31
Patent Foramen Ovale: Is Stroke Due to
Paradoxical Embolism?
D. Ranoux, MD; A. Cohen, MD; L. Cabanes, MD; P. Amarenco, MD;
M.G. Bousser, MD; and J.L. Mas, MD
Background and Purpose: A patent foramen ovale has been reported to be significantly more frequent in
stroke patients than in matched control subjects, and paradoxical embolism has been suggested as
the main mechanism of stroke in this situation. The present study was designed to test this hypothesis.
Methods: Sixty-eight consecutive patients under 55 years of age presenting with an ischemic stroke had
an extensive workup, including transesophageal echocardiography with contrast. We compared the
prevalence of criteria for the diagnosis of paradoxical embolism in patients with and without a patent
foramen ovale.
Results: A patent foramen ovale was found in 32 patients (47%). A Valsalva-provoking activity was
present at stroke onset in six patients with a patent foramen ovale and in eight patients with no patent
foramen ovale (X2=O.1, nonsignificant). Clinical/radiological features suggestive of an embolic mechanism were not more frequent in patients with a patent foramen ovale. Clinical evidence of deep vein
thrombosis was present in one patient with a patent foramen ovale and in none of the others. No occult
venous thrombosis was found in a subgroup of patients with a patent foramen ovale and no definite cause
for stroke who underwent venography (n=13).
Conclusions: Our results do not support the hypothesis that paradoxical embolism is the primary
mechanism of stroke in patients with a patent foramen ovale. (Stroke 1993;24:31-34)
KEY WORDs * cerebral ischemia * embolism * foramen ovale, patent
young
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
P aradoxical embolism, defined as the systemic embolism of venous or right atrial origin through a
right-to-left cardiac shunt, was formerly considered to be a rare cause of stroke. In almost all reported
cases, the conduit for paradoxical embolism was a
patent foramen ovale (PFO). Interest in paradoxical
embolism has been rekindled by the advent of contrast
echocardiography, a noninvasive and reliable technique
for the detection of small right-to-left cardiac shunts. In
1988, two case-control studies demonstrated a significantly higher prevalence of PFO in young patients with
ischemic stroke (40-50%) than in control subjects (1015%).1,2 Whether, and to what extent, the excess prevalence of PFO in young stroke patients is related to
paradoxical embolism remains unsettled.3 If one assumes that most strokes in patients with PFO are due to
paradoxical embolism, then a higher prevalence of overt
or occult deep venous thrombosis (DVT), or risk factors
for DVT, would be expected in these patients than in
stroke patients with no PFO. Likewise, an excess prevalence of Valsalva-provoking activities (which tranReview of this manuscript was directed by Editor-in-Chief Mark
L. Dyken.
From the Service de Neurologie (D.R., J.L.M.), H6pital SainteAnne, the Service de Neurologie (P.A., M.G.B.) and Service de
Cardiologie (A.C.), H6pital Saint-Antoine, and the Service de
Cardiologie (L.C.), H6pital Cochin, Paris.
Address for correspondence: Prof. J.L. Mas, Service de Neurologie, H6pital Sainte-Anne, Centre Raymond Garcin, 1 rue Cabanis, 75674 Paris, CEdex 14, France.
Received July 29, 1992; final revision received September 29,
1992; accepted September 29, 1992.
siently increase or provoke right-to-left shunting in
patients with PFO) at stroke onset and of features
suggesting cerebral embolism should theoretically be
found in these patients. We specifically addressed these
issues.
Subjects and Methods
Sixty-eight consecutive patients under 55 years of age
with ischemic stroke were included in this study. None
of these patients overlapped with those reported in a
previous study' by one of us. All patients had computed
tomography and/or magnetic resonance imaging of the
head. Cerebral angiography was performed in 62 patients, in 22 (35%) within 6 days after stroke onset.
Routine laboratory studies included complete blood
cell count; erythrocyte sedimentation rate; protein electrophoresis; antinuclear antibodies; glucose, serum cholesterol, and triglyceride levels; and prothrombin and
activated partial thromboplastin times. Extensive platelet, coagulation, and fibrinolysis studies were performed
in 44 patients.
Cardiac investigations. In addition to routine cardiac
examination (including electrocardiography and chest
roentgenography), all patients had transesophageal
echocardiography with contrast. The contrast was obtained by injecting rapidly 5-10 ml isotonic saline or
gelatin. We used two syringes, mounted on a three-way
stopcock, to mix the saline or gelatin with air. This
mixture was injected directly into an antecubital vein.
Three to six contrast injections were systematically
performed in each patient, in the resting state and
32
Stroke Vol 24, No 1 January 1993
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
during provocative maneuvers (Valsalva maneuver and
cough test). The echocardiographic diagnosis of PFO
was based on the appearance of more than five microcavitations into the left atrium within three cardiac
cycles after total opacification of the right atrium.
Search for deep venous thrombosis. Clinical signs and
risk factors for DVT (use of contraceptive agents,
obesity, prolonged bed rest within the month before
stroke onset) were systematically sought. Venography
was performed in a subgroup of patients (n = 13) according to the following criteria: 1) presence of PFO, 2) no
definite cause for stroke, and 3) examination within 4
weeks after stroke onset. None of the patients examined
by venography had severe motor (leg) deficit. For
ethical reasons, venography was not performed in patients with no PFO or with PFO and a definite cause for
stroke.
Circumstances of onset. All patients were asked about
their activities within minutes preceding stroke onset.
Sporting effort, straining at stool, intercourse, lifting a
heavy weight, getting up, laughing, and coughing were
considered as equivalents of Valsalva maneuvers.4,5
Information was obtained in 64 patients; in four patients
with no evidence of right-to-left shunting the information was not available because of aphasia.
Definite cause for stroke. It was defined as any disease
that has been clearly linked to the occurrence of stroke:
arteriopathies such as dissections, atherosclerosis with
stenosis of >50%, or angiitis; coagulopathies and systemic disorders such as thrombocythemia, disseminated
intravascular coagulation, or lupus erythematosus; and
cardiopathies such as mitral stenosis, atrial fibrillation,
or acute myocardial infarction. The presence of a
definite cause for stroke was assessed without knowledge of the results of contrast echocardiography. Atrial
septal aneurysm and mitral valve prolapse were not
considered to be definite causes for stroke, but only risk
factors for stroke. Other risk factors included hypertension, diabetes mellitus, cigarette smoking, hypercholesterolemia, migraine according to the criteria of the
International Headache Society,6 and previous or current use of contraceptive agents.
The patients were divided into two groups according
to the presence of a right-to-left shunt detected by
contrast transesophageal echocardiography. The two
groups were compared according to criteria consistent
with the diagnosis of paradoxical embolism: presence of
DVT or risk factors for DVT, Valsalva-provoking activity at stroke onset, and features suggestive of an embolic
mechanism. The presence of either arterial occlusion on
early angiography7 or at least one of sudden onset, early
loss of consciousness, and hemorrhagic or cortical infarct8-10 was considered to support the diagnosis of
cerebral embolism. These criteria were assessed in each
patient without knowledge of the results of contrast
echocardiography. Percentages were compared by the
X2 test with the appropriate degrees of freedom. Quantitative data were compared by t test.
Results
A positive contrast study was found in 32 patients
(47%): 21 at rest, and 11 only after provocative maneuvers. This percentage rose to 57% when only patients
with no definite cause for stroke were taken into account. In two patients, the right-to-left shunt was not
TABLE 1. Characteristics of Patients With and Without Patent
Foramen Ovale
With
Characteristic
Mean age (yr)
Sex
Female
Male
Definite cause for stroke
Dissection
Acute myocardial infarction
Cerebral angiitis
Thrombocythemia
Lupus erythematosus + antiphospholipid
antibodies
Disseminated intravascular coagulation
Atherosclerosis
Mitral stenosis with atrial fibrillation
Risk factors for stroke
Mitral valve prolapse
Atrial septal aneurysm
(n=32)
Without
(n=36)
36.8
40.2
13
19
1
1
0
0
0
12
24
13
5
1
1
2
0
0
0
01
1
1
1
1 (1)
7 (7)
6 (6)
3 (3)
2 (2)
0 (0)
8 (8)
5 (5)
13 (13)
5 (2)
4 (3)
Migraine
4 (2)
Oral contraceptive use
5 (2)
Diabetes mellitus
2 (1)
Hypertension
8 (8)
Cigarette smoking
15 (13)
Hypercholesterolemia
3 (3)
No risk factors
9 (4)
Numbers in parentheses concern patients with no definite cause
for stroke.
due to PFO. One patient had a patent ductus arteriosus
proven by aortography, and the other had an ostium
primum defect with a left-to-right shunt at rest and a
right-to-left shunt after Valsalva maneuver.
Characteristics of the patients with and without PFO
are shown in Table 1. Mean age and sex ratio in the
patients with PFO (36.8 years, male: female= 1.5) did
not differ significantly from those in patients with no
PFO (40.2 years, male: female=2). A definite cause for
stroke was found in one of the 32 patients with PFO and
in 13 of the 36 patients with no PFO (y2= 11.3, df=1,
p<O.OOl). Among the patients with no definite cause for
stroke, 18 (56%) with PFO and 19 (53%) with no PFO
had one or more risk factors for stroke (Table 1).
Table 2 shows the prevalence of criteria for paradoxical embolism in the patients with and without PFO.
Risk factors for DVT were present in five patients with
PFO and in 10 patients with no PFO (x2=l1.4 df=1,
nonsignificant). Only one patient with PFO had overt
clinical DVT. It occurred in a severely paralyzed leg 14
days after stroke onset and was confirmed by venous
echo Doppler. No latent DVT was found on venography
in the subgroup of 13 patients with PFO and no definite
cause for stroke.
Valsalva-provoking activities at stroke onset were not
more frequent in the patients with PFO than in the
patients with no PFO (x2=0.1 df= 1). Such activities
were present in six of the 32 patients with a right-to-left
shunt (six of the 31 with no or questionable cause for
their stroke) and in eight of the 32 patients with no PFO
Ranoux et al Patent Foramen Ovale
33
TABLE 2. Prevalence of Criteria for Paradoxical Embolism in Patients With and Without Patent
Foramen Ovale
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
Criteria
DVT
Clinical DVT
Positive venography
Risk factors for DVT
Circumstances of onset
Valsalva maneuver
During sleep
No Valsalva maneuver
Not available
Features suggesting embolic mechanism
At least one of sudden onset, early
loss of consciousness, and
hemorrhagic or cortical infarct
Arterial occlusion on early (<6 days)
angiogram
DVT, deep venous thrombosis.
With (N=32)
Without (N=36)
With cause
With cause
for stroke Without cause for stroke Without cause
(n = 13)
(n =23)
(n = 1)
(n=31)
0
0
0
1
0
5
...
3
7
0
0
1
0
6
5
20
0
7
1
12
2
11
5
6
2
1
25
12
15
...
4/13
1/2
2/7
(five of the 21 patients with no or questionable cause)
and no aphasia.
Clinical and radiological features suggestive of cerebral embolism did not differ significantly between the
two groups (Table 2).
Discussion
The prevalence of PFO in this study was 47% (57% in
those with strokes of undetermined etiology). A definite
cause for stroke was found in only one of the 32 patients
with a right-to-left shunt, compared with in 13 of the 36
patients with no PFO; the difference was significant.
These findings are in agreement with previous casecontrol studies demonstrating a higher prevalence of
PFO in young patients with stroke of unexplained
etiology than in control subjects.1"2
The main criterion for paradoxical embolism is the
presence of DVT. In our study, the only overt DVT was
likely a consequence of the stroke since it occurred 2
weeks after the stroke in a severely paralyzed leg. The
frequency of DVT secondary to stroke has been reported to be as high as 30% in studies using venous echo
Doppler and fibrinogen leg scanning.10 The risk of DVT
is related to the severity of the stroke and rises sharply
as early as the second day after stroke onset.10 No excess
of risk factors for DVT was found in our patients with
PFO. Finally, we did not find any occult DVT on
venography in 13 patients with PFO and no or mild to
moderate leg motor deficit. Although the number of
patients who underwent venography was relatively low,
the low prevalence of DVT and risk factors for DVT in
our study does not support a paradoxical embolic mechanism in stroke patients with PFO. Our results are
similar to those reported by Gautier et al"l in their
series of 29 patients with PFO and an unexplained
stroke or transient ischemic attack. Only three occult
DVT were found in 23 venographies performed within
2 days to 7 months after stroke onset. By contrast, in
another study12 venography disclosed occlusion of a leg
0
0
vein in 24 of 42 patients with PFO and cerebral (n =36)
or systemic (n = 6) ischemic events. However, no information is available concerning the severity of the motor
deficit or the precise time interval between stroke onset
and venography.
There are many pitfalls in the diagnosis of DVT that
may lead to an underrecognition of DVT. Even in
patients with documented pulmonary embolism, about
50% have no clinical signs of DVT. Furthermore, about
20% of patients with pulmonary embolism have no
evidence of venous thrombosis using conventional diagnostic tests.13 As pulmonary emboli may be of an occult
source, the lack of an identifiable source of many
paradoxical emboli is not unexpected. In a review of 29
cases of clinically diagnosed paradoxical embolism,
overt DVT was present in only three patients, and 14
were eventually found to have DVT.'4 The source of
emboli may remain undetected by conventional venography because of its location. In the two larger series of
pathologically proven systemic paradoxical embolism,'15"6 the sources of emboli (not stated in 20%) were
the leg veins in only 33%; the other sites were the right
atrium in 14%, pelvic veins in 25%, and miscellaneous
sites in 8%. Other diagnostic tools, such as immunoscintigraphy, will probably improve our ability to detect
DVT. Finally, DVT may disappear, either spontaneously or with anticoagulant therapy, before venography
is performed.
A sustained elevation of the right heart pressure,
usually due to pulmonary embolism,17 has long been
considered a precondition for paradoxical embolism.
However, contrast echocardiography has clearly demonstrated that the normal interatrial left-to-right pressure
gradient can be transiently reversed either spontaneously during early systole' or by Valsalva-provoking
activities.'8 In our study, Valsalva-provoking activities
were not more frequent in patients with PFO than in
patients with no PFO. Finally, we did not find any
difference between the two groups regarding features
34
Stroke Vol 24, No 1 January 1993
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
suggestive of cerebral embolism. However, the sensitivity and specificity of these clinical and radiological
criteria are poor.7
In conclusion, our study confirms that the prevalence
of PFO is high in young stroke patients and that stroke
in patients with PFO is not associated with the conventional definite causes for stroke. In addition our study
shows that conventional criteria for paradoxical embolism are not more frequent in patients with PFO than in
patients with no PFO, which suggests that paradoxical
embolism might not be the prevalent mechanism of
stroke in patients with PFO. This conclusion must be
tempered because of the relatively few patients who
underwent venography. In addition, it is possible that
the conventional criteria for paradoxical embolism on
which the present study is based may be insufficient or
inaccurate. The exact mechanism of stroke in patients
with PFO and no definite cause for stroke remains
unknown. The association between PFO and stroke
might be partly explained by other PFO-associated
cardiac or vascular abnormalities (confounding factors).
For instance, PFO has been reported to be associated
with mitral valve prolapse1 and atrial septal aneurysm.19
These potential confounding factors deserve further
studies.
References
1. Lechat P, Mas JL, Lascault G, Loron P, Theard M, Klimczac M,
Drobinski G, Thomas D, Grosgogeat Y: Prevalence of patent
foramen ovale in patients with stroke. N Engl J Med 1988;318:
1148-1152
2. Webster MWI, Smith HJ, Sharpe DN, Chancellor AM, Swift DL,
Bass NM, Glasgow GL: Patent foramen ovale in young stroke
patients. Lancet 1988;2:11-12
3. Mas JL: Patent foramen ovale, stroke and paradoxical embolism.
(editorial) Cerebrovasc Dis 1991;1:181-183
4. Biller J, Johnson MR, Adams HP Jr, Kerber RE, Corbett JJ, Bruno
A, Leth RJ: Further observations on cerebral or retinal ischemia in
5.
6.
7.
8.
9.
10.
patients with right-left intracardial shunts. Arch Neurol 1987;44:
740-743
Pedersen A, Sandoe E, Hvidberg E, Schwartz M: Studies on the
mechanism of the tussive syncope. Acta Med Scand 1966;179:
653-661
Headache Classification Committee of the International Headache
Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988;8:19-28
Cerebral Embolism Task Force. Cardiogenic brain embolism. Arch
Neurol 1986;43:71-84
Bogousslavsky J, Cachin C, Regli F, Despland PA, Van Melle G,
Kappenberger L: Cardiac sources of embolism and cerebral infarction -Clinical consequences and vascular concomitants: The Lausanne Stroke Registry. Neurology 1991;41:855-859
Kittner SJ, Sharkness CM, Sloan MA, Price TR, Dambrosio JM,
Tuhrim S, Wolf PA, Mohr JP, Hier DB, Caplan LR: Infarcts with
a cardiac source of embolism in the NINCDS Stroke Data Bank:
Neurologic examination. Neurology 1992;42:299-302
Landi G, D'Angelo A, Boccardi E, Candelise L, Mannuci PM,
Morabito A, Orazio EN: Venous thromboembolism in acute
stroke: Prognostic importance of hypercoagulability. Arch Neurol
1992;49:279-283
11. Gautier JC, Durr A, Koussa S, Lascault G, Grosgogeat Y: Paradoxical cerebral embolism with a patent foramen ovale. Cerebrovasc Dis 1991;1:193-202
12. Stollberger C, Karnik R, Schuster I, Kocher K, Leitner H, Langer
G, Slany J, Rudolfstiftung K: Diagnosing paradoxical embolism by
combined use of transesophageal contrast echocardiography
(TCE) and phlebography. (abstract) Circulation 1991;84(suppl II):
II-451
13. Hirsch J: Diagnosis of venous thrombosis and pulmonary embolism. Am J Cardiol 1990;65:45C-49C
14. Jones HR Jr, Caplan LR, Come PC, Swinton NW Jr, Breslin DJ:
Cerebral emboli of paradoxical origin. Ann Neurol 1983;13:
314-319
15. Corrin B: Paradoxical embolism. Br Heart J 1964;26:549-553
16. Johnson BI: Paradoxical embolism. J Clin Pathol 1951;4:316-332
17. Thompson T, Evans W: Paradoxical embolism. Q J Med 1930;23:
135-150
18. Dubourg 0, Bourdarias JP, Farcot JC, Gueret P, Terdjman M,
Ferrier A, Rigaud M, Bardet JL: Contrast echocardiographic visualization of cough-induced right-to-left shunt through a patent
foramen ovale. JAm Coll Cardiol 1984;4:587-594
19. Belkin RN, Waugh RA, Kisslo J: Interatrial shunting in atrial
septal aneurysm. Am J Cardiol 1986;57:310-312
Patent foramen ovale: is stroke due to paradoxical embolism?
D Ranoux, A Cohen, L Cabanes, P Amarenco, M G Bousser and J L Mas
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
Stroke. 1993;24:31-34
doi: 10.1161/01.STR.24.1.31
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1993 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/24/1/31
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/