: 285–320. blood lipids, lipoproteins, and cerebrovascular ath-
79 Castaigne P, L’hermitte F, Gautier JC, Escourolle R, erosclerosis. A review. Stroke 1988; 19: 423–430.
Derouesne C. Internal carotid artery occlusion. A 93 Iso H, Jacobs DR, Wentworth D, Neaton JD, Cohen
study of 61 instances in 50 patients with postmortem JD. Serum cholesterol and six-year mortality from
data. Brain 1970; 93: 231–258. stroke in 350,977 men screened for the multiple risk
80 Blackwood W, Hallpike JF, Kocen RS, Mair WG. factor intervention trial. N Engl J Med 1989; 320:
Atheromatous disease of the carotid arterial system 904–910.
and embolism from the heart in cerebral infarction: a 94 Khamashta MA, Cuadrado MJ, Mujic F, Taub NA,
morbid anatomical study. Brain 1969; 92: 897–910. Hunt BJ, Hughes GR. The management of thrombo-
81 Chaves CJ, Silver B, Schlaug G, Dashe J, Caplan LR, sis in the antiphospholipid-antibody syndrome. N
Warach S. Diffusion- and perfusion-weighted MRI Engl J Med 1995; 332: 993–997.
patterns in borderzone infarcts. Stroke 2000; 31: 95 Radack K, Deck C, Huster G. Dietary supplementa-
1090–1096. tion with low-dose fish oils lowers fibrinogen levels: a
82 Belden JR, Caplan LR, Pessin MS, Kwan E. randomized double-blind controlled study. Ann
Mechanisms and clinical features of posterior Intern Med 1989; 111: 757–758.
border-zone infarct. Neurology 1999; 53: 1312–1318. 96 Kobayashi S, Hirai A, Terano T, Hamazaki T, Tamura Y,
83 Angeloni U, Bozzao L, Fantozzi L, Bastianello S, Kumagai A. Reduction in blood viscosity by eicoso-
Kushner M, Fieschi C. Internal borderzone infarction pentanoic acid. Lancet 1981; 2: 197.
following acute middle cerebral artery occlusion. 97 Caplan LR. Lice, fleas and strokes. Arch Neurol 2000;
Neurology 1990; 40: 1196–1198. 57: 1113–1114.
84 Baird AE, Donnan GA, Saling M. Mechanisms and 98 Yamamoto H, Bogousslavsky J. Mechanism of second
clinical features of internal watershed infarction. and further strokes. J Neurol Neurosurg Psychiatry
Clin Exp Neurol 1991; 28: 50–55. 1998; 64: 771–776.
85 Caplan LR, Hennerici M. Impaired clearance of 99 Barnett HJM, Meldrum HE. Carotid endarterectomy:
emboli (washout) is an important link between a neurotherapeutic advance. Arch Neurol 2000; 57:
hypoperfusion, embolism, and ischemic stroke. Arch 40–45.
Neurol 1998; 55: 1475–1482. 100 CAPRIE Steering Committee. A randomised, blinded,
86 Darby DG, Barber PA, Gerraty RP et al. trial of clopidogrel versus aspirin in patients at risk of
Pathophysiological topography of acute ischemia by ischemic events (CAPRIE) Lancet 1996; 348:
combined diffusion-weighted and perfusion MRI. 1329–1339.
Stroke 1999; 30: 2043–2052. 101 Adams HP Jr, Davis PH, Leira EC et al. Baseline NIH
87 Solomon S, Lipton RB, Harris PY. Arterial stenosis in Stroke Scale score strongly predicts outcome after
migraine: spasm or arteriopathy? Headache 1990; 30: stroke: a report of the Trial of Org 10172 in Acute
52–61. Stroke Treatment (TOAST). Neurology 1999; 53:
88 Caplan LR. Migraine and posterior circulation stroke. 126–131.
In: ed. Caplan LR, ed. Posterior Circulation Disease: 102 Thijs VN, Lansberg MG, Beaulieu C, Marks MP, Moseley
14 John N. Fink and Louis R. Caplan

ME, Albers GW. Is early ischemic lesion volume on 105 Schwab S, Schwarz S, Spranger M, Keller E, Bertram
diffusion-weighted imaging an independent predictor M, Hacke W. Moderate hypothermia in the treatment
of stroke outcome? A multivariable analysis. Stroke of patients with severe middle cerebral artery infarc-
2000; 31: 2597–2602. tion. Stroke 1998; 29: 2461–2466.
103 Hacke E, Schwab S, Horn M, Spranger M, De Georgia 106 Schwab S, Steiner T, Schwarz S, Steiner HH, Jansen O,
M, von Klummer R. ‘Malignant’ middle cerebral Hacke W. Early hemicraniectomy in patients with
artery territory infarction: clinical course and prog- complete middle cerebral artery infarction. Stroke
nostic signs. Arch Neurol 1996; 53: 309–315. 1998; 29: 1888–1893.
104 Koh MG, Phan TG, Atkinson JLD, Wijdicks EFM. 107 Staessen JA, Gasowski J, Wang JG et al. Risks of
Neuroimaging in deteriorating patients with cerebel- untreated and treated isolated systolic hypertension
lar infarcts and mass effect. Stroke 2000; 31: in the elderly: meta-analysis of outcome trials.
2062–2067. Lancet 2000; 355: 865–872.
 2
Limitations of current brain imaging modalities in stroke

P. Alan Barber and Stephen M. Davis

Department of Neurology, Royal Melbourne Hospital, University of Melbourne, Parkville, Victoria, Australia

Introduction immediate hours after symptom onset. Fear, confu-

sion, language disturbance, and physical deficits
The successful management of stroke patients may all affect the ability to comprehend or follow
requires the ability to confirm the diagnosis, iden- commands. Patients are often restless, and fre-
tify the site, extent and age of the lesion, and deter- quently become more so with prolonged scanning
mine underlying pathophysiology. Several studies protocols. Many become claustrophobic within the
have made it clear that this cannot be done on the confined spaces of scanners. Head movement is a
basis of clinical findings alone.1–3 At the most basic considerable problem and can result in image
level, clinicians are unable to accurately differen- quality degradation. While this may be minimized
tiate between cerebral infarction and primary intra- by moulded head holders or masks, ill-fitting
cerebral hemorrhage.4 Cerebral imaging is therefore devices may be uncomfortable and increase patient
a prerequisite in the management of almost all agitation and movement. The use of sedatives to
stroke patients. reduce patient movement may mask neurological
Techniques such as computed tomography (CT), deterioration, and is therefore relatively contraindi-
magnetic resonance imaging (MRI), positron emis- cated in the acute stroke setting.
sion tomography (PET) and single photon emission
computed tomography (SPECT) have provided a
window onto the structural and functional changes Computed tomography
that occur during stroke can be examined and have
revolutionized our understanding of the pathophy- Computed tomography is more widely available
siology of ischemia. The advent of thrombolytic and is less expensive than other tomographic
therapy has exposed a need for imaging techniques imaging modalities and has been the investigation
that enable the more rational selection of patients of choice in stroke. CT images can be obtained
for potentially hazardous treatments. Each of these rapidly and non-invasively in unwell and frequently
currently used brain imaging techniques has limita- uncooperative patients. CT identifies cerebral
tions in such a role and these will form the focus of hemorrhage and is the investigation of choice for
this chapter. subarachnoid hemorrhage. In contrast to MRI, CT
can be performed in patients with ferromagnetic
metallic foreign bodies. CT is less affected by
Patient factors
patient motion than conventional MRI and also
There are a number of logistic and technical prob- allows full access to a patient during scanning,
lems related to the performance of all imaging making it of use in restless or critically ill patients.
studies in stroke patients. These difficulties relate to Bony anatomy is also well delineated. This is an
a patient’s emotional and physical state in the advantage when bony injury is suspected, but a

15
16 P. Alan Barber and Stephen M. Davis

Table 2.1. Advantages and limitations of cerebral computed tomography (CT) in acute stroke

Advantages of CT in stroke Limitations of CT in stroke

Widely available and relatively inexpensive Exposure to radiation

Non-invasive Low sensitivity for infarction at the hyperacute stage
Sensitive to hemorrhage, investigation of choice for Difficulties distinguishing acute from chronic infarction
subarachnoid hemorrhage
Good visualization of bony anatomy Poor imaging of posterior fossa structures
Good access to patients during scanning procedure Low interobserver agreement on the degree of ischemic
change
Rapid image acquisition, useful in restless patients Inability to stratify patients in acute stroke trials
CT angiography and CT perfusion imaging can be added Inability to delineate hypoperfused but viable cerebral
to conventional protocols tissue

limitation in the posterior cranial fossa where tening of cortical sulci, an asymmetry between the
‘beam hardening’ artefacts are produced by bone at sylvian fissures and compression of the ventri-
the base of the skull. A summary of the advantages cles.7–11
and limitations of CT in stroke can be found in These early ischemic changes are frequently
Table 2.1 and a full discussion of the role of CT in subtle and CT is often normal in the therapeutically
stroke can be found in Chapter 3. critical first hours after stroke onset. Furthermore,
The principle behind CT is similar to conven- these changes mean that it can be particularly diffi-
tional X-ray imaging. Radiation passes through cult to visualize small strokes on acute CT studies.
tissue from multiple directions; detectors measure This is a result of the small volume of hypodense
the degree of attenuation of the exiting radiation, ischemic tissue and the minimal compressive
and images are reconstructed in cross-section.5 effects on surrounding brain. Thus, CT is not par-
Thus, CT scanning requires exposure to radiation, ticularly effective in identifying small subcortical or
with the average dose equivalent to the background brainstem lacunar strokes.12–14
radiation received in 1 year.6 Repeated studies Over the first days following stroke onset,
should therefore be avoided unless absolutely nec- infarcted tissue becomes more hypodense and
essary. Pregnancy is also a relative contraindication clearly demarcated.15 Cerebral swelling also
to CT, although the risks to the fetus must be increases reaching a maximum between one and
weighed against the potential benefits of the infor- five days. During the second week, some patients
mation provided. may develop a progressive increase in the density of
the infarct, which may result in ‘fogging’ of the
infarct. This phenomenon may persist for up to a
Sequence of changes that are seen on CT following
fortnight and as a result strokes may become more
a stroke
difficult to identify and smaller strokes may be
The early CT changes of ischemia include paren- overlooked completely. The density of an infarct
chymal hypodensity and cerebral swelling. then decreases leaving a hypodense and atrophic
Parenchymal hypodensity corresponds to an lesion.16,17
increase in the intracellular water components of Despite this temporal sequence of changes, it is
affected brain tissue and is best seen as a loss of often difficult to differentiate between an acute or
grey–white matter differentiation in the cerebral subacute, and chronic infarction on CT. This is a
cortex, insular ribbon or basal ganglia (Figs. 2.1 and particular problem with small subcortical lesions
2.2). Cerebral swelling is the result of the accumula- that occur on a background of periventricular white
tion of extracellular water and can be seen as a flat- matter disease. The presence of such a lesion does
 Limitations of current brain imaging modalities in stroke 17

(a)

(b)

Fig. 2.1. Early ischemic changes on computed tomography (CT) may be subtle. (a) CT obtained at 2 hours 45 minutes in a patient
who presented with left hemineglect and hemiparesis, show a hyperdense right middle cerebral artery (MCA) sign (arrow), with a
subtle loss of distinction between the basal ganglia and surrounding white matter (arrowheads). (b) Magnetic resonance angio-
graphy and diffusion-weighted imaging in the same patient at 2 hours 15 minutes show an occluded right MCA and a hyperin-
tense region of ischemia in the right subcortical and basal ganglia regions.

not necessarily mean that the lesion is relevant to a tor (t-PA) be avoided in patients with CT evidence
patient’s clinical presentation. of major ischemia on a screening CT scan.22,23
The subtlety of the early changes of ischemia As these patients may have a greater risk of
results in a low sensitivity for cerebral infarction at hemorrhagic transformation and worse outcome
the hyperacute stage. Indeed the sensitivity of CT following t-PA.11,21,24
for stroke in patients imaged within 5 hours has In practical terms, these limitations mean that
been reported to be as low as 58%.18 The difficulty the presence, location and size of an infarct
identifying hyperacute ischemia on CT also results cannot be reliably determined in the first hours
in a substantial interobserver variability in the after stroke onset. As a consequence, many
detection of acute infarction.19,20 Furthermore, the patients are treated with thrombolytic therapy
reliability and reproducibility of CT in the estima- without definite pre-therapy cerebral CT confir-
tion of the degree of ischemic change is mation of diagnosis of stroke. The major role for
modest.11,21 This is of importance, as some guide- hyperacute CT is to exclude primary intracerebral
lines recommend that tissue plasminogen activa- hemorrhage and other non stroke pathology. CT
18 P. Alan Barber and Stephen M. Davis

(a)

(b)

Fig. 2.2. Early ischemic changes on computed tomography (CT) may be subtle. (a) CT obtained at 2 hours in a patient who pre-
sented with left hemineglect and hemiparesis, shows a hyperdense right middle cerebral artery (MCA) sign (arrow) and subtle loss
of distinction between cortical grey and white matter in the right middle cerebral artery territory (arrowheads). (b) Magnetic reso-
nance angiography and diffusion-weighted imaging in the same patient at 4 hours shows an occluded right MCA with a hyperin-
tense region of ischemia involving the whole of the right MCA territory.

may also lead to an underestimation of the volume Computed tomographic angiography and
of ischemic change so that patients are exposed to perfusion imaging
the risks of thrombolytic therapy, despite guide-
lines to the contrary. Conversely, excessive Computed tomographic angiography (CTA) is a
concern about the degree of ischemic change on rapid, reliable and safe method for imaging the
CT can lead to treatment being inappropriately intracranial vasculature. Data is acquired fol-
withheld.25 In acute stroke trials, CT cannot be lowing the administration of an intravenous
used to reliably stratify participants according to contrast agent. Reconstruction of this data pro-
infarct location or size before therapy is com- duces angiographic images that can be displayed
menced. Patients with similar strokes can only be in three dimensions. CTA is able to delineate
compared retrospectively after treatment has been the site of vessel occlusion and length of
given. occluded segment in acute stroke patients, and
 Limitations of current brain imaging modalities in stroke 19

Table 2.2. Advantages and limitations of conventional magnetic resonance imaging (MRI) in acute stroke

Advantages of MRI in stroke Limitations of MRI in stroke

Widely available Expensive

Rapid and non-invasive Contraindicated in patients with metal fragments,
pacemakers and other electrically or magnetically
active implanted devices
High soft tissue resolution Low sensitivity for infarction in the hyperacute stage
Sensitive to brainstem ischemia Difficulties estimating the degree of ischemic change
Can ‘weight’ images to augment tissue contrast Unable to delineate hypoperfused but viable cerebral
tissue
Multiple sequences can be obtained providing more Unable to stratify patients in acute stroke trials
information than from any one sequence alone

compares well with other vascular imaging tech- Magnetic resonance imaging
niques.26–29
CT can also be used to produce maps of cerebral Magnetic resonance imaging is the imaging modal-
perfusion by following the change in Hounsfield ity of choice for most neurologic conditions. MR
number that occurs with the passage of an intrave- images can be obtained rapidly, non-invasively and
nous contrast agent through the cerebral vascular with no exposure to ionizing radiation. Images can
bed. This change in density is proportional to the be generated in any orientation. MRI has high soft
concentration of the contrast agent, and can there- tissue resolution and sensitivity to tissue edema.
fore be used to determine flow information. CT per- MRI has a higher sensitivity than CT to brainstem
fusion-imaging techniques have been used to and cerebellar ischemia, making it of particular use
calculate maps of tracer transit time and, more in the posterior fossa. MR images can be ‘weighted’
recently, perfused cerebral blood volume.30,31 A to augment contrast between tissue types.
further CT perfusion imaging technique used in Furthermore, specialized pulse sequences have
stroke is xenon enhanced CT (XeCT).32–34 However, been developed to highlight specific tissue proper-
xenon is a mild short acting anesthetic and the use ties. A number of different sequences can be
of XeCT is not widespread. obtained at the same imaging session, providing
While the clinical role of CTA and CT perfusion more information than from any single sequence
imaging is still yet to be defined, these techniques alone. For example, information provided by T1-
have enormous potential in acute stroke. As almost and T2-weighted imaging can be augmented by that
all patients have CT scans, clinically useful vascular provided by susceptibility-weighted sequences
and blood flow information can be obtained with (sensitive to hemorrhage), diffusion-weighted
little added time or expense. In addition, CTA is less imaging (tissue destined to infarct without prompt
invasive than conventional angiography and both intervention), MR angiography (vascular anatomy),
CTA and perfusion CT can be performed in patients MR perfusion imaging (cerebral perfusion), and MR
who are critically ill or claustrophobic, or who have spectroscopy (concentration of specific cerebral
metallic foreign bodies such as pacemakers. metabolites). The technical aspects and role of MRI
Multimodal CT may therefore directly compete in stroke will be discussed in later chapters.
with the MR imaging techniques discussed in later However, MRI has a number of limitations in
chapters. However, CT is unable to delineate hypo- stroke (Table 2.2). First, MRI is approximately twice
perfused but potentially salvageable tissue of the as expensive as CT. Claustrophobic patients may
ischemic penumbra. have difficulties tolerating the studies. MRI is
contraindicated in patients with fragments of
metal, for example within the eyes, and in patients
20 P. Alan Barber and Stephen M. Davis

with intracranial aneurysm clips, cardiac pacemak-

ers or any other electrically or magnetically active
implanted devices that could interact with the mag-
netic field.

Early ischemic changes are subtle

(a)
Ischemic changes may be seen in conventional MR
studies within 1 to 2 hours of stroke onset. The ear-
liest findings include loss of the normal flow void
within major intracranial vessels and the presence
of arterial enhancement if contrast has been
used.35,36 The first morphological changes are due
to the development of parenchymal swelling with
effacement of cortical sulci and distortion of the
ventricular system. These changes are first seen in
(b)
T1-weighted sequences and may be present in up to
half of patients within 6 hours.36 This early swelling
occurs without signal change and is most likely
related to the onset of cytotoxic edema, which can
develop within minutes in experimental ischemia.37
Signal changes only appear with the development
of vasogenic edema and are not usually found
before eight hours on T2-weighted sequences or 16
(c)
hours on T1-weighted sequences.
As a consequence, the overall sensitivity of MRI
for ischemia is low in the first few hours following
the onset of symptoms.36,38,39 Conventional MRI is
also subject to the same limitations as cerebral Fig. 2.3. Conventional magnetic resonance imaging
sequences may not differentiate acute from chronic infarc-
CT when it comes to delineating acute from
tion. Conventional T2-weighted imaging (T2-WI) (A) and
chronic infarction, and stratifying patients
diffusion-weighted imaging (DWI) (B) in three patients with
according to infarct presence, location and size
subcortical infarction imaged at 11 hours (a), 10 hours (b)
prior to therapy or randomization in acute stroke and 10.5 hours (c) after the onset of symptoms. Ischemic
trials (Fig. 2.3). changes are seen in the periventricular and subcortical white
matter on T2-WI but there is no way of distinguishing acute
from chronic infarction. Hyperintense regions of acute
MRI in intracerebral hemorrhage
infarction are clearly seen on DWI.
There has been a widespread belief that conven-
tional MRI is less sensitive than CT in the detection
of intracerebral hemorrhage. Much of this pessi-
mism is based on a number of early trials using MR imaging may detect subarachnoid hemorrhage.45
scanners with low field strengths. However, there is However, until large randomized controlled trials
increasing evidence that echoplanar gradient-echo confirm the ability of MRI to detect intraparenchy-
T2*-weighted imaging, also termed susceptibility mal and subarachnoid hemorrhage, a screening CT
weighted imaging, is reliable in the detection of must be considered in all stroke patients. The
acute intraparenchymal hemorrhage.40–44 In addi- expense of any screening CT must be added to the
tion, fluid-attenuated inversion recovery (FLAIR) cost of performing MRI.
 Limitations of current brain imaging modalities in stroke 21

Table 2.3. Advantages and limitations of single photon emission computed tomography (SPECT) in stroke

Advantages of SPECT in stroke Limitations of SPECT in stroke

Generally well tolerated Technically and logistically complex

Images cerebral perfusion Exposure to ionizing radiation
Can demonstrate collateral flow or remote phenomena Still require CT or MRI imaging
such as diaschisis
Can demonstrate ischemic tissue while CT and No standardized image acquisition or analysis protocols
conventional MRI are normal
With 99Tc-HMPAO, images reflect perfusion at the time Unable to delineate the ischemic penumbra
of injection, so that scanning can be delayed
SPECT has no established clinical role in stroke

Echoplanar magnetic resonance imaging within the body. This raises concerns about the
generation of cardiac dysrhythmias or tetanic
Echoplanar MRI (EPI) offers advantages over con-
muscle contractions and some patients have
ventional MRI or CT. Echoplanar images can be
reported mild twitching or pain. However, the
obtained after a single measurement or shot, so that
current frequencies for switched gradient fields are
all data may be collected after a single excita-
usually well below the threshold for neuromuscular
tion.46–48 This is in contrast to conventional MR
stimulation.48–50
spin- and gradient-echo sequences in which only a
small portion of the data required to construct an
image is collected after each excitation. As a result,
EPI enables whole brain imaging in seconds. This in Single photon emission computed tomography
turn has facilitated the development of functional
brain imaging. EPI may be sensitized to flow and Single photon emission computed tomography
diffusion in the same way as conventional spin- (SPECT) is a rapid, relatively non-invasive perfu-
echo or gradient-echo imaging sequences.47 sion imaging modality that has been used since the
However, because of EPI’s very rapid imaging capa- late 1970s to study cerebral blood flow changes.
bilities, whole brain perfusion-weighted imaging SPECT is particularly well suited to the investiga-
(PWI) and diffusion-weighted imaging (DWI) can tion of stroke patients (Table 2.3). SPECT can
be obtained within minutes. This rapid imaging provide assessments of the degree and extent of
capability renders EPI relatively insensitive to ischemia, give an indication of likely underlying
movement artefact. pathogenesis, and demonstrate the presence of col-
EPI has several limitations in stroke patients. lateral flow and remote physiological phenomena
Firstly, it is as expensive and subject to the same such as diaschisis (Fig. 2.4, see colour plate
contraindications as conventional MRI. EPI is sen- section). However, SPECT is technically and logisti-
sitive to chemical shift artefacts that require the use cally complex, and at present has no established
of fat suppression corrections. Susceptibility arte- clinical role in stroke. A full discussion of the limita-
facts, which result in image distortions, are a partic- tions of SPECT in stroke requires some understand-
ular problem in the posterior fossa and around the ing of its underlying principles. These will be
paranasal sinuses. EPI also has the potential to gen- summarized in the following section.
erate electric currents, which is related to the rate of
switching of the magnetic field when rapidly alter-
Acquisition of SPECT studies
nating radiofrequency pulses are applied. A change
in magnetic field causes a change in electric field Perfusion SPECT acquires three-dimensional data
and results in the generation of electric currents that represent regional perfusion by localizing
22 P. Alan Barber and Stephen M. Davis

gamma ray-emitting radiotracers. These radiotracers isopropyl-p-iodoamphetamine (IMP, Spectamine)

are distributed in the brain according to the rate of and the inert gas 133xenon have been largely super-
CBF. The emitted photons are then detected using seded by the newer radiotracers.
gamma cameras or one of several different ring-
type gamma-ray detector systems. These may be
SPECT analysis
rotating gamma detector arrays, fixed detector
systems, or single or multihead camera systems. SPECT studies can provide both semiquantitative
Image acquisition time depends on the desired and qualitative assessments of cerebral perfusion.
quality of the image, as well as the imaging system This is in contrast to PET in which absolute meas-
and radiotracer used. In general, high-resolution ures of blood flow can be determined.
images can be obtained within 20–30 minutes.51 Semiquantitative SPECT analysis takes into account
After data acquisition, individual count rates at the number of radioactive ‘counts’ within a region of
each location are calculated and filtered to avoid interest, expressed as a fraction of the total counts
excessive amplification of noise. Computer gener- within the brain or with respect to some unaffected
ated reconstruction then enables data to be pre- region such as the contralateral hemisphere or cere-
sented three-dimensionally or as a series of bellum.59 This requires high-resolution whole brain
two-dimensional slices in the axial, coronal or sag- imaging, so that image acquisition and analysis may
ittal planes. An image consists of a matrix of pixels, take 30 minutes or more. Simpler qualitative acqui-
each reflecting the number of ‘counts’ emitted by sition protocols have been developed that require
the gamma rays.52 In areas of reduced perfusion, fewer slices with lower resolution. These protocols
less radiotracer is emitted and consequently the can be combined with visual analysis so that perfu-
region appears less bright. sion may be reported as simply being normal, low,
absent or high.60 This results in shorter acquisition
and analysis times. Qualitative SPECT measures
Radiolabelled tracers
have been shown to correlate with measured hypo-
The most commonly used tracer for cerebral perfu- perfusion volumes, and both stroke severity and
sion SPECT is 99m-Technetium hexamethylpropyle- outcome.60,61,62
neamine oxime (99Tc-HMPAO, exametazime,
Ceretec, Amersham International). 99Tc-HMPAO is
SPECT in stroke
a commercially available, low molecular weight,
lipophilic macrocyclic amine. It is less expensive With acute cerebral ischemia, up to 90% of SPECT
and offers better spatial resolution than earlier scans will be abnormal within eight hours of
radiotracers. 99Tc-HMPAO is unstable in vitro and symptom onset. SPECT may demonstrate the pres-
needs to be injected soon after preparation, which ence, location and degree of ischemia at a time
assuming technicians are available, takes approxi- when the changes on CT or MRI are subtle or
mately 30 minutes. 99Tc-HMPAO has an in vivo inconclusive.51,63 An exception to this is with
half-life of approximately 6 hours so that imaging lacunar infarction in which SPECT studies are often
may proceed immediately or be delayed for several normal. The degree of hypoperfusion or volume of
hours with images reflecting the state of cerebral a hypoperfused region also correlate with clinical
perfusion at the time of the injection.53–56 The in state, stroke outcome and final infarct size.64–69 This
vivo retention of 99Tc-HMPAO prevents early repeat prognostic information is over and above that pro-
imaging and results in a greater exposure to radia- vided by clinical scale scores alone.60,70–72
tion than other tracers.57,58 99Tc-ethyl cysteinate SPECT studies may facilitate patient targeted
dimer (ECD, Neurolite) is also used in perfusion acute stroke therapy. Ueda et al.73 found that
SPECT and has the advantage over 99Tc-HMPAO of ischemic tissue with CBF less than 35% of cerebel-
being stable in vitro for approximately 6 hours.51 lar flow may be salvaged if intra-arterial thromboly-
Other perfusion SPECT radiotracers such as 123I N- sis is initiated up to 5 hours after stroke onset, but is
 Limitations of current brain imaging modalities in stroke 23

at risk of hemorrhagic transformation after this no widespread and standardized SPECT image
time. SPECT may help exclude patients from partic- acquisition or analysis protocol. This has made it
ular acute stroke therapies. For example, patients difficult to compare results obtained by different
with increased isolated radioisotope uptake have a centres. As a consequence, the use of SPECT in
favourable outcome. These patients are likely to large multicentre acute stroke therapy trials has
have reperfused at the time of imaging and should been limited.80
avoid thrombolytic therapy. Patients with a focal SPECT is unable to assess cerebral metabolism. It
absence of perfusion are likely to have vessel occlu- is therefore unable to determine the presence of
sion without collateral flow and may be at risk of hypoperfused yet still potentially viable penumbral
hemorrhagic transformation or death following tissue (Fig. 2.6, see colour plate section). This infor-
thromobolytic therapy.74–77 The feasibility of using mation can only be inferred in retrospect by deter-
SPECT to help decide acute stroke therapy has been mining whether any reperfusion was maintained at
demonstrated by a number of groups. Serial SPECT outcome (nutritional reperfusion). Thus, SPECT
may also be used to monitor spontaneous changes cannot be used to limit therapy to only those with
or therapy-induced changes in perfusion over time potentially salvageable ischemic tissue.
(Fig. 2.5, see colour plate section).62,75,78–81

Positron emission tomography

SPECT limitations

Perfusion SPECT is a generally safe and well- Positron emission tomography (PET) has provided
tolerated procedure. One of the major concerns is major insights into the response of cerebral tissues
exposure to ionizing radiation. The International to reduced cerebral perfusion. In stroke, PET
Commission on Radiological Protection has devel- enables the simultaneous measurement of regional
oped a measure, the effective dose equivalent CBF, the consumption of oxygen with the cerebral
(EDE), which takes into account the relative radio- metabolic rate of oxygen (CMRO2), the consump-
sensitivity of each organ and tissue.82 The EDE for tion of glucose with the cerebral metabolic rate of
99
Tc-HMPAO of an injected dose of 500 MBq, with glucose (CMRglc), and the oxygen extraction frac-
bladder voiding every 3.5 hours, is 6.9 mSv. This is tion (OEF). These measures permit the delineation
similar to that received from a radionuclide bone of CBF thresholds for electrical and structural
scan and is 43% of the average annual background failure and as such PET has enabled the character-
radiation in the United States.51 SPECT is contrain- ization of the ischemic penumbra. The role of PET
dicated in pregnancy. in imaging the ischemic penumbra will be dis-
A further major limitation of perfusion SPECT is cussed in more detail in Chapter 15. However, while
poor image quality compared to CT or MR imaging. it has been a useful research tool, PET is not well
This is the result of the effects of photon attenua- suited to the acute investigation of stroke patients
tion and scatter, inadequate spatial resolution and (Table 2.4).
partial volume effects. Structural imaging with CT
or MRI must also be performed to exclude intracra-
Acquisition of PET images
nial hemorrhage and other non-stroke pathologies.
Thus two separate imaging sessions, usually in With PET imaging, patients are given radionuclide
different hospital departments, are required. This tracers that are radioisotope labelled biological
adds to the time needed for acute investigation and molecules, e.g. 11C-CO2, 15O-O2, 15O-H2O, 18F-
delays the institution of any therapy. fluorodeoxyglucose (18F-FDG), or drugs, e.g. 18F-
SPECT is more operator dependent than CT or fluoromisonidazole (18F-FMISO). As these tracer
MRI. Thus, the interpretation of results often isotopes decay, a positron is emitted. A positron is a
requires a close collaboration between nuclear subatomic particle with the same mass as an
medicine physicians and clinicians.51There is also electron but with a positive, rather than negative,
24 P. Alan Barber and Stephen M. Davis

Table 2.4. Advantages and limitations of positron emission tomography (PET) in stroke

Advantages of PET in stroke Limitations of PET in stroke

Quantification of cerebral blood flow Expensive

Demonstration of cerebral metabolism Exposure to ionizing radiation
Enables mapping of neuroreceptors Technically and logistically complex
Demonstration of the ischemic penumbra Limited to research or large tertiary institutions
Still require CT or MRI imaging
No standardized image acquisition or analysis protocols
PET has no established clinical role in stroke

charge. The positron then interacts with an electron rapidly converted to15O––H2O by red blood cell car-
in a matter–antimatter reaction, resulting in the bonic anhydrase. However, 15O––H2O based tech-
annihilation of both particles and the release of two niques may underestimate CBF as a result of the
gamma rays (photons). The gamma rays diverge incomplete permeability of the blood–brain barrier
from the site in opposite directions and are to H2O and incomplete tissue extraction of H2O at
detected by one of a large number of external high flow rates.84
detector pairs configured in one or more rings. The two main techniques for measuring CBF are
By configuring the system so that only the near the steady state and autoradiographic methods. In
simultaneous arrival of these photons is detected the steady state method, tracer is administered
(coincidence detection), only those photons arising until an equilibrium is reached. At this point, the
from between the detectors will be recorded. Not all amount of 15O––H2O entering the brain in arterial
photon pairs will reach the detectors because of a blood is equal to the amount lost to radioactive
change in direction (Compton scatter), or absorption decay and venous outflow. Simultaneous measure-
by intervening tissues. Accurate quantification of ment of arterial blood radioactivity enables quan-
images requires correction for this absorption and tification of regional CBF from PET imaging. The
scatter of photons. Two- or three-dimensional images autoradiographic method of CBF measurement
can then be generated according to the distribution of utilizes a single bolus of tracer (15O––H2O or
regional radioactivity. Mathematical models that 15
O––CO2) followed by PET scanning.85–87 Tracer
relate radionuclide tissue concentration to the activity in the arterial blood is also measured, with
physiological variable under study are then used. CBF values determined assuming a constant blood:
These models attempt to account for tracer delivery brain partition coefficient for the tracer. Each of
to tissues, tracer distribution and metabolism within these techniques have advantages and disadvan-
the tissues, the effects of recirculation of both metab- tages which are beyond the scope of this chapter
olized and non-metabolized tracer, and the amount and will not be discussed further.
of tracer remaining within the bloodstream.

Cerebral blood volume

Cerebral blood flow
Cerebral blood volume (CBV) can be measured
Cerebral blood flow (CBF) is the most useful using 11C or 15O labelled CO. The CO binds irrever-
physiological parameter to be measured by PET in sibly to hemoglobin forming labelled carboxyhe-
stroke (Fig. 2.7, see colour plate section). The meas- moglobin. Once equilibrium is reached, the
urement of CBF is based on the principles of inert, measured peripheral blood radioactivity will be
freely diffusible tracers.83 Most PET CBF studies are proportional to the regional carboxyhemoglobin
performed using either intravenously injected concentration, red cell mass and CBV. This enables
15
O––H2O or inhaled 15O––CO2; where 15O––CO2 is quantification of CBV after adjustments are made
 Limitations of current brain imaging modalities in stroke 25

for the hematocrit in peripheral blood and cerebral patients. Protocols typically require the placement
microvessels. CBV can then be used to correct other of an arterial catheter to monitor plasma concentra-
PET measurements such as the CMRO2 and OEF, for tions of radiolabelled tracers. Difficulties related to
radiotracer not extracted by the tissues. the use of inhalational tracers (e.g. 15O-labelled O2
or CO2) are common and may result from inability
to comprehend the required task. Other problems
Cerebral metabolic rate of oxygen and oxygen
arise from comorbidities affecting respiratory func-
extraction fraction
tion and unstable blood pressure.
Measures of cerebral oxygen metabolism are PET is expensive and requires specialized equip-
usually measured using a steady-state method after ment and staff. The short half-life of the commonly
the inhalation of 15O–O2.88 Tissue radioactivity is used PET radioisotopes (t ½ for 15O 2 min, 11C 20
proportional to the amount of oxygen extracted min, 18F 110 min) mean that they must be gener-
from the blood (OEF). The CMRO2 can then be cal- ated by a cyclotron that is either at or near the PET
culated from the relationship between OEF and scanner. This contributes to the significant techni-
CBF, described by the equation: cal and logistical complexities associated with
scanning patients at the acute stage. As a result,
CRMO2 ⫽CBF⫻OEF⫻arterial O2 content
PET scanners are limited to a small number of large
Tracer that remains bound to hemoglobin can tertiary hospitals and research centres. The acquisi-
result in an overestimation of OEF and CMRO2. This tion and analysis of PET images is also time con-
error is greatest at where CBF is low, such as suming. Additional time is needed with techniques
ischemic or infarcted regions but can be corrected such as 18F-FMISO PET, which requires a delay
by the independent measurement of CBV. Cerebral between tracer injection and imaging.
oxygen metabolism can also be measured using a As a result of these difficulties, it is unlikely that
single breath inhalation method,89 and a dynamic PET will have a place in the routine investigation of
method in which dynamic PET scans are obtained acute stroke patients in the foreseeable future. To
after single breath, multiple breath, or continuous date, there have been few studies investigating the
inhalation of labelled oxygen.90 potential role of PET in acute therapeutic interven-
tion. Most of these have examined small numbers of
Other PET techniques used to study cerebral patients with considerable variations in the time
ischemia interval between symptom onset and PET
Additional PET techniques used to study cerebral studies.95–97 The only data on the effect of thromb-
ischemia have included the measurement of olysis in humans studied with PET comes from
glucose metabolism (CMRglc) using a modification Heiss et al.,98 who reported on 12 stroke patients
of the autoradiographic method described by studied with PET either before or during the admin-
Sokoloff and colleagues.91 Benzodiazepine recep- istration of alteplase. None the less, PET has pro-
tors can be mapped using 11C-flumazenil.92 Studies vided an invaluable contribution to our knowledge
using the hypoxia marker 18F-fluoromisonidazole of stroke pathophysiology, particularly the charac-
have been reported as a potential method for terization of the ischemic penumbra.
directly identifying the ischemic penumbra.93

Conclusions
Limitations of PET

PET is a generally safe procedure. The radiation Of the imaging modalities described above, only CT
exposure in a typical study is similar to that received and MRI have established roles in the clinical
in many other routine imaging studies.94 A moder- setting. However, both CT and MRI have low sensi-
ate degree of patient cooperation is required, which tivities for acute ischemia in the hyperacute phase
can be a problem in unwell, aphasic or unconscious and neither imaging modality is able to identify
26 P. Alan Barber and Stephen M. Davis

hypoperfused but potentially salvageable ischemic 9 Tomura N, Uemura K, Inugami A, Fujita H, Higano S,
tissue. SPECT and PET have added significantly to Shishido F. Early CT finding in cerebral infarction:
our knowledge of the pathophysiology of stroke. obscuration of the lentiform nucleus. Radiology 1988;
However for logistic reasons, neither is particularly 168: 463–467.
suited to the acute evaluation of stroke patients. 10 von Kummer R, Nolte PN, Schnittger H, Thron A,
Ringelstein EB. Detectability of cerebral hemisphere
PET is the only established method of identifying
ischaemic infarcts by CT within 6 h of stroke.
the penumbra but its cost, technical complexity
Neuroradiology 1994; 38: 31–33.
and limited availability mean it is far from suitable
11 von Kummer R, Allen KL, Holle R. Acute stroke: use-
for routine clinical use. While SPECT is certainly fulness of early CT findings before thrombolytic
less expensive and more widely available than PET, therapy. Radiology 1997; 205: 327–333.
there is no accepted standardized image acquisi- 12 Donnan GD, Tress B, Bladin P. A prospective study of
tion and analysis protocol. lacunar infarction using computerised tomography.
Neurology 1982; 32: 49–56.
13 Bamford J, Sandercock P, Jones L, Warlow C. The
natural history of lacunar infarction: the Oxfordshire
AC K N OW L E D G E M E N T S Community Stroke Project. Stroke 1987; 18: 545–551.
14 Lindgren A, Norrving B, Rudling O, Johansson BB.
We thank Dr Stephen Read PhD, FRACP for his Comparison of clinical and neuroradiological findings
in first-ever stroke. A population-based study. Stroke
advice on the preparation of this chapter.
1994; 25: 1371–1377.
15 Hakim AM, Ryder-Cooke A, Melanson D. Sequential
computerized tomographic appearance of strokes.
Stroke 1983; 14: 893–897.
REFERENCES 16 Becker H, Desch H, Hacker H, Pencz A. CT fogging
effect with ischemic cerebral infarcts. Neuroradiology
1 Madden KP, Karanjia PN, Adams HP, Jr., Clarke WR. 1979; 18: 185–192.
Accuracy of initial stroke subtype diagnosis in the 17 Skriver EB, Olsen TS. Transient disappearance of cere-
TOAST study. Trial of ORG 10172 in acute stroke treat- bral infarcts on CT scan, the so-called fogging effect.
ment. Neurology 1995; 45: 1975–1979. Neuroradiology 1981; 22: 61–65.
2 Toni D, Fiorelli M, De Michele M et al. Clinical and 18 Horowitz SH, Zito JL, Dinnarumma R, Patel M, Alvir J.
prognostic correlates of stroke subtype misdiagnosis Computed tomographic-angiographic findings within
within 12 hours from onset. Stroke 1995; 26: the first five hours of cerebral infarction. Stroke 1991;
1837–1840. 22: 1245–1253.
3 Kidwell CS, Alger JR, Di Salle F et al. Diffusion MRI in 19 von Kummer R, Holle R, Gizyska U et al. Interobserver
patients with transient ischemic attacks. Stroke 1999; agreement in assessing early CT signs of middle cere-
30: 1174–1180. bral artery infarction. Am J Neuroradiol 1996; 17:
4 Warlow CP, Dennis MS, van Gijn J et al. Stroke: A 1743–1748.
Practical Guide to Management. 2nd edn. Oxford: 20 Schriger DL, Kalafut M, Starkman S, Krueger M, Saver
Blackwell Science Ltd; 2001. JL. Cranial computed tomography interpretation in
5 Gilman S. Imaging of the brain. First of two parts. N acute stroke: physician accuracy in determining eli-
Engl J Med 1998; 338: 812–820. gibility for thrombolytic therapy. J Am Med Assoc 1998;
6 Royal College of Radiologists. Making the Best Use of a 279: 1293–1297.
Department of Clinical Radiology. 3rd edn. London: 21 Marks MP, Holmgren EB, Fox A, Patel S, von Kummer
Royal College of Radiologists; 1995. R, Froehlich J. Evaluation of early computed tomo-
7 Garcia JH. Experimental ischemic stroke: a review. graphic findings in acute ischemic stroke. Stroke 1999;
Stroke 1984; 15: 5–14. 30: 389–392.
8 Unger E, Littlefield J, Gado M. Water content and 22 Adams HP, Brott TG, Furlan AJ et al. Guidelines for
water structure in CT and MR signal changes: possible thrombolytic therapy for acute stroke: A supplement
influence in detection of early stroke. Am J to the guidelines for the management of patients with
Neuroradiol 1988; 9: 687–691. acute ischemic stroke. A statement for healthcare pro-
 Limitations of current brain imaging modalities in stroke 27

fessionals from a special writing group of the stroke 35 Biller J, Yuh WTC, Mitchell GW, Bruno A, Adams HP,
council, American Heart Association. Stroke 1996; 27: Jr. Early diagnosis of basilar artery occlusion using
1711–1718. magnetic resonance imaging. Stroke 1988; 19:
23 Neurology Report of the Quality Standards Committee 297–306.
of the American Academy of Neurology. Practice advi- 36 Yuh WTC, Crain MR, Loes DJ, Greene GM, Ryals TJ,
sory: Thrombolytic therapy for acute ischemic stroke – Sato Y. MR imaging of cerebral ischemia: findings in
summary statement. Neurology 1996; 47: 835–839. the first 24 hours. Am J Neuroradiol 1991; 12: 621–629.
24 Mies G, Auer L, Ebhardt G, Traupe H, Heiss W. Flow 37 Schuier FJ, Hossmann K-A. Experimental brain
and neuronal density in tissue surrounding chronic infarcts in cats II: ischemic brain edema. Stroke 1980;
infarction. Stroke 1983; 14: 22–27. 11: 593–601.
25 Barber PA, Darby DG, Desmond PM et al. 38 Spetzler RF, Zabramksi JM, Kaufman B, Yeung HN.
Identification of major ischemic change: diffusion- Acute NMR changes during MCA occlusion: a prelimi-
weighted imaging versus computed tomography. nary study in primates. Stroke 1983; 14: 185–191.
Stroke 1999; 30: 2059–2065. 39 Brant-Zawadzki M, Pereira B, Weinstein P et al. MR
26 Knauth M, von Kummer R, Jansen O, Hahnel S, imaging of acute experimental ischemia in cats. Am J
Dorfler A, Sartor K. Potential of CT angiography in Neuroradiol 1986; 158: 701–705.
acute ischemic stroke. Am J Neuroradiol. 1997; 18: 40 Patel R, Edelman R, Warach S. Detection of hyper-
1001–1010. acute primary intraparenchymal hemorrhage by
27 Shrier DA, Tanaka H, Numaguchi Y, Konno S, Patel U, magnetic resonance imaging. Stroke 1996; 27:
Shibata D. CT angiography in the evaluation of acute 2321–2324.
stroke. Am J Neuroradiol 1997; 18: 1011–1020. 41 Atlas SW, Thulborn KR. MR detection of hyperacute
28 Na DG, Byun HS, Lee KH et al. Acute occlusion of the parenchymal hemorrhage of the brain. Am J
middle cerebral artery: early evaluation with triphasic Neuroradiol 1998; 19: 1471–1507.
helical CT–preliminary results. Radiology 1998; 207: 42 Schellinger PD, Jansen O, Fiebach JB, Hacke W, Sartor
113–122. K. A standardized MRI stroke protocol: comparison
29 Wildermuth S, Knauth M, Brandt T, Winter R, Sartor K, with CT in hyperacute intracerebral hemorrhage.
Hacke W. Role of CT angiography in patient selection Stroke 1999; 30: 765–768.
for thrombolytic therapy in acute hemispheric stroke. 43 Linfante I, Llinas RH, Caplan L, Warach S. MRI fea-
Stroke 1998; 29: 935–938. tures of intracerebral hemorrhage within 2 hours from
30 Muzelaar JP, Fatouros PP, Schroder ML. A new method symptom onset. Stroke 1999; 30: 2263–2267.
for quantitative regional cerebral blood volume meas- 44 Ebisu T, Tanaka C, Umeda M et al. Hemorrhagic and
urements using computed tomography. Stroke 1997; nonhemorrhagic stroke: diagnosis with diffusion-
28: 1998–2005. weighted and T2-weighted echoplanar MR imaging.
31 Hunter GJ, Hamberg LM, Ponzo JA et al. Assessment Radiology 1997; 203: 823–828.
of cerebral perfusion and arterial anatomy in hyper- 45 Singer MB, Atlas SW, Drayer BP. Subarachnoid space
acute stroke with three-dimensional functional CT: disease: a diagnosis with fluid-attenuated inversion
early clinical results. Am J Neuroradiol 1998; recovery MR imaging and comparison with gadolin-
19: 29–37. ium-enhanced spin-echo MR imaging – blinded
32 Gur D, Good WF, Wolfson SK, Jr., Yonas H, Shabason L. reader study. Radiology 1998; 208: 417–422.
In vivo mapping of local cerebral blood flow by xenon- 46 Mansfield P. Multi-planar image formation using NMR
enhanced computed tomography. Science 1982; 215: spin echoes. J Phys C 1977; 10: 349–352.
1267–1268. 47 Turner R, Le Bihan D, Chesnick AS. Echo-planar imaging
33 Yonas H, Darby JM, Marks EC, Durham SR, Maxwell C. of diffusion and perfusion. MRM 1991; 19: 247–253.
CBF measured by Xe-CT: approach to analysis and 48 Edelman R, Wielopolski P, Schmitt F. Echoplanar MR
normal values. J Cereb Blood Flow Metab 1991; 11: imaging. Radiology 1994; 192: 600–612.
716–725. 49 Kanal E. An overview of electromagnetic safety con-
34 Firlik AD, Kaufmann AM, Weschler LR, Firlik KS, siderations associated with magnetic resonance
Furkui MB, Yonas H. Quantitative cerebral blood flow imaging. Ann NY Acad Sci 1992; 649: 202–224.
determinations in acute ischemic stroke. Relationship 50 Athey T. Current FDA guidance for MR patient expo-
to computed tomography and angiography. Stroke sure and considerations for the future. Ann NY Acad
1997; 28: 2208–2213. Sci 1992; 649: 242–257.
28 P. Alan Barber and Stephen M. Davis

51 Report of the Therapeutics and Technology 64 Rango M, Candelise L, Perani D et al. Cortical
Assessment Subcommittee of the American Academy pathophysiology and clinical neurologic abnormal-
of Neurology. Assessment of brain SPECT. Neurology. ities in acute cerebral ischemia. A serial study with
1996; 46: 278–285. single photon emission computed tomography. Arch
52 Eberl S, Zimmerman RD. The computer and its appli- Neurol 1989; 46: 1318–1322.
cation in nuclear medicine. In: Murray I, Ell P, eds. 65 Raynaud C, Rancurel G, Samson Y et al.
Nuclear Medicine in Clinical Diagnosis and Pathophysiologic study of chronic infarcts with I-123
Treatment. Edinburgh: Churchill Livingstone; 1994: isopropyl iodo-amphetamine (IMP): the importance
1299–1314. of periinfact area. Stroke 1987; 18: 21–29.
53 Costa DC, Ell PJ, Cullum ID, Jarritt PH. The in vivo dis- 66 Bushnell DL, Gupta S, Micoch AG, Barnes WE.
tribution of 99Tc-HM-PAO in normal man. Nucl Med Prediction of language and neurologic recovery after
Commun 1986; 7: 647–658. cerebral infarction with SPECT imaging using N-iso-
54 Lassen NA, Andersen AR, Friberg L, Paulson OB. The propyl-p-(I-123) iodoamphetamine. Arch Neurol 1989;
retention of [99m-Tc]-,-HM-PAO in the human brain 46: 665–669.
after intracarotid bolus injection: a kinetic analysis. J 67 Launes J, Nikkinen P, Lindroth L, Brownell AL,
Cereb Blood Flow Metab 1988; 8: S13–S22. Liewendahl K, Ivanainen M. Brain perfusion defect
55 Neirinck RD, Burke JF, Harrison RC, Forster AM, size in SPECT predicts outcome in cerebral infarction.
Andersen AR, Lassen NA. The retention mechanism Nucl Med Commun 1989; 10: 891–900.
of technetium-99m-HMPAO: intracellular reaction 68 Giubilei F, Lenzi GL, Di Piero V et al. Predictive value
with glutathione. J Cereb Blood Flow Metab 1988; 8: of brain perfusion single-photon emission computed
S4–S12. tomography in acute ischemic stroke. Stroke 1990; 21:
56 Holman BL, Devous MD. Functional Brain SPECT: the 895–900.
emergence of a powerful clinical method. J Nucl Med 69 Davis SM, Chua M, Lichtenstein M, Rossiter SC, Tress
1992; 33: 1888–1904. BM, Hopper JL. Perfusion changes with recovery after
57 Masdeu JC, Brass LM, Holman BL, Kushner MJ. Brain stroke. Aust NZ J Med 1993; 23: 68.
single-photon emission computed tomography. 70 Laloux P, Richelle F, Jamart J, De Coster P, Laterre C.
Neurology 1994; 44: 278–285. Comparative correlations of HMPAO SPECT indices,
58 Fayad PB, Brass LM. Single photon emission com- neurological score, and stroke subtypes with clinical
puted tomography in cerebrovascular disease. Stroke outcome in acute carotid infarcts. Stroke 1997; 26:
1991; 22: 950–954. 816–821.
59 Infeld B, Binns D, Lichtenstein M, Hopper JL, Davis 71 Bowler JV, Wade DT, Jones BE, Nijran K, Steiner TJ.
SM. Volumetric analysis of cerebral hypoperfusion on Single-photon emission computed tomography using
SPECT: validation and reliability. J Nucl Med 1997; 38: hexamethylpropyleneamine oxime in the prognosis of
1447–1453. acute cerebral infarction. Stroke 1996; 27: 82–86.
60 Alexandrov AV, Black SE, Ehrlich LE et al. Simple visual 72 Lövblad K-O, Jakob PM, Chen Q et al. Turbo spin-echo
analysis of brain perfusion on HMPAO SPECT predicts diffusion-weighted MR of ischemic stroke. Am J
early outcome in acute stroke. Stroke 1996; 27: Neuroradiol 1998; 19: 201–208.
1537–1542. 73 Ueda S, Sakaki S, Yuh WTC, Nochide I, Ohta S.
61 Alexandrov AV, Bladin CF, Ehrlich LE, Norris JW. Outcome in acute stroke with successful intra-arterial
Noninvasive assessment of intracranial perfusion thrombolysis and predictive value on initial single-
in acute cerebral ischemia. J Neuroimaging 1995; 5: photon emission-computed tomography. J Cereb
76–82. Blood Flow Metab 1999; 19: 99–108.
62 Alexandrov AV, Grotta JC, Davis SM, Lassen NA. Brain 74 Alexandrov AV, Black SE, Ehrlich LE, Caldwell CB,
SPECT and thrombolysis in acute ischemic stroke: Norris JW. Predictors of hemorrhagic transformation
time for a clinical trial. J Nucl Med 1996; 37: occurring spontaneously and on anticoagulants in
1259–1262. patients with acute ischemic stroke. Stroke 1997; 28:
63 Baird AE, Austin MC, McKay WJ, Donnan GA. 1198–1202.
Sensitivity and specificity of 99Tc-HMPAO SPECT cere- 75 Ueda T, Hatakeyama T, Kumon Y, Sakaki S, Uraoka T.
bral perfusion measurements during the first 48 hours Evaluation of risk of hemorrhagic transformation in
for the localization of cerebral infarction. Stroke 1997; local intra-arterial thrombolysis in acute ischemic
28: 976–980. stroke by initial SPECT. Stroke 1994; 25: 298–303.
 Limitations of current brain imaging modalities in stroke 29

76 Davis SM, Chua M, Lichtenstein M, Rossiter SC, 88 Frackowiak RS, Lenzi GL, Jones T, Heather JD.
Hopper JL, Tress BM. Cerebral hypoperfusion in Quantitative measurement of regional cerebral blood
stroke prognosis and brain recovery. Stroke 1993; 24: flow and oxygen metabolism in man using 15O and
1691–1696. positron emission tomography: theory, procedure,
77 Limburg M, van Royen EA, Hijdra A, de Bruine JF, and normal values. J Comput Assist Tomogr 1980; 4:
Verbeeten B, Jr. Single-photon emission computed 727–736.
tomography and early death in acute ischemic stroke. 89 Mintun MA, Raichle ME, Martin WR, Herscovitch P.
Stroke 1991; 21: 1150–1155. Brain oxygen utilization measured with O-15 radio-
78 Hanson SK, Grotta JC, Rhoades H et al. Value of single- tracers and positron emission tomography. J Nucl Med
photon emission-computed tomography in acute 1984; 25: 177–187.
stroke therapeutic trials. Stroke 1993; 24: 1322–1329. 90 Ohta S, Meyer E, Thompson CJ, Gjedde A. Oxygen
79 Overgaard K, Sperling B, Boysen G et al. Thrombolytic consumption of the living human brain measured
therapy in acute ischemic stroke. A Danish pilot study. after a single inhalation of positron emitting oxygen. J
Stroke 1993; 24: 1439–1446. Cereb Blood Flow Metab 1992; 12: 179–192.
80 Alexandrov AV, Masdeu JC, Devous MD, Black SE, 91 Sokoloff L, Reivich M, Kennedy C et al. The
Grotta JC. Brain single-photon emission CT with [14C]deoxyglucose method for the measurement of
HMPAO and safety of thrombolytic therapy in acute local cerebral glucose utilization: theory, procedure,
ischemic stroke. Proceedings of the meeting of the and normal values in the conscious and anesthetized
SPECT safe thrombolysis study collaborators and the albino rat. J Neurochem 1977; 28: 897–916.
members of the brain imaging council of nuclear 92 Heiss WD, Graf R, Fujita T et al. Early detection of irre-
medicine. Stroke 1997; 28: 1830–1834. versibly damaged ischemic tissue by flumazenil posi-
81 Grotta JC, Alexandrov AV. tPA-associated reperfusion tron emission tomography in cats. Stroke 1997; 28:
after acute stroke demonstrated by SPECT. Stroke 2045–2051; discussion 2051–2052.
1998; 29: 429–432. 93 Read SJ, Hirano T, Abbott DF et al. The fate of hypoxic
82 Eberl S, Zimmerman RD. Radiation protection and tissue on 18F-fluoromisonidazole positron emission
dosimetry in clinical practice. In: Murray I, Ell P, eds. tomography after ischemic stroke. Ann Neurol 2000;
Nuclear Medicine in Clinical Diagnosis and 48: 228–235.
Treatment. Edinburgh: Churchill Livingstone; 1994: 94 Assessment: positron emission tomography. Report of
1367–1388. the Therapeutics and Technology Assessment
83 Kety SS, Schmidt CF. The nitrous oxide method for the Subcommittee of the American Academy of
quantitative determination of cerebral blood flow in Neurology. Neurology 1991; 41: 163–167.
man: theory, procedure and normal values. J Clin 95 Hakim AM, Evans AC, Berger L et al. The effect of
Invest 1946; 25: 476–482. nimodipine on the evolution of human cerebral
84 Eichling JO, Raichle ME, Grubb RL, Jr., Ter-Pogossian infarction studied by PET. J Cereb Blood Flow Metab
MM. Evidence of the limitations of water as a freely 1989; 9: 523–534.
diffusible tracer in brain of the rhesus monkey. Circ 96 Yamauchi H, Fukuyama H, Ogawa M, Ouchi Y, Kimura
Res 1974; 35: 358–364. J. Hemodilution improves cerebral hemodynamics in
85 Kety SS. Measurement of local blood flow by the internal carotid artery occlusion. Stroke 1993; 24:
exchange of an inert, diffusible substance. Med Res 1885–1890.
1960; 8: 228–236. 97 Heiss W-D, Holthoff V, Pawlik G, Neveling M. Effect of
86 Herscovitch P, Markham J, Raichle ME. Brain blood nimodipine on regional cerebral glucose metabolism
flow measured with intravenous H2(15)O. I. Theory in patients with acute ischemic stroke as measured by
and error analysis. J Nucl Med 1983; 24: 782–789. positron emission tomography. J Cereb Blood Flow
87 Raichle ME, Martin WR, Herscovitch P, Mintun MA, Metab 1990; 10: 127–132.
Markham J. Brain blood flow measured with intrave- 98 Heiss WD, Grond M, Thiel A, von Stockhausen HM,
nous H2(15)O. II. Implementation and validation. J Rudolf J. Ischaemic brain tissue salvaged from infarc-
Nucl Med 1983; 24: 790–798. tion with alteplase. Lancet 1997; 349: 1599–1600.
Random documents with unrelated
content Scribd suggests to you:
partly eaten, and wishes otherwise to dispose of—and the buried
game is so impregnated with his scent that no other creature will
touch it. He barks at night in mid-winter days—and spreads
uneasiness among sheep, as betrayed by the bleating of ewes. He
digs in a way all his own, throwing out the soil behind him in a
slovenly heap; he noses about mole-heaps and ant-hills, and his visit
is easily detected. On soft spots he leaves his footmarks—and he
always leaves his scent behind him. Pheasants without tails tell a
story of a young fox's spring that failed to bring him a supper. Heads
of rabbits, and nothing else, in snares, rejected maws lying near by
—the disinterment of poaching cats which the keeper has buried—
these show where hungry foxes have passed. By day their presence
is revealed if a cock pheasant cries a sudden, uneasy, short alarm-
note, by the screaming of jays, and by a particular blackbird note,
which, if it does not mean stoat or cat, certainly bespeaks a fox. A
crow may be seen suddenly swooping angrily as he passes over a
field—a fox lurks there. The hidden cause for the continuous uneasy
springing of partridges is often a fox, or at least a cub amusing
himself by partridge hunting.

Muzzled by a Snare
A fox does not grow very old without learning how to take
advantage of a snarer's catch. He learns to follow up runs and visit
places where the snarer has set his snares. And he often pays the
penalty, his feet falling foul of the noose. Hunting people commonly
suppose that traps—steel gins—are the chief cause of fox-maiming,
yet not once in a blue moon is a fox trapped. But if too clever to be
caught in a trap, he is not clever enough to keep his feet out of the
brass wire of the simple snare. We came across a curious instance
showing how a fox may suffer from a snare. Hounds found a fox
which ran to ground almost at once. Men were set to work to dig
him out, and they found he was merely skin and bone, and round
his muzzle they found part of a brass snare. The wire had fixed itself
in such a way that he could scarcely open his mouth, so that he was
handicapped both in catching food and eating it. From his
appearance it was thought that he had been in this miserable plight
for a month. It had been better for the fox if hounds had found him
a month earlier.

Cunning Rascals
A fox, in emergency, will sham death to perfection. A Master of
Hounds once noosed a fox in a whip as he bolted before a terrier
from an earth. The fox appeared to have been strangled—when held
up by the scruff of the neck his eyes were seen to be closed, his
jaws gaped, and the body hung limply down from the hand. He was
placed tenderly on the ground—only to dash off into covert. To be
over-cunning is a common fault. One fox entered a fowl-house, and
amused himself by killing every bird. In departing through the hole
by which he had entered, he stuck fast, and was found hanging
dead the next morning. Another sought refuge from hounds by
jumping on to the low roof of a thatched cottage, and crawling
beneath the rafters until he could crawl no farther. It was years
before his skeleton was discovered. Some of the foxes found dead
on railway lines, by the way, have been put there after death by
vulpicides. In olden days the punishment for the crime of fox-killing
was a spell in the stocks. Vulpicides remain, but the stocks—some
would say alas!—have gone from use for ever.

A Hunting Argument
The hunting man has a hundred reasons why hunting is a blessing
to the community. He argues that hunting circulates gold every year
to the tune of seven and a half million pounds—and that this is good
for the horse trade, the forage trade, for the blacksmith, the
harness-maker, and for an army of grooms. Then hunting tends to
keep at their homes in the country wealthy people, who might
winter abroad if there were no foxes to follow. This means that
many large establishments are kept open, servants are kept in food
and wages, local tradesmen stand to benefit. Further, it is claimed
that there is little to be said against hunting—we often hear how
riders, horses, hounds, and foxes all enjoy the sport; on this point,
however, we have no direct evidence from foxes. And it is claimed
that the amount of damage done to agriculture is infinitesimal—
though farmers who have had hounds over young corn, or seeds, or
fine fields of turnips, might bring conflicting evidence to bear on the
point. Perhaps the favourite argument in favour of hunting is that
the sport is good for horse-breeding, and that the hunting-field is
the finest training school for cavalry. Gamekeepers would be among
the first to lament the abolition of fox-hunting, for if it were not for
the existence of foxes and their preservation for the hounds, few
keepers would be required to protect game. Nor would there be
those useful little sums to the keeper's credit on account of litters,
finds, and stopping.

The Clever Terrier

Nobody can persuade a gamekeeper that dogs lack reasoning
powers. We were watching a terrier at work, and she gave us a
pretty example of something very like intelligence. A pheasant was
winged, fell on a bare field, and ran for a thick dell—the terrier in
pursuit. She made one or two ineffectual attempts to gather the
bird, until within a score of yards of the dell—then she raced ahead.
She seemed to realise that there was so much cover in the dell that
direct attempts to take the bird were risky—and she proceeded to
work the pheasant to a safe distance from the cover before tackling
it again, this time effectively.
When this little terrier has marked a rabbit or a rat in a patch of
grass or brambles, her common sense tells her that if she dives in
after her quarry it may dash out unseen by her, by reason of the
grass or brambles. So she stands by, and stamps, and otherwise
tries to make her game bolt, in a way which will allow her to see the
direction; and she is seldom baffled. It is difficult to decide whether
this terrier is more or less reasonable than her kennel companion, a
retriever, when feeding-time comes. If at feeding-time the retriever
has a biscuit left over from the last meal, which she has lightly
buried, on her master's approach she will promptly disinter the
treasure, holding it out as much as to say: "Thank you, I need no
biscuit." But experiments with the terrier show that she will ever
refuse to give the slightest indication of a buried hoard. Whether she
needs a biscuit or not, she always takes one when offered, as
though she desired nothing better in the world.
A good story in proof of a retriever's reasoning powers is told by an
old-time sportsman. He was shooting beside a frozen stream, and
winged a mallard, which fell in mid-stream. His dog crashed on to
the ice, broke through it, and fought her way to the middle, where
the ice only skimmed the water. She swam round for a moment,
then broke her way to the opposite bank, paused to give a knowing
look at the thin ice, and went down stream at full speed for about
eighty yards. Running down the bank, she broke a hole in the ice
with her fore-paws, then crouched back, watching the hole. In a few
moments she made a spring and plunged in, reappearing in mid-
stream with the mallard in her mouth. There was no doubt, at least
in her master's mind, that she had broken the hole for the purpose
of catching the bird when he came up to breathe.

Born Retrieving
A keeper owned two retriever puppies who were given a curious
start in life. Their mother was shut up at home, while her master
went to shoot some rooks. She was the proud mother of five new-
born puppies, but her litter was not complete. A few rooks had been
shot, and the keeper was waiting for others to appear, when up ran
the retriever carrying a rook in her mouth; somehow she had
managed to get out, and had followed to see the sport. She was
sent back to her puppies, and directly she reached home two new
puppies were born. They were born, as one might say, retrieving.

Some Sporting Types

The most common type of gunner is the man who kills frequently,
but is not a good shot because he does not know how to take his
birds. He would double his bag if he would put every shot a foot
farther forward—that golden foot forward—if he would not fire when
in the act of turning (which must depress the gun's muzzle), and if
he would remember that driven birds on seeing a man rise
immediately and instinctively, even at right angles to their line of
flight. The keeper detests the man who continually sends him to pick
up game which has never fallen. For these knowing gentlemen, he is
a wise keeper who carries a special bird or two in his pocket, against
the time when they say, in their haughty way, "Aw, my man, kindly
pick up my bird that fell tha-ar!"
The luckiest shot we ever met was a colonel who, one windy day,
happened to be stationed by himself on a road lined by telegraph-
wires. All the birds came his way, and with ten shots he killed one.
Startled by his volleys, a bunch of passing birds blundered into the
telegraph-wires which, more deadly than the gun, claimed nine
victims. The colonel was a study in modesty when he remarked a
little later that in ten shots he had been lucky enough to bag five
brace.
 Victims of Wire
Unfortunately the best stands for partridge-driving are often behind
hedges flanked by telegraph-wires. This is specially unfortunate
when the birds see the guns just before they pass beneath the
wires. Up they go, and a whole covey may be cut to pieces at the
moment when fingers were pulling triggers. Though a brace of birds
fall dead at the sportman's feet, evidently neatly taken in front, to
the sportsman this is not the same as a brace to his gun: he would
prefer, indeed, a good old-fashioned miss.

Stoat or Weasel?
Many country people who ought to know better are hazy on the
distinction between stoats and weasels. We can forgive the Cockney
uncertainty of this sort, as we forgive him for calling rooks, and even
starlings, crows. The countryman may well confuse crows and rooks;
his safest plan when in doubt about a big black bird is to name him
rook, for in most parts crows are now scarce to the point of
extermination. But those who live in the country have as little excuse
for speaking of stoats, when they should speak of weasels, as for
mixing rabbits with hares. It is easier to tell a weasel from a stoat
than a rabbit from a hare, if one is fairly close and has a clear view.
A weasel is quite a third of the size of a stoat and a third of the
weight: the males of both weasels and stoats are about twice the
size of the females. But the outstanding distinction between stoat
and weasel is the long, black-tufted tail of the stoat, and the short,
unassuming tail of the weasel—no more conspicuous than a mole's
tail.
 "The Horrid Badger"
We have come across many curious cases of ignorance on these
points. A countryman who had dwelt with stoats and weasels all his
life, and had killed hundreds by trap and gun, yet had no idea of the
true difference. Whichever he saw, or killed and hung up by a
twisted twig, he determined to be stoat or weasel according to its
size. Then we remember a lady who kept chickens, and suffered the
loss of half a brood. She called in a passing keeper to settle the
question of the thief. After waiting a while the keeper shot a weasel
in the act of returning for another chicken. The lady of the chickens
was overjoyed at this retribution, and presented the keeper with half
a crown. Her words in making the presentation have been treasured
by the keeper: "This," said she, "is for shooting the horrid badger."

Chalk-Pit Haunts
To the old chalk-pit, where the sun is trapped and the winds are kept
at bay, come all kinds of creatures for warmth and sanctuary.
However deserted the fields of winter seem to be—however silent
and sullen—signs of life are never wanting in the chalk-pits; they are
as inns to wayfarers who search the country for a living and lodging.
Creep silently, against the wind, to the chalk-pit's edge, and in
summer or winter, sunshine or shower, on a still day or a windy, you
will catch a glimpse of some wild creature, a visitor, or one of those
who have made their home in the pit for the sake of sustenance or
shelter.
 When the Fox sleeps
The sparrow-hawk may be caught napping on some favourite perch,
as on a stunted tree, in a sheltered nook. The partridge covey may
be seen for a moment, as the birds revel in the powdery soil, roofed
by an overhanging ledge—seeing you, they go whizzing off amid a
little cloud of dust. In the dead herbage a wily old cock pheasant
crouches, who long since denied himself the luxuries and the
dangers of social life in the big woodlands: he crouches as he sees
you, but not so quickly that you may not note the sinking of his
glossy neck. Two or three rabbits scuttle off to the doors of their
burrows. Through the bushes a hare steals away. No chalk-pit is
complete without a rabbit-burrow, a blackbird, and a robin. If
hounds came more often to the chalk-pits they would save
themselves many a blank hour. There is no peace for the fox in the
coverts, but the old chalk-pit is as quiet as a church.

When Ferret meets Fox

An exciting moment for rabbiters comes if a fox bolts from a burrow
when only a rabbit is expected—so exciting a moment that if there is
a man with a gun the fox is lucky to escape a shot—especially
should he have in his mouth the quivering body of a favourite ferret.
And the ferret is lucky to come alive from a hole if he meets the fox
in the only passage by which he can leave the burrow. But ferrets
often escape if the burrow is not a proper fox-earth, but has been
used only as a temporary shelter. Even if caught in the fox's jaws
there may be hope for the ferret; we heard of one who was none
the worse for a long ride between a fox's teeth. Like dogs and cats,
foxes can be soft-mouthed if they will. We have known a fox to deal
so tenderly with a captured rabbit that it ran about after the long
jaws had released their hold; and for some time it amused its captor
as a mouse amuses a cat. A fox, when he wishes, can carry an egg
without breaking the shell.

February Rabbits
Towards the end of January rabbits begin to fall off in condition. As
food becomes less nourishing their reserve supplies of fat gradually
dwindle. But with the end of the game season the price on their
heads begins to rise: and the keeper who has hard work to meet the
expenses of a shoot looks to the rabbit-catch of February to swell
the credit side of his accounts. Most people know that a hen
pheasant is more tender and delicate to eat than a cock, though
cock and hen may be of the same age. So with rabbits—those who
sell rabbits might well charge a penny or two more for the does than
for bucks. The countryman knows that the tenderest rabbits are
those that he may skin with the least difficulty.

The Moucher's Excuse

While the gamekeeper is seldom at fault in the matter of a ready
excuse, he meets many people who are his superior in carrying ever-
ready lies on their lips. From poachers and mouchers, as the
haunters of hedge-sides are called, he might learn the lesson that no
excuse is better than a fine excuse that is shallow. One Sunday
morning a keeper, dressed in his go-to-meeting clothes—a useful
disguise—came sauntering silently down a road bounded by
unkempt hawthorn hedges. His trained ear caught the sound of a
dog careering past him on the field-side of the road: then he saw
the dog's master, who, on seeing him, set up a sudden and energetic
whistling. Of this the dog took no notice; with his nose well down,
he rushed on to a rabbit-burrow and began digging furiously. "These
hedges are full of rats," remarked the dog's master. "My dog killed
five just now." Asked what had happened to their bodies, Mr.
Moucher replied calmly, "He swallowed 'em whole." On the keeper
suggesting that there was not much chance of finding a rat in the
rabbit's burrow, the moucher agreed, called off his dog, and went his
way. In the hedges there was no sign of a rat, but a few rabbits
managed to eke out an existence, though heavily persecuted by
gentlemen of the road.

When Hounds come

The opening of the hunting season proper brings a new anxiety to
the keeper. While it opens in early November, no date is recognised.
The keeper would like to see one fixed, and he would make it after
his coverts had been shot at least once. Many shooting men would
also like to see the idea established that hounds should not come to
their woods until after the first shoots, especially where there are
many hares. Often a landowner will refuse a master's request for
permission to come his way until he has done with his coverts. The
keeper does not so much object to the hounds merely passing
through when in full cry, for then the hounds run in a compact body,
and pay no attention to game. They only disturb a line about ten
yards wide right through the woods. What disturbs every game-bird
and hare in the place is drawing a covert, particularly when scent is
bad and foxes are in evidence, but not to be forced away. Unhappy
the keeper who must throw open his coverts at all seasons while
other neighbouring coverts are closed. The prohibition of one wood
often leads to the closing of many more; and hunt officials are well
advised to break down, by every power of persuasion, all restrictions
which favour one or two keepers at the expense of brother keepers.
At any rate, we think it would be an excellent idea that the keeper
whose coverts are always open to hounds should have double the
reward paid for a find to the keeper whose coverts are open only
after Christmas.

When Hounds are gone

Those who shoot in the wake of hounds are no sportsmen. To state
a case in illustration of this: A sportsman has the shooting of a wood
bounded on one side by another's fields. In days gone by he was
glad to keep a fox for hounds, and gladly he would throw open his
wood to the hunt, in a reasonable way. In the cause of sport, he was
content that his pheasants and hares should be driven out of his
wood into his neighbour's fields and hedgerows. But when he found
that his neighbour was the sort of man to shoot in the wake of
hounds, so that the evicted creatures were given no fair chance to
return to their home-wood, but instead were shot in the afternoon
following a morning visit of hounds—he felt compelled to close his
wood to the hunt, with the natural sequence that he was soon
compelled to bar the covert to foxes also. No shooting days in the
wake of hounds should be a golden rule for all neighbourly
neighbours.

Poachers' Weapons
Of poachers there are many types; and the worst are the organised
bands that hail chiefly from colliery and manufacturing districts.
These men are murderous ruffians, and the keeper who interferes
with them carries his life in his hand. Wives look anxiously indeed for
their husbands' return when such a band is about. The gangs chiefly
practise night shooting, and pheasants are their object. But they are
as ready to fire at a keeper as at pheasants. We were shown a
single-barrelled muzzle-loading gun which a keeper had taken from
such a poacher, who had shot a roosting pheasant under his very
eyes. After the shot, the keeper went up to the man, who pointed
the gun straight at his head, threatening to fire if he advanced
another yard. But the keeper knew his man—and his gun. He knew
there had been no time for the ruffian to reload. He knocked up the
barrel, and caught his man, who in due time was sentenced to nine
months' imprisonment. Had his gun been double-barrelled, it would
have been another story, and a tragic one. A favourite weapon, and
a deadly, in these poachers' hands is a heavy stone slung in a
stocking.

Moles' Skins for Furs

For moles' skins the keeper has no sentiment. He will not part with
his skins of rare birds—but will willingly barter the prospect of
wearing a moles' skin waistcoat for the price of an ounce of shag a
skin. By catching moles he pleases the farmers, who know no more
than he himself about any good work that moles do: he frees his
rides from unsightly heaps and raised tunnellings; and now and then
his mole-traps catch a weasel. Many keepers make a fair sum of
money each year by selling moles' skins; furriers will as readily give
twopence for a skin as others threepence or sixpence. The skins, cut
close round the head, are drawn from the moles' bodies as a man
draws stockings from his legs; they are pegged out, fur downwards,
on a board, to be dried and powdered with alum, then are stuffed
with meadow hay, and packed by scores or hundreds. Perhaps no fur
is quite so soft and beautiful as the mole's; and the keeper is always
well pleased to note how well the pelts of his enemies become
women-folk's faces.

Covert-shooting Problems
To shoot while there are still many leaves on the underwood and
trees, and while there is a full muster of pheasants, or to wait until
there are fewer leaves and fewer pheasants—that often is the
question. For there are many coverts in which pheasants will not
stay after the fall of the leaf. Then the shooting man who does not
own the coverts to which his birds will betake themselves must make
the best of things, and be content to bring down more leaves than
pheasants, and often nothing but leaves. What with the showering
of leaves and the crashing of shot-pruned boughs and dead wood,
he may imagine that a pheasant must be an extra heavy bird—only
to find that not a feather has been touched. To shoot pheasants
among a crowd of leafy oaks is no simple matter—it is more difficult
than to shoot a rocketer in the open valley. One thing may be said
for this aggravating pastime; it teaches the slow shooter to be quick.

"Cocks only"—to compromise

There are good reasons for shooting coverts for the first time before
the end of November, apart from the fear of a leakage of pheasants.
A sack of corn a day will quickly swell a bill to uncomfortable
proportions. Unshot coverts also mean that the whole time of
keepers and watchers is taken up, with a string of awkward
consequences. Thus, little can be done to thin the rabbits, for fear of
disturbing the other game in the coverts. Each night some of the
hares go out, never to return. Hunting must be curtailed in self-
defence. Then again, neighbours may be shooting, and it is very
certain that what goes into your neighbour's bag cannot go into
yours. The best compromise between shooting in woods still leafy
and waiting for the sporting Christmas pheasant to soar far above
the tops of the bare trees, is to shoot "cocks only" at the first covert
shoots. This may be a perplexing plan to those not accustomed to it
—either they include a good many hens, or they let off a good many
cocks which they mistake for hens. It is a plan to make the nervous
man shoot his worst. And the keeper, as a rule, will not be found to
favour it, unless the guns are discriminating and good, and
appreciate sport more than bag. But sooner or later the day of
"cocks only" must come—why should it not come at the beginning
and be done with?

What a Cat may kill

A strange confession was made by a cat-lover concerning the cat of
her fireside. The confession was made publicly; in fact, in the
columns of an obscure local paper. It was to the effect that the cat
had brought in to her kittens, in one week, twenty-six field-mice,
nineteen rabbits, ten moles, seven young birds, and two squirrels—
all of which passed through her mistress's hands; there may have
been others not taken account of. It never seemed to enter the head
of the cat's mistress that any hurt was being done to other people's
interests by this poaching of rabbits, nor that any neighbouring
gamekeeper might read her words. It would be unfair to argue that
all cats, with or without kittens, are as bad as this one; we have
heard of cats a great deal worse. Naturally a mother cat forages far
and wide for food; but she hunts chiefly for small things, and knows
that mice and birds are more suitable for her weaning kittens than
sitting partridges and pheasants. It is that arch old villain, Sir
Thomas, who commits the crimes for which mother cats are blamed.
But the keeper has no hesitation in bringing home to all cats a
reparation, sudden and effective, for Sir Thomas's sins.

A Cockney Story
A gamekeeper friend told us, with infinite delight, this quaint little
story. If we are to believe him, he was sitting one fine September
day behind the hedge of a cornfield, thinking about the coveys
hidden in the corn, when he became aware that a lover and his lass
were sitting on the road side of the hedge, directly behind him. They
were Cockneys, and this was the first of their days of country
holiday-making. Presently the lover speaks. "Emma," says he, "just
look at this pretty fly wot's settled on me 'and." "Lor'!" says Emma,
"ain't he a daisy?" A pause follows; the lovers are silently
contemplating the beauties of the fly. Emma suggests he is out for
an airing in his racing colours—yellow and black. Then the lover calls
out in a voice of mingled amusement and pain. "Crikey!" he cries,
"ain't 'is feet 'ot?"

Hares in Small Holdings

The hare that haunts a small holding has a slender chance of dying
a natural death in ripe old age. But we have a little story of how a
small-holder was converted from hare-shooting. He was a man who
rented a meadow on the outskirts of a large village; and it chanced
that hares were much attracted to this pleasant spot. The
gamekeeper of the shooting tenant was deeply troubled by the drain
on his stock of hares caused by the small-holder; but there was little
he could do to stop the slaughter that went on at all times and
seasons, and by all manner of means. He had the good sense to
keep on friendly terms with the troublesome sportsman, and at last
he thought that some improvement might be brought about by
arranging a laugh at his expense. He stuffed a hare, and one night
set up the skin in the meadow, at a fair range from a gap in the
hedge. Early next morning the news reached the small-holder that
there was a hare in his field. Off he started with gun and dog; saw
from the gap that the hare was sitting up, "jest about a pretty little
shot," took steady aim, and fired both barrels to make sure of a kill.
How his dog retrieved a hare-skin stuffed with hay was a story that
soon became public property in the village and the neighbourhood,
and from that day forward there has been no safer place for a hare
than this man's meadow.

The Sins of the Father

The gamekeeper often picks up hints about poachers in unexpected
ways. His wife, as a rule, takes no great interest in the affairs of
game; yet every now and again she is able to tell her husband some
news that may be at once bad and good. It happened that the wife
of a highly respected gardener fell ill, and one afternoon the
keeper's wife kindly offered to take charge of her children. The
eldest child, a boy of about six, seemed to have little to say for
himself; but, as the party was walking silently along a lane, he
suddenly said in a voice that promised well to be a bass some day:
"Our muver, she do make we some good dinners." "Indeed," said the
keeper's wife, "and what does she give you for dinner?" The boy
answered eagerly and proudly: "Bunny rabbits, m'm." "Indeed," said
the keeper's wife again, "and where does mother get the bunny
rabbits?" "Please, m'm, faither buys 'em off a man as brings 'em."
"Oh! in-deed!" said the keeper's wife, and it was not long before one
more receiver of stolen rabbits was brought to justice.

The Pheasants' Roosting-Trees

When the oaks shed their leaves night has a new danger for the
roosting pheasants. They become easy targets for the gun of the
night shooter. While the leaves remain the pheasants are well
screened—and they often owe their lives to their habit of roosting in
oaks, where the leaves give shelter long after beeches are bare. On
a night of bright moonshine beeches scarcely provide any cover for
the bulky form of a roosting pheasant. No doubt it is rather for
comfort than through cunning that pheasants choose a roosting-
place in oaks. They show no cunning in choosing their oak-tree, for
they will roost night after night on some low branch overhanging a
road. They seem naturally to prefer oaks to beeches for a lodging.
Unlike most trees, oaks throw out their branches horizontally, but
beeches' branches tend to rise vertically. Their bark is smooth and
cold, but oak bark is rough, easily gripped, and warm.
When oaks have lost all their leaves the beeches provide the better
cover; for their vertical lines form some sort of screen. Even with a
full moon it is not always easy to see sleeping pheasants which go to
roost in the lower branches. It may be more difficult to see a
roosting pheasant than to shoot it—though the hardest shot a
pheasant can give is when it flies by night. Fir-trees in a pheasant
covert have a special value to the roosting birds. While unsuitable as
sleeping-places, for the birds cannot fly up through the thick twiggy
branches, nor can they see where they are going, the firs make the
more suitable roosting-trees warm and cosy, and against their dark
background it is difficult to see the pheasants, and to shoot them.
The poacher has no liking for sporting shots.

The Fox in the Storm

Wet weather is often a benefit to the fox. Like all accomplished night
thieves he is more venturesome in attacking hand-reared birds when
the wind howls and rain beats heavily down. The storm drowns what
little noise there may be from his stealthy feet; and the scent of the
birds is stronger by reason of their steaming bodies. In wet autumns
foxes take their heaviest toll of the young birds that have grown to a
fair size—the dripping trees incline the birds to sleep on the ground
long after they are able to fly, and should be flying nightly to roost.
Grave risks are run by birds that sit on their nests through wet June
nights.
 Foxes at Pheasant Shoots
Foxes are sometimes found among pheasants where wire, or string
netting, has been set up at the flushing-places, to prevent the birds
running instead of flying, and to cause them to rise and fly at a
sporting height and pace. When it is too late, and the beaters have
come to the flushing-place, the indignant "cock-ups" of the
pheasants are heard, and then they rise in a great rush, too thick
and fast for the convenience of sport. We remember one case where
a stampede of pheasants so enraged a sportsman that he ordered
his loader to bowl over the old sinner of a fox. Should a fox show
himself during the beating of a wood, it would be wise to give him
every chance to escape. What usually happens is that the beaters
force him forward with sticks and curses, and the guns drive him
back with cries of "Tally-ho!"
But the fox's appearance is disconcerting; and there is a touch of
irony in the thought that a crafty old fox, who in his time has slain
more than his share of pheasants, should yet be in at the death of
those that escaped him.

Pheasants that go to Ground

The careful gamekeeper will stop all the rabbit-holes round about
the place where he hopes that many pheasants will fall—perhaps for
fifty yards before and behind the stands of the sportsmen. Many a
pheasant is lost through going to ground in a rabbit-burrow, and
there is seldom a spade and a grub-axe at hand. The pheasant may
be winged or otherwise wounded, and if it cannot be dug out may
die a lingering death. But many a crafty old cock has revealed his
hiding-place because, while he has taken the precaution of drawing
his body into a burrow, he has forgotten his tail. Only one partridge,
in our experience, has run to ground after being winged.

Pheasants' Doomsday
A wise pheasant would go abroad before the middle of November.
He would leave the fallen beech-mast for the pigeons, and turn a
deaf ear to the persuasive whistling of the maize-laden keeper. Since
the issue of his death-warrant on October 1, the pheasant has fared
well—he has never known the want of a hearty breakfast. But
sooner or later comes a morning when he must breakfast on the
remnants of a last good supper. If he wonders why, he never thinks
he has been denied his food because a big breakfast is not good to
fly on, because a full crop will lessen his value in the eyes of the
game-dealer, and because it is intended that he shall fly high, and
give a sporting shot. So he is kept short, like a pig whose time has
come to be made into pork. But no doubt even his short life has
been worth the living.

The Hungry Retriever

We have a story of a retriever who was forced to forego breakfast on
the morning of a shoot. Retrievers, as they grow old, often grow
cunning, and we saw this one getting the better of his master in a
novel and drastic way. The old dog had grown fat, and somebody
complained that he was inclined to be lazy in his work. It was
decided that he had too much to eat, and it was to improve his
activity during a day's partridge driving that his master kept him
without breakfast, usually a heavy meal. There was a cold partridge
that came within range of the dog's nose—but his longings were not
gratified. Out in the fields the dog was sent for the first bird his
master shot, a runner. Away went the dog with unusual speed; he
picked up the bird, and then quietly sat down and made a meal of it.
Having had his breakfast, he did his work handsomely for the rest of
the day.

The Old Wood

The first covert shoot has a peculiar charm for the sportsman—
especially when the shoot is in familiar woods. There has grown a
feeling of friendship for the old rides and trees, and they seem to
offer a warmer welcome every year. He comes to the historic corner
where he failed miserably to do justice to a rush of pheasants. Here
is the opening through which his first woodcock tried to glide—in
vain. He remembers, perhaps, that even now he has that
woodcock's two pen-feathers in the depths of some ancient purse.
Here was where he scored a double at partridges hurtling through
the tree-tops—only to be beaten a moment later by a hare, slowly
cantering. Nothing has changed in the woods. They wear the same
old look of nakedness; save for a hurrying pigeon, there is the same
desolate lifelessness. Nothing stirs, but the leaf fluttering to earth;
all is dead quiet. Then in the distance is heard the prelude of the
beaters' sticks—tap, tap, tapping. The sportsman dreams, musing of
past days and their great deeds. Then a lithe moving form catches
his eye—a hare has slipped out of sight. A shot rings out, echoes
and re-echoes; another, and doubles, and clusters of shots. The old
wood is the old wood still.

Memories of Muzzle-loaders
Perhaps not many shooting men remember much about the old days
of the muzzle-loader, or could recall all the items of the
paraphernalia necessary for a fair day's sport. In spite of their
drawbacks, wonderful feats were performed by the old guns; and
certainly there was a truer ring about the word sport in the good old
times. A fancy-dress shooting party, with the sportsmen in the old-
time shooting-suits, armed with muzzle-loaders, would be
entertaining—if dangerous. How many members of the party would
arrive on the scene of action with all the appliances necessary for
the firing of a fowling-piece—powder, shot, wads, and caps? And
who would know how to load his weapon, even with powder, shot,
wads, and caps at hand? The man who did not know how to load
would be in a bad way, for, of course, no valets could be allowed on
the scene, even supposing they might know more than their
masters. Short-tempered men would be exploding perpetually in
wrath at the delays caused by the process of loading, while birds
were rising and going away—we have heard powerful language
addressed even to the modern weapon when it has been responsible
for a hitch in shooting. It is shocking even to think of what a short-
tempered man might say if he flung away an open box full of copper
caps in mistake for an empty case, or if he applied his powder-flask
to his lips and swallowed a few drachms of treble strong black
powder instead of a few drops of sloe-gin. No doubt some of the
party would suffer the misfortune of upsetting their whole supply of
shot for the day's sport. Then the short-tempered man sooner or
later would break his ramrod—others would shoot ramrods, like
arrows, into the air. At the end of the day there would be headaches
and black-and-blue shoulders. And what would be the bag?

Relics of the Great Days

The old-time gunner went out in the morning with all manner of
contrivances and implements stowed about his person. He wore a
shot-belt for distributing the weight of his lead, he carried neat little
magazines, so that he might the more easily handle his copper
percussion-caps, and he wore a wallet of leather containing such
tools as a nipple-wrench and spare parts—the nipples in the gun
might break or blow out. The careful man carried a wad-punch, and
in emergency would punch wads for his muzzle-loader out of his felt
hat or his neighbour's—what could be a more neighbourly act than
to sacrifice a pair of leather gaiters in the cause of wads? A keeper
friend treasures many relics of the great days of the old squire—
among them a curious little mirror, the glass about the size of your
little-finger nail, set at the top of a tiny brass box, small enough to
slip into the barrel of a twenty-bore. The old squire would draw this
mirror from his waistcoat pocket before the first charge was poured
into the muzzle of his gun, dropping it glass upwards down each
barrel in turn, so that he could see by the reflected light if they were
well cleaned and polished.

Cleaning a Muzzle-loader
The cleaning of a muzzle-loader was an immense undertaking. First,
the barrels were removed from the stocks, then bucketfuls of hot
water were forced through them; out would pour a stream of black,
liquid filth, having no respect for clothes or person, and smelling
abominably. Heated water was used because it cleaned away all the
foulness of the black powder, and quickly dried off. After washing,
the barrels were fixed in vices carefully padded to prevent injury,
and then they were given a hearty polishing inside with a tow-
topped rod. Great attention was paid to the locks, which were not so
well protected from water as they are to-day—they were removed
every now and then, and taken apart by means of a neat little clamp
for holding the mainspring. In those days people spoke of how many
pounds of shot they had fired—not of how many cartridges. The old-
time bags were not to be despised. One keeper, who has been in his
present place for forty odd years, told us that he can always
remember his last day's shooting with muzzle-loaders, because they
bagged the same number of hares as pheasants—218—to say
nothing of 324 rabbits. They must have performed some wonderful
feats of loading as well as shooting.

The Knowing Beater

At covert shoots beaters often behave in unaccountable ways; but it
is not every day a beater is seen crawling about on hands and
knees. A guest at a covert shoot, surprised at such a sight, inquired
about the beater's object. "Beg pardon, sir; I thought as 'ow you
was the guv'ner," said the beater, rising. A further question as to
why the guv'ner should be met on all fours brought this answer:
"Well, you see, sir, 'tis this way like—the guv'ner, 'e don't allow no
game to git up 'igh, not if 'e can anyways 'elp it. Not 'e, for 'e wops it
into any birds as rises 'ardly afore they be got on their wings like. So
you see, sir, soon as I thinks I be gittin' dangerous near 'im, I allus
reckons to be a bit careful."

Old Friends
The shepherd and the gamekeeper are men in sympathy, for one is
dependent to some extent on the other. In the eyes of the keeper,
the shepherd is one of the most important persons on a farm. And
where there is not a good understanding between the two men the
keeper will suffer loss in game, and the shepherd not only in sheep,
but in rabbits. With rabbits to spare, the keeper's first thought is of
his friend the shepherd. The shepherd is vigilant by night as well as
by day, and may watch the interests of game without detriment to
his own charge. And it is a pleasure to the keeper to run his eye
over the fold when he passes that way to see if all is well. He comes
to the rescue of many a sheep on its back that would have remained
on its back until dead without his timely aid; and he saves the
shepherd many possible disasters through the flock breaking from
the fold, when the sheep might come to destruction by over-feeding
on green-stuffs. Through the long nights of the lambing-time the
keeper may give the shepherd his company over pipes of fragrant
shag, and pots of heart-cheering ale—hands, hearts, and ale alike
made warm by the little stove in the shepherd's movable house on
wheels. Look well at a shepherd's back, and you are likely to see a
keeper's old coat.

What Shepherds enjoy

Shepherds like their pot of beer—and some of them are wondrously
fond of a fight, and so may become useful allies to the keeper when
poachers are to be dealt with. We knew a shepherd who would
always be especially retained to help the keepers of an estate at
times when pheasants were liable to be shot at night. His
appointment came about in this way: the head keeper, during the
absence of an assistant, had employed the shepherd to watch, and
had dosed him with half a gallon of beer to keep the cold out before
sending him off on duty. The beer and the night air were not without
effect; and when presently a human form came stealthily along in
the shadow of a moon-lit ride, the shepherd was in grand fighting
trim and spirit. He waited his chance, then sprang like a lion on the
intruder, gripped his throat, bore him to the earth, and belaboured
him in hearty fashion. He was about to tie him hand and foot when
he saw that he was tackling his own master from the mansion, who,
having been dining with a neighbour, had chosen to walk home by
way of his woods. So impressed was the master with the shepherd's
valour on behalf of his pheasants that he gave him a sovereign, and
retained him on the night staff at five shillings a night—and half a
gallon of beer.
 Lives of Labour
Like most country workers, shepherds and gamekeepers may go
through a long life of labour without ever taking a holiday, possibly
without thinking of one. We hear of eight-hour days for factory
workers and discussions of an ideal work-day of six or even of four
hours; but seldom a word is spoken for those country labourers, the
length of whose toil is limited only by daylight—when it is not carried
on as a matter of course into the night. Farm hands may work
through all the days of the year; for where there is stock to be fed
work is never-ceasing. Yet it is reasonable to suppose that holidays
are as needful to the countryman as to the townsman, and that if
the farm labourer or the shepherd were sent away to the sea every
year for a fortnight's rest and change, he would work with a new
energy that would more than compensate for the work lost. It would
be something at least to break the deadly monotony of the daily
round, even if the labourer had no ideas for profitably spending a
holiday.

In the Folds
For the shepherd the days and nights of January are heavy with
responsibility—he counts himself lucky if he can find time for an
hour's sleep. It is wonderful how the shepherd of a large flock knows
all the ewes and the lambs over which he now watches. In his lambs
he has a personal interest, for there may be a sixpence in his purse
for each lamb that lives to be deprived of its tail. The shepherd's
knowledge of the lambs surpasses that of the ewes, whom
sometimes he deceives; for it is by scent rather than sight that the
mother recognises her offspring, while the shepherd believes only
what he sees. By fastening the skin of a dead lamb on to an orphan
he will induce a bereaved ewe to adopt the orphan, and she will
accept, guard, and love it as if it were her own.

Shepherds' Care
January is to the shepherd what June is to the gamekeeper. There is
more than common meaning to the shepherd in the greeting, "A
happy and a prosperous New Year." Be luck good or bad, the bleat
of the lamb is the sweetest sound of the year to shepherd ears: it
means as much as the pee-peep of the pheasant chick to the
gamekeeper. Keepers and shepherds are deeply attached to their
respective "coops"—a word used by the shepherd for the enclosures,
one hurdle square, made for the lambs. The experience of coop life
is briefer for the lamb than for the young pheasant. After enjoying a
few hours of privacy, the ewe and her lambs are turned into the
large general nursery, to fend for themselves among the baa-ing
crowd.

Winter Partridge-driving
Weather makes more difference to partridge-driving than to most
forms of shooting. The ideal day comes when the weather is mild,
and the air still. Then only can the movements of partridges be
controlled with some certainty—not that partridges ever can be
driven against their will. In high wind their speed is tremendous, and
a hundred birds do not give the chances of ten too tired to swerve.
In hard, frosty weather, when the fields are like rough paving-
stones, though the day is still, the birds are up and off before the
advancing driving-line can shape itself to influence their flight. But in
mild, still weather, the soft soil clogs the birds' feet, they are slow to
rise, and packs and coveys become split up and their ranks
disorganised—to the advantage of the sportsmen.
A mild day may open hopefully enough, but if driving rain comes
with blustering wind the sport is spoiled.
On a frosty day, when things have been going badly, the guns may
be congratulating themselves as they reach some big turnip-fields
for which the birds have been making. A turnip-field may be
expected to steady and control the departure and the direction of
birds; but in the grip of frost turnips are only a little better than the
bare, frozen field. For the leaves, that yesterday made luxuriant
cover, to-day are flattened to the ground by the frost. Even the
charlock, which may have done so much to make up for the thinness
of the turnips, has been shrivelled to a few brown stems. Why the
farmer leaves the late-grown charlock untouched is because he
knows that before it reaches seed-time the frost will have killed
every plant. On a small shoot, frost-flattened turnips may ruin the
hope of a full day's partridge-driving. On big shoots frost counts for
less, for long drives can be taken. Short drives in winter partridge-
driving are seldom profitable—whether a shoot be small or big.
 PEEWITS IN WINTER.
LONDON, EDWARD ARNOLD.

The Fear of Snow

By the very poor snow is regarded as among the most terrible
calamities of life. Many types of countrymen, rural publicans,
postmen, outdoor labourers, and small traders, speak of snow as the
worst of all possible weather, leaving the most serious after-effects.
And snow means calamity to many wild things. Lucky are the robins
of a garden who have a friend to stir the old hot-bed, and turn up
the worms from beneath the frozen top-soil; happy the grain-feeding
birds who find a rick that has been threshed. Thousands flock to the
corn-ricks, and there is food for all—pheasants, partridges, rooks,
jackdaws, starlings, sparrows, greenfinches, chaffinches, yellow-
hammers, and the bramble-finches, orange, white, and black in
plumage. To the holly-trees come the starving thrushes, and in hard
weather even the fieldfares will lose their extreme shyness to
besiege a holly-tree beside a door. The more delicate redwings die in
thousands, though the dying and dead are seldom seen.
To a few the snow means profit—for the hawks there is a carnival of
feasting, and the fox finds weak and hungry hares and rabbits an
easy prey, if ill-nourished on a diet of tree-bark and withered
herbage. As to the pheasants, they are well cared for—and the
keeper, in snowy weather, scatters his maize with a liberal hand.

Hard-Weather Prophets
By many signs wild creatures inform the gamekeeper of the
approach of hard weather. The wood-pigeons give him useful
warning. In most parts of the country flocks of pigeons take toll of
the greens and root-crops—a thousand pigeons may be seen rising
from a single field of roots. In mild weather they may return once or
twice during a day. When they are seen constantly streaming to the
root-fields, those disturbed returning again and again, it is a certain
sign that hard weather is near.

Weather-wise Beasts and Birds

Animals have a reputation as weather prophets—if their prophecies
strike the human observer as somewhat obvious. The cat washes
her face, and this is commonly held to be a sign of coming rain; in
summer it is thought to be a sign of a thunder-storm when cats are
remarkably lively. Dogs sometimes bury their bones when rain is in
the air—perhaps an inherited instinct to save food against days of
bad hunting. Horses by stretching their necks and sniffing the air
seem to be scenting distant rain; and donkeys have a way of braying
before the storm. Shepherds hold that if sheep turn their tails
windward rain will come; and cowherds read the same prophecy
when a herd of cows gathers at one end of a pasture, their tails to
the wind. Changes in weather mean much to wild life, and we are
prepared to believe they are forewarned. A storm may mean the loss
of a meal to a fox, a ruined nest to a bird, an end of all things to an
insect. The fox has done well that has eaten heartily before the
storm. Yet it appears that a change of weather must be near at hand
before wild creatures take notice. The pheasant crows before the
thunder-storm because he hears distant thunder. The wheatear, a
bird nervous of clouds, flies to shelter as the cloud drives up. It is
the first touch of cold weather that sets squirrels hiding nuts.
Weather has a marked effect on the moods of wild creatures. There
are days when hares or partridges seem overcome by oppression;
they move listlessly if disturbed, and lie or sit about as though all
energy had gone from them. Thunder in the air may be the cause,
or perhaps snow is coming; when the storm has blown over,
liveliness is restored, and new life inspires all things. Before a storm,
partridges in the stubble-fields set up their feathers, and in cold
weather the feathers of many birds have the appearance of being
puffed out, so that they look almost twice their usual size. Many
creatures feed at an unusually early hour if storms are coming. It is
a bad sign when rabbits are out feeding in the fields early on a
bright sunshiny afternoon. The birds of the open fields—rooks,
starlings, pigeons, or fieldfares—feed hungrily and hastily while rain-
clouds overshadow the sky; but it is a sign of good weather when
rooks fly to feed far from their roosting-trees, and fly high. Cock
pheasants will go to roost early before the storm, choosing low
branches, and trees that afford good protection. In bitter weather,
even the warm feathers of birds may become ice-bound.

Green Winters
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebooknice.com