Review Article Management of pulmonary embolus in pregnancy M. de Swiet Introduction

European Heart Journal (1999) 20, 1378–1385
Article No. euhj.1999.1584, available online at on
Review Article
Management of pulmonary embolus in pregnancy
M. de Swiet
Queen Charlotte’s Hospital for Women, London, U.K.
The importance of pulmonary embolus in pregnancy
cannot be exaggerated. In developed countries where
women no longer die from sepsis and haemorrhage, it is
the leading cause of maternal mortality[1,2]. The management of pulmonary embolus in pregnancy is difficult
because few internists have much experience of managing pregnant women; and few obstetricians have any
experience of pulmonary embolus. In addition there is
confusion about the safety of maternal investigations for
the fetus, particularly when these investigations involve
ionising radiation and also confusion about the effects
that any maternal therapy may have on fetal well being.
As in the non-pregnant state[3], pulmonary embolism is
both under- and over-diagnosed in pregnancy, exposing
women to the risks of further fatal pulmonary embolism
and unnecessary anticoagulation, respectively. Underdiagnosis usually occurs in the scenario of breathlessness, fever and pulmonary infiltration diagnosed as
pneumonia; over-diagnosis relates to ascribing all cases
of chest pain and collapse to pulmonary embolus.
Plethysmography has shown that veins are more distensible in pregnancy[5] and recent Doppler studies indicate
that lower limb blood flow is reduced in pregnancy[6].
These effects are likely to be due to a combination of
hormonal influences on the structure of vessel walls and
obstruction to blood flow caused by the enlarging
uterus. After delivery the effect is more marked in the
veins of the left leg than the right[7] supporting the
observation that in pregnancy deep vein thrombosis is
much more common in the left leg.
It has been known for some time that the composition
of the blood changes in pregnancy probably because of
high oestrogen levels. For example the levels of clotting
factors II, VII, VIII, X and fibrin all increase in pregnancy[8,9]; to a certain extent, this is offset by an increase
in thrombolysis[10]. However, recent interest in natural
anticoagulants has shown that pregnancy also decreases
some of their activities: specifically, activated protein
C resistance is increased (decreased activated protein C
resistance ratio) even in the absence of Factor V
Leiden[11,12] and protein S levels also decrease[13]. So it is
likely that changes in the composition of the blood in
pregnancy relate to both an increase in clotting factors
and a decrease in natural anticoagulation.
Definition and pathophysiology (see
also Risk Factors below)
Although several substances such as fat and air can
cause pulmonary embolus in pregnancy, this review
relates to pulmonary embolus caused by blood clot,
usually originating in the lower limb or pelvic veins. The
differential diagnoses with amniotic fluid embolus and
chorion carcinoma are also considered. The five- to
sixfold increased risk of pulmonary embolus due to
pregnancy[4] is due in part to changes in the composition
of the blood and in part to changes in blood flow.
Key Words: Pulmonary embolus, pregnancy, pregnancy complications, heparin, warfarin, low molecular
weight heparin.
The incidence of pulmonary embolism in pregnancy
varies between 1 per 1000 and 1 per 3000 deliveries[14–16].
This degree of variation is due, at least in part, to
differences in diagnostic methodology and to different
obstetric practices.
About 1 per 100 000 pregnant women die from pulmonary embolism[2]. The deaths are evenly distributed
between the antenatal period and the postnatal period,
conventionally taken as 6 weeks after delivery[2,15].
Within the antenatal period, thromboembolism occurs
equally within each trimester[17]. Therefore the postnatal
period is the most dangerous time in terms of deaths per
week but no period of pregnancy is without risk.
Revision submitted 20 January 1999, and accepted 8 February
Risk factors
Correspondence: Dr M. de Swiet, Queen Charlotte’s Hospital for
Women, Goldhawk Road, London W6 0XG, U.K.
Risk factors for pulmonary embolus in pregnancy
include those acting in the non-pregnant state. For
0195-668X/99/191378+08 $18.00/0
1999 The European Society of Cardiology
Table 1 Risk of thromboembolism per pregnancy (%) in patients who themselves
have had thromboembolism or who come from families with history of thromboembolism (High risk) compared to the risk in individuals found to have thrombophilia by
population screening (Low risk). (From Girling and de Swiet 1998[53]]
Low risk[22]
High risk families
Antithrombin III[54]
Protein C[56]
Protein S[56]
example, even within the reproductive age range, age is
still an important factor: the incidence of pulmonary
embolus is doubled in pregnant women aged over 35
years compared to those aged less than 35[16]; the
mortality from pulmonary embolus is nearly 100 times
greater in pregnant women aged over 40 years compared
to those age 20 to 25 years[18]. Parity is a further risk
factor, the risk increasing in those who have had more
than three pregnancies independent of age[18]; so too is
Operative delivery increases the risk of pulmonary
embolus between two and eightfold, the risk varying
depending on whether Caesarean section was an emergency or elective procedure and on the method of
anaesthesia: epidural block reduces the risk compared to
general anaesthesia[10].
Oestrogen administration to suppress lactation is
another risk factor, indeed the risk of pulmonary
embolus is the main reason why other drugs such
as bromocriptine are used if lactation has to be
Although bed rest is assumed to be a risk factor for
pulmonary embolus, the little evidence that there is
suggests that this is not the case[20]. Thrombophilia is an
important risk factor for pulmonary embolus in the
pregnant as well as in the non-pregnant state. For
example Hirsch et al.[21] found Factor V Leiden heterozygosity in 20% of cases of thromboembolism in pregnancy, whereas the prevalence in their community was
of the order of 5%. However, it is difficult to estimate the
risks for a given thrombophilia with precision, because
they vary depending on patient selection. Also figures
are usually quoted for all thromboembolism rather than
for pulmonary embolism alone.
For example, the risk for all thromboembolism in
antithrombin III deficiency varies between 12 and 60%
per pregnancy in groups of patients diagnosed because
they presented at some other time with thromboembolism. However, in patients found to have the condition
by population screening, the risk could be as low as
3%[22]. Figures for some other common thrombophilias
are given in Table 1.
Inherited and acquired thrombophilias are also
important, not only because of their interaction with
pregnancy to increase the risk of pulmonary embolism,
but also because nearly all thrombophilias are associated
with an increased risk of poor pregnancy outcome.
These adverse effects, which occur independently of
thromboembolism, vary from pre-eclampsia to miscarriage and intra-uterine death. Placental infarction is
a frequent feature so that the mechanism is likely to
involve thrombosis in the placental vessels.
Natural history and prognosis
In non-pregnant patients, many with pulmonary embolus die from the underlying cause rather than from
recurrent embolism[23]. By contrast, pregnant women
with pulmonary embolus are usually otherwise healthy.
They either die suddenly and unexpectedly or they have
an illness that lasts several days or weeks where the
diagnosis has not been considered. This emphasizes the
need for prophylaxis in some cases and above all for
correct diagnosis. Although there is much concern about
how to treat pulmonary embolus in pregnancy, failure of
treatment once instigated is uncommon.
Diagnosis and investigation
The clinical features of pulmonary embolus are no
different in pregnancy compared to the non-pregnant
state, where a recent review suggests that 90% of those
with pulmonary embolus have dyspnoea and tachypnoea i.e. the absence of these features will exclude 90%
of pulmonary emboli[3]. However, breathlessness is a
common feature of normal pregnancy. Women increase
their ventilation by about 40% in pregnancy and there is
considerable variation between women and in the same
women between pregnancies, in their awareness of this
increase in ventilation. Pregnant women often present
with breathlessness. It is difficult to know how far to go
to exclude pulmonary embolus if there are no other
clinical features. Depending on the severity of the symptom in a woman who does not present acutely, chest
Eur Heart J, Vol. 20, issue 19, September 1999
M. de Swiet
radiography and measurement of oxygen saturation at
rest and on exercise or measurement of arterial blood
gas tensions should suffice.
Basic tests (chest radiography, electrocardiography and
arterial blood gas tensions) to diagnose pulmonary
embolus and to exclude other conditions are well
known. With regard to pregnancy, it should be noted
that with modern equipment the radiation received by
the fetus from a single maternal chest radiograph (CXR)
is <10 ìGy[24], a trivial quantity compared to the
50 000 ìGy considered to be the upper limit with regard
to danger of injury for the fetus[24]. Mothers may also be
reassured by the comparison with flying: the radiation
from one CXR is about the same as the extra radiation
received in one transatlantic flight at the altitude flown
by modern jetliners.The ECG is non-specific for the
diagnosis of pulmonary embolus, even in the nonpregnant state. The lack of specificity is increased by the
changes in electrical axis that occur in normal pregnancy. However features of acute right heart strain are
common and specific in massive pulmonary embolus as
in the non-pregnant state.
Arterial blood gas values are not changed by pregnancy. However, the sample should be taken with the
patient sitting. When pregnant women lie flat, the
arterial oxygen tension falls by up to 2 kPa[25] for a
number of reasons including ventilation/perfusion
inequality and impaired venous return to the right heart.
Echocardiography is an attractive option in pregnancy because of the lack of radiation. It will show a
variety of abnormalities in patients with major central
pulmonary embolus[3] as well as exclude other causes of
collapse, in particular aortic dissection. Unfortunately
this investigation may not be available when most
needed in the emergency situation of the collapsed
patient, particularly in isolated maternity units.
Lung imaging
As in the non-pregnant state, lung imaging is mandatory
to confirm the abnormalities that may (or may not) have
been found in the non-specific tests above. Isotope
scanning is the technique that is most likely to be
available. The radiation from one perfusion scan using
Tc-labelled macroaggregates varies between 10 and
350 ìGy[24], again well within the 50 000 ìGy permitted
during pregnancy and the injected dose may be reduced
still further in pregnancy[26]. Nursing mothers should
not breast feed for the 15 h during which significant
quantities of 99mTc are secreted in their milk[3].
It is argued that ventilation scans add little diagnostic
precision[3,27]. However, even if a ventilation scan is
used, this only increases the radiation by a further
40–190 ìGy[24]. Pulmonary angiography is under-used
Eur Heart J, Vol. 20, issue 19, September 1999
in the non-pregnant state[3]. But the radiation from this
examination is considerable (500–3740 ìGy)[24] and in
pregnancy its use should be limited to patients who have
collapsed where there is no confirmed diagnosis. Similar
constraints apply to spiral CT scanning.
By contrast magnetic resonance imaging has recently
been shown to be of value in the diagnosis of pulmonary
embolus. If preliminary results are confirmed this could
well be the technique of choice for lung imaging in
pregnancy because of the lack of radiation.
The indications for leg vein imaging in suspected
pulmonary embolus (indeterminate V/Q scan) in pregnancy are the same as in the non-pregnant state. However, in pregnancy in particular, ultrasound should be
used rather than contrast venography because of the
lack of radiation. From about 20 weeks gestation such
studies should be performed in the left lateral position
to avoid false-positive results caused by obstruction to
venous flow from the gravid uterus[26]. Again, magnetic
resonance imaging of the femoral iliac and ovarian
veins[28] is likely to become very helpful. Plasma
D-dimer assay has been proposed as an under-used
test to exclude thromboembolism[3,29]. The levels are
elevated because of activation of fibrin degradation. If
this test is used in pregnancy it will need a different set of
cut-off values since fibrin degradation is part of normal
Differential diagnosis
Pregnancy-specific differential diagnoses are amniotic
fluid embolus and chorion carcinoma. The pathophysiology of amniotic fluid embolus is obscure. It is only
likely to cause confusion in the first hours of acute
vascular collapse. An important distinguishing feature is
disseminated vascular coagulopathy, which is almost
invariably present in amniotic fluid embolus but not in
pulmonary embolus before treatment.
Chorioncarcinoma can occur in pregnancy or be
present for several years after delivery from a normal
pregnancy or non-molar abortion. It can mimic pulmonary embolus due to the presence of tumour in pulmonary blood vessels or be a cause of pulmonary embolus.
Although the association is rare it is one that should not
be missed and ideally all women presenting with pulmonary embolus or suspected pulmonary embolus in
reproductive life should have a pregnancy test and
estimation of serum human chorionic gonadotropin if
Thrombophilia testing
As in the non-pregnant state, the presence or absence of
thrombophilia has prognostic value relating to recurrence risk and could be used to aid decision making with
regard to length of treatment. However, thrombophilia
status is also important with regard to obstetric
management since, in general, women with thrombophilia have a poorer pregnancy outcome. Unfortunately, testing for many thrombophilias is affected
both by pregnancy and by anticoagulant treatment.
Antiphospholipid antibodies such as anticardiolipin
antibody are not affected and these should be sought in
all women who have pulmonary embolus in pregnancy.
At present, individual units must develop their own
policies regarding the cost effectiveness of screening for
other thrombophilias.
Heparin, low molecular weight heparin and
Anticoagulation with heparin should be the initial treatment in pregnancy. Heparin, both low molecular
weight[31] and unfractionated[32] does not cross the placenta or the breast. It is therefore safe for the fetus and
for the breast-fed infant. However there is considerable
concern about the use of warfarin in pregnancy, chiefly
for fetal reasons. One review suggests that the pooled
incidence of adverse effects due to warfarin in pregnancy
is 26%[33]. In the first trimester warfarin is teratogenic[34,35] and causes abnormalities of cartilage and
bone formation, chondrodysplasia punctata. It also
increases the risk of miscarriage. In the second and third
trimester, maternal warfarin use has been associated
with major fetal central nervous system abnormalities,
particularly microcephaly and optic atrophy[34,35]. This
is probably because warfarin, in contrast to herparin,
does cross the placenta; also the fetus has a relatively
immature clotting system so that a dose of warfarin
which gives therapeutic anticoagulation in the mother is
likely to over-anticoagulate the fetus. If the mother
receives full anticoagulation with warfarin at the time of
delivery, there are the fetal risks of gross retroplacental
and intracerebral bleeding[34]. In addition, women
bleed easily in pregnancy, particularly from the
genital tract. While this risk may be no greater with full
anticoagulation using warfarin than with heparin,
reversal of anticoagulation is much easier with heparin
For all these reasons heparin is preferred to warfarin
for the treatment of pulmonary embolus for as long as
the woman is pregnant. (Certain patients, such as those
with artificial heart valves, heparin allergy or very active
antiphospholipid syndrome may still require treatment
with warfarin in pregnancy despite the fetal and maternal risks outlined above). No randomized controlled
trials have been performed for the treatment of pulmonary embolus in pregnancy. However, a reasonable strategy is to use intravenous heparin aiming for activated
partial thromboplastin time 1·5–2·5 times the control
value for 10 days and then to change to a lower intensity
subcutaneous regime. Patients can usually learn to inject
themselves with relatively little difficulty. In practice
10 000 units of unfractionated heparin twice daily, or
now more commonly low molecular weight heparin such
as enoxaparine 40 mg once daily, give adequate anticoagulation in this second phase. It is accepted that this
level of anticoagulation is less than that usually obtained
with warfarin maintaining the international normalized
ratio at 2·0–3·0.
Although it has been suggested that when administered subcutaneously, unfractionated heparin should be
given twice daily in doses (presumably in excess of
10 000 units) to maintain the activated partial thromboplastin time in the normal therapeutic range[26,36], this is
very difficult to achieve in practice and usually causes
unacceptable bruising at the injection site. There are no
comparative studies that indicate that this level of anticoagulation is necessary. Furthermore, although the aim
may be to achieve an activated partial thromboplastin
time ratio of 1·5, this is not necessarily achieved, leaving
the patient at risk from both under-and overanticoagulation. Additional anxiety will occur should
the patient need delivery or some other surgical procedure. In practice, the extra inconvenience of an
adjusted dose subcutaneous heparin regime does not
seem worthwhile.
An alternative and more practical strategy is to use
high dose subcutaneous low molecular weight heparin
e.g. enoxaparine 1 mg . kg 1 every 12 h for a further 6
weeks in patients judged to be at particularly high risk of
further embolization. Although high dose low molecular
weight heparin is not currently recommended for the
treatment of acute pulmonary embolism in pregnancy
(see below), there is no reason why it should not be used
in the treatment after the first week granted that lower
intensity anticoagulation has been satisfactory in current
clinical practice.
Whenever heparin is used for more than 2 months,
patients should be warned of the risk of bone demineralization. Most patients undergo some degree of demineralization as judged by bone densitometry studies;
however, the majority of these changes are reversed one
year after the cessation of therapy and breast-feeding[37].
About 2% of all patients taking heparin for 3 months or
more in pregnancy will have symptoms from bone
fracture due to demineralization. The incidence of such
symptoms may be less with low molecular weight
heparin, although strictly comparative data to support
this suggestion only come from animal studies. Other
complications of heparin therapy such as thrombocytopoenia are very uncommon in pregnant patients even if
they take heparin long term.
Occasionally patients develop erythematous itchy skin
rashes at the site of heparin injection, which can be so
severe as to limit treatment. These rashes are presumed
to be due to allergy. It is worth trying different forms
of heparin, both within and between the categories of
unfractionated heparin and low molecular weight
heparin, but often the patients react to all preparations.
The only option is then to use warfarin, notwithstanding
the fetal risks. It may be possible to use intravenous
heparin to cover the peripartum period, but this needs to
Eur Heart J, Vol. 20, issue 19, September 1999
M. de Swiet
be introduced very carefully because of the obvious risk
of anaphylaxis. It is possible that the risk of anaphylaxis
may be reduced by given a small dose of glucocorticoid
if intravenous heparin is being used[38].
The excess risk of thromboembolism associated with
pregnancy continues after delivery for an ill-defined
period, believed by most clinicians to be no longer than
6 weeks. Therefore, treatment for pulmonary embolus
occurring in relation to pregnancy should continue until
6 weeks after delivery or until 3 months after the initial
episode whichever is the longer.
To manage labour and delivery, one method has been
to reduce the dose of subcutaneous unfractionated
heparin to 7500 units twice daily in anticipation of the
contraction in circulating blood volume and to counter
any bleeding risk. This reduction seems unnecessary in
patients taking low molecular weight heparin in daily
doses of the order of enoxaparine 40 mg[39] or fragmin
5000 units. Providing the thrombin and activated partial
thromboplastin times are not prolonged by more than
5 s the patients do not bleed excessively and nor is
epidural haematoma a problem[39].
If there is real concern about the risk of bleeding in
patients taking unfractionated heparin, measurement of
the heparin level by anti Xa assay will give further
information; if the level is less than 0·4 units . ml 1 (as is
almost invariably the case using the doses of unfractionated heparin mentioned above), the patient should not
bleed because of heparin treatment[40]. Similar data have
not yet been established for low molecular weight
heparin. Since warfarin is not secreted in significant
quantities in breast milk, patients may convert to warfarin after delivery even if they are breast feeding[41].
There is also no problem with either unfractionated
heparin or low molecular weight heparin with regard to
The use of high dose low molecular weight heparin,
such as enoxaparine 1 mg . kg 1 every 12 h, for the
acute treatment of pulmonary embolus has recently been
studied in the non-pregnant state. However, the trials
that have been performed comparing high dose low
molecular weight heparin with unfractionated heparin in
acute thromboembolism have all compared low molecular weight heparin and near simultaneous warfarin with
unfractionated heparin and similar warfarin therapies.
The results of such trials cannot be extrapolated directly
to pregnancy where low molecular weight heparin would
be used alone. Therefore, at present high dose low
molecular weight heparin is not recommended for the
initial treatment of thromboembolism in pregnancy.
Nevertheless the practical advantages of low molecular
weight heparin are just as great if not greater for
pregnant patients compared to those who are not
pregnant. It is therefore likely that high dose low
molecular weight heparin will be used in acute
thromboembolism in pregnancy, as more experience is
gained in other situations such as in the management of
patients with artificial heart valves. It is also likely that,
as in the non-pregnant state, high dose low molecular
weight heparin will first be used in patients with acute
Eur Heart J, Vol. 20, issue 19, September 1999
deep vein thrombosis rather than acute pulmonary
There is published information on about 30 women
treated with thrombolytic agents in pregnancy, massive
pulmonary embolus being the indication in about one
third[42]. Streptokinase was the agent used most frequently. Streptokinase (and probably other thrombolytic drugs) does not cross the placenta because of its
high molecular weight. However in the mother, bleeding
is the major side effect, usually from the genital tract and
often severe; the overall incidence of bleeding is about
8%[42]. Because of the risk of bleeding, thrombolysis
should not be used routinely in pregnancy, but reserved
for those who are haemodynamically unstable, particularly with regard to systemic hypotension. It would not
be appropriate to extrapolate directly from studies in
non-pregnant patients[43] using echocardiographic variables as criteria for thrombolytic treatment. Because of
the risk of bleeding, thrombolytic treatment should not
be used at the time of delivery unless it appears that the
patient is likely to die.
Surgical procedures
Procedures such as embolectomy are only occasionally
indicated in pregnancy; one obvious situation would be
massive pulmonary embolus immediately after delivery
when thrombolysis should not be used. However, in
most maternity units thrombolytic treatment or even
transvenous catheter fragmentation or embolectomy[44]
are more likely to be available.
Caval filters
Filters have been used in pregnancy[45] but controversy
about the place of interrupt devices[46,47] is just as great
with regard to patients who are pregnant, if not more so.
Concerns that are specific to pregnancy are that free
floating clot in the inferior vena cava before delivery
may be ‘liberated’ at delivery, thus causing further
pulmonary embolus, and fear for the fetus because of
the additional radiation that is necessary for the placement of these devices. At present, interrupt devices
should only be placed where recurrent pulmonary
embolus has occurred despite adequate anticoagulation
or where patients cannot receive conventional
anticoagulant treatment.
Although the subject of thromboprophylaxis is too
general to be within the remit of this review,
thromboprophylaxis in pregnancy in women who have
previously had an episode of thromboembolism is a
specific problem, which should be addressed here. Again
there are no randomized trials to aid the clinician.
The following guidelines are suggested because they
are practical and because in general they have been
In women who have had thromboembolism in the
past, the recurrence risk in pregnancy has been estimated
to be about 12% whatever the circumstances of the
original clot and independent of risk factors such as
thrombophilia[49]. However, this study from 1974 was a
postal survey and no record was made of the method of
ascertainment of the subsequent episode. Recurrence
was probably over-diagnosed and the true recurrence
risk is probably less than 12%. Subsequent incidence
studies have been confounded by the use of some form
of prophylaxis. Granted the risk and granted the safety
for the fetus of maternal subcutaneous heparin therapy
it is tempting to recommend this throughout pregnancy
for all at risk of recurrence of thromboembolism. However, the risks of bone demineralization and other
heparin-induced side effects, plus the inconvenience of
at least daily injections make such a blanket recommendation inadvisable. Instead, patients may be categorized
as high and low risk for recurrence[10]. Low risk patients
are those who have only had one previous episode
of thromboembolism (pulmonary embolism or deep vein
thrombosis) no matter what the original circumstances
and who have no other high risk factors. High risk
patients have had more than one episode or have
thrombophilia or a family history of thromboembolism,
suggesting that they may have undiagnosed thrombophilia.
Low risk patients should take low dose aspirin, say
75 mg once daily from as soon as pregnancy has been
confirmed until delivery. Meta-analysis of many trials in
medical, surgical and orthopaedic (but not pregnant)
patients, indicates that aspirin reduces the incidence of
deep-vein thrombosis and pulmonary embolus by at
least 50%[50]. The safety of low dose aspirin has been
demonstrated in the numerous large randomized trials
e.g. CLASP (Collaborative Studies of Low Dose Aspirin
in Pregnancy)[51] where aspirin has been tested for the
prevention of pre-eclampsia. At delivery, patients should
take subcutaneous heparin[52] either unfractionated
heparin 7500 units twice daily, or low molecular weight
heparin such as enoxaparine 40 mg once daily, until
at least the first week after delivery, and then either
continue heparin for a further 5 weeks or switch to
warfarin[48]. High-risk patients take subcutaneous
heparin in doses as above rather than aspiring as soon as
pregnancy is confirmed. Management at delivery and
thereafter is as for low risk patients[39].
The safety of these regimes with regard to delivery,
epidural block and breast feeding has already been
discussed. Patients who are taking the high risk regime
should be warned of the risk of bone demineralization.
Individual risk factors and patients’ wishes should also
be considered e.g. an otherwise low risk woman aged
45 years should probably be treated with the high
risk regime.
As already indicated, none of the recommendations
given for therapy and prophylaxis of pulmonary embolus in pregnancy have been established on the basis of
clinical trial. It is therefore particularly difficult to be
dogmatic granted the inevitable wide variety of opinions
and clinical experience between different countries.
However, the guidelines suggested above are derived
from clinical practice and they have been found to work
well. They should therefore form a framework, which
can be adapted both in accordance with others’ clinical
experience and as a consequence of clinical trials, which
will eventually be performed.
Conclusions and recommendations
1. Take samples for blood gas analysis with the patient
2. Do not rely on axis changes in the electrocardiogram
3. Consider echocardiography
4. Chest X-ray and lung scan are safe in pregnancy
5. The diagnosis must be confirmed by lung imaging
6. Lung scan or magnetic resonance imaging are
preferred to pulmonary angiography
7. Exclude chorioncarcinoma
1. Warfarin should usually be avoided in the treatment
of pulmonary embolism in pregnancy
2. Use full anticoagulation with intravenous heparin for
the first 10 days
3. Subsequent treatment should be with subcutaneous
heparin either unfractionated heparin or low molecular weight heparin
4. Subcutaneous heparin does not need to be dose
5. Labour, surgical procedures and epidural block are
safe providing that the thrombin time, activated
partial thromboplastin time and platelet count are
6. Continue treatment for 6 weeks after delivery
7. Consider switching to warfarin one week after
8. Warfarin is safe when mothers are breast feeding
9. Thrombolysis is the most practical rescue therapy for
the patient who is haemodynamically unstable
following pulmonary embolus.
1. Low-risk patients have only had a single episode of
previous thromboembolism with no other high-risk
Eur Heart J, Vol. 20, issue 19, September 1999
M. de Swiet
2. High-risk patients have had previous multiple
episodes or have had at least one previous episode
and have documented thrombophilia or a family
history or thromboembolism.
3. Treat low risk patients with aspirin 75 mg once daily
from when pregnancy is confirmed to delivery. Then
switch to subcutaneous heparin, either unfractionated heparin 7500 units twice daily or low molecular
weight heparin e.g. enoxaparine 40 mg once daily.
One week after delivery consider switching to
4. Treat high risk patients as low risk except that they
should take subcutaneous heparin from as soon as
pregnancy is diagnosed.
Conflict of interest
Dr de Swiet has been supported with research and travel
grants by Rhône-Poulenc Rorer.
[1] Koonin LM, Atrash HK, Lawson HW, Smith JC. Maternal
mortality surveillance, United States, 1979–1986. MMWR
CDC Suveill Summ 1991; 40 (SS-2): 1–13.
[2] Report on Confidential Enquiries into Maternal Deaths in the
United Kingdom 1994–96. HMSO, 1998.
[3] British Thoracic Society Standards of Care Committee.
Thorax 1997; 52 (Suppl 4): S1–24.
[4] NIH 1986. Prevention of venous thrombosis and pulmonary
embolism. NIH Consensus Development. JAMA 1986; 256:
[5] Clarke-Pearson DL, Jelovsek FR. Alterations of occlusive cuff
impedance plethysmography results in the obstetric patient.
Surgery 1981; 89: 594–8.
[6] Macklon NS, Greer IA, Bowman AW. An ultrasound study
of gestational and postural changes in the deep venous system
of the leg in pregnancy. Br J Obstet Gynaecol 1997; 104:
[7] Macklon NS, Greer IA. The deep venous system in the
puerperium: an ultrasound study. Br J Obstet Gynaecol 1997;
104: 198–200.
[8] Bonnar J. Haemostasis and coagulation disorders in pregnancy. In: Bloom AL, Thomas DP, eds. Haemostasis
and thrombosis, 2nd edn. Edinburgh, Scotland: Churchill
Livingstone, 1987: 570–84.
[9] Weiner CP, Kwaan H, Hauck WW, Duboe FJ, Paul M,
Wallemark CB. Fibrin generation in normal pregnancy.
Obstet Gynecol 1984; 64: 46–8.
[10] de Swiet M. Thromboembolism. In: de Swiet M, ed. Medical
disorders in obstetric practice, 3rd edn. Oxford: Blackwell
Scientific Publications, 1995: 116–42.
[11] Hirsh J, Hoak J. Management of deep vein thrombosis and
pulmonary embolism. A statement for healthcare professionals. Council on Thrombosis (in consultation with the Council
on Cardiovascular Radiology), American Heart Association.
Circulation 1996; 93: 2212–45.
[12] Peek MJ, Nelson-Piercy C, Manning RA, de Swiet M, Letsky
EA. Activated protein C resistance in normal pregnancy. Br J
Obstet Gynaecol 1997; 104: 1084–6.
[13] Comp PC, Thurnau GR, Welsh J, Esmon CT. Functional and
immunologic protein S levels are decreased during pregnancy.
Blood 1986; 68: 881–5.
[14] de Swiet M, Fidler J, Howell R, Letsky E. Thromboembolism
in pregnancy. In: Jewell D, ed. Advanced Medicine. London:
Pitman Medical, 1981: 80: 309–17.
Eur Heart J, Vol. 20, issue 19, September 1999
[15] Rutherford S, Montoro M, McGehee W, Strong T. Thromboembolic disease associated with pregnancy: an 11 year
review. Am J Obstet Gynecol 1991; 164 (Suppl): 286.
[16] Macklon NS, Greer IA. Venous thromboembolic disease in
obstetrics and gynaecology: the Scottish experience. Scott
Med J 1996; 41: 83–6.
[17] Ginsberg JS, Brill-Edwards P, Burrows RF et al. Venous
thrombosis during pregnancy: leg and trimester of presentation. Thromb Haemost 1992; 67: 519–20.
[18] Report on Confidential Enquiries into Maternal Deaths in
England and Wales 1973–5. Department of Health and Social
Security. London: HMSO, 1979.
[19] Report on Confidential Enquiries into Maternal Deaths in the
United Kingdom 1988–90. Department of Health. London:
HMSO, 1994.
[20] Carr MH, Towers CV, Eastenson AR, Pircon RA, Iriye BK,
Adashek JA. Prolonged bedrest during pregnancy: does the
risk of deep vein thrombosis warrant the use of routine
heparin prophylaxis? J Matern Fetal Med 1997; 6: 264–7.
[21] Hirsch DR, Mikkola KM, Marks PW et al. Pulmonary
embolism and deep vein thrombosis during pregnancy or oral
contraceptive use: prevalence of factor V Leiden. Am Heart J
1996; 131: 1145–8.
[22] Friedreich PW et al. Frequency of pregnancy related venous
thromboembolism in anticoagulant factor deficient women:
implications for prophylaxis. Ann Int Med 1996; 125: 955–60.
[23] Carson JL, Kelley MA, Duff A et al. The clinical course of
pulmonary embolism. N Engl J Med 1992; 326: 1240–5.
[24] Ginsberg JS, Hirsh J, Rainbow AJ, Coates G. Risks to the
fetus of radiologic procedures used in the diagnosis of maternal venous thromboembolic disease. Thromb Haemost 1989;
61: 189–96.
[25] Ang CK, Tan TH, Walters WA, Wood C. Postural influence
on maternal capillary oxygen and carbon dioxide tension.
BMJ 1969; 4: 201–3.
[26] Toglia MR, Weg JG. Venous Thromboembolism during
Pregnancy. N Engl J Med 1996; 335: 108–14.
[27] Miniati M, Pistolesi M, Marini C et al. Value of perfusion
lung scan in the diagnosis of pulmonary embolism: results of
the Prospective Investigative Study of Acute Pulmonary
Embolism Diagnosis (PISA-PED). Am J Respir Crit Care
Med 1996; 154: 1387–93.
[28] Spritizer CE, Evans AC, Kay HH. Magnetic resonance imaging of deep venous thrombosis in pregnant women with lower
extremity edema. Obstet Gynecol 1995; 85: 603–7.
[29] Wells PS, Anderson DR, Bormanis J, Guy F, Mitchell M,
Lewandowski B. SimpliRED D-Dimer can reduce the diagnostic tests in suspected deep vein thrombosis. Lancet 1998;
351: 1405–6.
[30] Savage P, Roddie M, Seckl. A 28-year-old woman with
pulmonary embolus. Lancet 1998; 352: 30–1.
[31] Omri A, Delaloye JF, Andersen H, Bachmann F. Low
molecular weight heparin Novo (LHN-1) does not cross the
placenta during the second trimester of pregnancy. Thromb
Haemost 1989; 61: 55–6.
[32] Ginsberg JS, Hirsh J. Use of antithrombotic agents during
pregnancy. Chest 1995; 108 (Suppl): 305S–311S.
[33] Ginsberg JS, Hirsh J, Turner DC, Levine MN, Burrows R.
Risks to the fetus of anticoagulant therapy during pregnancy.
Thromb Haemost 1989; 61: 197–203.
[34] Hall JG, Pauli RM, Wilson KM. Maternal and fetal sequelae
of anticoagulation during pregnancy. Am J Med 1980; 68:
[35] Wong V, Cheng CH, Chan KC. Fetal and neonatal outcome
of exposure to anticoagulants during pregnancy. Am J Med
Genet 1993; 45: 17–21.
[36] Hirsh J, Hoak J. Management of deep vein thrombosis and
pulmonary embolism. Circulation 1996; 93: 2212–45.
[37] Dahlman T, Lindvall N, Hellgren M. Osteopenia in pregnancy during long-term heparin treatment: a radiological
study post partum. Br J Obstet Gynaecol 1990; 97: 221–8.
[38] Ojukwu C, Jenkinson SD, Obeid D. Deep vein thrombosis in
pregnancy and heparin hypersensitivity. Br J Obstet Gynaecol
1996; 103: 934–6.
[39] Nelson Piercy C, Letsky EA, de Swiet M. Low molecular
weight heparin for obstetric thromboprophyllaxis: experience
of 69 pregnancies in 61 women at high risk. Am J Obstet
Gynecol 1997; 176: 1062–8.
[40] Dahlman TC, Hellgren MS, Blomback M. Thrombosis
prophylaxis in pregnancy with use of subcutaneous heparin
adjusted by monitoring heparin concentration in plasma. Am
J Obstet Gynecol 1989; 161: 420–5.
[41] Orme ML, Lewis PJ, de Swiet M et al. May mothers given
warfarin breast-feed their infants? BMJ 1997; 1: 1564–5.
[42] Turrentine MA, Braems G, Ramirez MM. Use of thrombolytics for the treatment of thromboembolic disease during
pregnancy. Obstet Gynecol Surv 1995; 50: 534–41.
[43] Goldhaber SZ, Haire WD, Feldstein ML et al. Alteplase
versus heparin in acute pulmonary embolism: randomised trial
assessing right-ventricular function and pulmonary perfusion.
Lancet 1993; 341: 507–11.
[44] Greenfield J, Proctor MC. Role of catheter embolectomy in
treating pulmonary embolism Semin. Resp Crit Care Med
1996; 17: 71–85.
[45] Greenfield LJ, Cho KJ, Proctor MC, Sobel M, Shah S, Wingo
J. Late results of suprarenal Greenfield vena cava filter
placement. Arch Surg 1992; 127: 969–73.
[46] Chuu WM, Wang NY, Perry D. Vena caval filters for the
prevention of pulmonary embolism. N Engl J Med 1998; 339:
[47] Decousus H, Leizorovicz A, Parent F, Page Y, Tardy B,
Girard P et al. A clinical trial of vena caval filters in the
prevention of pulmonary embolism in patients with proximal
deep-vein thrombosis. N Engl J Med 1998; 338: 409–15.
[48] Tan J, de Swiet M. Use of aspirin for obstetric thromboprophylaxis in low risk women. J Obs Gynaecol 1998; 18
(Suppl 1): S46–7.
[49] Baddarraco MA, Vessey MP. Recurrence of venous thromboembolic disease and use of oral contraceptives. Br Med J
1974; 1: 215–7.
[50] Antiplatelet Trialists’ Collaboration. Collaborative overview
of randomised trials of antiplatelet therapy—111: Reduction
in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. Br Med
J 1994; 308: 235–46.
[51] CLASP Collaborative Group. (Writing Committee Redman
CWG, de Swiet M, Collins M & Grant A). A randomised trial
of low-dose aspirin for the prevention and treatment of
pre-eclampsia among 9364 pregnant women. Lancet 1994;
343: 619–29.
[52] Maternal and Neonatal Haemostasis Working Party of the
Haemostasis and Thrombosis Task. Guidelines on the prevention, investigation and management of thrombosis associated
with pregnancy. J Clin Pathol 1993; 46: 489–96.
[53] Girling J, de Swiet M. Inherited thrombophilia and pregnancy. Curr Opin Obstet Gynaecol 1998; 10: 135–44.
[54] Conard J, Horellou MH, Van Dreden P, Lecompte T,
Samama M. Thrombosis and pregnancy in congenital
deficiencies in AT III, protein C or protein S: study of 78
women. Thromb Haemost 1990; 63: 319–20.
[54] Hough RE, Makris M, Preston FE. Pregnancy in women
with thrombophilia: incidence of thrombosis and pregnancy
outcome (Abstr). Br J Haem 1996; 93 (Suppl 2): 136.
[55] Papinger I, Schneider B. Thrombotic risk in ereditary Antithrombin III, Protein C, or protein S deficiency. A cooperative
retrospective study. Study Group on Natural Inhibitors.
Arterioscler Thromb Vasc Biol 1996; 16: 742–8.
Eur Heart J, Vol. 20, issue 19, September 1999