Download Paradoxical Embolism* An Underrecognized

Paradoxical Embolism*
An Underrecognized Problem
Robert Ward, MD; David Jones, MD; and Edward F.
Despite reports of the clinical presentations and dev¬
astating consequences of paradoxical embolus (PDE)
to be
for more than a century, this diagnosis continues
missed. Because the prevalence of patent
frequentlyovale
foramen
(PFO) is 27 to 35% in the normal pop¬
ulation and the presence of deep vein thrombosis or
may not be clinically obvious, a
pulmonary embolus
in the event of un¬
high suspicion for PDE is needed
contrast echocar¬
While
occlusion.
arterial
explained and transcranial
Doppler ultrasound have
diography clinical
facilitated
recognition of PDE, the optimum
clarification. Primary
approachforto diagnosiswithrequires
is
PDE
anticoagulation, with
therapy patients
selected individ¬
in
considered
carefully
thrombolytics
uals, but there is little published information regard¬
ing long-term treatment and outcomes. Prevention
remains essential whenever possible. It is not yet de¬
"C^irst described by Cohnheim1 in 1877, paradoxical
-*" embolism (PDE) was defined as a venous throm¬
bosis causing systemic embolization through a rightto-left shunt. Multiple autopsy-proven cases were de¬
scribed in the French and German literature until 1930
when two patients diagnosed during life were report¬
ed.23 Since then, over 170 cases have been report¬
ed,4"13 with increasing numbers recognized antemortem.5,7'9"17 Recent investigations of patients with pe¬
cerebral arterial emboli ofunknown cause
ripheral and
PDE is a potentially treatable problem
that
suggest
that occurs more often than usual. We report four pa¬
tients with PDE, two of whom were diagnosed antemortem, and review the literature.
Case Reports
Case 1
A 30-year-old woman (gravida 2, para 1) was hospitalized while
in labor at 40 weeks' gestation. During delivery, she became cyan-
otic, apneic, and pulseless. Cardiopulmonary resuscitation was be¬
gun and an endotracheal tube was inserted. A viable 9 lb 10 oz in¬
fant was delivered via forceps. The patient developed ventricular
fibrillation and underwent defibrillation with conversion to sinus
rhythm but remained hypotensive. Arterial blood gas values dem¬
onstrated a profoundly increased alveolar-arterial O2 pressure dif¬
ference with a pH of 7.40, PaC02 of 9 mm Hg, and Pa02 of 168
mm Hg while breathing 100% O2. Initial chest roentgenogram
*From the Section on Pulmonary and Critical Care Medicine,
Bowman Gray School of Medicine, Winston-Salem, NC.
Haponik, MD, FCCP
fined whether prophylactic treatment of persons with
recognized predispositions to PDE (eg, PFO and pul¬
monary hypertension) is beneficial.
(CHEST 1995; 108:549-58)
AFE=amniotie fluid embolism; ASD=atrial septal defect;
CXR=chest x-ray; DVT=deep venous thrombosis; IVC=
inferior vena cava; LA=left atrium; PAP=pulmonary ar¬
tery pressure; PDE=paradoxical embolus; PE=pulmonary
embolus; PFO=patent foramen ovale; RA=right atrium;
atrial pressure; RV= right ventricle; TCD=
RAP=right
transcranial Doppler; TEE=transesophageal electrocar¬
diogram; TTE=transthoracic echocardiogram
Keywords: cryptogenic stroke; paradoxical embolus; patent
foramen ovale; pulmonary embolus; transcranial doppler;
transesophageal echocardiogram
(CXR) was normal and ECG revealed S-T segment elevation in
anterolateral leads. Laboratory studies included a platelet count of
81,000/mm3 and prolonged prothrombin and partial thromboplastin times. A transthoracic echocardiogram (TTE) showed increased
echogenicity in the inferior vena cava (IVC) and right atrium (RA),
which was believed to represent thrombus, and the patient received
thrombolytic therapy.
On transfer to our center the patient remained hypotensive, with
persistent coagulopathy, bleeding, and ECG changes. An intraaortic balloon pump was inserted, and cardiac catheterization showed
occlusion of the left anterior descending coronary artery. Left ven¬
tricular function was reduced and there was no evident thrombus.
Coagulopathy and shock continued despite maximal support and
the patient died. At autopsy,18 a cervical tear and disrupted
muscular wall were found. No atrial or coronary thrombi were seen,
but cardiac examination showed necrotic myocardium in the ante¬
rior left ventricular wall. The microvasculature in this region and
both lungs contained scattered squamous cells consistent with amniotic fluid emboli (AFE) (Fig 1). A pencil-thick 0.8-cm patent fo¬
ramen ovale (PFO) was found (Fig 2).
Case 2
A 49-year-old woman presented with a pulseless, cold left hand
10 days after open reduction of a left tibia-fibula fracture. She un¬
derwent thrombectomy of the left brachial and axial arteries. The
TTE suggested a left atrial thrombus. After heparinization, the pa¬
tient was transferred to our ICU with an enlarging hematoma of the
left arm. Physical examination revealed stable vital signs, a medial
incision of the left upper extremity with a large hematoma and a left
palsy. Irrigation and debridement of the arm were
was continued. Transesophageal
performed anticoagulation
revealed a PFO with
echocardiogram (TEE) with saline infusion
thrombus
but
no
intraatrial
shunt,
(Fig 3). Lower ex¬
right-to-left
tremity duplex Doppler showed no deep vein thrombosis (DVT).
On the third postoperative day, the patient developed hypoxemia
and left-sided pleurisy. Arterial blood gas determination revealed a
basilar atelectamedian
nerve
and
Pa02 of 67 mm Hg (FIo2, 0.6). The CXR showed
CHEST /108 / 2 / AUGUST, 1995
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
549
J?A
A
It
fr' 7 fell* fei?^
*
4/
.;<^Tft:<
.
la lit
mm
.'
^ %&'**'5
*JK8§
perfusion lung scan was indeterminate.
Pulmonary arteriogram demonstrated multiple bilateral pulmonary
emboli (PEs) and elevated pulmonary artery pressures (PAPs)
(50/19 mm Hg). An inferior vena cavagram was performed, a "bird's
nest" filter was inserted and the patient recovered uneventfully.
After 1 week, duplex Doppler studies remained negative, and she
was discharged on a regimen of warfarin therapy.
sis, and
a
ventilation
Case 3
A 53-year-old woman with hypertension and obesity with acute
respiratory failure was transferred to our institution. She had been
hospitalized for 11 days previously with right upper lobe pneumo¬
nia and now had vomiting and headache and had trouble keeping
her balance. In the emergency room, she developed a left
progressively dyspneic,
hemiparesis and right facial droop, became
and was intubated. Arterial blood gas determination revealed a pH
of 7.47, PaC02 of 42 mm Hg, Pa02 of 29 mm Hg while breathing
room air, and after intubation fraction of inspired oxygen [FIo2], 1.0;
positive end-expiratory pressure, 10 cm H20), apH of 7.58, PaC02
of 29 mm Hg, and a Pa02 of 31 mm Hg.
At transfer, examination revealed a blood pressure of 132/65 mm
Hg, constricted, reactive pupils, and no papilledema. There were
faint bilateral wheezes. The heart had a regular rate and rhythm with
a 3/6 systolic murmur along the left sternal border. The patient
Figure 2. A probe is placed through a PFO which is adjacent to
(Courtesy of P.E. Lantz, MD, Bowman Gray School of
Medicine, Wake Forest University.)
the RA.
case 1, fetal squamous cells
demonstrated in an intramyocardial arteriole. Ischemic myocytes are seen
lif}^ m ><%
I ~j« i, .<* j v adjacent to the arteriole (hematoxylin and
¦k,kVj^ eosin, original magnification x400).
l^'-^-
"'.
Figure 1. In
(arrow)
were
moved all extremities spontaneously and withdrew extremities
when pain stimulus was induced. The ECG showed anteroseptal
ventricular strain, and CXR
myocardial ischemia or possible rightatelectasis.
Massive PE was
revealed minimal right mid-lung
were not administered because of
suspected, but anticoagulants
a large right parietoneurologic changes. Cranial CT confirmed
area adjacent to the atrium of the lateral right
occipital, hypodense
ventricle (RV), mass effect, and associated edema. Lumbar punc¬
ture showed no infection or bleeding. The TTE revealed an ejec¬
tion fraction of 60%, RV volume and pressure overload with
RA, paradoxical septal movement, and tricuspid
enlarged RV and
an atrial septal defect (ASD) could not be
but
regurgitation,
excluded. There were no pericardial effusions, intracardiac masses,
or thrombi. Broad-spectrum antibiotics were started for possible
sepsis, and the patient worsened with fever (38.8°C), hypotension,
and progressive hypoxemia. She died within 48 h. Autopsy showed
multiple PEs with associ¬
right coronary occlusion by an embolus,
ated infarcts, a cerebral infarct, and a PFO.
Case 4
A 49-year-old man was admitted to
our
institution with left
leg
pain and swelling. He had a history of mesenteric arterial emboli
with bowel necrosis 2 years earlier, left middle cerebral arterial in¬
farction (with negative carotid ultrasound and TTE) 4 months ear¬
lier, right lower extremity DVT, and multiple bilateral segmental
defects by perfusion scan 2 months earlier. Medications included
and warfarin sodium (Coumadin). The blood
phenytoin (Dilantin) mm
Hg; heart rate, 80 beats per minute; res¬
pressure was 118/76
26
breaths
rate,
per minute; and temperature, 38.0°C. The
piratory
heart had a fixed, split S-2 without murmurs or gallops. The left
lower extremity was warm, swollen, and erythematous with a "cord"
medial thigh. Neurologic examination showed
palpable in theTheleftCXR
showed mild cardiomegaly. Arterial blood
hemiparesis.
gas determination showed a pH of 7.48; Pa02, 60 mm Hg; PaC02,
29 mm Hg. A ventilation perfusion scan was unchanged from pre¬
vious studies. The patient was heparinized, and leg edema improved
over 4 to 5 days. The TTE showed RV overload. At right heart
catheterization, passage of the catheter tip across the interatrial
septum suggested a PFO or a very small ASD. The mean PAP, 44
mm Hg, and right ventricular end diastolic pressures, 10 to 12 mm
Hg, were elevated. Hemoglobin oxygen saturations (95%, LA; 65%,
RA) did not suggest shunt. Pulmonary arteriogram showed multi¬
ple bilateral defects indicating previous embolization with partial
recanalization.
550
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
Reviews
Figure 3. Top, A two-dimensional image
taken from the transesophageal echocardio¬
gram TEE in case 2 demonstrates saline con¬
trast filling the RA with microbubbles in
transit across the PFO (arrow) to the LA.
Bottom, The two-dimensional color Doppler
image demonstrates blood flow from the RA
through the PFO (arrow) to the LA.
After 2 weeks of anticoagulation, he underwent a macroaggrea right-to-left shunt. Sub¬
gated albumin cardiac study confirming and
on
sequently, a caval umbrella was inserted he was discharged
a regimen of warfarin. It was concluded that the patient had had
PDEs from the venous system to his cerebral circulation and also
to his bowel 2 years previously. Six years later when hospitalized for
pneumonia, the patient developed small-bowel obstruction and
died. At autopsy, a PFO was found.
Discussion
The courses of these patients exemplify the clinical
presentations of PDE and dilemmas in its manage¬
ment. Four elements must be present for PDE: (1)
systemic embolism confirmed by clinical, angiographic,
or pathologic findings without an apparent source in
the left area of the heart or proximal arterial tree; (2)
an embolic source within the venous system; (3) an
abnormal intracardiac or intrapulmonary communica¬
tion between right and left circulations; and (4) a
pressure gradient that promotes right-to-left shunting
cardiac cycle. In 1951, Johnson19
"presumptive" in the presence of
the first three findings and "proved" if thrombus is
lodged within an intracardiac septal defect at autopsy.
Sources of Paradoxical Embolus
Some restrict their definition of PDE to include only
emboli due to thrombotic material,4'5'20,21 whereas
others include emboli of any kind.6 Most reports cite
venous thrombi from branches of the IVC as the ma¬
jor source. In Loscalzo's review,5 20 patients (66%) had
PE diagnosed by high-probability ventilation per¬
fusion scans, angiograms, or both, and 13 (43%)
had DVT documented at venography. Recently, Stollberger et al17 reported that 19 of 29 patients (66%)
with PFO and arterial embolism (without evident intraarterial or cardiac sources) had DVT at venography;
two thirds of the DVT were clinically silent. Thrombogenesis associated with Swan-Ganz catheters has
at some point in the
categorized PDE
as
CHEST /108 / 2 / AUGUST, 1995
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
551
been well documented.22 At least two cases6,23 of PDE
to a coronary artery have occurred during cardiac
catheterization, with thrombus forming along the
venous catheter.
Unusual PDE have included two cases of fat
embolism after long bone fracture,24,25 brain tissue
tumor emboli from a ma¬
complicating birth injury,26
teratoma,2 and the current case of AFE.
lignant
Definitive diagnosis of the latter rare obstetric com¬
fetal squamous
plication is based oronvernix
demonstrating
in the pulmonary vascula¬
cells, mucin, hair,
ture at autopsy.27 Extrapulmonary AFE have occurred
in capillaries of the heart, kidney, brain,
previously
small intestine, liver, spleen, pancreas, adrenal glands,
gallbladder, and the retina,28"32 attributed primarily to
passage of fetal debris via the pulmonary vascular bed.
We suspect that initial increased right-sided pressures
due to pulmonary AFE predisposed to PDE via a PFO
in
case
ARTERIAL EMBOLUS OF UNCERTAIN CAUSE
Stroke
orTIA
Peripheral
Embolus
TTE,
TTE,
Carotid
Ultrasound
Peripheral Arteriogram, and
Abdominal Aortic Ultrasound
Diagnostic
Diagnostic
Diagnostic
1.
Eleven cases of paradoxical air embolism have been
Accidental introduction of air may com¬
reported.6,33
intravenous injection, hemodialysis, induction
plicate
of pneumothorax owing to central line placement,
trauma, or any surgical procedure where the wound is
above the heart (most often during craniotomies per¬
formed with patients in the sitting position). Compli¬
cations include decreased right heart filling (potentially
with cardiac arrest), noncardiogenic pulmonary edema,
and PEs. With the latter, increased mean PAP and
right-heart pressures might predispose to PDE.34 The
true incidence of PDE due to air is unknown, and it is
controversial whether PDE occurs primarily via the
circulation or owing to an intracardiac de¬
pulmonary
fect. Among 11 case reports, 5 had intracardiac
defects.6,33
Cardiac and Vascular Communications
Most PDEs have been associated with a PFO, a
finding in all inourwhom
patients. Among 25 cases diagnosed
the interatrial defect was char¬
antemortem
acterized, 18 (72%) had a PFO, 3 had an ASD, 3 had
a pulmonary arteriovenous malformation, and one had
a ventricular septal defect.5 There also has been a case
with Ebstein's anomaly35 and patent truncus arteriosus.23 The PFO has been known to be a very common
since 1930, when Thompson and Evans3
finding
identified a "pencil patent" defect (0.6 to 1.0 cm in
diameter) in 6% of unselected autopsies and a "probe
patent" foramen (0.2 to 0.5 cm) in 29%. Although most
PDE are associated with the former, some have
occurred with a PFO smaller than 0.6 cm.2 Moreover,
PFOs have allowed passage of massive emboli that in
three cases resulted in aortic occlusion.36 Hagen et al37
found PFO in 27.3% of 965 autopsied hearts. The in¬
cidence and size of PFO did not differ with gender but
varied significantly with age: 34.3% occurred in the
Diagnostic
Non-Diagnostic
TEE
Contraindicated
Figure 4. Potential strategy for diagnosis of patient with arterial
embolus of uncertain cause. TIA=transient ischemic attack.
PARADOXICAL ARTERIAL EMBOLUS
? Candidate for Anticoagulation
IVC Filter
Placement
Anticoagulation Therapy
?
Impending PDE by Echocardiogram
JL
Yes
Consider
Intracardiac
?
Peripheral Embolus
with Limb Viability
in Jeopardy
No
?
Right Atrial Hypertensionand Acute PE
Embolectomy, Defect
Repair and IVC
Filter Placement
Consider
Thrombolytics
and/or
Embolectomy
Consider
Thrombolytic Therapy
or Pulmonary
Embolectomy if
if No Contraindications
Indicated
No:
Eisenmenger's Complex,
Other Inoperable Intracardiac Defect,
Unresectable Pulmonary AVM,
or COPD with Chronic Pulmonary HTN
PFO Present:
No Further Therapy
IVC Filter Placement
and
Chronic Anticoagulation
1.5:1 Pulmonary to
Systemic Bloodflow
ASD Present with
Surgical Repair
after Resolution of PE
Figure 5. Potential strategy for management of patients with ar¬
terial PDE. AVM=arteriovenous malformation; HTN= hyperten-
552
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
Reviews
first three decades, 25.4% in the 4th to 8th decades,
and 20.2% in the 9th and 10th decades. The PFO
to 19 mm (4.9 mm, mean) and increased
ranged from 1from
a mean of 3.4 mm in the first de¬
progressively
mm
in
the 10th, perhaps because smaller
cade to 5.8
defects seal with age. Among 30 patients diagnosed
life, Loscalzo5 found the mean age to be
during
44.3±16.1 years, with a 2:1 female-to-male predomi¬
nance. The precise frequency of PDE complicating
PFO is unknown. The PDE occurs in a minority of the
27 to 35% of patients with venous thromboembolic
disease who also have a PFO, because normally the
foramen ovale is closed by the left atrial pressure-right
atrial pressure (RAP) gradient.37
The ASDs are the most common form of congeni¬
tal heart disease in adults, comprising one third of
cases. The direction and severity of shunt depends on
ASD size, left ventricular and right ventricular com¬
diastole, and the ratio of pulmonarypliance during
de¬
to-systemic vascular resistance. Size1 is 2not a major
terminant for defects more than to cm.2 Patients
with thromboembolism and ASD have a small but
definite risk of PDE, depending on the degree of
shunt.38 In patients with left-to-right shunt,
right-to-left
dilutional studies also have shown small right-to-left
shunt due to streaming of venous blood from the IVC
toward the fossa ovalis and across the defect.39 Even in
the absence of ventricular hypertension, the pressure
gradient may reverse transiently before or during atrial
systole.40
Hunter41 first
hypothesized that PDE would be fa¬
cilitated by shunting through the low resistance com¬
munication of a pulmonary arteriovenous malforma¬
tion. The latter are very uncommon, with only 3 PDE
reported.5 Thrombosis within the fistula is common
and 10% are reported to cause bland and septic cere¬
bral emboli.42 However, there is a 37% incidence of
abnormalities ranging from mild sensory
neurologic
and motor deficits to severe hemiplegia, often ascribed
to polycythemia and hypoxia.43
Mechanisms of Right-to-Leji Pressure Gradients
The PE is the most common cause of acutely
elevated RAP and right-to-left shunt in patients with
PFO or ASD, and occurs in at least 60% of PDE (5).
The frequency with which PDE occurs in patients with
PE is unknown. In the investigation of Sharma and
concerning acute hemodynamic alter¬
colleagues44
ations after PE, PAP rose in proportion to the extent
of pulmonary vascular obstruction in 21 patients with¬
out preexisting pulmonary vascular disease. Obstruc¬
tion of 25 to 30% of the pulmonary vasculature was
associated with pulmonary hypertension, although no
patient had a mean PAP over 40 mm Hg despite mas¬
sive (>50%) embolic obstruction. The RAP was ele¬
vated only occasionally for mean PAP less than 30 mm
Hg, but it rose consistently for mean PAP more than
30 mm Hg. Similar observations have been made by
other investigators.45"47 Such data suggest that in
patients without preexisting pulmonary vascular dis¬
ease, acute PE may result in PDE if mean PAP
increases to at least 30 mm Hg and there is a 35 to 40%
vascular obstruction and an intracardiac
pulmonary
defect such as an ASD or PFO.
When PDE occurs in the absence of PE, chronic
lung disease with pulmonary hypertension has usually
been present.4 The frequency of PDE in this setting is
unknown. In the report of Sharma et al44 on 30 patients
with chronic pulmonary vascular disease and acute PE,
mean PAP of 40 mm Hg was seen, reflecting preex¬
isting RV hypertrophy. Also, acute elevations of RAP
in a Valsalva maneuver or cough have been associated
with 15% of PDE cases.5 On release of the Valsalva
maneuver, RAP momentarily exceeds left atrial pres¬
sure due to a sudden rush of blood into the RV and has
been useful in diagnosing intracardiac defects. In some
circumstances, superimposed acute respiratory illness
and its treatment may accentuate the potential hazard
of PFO: patients with acute respiratory failure who
receive positive end expiratory pressure may experi¬
ence
shunting across a PFO. Cujec et al48 reported
these effects of positive end-expiratory pressure in
patients with acute respiratory failure and compared
patients with PFO with those without PFO. The TEE
showed increased right-to-left shunt in 28% of 46 pa¬
tients; nearly half of these individuals had a PFO.
Other causes of elevated RAP facilitating right-to-left
shunt include idiopathic pulmonaryhypertension, pul¬
monary valve stenosis, congestive heart failure, after¬
math of an RV infarction, cardiopulmonary bypass, air
embolism, and platypnea orthodeoxia.49 Stroke and
other neurologic deficits are common in patients with
COPD; the role of PDE in those with pulmonary hy¬
pertension is unknown.
Sequelae of Paradoxical Embolus
As seen in our patients, PDE can have catastrophic
review of 30 patients by Loscalzo,5
sequelae. In theemboli
included 20 peripheral (49%),
sites of arterial
15 cerebral (37%), 4 coronary (9%), 1 renal (1%), and
1 splenic (1%). In addition, PDE have been reported
to cause brain abscess.43 The relationship of PDE to
embolic stroke, present in two of our patients, merits
particular attention. Forty percent of cerebral infarcts
are
of undetermined
cause
despite comprehensive
evaluation.50 The PDE via a PFO has been suggested
persons. In 1988, Webster
potential causehalfinofsuch
40 stroke patients less than 40
found that
years of age had a PFO, in contrast to only 15% of
control subjects. Among 34 patients with cryptogenic
56%
PFO. Lechat et al15 found PFO in
as a
et al
stroke,
had
CHEST /108 / 2 / AUGUST, 1995
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
553
Table 1.Prevalence of Patent Foramen Ovale
(by Transthoracic Echocardiogram With Contrast and Valsalva
Maneuver) in Stroke Patients
Variable
Investigation Reference
I-
DiTullio et al16
Webster et al14
Age, yr (x±SD)
<40
Patients, No.
40
20/40 (50)
1/6 (17)
NI*
19/34 (56)
6/40 (15)
Overall, No. (%)
Stroke, determined origin
Stroke, cryptogenic controls with risk factors for stroke
Stroke, cryptogenic
Control subjects
(31±6)
Lechat et al15
<55(36±10)
60
24/60 (40)
4/19 (21)
6/15 (40)f
14/26 (56)
10/100 (10)
45
11/45 (24)
1/24 (4)
NI*
(48)
10/21
101
15/101 (15)
6/77 (8)
NI*
9/24 (38)
*NI=not identified.
*Risk factors in this study included mitral valve prolapse, migraine, use of contraceptives. None presented with hypertension or diabetes.
+Patent foramen ovale was inversely associated with the presence of hypertension and diabetes, but the number of cryptogenic stroke patients with
combination of risk factors and PFO was not specifically provided.
40% of 60 patients aged less than 55 years with stroke
10% of control subjects. In the first PFO prevalence
report in older persons with stroke, Di Tullio et al16
evaluated 146 stroke patients of all ages with contrast
TTE and found PFO in 18%. Thirty-one percent of
vs
cryptogenic; these were associated with
higher
frequencies of PFO in both younger (<55 years;
48 vs 4%; p<0.001) and older (>55 years; 38 vs 8%;
p<0.001) patients when compared with those with
strokes were
identifiable causes of stroke. The differences remained
significant after controlling for hypertension and dia¬
betes. These associations of PFO with cryptogenic
stroke suggest that a larger proportion of such events
maybe secondary to PDE than is generally appreciated
(Table 1).
Paradoxical coronary embolism is a rare diagnosis
first reported by Marchand in 1894.51 In the recent
review of 27 cases by Jungbluth et al,6 9 were diagnosed
antemortem. Of the 18 thrombotic PDE, the rightto-left communication was via PFO in 13, an ASD in
4, and a patent truncus arteriosus in 1. The remaining
cases were caused by air embolism, and an
eight
abnormal intracardiac defect was found in only three.
Five air emboli involved the left coronary artery, per¬
haps related to its anterocranial position.
Diagnosis of Paradoxical Embolism
potentially devastating effects of
PDE, early diagnosis and treatment are essential in
order to manage the current event and to prevent re¬
currence. The high prevalence of anatomic (eg, PFO)
and hemodynamic (eg, pulmonary hypertension due to
acute or chronic respiratory illness or both) predispo¬
sitions to PDE present unresolved diagnostic and
dilemmas. The PDE should be suspected
therapeutic
in any patient with unexplained arterial embolism. Al¬
obvious, simultaneous pulmonary and
though clinically
embolism
should strongly suggest PDE, but
systemic
it is not prerequisite for the diagnosis. The PE may go
Because of the
DVTs are often
undiagnosed, and as noted previously,
also
is
hindered
asymptomatic.17,52 Diagnosis
by the
variable manifestations of systemic emboli, which also
maybe asymptomatic,53 and the presumptive nature of
the diagnosis of PDE when it occurs. Moreover,
shunt may be transient and occur only
right-to-left
Valsalva or other inciting maneuvers; if the pa¬
during
tient is studied after the acute event, both the PE and
the right-to-left shunt may have resolved.9 The weight
to assign to coexisting carotid disease in the decision
whether or not to pursue PDE as a diagnostic consid¬
eration is also unclear.
The differential diagnosis of arterial embolism in
patients with temporally associated PE or DVT in¬
cludes (1) cardiomyopathy with right ventricular and
left ventricular mural thrombi; (2) emboli from biatrial
myxomas; (3) mitral and tricuspid valve disease (with
or without atrial fibrillation); (4) myocardial infarction
with a left ventricular thrombus and congestive heart
failure predisposing to DVT; (5) emboli via right- and
left-sided infective endocarditis; and (6) emboli related
to prosthetic replacement of both mitral (and/or aor¬
tic) and tricuspid valves.9 In patients with cyanotic
heart disease and arterial embolism, PDE is
congenital
in
other settings, the diagnosis may be difficult
likely;
to establish. Because of the importance of DVT as or¬
igin of PDE, vigorous exclusion of DVT (eg, with
contrast venography) assumes special importance in
patients with unexplained stroke or other embolic
events. Stroke patients with COPD and pulmonary
have vigorous assessment for
hypertension also shouldcommunication
and venous
coexisting intracardiac
disease.
Diagnostic
Techniques
Before availability of contrast TTE and TEE, many
were used to demonstrate right-to-left
approaches
shunt. Following the use of oximetry to document a
drop in oxygen saturation during the Valsalva maneu-
554
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
Reviews
ver, in
cators
the study by Lee and Gilmette,54 various indi¬
including ascorbate, hydrogen, krypton, freon,
and indocyanine green were injected into the right side
of the circulation, with their early systemic appearance
signifying a right-to-left shunt.4'21,55,56 Accurate antecon¬
diagnosis has beenSixfacilitated
greatly byPDE
of
cases
echocardiography.57
impending
(with the clot visualized within the intracardiac defect),
"embolism interruptus," have been reported during
life, three each by TTE5,7 and TEE.913
The contrast effect is achieved by microcavitation
bubbles produced during rapid injection of 10 mL of
agitated contrast solution (eg, 9 mL of saline and 1 mL
of air) into a peripheral vein. Various contrast agents
may be used, including indocyanine green, dextrose
water, normal saline solution, or autologous blood.
Usually, contrast TTE is obtained using an apical
four-chamber view. At least four saline injections are
in quiet breathing and after the Valsalva
performed
maneuver. The study is considered positive if microbubbles appear either in the LA or left ventricle no
later than two to three cycles after their initial appear¬
ance in the RA.58 Another maneuver, the cough test,
has been reported to be superior to the Valsalva ma¬
neuver in diagnosis of PFO in TEE.59 Three to five
rapid, successive coughs are performed immediately
after opacification of the RA with contrast. In 73 con¬
secutive patients, PFO was detected during the cough
test in 32 (43.8%), significantly more often than in the
Valsalva maneuver (32.9%) or quiet breathing (24.7%).
Contrast TTE is more sensitive than oximetry and
dye curves, but its sensitivity is only 64%.60 Technical
limitations relate to spontaneous variations in right-toleft shunt during normal breathing, the quality of the
Valsalva maneuver, uneven distribution of contrast
solution, and missing the sampling site of the M-mode
beam. Contrast shunting is apparent immediately in
patients with cyanotic heart defects but is often subtle
in patients with uncomplicated ASD. Contrast shunt¬
ing may become evident during the Valsalva maneuver,
but this part of the tracing often is technically unsat¬
and difficult to interpret. The most important
isfactory
moment is thus the first diastole following the Valsalva
maneuver release. The TTE also frequently demon¬
strates echo dropouts because the ultrasonic beams do
not generally hit the interatrial septum with a perpen¬
dicular orientation and may result in false-positive ap¬
pearances of ASDs. Since the degree of contrast shunt
is independent of defect size, TTE cannot reliably
differentiate ASD from PFO.60 By contrast, TEE is
unrestricted by lung tissue or thoracic malformations,
providing
superior imaging of the atrial septum and a
clear view of the foramen ovale.61
The prevalence of PFO detected by TEE has ranged
from 8 to 63% because of varying study populations,
techniques, and defining criteria.14,58"64 In the first
mortem
trast
report validated by catheterization or operative results,
Chen and coworkers58 documented PFOs with con¬
trast TTE and TEE in 32 patients. A right-to-left shunt
was seen by TEE in 14 in normal breathing and 20 with
the Valsalva maneuver and by TTE in 8 with normal
and 12 with the Valsalva maneuver. All PFOs
breathing
seen by TTE were also seen by TEE, and all but one
of the PFOs seen by TEE were confirmed by cathe¬
terization or surgery. The false-positive patient had a
right atrial myxoma, the highest RAP of the group;
shunt was accentuated by
presumably, right-to-left
"communications other than PFO." The diagnostic
sensitivity (100 vs 63%) and accuracy (97 vs 78%) of
TEE were higher than with TTE. Although such ini¬
tial observations suggest that TEE provides superior
imaging, even in the strain phase, and higher sensitiv¬
ity in the detecting PFO, TEE has the disadvantages
of semiinvasiveness, higher cost, and limited availabil-
ity.
Transcranial Doppler (TCD) ultrasound of the
middle cerebral artery in contrast injection is another
noninvasive method that may enhance detection of
interatrial right-to-left shunt.65 Nemec et al66 com¬
TTE, TEE, and TCD in 32 patients, 21 ofwhom
pared
had unexplained arterial emboli; TEE showed intera¬
trial shunt in 13 and intrapulmonary shunt in 6. Using
TEE as the gold standard, TCD had a sensitivity and
of 100% and identified 3 of the intrapul¬
specificityshunts.
Thus, TCD provides a useful alterna¬
monary
tive for excluding interatrial right-to-left shunt when
TEE is unavailable. This role of TCD warrants further
investigation because of its wide availability.66,67
Diagnostic Approaches
Evaluation
Integrated use of these tools in diagnostic evaluation
of an arterial embolus varies considerably with insti¬
tutional resources and clinicians' preferences. Because
arterial emboli most often originate in the heart or
arteries, evaluation often begins with Mproximal
mode two-dimensional TTE, arteriogram or abdomi¬
nal aortic ultrasound (for abdominal or lower extrem¬
ity peripheral emboli),59 or carotid ultrasound (for
stroke or transient ischemic attack [Fig 4]). If no source
is found, or if DVT or PE or both are associated, then
contrast TTE with Valsalva maneuver or cough test or
contrast TEE
rule out
(if TTE is negative)
are
performed to
shunt. If TEE cannot be per¬
right-to-left
These approaches focus¬
TCD
is
considered.
formed,
and
the
embolus
possible intracardiac com¬
ing upon
munications should be coordinated with vigorous
exclusion of associated venous disease. The optimum,
most cost-effective means of detecting PFO requires
further clarification.
CHEST /108 / 2 / AUGUST, 1995
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
555
Management Issues and Challenges
Various therapeutic strategies have been used in
patients with PDE (Fig 5). A multifaceted approach is
necessary since no single algorithm is appropriate for
all individuals; the heterogeneity and complexity of
patients, together with their acuity of illness, makes
management highly individualized. There are no pub¬
lished investigations comparing patient outcomes or
showing the superiority of any one strategy. Most au¬
thorities5,9,36 agree that anticoagulation should be ini¬
tiated immediately on diagnosis of arterial emboli un¬
less there are major contraindications. Interestingly,
strategies for long-term management
anticoagulation
PDE
or for these diverse issues related
of patients with
PDE
are
seldom addressed. Although
to
specifically
and
risk of recurrent emboliza¬
the long-term courses
PDE
are
tion in patients with
unknown, it seems rea¬
should
have lifelong antico¬
that
such
sonable
patients
in
of
the
absence
contraindications. Even
agulation
to
difficult
is
the
more
approach persons with DVT and
a documented intracardiac communication or that to
patients with other combinations of predispositions to
PDE (eg, PFO and pulmonary hypertension). Prophy¬
lactic anticoagulation appears reasonable in such indi¬
viduals but data documenting its benefit are unavail¬
able. In this regard, the risk posed by echocardiodetected PFO in the absence of PDE
graphicallydelineation.
Without the benefit of large
requires
clinical trials, it is unclear whether long-term antico¬
would benefit stroke patients with
therapy
agulant
PFO and no cardiac source of embolism, particularly
in the presence of COPD and pulmonary hyperten¬
sion.
Because PE is a major problem likely to accompany
PDE, its management is another important related is¬
In the special circumstance ofcombined PDE and
PE, the greatest danger of further PDE seems to oc¬
cur in the few hours after the initial episode, when RAP
is highest. Thus, thrombolytic agents might be con¬
sue.
sidered in carefully selected patients with suspected
PDE and significant RA hypertension due to PE even
when systemic hypotension is absent, in an effort to
reduce RAP and the likelihood of recurrent PDE.5,36
Evidence ofcerebral emboli may be a contraindication
to thrombolytic therapy, although use of lytic therapy
for such cases has been reported.62 Thrombolytics and,
pe¬
occasionally, embolectomy are consideredlimbwhen
ripheral arterial obstruction jeopardizes viabili¬
ty.36The role of caval
interruption in patients with PDE
are
If
unclear.
is also
thrombolytics or ofanticoagulation
filter
is
an
IVC
then
contraindicated,
placementof DVT.
in
the
recommended
Occasionally,
presence
considered. In the rare
embolectomy isPDE
pulmonary
discovered at echo¬
case of an "impending"
intracardiac
embolectomy may be at¬
cardiography,
tempted,
together with correction of the intracardiac
defect and placement of an IVC filter.5 If PDE occurs
via an unresectable pulmonary arteriovenous malfor¬
mation or in the presence of inoperable, irreversible
intracardiac right-to-left shunt, then an IVC filter is
considered. Long-term anticoagulant administration
as well because of the po¬
generally is recommended
effects
of even very small PDE.36
tentially devastating
of
the
is unnecessary after PDE,
closure
PFO
Usually, if RA
particularly to be hypertension and right-to-leftofshunt
are thought
temporary.36 Surgical repair iso¬
lated ASD has been advocated when the ratio of pul¬
monary to systemic blood flow is 1.5:1. In persons with
Eisenmenger's complex (irreversibly elevated mean
PAP and right-to-left shunt), surgery is contraindicated
because the severe pulmonary hypertension causes
heart failure upon closure of the ASD.68 If an ASD
patient has a right-to-left shunt due to an acute PE, RA
may be self-limited and surgical repair
hypertension
still be
may
possible.36
Comment
of PDE have been
descriptive, pro¬
reports
information about its natural history, out¬
viding little
comes, and effects of various therapeutic approaches.
The long-term management of survivors of the acute
PDF is unclear from the literature. More epidemio¬
logic data are needed to delineate these issues. The 27
to 35% prevalence of PFO in the general popula¬
tion,3,37 and the association of cryptogenic strokes with
PFOs14"16 suggests that PDE is frequently missed. This
impression is supported by the review ofAbuRahma et
had a
al9 of 41 cases of arterial embolism: 20
Most
(49%)
cardiac thrombus or other cardiac disease, 7 (17%) had
a proximal arterial thrombus by arteriogram or ab¬
dominal aortic ultrasound, 5 (12%) had "possible" or
"probable" PDE, and another 9 individuals (22%) had
emboli of unknown source after negative or incom¬
plete evaluations. Five of 9 patients (56%) less than 50
years old had "probable" or "possible" PDE.
To improve recognition of PDE, clinicians should
maintain a high suspicion for this event whenever
neurologic
changes complicate cardiovascular events,
DVT or PE, or any unexplained arterial occlusion oc¬
curs (particularly in young or postoperative patients).
When the source of an embolus is not identified by
M-mode and two-dimensional echocardiogram, ab¬
dominal ultrasound, or arteriogram, contrast TTE with
cough test or Valsalva maneuver should be considered,
followed by a TEE or TCD if necessary. The most
cost-effective diagnostic strategy remains to be deter¬
mined, and approaches will vary with the institutional
resources available to the clinician. Anticoagulation is
the mainstay of therapy. Unless contraindicated, a
lower threshold for use of thrombolytics may be con¬
sidered in selected cases. It seems reasonable that an-
556
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
Reviews
for patients predisposed to PDE should
ticoagulation
continue as long as the predisposition is present. For
many patients (eg, individuals with COPD-induced
hypertension, PFO) this recommendation
pulmonary
would be lifelong. The benefits of this approach are
undocumented and require definition before making
firm guidelines.
Prevention of PDE is especially desirable. Although
there is no information regarding specific measures to
reduce PDE risk, improved application of already val¬
idated prophylaxis when predispositions to DVT occur
in patients with known PFO or other intracardiac
communications would appear especially worthwhile.
Even less is known if prophylactic anticoagulation of
persons with recognized risks for PDE (eg, PFO and
chronic pulmonary hypertension) is beneficial. If fu¬
ture reports confirm that PDE occurs as often as the
might suggest, then
high frequency of predispositions
of
current
thresholds
and antico¬
major adjustments
In
will
be
the
absence of
necessary.
agulation strategies
contraindications, lower thresholds for anticoagulating
patients at risk might be needed, but documentation of
benefits of this approach may be difficult.
References
J. Thrombose und Embolie. In: Vorlesungen iiber
Allgemeine Pathologie, vol 1. Berlin: Hirschwald, 1877; 134
2 Dahl-Iverson E. Embolie paradoxale de l'artere iliaque externe
gauche-embolectomie. Lyon Chir 1930; 1:39-42
3 Thompson T, Evans W. Paradoxical embolism. Quart J Med 1930;
1 Cohnheim
23:135-0.
4 Meister SG, Grossman W, Dexter L, et al. Paradoxical embo¬
lism.diagnosis during life. Am J Med 1972; 53:292-98
5 Loscalzo J. Paradoxical embolism: clinical presentation, diagnos¬
tic strategies, and therapeutic options. Am Heart J 1986; 112:
141-45
6 Jungbluth A, Erbel R, Darius H, et al. Paradoxical coronary em¬
bolism: case report and review ofliterature. Am Heart J1988; 116:
879-85
7 Spechly-Dick ME, Middleton SJ, Foale RA. Impending para¬
doxical embolism: a rare but important diagnosis. Br Heart J 1991;
65:163-65
8 Nelson CW, Snow FR, Barnett M, et al. Impending paradoxical
embolism: echocardiographic diagnosis of an intracardiac throm¬
bus crossing a patent foramen ovale. Am Heart J 1991; 122:859-62
9 AbuRahma AF, Lucente FC, Boland JP. Paradoxical embolism:
an underestimated entity.a plea for comprehensive workup. J
Cardiovasc Surg 1990; 31:685-92
10 Movsowitz C, Podolsky LA, Meyerowitz CB, et al. Patent foramen
ovale: a nonfunctional embryological remnant or a potential cause
of significant pathology? J Am Soc Echocardiologists 1992;
5:259-70
11 Nellessen U, Daniel WG, Matheis G, et al.
14 Webster MW, Chancellor AM, Smith HJ, et al. Patient foramen
ovale in young stroke patients. Lancet 1988; 2:11-12
15 Lechat P, Mas JL, Lascault G, et al. Prevalence of patent foramen
ovale in patients with stroke. N Engl J Med 1988; 318:1148-52
16 Di Tullio M, Sacco RL, Gopal A, et al. Patient foramen ovale as
a risk factor for cryptogenic stroke. Ann Intern Med 1992; 117:
461-65
17 Stollberger C, Slany J, Schuster I, et al. The prevalence of deep
thrombosis in patients with suspected paradoxical embo¬
lism. Ann Intern Med 1993; 119:461-65
18 Prahlow JA, Lantz PE. Amniotic fluid embolism with paradoxi¬
cal embolization. Forensic Pathol 1993; 35:1-4
19 Johnson BJ. Paradoxical embolism. J Clin Pathol 1951; 4:316-32
20 Silverman ME. Paradoxical embolus. N Engl J Med 1993;
venous
329:930
Cheng TO. Paradoxical embolism.a diagnostic challenge and its
detection during life. Circulation 1976; 53:565-68
22 Hoar PF, Stone JG, Wickes AE, et al. Thrombogenesis associated
with Swan-Ganz catheters. Anesthesiology 1978; 48:445-47
23 From AHL, Wang Y, Eliot RS, et al. Coronary arterial embolism
21
following cardiac
24 Pell ACH, Hughes D, Keating J, et al. Brief report: fulminating
fat embolism syndrome caused by paradoxical embolism through
a patent foramen ovale. N Engl J Med 1993; 329:926-29
25 Etchells
in persistent truncus arteriosus, report of a case
catheterization. N Engl J Med 1965; 272:1204
EE, Wong DT, Davidson G, et al. Fatal cerebral fat
embolism associated with a patent foramen ovale. Chest 1993;
104:962-63
26 Abrikossoff
AJ. Zur kasuistik der Parenchymembolien: Kleinhirngewebeembolie der arteria coronaria cordis beim neugeborenen. Zentrabl Allg Pathol 1913; 23:244
27 Clark SL, Pavlova Z, Greenspoon J, et al. Squamous cells in the
maternal pulmonary artery circulation. Am J Obstet Gynecol
1986; 154:104-06
28 Liban E, Raz S. A clinicopathologic study of fourteen cases of
amniotic fluid embolism. Am
J Clin Pathol 1969; 51:477-86
29 Price TM, Baker W, Cefalo RL. Amniotic fluid embolism: three
case reports with a review of the literature. Obstet Gynecol Surv
1985; 40:462-75
30 Chang M, Herbert WNP. Retinal arteriolar occlusions following
Ophthalmology 1984; 91:1634-37
EP, Taylor HB. Amniotic fluid embolism. Obstet
Gynecol 1970; 35:787-93
amniotic fluid embolism.
31 Peterson
32 Attwood HD. Fatal pulmonary embolism by amniotic fluid. J Clin
Pathol 1956; 9:38-46
33 Marquez J, Sladen A, Gendell H, et al. Paradoxical cerebral air
embolism without an intracardiac defect. J Neurosurg 1981; 55:
997-1000
34
Brogno D, Lancaster G, Rosenbaum M. Embolism interruptus.
N Engl J Med 1994; 330:1761-62
1982;
JL, Pennington WS, Isobe JH, et al. Paradoxical
embolization with Ebstein's anomaly. Arch Surg 1983; 118:1101
Leonard RCF, Neville E, Hall RJC. Paradoxical embolism: a re¬
view of cases diagnosed during life. Eur Heart J 1982; 3:362-70
Hagen PT, Scholz DG, Edwards WD. Incidence and size of
patent foramen ovale during the first 10 decades oflife: an autopsy
study of 965 normal hearts. Mayo Clin Proc 1984; 59:17-20
Kopf GS, Laks H. Atrial septal defects and cor triatriatum. In:
Baue AE, ed. Glenn's thoracic and cardiovascular surgery. 5th ed,
vol 2. East Norwalk, Conn: Appleton and Lange, 1991; 995-1005
Swan HJC, Burchell H, Wood E. The presence of venoarterial
shunts in patients with interatrial communications. Circulation
35 Mathews
36
37
38
39
1954; 10:705-13
40
1991; 121:1552-54
13
Venous air embolism. Arch Intern Med
142:2173-76
Impending paradox¬
ical embolism from atrial thrombus: correct diagnosis by transe¬
sophageal echocardiography and prevention by surgery. J Am Coll
Cardiol 1985; 5:1002-04
12 Nagelhout DA, Pearson AC, Labovitz AJ. Diagnosis of paradox¬
ical embolism by transesophageal echocardiography. Am Heart J
O'Quin RJ.
41
Steiger BW, Libanoff AJ, Springer EB. Myocardial infarction due
to paradoxical embolism. Am J Med 1969; 47:995-98
Hunter DD. Pulmonary arteriovenous malformation: an unusual
case of cerebral embolism. Can Med Assoc J 1965; 93:662-65
CHEST /108 / 2 / AUGUST, 1995
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
557
Pulmonary hypertension and cor pulmonale. In:
Fishman AP, ed. Pulmonary diseases and disorders. 2nd ed, vol
2. New York: McGraw-Hill, 1988; 1043-45
43 Sisel RJ, Parker BM, Bahl OP. Cerebral symptoms in pulmonary
arteriovenous fistula. Circulation 1970; 41:123-28
44 Sharma GVRK, Mclntyre KM, Sharma S, et al. Clinical and he¬
modynamic correlates in pulmonary embolism. Clin Chest Med
42 Fishman AP.
1984; 5:421-37
45 Dalen JE, Banas JS, Brooks HL, et al. Resolution rate of acute
pulmonary embolism in man. N Engl J Med 1969; 280:1194-99
46 Mclntyre KM, Sasahara AA. The hemodynamic response to
pulmonary embolism in patients without prior cardiopulmonary
disease. Am J Cardiol 1971; 28:288-94
47 Jardin F, Duboura O, Gueret P, et al. Quantitative two-dimen¬
sional echocardiography in massive pulmonary embolism: em¬
phasis on ventricular interdependence and lefthand septal dis¬
placement. J Am Coll Cardiol 1987; 10:1201-06
48 Cujec B, Polasek P, Mayers I, et al. Positive end-expiratory pres¬
sure increases the right-to-left shunt in mechanically ventilated
patients with patent foramen ovale.
Ann Intern Med
1993;
119:887-94
49 Seward JB,
50
Hayes DL, Smith HC, et al. Platypnea-orthodeoxia:
clinical profile, diagnostic workup, management, and report of
seven cases. Mayo Clin Proc 1984; 59:221-31
Sacco RL, Ellenberg JH, Mohr JP, et al. Infarcts of undetermined
cause:
the NINCDS Stroke Data Bank. Ann Neurol 1989; 25:
382-90
51 Marchand F. Zur Kennthis der embolie und thrombose der Ge-
hirnarterien, zugleich
ein beitrag zur casuistik der primaren
herztumoren und der gekreuzten embolien. Berl Klin Wochenschr 1894; 31:36
52 Moser KM. Venous thromboembolism. Am Rev Respir Dis 1990;
141:235-49
53 Edwards EA, Tilney N, Lindquist RR. Causes of peripheral em¬
bolism and their significance. JAMA 1966; 196:119-24
54 Lee G, Gimlette T. A simple test for interatrial communication.
BMJ 1957; 1:1278-81
55 Tracy GP, Smulyan H, Potts JL, et al. One angiographic diagno¬
sis of patent foramen ovale in paradoxical embolism. Catheter¬
ization Cardiovasc Diagnosis 1976; 2:137-42
56 Banas J, Meister S, Gazzaniga A, et al. A
technique for
detecting small defects of the atrial septum. Am J Cardiol 1971;
28:467-71
57 Valdez-Cruz LM, Pieroni DR, Roland JMA, et al. Echocardio¬
of intracardiac right-to-left shunts following
graphic detection
vein
peripheral injections. Circulation 1976; 54:558-62
58 Chen W, Kuan P, Lien W, et al. Detection of patent foramen
ovale by contrast transesophageal echocardiography. Chest 1992;
101:1515-20
59 Stoddard MF, Keedy DL, Dawkins PR. The cough test is supe¬
rior to the Valsalva maneuver in the delineation of right-to-left
ovale during contrast
shunting through the patent foramen
Am Heart J 1993; 125:185-89
transesophageal echocardiography.
60 Kronik G, Slany J, Moesslacher H. Contrast M-mode echocardi¬
ography in diagnosis of atrial septal defect in acyanotic patients.
Circulation 1979; 59:371-78
Langenstein BA, et al. Detection of osseptal defects by transesophageal crosssectional echocardiography. Br Heart J 1983; 49:350-58
62 Wittlich N, Kramer M. Diagnosis of patent foramen ovale by
transesophageal contrast echocardiography [abstract]. Circula¬
61 Hanrath P, Schliiter M,
tium secundum atrial
tion
63 Lee
1988; 78:11-441
RJ, Bartzokis T, Yeoh T, et al. Enchanced detection of int¬
sources of cerebral emboli by transesophageal echo¬
cardiography. Stroke 1991; 22:734-39
64 Archer SL, Kvernen LR, James K, et al. Does warfarin reduce the
atrial fibrillation? a
prevalence of left atrial thrombus in chronic
double blind, placebo controlled study. Circulation 1991; 84:11693
65 Teague SM, Sharma MK. Detection of paradoxical cerebral echo
contrast embolization by transcranial doppler ultrasound. Stroke
racardiac
1991; 22:740-45
66 Nemec
JJ, Marwick TH, Lorig RJ. Comparison of transcranial
and transesophageal contrast echocardiog¬
Dopplerin ultrasound
of interatrial right-to-left shunts. Am J
the
detection
raphy
Cardiol 1991; 68:1498-1502
67 Herderschee D, Limburg M, Hijdra A, et al. Recombinant tissue
basilar artery occlusion.
plasminogen activator in two patients with
54:71-3
Neurol
1991;
Neurosurg Psychiatry
J
68 Borow KM, Karp R. Atrial septal defect: lessons from the past,
directions for the future [editorial]. N Engl J Med 1990;
323:1698-1700
simple
558
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21718/ on 05/10/2017
Reviews