Download Cerebral Embolization: Prospective Clinical Analysis of 42 Cases

Cerebral Embolization: Prospective
Clinical Analysis of 42 Cases
BY JOHN STIRLING MEYER, M.D., JONATHAN Z. CHARNEY, M.D.,
VICTOR M. RIVERA, M.D., AND NINAN T. MATHEW, M.D.
Abstract:
Cerebral
Embolization:
Prospective
Clinical
Analysis of
42 Cases
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
• The sites, sources, and incidence of cerebral embolization were established in
an unselected series of hospitalized patients using recommended epidemiological
criteria. Criteria for diagnosis of cerebral embolization as a cause of stroke are
defined.
Forty-two (15%) from a total of 280 patients with various forms of
cerebrovascular disease were diagnosed as having cerebral embolization. The
epidemiology has been altered by antibiotics and cardiac surgery, particularly
the use of prosthetic heart valves. In this series, the mortality rate was 9.5%,
which is lower than in previously reported series probably due to emergency
medical treatment. The heart was the most common source. Prosthetic heart
valves emerged as the most common source of emboli in this scries. Ulcerated
plaques of the carotid arteries in the neck ranked second. Seizures occurred in
42.8% of cases. The left cerebral hemisphere was involved approximately twice
as frequently as the right one.
Possible explanations based on the anatomy of the great vessels are described
which predispose to embolization of the left carotid artery. Recommendations
are suggested for preventing further embolization and for treatment of acute
embolic cerebral infarction by the use of anticonvulsant drugs and
hyperosmolar agents designed to reduce cerebral edema.
cerebral emboli
ADDITIONAL KEY WORDS
papaverine
diphenylhydantoin
glycerol
• Cerebral embolization has long been
recognized as a common cause of death and
disability; yet, surprisingly little has been
written concerning its prevalence, diagnosis,
and treatment because until recently clinical
proof of its occurrence was often lacking.
The potential epidemiological and therapeutic usefulness of case-fatality ratios in
hospitalized populations, provided rigid criteria
for case selection are used, has recently been
reviewed.1 Development of arteriographical
From the Department of Neurology, Baylor College
of Medicine, and the Baylor-Methodist Center for
Cerebrovascular Research, Houston, Texas.
Supported by National Institute for Neurological
Diseases and Stroke Grant No. NS 09287-01.
Reprint requests to: Dr. lohn Stirling Meyer,
Department of Neurology, Baylor College of Medicine, 1200 Moursund, Houston, Texas, 77025.
Sfroke, Vol. 2, Novomber-D»cembor 1971
cerebral edema
seizures
prosthetic valve replacement
techniques for visualizing obstruction of the
cerebral arteries, and advances in surgical
treatment such as carotid endarterectomy,
prosthetic valve replacement, electronic pacemakers, and coronary bypass procedures,
together with new methods of monitoring the
electrocardiogram, have made it possible to
distinguish cerebral embolism from other forms
of cerebrovascular disease with considerable
precision.
For these reasons, few planned prospective clinical studies of the relative frequency,
symptomatology, pathogenesis, and treatment
of embolism in a large series of hospitalized
patients suffering from stroke are cited in the
literature. Available reports are almost invariably concerned with retrospective analysis of
series of cases diagnosed at necropsy as
embolism with retrospective inspection of
541
MEYER, CHARNEY, RIVERA, MATHEW
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
hospital records.2^* The single exception is the
prospective study of a population group living
in Framingham, Massachusetts.5
The Framingham study consisted of a
population group of 5,106 adults without
cerebrovascular symptoms followed for 12
years who were initially free of coronary heart
disease and cerebrovascular disease. During
this period, 11 of the 90 of this population
later had strokes presumed to be due to
cerebral embolism. In the Framingham study,
the diagnosis of embolism was based, in the
majority of cases, on clinical data without
detailed ancillary diagnostic tests. The diagnosis of embolism in these cases was usually
based on retrospective analysis of available
clinical data provided by the patient's physician. Arteriograms were seldom carried out.1
Fisher et al.,8 in 17 of 57 brains studied at
autopsy, reported infarction presumed to be
due to cerebral embolism. The thought should
be borne in mind that the diagnosis of cerebral
embolism is often difficult to make on the basis
of arteriographical findings during life since the
embolus often becomes lysed or fragmented;
hence, it is not surprising that these investigators failed to find the suspected embolism in
the majority of cases. To overcome this
difficulty, the majority of their diagnoses of
cerebral embolism were based primarily on the
hemorrhagic nature of the infarction.
Castaigne et al.,4 in a similar autopsy
series, used one or more of the following
criteria for diagnosing cerebral embolism: (1)
the lack of disease in the arterial wall at the
site of occlusion, (2) identification of the
probable source of the embolus, and (3) the
presence of hemorrhagic infarction of the
brain.
Fisher et al.,8 in their necropsy study,
employed one or more similar but not identical
criteria: (1) the presence of embolic material
in the occluded vessel, (2) atrial fibrillation
prior to death, (3) acute myocardial infarction
with mural thrombosis, (4) vegetative endocarditis, (5) recent surgical procedures on the
heart or the great vessels supplying the brain,
and (6) the presence of hemorrhagic cerebral
infarction. As stated, their diagnosis of
cerebral embolism was often based on an
assumption, namely, criterion 6. Investigators
who have made retrograde analyses from
postmortem data agree that hemorrhagic
infarction may be caused by factors other than
542
embolism. For this reason, in addition to the
fact that all their patients expired, their
autopsy statistics have limited epidemiological
or therapeutic value.
The present series of cases were analyzed
prospectively using rigid clinical criteria.1 The
majority of these patients survived and the data
to be reported may be of interest from a case
mortality and epidemiological point of view.
For example, the information obtained from
the study of these patients indicates that certain
types of recurrent embolism are characterized
by ischemia in the distribution of small
cerebral vessels, resulting in transient or
minimal neurological deficits which do not
cause death or permanent disability, particularly if adequate treatment is instituted.
Case Material and Experimental
Design
Prospective clinical analyses were carried out in
42 patients in whom the clinical diagnosis of
cerebral embolism appeared to be well established
(table 1). These patients were identified from a
group of 280 with cerebrovascular disease
examined over a period of 14 months by at least
two of the authors in a large cerebrovascular
research center. The diagnosis of cerebral embolism was based on the following clinical criteria:
1. In every case a probable or certain source
of cerebral embolization existed including:
(a) atherosclerotic or rheumatic heart disease
with atrial fibrillation confirmed by cardiological
consultation and electrocardiograms, supplemented as needed by coronary arteriograms and
cardiac catheterization if the diagnosis was in
doubt, (b) prosthetic valve replacement, (c) insertion of pacemakers for heart block due to
myocardial infarction associated with mural
thrombosis of the left side of the heart,
(d) ulcerated atheromatous plaques or stenosis of
the internal carotid artery proved arteriographically and confirmed by biopsy and microscopical
examination of the surgical specimen, (e) vertebral stenosis with ulcerated atheromatous plaques
demonstrated by arteriography only, and (f) aneurysms of the aortic arch with mural thrombi
demonstrated angiographically.
2. In every case there was sudden onset of a
focal neurological deficit, often preceded or
accompanied by a seizure or seizures.
3. Demonstration of embolic occlusion of
the suspected intracranial vessel by arteriography
(26 cases).
4. Localized EEG abnormality consisting of
slow waves or spike activity or both in the
appropriate cerebral region supplied by the vessel
Stroke, Vol. 2, November-Dtctmber 1971
CEREBRAL EMBOLIZATION
TABLE 1
PoHant
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
Ag.
Sax
1
2
3
4
5
6
7
8
9
71
31
65
59
71
55
65
56
75
F
M
M
F
F
F
F
M
F
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
56
65
63
79
72
66
64
60
76
57
54
45
74
59
60
66
F
M
M
M
F
M
M
M
M
M
M
M
M
M
M
M
26
47
M
27
86
M
28
29
18
67
F
M
30
31
32
33
34
64
48
65
23
23
M
M
M
F
M
35
61
M
36
37
38
39
40
57
81
76
63
67
M
M
M
M
F
41
42
46
67
M
M
Sourc* of •mboll
L carotid plaque
Cardiac dysrhyttimia
L carotid plaque
L carotid plaque
Atrial fib. RHD
R carotid stenosis
Ml mural thrombosis
R carotid plaque (ULC)
Ml heart block,
pacemaker failure
L carotid plaque
Plaque
RHD, pros, valve (A)
Ml atrial fib.
Ml heart block, pacemaker
L carotid plaque
Pros, valve (AJ
L carotid plaque (ULC)
Ml, ASHD, fib.
R carotid plaque
Pros, valve (M)
Carotid plaque, kink
L carotid plaque
Pros, valve (A)
Pros, valve (M)
Pros, valve (M)
Pros, valve (A)
Pros, valve (M)
Pros, valve (A)
Bilateral carotid
plaques (ULC)
Pros, valve (M)
Mural thrombosis
A arch aneurysm
Ml, dysrhythmia
L carotid plaque
R carotid plaque
Pros, valve (M)
Pros, valve (A)
Pros, valve (M)
Mural thrombosis
A arch aneurysm
L carotid plaque (ULC)
R carotid plaque
R carotid plaque
RHD, mitral stenosis
Vert, artery occlusion
with embolization
Pros, valve (A)
L carotid plaque (ULC)
Location of vmboD
L hemisphere
L hemisphere
L hemisphere
L hemisphere
L hemisphere
R hemisphere
R hemisphere
R hemisphere
L hemisphere
Stuporous
Obtunded
Alert
Lethargic
Lethargic
Alert
Stuporous
Alert
Lethargic
L hemisphere
L hemisphere
L hemisphere
L hemisphere
L hemisphere
L hemisphere
L hemisphere
L hemisphere
L hemisphere
R hemisphere
R hemisphere
R hemisphere
L hemisphere
L hemisphere
L hemisphere
R hemisphere
Alert
Lethargic
Alert
Alert, aphasic
Lethargic
Confused
Dysphasic
Alert, dysphasic
Lethargic
Lethargic
Lethargic
Dysphasic
Dysphasic, alert
Postictal
Lethargic
Obtunded
R hemisphere
Alert
R and L
hemispheres
L hemisphere
L hemisphere
Alert
R hemisphere
L hemisphere
R hemisphere
L hemisphere
R and L
hemispheres
L hemisphere
Lethargic
Obtunded
Alert
Alert
Comatose
L hemisphere
R hemisphere
R hemisphere
L hemisphere
R hemisphere
Alert
Comatose
Alert
Alert, dysphasic
Drowsy
L hemisphere
L hemisphere
Alert
Asphasic
Dysphasic
Lethargic
Coma
M = male; F = female; L = left; R = right; (A) = aorta; (M) = mitral; Vert. = vertebral; RHD = rheumatic heart disease; (ULC) = ulcerated plaque found at surgery; MI := myocardial infarction; Pros. = prosthetic; ASHD = arteriosclerotic heart disease.
Strok;
Vol. 2, Nov»mfaer-D«cemfa«r 1971
543
MEYER, CHARNEY, RIVERA, MATHEW
two who were seen within 48 hours of onset.
Severely ill patients were admitted or transferred
to the intensive care units of The Methodist
Hospital where blood pressure, electrocardiograms, blood gases, and electrolytes were monitored.
Detailed records were kept from September,
1969, through November, 1970 (14 months), of
280 otherwise unselected consecutive patients
admitted or treated by the authors with all types
of cerebrovascular disease. Fifteen percent of the
patient population were Negro, 84% white, and
1% Asian Indian. The criteria cited for admission
to the study with a diagnosis of embolism were
employed within two days of hospitalization.
Forty-two (15%) of the patients met the
criteria cited and were diagnosed as having
cerebral embolism (tables 1-5). In 43% of cases,
the neurological deficit was preceded or accompanied by a seizure (table 6). In the majority of
patients, occlusion of the vessel thought to be
embolized was confirmed by arteriography (table
3). In all those in whom EEGs were recorded,
focal abnormality was observed in the area
supplied by the vessel thought to be involved, and
in many the brain scans showed an abnormal
uptake of radioisotope in the same areas. A few
patients had multiple episodes of cerebral embolization, and in a few others embolism took place in
other parts of the body. The remainder of the 280
patients suffered from acute signs and symptoms
due to cerebral thrombosis, intracranial hemorrhage, and subarachnoid hemorrhage.
Complete medical, cardiological, and neurological examinations were performed in all
patients. The most frequently observed neurological sign in patients with cerebral embolism was
asymmetrical deep tendon reflexes (83.3%) (table
2). Motor deficits consisting of hemiparesis or
hemiplegia were present in 73.8%, and the cranial
nerves were involved in 71.4% (fig. 1). Homonymous hemianopia was observed in 23.8% of
those with cranial nerve involvement and was
attributed to the high frequency of lesions at the
junction of the middle cerebral and posterior
cerebral arteries. The sensory examination was
abnormal in 42.9% of these patients.
TABLE 2
Abnormal Neurological Findings
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
Abnormal tendon jerks
Motor
Cranial nerve involvement
Second
Homonymous hemianopia
Monocular blindness
Fifth
Seventh
Ninth, tenth
Twelfth
Extensor plantar response
Disturbed consciousness
Sensory
Numbtr
Percent
(41 tat..)
35
31
30
83.33
73.80
71.40
10
4
7
20
2
2
25
29
18
23.80
9.50
16.70
47.60
4.80
4.80
59.50
69.00
42.90
considered to be occluded by the embolus (27
cases).
5. An abnormal uptake of technetium in
brain scans in the area thought to be infarcted
(ten cases).
6. Evidence of multiple cerebral embolic
episodes (nine cases).
7. Evidence of embolism to other parts of
the body (three cases).
All of the patients examined were admitted
to the Baylor-Methodist Center for Cerebrovascular Research or were seen in neurological
consultation on the medical and cardiovascular
services of The Methodist Hospital and treated by
the authors in collaboration with the referring
physicians.* Two or more of these criteria were
present in each patient (tables 1-6).
All patients were examined within 24 hours
of the onset of symptoms with the exception of
•The following persons kindly cooperated: Dr.
Michael E. DeBakey, Dr. E. Stanley Crawford, and
Dr. James Alexander, Baylor College of Medicine,
Houston, Texas; Dr. John Oschner, the Oschner
Clinic, New Orleans, Louisiana; and Dr. H. S. Roane,
the Green Clinic, Ruston, Louisiana.
TABLE 3
Laboratory Findings
T*«t pmr1ormm4
Arteriogram
Electroencephalogram
Brain scan
CSF pressure exceeding
180 mm H,O
544
No. cases •tadlod
% Normal
% Abnormal
28
30
11
7
10
9
93
90
91
20
50
50
Sfrokn, Vol. 2, Nov«mber-Dec«mb»r 1971
CEREBRAL EMBOLIZATION
TABLE 4
Source* of Cerebral Embolization
Porcvnt
(41cu«)
Nurabar of
CCUM
1.
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
Cardiac
Prosthetic valves (75% RHD)
Mitral stenosis
(rheumatic heart disease)
Arteriosclerotic heart disease
Myocardial infarction
Pacemaker failure
Arrhythmias
II.
Carotid plaques (ulcerated)
Right
Left
III. Carotid stenosis (nonulcerated)
IV. Aortic aneurysms with mural thrombi
V. Carotid kinking
VI. Vertebral thrombosis with embolization
Electroencephalography was performed in 30
patients, arteriography in 28, and brain scans in
11. Lumbar punctures were performed in half the
patients (table 3).
THERAPEUTIC MEASURES
Therapy was instituted in all patients within 48
hours of the neurological deficit. Since seizures
increase cerebral metabolism and are likely to
worsen the neurological deficit or prove fatal in
patients with heart disease, the 43% of patients
47.6
20
ii
26.2
l
2.4
14.3
6
7.1
4.8
2.4
%
2
1
33.3
14
S
9
4
2
1
1
11.9
21.4
9.5
4.8
2.4
2.4
who had seizures were treated with intramuscular
diphenylhydantoin (100 mg three times daily)
and phenobarbital (32 mg three times daily).
Eighteen patients were given diazepam, 5 to 10
mg intravenously, which successfully arrested
seizures immediately after administration. Two
additional patients received oral phenobarbital
and diphenylhydantoin prophylactically because
of spike activity in the EEG and the assumption
of imminent clinical seizures. In neither of these
two patients did seizures occur.
FIGURE 1
Microembolus of right retina from aortic valve prosthesis. The white, nonrefractile appearance is characteristic of platelet material.
Sfrok*. Vol. 2, Novamb»r-0etember T97J
545
MEYER, CHARNEY, RIVERA, MATHEW
TABLE 5
Sources and Sites of Cerebral Embolization (42
Cases)
Numbar
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
Total (all sources)
Left hemisphere
Right hemisphere
Bilateral
Cardiac origin
Left hemisphere
Right hemisphere
Bilateral
Carotid plaque origin
Left hemisphere
Right hemisphere
Bilateral
Porcant
(41 casai)
27
13
2
64
31
5
13
4
3
65
20
15
9
5
0
64
36
0
percent received papaverine hydrochloride intramuscularly in doses of 60 mg every six hours for
five days or 450 mg daily by mouth in order to
increase the cerebral collateral circulation to the
infarcted area.8 Ten percent were treated with low
molecular weight dextran.6 All patients with
embolism from prosthetic heart valves (26%) were
treated with dipyridamole, 100 mg four times
daily, since this drug has been shown to be
effective in preventing further embolization.
Anticoagulants were not used during the first
three weeks of the acute ictus because of the
danger of increasing the risk of cerebral hemorrhage. Anticoagulants had been used in the
management during the chronic stage in patients
with recurrent cerebral embolism due to valvular
heart disease and prosthetic replacements.
Results
Forty-three percent of patients were treated
with continuous or intermittent inhalation of 40%
to 100% oxygen since this minimizes anoxia in the
ischemic brain.0 In 38% intermittent 5% CO.
plus oxygen inhalation was administered for 15
minutes every hour in order to enhance the
cerebral collateral circulation and oxygen delivery
to the ischemic zone.7 In an effort to reduce
cerebral edema, 33% were treated for four days
with 10% glycerol either intravenously (1.2 gm/kg
of body weight daily) or orally (1.5 gm/kg of
body weight daily).* Forty-three percent received
dexamethasone, 12 mg intravenously, followed by
intramuscular injection of 4 mg every four hours
for five days to achieve the same end. Fifty-two
*U.S.P. commercial laboratory glycerol was used.
When given orally, it was mixed with orange or lemon
juice for flavoring. The intravenous preparation was
made by dissolving 50 gm of glycerol in 500 ml of
commercially available 5% glucose or normal saline.
The bottles were steam autoclaved for 20 minutes at
250° F in the HiVac autoclave on the liquid cycle
without drying time. During autoclaving, the tops of
the bottles were loosened and taped with masking
autoclave tape. When the cycle was finished, the tops
were tightened and sealed immediately. Commercially
available intravenous Soluset plastic tubing was
connected to an indwelling intravenous catheter.
TABLE 6
Incidence of Seizures in 42 Cases of Cerebral
Embolization
Total with seizures
Focal
Generalized
546
18
14
4
(42.8%)
(78.0%)
(22.0%)
The age and sex of 42 patients with cerebral
embolism are shown in table 7. Men exceeded
women by a ratio of approximately 3:1. The
mean age for the males was 61 years and for
the females 56 years. The mean age for the
entire group was 60 years and the median age
was 63.5 years.
Thirty-nine of the patients with cerebral
embolism were white, two were Negro, and one
was an Asian Indian. This disproportionate
ratio of Caucasian to Negro probably represents the higher incidence of hypertensive
intracranial vascular disease in the Negro race
and a higher incidence of extracranial ulcerative plaques in the carotid arteries in the white
population. 10 ' 11 Six (14.2%) of the patients
TABLE 7
Age and Sex Distribution
Aga
Mala
Pemala
0
1
1
4
7
12
4
2
31
1
1
0
0
4
1
4
0
11
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
Total
Mean.Female
Male
Total
56.45
61.00
59.80
Total
(parcant of 41)
1
2
1
4
11
13
8
2
42
Median-.
Female
Male
Total
(2.4)
(4.8)
(2.4)
(9.5)
(2.62)
(31.0)
(19.0)
(4.8)
59
64
63.5
Stroka, Vol. 2, Novamber-Decamber 7971
CEREBRAL EMBOLIZATION
LET
v
LT P
-
^^^^A/\\yK^'^V^^\j^^^'^
RF-LF
l5O
I Sec
FIGURE 2
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
0/ patient with left focal seizures showing epileptogenic focus arising in right posterior
quadrant of the head regions examined.
were diabetic and 21 ( 5 0 % ) were hypertensive, which is a low incidence compared with
cases of cerebral infarction due to thrombosis.8' 12 Probably the high incidence of embolism caused by rheumatic and valvular heart
disease accounts for the difference.18
The electroencephalograms, brain scans,
and arteriograms were abnormal in more than
90% of the patients tested (table 3 ) . Focal
abnormalities were noted in the majority of
EEGs (fig. 2) and in brain scans in areas
diagnosed clinically as being infarcted. Arteriograms confirmed occlusion of the vessel
thought to be occluded by embolization in
9 3 % of patients in whom the test was carried
out (fig. 3 ) . In 50% of cases, the initial CSF
pressure was more than 180 mm H 2 O, which
was attributed to cerebral edema.
Table 4 shows the sources of cerebral
emboli in the present series of patients. The
heart was the most common source (47.6%)
followed by emboli from ulcerated plaques in
the carotid arteries ( 3 3 . 3 % ) . Other sources
were carotid stenosis without ulceration
(9.5%) and aortic aneurysm with mural
thrombi ( 4 . 8 % ) . Rare causes were carotid
kinking due to atherosclerosis and proximal
vertebral thrombosis.
The majority of emboli ( 6 4 % ) lodged in
the left cerebral hemisphere, 31 % were in the
right hemisphere, and 5% were bilateral (table
5). When embolization was of cardiac origin, it
appeared to have a predilection for the left
hemisphere. In contrast, a high percentage
Stroke, Vol. 2, Novembtr-December 1971
(although not all) of emboli from carotid
plaques lodge in the ipsilateral hemisphere.
CONVULSIVE SEIZURES DUE TO
CEREBRAL EMBOLIZATION
Eighteen (42.8% ) of the patients in our series
with cerebral embolization had convulsive
seizures (table 6 ) , the majority ( 7 8 % ) of
which were focal seizures of the Jacksonian
type. The remainder were generalized. This
incidence of seizures is four times higher than
that reported by Dodge et al.,14 in a
retrospective analysis of autopsied cases of
cerebrovascular disease. In all patients with
seizures, this was the first evidence of
neurological deficit. In a few patients, seizures
recurred after a three-week interval, but one
patient had recurrent seizures after six weeks
and another one at 12 months following the
embolic episode. In the latter two patients, the
seizures coincided with withdrawal of anticonvulsive drugs.
Discussion
The reported incidence of cerebral embolism as
a cause of stroke found at autopsy or from
clinical evaluation without benefit of electrocardiography, EEG, and brain scan varies from
a low of 8% reported by Carter15 in a
retrospective study of patients with cerebrovascular disease to 46% observed by Jorgensen
and Torvik 10 and 50% reported by Adams and
Vander Eecken. 17 The latter two studies were
based on autopsy findings in which hemorrhagic infarction was the main criterion. In
another autopsy series in which rigid criteria
547
MEYER, CHARNEY, RIVERA, MATHEW
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
FIGURE 3
Carotid arteriogram showing occlusion of the left middle cerebral artery from embolus arising from mural thrombosis of the heart after pacemaker was inserted for complete heart
block resulting from myocardial infarction.
other than hemorrhagic infarction were used,18
the incidence was only 8.7%. Kannel et al.6
provided the only available prospective study
of a population group and reported an
incidence of 15% of strokes due to cerebral
embolism, which is identical with the incidence
in our selected series of hospitalized patients
suffering from all types of stroke in which the
embolic nature was confirmed using the criteria
described.
Males predominated in our study, and
548
other studies have also shown a male preponderance of cerebral embolism,19 although Kane
and Aronson 18 and Kannel et al. s found a
higher female incidence. In the productive
years of life (before age 65) incidence of
arteriosclerotic disease of the intracranial and
extracranial vessels is highest in males. After
the age of 65 and the menopause, the incidence
of cerebrovascular disease is higher in females
than in males.10 In the present series, rheumatic heart disease was a major cause of cerebral
Stroke, Vol. 2, Novemb«r-O«c«mber 7971
CEREBRAL EMBOLIZATION
FIGURE 4
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
Photograph of a variant ball valve inside the cage (a) and out of the cage (b). The patient
had numerous episodes of platelet emboli (two to the right and three to the left cerebral
hemisphere) which were treated with dipyridamole. This valve was replaced by Dr. E. Stanley
Crawford because of the possibility of "sticking" in the cage. Since surgical replacement
four months ago, the patient has been asymptomatic. Note the platelet depositions and the
swollen, irregular surface. Also note the fibrous overgrowth around the seating of the valve
cage. (This photograph was kindly made available by Dr. Crawford.)
embolism and the incidence was highest in
females.
The finding that the heart is the most
common source of emboli is in agreement with
all other prospective studies;11 however, the
development of improved surgical techniques
for patients with chronic rheumatic heart
disease has resulted in changes in the epidemiology of cerebral embolism. Prosthetic heart
valves now have become the most common
source of cerebral embolization. Rheumatic
valvular disease usually with prosthetic replacement of diseased valves was the source of
embolization in more than half of the patients
with cerebral emboli of cardiac origin. This is
in contrast with the series of patients reported
ten years ago by Carter18 and others 20 in whom
untreated rheumatic valvulitis was the most
common cause of cerebral embolization.
The prophylactic use of antibiotics in
treating rheumatic heart disease brought about
a reduction in incidence of this disorder as well
as subacute bacterial endocarditis as a cause of
stroke,12 but now that chronic cases are being
treated by inserting prosthetic valves it is again
becoming a common source of cerebral emboli.
In our series, prosthetic aortic valves were the
source in 22% and prosthetic mitral valves in
44% of patients with cerebral emboli of
cardiac origin. Ball valve variance occurs most
commonly in the aortic prosthetic valves,
Stroke, Vol. 2, Novemb«r-D»ctmb»r 7971
presumably due to the high pressure in this
system which causes distortion and lipid and
platelet deposition on the ball of the valve.21-2This was particularly true of the early silastic
ball valves of the Starr-Edwards type. In one
patient, not included in this series, with
cerebral embolism in whom the valve was
replaced because of repeated small vessel
emboli, white deposits having a thorny, linear
appearance were observed on the circumference of the valve (fig. 4 ) . These appeared to
be due to aggregated platelets and did not have
the characteristic appearance of red organized
thrombus which tends to occur soon after
insertion of the valves, usually at the site of the
suture line to the myocardium or on the cage of
the valve. An example of a ball from a valve is
shown in figure 5 on which may be seen white
platelet deposits.
From a neurological and therapeutic point
of view, emboli from prosthetic valves cause
two distinctly different types of cerebral
embolization:
1. Recurrent small cerebral emboli, best
illustrated in case No. 20 in this series. Four
small cerebral emboli had lodged in the left
hemisphere and three in the right hemisphere
in this patient. In patients with emboli of this
type, the neurological symptoms consist of
typical transient ischemic attacks of less than
24 hours' duration with hemiparesthesias,
549
MEYER, CHARNEY, RIVERA, MATHEW
FIGURE 5
Aortic valve showing variance removed from a patient on February 5, 1971. The prosthetic
valve had been originally inserted on April 1, 1966. Irregular lipid deposits can be seen on
the valve (arrows). (Obtained through the kindness of Dr. George Magovern, Allegheny
General Hospital, Pittsburgh, Pennsylvania.)
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
hemiparesis, and minor episodes of dysphasia.
These episodes may or may not be associated
with ball valve variance, and in several of our
patients murmurs of the type due to insufficiency of the prosthetic valve were not present.
These small platelet emboli tend to occur three
to five years after insertion of the StarrEdwards type of ball valve, particularly those
fabricated from silastic material. Such emboli
are believed to be fibrinoplatelet in nature, and
dipyridamole appears to be more effective in
their prevention than anticoagulant therapy
alone.28
2. Larger cerebral emboli, which cause
more severe and persistent neurological deficits
that last for weeks with incomplete recovery,
are believed to be due to dislodged fragments
of red thrombus from accumulations on the
cage of the valve or at its seating in the
myocardium. They occur most commonly
during the immediate postoperative period and,
in rare instances, three to five years later when
granulation tissue invades the struts of the
cage. Anticoagulant therapy is useful in this
group of patients. This was found to be the
case in one of our patients not included in the
present series in whom the valve was removed
and replaced due to cerebral embolism with a
protracted neurological deficit. Following replacement of the valve in this patient, there
have been no further neurological symptoms.
Cardiac dysrhythmias occurred in the vast
majority but not all of our patients with
cerebral embolism of cardiac origin. Atrial
fibrillation occurred in 8 5 % , but regular sinus
550
rhythm was present at the time of embolization
in 15%. This conclusion was based not only on
physical examination and repeated electrocardiograms but also on EEG monitoring in the
intensive care unit. There is no doubt that
cerebral embolization of cardiac origin is
usually associated with cardiac dysrhythmia;
however, it should be stressed that cerebral
embolization may occur in rare instances when
the sinus rhythm is normal, e.g., following
myocardial infarction or prosthetic replacement of a diseased valve. While it could be
argued that a brief episode of fibrillation may
have been missed at the time of embolism, this
seems unlikely since the ECG was continuously
monitored in an intensive care unit.
Recurrent small cerebral emboli from
prosthetic valves occurred in 11 of our patients
who were on anticoagulant therapy. Institution
of dipyridamole therapy in those with recurring
embolism, despite treatment with anticoagulants, was effective in decreasing the incidence
of further neurological deficit due to embolization from prosthetic valves, even though
anticoagulation therapy was discontinued. The
results obtained with this drug thus tend to
confirm the findings of Sullivan et al.,23 who
found in a controlled study that the combination of dipyridamole and anticoagulants was
more effective than anticoagulants alone and
also suggested that dipyridamole alone without
anticoagulants may be effective in preventing
cerebral embolism of the platelet type from
prosthetic valves. All of the present series of
patients have been followed for six to 15
Strok;
Vol. 2, November-December 797)
CEREBRAL EMBOLIZATION
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
months, and none of those given large doses of
dipyridamole, 100 mg four times daily, have
had further cerebral embolic episodes.
The incidence of cerebral emboli from
mural thrombosis due to myocardial infarction
reported by Carter16 was higher than in our
series. As stated, cerebral emboli from this
source tend to be large and produce a severe
neurological deficit. It is doubtful that dipyridamole would prevent these large emboli of
the red thrombus type. Unfortunately, anticoagulants are contraindicated during the acute
stage (three weeks) of cerebral infarction
because of the risk of intracranial hemorrhage.
In such cases, anticoagulant therapy was
delayed for three weeks and then instituted.
Ulcerated atherosclerotic plaques at the
bifurcation of the internal carotid artery were
the second most frequent cause of cerebral
embolization in our patients (33.3%). The
presence of these lesions was confirmed in
every case by arteriography and biopsy of the
surgical specimen. Carotid endarterectomy now
has been proved to be the most appropriate
form of treatment in these patients to lessen
both the risk of further embolization or
death.11
The relatively low incidence of myocardial infarction as a source of cerebral embolism
in the present series of patients is probably due
to the increased incidence of cerebral emboli as
a result of new techniques of treating heart
disease such as insertion of prosthetic valves
and pacemakers. Pacemakers are usually
inserted in patients with heart block from
myocardial infarction, and emboli from mural
thrombi may be dislodged in the heart as the
cardiac rate is changed. While these are
lifesaving procedures, they still carry a calculated and acceptable risk of cerebral embolization which can be reduced by medical
treatment.
The left hemisphere was involved more
frequently than the right one when emboli were
of cardiac origin. Brain stem embolization
occurred simultaneously with cerebral embolism in only two patients.
Embolization of the vertebral and basilar
arteries due to myocardial infarction without
concurrent embolization of the carotid arteries
may be a cause of cortical blindness due to
"saddle embolism" of the basilar artery where
it bifurcates into both posterior cerebral
Strck;
Vol. 2, Nov«mb«r-Dectmber 7977
arteries. This did not occur in a single patient
in our series and is admittedly a rare
complication.24' 25
The higher incidence of cerebral embolism via the left carotid artery compared to the
right one has never been satisfactorily explained. Two factors are suggested which may
account for this: (1) cerebral embolization of
the left hemisphere, which is the dominant
hemisphere in more than 90% of people,
causes more pronounced, recognizable clinical
signs and symptoms than embolization of the
nondominant side. Hence, the patient is more
likely to seek medical attention and the
neurological deficit is more easily recognized,
and (2) the innominate artery usually arises
at an oblique angle from the aortic arch, the
angle being in the direction opposite the
curvilinear trajection of the ejected bolus of
blood with each systole (fig. 6), while the left
common carotid artery normally rises vertically
from the left side of the aortic arch. Hence,
FIGURE 6
Diagram of aortic arch and major branches showing
direction of blood flow after each systole.
551
MEYER, CHARNEY, RIVERA, MATHEW
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
emboli are more likely to pass through the
latter vessel. Furthermore, if an embolus
should enter the innominate artery, the
trajectory of force in the reverse direction tends
to direct it toward the subclavian rather than
the carotid artery. In our view, the latter
explanation seems the most important since all
our patients had detailed neurological evaluations and it is doubtful that right hemispheric
lesions were missed. Other investigators have
confirmed the preponderance of left hemispheric infarction.18 Reports to the contrary
are rare.15
The unusually high incidence of seizures
(42.8%) as a manifestation of cerebral
embolization found in our patients was
striking, regardless of the source of emboli. A
seizure, usually focal, involving the face, hand,
and arm was frequently the initial symptom.
Seizures of such a focal nature occurred in
78% of our patients; the remainder were
generalized. Hence, seizures that occur as a
manifestation of cerebrovascular disease should
raise the question of embolism in the differential diagnosis of their cause. Such acute
seizures were invariably controlled with intravenous diazepam and repeated seizures were
prevented with diphenylhydantoin and phenobarbital. Rapid institution of therapy is
undoubtedly lifesaving in patients with severe
heart disease, particularly in the postoperative
period. The importance of long-term anticonvulsive therapy was illustrated by two patients
who had recurrent focal seizures several weeks
or months after withdrawal of diphenylhydantoin and phenobarbital.
In addition to controlling seizures and
preventing further embolization, it seems
logical that therapy should be directed toward
reducing cerebral edema and promoting increased collateral blood flow to the infarcted
area. Such therapy may decrease the size of the
infarction, thereby enhancing the dissolution
and removal of the embolic fragments. Regional edema was extensive enough in 50% of our
patients to increase the intracranial pressure.
Glycerol, 500 ml in 10% solution of saline,
infused daily for four days was found to be
extremely effective in reducing cerebral edema
as judged by the clinical course of the patient
and repeated measurement of CSF pressure.
This innovation in therapy was believed to
contribute to the low mortality rate of 9.5%
(four of 42 patients) and rapid recovery in the
552
present series. Carter,20 in a series of patients
with untreated cerebral embolism, reported that
one-third (33.34%) died from the immediate effects of embolism. In another series, Carter11"'
reported a higher incidence of myocardial infarction (31%) versus 14% as the cause of
cerebral embolism. Although such differences
in case series may influence case-mortality
ratios, infarction of the brain was responsible
for death in only 7% of the series reported in
this communication. The other causes of death
were cardiac in origin. The neurological deficits
in the present series were comparable in
severity with those found in other series. Harris
and Levine27 also reported immediate mortality
of one-third of their patients with cerebral
embolism. Wells,28 in a series of patients with
acute cerebral embolism, reported that 16 of
53 (30.84%) patients died as a result of
embolism and its immediate complications.
Whisnant,29 in a series of patients with cerebral
infarction due to thromboembolism, reported
an acute mortality rate of 27 %.
Cerebrospinal fluid pressure was reduced
in all four patients in whom lumbar puncture
was repeated at the completion of treatment. In
other patients with acute cerebral infarction
not included in the present series, this form of
therapy has resulted in a measurable increase
in regional cerebral blood flow.80 Dexamethasone, although less effective than glycerol, has
been shown recently to reduce cerebral edema
and to improve the prognosis of stroke
patients.S1'32 Low molecular weight dextran
increases cerebral blood flow and oxygen
delivery to infarcted brain presumably by
decreasing cerebral edema and platelet adhesiveness.9
Intermittent inhalation of 5% CO2 in
oxygen has been shown to increase regional
cerebral blood flow in areas of infarction by
increasing the collateral circulation.88'84 The
effect of 5% CO2 inhalation in clearing the
brain of fat emboli has been demonstrated
clinically by Harnett et al.3B Likewise, papaverine administered by intramuscular or intravenous injection increases total and regional
blood flow following hemispheric infarction
and has been shown to decrease the mortality
rate and improve the quality of survival in controlled clinical trials.8'36
It appears from the present clinical study
that diagnostic methods are now available to
Stroke, Vol. 2, Novemb*r-D*cembsr 1971
CEREBRAL EMBOLIZATION
accurately diagnose cerebral embolization during life. Therapeutic measures are discussed
which may prevent further cerebral emboli,
reduce seizure activity which is common, and
minimize the neurological deficit due to acute
cerebral embolism.
14.
15.
References
1.
2.
3.
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
4.
5.
6.
7.
8.
9.
10.
1 1.
12.
13.
Borhani NO: Differences among hospitals in
case-fatality ratios of cerebrovascular disease.
Ann Int Med 72: 953-955, 1970
Wells CE: Premonitory symptoms of cerebral
embolism. Arch Neurol 5: 490-496, 1961
Fisher CM, Gore I, Okabe N, et a l : Atherosclerosis of the carotid and vertebral arteries—extracranial and intracranial. J Neuropath Exp
Neurol 4 : 455-476, 1965
Castaigne P, Lhermitte F, Gautier J-C, et a l :
Internal carotid occlusion. A study of 61
instances in patients with post-mortem data.
Brain 93: 231-258, 1970
Kannel WB, Dawber TR, Cohen ME, et a l :
Vascular disease of the brain—epidemiologic
aspects: The Framingham study. Amer J
Public Health 55: 1355-1366, 1965
Meyer JS, Fang HC, Denny-Brown D: Polarographic study of cerebral collateral circulation.
Arch Neurol Psychiat 72: 296-312, 1954
Gotoh F, Meyer JS, Tomita M : Carbonic
anhydrase inhibition and cerebral venous blood
gases and ions in man. Demonstration of increased oxygen availability to ischemic brain.
Arch Int Med 117: 39-46, 1966
Meyer JS, Gotoh F, Gilroy J, et a l : Improvement of brain oxygenation and clinical improvement in patients with stroke associated
with papaverine hydrochloride. JAMA 194:
957-961, 1965
Gilroy J, Barnhart M l , Meyer JS: Treatment of
acute stroke with dextran 40. JAMA 210: 293298, 1969
Fields WS, North RR, Hoss WK, et a l : Joint
study of extracranial arterial occlusion as a
cause of stroke. I. Organization of study and
survey of patient population. JAMA 203: 955968, 1968
Fields WS, Maslenikov V, Meyer JS, et a l : Joint
study of extracranial arterial occlusion. V.
Progress report of prognosis following surgery or
nonsurgical treatment for transient cerebral
ischemic attacks and cervical carotid artery
lesions. JAMA 211:1993-2003, 1970
Meyer JS: Acute stroke; biochemical and
therapeutic studies. Minn Med 47: 2 6 5 - 2 7 1 ,
1964
Dyken M L : Precipitating factors, prognosis,
and demography of cerebrovascular disease in
an Indiana community. A review of all patients
Sfroke, Vol. 2, November-December 1971
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
hospitalized from 1963 to 1965 with neurological examination of survivors. Stroke 1 : 2 6 1 269 (July-Aug) 1970
Dodge PR, Richardson ED, Victor M : Recurrent convulsive seizures as a sequel to cerebral
infarction. Brain 77: 610-638, 1954
Carter AB: Immediate treatment of cerebral
infarction. Seven years' personal experience.
Lancet 1 : 3 4 5 - 3 5 1 , 1960
Jorgensen L, Torvik A : Ischaemic cerebrovascular diseases in an autopsy study. Part 1 :
Prevalence, location, and predisposing factors
in verified thrombo-embolic occlusions, and
their significance in the pathogenesis of
cerebral infarction. J Neurol Sci 3: 490-509,
1966
Adams RD, Vender Eecken H M : Vascular
diseases of the brain. Ann Rev Med 4 : 213252, 1953
Kane WC, Aronson SM: Cardiac disorders
predisposing to embolic stroke. Stroke 1 : 164172 (May-June) 1970
Leussenhop A J : Occlusive disease of the
carotid artery. Observations on the prognosis
and surgical treatment. J Neurosurg 16: 705730, 1959
Editorial: Neurological complications of infective endocarditis. Brit Med J 2 : 619, 1970
Lee SJK, Haraphongse M, Callaghan JC, et a l :
Hemodynamic changes following correction of
severe aortic stenosis using the Cutter-Smeloff
prosthesis. Circulation 4 2 : 719-728, 1970
McHenry M M , Smeloff EA, Fong WY, et a l :
Critical obstruction of prosthetic heart valves
due to lipid absorption by silastic. J Thorac
Cardiovas Surg 59: 413-425, 1970
Sullivan JM, Harken DE, Gorlin R: Effect of
dipyridamole on the incidence of arterial
emboli after cardiac replacement. Circulation
39 and 40 (suppl 1) : 1149-1153, 1969
Gilroy J, Meyer JS: Medical
Neurology
London, Macmillan Co, p 515-516, 1969
Loeb C, Meyer JS: Strokes Due to VertebroBasilar Disease. Infarction, Vascular Insufficiency and Hemorrhage of the Brain Stem and
Cerebellum. Springfield, Illinois, Charles C
Thomas, 1965
Carter AB: Prognosis of cerebral embolism.
Lancet 2 : 514-519, 1965
Harris AW, Levine SA: Cerebral embolism in
mitral stenosis. Ann Int Med 15: 637-643,
1941
Wells CE: Cerebral embolism; the natural
history, prognostic signs and effects of anticoagulation. Arch Neurol Psychiat 81 : 661-611,
1958
Whisnant JP: Epidemiology of stroke in
Rochester, Minnesota. American Heart Association, Joint meeting of the Council on
553
MEYER, CHARNEY, RIVERA, MATHEW
Cerebrovascular Disease and the Council on
Epidemiology on the Epidemiology of Cerebrovascular Disease, New Orleans, Louisiana,
March 4, 1970
30. Mathew NT, Meyer JS, Ericsson AD, et a l : New
method for measuring regional cerebral blood
flow and blood volume -in man using the
gamma camera. Trans Amer Neurol Assoc, in
press, 1971
3 1 . Patten BM, Mandel J, Bruun B, et a l : Double
blind study of the effect of dexamethasone on
acute stroke. Neurology 2 1 : 403, 1971 (abstract)
32. Russek H I , Russek AS, Zohman BL: Cortisone
in immediate therapy of apoplectic stroke.
JAMA 159: 102-105, 1955
33. Cooper ES, West JW, Jaffe ME, et a l : Effect of
CO2 inhalation on cardiac function and
cerebral blood flow in stroke patients. Circulation 42 (suppl 3) : 120, 1970 (abstract)
34. Meyer JS, Fukuuchi Y, Kanda T, et a l :
Regional measurements of cerebral blood flow
and metabolism using intracarotid injection of
hydrogen, with comments on intracerebral
steal. In Ross Russell RW (ed) : Brain and
Blood Flow. London, Pitman Medical and
Scientific Publishing Co., p 71-79, 1971
35. Harnett RW, Lowe KG, Paterson JR, et a l :
Treatment of cerebral-fat embolism. Lancet 2 :
762-764, 1959
36. McHenry LC Jr, Jaffe ME, Kawamura J, et a l :
Effect of papaverine on regional blood flow in
focal vascular disease of the brain. New Eng J
Med 282: 1 167-1 170, 1970
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
554
Strokm, Vol. 2, Novmbar-Dac»mber 1971
Cerebral Embolization: Prospective Clinical Analysis of 42 Cases
JOHN STIRLING MEYER, JONATHAN Z. CHARNEY, VICTOR M. RIVERA and NINAN T.
MATHEW
Stroke. 1971;2:541-554
doi: 10.1161/01.STR.2.6.541
Downloaded from http://stroke.ahajournals.org/ by guest on June 12, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1971 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/2/6/541
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/