1. No category

CERVICOCEPHALIC ARTERY
DISSECTION
Radiological study with clinical outcome
O U TI
PELKONEN
Department of Diagnostic Radiology,
Department of Neurology,
University of Oulu
OULU 2004
OUTI PELKONEN
CERVICOCEPHALIC ARTERY
DISSECTION
Radiological study with clinical outcome
Academic Dissertation to be presented with the assent of
the Faculty of Medicine, University of Oulu, for public
discussion in the Auditorium 7 of Oulu University
Hospital, on January 30th, 2004, at 12 noon.
O U L U N Y L I O P I S TO, O U L U 2 0 0 4
Copyright © 2004
University of Oulu, 2004
Supervised by
Professor Juhani Pyhtinen
Docent Tapani Tikkakoski
Reviewed by
Docent Pekka Keto
Professor Juhani Sivenius
ISBN 951-42-7269-2 (nid.)
ISBN 951-42-7270-6 (PDF) http://herkules.oulu.fi/isbn9514272706/
ISSN 0355-3221
OULU UNIVERSITY PRESS
OULU 2004
http://herkules.oulu.fi/issn03553221/
Pelkonen, Outi, Cervicocephalic artery dissection. Radiological study with clinical
outcome
Department of Diagnostic Radiology, Department of Neurology, University of Oulu, P.O.Box 5000,
FIN-90014 University of Oulu, Finland
2004
Oulu, Finland
Abstract
The aim of this study was to analyze angiographic findings and the presence and topography of
cerebral ischemic and/or hemorrhagic lesions in cerebral CT or MRI, and to assess the long-term
clinical outcome of a series of 136 consecutive cervicocephalic artery dissection (CCAD) patients.
Pulsatile tinnitus was evaluated as a symptom of CCAD. Medical records and films were reviewed
retrospectively.
Irregular stenosis was found in angiography in 50% and occlusion in 33% of the dissected
cervicocephalic arteries. Irregular stenosis normalized in 81% and occlusion recanalized in 34%.
Other findings, such as pseudoaneurysms, intimal flaps, double lumens, and irregular dilatations were
rare and often remained unchanged in follow-up.
Pulsatile tinnitus was a presenting symptom in 12% of the CCAD patients, but the majority of
patients had concomitant head or neck pain, ischemic brain symptoms, Horner's syndrome, or cranial
neuropathies.
Of the 131 patients who underwent brain imaging, 73 (56%) had signs of infarction in cerebral CT
or MRI. Occlusion of the dissected vessel was accompanied by infarction in 76%, irregular stenosis
in 40%, and other findings only rarely. Of the anterior circulation infarctions, 95% (39/41) were
territorial, subcortical, or territorial infarctions with fragmentation and could thus be considered
embolic. Subarachnoid hemorrhage was found in CT in 5 of the 22 patients (23%) with intracranial
dissection.
The patient's long-term clinical outcome was assessed using two methods: a classification into
categories based on neurological symptoms and defects and the modified Rankin Scale (mRS). Of the
136 CCAD patients, 60% recovered with no or mild disability and 79% scored 0–2 on mRS. In the
case of dissection of one or more cervicocephalic arteries without occlusion, the figures were 75%
and 89%. In the case of occlusive dissection of one or more arteries, only about 35% of the patients
recovered well, having no or mild disability, and 61% scored 0–2 on mRS. No significant differences
were seen in recovery after intra- and extracranial dissections.
In conclusion: irregular stenosis, which is the most common angiographic finding in CCAD, is
associated with brain infarction less frequently than occlusion, and the long-term clinical outcome is
good in most cases. Occlusion of the dissected vessel causes more brain infarctions, and only about
35% of the patients recover well, having no or mild disability. More than 10% of CCAD patients have
pulsatile tinnitus as a presenting, and sometimes the only symptom.
Keywords: angiography, cerebral infarction, dissection, internal carotid artery, outcome,
pulsatile tinnitus, stroke, vertebral artery
To Hannu and Jere
Acknowledgements
This study was carried out at the Department of Diagnostic Radiology, University of
Oulu, during the years 1996–2003.
I owe my deepest gratitude to Professor Ilkka Suramo, M.D., who introduced me to the
field of radiology and scientific research and encouraged and supported me in the most
friendly way throughout my work. He made his most valuable time and effort available to
me whenever I needed it, and his enthusiasm and optimism inspired me and helped me to
complete this work.
I want to express my sincere gratitude to Docent Osmo Tervonen, M.D., head of the
Department of Diagnostic Radiology, who gave me the opportunity to carry out this
research in an encouraging and inspiring atmosphere.
I owe my warmest gratitude to my supervisor, Professor Juhani Pyhtinen, M.D., whose
expert guidance in the field of neuroradiology has been essential in the planning and
accomplishment of this study. He always had a new idea of what to do next.
I am deeply indebted to my supervisor, Docent Tapani Tikkakoski, M.D.. His wide
experience and skills in scientific work have impressed me. He helped me even with the
tiniest problems in the course of this work. His optimistic attitude towards science was
inspiring, and his encouragement and valuable advice were essential for the completion of
this work.
I am deeply grateful to Docent Kyösti Sotaniemi, M.D., from the Department of
Neurology. His expertise in clinical neurology was essential at all phases of this work,
and without his help a clinical study of this kind would not have been possible. He
patiently taught me scientific writing and encouraged me during these years.
I give my warm thanks to Docent Sami Leinonen, M.D., who was the first to suggest
this topic for research. I also thank him for teaching me the principles of angiography.
I wish to express my gratitude to my co-authors Docent Jukka Luotonen, M.D., from
the Department of Otorhinolaryngology and Docent Martti Lepojärvi, M.D., from the
Department of Surgery, who made their expert clinical contributions to the studies.
I am grateful to Professor Juhani Sivenius, M.D., and Docent Pekka Keto, M.D., the
reviewers of this thesis. Their constructive criticism and encouraging comments are
appreciated.
I express my gratitude to Mrs. Marianne Haapea, M.Sc., for her friendly help in
statistical analysis and Mrs. Sirkka-Liisa Leinonen, Lic. Phil., for revision of the English
language of this thesis and the original papers. Ms Helena Saari deserves my gratitude for
revision of the final layout. I wish to thank the Photographic Laboratory, especially Mrs.
Seija Leskelä, for friendly and excellent service. I wish to thank Ms Kaisa Punakivi for
her kind and valuable help in many practical matters. I owe my warm thanks to my friend
Mrs. Heli Reinikainen, M.D., Ph.D., for her ever so friendly help and support at the most
hectic times. Additionally, I want to thank Antero Koivula, Ph.D., for his help with
technical matters.
I thank the staff of the Department of Diagnostic Radiology, especially those working
in the Northern Central Radiology Department. I have got so much help from you during
these years. I express my thanks to all my colleagues for their friendship and support. I
wish to thank the members of CC-Oulu for providing many joyful and unforgettable
moments.
My warmest thanks go to my dear parents, Seija and Seppo, and to my dear sisters,
Heli, Ulla and Minna, who have loved and supported me in all phases of my life and
encouraged me in the accomplishment of this work.
Finally, I express my love and gratitude to my husband Hannu for his love, patience
and support during this work and to our son Jere for filling my heart with joy.
This study was financially supported by grants from the Radiological Society of
Finland, which is gratefully acknowledged.
Oulu, January 2004
Outi Pelkonen
Abbreviations
AICA
ACA
ACAD
ASA
CBF
CBV
CCAD
CSF
CT
DSA
DWI
EPI FLAIR
FLAIR
FMD
HU
ICA
ICAD
IVUS
MCA
MCAD
MDCT
MIP
MR
MRA
MRI
mRS
PC
PCA
PCAD
PD
PICA
anterior inferior cerebellar artery
anterior cerebral artery
anterior cerebral artery dissection
acetylsalisylic acid
cerebral blood flow
cerebral blood volume
cervicocephalic artery dissection
cerebrospinal fluid
computed tomography
digital subtraction angiography
diffusion-weighted imaging
echo planar imaging, fluid-attenuated inversion recovery
fluid-attenuated inversion recovery
fibromuscular dysplasia
Hounsfield Unit
internal carotid artery
internal carotid artery dissection
intravascular ultrasound
middle cerebral artery
middle cerebral artery dissection
multidetector computed tomography
maximum-intensity projection
magnetic resonance
magnetic resonance angiography
magnetic resonance imaging
modified Rankin Scale
phase contrast
posterior cerebral artery
posterior cerebral artery dissection
proton density
posterior inferior cerebellar artery
PICAD
PWI
RF
SAH
SPECT
T
TCD
TIA
TOF
TR
T1
T2
VA
VAD
VENC
V1
V2
V3
V4
2D
3D
posterior inferior cerebellar artery dissection
perfusion-weighted imaging
radio frequency
subarachnoid hemorrhage
single photon emission computed tomography
tesla
transcranial Doppler ultrasound
transient ischemic attack
time-of-flight
time of repetition
longitudinal relaxation time
transversal relaxation time
vertebral artery
vertebral artery dissection
velocity encoding
vertebral artery from origin to entry into the foramen transversarium of a
cervical vertebra, usually C6
C6 to C2 level of the vertebral artery
vertebral artery from the C2 foramen transversarium to the dura mater
intradural segment of the vertebral artery
two-dimensional
three-dimensional
List of original publications
This thesis is based on the following articles, which are referred to in the text by their
Roman numerals:
I
Pelkonen O, Tikkakoski T, Leinonen S, Pyhtinen J & Sotaniemi K (1998) Intracranial arterial dissection. Neuroradiology 40:442–447.
II
Pelkonen O, Tikkakoski T, Leinonen S, Pyhtinen J, Lepojärvi M & Sotaniemi K
(2003) Extracranial internal carotid and vertebral artery dissections: Angiographic
spectrum, course and prognosis. Neuroradiology 45:71–77.
III
Pelkonen O, Tikkakoski T, Luotonen J & Sotaniemi K (2003) Pulsatile tinnitus as a
symptom of cervicocephalic arterial dissection. The Journal of Laryngology &
Otology, in press.
IV
Pelkonen O, Tikkakoski T, Pyhtinen J & Sotaniemi K (2003) Cerebral CT and MRI
findings in cervicocephalic artery dissection. Acta Radiologica, in press.
V
Pelkonen O, Sotaniemi K, Pyhtinen J & Tikkakoski T (2003) Clinical outcome of
patients with cervicocephalic artery dissection. Submitted for publication.
Contents
Abstract
Acknowledgements
Abbreviations
List of original publications
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 Review of the literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1 Anatomy of internal carotid and vertebral arteries and pathology of CCAD . .
2.2 Epidemiology of CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3 Pathogenesis of CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1 Genetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.2 Environmental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4 Clinical manifestations in CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.1 Extracranial internal carotid artery dissection . . . . . . . . . . . . . . . . . . . . .
2.4.2 Intracranial carotid system dissection . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.3 Extracranial vertebral artery dissection . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.4 Intracranial vertebrobasilar system dissection . . . . . . . . . . . . . . . . . . . .
2.5 Vascular imaging of CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1 Digital subtraction angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1.1 Imaging technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1.2 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2 Magnetic resonance imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2.1 Imaging technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2.2 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.3 Computed tomography angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.3.1 Imaging technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.3.2 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.4 Ultrasound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.4.1 Imaging technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.4.2 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6 Imaging of brain manifestations in CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15
16
16
18
19
19
19
20
20
21
21
22
22
22
22
23
24
24
26
27
28
28
28
29
29
30
2.6.1 Computed tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.2 Magnetic resonance imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.3 Magnetic resonance spectroscopy, single photon emission
computed tomography, and positron emission tomography . . . . . . . . . .
2.6.4 Topography and etiology of cerebral infarction in CCAD . . . . . . . . . . .
2.7 Treatment of CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.8 Prognosis of CCAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3 Purpose of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 Patients and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Clinical information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2 Vascular imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2.1 Initial angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2.2 Initial MRI + MRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2.3 Follow-up angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.3 Cerebral CT and MRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.4 Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1 Angiographic spectrum and course of CCAD (I,II) . . . . . . . . . . . . . . . . . . . . .
5.2 Pulsatile tinnitus as a symptom of CCAD (III) . . . . . . . . . . . . . . . . . . . . . . . .
5.3 Cerebral CT and MRI findings in CCAD (IV) . . . . . . . . . . . . . . . . . . . . . . . . .
5.4 Long-term clinical outcome in CCAD and the correlation between
outcome and initial angiographic or MRA finding (I,V) . . . . . . . . . . . . . . . . .
6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1 Imaging findings in dissected vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2 Pulsatile tinnitus as a presenting symptom in CCAD . . . . . . . . . . . . . . . . . . .
6.3 Cerebral CT and MRI findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.4 Long-term clinical outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.5 Future prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References
30
31
32
32
33
34
36
37
37
39
39
40
40
41
41
41
42
43
43
44
45
46
49
50
52
53
55
57
59
1 Introduction
Dissection occurs when the intima or media of the arterial wall disrupts, causing an
intramural hematoma to develop in the subintimal, medial, or subadventitial layers (Hart
& Easton 1983, de Bray et al. 1994, Lal et al. 1994, Provenzale et al. 1995). Spontaneous
dissection of the internal carotid artery was first described in 1954 by Jenzer (1954). In
1983, Hart & Easton (1983) reviewed over 180 cervicocephalic artery dissections
(CCAD) reported in the literature. Most CCADs, however, remained undiagnosed at that
time or were diagnosed postmortem. Along with the development of imaging modalities,
CCAD has become an increasingly well known cause of acute strokes and transient
ischemic attacks (TIA) in young adults, explaining up to 22–32% of strokes in patients
under 40 years of age (Norrving et al. 1986, Bogousslavsky & Regli 1987). The average
annual incidence of spontaneous internal carotid artery dissections (ICAD) in all age
groups has been estimated to be 2.6 per 100,000 (Schievink et al. 1993). CCAD may
occur spontaneously, but extracranial dissections may be caused by many everyday
activities, including sudden severe stretching, compression, or both or prolonged
extension, flexion, or rotation of the neck. Traumas to the neck or head, common vascular
risk factors, and arteriopathies are considered to be associated with CCAD. Some
dissections occur as clinically silent. Some patients may have only mild symptoms, such
as pulsatile tinnitus or Horner’s syndrome, but CCAD may also cause brain infarctions
with persistent neurologic symptoms and signs or even death. Angiography has long been
the golden standard in diagnosing CCAD, showing indirect signs of dissection in the
vessel lumen. It is still the best imaging modality to show filiform stenosis with very slow
blood flow, changes due to fibromuscular dysplasia (FMD), and other irregularities of the
vessel wall (Schulze et al. 1992, Link et al. 1996). There are no randomized trials of the
different treatment regimens in CCAD, and both treatment and prognosis remain
controversial.
The purpose of this retrospective study was to evaluate the angiographic spectrum and
course and the long-term clinical outcome of CCAD. Pulsatile tinnitus was evaluated as a
presenting symptom. Brain manifestations were analyzed and anterior circulation
infarctions evaluated in order to find out if the etiology of CCAD-related brain infarctions
is embolic or hemodynamic. Long-term clinical outcome was assessed by using two
methods: classification into categories based on neurological symptoms and defects and
the modified Rankin Scale (mRS). Comparisons were made between the initial findings
in the dissected vessels and the frequency and pattern of infarctions and also with the
clinical outcome.
2 Review of the literature
2.1 Anatomy of internal carotid and vertebral arteries
and pathology of CCAD
The arterial wall is composed of three coats: intima (an internal or endothelial coat),
media (a middle or muscular coat), and adventitia (an external or connective-tissue coat).
The intima consists of a layer of pavement endothelium, a subendothelial layer, and an
internal elastic membrane. The thickness of the arterial wall is mainly due to the media,
which consists of elastic fibers alternating with layers of muscular fibers. The adventitia
consists mainly of collagen fibers, but also contains elastic fibers, i.e. there is an external
elastic membrane between the adventitia and the media. Arteries situated in the cavity of
the cranium and the vertebral canal have extremely thin walls relative to their size, the
difference being due to the thinness of the external and middle coats. Thin fibro-areolar
investments form the sheaths of the arteries and usually enclose the accompanying veins,
sometimes also a nerve. The arteries in the cranium are not included in sheaths. All larger
arteries are supplied by nutrient blood vessels called the vasa vasorum. Arteries are also
supplied by nerves, which are derived from the sympathetic system, but may pass through
the cerebrospinal nerves. They form intricate plexuses upon the surfaces of the larger
trunks. (Williams et al. 1989.)
Dissection occurs when the intima or media of the arterial wall disrupts, causing an
intramural hematoma to develop in the subintimal, medial, or subadventitial layers (Hart
& Easton 1983, de Bray et al. 1994, Lal et al. 1994, Provenzale et al. 1995). Whether a
primary intimal tear allows blood to pass from the lumen into the arterial wall or whether
a primary intramedial hematoma secondarily ruptures into the true lumen is unclear. Most
probably, the two mechanisms co-occur. Histological examinations of CCAD are usually
done postmortem, and reports are thus rare (Jenzer 1954, Petro et al. 1986,
Bogousslavsky et al. 1987, Steinke et al. 1996, Hamada et al. 2001). Intimal tear has not
been visible in all cases.
The internal carotid artery (ICA) supplies the anterior part of the brain, the eye and its
appendages, and sends branches to the forehead and nose. It may be divided into four
portions. The cervical portion begins at the bifurcation of the common carotid artery and
17
runs almost vertically in front of the transverse processes of the upper three cervical
vertebrae to the carotid canal at the base of the skull. It is closely related to the
glossopharyngeal (IX), vagus (X), accessory (XI), and hypoglossal (XII) nerves, the
superior cervical ganglion of the sympathetic trunk, the superior laryngeal nerve, the
internal jugular vein, the pharynx, and the ascending pharyngeal artery. In the petrous
portion, ICA enters the canal in the petrous portion of the temporal bone, ascends a short
distance, then turns forward and medially, and again ascends as it leaves the canal to enter
the cavity of the skull between the lingula and the petrosal process of the sphenoid. A
prolongation of the dura mater separates the artery from the bony wall of the carotid
canal. The petrous portion of ICA is surrounded by a number of small veins and by
filaments of the carotid plexus, derived from the ascending branch of the superior cervical
ganglion of the sympathetic trunk. In the cavernous portion, ICA is situated between the
layers of the dura mater, forming the cavernous sinus, and in the upper part it perforates
the dura mater, forming the roof of the sinus. It is surrounded by filaments of the
sympathetic nerve, and it is in a close topographical relationship with the oculomotor
(III), trochlear (IV), and abducent (VI) nerves and the ophthalmic and maxillary divisions
of the trigeminal nerve (V). In the cerebral portion, having perforated the dura mater, the
ICA passes between the optic (II) and oculomotor (III) nerves to the anterior perforated
substance at the medial extremity of the lateral cerebral fissure, where it gives off its
terminal or cerebral branches. The cervical portion of the internal carotid gives off no
branches, but the main branches arising from the cerebral portion are the middle cerebral
artery (MCA) and the anterior cerebral artery (ACA). (Williams et al. 1989, Uflacker
1997.)
Extracranial ICAD may be caused by a sudden stretching of the ICA over the
transverse processes of the upper cervical vertebrae in hyperextension and flexion or
rotation of the neck (Stringer & Kelly 1980). The most usual localization of extracranial
ICAD is the cranial part of the bulbus up to the skull base, i.e. foramen lacerum (Mokri et
al. 1986, Ast et al. 1993, Klufas et al. 1995).
The vertebral artery (VA) is the first branch of the subclavian artery. It can be divided
into four parts. The first part, often referred to as V1, is the part from the origin to the
entry into the foramen transversarium of a cervical vertebra, usually C6. Behind it are the
transverse process of the seventh cervical vertebra, the sympathetic trunk, and its inferior
cervical ganglion. The second part, V2, courses cranially through the foramina in the
transverse processes of the C6-C2. It is in close contact with the trunks of the cervical
nerves and surrounded by branches from the inferior cervical sympathetic ganglion and
by a plexus of veins. The third part, V3, is the part from the C2 foramen transversarium to
the dura mater. V3 issues from the latter foramen and curves backward behind the
superior articular process of the atlas. The anterior ramus of the first cervical nerve is on
its medial side. V3 runs in the groove on the upper surface of the posterior arch of the
atlas, and enters the vertebral canal by passing beneath the posterior atlanto-occipital
membrane. The first cervical or suboccipital nerve lies between the artery and the
posterior arch of the atlas. The fourth part, V4, is the intradural segment of the vertebral
artery. It is placed between the hypoglossal nerve (XII) and the anterior root of the first
cervical nerve. At the lower border of the pons, it joins the vessel on the opposite side to
form the basilar artery. (Williams et al. 1989, Uflacker 1997.)
18
Vertebral artery dissection (VAD) may be caused by stretching and compression of the
VA over the lateral masses of the CI and CII vertebral bodies (Sherman et al. 1981, Sheth
et al. 2001) in hyperextension and flexion or rotation of the neck. The most frequently
affected level was C1-C6 in one study (Chiras et al. 1985), the C1-C2 level in one (Mokri
et al. 1988a), and the Th1-C2 level in one (Bartels & Flügel 1996), while one study
(Provenzale et al. 1996) failed to establish any predominant site of extracranial VAD.
Extracranial dissections most often involve the medial layer or occur subadventitially and
expand either outward, causing an increase in the vessel diameter, or inward, causing
narrowing or occlusion of the lumen (Petro et al. 1986, Sturzenegger & Huber 1993, de
Bray et al. 1994, Klufas et al. 1995).
Intracranial dissections usually involve the intimal and medial layers, and the most
common location of intramural hematoma is between the internal elastic lamina and the
medial layer (Hart & Easton 1983, Berger & Wilson 1984, O’Connell et al. 1985, Rawat
et al. 1992). Intracranial dissections tend to extend inward, causing narrowing or
occlusion of the vessel lumen (Luken et al. 1979, Hart & Easton 1983, Sturzenegger &
Huber 1993). Pseudoaneurysms develop when the dissection proceeds through the media
to the subadventitial layer and causes dilatation of the outer vessel wall (Luken et al.
1979, Hart & Easton 1983, Petro et al. 1986, Lepojärvi et al. 1988, Mokri et al. 1988a).
Intracranial dissections may rupture through the adventitia and cause subarachnoid
hemorrhage (SAH) (Mokri et al. 1988a, Klufas et al. 1995). Histological studies have
shown that intracranial arteries lack an external elastic membrane and have a thinner
adventitia, fewer elastic fibers in the media, and generally a thicker internal elastic lamina
than extracranial arteries (Wilkinson 1972, Berger & Wilson 1984). The most usual
localization of dissection in anterior circulation is the supraclinoidal part of ICA (Rawat
et al. 1992, Schievink et al. 1994b). Dissections of the anterior (Ohkuma et al. 2003a) and
middle cerebral arteries (Ohkuma et al. 2003b) are less common. In the posterior
circulation, the most usual localization of intracranial dissection is the intracranial part of
the VA at or near the origin of the posterior inferior cerebellar artery (PICA) (Saver &
Easton 1998, Iihara et al. 2002). Dissections of the basilar artery (Leibowitz et al. 2003),
PICA (Tikkakoski et al. 1997), posterior cerebral artery (PCA) (Lazinski et al. 2000), and
anterior inferior cerebellar artery (AICA) (Hancock & Millar 2000) are rare.
2.2 Epidemiology of CCAD
Cerebrovascular diseases were the third leading cause of death in Finland in the year
2000 after heart disease and cancer (Statistics Finland 2002). The annual incidence of
stroke in two Finnish studies was 235 (Sivenius 1982) and 236 (Rissanen 1992) per
100,000 population, including the first and recurrent strokes in all age groups.
Spontaneous CCAD may explain up to 22–32% of strokes in young patients under 40
years of age (Norrving et al. 1986, Bogousslavsky & Regli 1987) and about 1–2% of all
instances of cerebral ischemia (Saver & Easton 1998). There are, however, no reliable
epidemiologic reports on the occurrence of CCAD. In a community-based population
study carried out in Rochester, Minnesota, the average annual incidence of the most
common spontaneous CCAD, ICAD, for all age groups was 2.6 per 100,000 and that for
19
persons aged 20 years or older 3.5 per 100,000 (Schievink et al. 1993). CCAD occurs
predominantly in the mid-40s, but the mean age for intracranial dissections tends to be
slightly lower, in the 30s (Saver & Easton 1998). There is no sex predilection.
2.3 Pathogenesis of CCAD
Blunt or penetrating trauma to the neck may cause traumatic CCAD. Spontaneous CCAD
is more common, and it is reported to be associated with several genetic and
environmental factors. In most cases, however, the pathogenesis remains unclear, and
even patients with no apparent risk factors may have CCAD. An arterial dissection is
probably the endpoint of a complex and possibly heterogeneous group of vasculopathies
developing under the influence of various genetic and environmental factors (Brandt &
Grond-Ginsbach 2002).
2.3.1 Genetic
Patients with spontaneous CCAD are thought to have an underlying structural defect of
the arterial wall, although the exact type of arteriopathy remains elusive in most cases.
FMD, cystic medial necrosis, Marfan’s syndrome, and Ehlers-Danlos syndrome type IV
are reported in 14–16% of patients with CCAD (Houser et al. 1984, Chiras et al. 1985,
Mokri et al. 1986, Ast et al. 1993, Schievink et al. 1996). FMD shows a female
preponderance. Approximately 5% of patients with CCAD have a family history of
arterial dissection, which also increases the risk of recurrent dissection (Schievink et al.
1996). Internal carotid artery redundancies – coils, kinks, and loops – are reported to be
possible predisposing factors in dissection (Sturzenegger & Huber 1993, Ozodoba et al.
1996). Ultrastructural abnormalities of dermal connective-tissue components have been
detected in up to two thirds of patients with spontaneous CCAD (Brandt et al. 1998,
Brandt et al. 2001).
2.3.2 Environmental
Because of the anatomic conditions of ICA and VA mentioned in 2.1, sudden severe
stretching, compression, or both or prolonged extension, flexion, or rotation of the neck
may cause extracranial dissections. The precipitating events reported include motor
vehicle accidents, all varieties of sports activity, cervical manipulation, painting a ceiling,
anesthesia administration, head turning while reversing a car, and even coughing,
vomiting, and sneezing (Schellhas et al. 1980, Stringer & Kelly 1980, Sherman et al.
1981, Zelenock et al. 1982, Chiras et al. 1985, Mokri et al. 1988b, Hoffmann et al. 1993,
Leys et al. 1995).
20
Common vascular risk factors considered to be associated with CCAD are
hypertension, smoking, diabetes mellitus, hyperlipidemia, and use of oral contraceptives
(Chiras et al. 1985, Mokri et al. 1986, Mokri et al. 1988a, Eljamel et al. 1990, Ast et al.
1993, Biousse et al. 1994, Sturzenegger 1994, Provenzale et al. 1996). In a recent study,
mild hyperhomocyst(e)inemia was also found to be a possible risk factor (Gallai et al.
2001). A recent history of infection is a risk factor for CCAD (Grau et al. 1999, Guillon
et al. 2003), and seasonal variation in the incidence of spontaneous CCAD was noted in
one study, which reported a peak incidence in the fall (Schievink et al. 1998). Migraine
was shown to have a significant positive association with non-traumatic CCAD
independently of type and treatment regimen (D’Anglejan-Chatillon et al. 1989).
2.4 Clinical manifestations in CCAD
CCAD has a wide spectrum of symptoms and signs. CCAD may occur as clinically
silent, cause minor symptoms, or result in a major stroke leading to severe permanent
disability or even death. Some clinical features differentiate ICAD from VAD and
intracranial dissections from extracranial dissections.
2.4.1 Extracranial internal carotid artery dissection
The classic triad of symptoms of extracranial ICAD is head or neck pain and Horner’s
syndrome followed by focal motor or sensory deficits appearing hours or days later
(Houser et al. 1984, Mokri et al. 1986, Schulze et al. 1992, Biousse et al. 1994). The
presence of any two elements of this triad should strongly suggest the diagnosis.
Pain is the most common, and often the initial, manifestation of ICAD, being present
in more than 70% of patients. It is considered to be due to distension of the artery, which
stimulates pain-sensitive receptors in the vessel wall. Typical pain is an ipsilateral
headache in the frontotemporal area, occasionally in the entire hemicranium or the
occipital area. In approximately 10% of cases pain manifests as isolated local neck pain,
sometimes as isolated facial or orbital pain, but usually there is simultaneous headache.
The median duration of pain in CCAD is 5 days. (Biousse et al. 1994, Silbert et al. 1995.)
Ipsilateral oculosympathetic palsy, called Horner’s syndrome, usually consisting of
ptosis, pupillary miosis, and facial anhidrosis, is found in one third of ICAD patients. It is
considered to be due to compression of the sympathetic fibers of the internal carotid
plexus running along the distended vessel wall. (Saver & Easton 1998.)
Reportedly, 49–86% of extracranial ICAD patients suffer from symptoms of cerebral
or retinal ischemia (de Bray et al. 1994, Biousse et al. 1995, Silbert et al. 1995, Engelter
et al. 2000, Baumgartner et al. 2001). Approximately one third of them have transient
ischemic attacks, while the rest have stroke (Biousse et al. 1995). Retinal infarctions are
rare (Baumgartner et al. 2001). In almost 90% of cases, stroke occurs within the first 7
days following the onset of the symptoms (Biousse et al. 1995).
21
Cranial nerve deficits occur in more than 10% of extracranial ICAD patients, and they
are due to ischemia or compression caused by dilatation of the dissected vessel
(Sturzenegger & Huber 1993, Klossek et al. 1994, Gobert et al. 1996, Mokri et al. 1996).
The most common affected nerve is the XII nerve, such deficits being present in 5% of
ICAD patients (Mokri et al. 1996), followed by other lower cranial nerves (IX, X, XI) and
the V nerve.
Pulsatile tinnitus is almost always due to the sound of nonlaminar blood flow that is
transmitted to the inner ear. Extracranial ICAD may cause local disorders anatomically
close to or within the petrous bone and alteration of the hemodynamics leading to
pulsatile tinnitus. In a review of 140 extracranial ICADs from 1974–1983, pulsatile
tinnitus was the only symptom in 4% and an associated symptom in 35% of patients (Hart
& Easton 1983). Ast et al. (1993) reported subjective tinnitus as a symptom in 7.5% (5/
68) of ICAD patients. In another study, subjective or objective pulsatile tinnitus was
reported in 50% (18/36) of patients with ICAD (Mokri et al. 1986). Steinke et al. (1994)
reported pulsatile tinnitus in 12% of 48 patients. In two previous studies, the frequency of
pulsatile tinnitus was 16% (Baumgartner et al. 2001) and 27% (Silbert et al. 1995) out of
181 and 161 patients, respectively. Pulsatile tinnitus was an associated symptom in all
cases.
2.4.2 Intracranial carotid system dissection
Some clinical features differentiate intracranial carotid system dissections from
extracranial ICADs. Headache is usually more severe in the former group. Almost all
patients have symptoms of cerebral or retinal ischemia, often a major stroke developing
over several days. In one fifth of cases, dissection presents as SAH. Intracranial
dissections do not cause Horner’s syndrome, cranial nerve palsies, or pulsatile tinnitus.
(Saver & Easton 1998.)
2.4.3 Extracranial vertebral artery dissection
The most typical symptoms of VAD are sudden occipital headache or posterior neck ache
on the same side as the dissected vessel and symptoms typical of a vertebrobasilar TIA or
infarction. Headache, present in up to 88 % of VAD patients (Sturzenegger 1994, Saeed
et al. 2000), may also be located temporally, retroorbitally, or bilaterally. Ischemic
symptoms of vertebrobasilar circulation, present in about two thirds of patients
(Provenzale et al. 1996), include vertigo, nausea, vomiting, more or less complete
Wallenberg syndrome, the locked-in syndrome, cerebellar syndrome, vestibular
syndrome, transient amnesia, blurred vision, and hemianopia. TIA is less common in
VAD than in ICAD. Uncommon manifestations of VAD are isolated spinal manifestations
and spinal epidural hematomas (Crum et al. 2000).
Horner’s syndrome has been reported in 36% of VAD patients (Sturzenegger 1994).
Rarely, VAD presents with pulsatile tinnitus: in two studies VAD was never the cause of
22
pulsatile tinnitus (Silbert et al. 1995, Baumgartner et al. 2001), while in a third study only
one patient out of 25 with VAD presented with pulsatile tinnitus (Mokri et al. 1988a).
Cranial nerve palsies of the nerves V, VII, VIII, IX, X, XII have been reported
(Sturzenegger 1994).
2.4.4 Intracranial vertebrobasilar system dissection
The main clinical feature distinguishing intracranial VAD from extracranial is the quite
frequent occurrence of SAH. Hosoya et al. (1999) found SAH in 3 of 31 patients with
intracranial VAD, while in another study more than half of patients (4/7) with intracranial
VAD had SAH (Provenzale et al. 1996). Intracranial VADs do not cause Horner’s
syndrome or cranial nerve palsies.
2.5 Vascular imaging of CCAD
2.5.1 Digital subtraction angiography
Angiography has long been the gold standard in diagnosing CCAD, but being an invasive
method, it carries a potential, though minimal, risk of severe complications. Digital
subtraction angiography (DSA) requires the use of intravascular, usually iodine-based
contrast agents, but their use is limited or contraindicated in patients with renal
insufficiency or a history of reaction to iodinated contrast media. Angiography shows the
arterial lumen but does not enable assessment of the arterial wall. Signs of CCAD are
indirect, but often typical and diagnostic. Pathognomonic signs are rare. Angiographic
findings depend on the degree of luminal stenosis and the depth at which the false lumen
is situated in the vessel wall. When dissection occurs in the subadventitial layer without
relevant narrowing of the arterial lumen, or when an aneurysm is thrombosed,
angiography does not yield the diagnosis (Sturzenegger & Huber 1993). Technical
developments in non-invasive modalities (magnetic resonance imaging (MRI), magnetic
resonance angiography (MRA), ultrasound, and computed tomography (CT)
angiography) are replacing angiography in many centers.
2.5.1.1 Imaging technique
In the past decades, DSA techniques have replaced conventional angiography in
cerebrovascular imaging. In DSA, an initial image acquired before the introduction of
contrast agent is a mask that is subtracted from the subsequent images with contrast
agent. This results in a subtracted image of contrast only. This technique has certain
23
advantages, such as instant acquisition, real-time information, increased contrast
sensitivity, and the possibility to electronically manipulate images. 1024 x 1024 matrix
size is required for sufficient resolution (Katzen 1995). Usually, transfemoral bilateral
carotid and bilateral vertebral artery angiography in at least two orthogonal directions is
performed, sometimes with aortic arch injections in postero-anterior and left and right
oblique projections. In recently introduced rotational three-dimensional (3D)
angiography, two multidirectional C-arms in a single system allow simultaneous
projection from two directions. The 3D capability provides added visual depth perception
for more precise images and reduction of radiation exposure.
2.5.1.2 Findings
Double lumen develops when an intramural hematoma ruptures back into the real lumen,
allowing blood to flow through both lumens. Intimal flap is a floating membrane in the
lumen near the proximal margin of dissection. These angiographic signs are considered
pathognomonic of dissection as seen also in the case of aortic dissection, but they are rare
and present in less than 10% of CCAD patients. (Houser et al. 1984.)
Indirect, nonspecific signs of CCAD in angiography are most common, and absence of
atheroma is typical of CCAD patients. Subintimal hematoma tends to expand inward,
causing narrowing or occlusion of the lumen. The most common sign of extracranial
ICAD is irregular stenosis, visible as a “wavy ribbon” or a “string sign”. It is found in 58–
75 % of extracranial ICAD patients (Houser et al. 1984, Ast et al. 1993). Usually, stenosis
begins 2–3 cm distal to the carotid bulb, extending over a variable distance. Dissection
terminates, in all but exceptional cases, before the entry of ICA into the petrous part of
the temporal bone, where mechanical support seems to limit further expansion and where
the lumen is abruptly reconstituted. (Hart & Easton 1983, Houser et al. 1984, Mokri et al.
1986, Mokri 1990, Ast et al. 1993, Saver & Easton 1998.) Gradually tapering occlusion
of ICA is less specific. It is found in 17–35 % of extracranial ICAD patients (Mokri et al.
1986, Ast et al. 1993, Saver & Easton 1998). In the acute phase, a flame-like occlusion at
least 2 cm above carotid bifurcation is characteristic of dissection. Dissecting aneurysms,
i.e. pseudoaneurysms, develop when the dissection proceeds through the media to the
subadventitial layer and causes dilatation of the outer vessel wall. They are usually
located near the base of the skull and may be present in up to 37 % of extracranial ICAD
patients. Irregular dilatation is rare. Embolic occlusions of distal branches may be present
in 11 % (Mokri et al. 1986). In a canine model, morphologic changes in angiography after
experimental carotid dissection were closely related to the size of the intimal entry zone;
small entry zones resulted in spontaneous healing and large ones in stenotic lesions, but
medium entry zones (4–6 mm) potentially induced aneurysm formation (Okamoto et al.
2002).
Extracranial VAD is less common than extracranial ICAD. It has the same set of
patterns in angiography as extracranial ICAD, but the findings are less specific. Irregular
stenosis is the most common finding present in about 80 % (Mokri et al. 1988a, Saver &
Easton 1998). The previous reports on the most typical location of VAD are controversial.
The most frequently affected level was C1-C6 in one study (Chiras et al. 1985), the
24
C1-C2 level in one (Mokri et al. 1988a), and the Th1-C2 level in one (Bartels & Flügel
1996), while one study (Provenzale et al. 1996) failed to establish any predominant site of
extracranial VAD. VA enters the skull through the foramen magnum, which explains why
up to 20 % of VAD may extend intracranially (Mokri et al. 1988a).
Intracranial CCAD is rare. Most intracranial dissections occur in the posterior
circulation in the distal VA or basilar artery (Mokri et al. 1988a, Kitanaka et al. 1994,
Hosoya et al. 1999). Dissections of PICA (Berger & Wilson 1984, Tikkakoski et al. 1997)
and PCA are rare (Lazinski et al. 2000). Intracranial carotid system dissection is usually
located in the supraclinoid part of ICA (Rawat et al. 1992, Schievink et al. 1994b) or
MCA (Ohkuma et al. 2003b), sometimes ACA (Ohkuma et al. 2003a). Findings in
angiography are usually nonspecific, typically irregular stenosis with or without
dissecting aneurysm or total occlusion following tapering stenosis. Double lumen and
intimal flap are rare but specific. Intracranial dissections may rupture through the
adventitia and cause SAH. (Mokri et al. 1988a, Klufas et al. 1995, Lazinski et al. 2000,
Ohkuma et al. 2003a, Ohkuma et al. 2003b.)
Multivessel dissections are quite common, being found in 16–28 % of CCAD patients
(Schievink et al. 1994a, Mokri et al. 1996, Ozodoba et al. 1996). FMD is seen in
angiography as an additional finding in about 25% of CCAD patients (Mascalchi et al.
1997). String of beads is a characteristic finding.
Up to 80% of dissections normalize / recanalize during follow-up (Houser et al. 1984,
Mokri et al. 1986, Mokri et al. 1988a, Steinke et al. 1994). Occlusions recanalize in only
8% (Mokri et al. 1988a, Mokri 1990). Half of the pseudoaneurysms remain unchanged
(Touze et al. 2001).
2.5.2 Magnetic resonance imaging
The major advantage of MRI is that it can demonstrate the intramural crescentic
hematoma itself. The resolution of MRA is now approaching that of angiography, and
nonspecific changes in the calibre of the vessel can be seen. MRI and MRA are non- /
mini-invasive and free of radiation, but not readily available in all centers. (Levy et al.
1994, Ozodoba et al. 1996, Mascalchi et al. 1997, Lecrerc et al. 1999.)
2.5.2.1 Imaging technique
The sequences used in MRI (0.5–1.5 Tesla (T) system) of cervicocephalic arteries usually
include T1, T2, and proton density (PD)-weighted images of the neck in an axial plane.
Caudal spatial presaturation pulse is often used to suppress arterial inflow. Fat saturation
techniques are used in T1-weighted images to differentiate hyperintense intramural
hematoma from the surrounding tissues. Section thickness of 4 mm is recommended for
good spatial resolution. (Levy et al. 1994, Ozodoba et al. 1996, Mascalchi et al. 1997,
Lecrerc et al. 1999.)
25
Two types of MRA techniques are used routinely in neurovascular imaging.
Flow-based techniques utilize the fact that flowing blood moves through the vessels
during the measurement. Such techniques include time-of-flight (TOF) and phase contrast
(PC) imaging techniques, which can be run in most magnetic resonance (MR) scanners.
In the contrast-enhanced technique, intravenously injected contrast agents shorten the T1
of blood, and scanning is done during their passage through the vessels. This technique
requires a state-of-the-art system with high slew rates and dedicated tools for accurate
timing of acquisition during the first pass of the contrast bolus. (Bosmans et al. 2001.)
In the time-of-flight technique, stationary tissues are supressed by subjecting their
spins to the complete series of radio frequency (RF) pulses in an imaging sequence. With
a short time of repetition (TR), the signal intensity in stationary tissue during a series of
RF pulses decreases, resulting in a low signal. Flowing spins in the blood, however,
experience only few RF pulses, with blood continuously entering and leaving the imaging
plane. Therefore, the hypointense steady-state value is never reached in blood, while
positive contrast with the surrounding tissues is obtained. The usability of TOF depends
on the velocity of blood in the vessel, the length of the vessel in the imaging slab, the
flow pattern, and the sequence parameter setting. Two-dimensional (2D) and 3D
applications are available, 3D TOF being mostly recommended for the diagnosis of
CCAD. (Bosmans et al. 2001.) However, the field of view is limited in 3D TOF and also
dependent on the flow direction. For example, in the case of subclavian steel VA cannot
be seen.
Phase contrast imaging is based on the fact that there is a direct relationship between
the velocity of the spins and the phases they acquire when moving along a magnetic field.
In the PC imaging technique, bipolar pulses are used to achieve flow-induced calculable
phase in spins that move along the direction of the gradients, while pulses have no effect
on stationary spins. Minimally, two measurements are made, a flow-compensated (TOF)
acquisition and a second acquisition with bipolar pulses in a particular direction.
Complete flow data can be obtained for bipolar pulses in all of the three dimensions.
Measurement in PC imaging takes quite long, and in order to get optimal vessel contrast,
velocity encoding (VENC) must be done. 2D PC imaging with different VENC is more
practical than 3D PC, which takes even longer and should be done with the optimal
VENC. In practice, PC imaging is used most commonly in the visualization of venous
pathologies, but it can also be used, for example, to show possible absence of flow in
MCA in acute ischemic stroke, providing complementary information to diffusion and
perfusion-weighted imaging in predicting the outcome of acute stroke patients (Liu et al.
in press). It also gives information about flow dynamics, such as velocities. (Bosmans et
al. 2001.)
In MRA, contrast agents, usually gadolinium-based contrast agents, are used to
achieve higher signal intensity in blood than in other tissues in T1-weighted images, as
the T1 of blood is shortened. Contrast enhancement helps to overcome the saturation
effect in TOF techniques in the case of slow or turbulent flow, as in aneurysms.
Contrast-enhanced MRA is not dependent on the direction of flow. Recently, new
ultrafast acquisitions have been introduced. They enable the use of stronger gradients in a
shorter time, giving rise to 3D techniques with very short TR. Ultrafast MRA must be
performed during the first pass of the contrast bolus. Techniques used to attain accurate
timing include test bolus acquisition, semiautomatic or automatic bolus tracking,
26
acquirement of series of ultrashort acquisitions, and use of multiple boluses and dedicated
post-processing tools. (Bosmans et al. 2001, Wilms et al. 2001.) The coverage in
contrast-enhanced MRA is better than that in flow-based MRA techniques, allowing, for
example, good overall imaging of the aortocervical vessels.
Both flow-based MRA techniques tend to overestimate stenosis both in diameter and
in length, and severe stenosis may stimulate occlusion due to the slow or turbulent flow.
For the same reason, aneurysms may be invisible in MRA. Contrast agents help to
overcome these effects. The most widely recommended technique of MRA in CCAD is
3D TOF and contrast-enhanced ultrafast MRA, when available. The hyperintense
intramural hematoma can mimic flowing blood on a maximum intensity projection
(MIP), and correlation with the spin echo MR images is necessary. (Levy et al. 1994,
Ozodoba et al. 1996, Mascalchi et al. 1997, Lecrerc et al. 1999, Bosmans et al. 2001,
Wilms et al. 2001.)
2.5.2.2 Findings
A pathognomonic sign of CCAD in MRI is an eccentrically narrowed lumen with an
adjacent crescent-shaped increased signal, representing the intramural hematoma, i.e. the
extravasated blood in the vessel wall. It is best demonstrated in T1-weighted
fat-suppressed MR images with caudal spatial presaturation, where the high-intensity
intramural hematoma can be distinguished from the surrounding fat and cerebrospinal
fluid (CSF). (Provenzale et al. 1995, Ozodoba et al. 1996, Mascalchi et al. 1997, Saver &
Easton 1998.) In the first few days, the intramural hematoma is intermediate on
T1-weighted images and high on PD and T2-weighted images. After that, the signal is
high on T1, PD, and T2-weighted images, before the changes gradually resolve. A
hyperacute intramural hematoma would have a low signal on T2-weighted MR images
due to susceptibility effects of deoxyhemoglobin, but no such cases have been reported.
(Mascalchi et al. 1997.)
Less specific signs of dissection in MRI are increased signal or poor or absent
visualization of the entire vessel, marked narrowing of the vessel lumen by adjoining
tissue with abnormally increased signal intensity, and an enlarged vessel diameter, which,
in the early stage, may be the only indicator of dissection (Ozodoba et al. 1996, Saver &
Easton 1998). MRI is superior to DSA and MRA in cases of nonspecific occlusion and
dissection without associated luminal abnormalities.
The sensitivity of MRI in ICAD is 84% and specificity 99%, but in VAD its sensitivity
is only 60% and specificity 98% (Levy et al. 1994). The detectability of the VA
intramural hematoma on MRI depends on the segment involved. The surrounding tissues
have different signal characteristics at different levels. On T1-weighted images, the
intermediate or high-signal intramural hematoma is best visualized in V4 (intradural
segment), where VA is surrounded by low-signal CSF. In V3 (from C2 foramen
transversarium to the duramater) and V2 (C6 to C2 level), the intermediate to high-signal
perivascular structures and the slow flow proximal and distal to the intramural hematoma
make the intramural hematoma more difficult to detect. On PD images, intramural
hematoma can be easily detected only in V4. In T2-weighted images, CSF in V4 and
27
perivascular structures in V2 and V3 have a similar high signal as intramural hematoma.
(Mascalchi et al. 1997.) Additionally, VA is more sensitive than ICA to the technical
artefacts of MR due to the smaller and variable diameter.
Intracranial arteries are small and sensitive to artefacts, but the crescent-shaped
high-signal change representing the intramural hematoma may sometimes be seen. In the
study of Hosoya et al. (1999), the positive rate of intramural hematomas in intracranial
vertebrobasilar artery dissection was 32%.
MRA shows changes in the calibre of the vessel, which are indirect and nonspecific
signs of dissection. The most common findings in dissection are irregular stenosis and
absence of the arterial signal. An abnormal signal indicating intramural hematoma
adjacent to the dissection may be seen. The sensitivity of MRA in ICAD is 95% and
specificity 99%. In VAD, the corresponding figures are 20% and 100% (Levy et al.
1994). In another study, MRA detected signal abnormalities within the dissected
extracranial vertebral artery in 94% (16/17), and MRI was specific for dissection in 29%
(5/17) in the acute and subacute stages, but specificity (true negative rate in subjects free
of disease) was not considered in this study because all patients had vertebral artery
dissection (Auer et al. 1998). Better results in VAD can be achieved by using contrast
agent, which allows acquisition of a large volume in a short scan time without major
artefacts (Leclerc et al. 1999). Contrast agents shorten the T1 relaxation time of blood,
giving a high signal and reducing in-plane saturation.
Due to the artifacts that can mimic the MR appearance of FMD and thus decrease the
sensitivity and specificity of MRA in its detection, MRA is inadequate, similarly to MRI,
for the demonstration of FMD and aneurysms associated with dissection (Heiserman et
al. 1992, Levy et al. 1994). MRA is suitable for the detection of FMD when the
characteristic pattern is combined with moderate stenosis of the vessel, but DSA remains
the standard of reference (Link et al. 1996). Angiography is also needed in the detection
of filiform stenosis with very slow blood flow and FMD changes and other irregularities
of the vessel wall (Schulze et al. 1992). MRI and MRA are good in the follow-up of
CCAD (Kasner et al. 1997, Leclerc et al. 1999).
2.5.3 Computed tomography angiography
CT angiography is minimally invasive and, with the improved technology, able to
provide high-resolution images of the arterial lumen as well as the vessel wall. However,
similarly to DSA, CT angiography requires the use of intravascular iodine-based contrast
agents, and their use is limited or contraindicated in patients with renal insufficiency or a
history of reaction to iodinated contrast agent. The experience of carotid and vertebral as
well as cerebral CT angiography is so far limited.
28
2.5.3.1 Imaging technique
Helical CT angiography was introduced in the late 1980s. It combines continuous gantry
rotation with simultaneous displacement of the examination table throughout the
acquisition, providing a registered volume data set with thinner images and more rapid
acquisition compared to pre-helical scanners. Multirow / multidetector CT (MDCT) was
introduced in 1998. MDCT systems with 4, 8 or 16 channels have improved both the
coverage speed and the z-axis resolution of CT angiography and increased the range of
clinical applications. (Leclerc et al. 1996, Foley & Karcaaltincaba 2003.)
The three aims of CT angiography are 1) to achieve an adequate level of arterial
contrast enhancement during acquisition, 2) to provide cephalocaudad coverage of the
targeted anatomy during the first circulation of the injected bolus and during an easily
sustainable breath hold interval, and 3) to time the onset of CT acquisition after a
peripheral contrast injection in such a way that the first circulation enhancement is
obtained from the beginning to the end of the acquisition. Timing of circulation can be
achieved using a preliminary mini-test bolus or online bolus tracking. (Foley &
Karcaaltincaba 2003.) The thickness of the arterial wall and the size of the opacified
lumen are usually evaluated on the axial CT images, but axial scan data can also be
reformatted in various paraxial and oblique planes.
2.5.3.2 Findings
Helical CT is a reliable method for evaluating extracranial ICAD. The most sensitive and
specific sign of dissection is a narrowed eccentric lumen in the upper portion of ICA.
Mural thickening, occlusion, and aneurysms can be evaluated well with CT angiography,
but are not specific to dissection. The target picture is a rare and less reliable sign.
(Leclerc et al. 1996.) Helical CT can be used in the follow-up of extracranial ICAD
(Leclerc et al. 1998) and in the diagnosis of extracranial VAD with similar findings as in
ICAD (Soper et al. 1995, Kurokawa et al. 2000).
2.5.4 Ultrasound
Being a non-invasive and readily available procedure, ultrasound in often the first
screening method in suspected carotid or vertebral artery obstructive disease. The result
of the ultrasound examination strongly depends on the diligence and experience of the
examiner.
29
2.5.4.1 Imaging technique
Before the introduction of the duplex technology, continuous-wave Doppler and
high-resolution gray-scale (B-mode) imaging were the ultrasound techniques used to
examine carotid and vertebral arteries. Nowadays, color duplex ultrasound allows
simultaneous visualization of vascular lesions in the gray-scale image (plaques, stenosis,
occlusion) and the associated abnormalities in flow in the color-encoded image
(intrastenotic velocity increase, poststenotic flow disturbances, lack of flow signal due to
occlusion). Hemodynamic quantification of the pathology is achieved by analysis of
Doppler spectral wave forms. Examinations should be done in both transverse and
sagittal planes, using linear transducers (5–7.5 MHz). Power Doppler and the use of
ultrasound contrast agents are helpful in differentiating high-grade stenosis from
occlusion. (Landwehr et al. 2001.) Transcranial Doppler ultrasound (TCD) can show
intracranial collateral blood flow (Steinke et al. 1994) and distal microemboli (Srinivasan
et al. 1996).
2.5.4.2 Findings
The signs considered specific to dissection in high-resolution gray-scale (B-mode) and
color duplex ultrasound are intramural hematoma and intimal flap, which, however, are
seen rarely, in less than one fifth of patients with extracranial ICAD (de Bray et al. 1994,
Steinke et al. 1994, Bartels & Flügel 1996). Intramural hematoma is seen as an
echolucent matrix causing an increase of the arterial diameter. Intimal flap is seen as a
moving, thin, echogenic intravascular structure. Suggestive but not specific signs of
extracranial ICAD or VAD are tapering stenosis, tubular vessel, segmental ectasis,
occlusion without atheroma, double lumen, and pseudoaneurysm (de Bray et al. 1994,
Steinke et al. 1994, Bartels & Flügel 1996).
In Doppler ultrasound, a high-resistance flow pattern with bidirectional signal
components and absent diastolic flow is characteristic of dissection. Steinke et al. (1994)
found it in 68 % of angiographically confirmed extracranial ICAD patients (n=48). In
statistical analysis, sensitivity was 68%, specifity 99%, positive predictive value 94% and
negative predictive value 99% based on over 3000 carotid artery angiograms and
corresponding duplex Doppler examinations. Bakke et al. (1996), however, found it in
only one of twelve extracranial ICAD patients. Suggestive indirect signs of more distal
extra- or intracranial CCAD are markedly reduced systolic and diastolic blood flow
velocities, absence of any Doppler signal, retrograde ophthalmic blood flow (in ICAD),
spectral broadening, and retrograde flow (in VAD). (Rawat et al. 1992, de Bray et al.
1994, Steinke et al. 1994, Bakke et al. 1996, Bartels & Flügel 1996.)
Transcranial Doppler ultrasound rarely shows stenosis suggestive of dissection, but
collateral blood flow via the anterior and posterior communicating arteries is seen more
often (Steinke et al. 1994, Srinivasan et al. 1996). TCD is useful in the detection and
follow-up of microemboli in the MCA distal to the ICAD (Srinivasan et al. 1996).
Ultrasound has several limitations in the diagnosis of CCAD. The distal and
intracranial parts of the arteries cannot be seen satisfactorily or at all. The intramural
30
hematoma is hypoechoic and difficult to analyze. Severe stenosis may mimic occlusion.
VA may be hypoplastic or have changes in lumen diameter mimicking pathologies. VA is
also partly hidden behind the transverse processes of the C6-C2 vertebrae. The
abnormalities are usually indirect and not specific. In the follow-up of known CCAD,
however, ultrasound is useful.
2.6 Imaging of brain manifestations in CCAD
The main purpose of brain imaging in CCAD is to assess the presence of an infarction
and, in intracranial CCAD, also the presence of SAH and to exclude other diseases
mimicking ischemia (tumor, intracerebral hemorrhage, subdural hematoma, vascular
malformation). Brain imaging plays a vital role in the management of stroke.
2.6.1 Computed tomography
Cerebral CT is still the first method in the differential diagnosis of acute stroke. It is
widely available, quick, cheaper than MRI, and sensitive in excluding hemorrhage.
Cerebral CT is usually done without intravenous contrast agent in transverse sections
with a slice thickness of 4–5mm infratentorially and 8–10 mm supratentorially (Dippel et
al. 2000, Urbach et al. 2000). Intravenous contrast agent can be used to help in
differential diagnosis, but it has not been demonstrated to increase the detectability of
ischemic lesions within the first 24 hours and it may even hide hemorrhage or infarction
by enhancing it to the same Hounsfield Unit (HU) level as the surrounding brain.
Cerebral CT is often normal during the first few hours after acute ischemic stroke (Dippel
et al. 2000). The early CT signs of brain infarction are subtle, but a slight decrease in the
density of gray matter, leading to a loss of precise delineation between gray and white
matter, or sulcal effacement may be seen. These findings are typical early CT signs of
infarction, and they are due to an increase in the brain tissue water content and cerebral
edema. In the middle cerebral artery territory, the reported early CT signs of infarction are
attenuation of the lentiform nucleus, loss of the insular ribbon, and the hyperdense middle
cerebral artery sign (Tomura et al. 1988, Truwit et al. 1990, Moulin et al. 1996). The
decrease in the density of the infarction area becomes prominent over time, and
increasing edema may also cause a mass effect. About half of CCAD patients have signs
of infarction in brain imaging (Saver et al. 1992, Steinke et al. 1996), but other findings,
such as hyperdense artery, hemorrhage, and SAH, are rare.
Disturbances in cerebral capillary perfusion can be quantitatively evaluated with
perfusion CT, using an iodine-based contrast agent. Perfusion CT helps in the early
differential diagnosis of acute ischemic stroke. Dynamic scanning is performed about 5
seconds after the initiation of rapid contrast agent infusion. Cerebral blood flow (CBF),
cerebral blood volume (CBV), mean transit time, and time to peak are perfusion
parameters that can be calculated and presented in images or maps, allowing assessment
of the type and extent of cerebral perfusion disturbances. Perfusion CT can also be done
31
with the subtraction technique, in which the initial images acquired before the
introduction of contrast agent are subtracted from the subsequent images with contrast
agent, leaving only the change in the attenuation values, i.e. perfusion data. (Hamberg et
al. 1996, Hunter et al. 1998, Aksoy & Lev 2000, Wityk & Beauchamp 2000.) A 4-row
CT scanner enables the acquisition of sections about 2cm thick and a 16-row CT scanner
sections about 3cm thick, but whole brain perfusion studies are not yet possible (Tomandl
et al. 2003). Multi-row CT allows the combined use of three imaging modalities,
nonenhanced CT, perfusion CT, and CT angiography, to rapidly obtain comprehensive
information regarding the extent of ischemic damage in acute stroke patients (Tomandl et
al. 2003). A less frequently used technique is xenon-CT, in which the patient breathes
xenon gas (Jungreis et al. 1999). The gas then dissolves into blood and passes into the
brain, where it readily diffuses through the blood brain barrier. Its distribution in the brain
is measured with CT based on changes in attenuation.
2.6.2 Magnetic resonance imaging
When infarction is suspected, conventional MRI of brain usually includes T1-weighted
images in the sagittal plane and T1, PD, and T2-weighted images in the axial plane.
Section thickness of 5mm with a 1mm gap is commonly used (Mullins et al. 2002).
Conventional MRI has been reported to be more sensitive than CT in acute ischemic
changes (Bryan et al. 1991), but due to the facts that MRI is expensive and not readily
available, that the scanning takes a relatively long time, and that the sensitivity of
detecting hemorrhage with routine MRI sequences is lower than in CT, it has not been
widely used as the first imaging modality in acute stroke. Gadolinium-based contrast
agent may help in differential diagnosis and also improve the detection of brain infarction
in T1-weighted images. Transverse fluid-attenuated inversion recovery (FLAIR) images
visualize edema (Mullins et al. 2002). MRI may also show an abnormal arterial flow void
or the presence of an intramural hematoma indicating intracranial dissection. In the past
decade, new techniques allowing the evaluation of functional parameters have been
introduced. Such techniques include MR diffusion-weighted imaging (DWI) and MR
perfusion-weighted imaging (PWI). DWI exploits the transitional mobility of water
molecules to obtain information of the microscopic behavior of tissues, e.g. the changes
in intracellular vs. extracellular water balance in acute stroke. PWI makes use of
endogenous and exogenous tracers in monitoring regional hemodynamic quantities, such
as cerebral blood volume, cerebral blood flow, and mean transit time. (Luypaert et al.
2001.) DWI and PWI are superior to CT and conventional MRI in the diagnosis of acute
stroke during the first 12 hours after presentation (Mullins et al. 2002), but these
techniques are not readily available in all centers.
32
2.6.3 Magnetic resonance spectroscopy, single photon emission
computed tomography, and positron emission tomography
Magnetic resonance spectroscopy utilizes the fact that different chemicals vibrate at
different frequencies when stimulated by a magnet. The signals are detected with MRI,
and a ”signature” is produced to show what chemicals and in what amounts are present.
Magnetic resonance spectroscopy can be used to evaluate the metabolic abnormalities
associated with focal brain ischemia by specific biochemical measurements. The level of
N-acetylaspartate drops and the lactate level rises in infarction during the first few hours
of ischemia already. (Saunders 2000, Liu et al. 2003.)
In single photon emission computed tomography (SPECT), a radioisotope, such as
technitium-99m (99mTc), is added to a delivery compound that passes through the blood
brain barrier after an intravenous injection and is metabolized by neuronal and glial cells.
Its uptake in brain is proportional to CBF at the moment of passage. Imaging with a
2-headed or 3-headed SPECT imaging system can be performed within a few hours after
the injection. SPECT is a relatively low-resolution method for imaging cerebral perfusion,
without the inherently high spatial resolution of CT or MRI. (Holman & Devous 1992.)
Positron emission tomography is a technique for measuring the concentrations of
positron-emitting radioisotopes within the tissues of living subjects. It enables
measurement of the function of cerebral tissue, which depends critically on the use of
oxygen, and impairment in its rate of consumption often constitutes a pathological
condition. Using 15O2 and H215O with positron emission tomography, oxygen
consumption and cerebral perfusion can be quantitatively measured in ischaemic stroke.
However, its availability and practicality in acute clinical settings are limited. (Heiss
2000.)
2.6.4 Topography and etiology of cerebral infarction in CCAD
Using the current pathogenetic concepts of the topography of cerebral infarction, the
etiology can be considered either hemodynamic or embolic (Ghika et al. 1989,
Bogousslavsky 1991, Fisher 1991). Territorial infarctions (involving the cerebral cortex
of one or more major cerebral artery territories), subcortical infarctions (affecting the
territory of deep perforating branches originating from the distal ICA or the MCA,
involving the basal ganglia, thalamus, internal capsule, or centrum ovale), and territorial
infarctions with fragmentation (representing a lesion similar to territorial infarction with
additional smaller lesion(s) in cortical regions) are considered embolic, whereas
watershed infarctions (located between two arterial territories in regions considered to
constitute a hemodynamic risk zone) are considered hemodynamic. Most infarctions in
ICAD patients are considered embolic, as only about 5% of all infarctions are located in
hemodynamic risk zones (Steinke et al. 1996, Lucas et al. 1998, Baumgartner et al. 2001,
Milhaud et al. 2002). Weiller et al. (1991) reported a nearly 50% frequency of
hemodynamic infarctions in their study of fifteen patients with eleven infarctions.
33
2.7 Treatment of CCAD
There are no randomized trials of the different treatment regimens in CCAD, and
treatment hence remains controversial. The natural course of dissection is favorable, and
in some centers, CCAD patients, especially those without ischemic symptoms, are treated
with antiplatelets only. Dissection-related strokes usually occur in the first few days after
the onset of the symptoms, but may occur up to one month later (Biousse et al. 1995).
Early diagnosis and treatment are therefore essential. While a great majority of
dissection-related infarctions are considered thromboembolic (Steinke et al. 1996, Lucas
et al. 1998, Baumgartner et al. 2001), and while there is a high frequency of intracranial
microemboli detectable with transcranial Doppler (Srinivasan et al. 1996), treatment aims
at the prevention of further cerebral embolism. The most frequently used empiric therapy
is parenteral anticoagulation for about a week followed by warfarin for 3–6 months
(Srinivasan et al. 1996), often followed by acetylsalisylic acid (ASA). Koch et al. (2001)
demonstrated a decline of microembolic signals in transcranial Doppler with heparin
anticoagulation in the case of ICAD. Many dissections recanalize / normalize during the
first three months, but whenever there is persistent abnormality in follow-up angiography
or MRI + MRA, anticoagulation is often continued for three more months. Recurrence of
symptoms is rare after 6 months, and despite possible abnormalities in follow-up
angiography or MRI + MRA, warfarin is usually discontinued and replaced by
antiplatelet therapy, generally aspirin. (Schievink 2000.) Anticoagulation is contraindicated when there is a large infarction involving more than one third of the middle
cerebral artery territory (Adams et al. 1996) with an associated mass effect, hemorrhagic
transformation of the infarcted area, an intracranial aneurysm, or intracranial dissection
with SAH. Intravenous and intraarterial thrombolysis with urokinase or recombinant
tissue plasminogen activator have been used with promising results in the treatment of
patients with ICAD or VAD-related acute stroke and no contraindications for
thrombolysis (Derex et al. 2000, Arnold et al. 2002, Restrepo et al. 2003). Both
anticoagulation and thrombolysis, however, entail a minor risk of intracranial
hemorrhage, but the risk of the extension of intramural hematoma during treatment is
mainly theoretical (Mokri et al. 1986, Arnold et al. 2002).
Endovascular treatment should be considered in patients who remain symptomatic
(thromboembolic events or progression) despite therapeutic anticoagulation, in cases of
persistent pseudoaneurysms, and when anticoagulation is contraindicated (Liu et al. 1999,
Simionato et al. 1999, Saito et al. 2000, Schievink 2000). The most commonly used
technique in the case of persistent, hemodynamically significant stenosis is percutaneous
balloon angioplasty followed by placement of one or more balloon-expandable or,
preferably, self-expanding metallic stents. It has been successfully applied and reported in
extracranial ICAD (Liu et al. 1999, Simionato et al. 1999, Malek et al. 2000,
Albuquerque et al. 2002). Dissecting aneurysms may require coil embolization or
placement of a covered stent (Manninen et al. 1997). Vanninen et al. (2003) reported
successful treatment of an intrasellar iatrogenic ICA pseudoaneurysm with a
polytetrafluoroethylene-covered stent. Dissecting aneurysm, especially in a vertebral
artery and in cases of SAH, can be treated with endovascular occlusion of the parent
artery at or just proximal to the dissection site (Halbach et al. 1993, Iihara et al. 2002,
Leibowitz et al. 2003). Therapeutic intervention should be performed as soon as possible
34
because of the high risk of rebleeding. Internal trapping of the dissected site is preferable
to proximal occlusion, while proximal occlusion does not completely eliminate the risk of
rebleeding (Kitanaka et al. 1992, Iihara et al. 2002). When the dissection of VA involves
PICA, proximal endovascular occlusion performed during the acute stage followed by
internal trapping of the dissected site, if possible, based on the results of a balloon test
occlusion of the distal VA segment of the dissected site, may be the treatment of choice
(Iihara et al. 2002). All endovascular treatments carry a potential risk of thrombosis and
consequent ishemic complications, and the long-term results of stenting are unknown,
which is why endovascular treatment should be used in selected cases only.
Surgical treatment of CCAD is rarely needed and should be considered only in patients
with persistent symptoms despite maximal medical therapy and who are not candidates
for endovascular treatment. The techniques used include arterial ligation, an in situ
interposition graft, extracranial to intracranial bypass, and rarely, surgical clipping of the
dissecting aneurysm (Schievink 2000).
2.8 Prognosis of CCAD
Prognosis in extracranial CCAD is mainly related to the possible presence of cerebral
ischemic complications, while local manifestations rarely significantly disable patients.
The reports on outcomes in extracranial ICAD are controversial. Many authors have
reported excellent or good recovery in 71–85% of extracranial ICAD patients (Mokri et
al. 1986, Leys et al. 1995, Saver & Easton 1998, Engelter et al. 2000). Poor outcomes
have been reported in about half of patients with occlusive dissection of extracranial
ICAD and acute stroke (Bobousslavsky et al. 1987, Ast et al. 1993). Pozzati et al. (1990)
reported high mortality (23%) and poor outcome (37%) in patients with cerebral
infarction due to occlusive dissection of ICA, while the rate of early death due to brain
infarction was 2–5 % in other reports (Schievink et al. 1994a, Leys et al. 1995, Saver &
Easton 1998). Milhaud et al. (2002) reported worse outcome in stroke patients with
occlusion of ICA due to dissection than in stroke patients with occlusion of ICAD due to
atherothrombosis. Traumatic dissection of ICA may have a worse prognosis than
spontaneous ICAD (Mokri 1990).
Extracranial VAD is considered to have good prognosis, with excellent or good
recovery in 67–87 % of the patients (Chiras et al. 1985, Mokri et al. 1988a, de Bray et al.
1994, Leys et al. 1995, Saver & Easton 1998, Saeed et al. 2000, Czechowsky & Hill
2002). Mortality rate is estimated to be about 6 % (Saver & Easton 1998).
The prognosis of intracranial dissection is related to ischemic complications and
bleeding. Worse outcomes in intracranial compared to extracranial ICAD, with mortality
rates up to 75 % (Hart & Easton 1983, Schievink et al. 1994b), have been reported. In
intracranial VAD, prognosis is controversial. de Bray et al. (1997) reported poor outcome
in 44 % of patients with intracranial VAD, while others have reported 84–88 % of patients
to have good or excellent recovery (Mokri et al. 1988a, Hosoya et al. 1999, Nakagawa et
al. 2000). Reports on outcomes in rare intracranial dissections of basilar artery, ACA,
MCA, and PCA are scant. Anterior cerebral artery dissection (ACAD) seems to have
good prognosis, with 78 % of patients recovering well (Ohkuma et al. 2003a). In a report
35
of 13 middle cerebral artery dissection (MCAD) patients (cerebral ischemia in 4 and SAH
in 9), good recovery was only seen in 3 patients, while 5 had severe disability, 1 remained
in vegetative state, and 4 died (Ohkuma et al. 2003b). In the same study, about 50 % of
patients with SAH suffered from rebleeding, which impaired their prognosis. There are no
large reports on long-term outcomes in posterior cerebral artery dissection (PCAD), but in
a review by Berger & Wilson (1984), one patient with isolated PCAD had a good
outcome, as did three patients in another review (Pozzati et al. 1994).
No correlation between the clinical outcome of patients and angiographic outcome of
the dissected artery has been reported (Steinke et al. 1994, Provenzale et al. 1996, de
Bray et al. 1997). Occlusion of extracranial ICAD at the initial angiography seems to be
associated with poor outcome (Bogousslavsky et al. 1987, Milhaud et al. 2002). Touze et
al. (2001) reported a very small risk of ischemic events or other complications under
antiplatelet treatment in aneurysmal forms of ICAD and VAD. Guillon et al. (1999)
described a benign clinical course in extracranial ICAD with dissecting aneurysm.
The risk of recurrent dissection in an initially unaffected artery is about 2% during the
first month and about 1% per year (Schievink et al. 1994a). In a large series, recurrent
events in other vessels were found in 8%, and a family history of arterial dissections was
found to be the only factor increasing the risk (Schievink et al.1996). However,
recurrence of dissection in the same vessel is extremely rare (Ast et al. 1993, Schievink et
al. 1996).
3 Purpose of the study
The purpose of the present study was:
1. to investigate the angiographic spectrum and course of cervicocephalic artery
dissection. (I, II)
2. to evaluate pulsatile tinnitus as a symptom of cervicocephalic artery dissection. (III)
3. to assess the frequency of brain infarctions and subarachnoid hemorrhage detectable
by CT/MRI in cervicocephalic artery dissection patients, to evaluate the correlation
between initial vessel wall findings and brain infarctions, and to explore the patterns
of cerebral infarctions in oder to find out if they are of embolic or hemodynamic
origin. (IV)
4. to examine the long-term clinical outcome of cervicocephalic artery dissection
patients, to find out if there are differences in outcome between extra- and intracranial
CCAD patients, and to examine the predictive clinical value of the angiographic
finding of the dissected vessel at the time of diagnosis. (I, V)
4 Patients and methods
4.1 Patients
The total number of patients included in these studies was 136 (42 women). They were
consecutive patients diagnosed with CCAD at Oulu University Hospital between August
1982 and March 2002. Diagnosis was based on clinical presentation, angiographic (124
patients) or MRI and MRA (12 patients) findings, and exclusion of other specific arterial
wall pathologies, e.g. atherosclerosis and arteritis. The mean age of the patients was 42.9
years, range 1–72 years. Potential precipitating events, vascular risk factors, and
concomitant diseases are listed in Table 1.
Table 1. Potential precipitating events, vascular risk factors, and concomitant diseases in
the 136 patients participating in the present study with cervicocephalic artery dissection.
Precipitating event
n
Activities involving physical strain
35
Head or neck trauma
17
Forced head turning
4
Iatrogenic, complication of angiography
2
Cesarean section
1
Caldwell-Luc operation
1
Sneezing
1
Vascular risk factors or diseases¹
n
Arterial hypertension²
60
Elevated serum cholesterol levels
40
Diabetes mellitus
9
Fibromuscular dysplasia
9
Family history of cerebral infarction or intracranial hemorrhage
9
Family history of aortic or cervicocephalic artery dissection
2
Arteritis
1
Klinefelter syndrome
1
Essential thrombocytopenia
1
¹ patients may have more than one vascular risk factor or disease, ² 24 patients had medication at the time of the
index event
38
The 136 patients had a total of 165 CCADs. Of them, 139 dissections (found in 114
patients) were extracranial and 26 dissections (found in 22 patients) intracranial (Table 2).
Table 2. Dissected vessels in 136 consecutive CCAD patients.
Dissected vessel
Extracranial dissections
Intracranial dissections
Unilat
Bilat
Other
Unilat
Bilat
n
n
n
n
n
ICA
62
10
VA
30
9
Other
n
6
6
ICA + VA
1
1
ICA + bilateral VA
1
Bilateral ICA + bilateral VA
1
ICA+ MCA
3
MCA
3
PICA
1
PCA
1
basilar artery
1
Unilat = unilateral, Bilat = bilateral, n = number of patients
Paper I describes the first nine consecutive patients (two women) diagnosed to have ten
angiographically proved intracranial cervicocephalic artery dissections. The numbers of
patients covered in the original papers are presented in Table 3. The patients were aged
from 5 to 49 years (mean 29 years). Five dissections were in ICA (in one case the
extracranial part of ICA was also affected), four in VA (in one case the extracranial part
of VA was also affected), and one in the posterior inferior cerebellar artery.
Table 3. Numbers of patients in the original papers.
Paper
Number of patients
Period of data collection
I Intracranial dissections
9
II Extracranial dissections
93
August 1982 – September 1996
August 1982 – August 1999
III Pulsatile tinnitus
16 ¹
August 1982 – March 2002
IV Brain manifestations
131 ²
August 1982 – March 2002
V Clinical outcome
136 ³
August 1982 – March 2002
¹ A subgroup of patients with pulsatile tinnitus derived from the total number of 136 patients in paper V.
² Five patients without brain imaging were excluded from the total number of 136 patients in paper V.
³ The patients covered in the papers I–IV are included.
Paper II covered the first 93 consecutive patients (32 women) with 111 angiographically
(one MRA) proved extracranial internal carotid and vertebral artery dissections and one
concomitant intracranial vertebral artery dissection. The patients’ ages ranged from 3 to
72 years (mean 45.2 years). There were 55 unilateral ICADs, 22 unilateral VADs, 9
bilateral ICADs, 5 bilateral VADs, 1 unilateral ICAD + bilateral VAD, and 1 bilateral
ICAD + bilateral VAD.
Paper III deals with 16 of the 136 CCAD patients (9 women, mean age 43.8 years,
range 33–55 years), who had pulsatile tinnitus as a presenting symptom. They had 11
39
unilateral ICADs, 2 bilateral ICADs, 2 unilateral VADs, and 1 bilateral ICAD + bilateral
VAD, all dissections being extracranial.
Of the 136 patients described in paper V, 5 patients with extracranial dissections (3
unilateral ICADs, 1 bilateral ICAD, and one unilateral VAD) were excluded from paper
IV, because they had no brain imaging. The remaining 131 patients (42 women, mean age
42.7 years, range 1–62 years) had a total of 159 CCADs. One hundred and nine patients
had extracranial dissections and 22 intracranial dissections.
4.2 Methods
4.2.1 Clinical information
Information on the 136 patients was gathered from the medical records of our hospital
and reviewed retrospectively by a radiologist (O.P.). The patients’ symptoms and signs,
their treatment, and their clinical outcome were analyzed. All patients had undergone a
clinical examination by a staff neurologist. The symtoms and signs listed were cerebral
infarction, cerebellar infarction, carotid TIA, vertebrobasilar TIA, head or neck pain,
Horner’s syndrome, pulsatile tinnitus, cranial nerve palsy, dizziness / disturbance of
equilibrium, seizure / loss of consciousness, amaurosis fugax, diplopia, dysphagia,
disorientation, visual field defect, loss of hearing, and no symptoms. Pulsatile tinnitus
was analyzed in more detail in paper III, where it was evaluated as the presenting or sole
symptom of CCAD, and its duration was assessed to be days, weeks, months, years, or
persistent. The follow-up time ranged from five days in a patient who died in the acute
phase up to 11 years and 7 months, median 11 months. In paper III, follow-up data of 13
patients were also collected by telephone interviews (O.P.) in November 2002, and one of
these patients was followed up for 14 years and 3 months. Two methods used to assess
long-term clinical outcome are presented in Table 4. Based on the neurological defects
and symptoms, the clinical outcome was classified into five categories by a radiologist
(O.P.) and a neurologist (K.S.). The categories ranged from 1 (no neurological defects, no
symptoms) to 5 (death). Outcomes were considered good in the categories 1–3. This
classification takes into account functional outcome, but also symptoms and disabilities,
such as visual defects, neuropsychological defects, and problems in the cognitive domain.
Functional outcome was assessed by using the universally known modified Rankin Scale
(mRS), with scores ranging from 0 (no symptoms) to 6 (dead) (Wade 1992). A good
outcome was defined as mRS 0–2. The differences in outcome between extra- and
intracranial CCAD patients were evaluated, and the predictive clinical value of the
dissection-related vessel finding at angiography at the time of diagnosis was assessed
using Fisher’s exact test. p-values of less than 0.05 were considered significant.
40
Table 4. The two methods used to assess long-term clinical outcome in 136 patients with
cervicocephalic artery dissection.
Five-category classification based on
neurological defects and symptoms
modified Rankin Scale
1 = No neurological defects,
no symptoms.
0 = No symptoms at all.
2 = No neurological defects,
symptoms present.
1 = No significant disability despite symptoms; able to carry
out all usual duties and activities.
3 = Persistent neurological defects,
no or mild disability.
2 = Slight disability; unable to carry out all previous activities,
but able to look after own affairs without assistance.
4 = Persistent neurological defects,
moderate or severe disability.
3 = Moderate disability; requiring some help, but able to walk
without assistance.
5 = Death.
4 = Moderately severe disability; unable to walk without
assistance, and unable to attend to own bodily needs without
assistance.
5 = Severe disability; bedridden, incontinent, and requiring
constant nursing care and attention.
6 = Death.
4.2.2 Vascular imaging
4.2.2.1 Initial angiography
Intra-arterial DSA with transfemoral catheterization was performed initially on 125
patients (until 1989 a 512 x 512 matrix, Angiotron, thereafter a 1024 x 1024 matrix,
Polytron, Siemens, Erlangen, Germany). One hundred and five patients underwent aortic
arch injections in postero-anterior and left and right oblique projections. Seventy-one of
them and further 20 patients had selective common carotid or vertebral injections in at
least two orthogonal directions. Angiography confirmed the diagnosis in 124/125
patients, but in one case, in which dissection did not cause any luminal abnormalities,
angiography was negative. All the initial angiograms were analyzed retrospectively from
films by a radiologist (O.P.) and another radiologist (S.L.) or a neuroradiologist (T.T.).
The findings were classified as irregular stenosis, occlusion, pseudoaneurysm, fusiform
aneurysm, intimal flap, irregular dilatation, double lumen, and irregularity of vessel wall.
In each dissected vessel, one major finding was considered to be the main finding, while
the other findings were considered additional. A detailed overall analysis of the initial
angiographic (one MRA) findings of the 121 dissected cervicocephalic arteries from the
studies I and II and the levels of VADs was done (I,II).
41
4.2.2.2 Initial MRI + MRA
The dissection was diagnosed with MRI + MRA in ten patients, including the one with a
previous negative angiography. In two further patients, MRI without MRA was sufficient
to confirm the diagnosis. MRI and MRA were performed by using a 1.5 T scanner (Signa
Echospeed 1.5 T, General Electric, USA). In all of the twelve patients, MRI of the head
included spin echo T1-weighted sagittal and/or coronal images and T2 and/or
PD-weighted axial images. Slice thickness was 4–5 mm. Contrast-enhanced T1, T2*, EPI
FLAIR, FLAIR, and T1 fat sat images were acquired from only a few/single patients. In
two cases, DWI was done in the axial plane. MRA was performed initially in ten cases,
3D TOF in three, 3D TOF and 2D PC in two, 2D TOF and 2D PC in two, and 3D TOF,
2D TOF, and i.v. bolus angiography in three. MRA confirmed the diagnosis in all of the
ten patients. All the MRI and MRA images were reviewed retrospectively by a
radiologist (O.P.) and a neuroradiologist (T.T. or J.P.). A narrowed lumen with an adjacent
crescent-shaped increased signal in T1-weighted images representing an intramural
hematoma was considered pathognomonic. The other findings were occlusion seen as an
increased signal or absent visualization of the entire vessel, marked narrowing of the
vessel, and enlarged vessel diameter. MRI and MRA findings of the dissected vessels
were needed when the correlation between the initial vessel wall findings and the brain
infarctions as well as the long-term clinical outcome was assessed.
4.2.2.3 Follow-up angiography
In paper I, follow-up angiograms were obtained 1–6 months (mean 3.2 months) after the
initial angiography and were available for review in all of the nine cases. One patient had
a new dissection 7 months after the first dissection, but this was not followed up with
angiography. In paper II, follow-up angiograms performed within 6 days to 19 months
(mean 4.0 months) after the initial angiography were available for review in 77/93 cases
(83%). The follow-up findings of the dissected vessels were classified as total or partial
normalization / recanalization, slight improvement, occlusion of an initially stenosed
artery, no change, growth of a pseudoaneurysm, and other occasional findings.
4.2.3 Cerebral CT and MRI
All but five of the 136 patients underwent either cerebral CT (127) (Somatom Plus,
Siemens, Germany, Highspeed, General Electric, USA or Aquilion, Toshiba, Japan) or
MRI (4) (Magnetom 1 T, Siemens, Erlagen Germany or Signa Echospeed 1.5 T, General
Electric, USA) a mean of 2.1 days after the onset or distinct aggravation of symptoms
(range 0–20 days). Conventional spin-echo T2-weighted and T1-weighted images in the
axial, sagittal, and coronal planes were included. Follow-up brain imaging was done on
73 patients (51 CT, 22 MRI) a mean of 13.2 days after the onset or aggravation of
42
symptoms (range 0–120), and a third imaging was done on 29 patients (21 CT, 8 MRI), a
fourth imaging on 14 (9 CT, 5 MRI), and a fifth imaging on 6 (1 CT, 5 MRI). All of the
CT and MRI scans were analyzed by a radiologist (O.P.) and a neuroradiologist (J.P. or
T.T.) to evaluate the presence and topography of cerebral ischemic and/or hemorrhagic
lesions. Of the 74 infarctions detected in CT / MRI, the films of 70 were available for
detailed analysis.
Infarctions involving the anterior circulation were differentiated into four patterns
according to Szabo et al. (2001). Territorial infarction involves the cerebral cortex of one
or more major cerebral artery territories. Subcortical infarction affects the territory of
deep perforating branches originating from the distal ICA or the MCA, involving the
basal ganglia, thalamus, internal capsule, or centrum ovale. Territorial infarction with
fragmentation represents a lesion similar to territorial infarction with additional smaller
lesion(s) in cortical regions. Watershed infarctions are located between two arterial
territories in regions considered to constitute a hemodynamic risk zone.
Of the posterior circulation infarctions, only the side of the infarction and the presence
of hemorrhagic lesions were determined.
4.2.4 Statistical analysis
The data in the present study were analyzed using cross-tabulation. In study V, Fisher’s
exact test was used in statistical evaluations, and p-values of less than 0.05 were
considered significant.
5 Results
5.1 Angiographic spectrum and course of CCAD (I,II)
The main initial and follow-up angiographic findings of the 121 dissected vessels from
the studies I and II are presented in Table 5. There were 10 intracranial dissections (nine
in study I and one in study II) and 111 extracranial dissections (study II). Recurrent
dissections were detected in follow-up in five previously unaffected arteries (in four of
the 102 patients, 4 %), but they are not included in the table.
Table 5. Main initial and follow-up angiographic findings in 10 intracranial and 111
extracranial cervicocephalic artery dissections (I,II).
Initial angiography
Follow-up angiography
OccTotal or
No. of
No. of
lusion
partial
angiodissecgraphically normalization
ted
or
arteries controlled
recanalization
arteries
Irregular stenosis
61
54
44
Occlusion
40
32
11
Pseudoaneurysm
14
13
1
Irregular dilatation
5
4
2
Double lumen
1
1
1
121
104
59
Total
3
Slight
improvement
3
No
change
Growth
Other
1
1**
4
20
2
8
1*
1
3
5
33
1***
1
3
* further thrombosis of the artery, ** artery ligated, *** irregular dilatation normalized, new intimal flap
The most common main finding in initial angiography (one MRA) was irregular stenosis,
which was seen in 50% (6/10 of the affected intracranial and 55/111 of the affected
extracranial arteries). In follow-up, 81% of the stenoses showed total or partial
normalization. Occlusion was the second most common main finding, being detected in
33% (2/10 and 38/111). Occlusion was persistent in follow-up in 66%, showing even
further thrombosis of the artery in one case, while 34% showed total or partial
recanalization. Pseudoaneurysm was the main initial angiographic finding in 12% (1/10
44
and 13 /111). No change was seen in follow-up in 62%, and growth was seen in only one
case. Irregular dilatation (5/121) and double lumen (1/121) were rare findings.
Additional findings were detected in initial angiography in two intracranial (1 ICA, 1
VA) and 19 extracranial (18 ICA, 1 VA) vessels: an intimal flap (10), a pseudoaneurysm
(10), irregular stenosis (2), and irregular dilatation (2).
Among the 39 VADs in the studies I and II, the most often and almost equally
frequently affected levels of VAD were the C6-Th1 and C1-C2 levels (Table 6). Of the
four intracranial VADs, three were in the post-PICA segment.
Table 6. Levels of 39 vertebral artery dissections (I, II).
Level
n
Intracranial
4
Skull base - intracranial
2
skull base - C1
6
skull base - Th1
1
C1 - C2
13
C5 - C6
1
C4 - Th1
1
C6 - Th1
11
C = cervical vertebrae, Th = thoracal vertebrae.
5.2 Pulsatile tinnitus as a symptom of CCAD (III)
Pulsatile tinnitus was a presenting symptom in 12% of the CCAD patients (16/136). On
admission, ten patients presented with subjective tinnitus and five with objective tinnitus.
One ICAD patient had objective tinnitus as the only presenting symptom, while the
others had concomitant head or neck pain, ischemic brain symptoms, Horner’s syndrome,
or cranial nerve palsies. In the remaining one case, subjective tinnitus appeared three
months after the first symptoms of bilateral ICAD, although contralateral cervical bruit
was evident on admission. Unilateral ICAD was the cause of pulsatile tinnitus in eleven
patients, unilateral VAD in two, bilateral ICAD in two, and bilateral ICAD+bilateral VAD
in one patient. All patients with objective tinnitus had ICAD. All dissections were
extracranial, and the most common finding was irregular stenosis in eight patients. Two
patients had irregular stenosis and pseudoaneurysm, two pseudoaneurysm and intimal
flap, two pseudoaneurysm, one double lumen, and one occlusion. According to telephone
interviews, pulsatile tinnitus lasted for five minutes in one patient, for 1–7 days in three,
for 1–8 weeks in 7, and for six months in two. Three patients could not be reached by
telephone, and the duration of pulsatile tinnitus could not be defined based on medical
records.
45
5.3 Cerebral CT and MRI findings in CCAD (IV)
Of the 131 patients who underwent brain imaging, 73 (56%) had signs of altogether 74
cerebral or cerebellar infarctions. Infarction was already seen in the first scan performed a
mean of 2.1 days, range 0–20 days, after the onset or distinct aggravation of the
symptoms in 55 patients. In one case cerebral CT was performed at 10 months, and in
two cases the onset or aggravation of symptoms could not be defined. A repeated scan
(51 CT and 22 MRI) (0–120 days, mean 13.2 days, after the onset or aggravation of
symptoms) showed 17 new infarctions, and in one case a fourth scan showed the first
signs of infarction. Based on a correlation with the dissected vessels, 47 % of the
dissected vessels caused a detectable brain infarction in a relevant territory, while 53 % of
the dissected vessels did not cause signs of infarction (Table 7). Occlusion of the
dissected vessel was accompanied by infarction in 76 %, irregular stenosis in 40 %, and
other findings, such as pseudoaneurysm, fusiform aneurysm, irregular dilatation, double
lumen, and irregularity of the vessel wall, only rarely (Table 7). Not all clinically evident
infarctions could be detected in brain imaging: 88 patients (67%) had symptoms of
cerebral (57) or cerebellar (31) infarction, but brain imaging showed signs of infarction in
46 and 17 of them. Thirteen of the 88 patients had no signs of infarction in brain imaging
done at least three days after the onset of the symptoms.
Table 7. Frequency of infarctions in brain imaging correlated with vessel findings in 156
dissected cervicocephalic arteries (in three cases with simultaneous ICAD + MCAD on
the same side, the number of vessels is counted as one).
Vessel finding
Infarction at CT / MRI (n)
No infarction (n)
Irregular stenosis
40 % (32)
60 % (48)
Occlusion
76 % (39)
24 % (12)
Other *
12 % (3)
88 % (22)
Total
47 % (74)
53 % (82)
* pseudoaneurysm, fusiform aneurysm, irregular dilatation, double lumen, or irregularity of the vessel wall
Fourty-four of the 74 detected infarctions were in the anterior and 30 in the posterior
circulation. Of the 30 posterior circulation infarctions, 17 were related to occlusion, 12 to
irregular stenosis, and one to another nonocclusive finding of the dissected vessel.
Intracranial posterior circulation dissection was associated with cerebral infarction rarely
(in 1/11 of the dissected vessels), whereas intracranial anterior circulation dissections
were accompanied by infarctions more commonly (in 9/12 of the dissected vessels). The
stroke patterns of 41 anterior circulation infarctions were analyzed, and the results are
presented in Table 8.
46
Table 8. Stroke patterns of 41 infarctions in anterior circulation compared to angiographic
or MRA findings of the dissected vessels.
Territorial
infarction (n)
Subcortical
infarction (n)
Territorial infarction with
fragmentation (n)
Watershed
infarction (n)
Irregular stenosis
37 % (7)
32 % (6)
26 % (5)
5 % (1)
Occlusion
35 % (7)
20 % (2)
50 % (10)
5 % (1)
50 % (1)
50 % (1)
–
22 % (9)
39 % (16)
5 % (2)
Other *
Total
–
34 % (14)
* pseudoaneurysm, fusiform aneurysm, irregular dilatation, double lumen, and irregularity of the vessel wall
Since territorial and subcortical infarctions and territorial infarctions with fragmentation
are considered embolic, 95% (39/41) of the cerebral infarctions in the present study were
of embolic origin. Hemodynamic etiology was indicated in only two cases (5 %) with
watershed infarctions. The stroke patterns did not correlate with the angiographic
findings, and only territorial infarctions with fragmentation occurred slightly more often
in cases of occlusion (10/20) than in cases of stenosis (5/19) (Table 8).
Subarachnoid hemorrhage was found in CT in 5 of the 22 patients (23%) with
intracranial dissection, all the dissections being in the posterior circulation vessels (two
VADs, one PCAD, one posterior inferior cerebellar artery dissection (PICAD), and one
dissection of the basilar artery). One of them also involved signs of infarction.
Hemorrhagic transformation of cerebral infarction was present in 5/131 patients at the
initial or follow-up brain imaging, and the dense middle cerebral artery sign in CT in six.
5.4 Long-term clinical outcome in CCAD and the correlation between
outcome and initial angiographic or MRA finding (I,V)
The clinical manifestations of the 136 CCAD patients are listed in Table 9.
47
Table 9. Clinical manifestations in 136 cervicocephalic artery dissection patients.
Symptom¹
Total
(n=136)
%
(n)
ICAD²
(n=84)
%
(n)
VAD³
(n=49)
%
(n)
Head or neck pain
76 % (103)
70 % (59)
Cerebral infarction
42 % (57)
63 % (53)
Cerebellar infarction
23 % (31)
Horner’s syndrome
21 % (28)
20 % (17)
22 % (11)
Dizziness / disturbance of
equilibrium
19 % (26)
4%
45 % (22)
ICAD + VAD
(n=3)
%
(n)
88 % (43)
33 % (1)
4%
67 % (2)
(2)
63 % (31)
(3)
Cranial nerve palsy
13 % (18)
14 % (12)*
Pulsatile tinnitus
12 % (16)
15 % (13)
Carotid TIA
11 % (15)
18 % (15)
Seizure / loss of consciousness
7%
(9)
4%
Vertebrobasilar TIA
5%
(7)
Amaurosis fugax
3%
(4)
Diplopia
2 % (3)
Disorientation
2 % (3)
4 % (3)
Dysphagia
1 % (2)
2 % (2)
Visual field defect
1%
Loss of hearing
1 % (1)
Asymptomatic
2 % (3)
(3)
33 % (1)
12 % (6)**
4%
(2)
33 % (1)
12 % (6)
14 % (7)
4%
(3)
2%
(1)
6 % (3)
(2)
4%
(2)
2 % (1)
2 % (2)
2 % (1)
n = number of patients, ¹ multiple symptoms present in several patients, ² includes patients with middle cerebral
artery dissection with or without simultaneous ICAD, ³ includes patients with dissection of posterior inferior
cerebellar artery, posterior cerebral artery or basilar artery, * six V, four IX, one X, one XII, ** three VII, one V,
VI and IX.
One hundred and eleven patients were treated with anticoagulants (heparin and / or
warfarin). The mean duration of oral anticoagulation was 5.7 months. Seventy-three
patients had acetylsalisylic acid (ASA) after warfarin. Eight patients were treated with
acetylsalisylic acid only, 7 with surgery or endovascular treatment, and 10 with no
specific treatment.
Nine patients died during follow-up, four due to massive infarction within 9 days after
the onset of symptoms, one due to status epilepticus at 3 years, and four due to causes
unrelated to dissection (two myocardial infarctions, one pulmonary embolism after hip
fracture, and one death at over three years from an unknown cause).
When outcomes were classified into five categories based on neurological defects and
symptoms, 60% (82/136) of all CCAD patients in the present study recovered well,
having no or mild disability (Table 10). When dissection affected only one
cervicocephalic artery and did not cause vessel occlusion, the prognosis was good; 79%
of the patients recovered with no or mild disability. When non-occlusive dissection
affected two or more arteries, the outcome was poorer; 50% of the patients had at least
moderate disability. In cases of occlusive dissection in one or more arteries, the outcome
was even less advantageous, with only about 35% of the patients recovering enough to
have no or mild disability. There was a statistically highly significant difference
(p<0.0001) in outcome between the patients with non-occlusive dissection in one or more
48
arteries and those with occlusive dissection in one or more arteries; 75% of the former but
only 35% of the latter recovered well, having no or mild disability.
Table 10. Initial angiographic vessel findings and long-term clinical outcome based on the
neurological defects and symptoms in 136 patients with cervicocephalic artery dissection.
Number of dissected
vessels and
angiographic finding
Clinical outcome
No neurol.
defects, no
symptoms
%
(n)
No neurol.
defects, symptoms present
%
(n)
1 vessel, no occlusion
30.1 % (22)
13.7 % 10)
35.6 % (26)
15.1 % (11)
5.5 % (4)
1 vessel, occlusion
13.5 % (5)
8.1 % (3)
13.5 % (5)
54.1 % (20)
10.8 % (4)
≥ 2 vessels, no
occlusion
28.6 % (4)
7.1 % (1)
14.3 % (2)
50.0 % (7)
–
8.3 % (1)
8.3 % (1)
16.7 % (2)
58.3 % (7)
8.3 % (1)
23.5 % (32)
11.0 % (15)
25.7 % (35)
33.1 % (45)
6.6 % (9)
≥ 2 vessels, occlusion
Total
Persistent neurol. Persistent neurol.
Death
defects, disability defects, disability
no or mild
moderate or severe
%
(n)
%
(n)
%
(n)
When mRS was used to assess outcome, the results were better. Good recovery, defined
as mRS 0–2, was found in about 79% of all the patients and in 89% of the patients with
non-occlusive dissection of one cervicocephalic artery. In case of non-occlusive
dissection of two or more arteries 86% and in case of occlusive dissection of one or more
arteries 61% of the patients recovered well. Also, when mRS was used, there was a
statistically highly significant difference (p=0.0004) in outcome between the patients with
non-occlusive dissection in one or more arteries and those with occlusive dissection in
one or more arteries; 89% of the former but only 61% of the latter recovered well, scoring
0–2 on mRS.
No significant differences in recovery between intra- and extracranial dissections were
seen in either of the two outcome classifications (p=0.481 / p=0.785). Outcomes were
also similar in the patients with anterior and posterior circulation dissections, except that
all the four patients who died due to large dissection-related infarctions had dissection in
the anterior circulation.
All the five patients with SAH recovered well: three patients had persistent defects but
only mild or no disability, and the remaining two cases had no defects or symptoms. The
hyperattenuating middle cerebral artery sign in CT seen in six patients predicted a poor
outcome; two died within a week, and the remaining four had persistent neurological
defects and disability.
6 Discussion
In the present study, the angiographic spectrum and the course of cervicocephalic artery
dissection were studied. Initial and follow-up angiograms were retrospectively analyzed,
and the intial and follow-up vessel findings were categorized. Clinical manifestations,
especially pulsatile tinnitus as a presenting symptom, were evaluated. In addition, the
current pathogenetic concepts of topography of cerebral infarction were used as
indicators of hemodynamic versus embolic etiology. Anterior circulation infarctions were
analyzed, and territorial, subcortical, and territorial infarctions with fragmentation were
considered embolic. Comparisons were made between the different vessel findings, the
frequencies and patterns of brain infarction, and the long-term clinical outcome.
Irregular stenosis, which normalized in 81%, was the most frequent finding, being
present in half of the patients in the studies I and II. Occlusion, which recanalized in
about 34%, was found initially in one third of the patients. The symptoms showed a wide
range of variation, and pulsatile tinnitus was the presenting symptom, sometimes the only
symptom, in more than one out of ten patients with CCAD (III). Ninety-five per cent of
the analyzed anterior circulation infarctions were considered embolic (IV). Occlusion was
accompanied by infarction more often than other vessel findings (IV), and in long-term
clinical follow-up, the prognosis was worse in occlusive dissection of one or two
cervicocephalic arteries (V).
The 136 patients in the present study were consecutive patients diagnosed with CCAD
at Oulu University Hospital between August 1982 and March 2002. Diagnosis was based
on clinical presentation, angiographic (124 patients) or MRI and MRA (12 patients)
findings, and exclusion of other specific arterial wall pathologies, e.g. atherosclerosis and
arteritis. Both intra- and extracranial dissections of ICA and VA were included together
with fewer dissections in intracerebral arteries. There may have been some selection bias
in the sense that the patients with mild symptoms may not have consulted the hospital or
undergone extensive investigations, such as angiography, to confirm the diagnosis.
Furthermore, some dissections, especially intracranial dissections, have probably gone
undiagnosed, especially when angiography without selective injections with a 512x512
matrix was used. Patients with unspecific occlusion in angiography showing no
recanalization on follow-up angiography were excluded from the present study, and some
of these occlusions are likely to have been caused by dissection. However, awareness of
50
this medical condition has increased and imaging modalities have improved during the
20-year period of data collection, and most missed cases are from the early years. The
present study population is relatively large compared to previous reports. To my
knowledge, the biggest series from a single institution was gathered at the Mayo Clinic
between 1970 and 1990. That series consisted of 200 patients with spontaneous ICAD.
The reports on that series focused on recurrent dissections. (Shievink et al. 1994a,
Schievink et al. 1996). Silbert et al. (1995) reported characteristics of headaches in 161
symptomatic patients with spontaneous dissections of ICA or VA, and they also belonged
to the series gathered at the Mayo Clinic between 1970 and 1990. Baumgartner et al.
(2001) reported on 181 consecutive patients with spontaneous ICAD, but the patients
were from two different hospitals, as was also the case in another report on the follow-up
of 105 consecutive patients with ICAD or VAD (Leys et al. 1995). The populations
reported in other studies do not exceed one hundred, are often selected, and sometimes
come from many, up to 46 institutions (Ohkuma et al. 2003b).
6.1 Imaging findings in dissected vessels
Angiography has long been the golden standard in diagnosing cervicocephalic artery
dissection, as in the present study. First conventional angiography and thereafter digital
subtraction angiography have confirmed the diagnosis by showing indirect signs of
dissection in the vessel lumen. Angiography is available in most centers. It is still the best
imaging modality that can show filiform stenosis with very slow blood flow, FMD
changes, and other irregularities of the vessel wall (Schulze et al. 1992, Link et al. 1996).
In addition, contrary to MRI and MRA, only angiography enables dural arteriovenous
fistulas to be excluded in a patient with pulsatile tinnitus and normal otoscopy (Shah et
al. 1999). Angiography can be, and nowadays usually is, done with selective injections,
giving better images and real-time information of flow dynamics. The relatively high
incidence of multivessel dissections, 19 % (26/136) in the present study and 16–28 % in
previous reports (Schievink et al. 1994a, Mokri et al. 1996, Ozodoba et al. 1996),
emphasizes the importance of four-vessel imaging. In endovascular treatments,
angiography is the only universally used guiding method.
There are, however, many disadvantages in the angiographic imaging of CCAD, which
have led to the developement and increasing use of other imaging modalities, such as
MRI and MRA, CT angiography, and Dupplex ultrasound. Firstly, angiography is
invasive, being associated with a 1% overall incidence of neurologic deficits and a 0.5%
incidence of persistent deficits (Warnock et al. 1993, Heiserman et al. 1994). In the
present study and, to my knowledge, in other studies on CCAD, no complications were
encountered after angiography. However, two recent studies reported clinically silent
embolism and ischemic lesions after angiography (Bendszus et al. 1996, Kato et al.
2003). Secondly, angiography can only show indirect signs of dissection in the vessel
lumen, and when the dissection occurs in the subadventitial layer without relevant
narrowing of the arterial lumen or when the aneurysm is thrombosed, conventional
angiography does not yield the diagnosis (Sturzenegger & Huber 1993). One of the
patients in the present study had left-sided Horner’s syndrome and retroorbital pain, but
51
no abnormalities in DSA. MRI and 3D TOF MRA done seven days later showed a typical
hyperintense crescent-shaped intramural hematoma in the extracranial part of the left ICA
with only minimal irregularity of the lumen. It is possible that some other cases in the
present study may have gone undiagnosed in normal initial angiography. When CCAD is
clinically strongly suggested but angiography is normal, or when there is an unspecific
occlusion in angiography, MRI and MRA should be performed to improve diagnostic
accuracy. There are also differential diagnostic problems in angiography. Atherosclerotic
disease, arteritis, arterial spasm, and partially recanalized emboli may cause similar
arterial narrowing and irregularity as dissection (Hoffman et al. 1993, Lazinski et al.
2000). Follow-up angiography approximately three months later may help in differential
diagnosis by showing changes in the vessel finding, e.g. recanalization / normalization or
development of a dissecting aneurysm, and thus confirming the diagnosis of CCAD. In
follow-up angiography of CCAD, irregular stenosis is often normalized, while that never
happens when atherosclerotic changes are involved. Clinical presentation may give more
information of the etiology. Mini-invasive imaging modalities, i.e. MRI, MRA, and CT
angiography, may also resolve some of the aforementioned problems in initial
angiography, but may also be negative in CCAD. Helical CT has higher resolution than
MRI and no flow-related artifacts, but near the skull base, artifacts from bone impair
image quality (Leclerc et al. 1998), and the use of this method may be limited by the
requirements for intravenous contrast and ionizing radiation (Kasner et al. 1997). MRI
with MRA (3D TOF) is sensitive in diagnosing ICAD (Levy et al. 1994, Bakke et al.
1996, Ozodoba et al. 1996), but its sensitivity to VAD is only 20% (Levy et al. 1994), and
it is not readily available in all hospitals.
Angiographic findings depend on the degree of luminal stenosis and the depth at which
the false lumen is situated in the vessel wall. Furthermore, dissection is a dynamic
process, and the degree of luminal stenosis may change within a short time; stenosis may
progress to occlusion, or contrariwise, occlusion may recanalize. Compared to previous
reports, no significant differences in initial vessel findings were found in the present
study. The most common main initial finding, which was present in 50% of the present
patients was irregular stenosis. This is in line with two previous reports, where stenosis
was found in 55–58% of cases (Ast et al. 1993, de Bray et al. 1994). Some authors have
reported figures of 70–80%, especially in VAD (Houser et al. 1984, Mokri et al. 1986,
Mokri et al. 1988a, Levy et al. 1994, Schievink et al. 1994b). The size of the population
varies in different studies, which may influence the incidence of stenosis, but all studies
have established stenosis as the most common vessel finding. Occlusion, found in 33% of
the present cases, has been reported in 11–35% previously (Hart & Easton 1983, Houser
et al. 1984, Mokri et al. 1986, Mokri et al. 1988a, Ast et al. 1993, deBray et al. 1994,
Levy et al. 1994, Schievink et al. 1994b). Pseudoaneurysm was the main initial finding in
the present study in 12%, but when additional findings were included, it was found in
20%, as in previous studies (Mokri et al. 1988a, Ast et al. 1993, de Bray et al. 1994, Levy
et al. 1994). Irregular dilatations and double lumens are rare. Despite the fact that there
are differences in the vessel wall structure between intra- and extracranial arteries, no
significant differences in the distribution of different vessel findings were found between
them in the present study.
All nine patients in study I and 77/93 patients in study II underwent follow-up
angiography. The total follow-up rate was thus 84%, which is clearly higher than the 38–
52
64% in previous studies (Houser et al. 1984, Mokri et al. 1986, Mokri et al. 1988a, Mokri
1990, Steinke et al. 1994). Stenoses have been reported to normalize totally or partially in
77–85% (Houser et al. 1984, Mokri et al. 1986, Mokri et al. 1988a, Mokri 1990), which
was also the case in the present study, where a 81% normalization rate was seen. More of
the occlusions than previously reported recanalized totally or partially, namely 34%. In
the study of Mokri (1990), only one (8%) of the initially occluded twelve ICAs
recanalized during follow-up, and the result was similar in dissected VAs, while 1/13
(8%) recanalized (Mokri et al. 1988a). Houser et al. (1984) found recanalization in 14 %
of the occluded arteries, but only 29/57 of all arteries were followed up.
Pseudoaneurysms rarely increase in size during follow-up, but this may happen, as was
the case in one of the present patients described in more detail in the study of Tikkakoski
et al. (1997). More than half of the pseudoaneurysms resolve or decrease in size (Houser
et al. 1984, Mokri et al. 1986, Mokri et al. 1988a, Touzé et al. 2001). In the present study,
no change was seen in follow-up in 62 % of the pseudoaneurysms, while in the study of
Touzé et al. (2001), 46% of the aneurysms persisted.
Previous reports claim that there is no correlation between the angiographic outcome
of the dissected artery and the clinical outcome the patient (Steinke et al. 1994,
Provenzale et al. 1996). Furthermore, disabling infarctions usually occur within the first
few days after the onset of symptoms (Biousse et al. 1995). Why, then, should we control
the vessel finding? Follow-up angiography may confirm the diagnosis of dissection.
Infarctions may occur as much as one month later (Biousse et al. 1995), and persistent
changes in follow-up angiography increase the risk of further ischemic complications.
Therefore, follow-up angiographic findings often affect therapeutic decisions. The risk of
recurrent stroke under antithrombotic medication (ASA or warfarin) remains low in both
stenotic or occlusive dissections (Leys et al. 1995, Kremer et al. 2003) and in aneurysmal
forms (Touzé et al. 2001). Pseudoaneurysms cannot be seen in initial angiography if they
are thrombosed, and they may only be visible in follow-up angiography. Pseudoaneurysm
may also increase in size and require endovascular treatment to prevent ischemic
complications or bleeding when intracranial. Recurrent dissections are rare, but a new
dissection in another vessel may be found in follow-up, especially in patients with a
family history of arterial dissections (Schievink et al. 1996) or in patients with new
symptoms. Mini-invasive imaging modalities, i.e. MRI, MRA, and CT angiography, can
also be used in follow-up.
6.2 Pulsatile tinnitus as a presenting symptom in CCAD
Tinnitus is defined as pulsatile when the patient describes a sound synchronous with heart
beat. It is classified as subjective when heard only by the patient and as objective when
the sound is also audible by auscultation. Pulsatile tinnitus is almost always due to the
sound of nonlaminar blood flow transmitted to the inner ear. This may occur in systemic
disease causing general alteration of the hemodynamics or in local disorders that are
anatomically close to or within the petrous bone. In a review of 140 extracranial ICADs
from 1975–1983, pulsatile tinnitus was the only symptom in 4% and an associated
symptom in 35% of patients (Hart & Easton 1983). Ast et al. (1993) reported subjective
53
tinnitus as a symptom in 7.5% (5/68) of ICAD patients. In another study, subjective or
objective pulsatile tinnitus was reported in 50% (18/36) of patients with ICAD (Mokri et
al. 1986). Steinke et al. (1994) reported pulsatile tinnitus in 12% of 48 ICAD patients. In
two studies, the frequency of pulsatile tinnitus was 16% (Baumgartner et al. 2001) and
27% (Silbert et al. 1995), out of 181 ICAD and 161 ICAD or VAD patients, respectively.
Pulsatile tinnitus was an associated symptom in all. VAD has been thought to be a very
rare cause of pulsatile tinnitus: in two studies VAD was never the cause of pulsatile
tinnitus (Silbert et al. 1995, Baumgartner et al. 2001), while in a third study only one
patient out of 25 with VAD presented with pulsatile tinnitus (Mokri et al. 1988a). In the
present study, nearly 12% of the patients (16/136) had pulsatile tinnitus. In all of the 16
cases, dissection was extracranial and located in ICA in 13 cases, in VA in 2 cases, and in
both ICA and VA in one case. Thus, of the present extracranial CCAD patients, 14% had
pulsatile tinnitus. Contrary to previous works (Hart & Easton 1983, Ast et al. 1993,
Baumgartner et al. 2001), the present patient series was collected in a single centre and
represented consecutive angiographically confirmed dissections. The present result is best
comparable with the results reported by Silbert et al. (1995), who had 161 symptomatic
patients with spontaneous dissections of ICA or VA at the Mayo Clinic between 1970 and
1990. The overall incidence of pulsatile tinnitus in the present study was lower, but
contrary to Silbert et al. (1995), three of the present VAD patients had pulsatile tinnitus.
Compared to the 12% frequency of pulsatile tinnitus in 48 consecutive extracranial ICAD
patients from a single hospital in the report of Steinke et al. (1994), slightly more (19%,
14/75) of the extracranial ICAD patients in the present study had pulsatile tinnitus. The
majority of the present patients with pulsatile tinnitus had concomitant head or neck pain,
ischaemic brain symptoms, Horner’s syndrome, or cranial neuropathies, and only one had
it as the only presenting symptom. Although pulsatile tinnitus overall is an uncommon
otologic symptom and mostly subsided in less than two months in the present study, it is
found quite frequently, and should arouse active suspicion of CCAD or other serious and
potentially life-threatening pathology, such as dural arteriovenous fistula,
carotid-cavernous fistula, extracranial and intracranial arteriovenous malformations,
atherosclerotic carotid stenosis, cerebral aneurysms, fibromuscular dysplasia, or
paraganglioma.
6.3 Cerebral CT and MRI findings
CT or MRI signs of cerebral infarction have been reported in 48–55% of CCAD patients
(Saver et al. 1992, Steinke et al. 1996). In agreement with previous reports, the frequency
of cerebral and cerebellar infarctions in the present study was 54% (73/136). Due to its
retrospective quality, no consistent imaging protocol was used in the present study.
However, all but five patients underwent at least one CT (127 patients) or MR (4 patients)
imaging, and follow-up brain imaging was done on 73 patients (51 CT, 22 MRI). Due to
the fact that CT was the modality used in most cases, some infarctions in the acute phase
have not been detected, and the use of MRI, or preferably diffusion-weighted MRI, would
have improved accuracy. CT is known to be less sensitive than diffusion-weighted MRI
in detecting signs of infarction during the first 12 hours after stroke (Mullins et al. 2002)
54
and also in posterior circulation territories (Sturzenegger 1994). Some clinically evident
infarctions could not be detected in brain imaging in the present study. Eighty-eight
patients (65%) had symptoms of cerebral (57) or cerebellar (31) infarction, but brain
imaging showed signs of infarction in 63 (46 and 17) of them. However, in 13 of them,
brain imaging was done at least three days after the onset of the symptoms, and no signs
of infarction were present even then.
The patterns of anterior circulation infarctions in ICAD patients were explored in order
to find out if the etiology of infarction is embolic or hemodynamic. Of the 44 infarctions
in anterior circulation, films of 41 were available for detailed analysis. The current
pathogenetic concepts of the topography of cerebral infarction were used as an indicator
of hemodynamic versus embolic etiology (Ghika et al. 1989, Fisher 1991, Bogousslavsky
1991). Territorial, subcortical, and territorial infarctions with fragmentation were
considered embolic, whereas watershed infarctions were considered hemodynamic. The
results of the present study strongly suggest that most infarctions in ICAD patients are of
embolic origin; only 5% of the infarctions (2/41) were in hemodynamic risk zones, which
is in agreement with the corresponding 5% rate in three previous studies (Steinke et al.
1996, Lucas et al. 1998, Baumgartner et al. 2001). Weiller et al. (1991) reported a nearly
50% frequency of hemodynamic infarctions in their study of fifteen patients with eleven
infarctions. Embolic etiology is also supported by reports of microembolic signs in
transcranial Doppler in CCAD patients (Srinivasan et al. 1996, Kimura et al. 2001) and
by histologic reports on thrombus formation in the lumen of the dissected vessel
(Bogousslavsky et al. 1987, Steinke et al. 1996). The etiology of brain infarction in ICAD
may, however, be a more complicated issue. The collateral blood supply and the
territories of the major cerebral arteries may vary. Furthermore, the CCAD itself is a
dynamic process. The degree of luminal stenosis may change within a short time, stenosis
may progress to occlusion, or contrariwise, occlusion may recanalize. Distal embolism
may be associated with all kinds of vessel wall abnormalities or develop without them.
One limitation of this technique is the lack of direct embolus detection, which could be
achieved to some extent by using transcranial Doppler ultrasound with or without
diffusion-weighted MRI (Srinivasan et al. 1996, Kimura et al. 2001).
The patterns of cerebral infarctions in the present study did not correlate with the
angiographic findings, which is in agreement with Steinke et al. (1996) and Lucas et al.
(1998). Only territorial infarctions with fragmentation were found slightly more often in
cases of occlusion (10/20) than in cases of stenosis (5/19) in the present study. However,
contrary to the results of Biousse et al. (1995) and Steinke et al. (1996), the present vessel
occlusions correlated well with the presence of infarction. Infarction was found in more
than 70% of the cases with occlusion, but in only about 40% of the cases with irregular
arterial stenosis and in about 10% of the cases with other, non-occlusive angiographic
findings. The present results are in line with a recent study, where, however, stenosis of
80% or more was not differentiated from total occlusion (Baumgartner et al. 2001).
Occlusion of the dissected vessel is most probably due not only to the intramural
hematoma but also to the intraluminal thrombosis, which may cause distal embolism and
infarctions.
Intracranial arteries have fewer elastic fibres in the media, thinner adventitia, and no
external elastic membrane compared to extracranial arteries (Wilkinson 1972, Berger &
Wilson 1984). These characteristics increase the risk of rupture through the adventitia and
55
SAH (Mokri et al. 1988a, Klufas et al. 1995, Iihara et al. 2002). Also, mucoid
degeneration of the intima in a ruptured dissecting aneurysm leading to SAH has been
reported (Hamada et al. 2001). In the present study, SAH was found in CT in 5 of the 22
patients (23%) with intracranial dissection, all the dissections being in the posterior
circulation vessels. Ruptured dissecting aneurysms have a high risk of rebleeding, and
therapeutic interventions in carefully selected cases should be performed as soon as
possible. Internal trapping of the dissected site is preferable, while proximal occlusion
does not completely eliminate the risk of rebleeding (Kitanaka et al. 1992, Iihara et al.
2002).
Hemorrhagic transformation occurs rarely in brain infarctions caused by CCAD.
Steinke et al. (1996) reported it in 5.4% of their ICAD patients, which is in agreement
with the present figure of 6.8%. The overall incidence of hemorrhagic transformation in
acute ischemic stroke has been reported to be of the order of 40% (Larrue et al. 1997).
Okada et al. (1989) studied the incidence of hemorrhagic transformation after acute
cerebral embolism and found it to be 40.6% (65/160). In both studies, advanced age, large
size of the infarction, and the presence of atrial fibrillation were associated with an
increased risk of hemorrhagic transformation. The lower prevalence of hemorrhagic
transformations associated with CCAD may be due to the smaller infarction size and the
lower age of the patients.
6.4 Long-term clinical outcome
Nearly 82% of the patients in the present study were treated with anticoagulants,
followed by acetylsalisylic acid in most cases. The treatment was acetylsalisylic acid only
in 6%, surgery or endovascular treatment in 5%, and no specific treatment in 7%. The
present study does not therefore allow any comparisons between different treatments and
outcomes. It is notable that, with the exception of four patients who died within about one
week after the onset of the symptoms due to massive cerebral infarctions, no further
patients had any progression of symptoms regardless of the treatment regimen.
Furthermore, with the exception of one peripheral bleeding complication related to
anticoagulation treatment, no other complications were encountered.
Local manifestations of CCAD, such as Horner’s syndrome and cranial nerve palsies,
seldom significantly disable patients, and the prognosis of CCAD is mainly related to the
cerebral ischemic complications, which usually occur within the first few days after the
onset of symptoms (Biousse et al. 1995). In assessing long-term clinical outcome, two
methods measuring slightly different aspects were used in the present study. One was
based on the neurological defects and symptoms, and the clinical outcome was classified
into five categories: 1) no neurological defects, no symptoms, 2) no neurological defects,
symptoms present, 3) persistent neurological defects, no or mild disability, 4) persistent
neurological defects, moderate or severe disability, and 5) death. The outcomes in the
groups 1–3 were considered good. The advantage of this classification, though it is not
universally known, is that it also takes into account symptoms and disabilities, such as
visual defects, neuropsychological defects, and problems in the cognitive domain, that
may have a major influence on the patients’ long-term outcome. The other classification
56
used in the present study was the universally known modified Rankin Scale (Wade 1992),
which assesses functional outcome. Good outcome was defined as mRS 0–2, while mRS
6 refers to death. The limitation of the present study is that it was retrospective and
outcome classification was based on information gathered from medical records. A new
clinical examination by a neurologist along with a patient interwiev would have given
additional, more accurate information on outcome. Moreover, quality of life, which is an
important patient-centered outcome measure (Chechowsky & Hill 2002), cannot be
assessed from medical records. The median follow-up time of 11 months (range 5 days to
11 years 7 months) seemed sufficient, however.
Long-term clinical outcomes assessed using the modified Rankin Scale were excellent
in the present study. Good functional recovery was shown by about 79 % of all patients.
There was a major difference in outcome between non-occlusive and occlusive
dissections (p=0.0004). In non-occlusive dissection of one or more cervicocephalic
arteries, good recovery defined as mRS 0–2 was seen in 89%, but in the case of occlusive
dissection of one or more arteries, only 61% of the patients recovered well. When
outcome was assessed using classification into categories based on the neurological
defects and symptoms, approximately 60% of all patients in the present study recovered
well. This is slightly less than in some previous reports with favorable outcomes in 70–
90 % of CCAD patients (Mokri et al. 1986, Mokri et al. 1990, Engelter et al. 2000, Saeed
et al. 2000). In one further study on 105 CCAD patients, 73% of those discharged alive
after CCAD remained fully independent, but patients with early death were thus excluded
(Leys et al. 1995). Comparison between different studies is difficult because of the
heterogeneous definitions of outcome, the small and often selected populations, and the
variable imaging reference standards and follow-up times. Furthermore, only the study of
Saeed et al. (2000) also included intracranial CCADs, but on the other hand, excluded
ICADs. The difference in long-term clinical outcome between non-occlusive and
occlusive CCADs in the present study was even more evident when a classification into
five categories was used. Only 35% of the patients with occlusive CCAD of one or more
arteries recovered well, while 75% of the patients with non-occlusive dissection of one or
more arteries recovered well, having no or only mild disability (p<0.0001). The present
findings support the reports of some other authors, in which occlusive CCAD has been
found to have worse outcome than previously suggested. Ast et al. (1993) reported poor
outcome in about half of the patients with occlusive dissection of extracranial ICAD. Half
of the patients with occlusive extracranial ICAD also had poor outcome in the study of
Bogousslavsky et al. (1987), but all patients had acute stroke, and traumatic dissections
were presumably excluded.
In the present study, both intra- and extracranial dissections of anterior and posterior
circulation arteries were included, regardless of etiology, and this gives better information
of the clinical outcome in CCAD in general. No differences between intra- and
extracranial or anterior and posterior circulation dissections were found in the present
study contrary to some other reports with poor outcomes in up to 75% of the patients with
intracranial CCAD (Hart & Easton 1983, Schievink et al. 1994b). The reports on outcome
in intracranial VAD are controversial. de Bray et al. (1997) reported poor outcomes in
44% of their patients with intracranial VAD, while other reports have shown 84–88% of
the patients to have good or excellent recovery (Mokri et al. 1988a, Hosoya et al. 1999,
57
Nakagawa et al. 2000). The patients with SAH also recovered well in the present study,
and no rebleeding was encountered.
The initial vessel finding has not been reported to correlate either with the patterns of
brain infarction or with the frequency of infarctions in CCAD patients (Biousse et al.
1995, Steinke et al. 1996, Lucas et al. 1998). In the present study, however, occlusions
correlated well with the presence of infarction but not with the stroke patterns. Therefore,
the more severe hemodynamic compromise in occlusive dissections does not seem to
explain the worse outcome. More likely, the overall higher frequency of infarctions in
occlusive dissection seems to be the main reason for the worse outcome. In multivessel
dissections, the extent of collateral circulation may be compromised. Moreover, compared
to patients with atherothrombotic infarctions, CCAD patients have been hypothesized to
have poorer functioning and mobilization of the secondary collateral leptomeningeal
pathways (Milhaud et al. 2002).
6.5 Future prospects
Radiologic imaging of vessels is shifting more and more from imaging of the vessel
lumen toward direct imaging of the pathologic vessel wall. Reports concentrate on
atherosclerosis, but new techniques might also be feasible in imaging CCAD.
Intravascular ultrasound (IVUS) has already been shown to be feasible in carotid arteries
with mild atherosclerosis, often revealing intimal thickenings and concentric plaques in
angiographically normal segments (Manninen et al. 1998). Imaging of arterial calcium is
possible using CT, whereas MRI has the potential to visualize the lipid and fibrous
components of plaque acquisition and 3-dimensional slice registration (Rumberger 2001).
Recently, a new intravascular magnetic resonance device for in vivo high-resolution
imaging (Worthley et al. 2003) and spectroscopy of atherosclerotic plaques has been
developed. One of the most promising new technologies is optical coherence tomography,
which is basically similar to IVUS, but measures back-reflected infrared light rather than
sound. It has an extremely high axial resolution, 10–15 µm in standard-resolution
imaging and up to 1–2 µm in ultrahigh-resolution imaging (Fujimoto 2003).
MRI and MRA will displace, and in many centers already have displaced, angiography
in the diagnogstic imaging of CCAD. Due to the radiation, the use of CT angiography
will remain limited. In order to reach the diagnosis of CCAD before disabling ischemic
complications develop, early imaging is mandatory. Along with the mini-invasive
imaging techniques, patients with mild symptoms should also be investigated thoroughly
with multiple methods, including imaging. In a modern hospital, neuroradiologic imaging
could be done with MRI (diffusion and perfusion) of the head and MRI and
contrast-enhanced MRA of the cervicocephalic arteries. If a double lumen or an
intramural hematoma with stenosis or aneurysmal dilatation is detected, DSA can be
omitted. If the MRI/MRA findings do not correlate with the clinical findings or are
unspecific, selective DSA should be done to improve diagnostic accuracy, especially in
VAD and intracranial dissections. Early treatment will decrease further ischemic
complications, but also improve prognosis in patients who already suffer from brain
infarction. Perfusion CT helps to detect early ischemic signs, and perfusion- and
58
diffusion-weighted MRI can be used to select candidates for aggressive stroke therapy
(Restrepo et al. 2003). Angiography will, at least for now, maintain its position in
endovascular treatments, which will, however, require more investigations to define their
indications and long-term outcome.
7 Conclusions
1. In cervicocephalic artery dissection, irregular stenosis is the most common
angiographic finding, and 81% of these stenoses normalize. Occlusion is the second
most common finding, and 33% of the occlusions recanalize. Pseudoaneurysms,
intimal flaps, double lumens, and irregular dilatations are rare findings and often
remain unchanged in follow-up.
2. Over 10% of the patients with cervicocephalic artery dissection present with pulsatile
tinnitus, which may even be the only symptom, although the majority of patients also
have concomittant head or neck pain, ischemic brain symptoms, Horner’s syndrome,
or cranial neuropathies.
3. More than half of CCAD patients have cerebral or cerebellar infarction at CT or
conventional MRI. Initial occlusion of the dissected vessel is accompanied by
infarction more often than other vessel wall abnormalities. Most cerebral infarctions
caused by arterial dissections are of embolic origin. Intracranial dissections cause
subarachnoid hemorrhage in more than 20% of patients.
4. In the case of dissection of one or more cervicocephalic arteries without occlusion,
75% of patients recover well, having no or mild disability, and when good recovery is
defined as mRS 0–2, 89% of patients recover. Contrary to that, only 35% of patients
with occlusive dissection of one or more arteries recover well, having no or mild
disability, and 61% of the patients score 0–2 on mRS. There seem to be no significant
differences in recovery after intra- and extracranial dissections.
References
Adams HP Jr, Brott TG, Furlan AJ, Gomez CR, Grotta J, Helgason CM, Kwiatkowski T, Lyden PD,
Marler JR, Torner J, Feinberg W, Mayberg M & Thies W (1996) Guidelines for thrombolytic
therapy for acute stroke: a supplement to the guidelines for the management of patients with acute
ischemic stroke. A statement for healthcare professionals from a special writing group of the
stroke council, American Heart Association. Stroke 27:1711–1718
Aksoy FG & Lev MH (2000) Dynamic contrast-enhanced brain perfusion imaging: technique and
clinical applications. Semin Ultrasound CT MRI 21:462–477
Albuquerque FC, Han PP, Spetzler RF, Zabramski JM & McDougall CG (2002) Carotid dissection:
Technical factors affecting endovascular therapy. Can J Neurol Sci 29:54–60
Arnold M, Nedeltchev K, Sturzenegger M, Schroth G, Loher TJ, Stepper F, Remonda L, Bassetti C
& Mattle HP (2002) Thrombolysis in patients with acute stroke caused by cervical artery
dissection. Analysis of 9 patients and review of the literature. Arch Neurol 59:549–553
Ast G, Woimant F, Georges B, Laurian C & Haguenau M (1993) Spontaneous dissection of the
internal carotid artery in 68 patients. Eur J Med 2:466–472
Auer A, Felber S, Schmidauer C, Waldenberger P & Aichner F (1998) Magnetic resonance
angiographic and clinical features of extracranial vertebral artery dissection. J Neurol Neurosurg
Psychiatry 64:474–481
Bakke SJ, Smith H-J, Kerty E & Dahl A (1996) Cervicocranial artery dissection. Detection by
Doppler ultrasound and MR angiography. Acta Radiol 37:529–534
Bartels E & Flügel KA (1996) Evaluation of extracranial vertebral artery dissection with duplex
color-flow imaging: Stroke 27: 290–295
Baumgartner RW, Arnold M, Baumgartner I, Mosso M, Gönner F, Studer A, Schroth G, Schuknecht
B & Sturzenegger M (2001) Carotid dissection with and without ischemic events. Local
symptoms and cerebral artery findings. Neurology 57:827–832
Bendszus M, Koltzenburg M, Burger R, Warmuth-Metz M, Hofmann E & Solymosi L (1999) Silent
embolism in diagnostic cerebral angiography and neurointerventional procedures: a prospective
study. Lancet 354:1594–1597
Berger MS & Wilson CB (1984) Intracranial dissecting aneurysms of the posterior circulation. Report
of six cases and review of the literature. J Neurosurg 61:882–894
Biousse V, D’Anglejan-Chatillon J, Massiou H & Bousser M-G (1994) Head pain in non-traumatic
carotid artery dissection: a series of 65 patients. Cephalalgia 14:33–36
Biousse V, D’Anglejan-Chatillon J, Touboul P-J, Amarenco P & Bousser M-G (1995) Time course
of symptoms in extracranial carotid artery dissections. A series of 80 patients. Stroke 26:235–239
Bogousslavsky J (1991) Topographic patterns of cerebral infarcts. Cerebrovasc Dis 1(suppl 1):61–68
61
Bogousslavsky J, Despland P-A & Regli F (1987) Spontaneous carotid dissection with acute stroke.
Arch Neurol 44:137–140
Bogousslavsky J & Regli F (1987) Ischemic stroke in adults younger than 30 years of age. Cause and
prognosis. Arch Neurol 44:479–482
Bosmans H, Wilms G, Dymarkowski S & Marchal G (2001) Basic principles of MRA. Eur J Radiol
38:2–9
Brandt T & Grond-Ginsbach C (2002) Spontaneous cervical artery dissection. From risk factors
toward pathogenesis. Stroke 33:657–658
Brandt T, Hausser I, Orberk E, Grau A, Hartschuh W, Anton-Lamprecht I & Hacke W (1998)
Ultrastructural connective tissue abnormalities in patients with spontaneous cervicocerebral artery
dissections. Ann Neurol 44:281–285
Brandt T, Orberk E, Weber R, Werner I, Busse O, Muller BT, Wigger F, Grau A, Grond-Ginsbach C
& Hausser I (2001) Pathogenesis of cervical artery dissections: association with connective tissue
abnormalities. Neurology 57:24–30
de Bray J-M, Lhoste P, Dubas F, Emile J & Saumet J-L (1994) Ultrasonic features of extracranial
carotid dissections: 47 cases studied by angiography. J Ultrasound Med 13:659–664
de Bray JM, Penisson-Besnier I, Dubas F & Emile J (1997) Extracranial and intracranial
vertebrobasilar dissections: diagnosis and prognosis. J Neurol Neurosurg Psychiatry 63:46–51
Bryan RN, Levy LM, Whitlow WD, Killian JM, Preziosi TJ & Rosario JA (1991) Diagnosis of acute
cerebral infarction: comparison of CT and MR imaging. Am J Neuroradiol 12:611–620
Chiras J, Marciano S, Molina JV, Touboul J, Poirier B & Bories J (1985) Spontaneous dissecting
aneurysm of the extracranial vertebral artery (20 cases). Neuroradiology 27:327–333
Crum B, Mokri B & Fulgham J (2000) Spinal manifestations of vertebral artery dissection. Neurology
55:304–306
Czechowsky D & Hill MD (2002) Neurological outcome and quality of life after stroke due to
vertebral artery dissection. Cerebrovasc Dis 13:192–197
D’Anglejan-Chatillon J, Ribeiro V, Mas JL, Youl BD & Bousser MG (1989) Migraine – A risk factor
for dissection of cervical arteries. Headache 29:560–561
Derex L, Nighoghossian N, Turjman F, Hermier M, Honnorat J, Neuschwander P, Froment JC &
Trouillas P (2000) Intravenous tPA in acute ischemic stroke related to internal carotid artery
dissection. Neurology 54:2159–2161
Dippel DWJ, Du Ry van Beest Holle M, van Kooten F & Koudstaal PJ (2000) The validity and
reliability of signs of early infarction on CT in acute ischaemic stroke. Neuroradiology 42:629–
633
Eljamel MSM, Humphrey PRD & Shaw MDM (1990) Dissection of the cervical internal carotid
artery. The role of Doppler / Duplex studies and conservative management. J Neurol Neurosurg
Psychiatry 53:379–383
Engelter ST, Lyrer PA, Kirsch EC & Steck AJ (2000) Long-term follow-up after extracranial internal
carotid artery dissection. Eur Neurol 44:199–204
Fisher CM (1991) Lacunar infarcts: A review. Cerebrovasc Dis 1:311–320
Foley WD & Karcaaltincaba M (2003) Computed tomography angiography: Principles and clinical
applications. J Comput Assist Tomogr 27 (Suppl.1):S23–S30
Fujimoto JG (2003) Optical coherence tomography for ultrahigh resolution in vivo imaging. Nat
Biotechnol 21:1361–1367
Gallai V, Caso V, Paciaroni M, Cardaioli G, Arning E, Bottiglieri T & Parnetti L (2001) Mild
hyperhomocyst(e)inemia. A possible risk factor for cervical artery dissection. Stroke 32:714–718
Ghika J, Bogousslavsky J & Regli F (1989) Infarcts in the territory of the deep perforators from the
carotid system. Neurology 39:507–512
Gobert M, Mounier-Vehier F, Lucas Ch, Leclerc X & Leys D (1996) Cranial nerve palsies due to
internal carotid artery dissection: seven cases. Acta Neurol Belg 96:55–61
62
Grau AJ, Brandt T, Buggle F, Orberk E, Mytilineos J, Werle E, Conradt, Krause M, Winter R &
Hacke W (1999) Association of cervical artery dissection with recent infection. Arch Neurol
56:851–856
Guillon B, Berthet K, Benslamia L, Bertrand M, Bousser M-G & Tzourio C (2003) Infection and the
risk of spontaneous cervical artery dissection: a case-control study. Stroke 34:e79–e81
Guillon B, Brunereau L, Biousse V, Djouhri H, Lévy C & Bousser M-G (1999) Long-term follow-up
of aneurysms developed during extracranial internal carotid artery dissection. Neurology 53:117–
122
Halbach VV, Higashida RT, Dowd CF, Fraser KW, Smith TP, Teitelbaum GP, Wilson CB &
Hieshima GB (1993) Endovascular treatment of vertebral artery dissections and
pseudoaneurysms. J Neurosurg 79:183–191
HamadaY, Mannoji H & Kaneko Y (2001) A ruptured dissecting aneurysm of the vertebral artery:
comparison of angiographic and histological findings. Neuroradiology 43:375–378
Hamberg LM, Hunter GJ, Kierstead D, Lo EH, Gonzáles RG & Wolf GL (1996) Measurement of
cerebral blood volume with subtraction three-dimensional functional CT. Am J Neuroradiol
17:1861–1869
Hancock JH & Millar JS (2000) Spontaneous dissection of the anterior inferior cerebellar artery.
Neuroradiology 42:535–538
Hart RG & Easton JD (1983) Dissections of cervical and cerebral arteries. Neurol Clin 1:155–182
Heiserman JE, Dean BL, Hodak JA, Flom RA, Bird CR, Drayer BP & Fram EK (1994) Neurologic
complications of cerebral angiography. Am J Neuroradiol 15:1401–1407
Heiserman JE, Drayer BP, Fram EK & Keller PJ (1992) MR angiography of cervical fibromuscular
dysplasia. Am J Neuroradiol 13:1454–1457
Heiss WD (2000) Ischemic penumbra: evidence from functional imaging in man. J Cereb Blood Flow
Mateb 20:1276–1293
Hoffman M, Sacco RL, Chan S & Mohr JP (1993) Noninvasive detection of vertebral artery
dissection. Stroke 24:815–819
Holman BL & Devous MD Sr (1992) Functional brain SPECT: the emergence of a powerful clinical
method. J Nucl Med 33:1888–1904
Hosoya T, Adachi M, Yamaguchi K, Haku T, Kayama T & Kato T (1999) Clinical and
neuroradiological features of intracranial vertebrobasilar artery dissection. Stroke 30:1083–1090
Houser OW, Mokri B, Sundt TM Jr, Baker HL Jr & Reese DF (1984) Spontaneous cervical cephalic
arterial dissection and its residuum: Angiographic spectrum. Am J Neuroradiol 5:27–34
Hunter GJ, Hamberg LM, Ponzo JA, Huang-Hellinger FR, Morris PP, Rabinov J, Farkas J, Lev MH,
Schaefer PW, Ogilvy CS, Schwamm L, Buonanno FS, Koroshetz WJ, Wolf GL & González RG
(1998) Assessment of cerebral perfusion and arterial anatomy in hyperacute stroke with
three-dimensional functional CT: early clinical results. Am J Neuroradiol 19:29–37
Iihara K, Sakai N, Murao K, Sakai H, Higashi T, Kogure S, Takahashi JC & Nagata I (2002)
Dissecting aneurysms of the vertebral artery: a management strategy. J Neurosurg 97:259–267
Jentzer A (1954) Dissecting aneurysm of the left internal carotid artery. Angiology 5:232–234
Jungreis CA, Yonas H, Firlik AD & Wechsler LR (1999) Advanced CT imaging (functional CT).
Neuroimaging Clin N Am 9(3): 455–464
Kasner SE, Hankins LL, Bratina P & Morgenstern LB (1997) Magnetic resonance angiography
demonstrates vascular healing of carotid and vertebral artery dissections. Stroke 28:1993–1997
Kato K, Tomura N, Takahashi S, Sakuma I & Watarai J (2003) Ischemic lesions related to cerebral
angiography: evaluation by diffusion weighted MR imaging.
Neuroradiology 45:39–43.
Katzen BT (1995) Current status of digital angiography in vascular imaging. Radiol Clin North Am
33:1–14
63
Kimura K, Minematsu K, Koga M, Arakawa R, Yasaka M, Yamagami H, Nagatsuka K, Naritomi H
& Yamaguchi T (2001) Microembolic signals and diffusion-weighted MR imaging abnormalities
in acute ischemic stroke. Am J Neuroradiol 22:1037–1042
Kitanaka C, Morimoto T, Sasaki T & Takakura K (1992) Rebleeding from vertebral artery dissection
after proximal clipping. Case report. J Neurosurg 77:466–468
Kitanaka C, Tanaki J-I, Kuwahara M, Teraoka A, Sasaki T & Takakura K (1994) Nonsurgical
treatment of unruptured intracranial vertebral artery dissection with serial follow-up angiography.
J Neurosurg 80: 667–674
Klossek J-M, Neau JP, Vandenmark P & Fontanel JP (1994) Unilateral lower cranial nerve palsies
due to spontaneous internal carotid artery dissection. Ann Otol Rhinol Laryngol 103:413–415
Klufas RA, Hsu L, Barnes PD, Patel MR & Schwartz RB (1995) Dissection of the carotid and
vertebral arteries: Imaging with MR angiography. AJR 164:673–677
Koch S, Romano JG, Bustillo IC, Concha M & Forteza AM (2001) Anticoagulation and
microembolus detection in a case of internal carotid artery dissection. J Neuroimaging 11:63–66
Kremer C, Mosso M, Georgiadis D, Stöckli E, Benninger D, Arnold M & Baumgartner RW (2003)
Carotid dissection with permanent and transient occlusion or severe stenosis. Long-term outcome.
Neurology 60:271–275
Kurokawa Y, Yonemasu Y, Kano H, Sasaki T & Inaba K (2000) The usefulness of 3D-CT
angiography for the diagnosis of spontaneous vertebral artery dissection – report of two cases.
Comput Med Imag & Graph 24:115–119
Lal V, Sawhney IMS, Bansal SK & Prabhakar S (1994) Abducens nerve palsy due to spontaneous
internal carotid artery dissection – A case report. Comput Med Imaging Graph 18:381–383
Landwehr P, Schulte O & Voshage G (2001) Ultrasound examination of carotid and vertebral
arteries. Eur Radiol 11:1521–1534
Larrue V, von Kummer R, del Zoppo G & Bluhmki E (1997) Hemorrhagic transformation in acute
ischemic stroke. Potential contributing factors in the European cooperative acute stroke study.
Stroke 28:957–960
Lazinski D, Willinsky RA, TerBrugge K & Montanera W (2000) Dissecting aneurysms of the
posterior cerebral artery: angioarchitecture and a review of the literature. Neuroradiology 42:128–
133
Leclerc X, Godefroy O, Salhi A, Lucas C, Leys D & Pruvo JP (1996) Helical CT for the diagnosis of
extracranial internal carotid artery dissection. Stroke 27:461–466
Leclerc X, Lucas C, Godefroy O, Nicol L, Morretti A, Leys D & Pruvo JP (1999) Preliminary
experience using contrast-enhanced MR angiography to assess vertebral artery structure for the
follow-up of suspected dissection. Am J Neuroradiol 20:1482–1490
Leclerc X, Lucas C, Godefroy O, Tessa H, Martinat P, Leys D & Pruvo J-P (1998) Helical CT for the
follow-up of cervical internal carotid artery dissections. Am J Neuroradiol 19:831–837
Leibowitz R, Do HM, Marcellus ML, Chang SD, Steinberg GK & Marks MP (2003) Parent vessel
occlusion for vertebrobasilar fusiform and dissecting aneurysms. Am J Neuroradiol 24:902–907
Lepojärvi M, Tarkka M, Leinonen A & Kallanranta T (1988) Spontaneous dissection of the internal
carotid artery. Acta Chir Scand 154:559–566
Levy C, Laissy JP, Raveau V, Amarenco P, Servois V, Bousser MG & Tubiana JM (1994) Carotid
and vertebral artery dissections: Three-dimensional time-of-flight MR angiography and MR
imaging versus conventional angiography. Radiology 190:97–103
Leys D, Moulin Th, Stojkovic T, Begey S, Chavot D & DONALD investigators (1995) Follow-up of
patients with history of cervical artery dissection. Cerebrovasc Dis 5:43–49
Link J, Steffens JC, Muller-Hulsbeck S, Brossmann J & Heller M (1996) MR angiography in
fibromuscular dysplasia of the cervical arteries. Rofo Fortschr Geb Roentgenstr Neuen Bildgeb
Verfahr 164:201–205
64
Liu YJ, Chen CY, Chung HW, Huang IJ, Lee CS, Chin SC & Liou M (2003) Neuronal damage after
ischemic injury in the middle cerebral arterial territory: deep watershed versus territorial
infarction at MR perfusion and spectroscopic imaging. Radiology 229(2):366–374
Liu Y, Karonen JO, Vanninen RL, Nuutinen J, Koskela A, Soimakallio S & Aronen HJ Predictive
value of 2D-phase contrast MRA in acute ischemic stroke: a serial study with combined diffusion
and perfusion MRI. Radiology, in press
Liu AY, Paulsen RD, Marcellus ML, Steinberg GK & Marks MP (1999) Long-term outcomes after
carotid stent placement for treatment of carotid artery dissection. Neurosurgery 45:1368–1374
Lucas C, Moulin T, Deplanque D, Tatu L, Chavot D & the DONALD investigators (1998) Stroke
patterns of internal carotid artery dissection in 40 Patients. Stroke 29:2646–2648
Luken MG, Ascherl GF Jr, Correll JW & Hilal SK (1979) Spontaneous dissecting aneurysms of the
extracranial internal carotid artery. Clin Neurosurg 26:353–375
Luypaert R, Boujraf S, Sourbron S & Osteaux M (2001) Diffusion and perfusion MRI: basic physics.
Eur J Radiol 38:19–27
Malek AM, Higashida RT, Phatouros CC, Lempert TE, Meyers PM, Smith WS, Dowd CF & Halbach
VV (2000) Endovascular management of extracranial carotid artery dissection achieved using
stent angioplasty. Am J Neuroradiol 21:1280–1292
Manninen HI, Koivisto T, Saari T, Matsi PJ, Vanninen RL, Luukkonen M & Hernesniemi J (1997)
Dissecting aneurysms of all four cervicocranial arteries in fibromuscular dysplasia: treatment with
self-expanding endovascular stents, coil embolization, and surgical ligation. Am J Neuroradiol
18:1216–1220
Manninen HI, Rasanen H, Vanninen RL, Berg M, Hippelainen M, Saari T, Yang X, Karkola K &
Kosma VM (1998) Human carotid arteries: correlation of intravascular US with angiographic and
histopathologic findings. Radiology 206:65–74
Mascalchi M, Bianchi MC, Mangiafico S, Ferrito G, Puglioli M, Marin E, Mugnai S, Canapicchi R,
Quilici N & Inzitari D (1997) MRI and MR angiography of vertebral artery dissection.
Neuroradiology 39:329–340
Milhaud D, de Freitas GR, van Melle G & Bogousslavsky J (2002) Occlusion due to carotid artery
dissection. A more severe disease than previously suggested. Arch Neurol 59:557–561
Mokri B (1990) Traumatic and spontaneous extracranial internal carotid artery dissections. J Neurol
237:356–361
Mokri B, Houser OW, Sandok BA & Piepgras DG (1988a) Spontaneous dissections of the vertebral
arteries. Neurology 38:880–885
Mokri B, Piepgras DG & Houser OW (1988b) Traumatic dissections of the extracranial internal
carotid artery. J Neurosurg 68:189–197
Mokri B, Silbert PL, Schievink WI & Piepgras DG (1996) Cranial nerve palsy in spontaneous
dissection of the extracranial internal carotid artery. Neurology 46:356–359
Mokri B, Sundt TM Jr, Houser OW & Piepgras DG (1986) Spontaneous dissection of the cervical
internal carotid artery. Ann Neurol 19:126–138
Moulin T, Cattin F, Crépin-Leblond T, Tatu L, Chavot D, Piotin M, Viel JF, Rumbach L &
Bonneville JF (1996) Early CT signs in acute middle cerebral artery infarction: Predictive value
for subsequent infarct locations and outcome. Neurology 47:366–375
Mullins ME, Schaefer PW, Sorensen AG, Halpern EF, Ay H, He J, Koroshetz WJ & Gonzalez RG
(2002) CT and conventional and diffusion-weighted MR imaging in acute stroke: Study in 691
patients at presentation to the emergency department. Radiology 224:353–360
Nakagawa K, Touho H, Morisako T, Osaka Y, Tatsuzava K, Nakae H, Owada K, Matsuda K &
Karasawa J (2000) Long-term follow-up study of unruptured vertebral artery dissection: clinical
outcomes and serial angiographic findings. J Neurosurg 93:19–25
Norrving B, Nilsson B & Cronqvist S (1986) Dissection of cerebral arteries – a major cause of stroke
in the young. Acta Neurol Scand 73:517
65
O’Connell BK, Towfighi J, Brennan RW, Tyler W, Mathews M, Weidner WA & Saul RF (1985)
Dissecting aneurysms of head and neck. Neurology 35:993–997
Ohkuma H, Suzuki S, Kikkawa T & Shimamura N (2003a) Neuroradiologic and clinical features of
arterial dissection of the anterior cerebral artery. Am J Neuroradiol 24:691–699
Ohkuma H, Suzuki S, Shimamura N & Nakano T (2003b) Dissecting aneurysms of the middle
cerebral artery: neuroradiological and clinical features. Neuroradiology 45:143–148
Okada Y, Yamaguchi T, Minematsu K, Miyashita T, Sawada T, Sadoshima S, Fujishima M & Omae
T (1989) Hemorrhagic transformation in cerebral embolism. Stroke 20:598–603
Okamoto T, Miyachi S, Negoro M, Otsuka G, Suzuki O, Keino H & Yoshida J (2002) Experimental
model of dissecting aneurysms. Am J Neuroradiol 23:577–584
Ozodoba C, Sturzenegger M & Schroth G (1996) Internal carotid artery dissection: MR imaging
features and clinical-radiologic correlation. Radiology 199:191–198
Petro GR, Witwer GA, Cacayorin ED, Hodge CJ, Bredenberg CE, Jastremski MS & Kieffer SA
(1986) Spontaneous dissection of the cervical internal carotid artery: Correlation of arteriography,
CT, and pathology. Am J Neuroradiol 7:1053–1058
Pozzati E, Andreoli A, Limoni P & Casmiro M (1994) Dissecting aneurysms of the vertebrobasilar
system: study of 16 cases. Surg Neurol 41:119–124
Pozzati E, Giuliani G, Acciarri N & Nuzzo G (1990) Long-term follow-up of occlusive cervical
carotid dissection. Stroke 21:528–531
Provenzale JM, Barboriak DP, Taveras JM (1995) Exercise-related dissection of craniocervical
arteries: CT, MR, and angiographic findings. J Comput Assist Tomogr 19:268–276
Provenzale JM, Morgenlander JC & Gress D (1996) Spontaneous vertebral dissection: Clinical,
conventional, angiographic, CT and MR findings. J Comput Assist Tomogr 20:185–193
Rawat B, Gerroti M, Ashby P & Simons M (1992) Application of duplex Doppler sonography in
intracranial carotid dissection. J Ultrasound Med 11:121–123
Restrepo L, Pradilla G, Llinas R & Beauchamp NJ (2003) Perfusion- and diffusion-weighted MR
imaging–guided therapy of vertebral artery dissection: Intraarterial thrombolysis through an
occipital vertebral anastomosis. Am J Neuroradiol 24:1823–1826
Rissanen A (1992) Cerebrovascular disease in the Jyväskylä region, Central Finland. Kuopio:
University of Kuopio
Rumberger JA (2001) Tomographic (plaque) imaging: state of the art. Am J Cardiol 88(suppl):66E–
69E
Saeed AB, Shuaib A, Al-Sulaiti G & Emery D (2000) Vertebral artery dissection: Warning
symptoms, clinical features and prognosis in 26 patients. Can J Neurol Sci 27:292–296
Saito R, Ezura M, Takahashi A & Yoshimoto T (2000) Combined neuroendovascular stenting and
coil embolization for cervical carotid artery dissection causing symptomatic mass effect. Surg
Neurol 53:318–322
Saunders DE (2000) MR spectroscopy in stroke. Br Med Bull 56:334–345
Saver JL, Easton JD & Hart RG (1992) Dissections and trauma of cervicocerebral arteries. In: Barnett
HJM, Mohr JP, Stein BM & Yatsu FM, eds. Stroke: Pathophysiology, diagnosis and management.
Churchill Livingstone, New York, 671–688
Saver JL & Easton JD (1998) Dissections and trauma of cervicocerebral arteries. In: Barnett HJM,
Mohr JP, Stein BM & Yatsu FM, eds. Stroke: Pathophysiology, diagnosis and management.
Churchill Livingstone, New York, 769–786
Schellhas KP, Latchaw RE, Wendling LR & Gold LHA (1980) Vertebrobasilar Injuries Following
Cervical Manipulation. JAMA 244:1450–1453
Schievink WI (2000) The treatment of spontaneous carotid and vetebral artery dissections. Curr Opin
Cardiol 15:316–321
Schievink WI, Mokri B & O’Fallon WM (1994a) Recurrent spontaneous cervical-artery dissection.
N Engl J Med 330:393–397
66
Schievink WI, Mokri B & Piepgras DG (1994b) Spontaneous dissections of cervicocephalic arteries
in childhood and adolescence. Neurology 44:1607–1612
Schievink WI, Mokri B, Piepgras DG & Kuiper JD (1996) Recurrent spontaneous arterial dissections.
Risk in familial versus nonfamilial disease. Stroke 27:622–624
Schievink WI, Mokri B & Whisnant JP (1993) Internal carotid artery dissection in a community.
Rochester, Minnesota, 1987–1992. Stroke 24:1678–1680.
Schievink WI, Wijdics EFM & Kuiper JD (1998) Seasonal pattern of spontaneous cervical artery
dissection. J Neurosurg 89:101–103
Schulze H-E, Ebner A & Besinger UA (1992) Report of dissection of the internal carotid artery in
three cases. Neurosurg Rev 15:61–64
Shah SB, Lalwani AK & Dowd CF (1999) Transverse / sigmoid sinus dural arteriovenous fistulas
presenting as pulsatile tinnitus. Laryngoscope 109:54–58
Sherman DG, Hart RG & Easton JD (1981) Abrupt change in head position and cerebral infarction.
Stroke 12:2–6
Sheth TN, Winslow JL & Mikulis DJ (2001) Rotational changes in the morphology of the vertebral
artery at a common site of artery dissection. Can Assoc Radiol J 52(4):236–241
Silbert PL, Mokri B & Schievink WI (1995) Headache and neck pain in spontaneous internal carotid
and vertebral artery dissections. Neurology 45:1517–1522
Simionato F, Righi C & Scotti G (1999) Post-traumatic dissecting aneurysm of extracranial internal
carotid artery: endovascular treatment with stenting. Neuroradiology 41:543–547
Sivenius J (1982). Studies on the rehabilitation, epidemiology and clinical features of stroke in East
Central Finland. Kuopio: University of Kuopio
Soper JR, Parker GD & Hallinan JM (1995) Vertebral artery dissection diagnosed with CT. Am J
Neuroradiol 16:952–954
Srinivasan J, Newell DW, Sturzenegger M, Mayberg MR & Winn HR (1996) Transcranial Doppler
in the evaluation of internal carotid artery dissection. Stroke 27:1226–1230
Statistics Finland (2002) Causes of death in 2000. In: Statistical yearbook of Finland 2002. Keuruu,
Otava
Steinke W, Rautenberg W, Schwartz A & Hennerici M (1994) Noninvasive monitoring of internal
carotid artery dissection. Stroke 25:998–1005
Steinke W, Schwartz A & Hennerici M (1996) Topography of cerebral infarction associated with
carotid artery dissection. J Neurol 243:323–328
Stringer WL & Kelly DL Jr (1980) Traumatic dissection of the extracranial internal carotid artery.
Neurosurgery 6:123–130
Sturzenegger M (1994) Headache and neck pain: The warning symptoms of vertebral artery
dissections. Headache 34:187–193
Sturzenegger M & Huber P (1993) Cranial nerve palsies in spontaneous carotid artery dissection. J
Neurol Neurosurg Psychiatry 56:1191–1199
Szabo K, Kern R, Gass A, Hirsch J & Hennerici M (2001) Acute stroke patterns in patients with
internal carotid artery disease. A diffusion-weighted magnetic resonance imaging study. Stroke
32:1323–1329
Tikkakoski T, Leinonen S, Siniluoto T & Koivukangas J (1997) Isolated dissecting aneurysm of the
left posterior inferior cerebellar artery: endovascular treatment with Guglielmi detachable coil.
Am J Neuroradiol 18:936–938
Tomura N, Uemura K, Inugami A, Fujita H, Higano S & Shishido F (1988) Early CT finding in
cerebral infarction: obscuration of the lentiform nucleus. Radiology 168:463–467
Tomandl BF, Klotz E, Handschu R, Stemper B, Reinhardt F, Huk WJ, Eberhardt KE &
Fateh-Moghadam S (2003) Comprehensive imaging of ischemic stroke with multisection CT.
RadioGraphics 23:565–592
67
Touzé E, Randoux B, Méary E, Arquizan C, Meder J-F & Mas J-L (2001) Aneurysmal forms of
cervical artery dissection. Associated factors and outcome. Stroke 32:418–423
Truwit CL, Barkovich AJ, Gean-Marton A, Hibri N & Norman D (1990) Loss of the insular ribbon:
another early CT sign of acute middle cerebral artery infarction. Radiology 176:801–806
Uflacker R (1997) Atlas of vascular anatomy. An angiographic approach. Williams and Wilkins,
Baltimore, pp 7–8, 17–18
Urbach H, Flacke S, Keller E, Textor J, Berlis A, Hartmann A, Reul J, Solymosi L & Schild HH
(2000) Detectability and detection rate of acute cerebral hemisphere infarcts on CT and
diffusion-weighted MRI. Neuroradiology 42:722–727
Vanninen RL, Manninen HI, Rinne J (2003) Intrasellar iatrogenic carotid preudoaneurysm:
endovascular treatment with a polytetrafluoroethylene-covered stent. Cardiovasc Intervent Radiol
26:298-301
Wade DT (1992) Measurement in neurological rehabilitation. Curr Opin Neurol Neurosurg 5:682–
686
Warnock NG, Gandhi MR, Bergvall U & Powell T (1993) Complications of intraarterial digital
subtraction angiography in patients investigated for cerebral vascular disease. Br J Radiol 66:855–
858
Weiller C, Müllges W, Ringelstein EB, Buell U & Reich W (1991) Patterns of brain infarctions in
internal carotid artery dissections. Neurosurg Rev 14:111–113
Wilkinson IMS (1972) The vertebral artery: extracranial and intracranial structure. Arch Neurol
27:392–396
Williams PL, Warwick R, Dyson M & Bannister LF (eds) (1989) Gray’s anatomy. Thirty-seventh
edition. Churchill Livingstone, London, pp 684–686, 743–745, 750–751
Wilms G, Bosmans H, Demaerel Ph & Marchal G (2001) Magnetic resonance angiography of the
intracranial vessels. Eur J Radiol 38:10–18
Wityk RJ & Beauchamp NJ Jr (2000) Diagnostic evaluation of stroke. Neurol Clin 18:357–378
Worthley SG, Helft G, Fuster V, Fayad ZA, Shinnar M, Minkoff LA, Schechter, Fallon JT &
Badimon JJ (2003) A novel nonobstructive intravascular MRI coil: in vivo imaging of
experimental atherosclerosis. Arterioscler Thromb Vasc Biol 23:346–350
Zelenock GB, Kazmers A, Whitehouse WM Jr, Graham LM, Erlandson EE, Cronenwett JL,
Lindenauer SM & Stanley JC (1982) Extracranial internal carotid artery dissections.
Noniatrogenic traumatic lesions. Arch Surg 117:425–432