www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1 1
April 10, 2012
Brigham and Women’s
Hospital, Department of
Medicine, Harvard Medical
Division, Boston, MA, USA
(Prof S Z Goldhaber MD); and
Faculty of Medicine of Geneva,
Division of Angiology and
Hemostasis, Department of
Medical Specialties, University
Hospitals of Geneva, Geneva,
(Prof H Bounameaux MD)
Prof Samuel Z Goldhaber,
Cardiovascular Division, Brigham
and Women’s Hospital, Boston,
MA 02115, USA
Pulmonary embolism and deep vein thrombosis
Samuel Z Goldhaber, Henri Bounameaux
Pulmonary embolism is the third most common cause of death from cardiovascular disease after heart attack and
stroke. Sequelae occurring after venous thrombo embolism include chronic thromboembolic pulmonary hypertension
and post-thrombotic syndrome. Venous thromboembolism and atherothrombosis share common risk factors and the
common pathophysiological characteristics of infl ammation, hypercoagulability, and endothelial injury. Clinical
probability assessment helps to identify patients with low clinical probability for whom the diagnosis of venous
thromboembolism can be excluded solely with a negative result from a plasma D-dimer test. The diagnosis is usually
confi rmed with compression ultrasound showing deep vein thrombosis or with chest CT showing pulmonary embolism.
Most patients with venous thromboembolism will respond to anticoagulation, which is the foundation of treatment.
Patients with pulmonary embolism should undergo risk stratifi cation to establish whether they will benefi t from the
addition of advanced treatment, such as thrombolysis or embolectomy. Several novel oral anticoagulant drugs are in
development. These drugs, which could replace vitamin K antagonists and heparins in many patients, are prescribed in
fi xed doses and do not need any coagulation monitoring in the laboratory. Although rigorous clinical trials have reported
the eff ectiveness and safety of pharmacological prevention with low, fi xed doses of anticoagulant drugs, prophylaxis
remains underused in patients admitted to hospital at moderate risk and high risk for venous thromboembolism. In
this Seminar, we discuss pulmonary embolism and deep vein thrombosis of the legs.
Deep vein thrombosis and pulmonary embolism consti tute
venous thromboembolism. Deep vein thrombosis occurs
most often in the legs, but can form in the veins of the
arms,1 and in the mesenteric and cerebral veins. We focus
on deep vein thrombosis of the legs and pulmonary
embolism. Although these disorders are part of the same
syndrome, important diff erences in epidemiology, diag-
nosis, and treatment exist between them.
In population-based studies, no consensus exists about
whether the incidence of venous thromboembolism
varies according to sex. In a Norwegian study,2 the inci-
dence of all fi rst events of venous thromboembolism was
1·43 per 1000 person-years, and was slightly higher in
women than in men. In a Swedish study,3 incidence was
equal for both sexes. In a community-based study,4
incidence was higher for men than for women (1·14 per
1000 patient-years vs 1·05 per 1000 patient years). In
the International Cooperative Pulmonary Embolism
Registry,5 the primary outcome—all-cause mortality rate
at 3 months—associated with acute pulmonary embolism
was 17%. This registry, which had no exclusion criteria,
enrolled 2454 consecutive patients from 52 hospitals
in seven countries in Europe and North America.
Pulmonary embolism was considered to be the cause of
death in 45% of patients. Important prognostic factors
associated with death from pulmonary embolism were
age older than 70 years, cancer, congestive heart failure,
chronic obstructive pulmonary disease, systolic arterial
hypotension, tachypnoea, and right ventricular hypo-
kinesis on echocardiography.
In the Worcester, Massachusetts metropolitan area,
patients presenting with pulmonary embolism from the
outpatient setting had an all-cause mortality rate of 11·1%
at 90 days;6 however, some estimates of case fatality rate
are lower. For example, in the Registry of Patients with
Venous Thromboembolism (RIETE)7 of 6264 patients
with pulmonary embolism, the cumulative overall mor-
tality rate was 8·6% at 3 months and the case fatality rate
was 1·7%. Mortality rates were low among 1880 patients
diagnosed with acute pulmonary embolism in 22 US
emergency departments: the all-cause mortality rate was
5·4% at 30 days, and the mortality rate directly attributable
to pulmonary embolism was only 1·0%.8,9 Although some
studies report low rates of short-term mortality, long-
term mortality associated with acute pulmonary embol-
ism seems to be high. In an Australian registry9 of
1023 patients with confi rmed pulmonary embolism
followed up for a mean of 4 years, 36% of patients died,
but only 3% died in hospital during the index admission
for pulmonary embolism. The mortality after discharge
of 8·5% per patient-year was 2·5 times higher than that
in an age-matched and sex-matched general population.
Of the 332 deaths occurring after discharge, 40% were
because of cardiovascular causes.
Many individuals who have a fi rst episode of deep
vein thrombosis or pulmonary embolism will have a
recurrent event. For some, the fi rst event of venous
thrombo embolism is not diagnosed, whereas for others,
venous thromboembolism recurs after anticoagulation
treatment is stopped. Two associated illnesses arise after
pulmonary embolism or deep vein thrombosis: chronic
thrombo embolic pulmonary hypertension10 and post-
thrombotic syndrome.11 The term chronic venous
insuffi ciency encompasses post-thrombotic syndrome
but can be idiopathic or caused by disorders other than
thrombosis. Chronic thromboembolic pulmonary
hyper tension is defi ned as a mean pulmonary artery
pressure greater than 25 mm Hg that persists 6 months
after diagnosis of pulmonary embolism. The disorder
occurs in 2–4% of patients after acute pulmonary
embolism and results in disabling dyspnoea, both at
Volume 379, Issue
9828, Pages 1835 -
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1
rest and with exertion. Life expectancy is often shortened
and patients frequently die of sudden cardiac death.
Death is usually due to progressive pulmonary hyper-
tension culminating in right ventricular failure. Post-
thrombotic syndrome can result in chronic calf swelling,
which might lead to brownish skin pigmentation of the
lateral medial malleolus and, in extreme circumstances,
to venous ulceration of the skin. Only mild to moderate
forms of the post-thrombotic syndrome are usually
seen; severe forms are rare. In a prospective multicentre
cohort study12 of 387 patients newly diagnosed with
symptomatic deep vein thrombosis of the leg who were
followed up for 2 years, post-thrombotic syndrome
developed in 43% of patients and was mild in 30%,
moderate in 10%, and severe in 3%.
The traditional concept of separation of risk factors and
pathophysiology for venous thromboembolism and
coronary artery disease is being reconsidered. Labelling
of venous thromboembolism as a venous disease with
red thrombus, by contrast with coronary artery disease as
a separate arterial disease with white platelet plaque,
might be an oversimplifi cation. For example, 4 years after
the onset of acute pulmonary embolism, fewer than half
of those who initially survive will remain free of
myocardial infarction, stroke, peripheral arterial disease,
recurrent venous thrombo embolism, cancer, or chronic
thromboembolic pul monary hypertension.13
Venous thromboembolism and atherothrombosis
have shared risk factors and a common pathophysiology
that includes infl ammation, hypercoagulability, and
endothelial injury.14 A novel approach reframes venous
thromboembolism as a disease that contributes to a pan-
vascular syndrome that consists of coronary artery
disease, peripheral arterial disease, and cerebrovascular
disease. Risk factors for venous thromboembolism, such
as cigarette smoking, hypertension, diabetes, and obesity,
are often modifi able and overlap with risk factors for
atherosclerosis.15 Infl ammatory disorders, such as infl am-
matory bowel disease and systemic vasculitis, have been
associated with venous thromboembolism. In the
Atherosclerosis Risk In Communities (ARIC) study,16
concentrations of C-reactive protein (a marker of
infl ammation) above the 90th percentile were associated
with a substantial increase in risk of venous thrombo-
embolism compared with lower percentiles.
Venous thromboembolism can be categorised, some-
what arbitrarily, as idiopathic and primary or as provoked
and secondary (panel 1). This dichotomy is often unclear
and, at times, does not seem to have consistent logic. For
example, venous thromboembolism resulting from long-
haul travel is usually assigned idiopathic, whereas that
caused by oral contraceptives is usually assigned
provoked. Patients with idiopathic and primary disease
are much more likely to suff er recurrence than are
those with the provoked and secondary form if
anticoagulation is discontinued. Whether patients with
venous thromboembolism should be screened for
thrombophilia remains controversial.24 Hypercoagulable
states—eg, factor V Leiden or prothrombin gene
mutation—can be associated with an initial episode of
venous thromboembolism. Factor V Leiden has a much
stronger association with deep vein thrombosis than with
pulmonary embolism;25 this observation is the Leiden
paradox. Neither the factor V Leiden nor the prothrombin
gene mutation is a strong predictor of recurrent venous
Clinical probability assessment
Diagnosis of deep vein thrombosis and pulmonary
embolism is dependent on several, mainly non-invasive,
diagnostic techniques that should be used sequentially.
Because use of a validated diagnostic work-up is
associated with a substantially diminished risk of com-
plications,27 implementation of such standardised
approaches is highly recommended. Massive pulmonary
embolism should be diagnosed quickly; its clinical
features include shock or haemodynamic instability.
Clinical probability assessment aims to identify patients
with a high or intermediate clinical probability who
need anticoagulant treatment while awaiting the results
of diagnostic tests. In patients with a low clinical
probability, the diagnosis of venous thromboembolism
can be ruled out solely with a normal D-dimer test
(fi gure 1). Clinical probability incorporates clinical
history (including personal and familial features) and
symptoms, signs, and abnormalities of oxygen satura-
tion, chest radiography, and electrocardiography. The
probability can be assessed empirically or with prediction
rules or scores.
Panel 1: Major risk factors for pulmonary embolism
Idiopathic, primary, and unprovoked
• No apparent cause
• Old age (>65 years)
• Long-haul travel17
• Associated with thrombophilia (eg, factor V Leiden or
prothrombin gene mutation)
• Cigarette smoking18
• Metabolic syndrome19
• Air pollution20
Secondary and provoked
• Oral contraceptives,21 pregnancy, postmenopausal
• Acute medical illness (eg, pneumonia, congestive
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1 3
Scoring systems have clinical use and are useful
educational methods for clinicians and medical students
attempting to diagnose or exclude venous thrombo-
embolism. For suspected pulmonary embolism, two scores
are widely used: the Wells score28 and the revised Geneva
score29 (table 1). The Wells score can be used to diagnose
suspected deep vein thrombosis.30 The Wells score for
pulmonary embolism is now mostly used with a cutoff of
four points,31 which allows a dichotomous classifi cation of
likely or unlikely pulmonary embolism. According to a
meta-analysis32 of the performance of all available clinical
prediction rules for suspected pulmonary embolism, these
rules have similar accuracy, but are not totally equivalent.
The choice among various prediction rules and
classifi cation schemes should be guided by the local
prevalence of pulmonary embolism, the type of patients
being assessed (outpatients or inpatients), and the type of
D-dimer assay used. For example, the revised Geneva score
should be used in populations with a prevalence of
pulmonary embolism of more than 20%, whereas the
Wells score is the only validated score for patients admitted
to hospital. The results of arterial blood gas oxygen
saturation, electrocardiography (ECG), and chest radi-
ography have low sensitivity and specifi city for the
diagnosis of pulmonary embolism, and are incorporated
in neither the Wells nor the revised Geneva score.
Conversely, ECG might be useful to exclude pulmonary
embolism (and to suggest acute coronary syndrome, for
example), but chest radiography and arterial blood gas
saturation should not be used routinely.
Measurement of fi brin D-dimer
Fibrin D-dimer is a degradation product of cross-linked
fi brin, and its concentration increases in patients with
acute venous thromboembolism. When assayed by a
quantitative ELISA or by some automated turbidimetric
assays, D-dimer is highly sensitive (more than 95%) in
excluding acute deep vein thrombosis or pulmonary
embolism, usually below a threshold of 500 μg/L.
Hence, a concentration lower than this value rules out
acute venous thromboembolism, at least in patients
with low or intermediate clinical probability.33 According
to a meta-analysis,34 the VIDAS D-dimer exclusion test
(an ELISA assay, bioMérieux) has now been reported in
5060 patients with suspected pulmonary embolism and
is associated with a very low (less than 1%) thrombo-
embolic risk at 3 months. The Tinaquant test (an
immunoturbimetric assay, Roche) has been validated in
more than 2000 patients,34 and showed a similarly low
thromboembolic risk at 3 months. Finally, the SimpliRed
assay (a whole blood bedside latex assay, Agen
Biomedical) is well validated,34 but interobserver
variability might be an issue.35
D-dimer tests have restricted specifi city and are less
useful than other measures in some groups of patients,
including in those with high clinical probability, those
admitted to hospital for another reason in whom the
CUS or MDCTA
High (or likely)
Figure 1: A diagnostic algorithm for clinically suspected deep vein thrombosis or pulmonary embolism
Use of CUS with suspected deep vein thrombosis, and of multidetector CT angiography with pulmonary embolism.
CUS=compression ultrasonography. MDTCA=multidetection CT angiography.
Wells score for DVT*
Paralysis or recent plaster cast
Bed rest >3 days or surgery <4 weeks
Pain on palpation of deep veins
Swelling of entire leg
Diameter diff erence on aff ected calf >3 cm
Pitting oedema (aff ected side only)
Dilated superfi cial veins (aff ected side)
Alternative diagnosis at least as probable as DVT
Wells score for PE†
Previous PE or DVT
Heart rate >100 beats per min
Recent surgery or immobilisation
Clinical signs of DVT
Alternative diagnosis less likely than PE
Revised Geneva score for PE‡
Age >65 years
Previous DVT or PE
Surgery (under general anaesthesia) or fracture (of the lower
limbs) within 1 month
Active malignancy (solid or haematological malignancy,
currently active or considered as cured since less than 1 year)
Unilateral leg pain
Heart rate 75–94 beats per min
Heart rate ≥95 beats per min
Pain on deep vein palpation in leg and unilateral oedema
Scoring systems to assess probability of suspected DVT or PE on the basis of
item and assigned points. DVT=deep vein thrombosis. PE=pulmonary
embolism. *Patients with a score of 0 are low risk, 1–2 are intermediate risk,
and ≥3 are high risk. †For the initial rule, patients with a score of 0–1 are low
risk, 2–6 are intermediate risk, and ≥7 are high risk; for the dichotomised rule,
patients are unlikely or likely to have PE if they have scores ≥4 and ≤4,
respectively. ‡Patients with a score <2 are low risk, 2–6 are intermediate risk,
and ≥6 are high risk.
Table 1: Clinical probability assessment
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1
suspicion of pulmonary embolism is raised during
hospital stay, individuals older than 65 years, and
pregnant women.33,36 A proposed age-adjusted diagnostic
threshold for suspected pulmonary embolism increases
the usefulness of D-dimer measurement in elderly
patients,37 but clinical implementation should await
prospective external validation.
Compression ultrasonography for diagnosing deep
Compression ultrasonography has largely replaced
venography as the main imaging procedure to diagnose
deep vein thrombosis (fi gure 2). Three options are presently
used in clinical practice. Some groups look only at proximal
(above the calf) veins and, in patients with a negative fi rst
exam, repeat the exam 1 week later to detect clinically
relevant distal thrombi that might have progressed
proximally. This method is resource demanding,
cumbersome, and has a very low yield (about 1–2% of
results are positive in the second exam). A second approach
is to assess proximal and distal veins with complete
compression ultrasonography, which is associated with a
low rate of thromboembolism at 3 months.38 However, this
approach leads to anti coagulation of many patients with
isolated deep vein thrombosis of the distal calf and might
increase risk of bleeding in some patients undergoing this
procedure.39 A third approach consists of use of a single
proximal compression ultrasonography. Deep vein
thrombosis can be excluded with this technique if results
are negative in patients with a low or intermediate clinical
probability, whereas those with a high clinical probability
and a negative proximal compression ultrasonography
would qualify for additional imaging (distal veins
ultrasound imaging or venography) or serial ultrasound
surveillance (fi gure 1). This approach seems to be associated
with a 3-month risk of venous thromboembolism that is
very similar to that of complete compression ultra-
sonography, with 30–50% fewer patients prescribed
anticoagulant treatment compared with the complete
compression ultrasonography strategy.40
In 2012, revised American College of Chest Physicians
(ACCP) guidelines recommend against routine treatment
of asymptomatic isolated calf deep venous thrombosis.
Previous guidelines41 recommended that treatment of
distal deep vein thrombosis be similar to that of proximal
deep vein thrombosis.
For suspected pulmonary embolism, diagnosis of
proximal deep vein thrombosis in a symptomatic patient,
or in an asymptomatic patient who has contraindications
to CT angiography, is considered suffi cient to rule in
Multidetector CT angiography for diagnosing
CT angiography (fi gure 2) has largely replaced ventilation-
perfusion (V/Q) lung scintigraphy as the main imaging
modality in suspected pulmonary embolism. Single-
detector CT angiography has a sensitivity of only about
70%42 and needs combination imaging with compression
ultra sonography of the proximal veins of the leg if
negative.43,44 Multidetector CT angiography is more
sensitive than single-detector CT angiography.31,45,46 This
technological advance allows exclusion of pulmonary
embolism without additional compression ultrasono-
graphy of the leg.47 Overall, a meta-analysis48 that
compiled 23 studies with 4657 patients with a negative
CT angiography (mainly single detector) who did not
receive anti coagulation showed a 3-month rate of
subsequent venous thromboembolism of 1·4% (95% CI
1·1–1·8) and a 3-month rate of fatal pulmonary embolism
of 0·51% (0·33–0·76), which compares favourably with
the results noted after a normal invasive contrast
Notably, the increased use of CT angiography might
cause an increased incidence of cancer attributable
to radiation.50 Dangers of radiation mean that protocols
for CT angiography should be optimised. For this
reason, combined use of CT pulmonary angiography and
CT venography should be questioned. In the Prospective
Figure 2: Contemporary imaging of deep vein thrombosis with compression
ultrasound or pulmonary embolism with CT angiography
Compression ultrasound (A): upper series, from left to right; representation of
vein and artery without and with (arrow) gentle compression with the
echocardiographic probe; lower series, corresponding echocardiographic
fi ndings. The third image from the left show a thrombus in the vein (vein not
compressible by the probe). CT angiography (B): CT angiography showing
several emboli (arrows) in the main right pulmonary artery and in left lobar and
segmental arteries. A=artery. V=vein.
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1 5
Investigation of Pulmonary Embolism Diagnosis
(PIOPED) II study,51 no patient with pul monary embolism
or deep vein thrombosis would have been undiagnosed if
imaging of the pelvic veins had been omitted. The
radiation risk is particularly important in pregnant
women in whom the respective advantages of CT
angiography versus ventilation-perfusion or perfusion-
only lung scintigraphy are still debated.
Other diagnostic imaging modalities for suspected deep
vein thrombosis and pulmonary embolism
Gadolinium-enhanced magnetic resonance pulmonary
angiography (MRA) could be used to diagnose pulmonary
embolism because it is devoid of radiation. The accuracy
of this technique combined with magnetic resonance
venography (MRV) has been studied in the prospective,
multicentre PIOPED III accuracy study.52 The proportion
of technically inadequate images ranged from 11% to
52% across the seven participating centres. Technically
adequate MRA had a sensitivity of 78% and a specifi city
of 99%, whereas technically adequate MRA and MRV
had a sensitivity of 92% and a specifi city of 96%. However,
194 (52%) of 370 patients had technically inadequate
results, which substantially restricts its clinical use.
Conventional pulmonary angiography and venography
remain the gold standards for diagnosis of pulmonary
embolism and deep vein thrombosis, respectively. Because
these exams are invasive, they should be restricted to
patients in whom a clinically likely diagnosis cannot be
confi rmed by other means, or in whom endovascular
treatment of pulmonary embolism is being considered.
Table 2 summarises the performance of some diagnostic
tests or algorithms to rule in or rule out pulmonary
embolism on the basis of a systematic review.53
Prognostic stratifi cation of patients with
Patients with pulmonary embolism should be stratifi ed
according to prognosis.54 The Pulmonary Embolism
Severity Index55 and its simplifi ed version56 allow such
stratifi cation on a clinical basis (table 3). Several thera-
peutic implications exist for patients with pulmonary
embolism: (1) high-risk patients (who represent about 5%
of all symptomatic patients, with about a 15% short-term
mortality) should be treated aggressively with thrombolytic
drugs or surgical or catheter embolectomy;57 (2) low-risk
patients (most patients with pulmonary embolism), with a
short-term mortality of about 1% might benefi t from early
discharge or even outpatient treatment;58 (3) intermediate-
risk patients (who represent about 30% of all symptomatic
patients) should probably be admitted to hospital, with
potential benefi t of thrombolytic treatment, pending
results of ongoing clinical trials. Low-risk and intermediate-
risk categories are referred to as non-massive pulmonary
embolism. Echocardiography or measurement of bio-
markers, such as troponin or pro-brain natriuretic peptide,
might refi ne prognostic stratifi cation,59 but whether their
addition to the risk stratifi cation work-up is cost-eff ective
remains to be established.
To rule in PE
High-probability ventilation perfusion lung
Positive proximal vein CUS of the leg
To rule out PE
Normal or near normal ventilation perfusion lung
Negative CTA (mainly single detector)
Negative CTA and proximal vein CUS of the leg
Negative proximal vein CUS of the leg
Quantitative ELISA D-dimer assay less than 500 μg/L
Likelihood ratios to rule in PE are positive and to rule out PE are negative. The
likelihood ratio is the likelihood that a given test result would be expected in a
patient with the target disorder compared with the likelihood that that same result
would be expected in a patient without the target disorder—eg, a positive ratio of
20 means that, with the given test result, the patient is 20 times more likely to have
PE than not to have PE. Conversely, with a negative ratio of 0·10, with the given test
result, the patient is 10 times less likely to have PE than to have PE. PE=pulmonary
embolism. CTA=CT angiography. CUS=compression ultrasonography.
Table 2: Performance of some tests or diagnostic algorithms to rule in or
rule out PE46
Pulmonary embolism severity index*
Age >80 years
History of cancer
History of heart failure
History of chronic lung disease
Heart rate ≥110 beats per min
Systolic blood pressure <100 mmHg
Respiratory rate ≥30 breaths per min
Altered mental status
Arterial oxygen saturation <90%
Simplifi ed pulmonary embolism severity index
according to RIETE†
Age >80 years
History of cancer
History of heart failure or chronic lung disease
Heart rate ≥110 beats per min
Systolic blood pressure <100 mm Hg
Arterial oxygen saturation <90%
Age in years
In the pulmonary embolism severity index score, classes 1 and 2 are considered
low risk, and classes 3–5 high risk. RIETE=Registry of Patients with Venous
Thromboembolism. PE=pulmonary embolism. *Class 1=≤65; class 2=66–85;
class 3=86–105; class 4=106–125; class 5=>125. †Patients with a score of 0 are low
risk; those with scores ≥1 are high risk.
Table 3: Prognostic stratifi cation of PE
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1
Standard treatment of deep vein thrombosis and
Treatment of non-massive venous thromboembolism
has three phases: the initial phase, the early mainten-
ance phase, and the long-term secondary prevention
phase (fi gure 3). Low-molecular-weight heparin and
fondaparinux are the cornerstones of initial treatment
for patients with deep vein thrombosis and pulmonary
embolism.41 Heparins act by binding to the natural
anticoagulant antithrombin, thereby substantially accel-
erating the inactivation of thrombin by antithrombin
and of several other activated coagulation factors
(including activated factor X [FXa]). Unfractionated
heparin is usually administered as an initial bolus,
followed by a continuous intravenous infusion. Because
of a large individual diff erence in the binding of heparins
to plasma proteins, the doses should be adjusted to the
results of blood tests, such as the activated partial
thromboplastin time or the anti-FXa activity.
The main advantage of low-molecular-weight heparins
is that they can be administered subcutaneously in fi xed
weight-adjusted doses without needing monitoring in
most cases.62 The mechanism of action of these heparins
is similar to that of unfractionated heparin, but with
a more pronounced eff ect on FXa compared with
thrombin. The clinical equivalence of low-molecular-
weight heparin and unfractionated heparin for treating
deep vein thrombosis has been confi rmed in a meta-
analysis.63 One study confi rmed this conclusion for
pulmonary embolism.64 Fondaparinux, a pentasaccharide,
is almost identical to the smallest natural component of
heparin that can still bind to antithrombin to specifi cally
inhibit FXa. By contrast with un fractionated heparin
and low-molecular-weight heparins, which are derived
from the porcine intestinal tract, fondaparinux is a
synthetic compound. This drug is non-inferior to low-
molecular-weight and unfractionated heparin in patients
with deep vein thrombosis65,66 and pulmonary embol-
Low-molecular-weight heparin and fondaparinux are
mainly cleared by the kidney. Particular caution is
advised when the calculated creatinine clearance is less
than 30 mL/min. In such cases, anticoagulation options
include dose reduction, increase of the interval between
injections, monitoring of FXa activity, or use of
unfractionated heparin.62 Administration of heparins or
fondaparinux should overlap during at least 5 days with
that of vitamin K antagonists. The parenteral drug can
be stopped when the anticoagulant concentration
induced by the vitamin K antagonist has reached an
international normalised ratio of 2·0. Patients with
cancer have been recommended to be treated for at least
3 months with low-molecular-weight heparin rather
than with vitamin K antagonists.41 These antagonists
block a late step in the biosynthesis of four plasma
coagulation factors (pro thrombin or factor II, and factors
VII, IX, and X) by the liver. Because of the diff erent half-
lives of circulating factors, steady-state anticoagulation
cannot be reached before 4–7 days. Vitamin K antagonists
include substances with a short (acenocoumarol),
intermediate (warfarin, fl uindione), or long (phenpro-
coumone) half-life. For this reason, and because of
genetically induced metabolic variability,67,68 the variable
vitamin K content of food, a narrow therapeutic index,
and several interactions with other drugs, treatment
with vitamin K antagonists needs close monitoring with
the international normalised ratio; the targeted
therapeutic level is 2·5 (range 2·0–3·0). Although
thrombolysis, regardless of mode of admini stration, is
not better than standard treatment, it could be used in
selected patients (especially those with iliac or iliofemoral
vein thrombosis and massive pulmonary embolism) if
experience and resources are available.
Safety of anticoagulant treatment
All anticoagulant drugs can produce bleeding, especially
at the start of treatment (eg, caused by unmasking of
lesions). Major bleeding associated with vitamin K
antagonists increases with age. Clinical scores—eg, the
HEMORR2HAGES score69 and the RIETE score70
(table 4)—have been prospectively validated (not in
venous thromboembolism for the HEMORR2HAGES
score), and could guide estimation of the haemorrhagic
risk. The safety of treatment with vitamin K antagonists
can be improved by encouragement of patients’ com-
pliance, avoidance of concurrent drugs with potential
New potential treatment schemes with the novel oral anticoagulant drugs
VKA INR 2·0–3·0
Long-term secondary prevention
VKA INR 2·0–3·0*
>3 months, years, or indefinite
with periodic reassessment
Single drug approach
Figure 3: Three phases of the disease with the corresponding
A and B depict potentially new treatment schemes that are based on the
regimen studied in the RECOVER with dabigatran etexilate (A)60 or EINSTEIN
DVT with rivaroxaban (B)61 studies. UFH=unfractionated heparin.
LMWH=low-molecular-weight heparin. VKA=vitamin K antagonist.
INR=international normalised ratio. *In some patients in whom less frequent
INR monitoring is requested, an INR of 1·5–2·0 can also be targeted (American
College of Chest Physicians Grade 1A recommendation).41
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1 7
interactions, restriction of alcohol ingestion, and, in
some patients, with use of self-monitoring or even self-
management,71 which remains debated.72,73 Additionally,
large loading doses should be avoided to prevent
development of a paradoxical prothrombotic state due to
the depletion of protein C, a vitamin K-dependent coagu-
lation inhibitor with a very short half-life. Whether rapid
turnaround genetic testing will be clinically useful to
guide warfarin dosing remains to be established.68
Heparin-induced thrombocytopenia is a feared compli-
cation of treatment with heparin and low-molecular-
weight heparin. Although this complication is rare
(extremely rare with fondaparinux), it can provoke
devastating venous and arterial thromboembolic con-
sequences.74 However, monitoring of platelet counts
during treatment with unfractionated and low-molecular-
weight heparin has become controversial because of
overdiagnosis simply on the basis of a positive heparin-
PF4 test. Monitoring of platelet function should not be
routinely pursued after 14 days, and should always be
combined with clinical risk assessment for heparin-
Treatment duration after deep vein thrombosis and
The duration of anticoagulation treatment should be
dictated by the balance between the risk of recurrent
venous thromboembolism with and without treatment,
and the risk of treatment-induced haemorrhage. In
RIETE,75 the rate of recurrent venous thromboembolism
while patients were receiving anticoagulant treatment
was 7·0%. In the literature review that supports the
treatment durations recommended by the 8th ACCP
consensus guidelines (table 5), Kearon and colleagues41
reported that a 3-month course of anticoagulant treatment
was as eff ective as a course of 6–12 months, and that
venous thromboembolism related to transient (reversible)
risk factors (eg, surgery, trauma) is associated with a
reduced risk of recurrence.
The decision about the optimum duration of anti-
coagulation can be approached on an individual basis
that recognises clinical variables,76 D-dimer concentration
1 month after stopping of anticoagulant treatment,77 or
presence of residual thrombi in the leg veins.78 These
potential methods have not gained widespread atten-
tion.79 Presently, all patients with deep vein thrombosis
or pulmonary embolism should be treated for at least
3 months. In case of a transient or reversible risk factor,
especially if this risk factor was the clear precipitant of
venous thromboembolism, anticoagulant treatment
might then be stopped. In patients with no triggering
risk factor (the so-called idiopathic or unprovoked
events), anticoagulant treatment should be continued as
long as the benefi t–risk balance is favourable, whereas
patients with venous thrombo embolism and cancer
should receive anticoagulant treatment until the cancer
is considered under control and possibly cured.
Advances in anticoagulant treatment
Several new oral anticoagulant drugs are under develop-
ment.80 These direct (ie, antithrombin-independent)
inhibitors of FXa (eg, rivaroxaban, apixaban) or thrombin
(eg, dabigatran) avoid most of the drawbacks of heparin
and could replace vitamin K antagonists and heparins in
HEMORR2HAGES bleeding risk score*
Hepatic or renal disease
Age >75 years
Excessive risk of fall
Reduced platelet count or function
RIETE bleeding risk score†
Recent major bleed
Creatininaemia >1·2 mg/dL
Haemoglobin <13 g/dL (male) or 12 g/dL (female)
Clinically overt PE
Age >75 years
RIETE=Registry of Patients with Venous Thromboembolism. PE=pulmonary
embolism. *Patients with a score of 0 have a major bleeding rate (per 1000
patient-years) of 1·9, scores of 1 have a bleeding rate of 2·5, scores of 2 have a
bleeding rate of 5·3, scores of 3 have a bleeding rate of 8·4, scores of 4 have a
bleeding rate of 10·4, and those with scores >5 have a bleeding rate of 12·3.
†Patients with a score of 0 have a major bleeding rate (per 1000 patient-years) of
0·3, scores of 1–3 have a bleeding rate of 2·6, and those with scores ≥4 have a
bleeding rate of 7·3.
Table 4: Clinical scores to predict bleeding with anticoagulant treatment
Recommended treatment durationGrade of
First DVT or PE secondary to a transient
(reversible) risk factor (provoked event)
First idiopathic (unprovoked) DVT or PE
At the end of the initial 3-month period
In the absence of contraindication
During long-term Rx
Recurrent DVT or PE or strong
DVT or PE secondary to cancer
At least 3 months
Assess for long-term Rx
Assess risk–benefi t balance periodically
Long-term Rx, preferentially with LMWH
during the fi rst 3-6 months, then
anticoagulate as long as the cancer is
Recommendation according to the eighth American College of Chest Physicians evidence-based clinical practice guidelines.41
Grade 1 recommendations pertain to a situation in which the desirable eff ects clearly outweigh the undesirable eff ects. A and
C qualify the methodological quality of the supporting evidence: A=consistent evidence is available from several randomised
controlled trials. C=evidence is available from at least one critical outcome from observational studies, cases series, or
randomised controlled trials with fl aws. DVT=deep vein thrombosis. PE=pulmonary embolism. Rx=treatment.
Table 5: Recommended duration of anticoagulant treatment for events of venous thromboembolism
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1
many patients. These drugs are administered in fi xed
doses, do not need coagulation monitoring in the
laboratory, and have very few drug–drug or drug–food
interactions. In the randomised, double-blind RE-
COVER trial,60 which involved patients with acute venous
thromboembolism who were initially given parenteral
anticoagulation treatment for a median of 9 days (IQR
8–11), oral dabigatran etexilate 150 mg twice a day with
no monitoring was non-inferior to warfarin (target
international normalised ratio of 2·0–3·0) bridged with
low-molecular-weight heparin, with a similar safety
profi le.81 In the multi centre, randomised, EINSTEIN-
DVT and EINSTEIN-EXTENSION studies,61 rivaroxaban
(15 mg twice a day for 3 weeks followed by 20 mg once a
day with no monitoring) was non-inferior to a vitamin K
antagonist bridged with low-molecular-weight heparin,
with a similar safety profi le. For long-term secondary
prophy laxis, rivaroxaban (20 mg once a day) was better
than placebo, with 82% (HR 0·18, 95% CI 0·09–0·39;
p<0·001) relative risk reduction of recurrent thrombo-
embolic events and no increase in the risk of major
bleeding. However, the rate of clinically relevant non-
major bleeding diff ered signifi cantly between the two
groups, increasing from 1·2% in the placebo group to
5·4% in the rivaroxaban group. Figure 3 shows the
potential of these new drugs to aff ect the therapeutic
concept of acute venous thromboembolism.
Findings from rigorous clinical trials have shown the
eff ectiveness and safety of pharmacological prevention
with low, fi xed doses of anticoagulant drugs (panel 2).
For patients undergoing orthopaedic surgery—eg, total
hip or knee replacement—novel oral anticoagulant
drugs have been approved for thromboprophylaxis and
are available instead of warfarin, heparins, and fonda-
parinux. Mechanical prophylactic measures, including
graduated compression stockings and intermittent
pneumatic compression devices, should be considered in
at-risk patients who are not candidates for pharmacological
thromboprophylaxis. Inferior vena caval fi lters can also
be used for the primary or secondary prevention of
pulmonary embolism, but they will not halt the
thrombotic process. In the USA, use of inferior vena
caval fi lters seems to have substantially increased for
primary prevention of venous thromboembolism.82
Although prophylaxis for venous thromboembolism is
mandated for moderate-risk and high-risk patients at the
time of hospital admission,83 the decision to continue
prophylaxis after discharge remains diffi cult. The risk of
venous thromboembolism during admission rarely abates
by the time a patient is ready for discharge home or to a
skilled nursing facility. Rapid transition of patients to
skilled nursing or rehabilitation facilities and rapid
discharge home with home services have shortened
lengths of hospital stay. During admission to hospital,
nurses and therapists encourage patients to ambulate and
minimise immobilisation. Patients often receive less
physical therapy after discharge than during admission,
which leads to a paradoxical increase in immobility and a
presumed rise in risk of venous thromboembolism. Early
hospital discharge minimises the hospital length of stay
but blurs the traditional concept of inpatient versus
ambulatory care. For example, the risk of venous
thromboembolism remains increased in women for the
fi rst 12 weeks after surgery.84
A contemporary approach to prevention of venous
thromboembolism focuses on the continuum of care from
hospital to the community. Thus, extended prophylaxis up
to 5 weeks is recommended after total hip arthroplasty.83
The MAGELLAN trial85 of medical patients admitted to
hospital (presented at the 2011 American College of
Cardiology Scientifi c Sessions) reported that in those
receiving traditional enoxaparin prophylaxis for 6–14 days
for disorders such as heart failure, respiratory failure, or
pneumonia, the incidence of death related to venous
thromboembolism at 5 weeks was 1·0%, with most deaths
occurring after hospital discharge. Findings from a
review86 of 1897 patients with venous thromboembolism
in the Worcester, Massachusetts health-care system
showed that 74% of patients suff ered deep vein thrombosis
or pulmonary embolism in the outpatient setting, not
during a hospital admission. 37% of patients with venous
thromboembolism had recently been admitted to hospital,
and 23% had undergone major surgery in the 3 months
before developing acute venous thromboembolism. Of
the episodes of venous thromboembolism occurring
within 3 months of a previous admission, 67% occurred
within the fi rst month after discharge. The median length
of admission was 4 days.
In the EXCLAIM Trial,87 extended duration prophylaxis
for venous thromboembolism was tested after hospital
discharge in high-risk medical patients with heart
failure, respiratory insuffi ciency, infection, or reduced
mobility. Incidence of venous thromboembolism was
reduced in patients receiving extended prophylaxis after
Panel 2: Pharmacological prophylaxis for venous
• Low-dose unfractionated heparin twice or three times a day
• Low-molecular-weight heparins
• Fondaparinux 2·5 mg per day for orthopaedic surgical or
general surgical procedures or, in some countries, for
acute medical illness (also often used off label when
heparin-induced thrombocytopenia is suspected)
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1 9
discharge with enoxaparin 40 mg/day. However, a
substantial methodological issue with EXCLAIM was
the change in enrolment eligibility halfway through the
study;88 the inclusion criteria were made more restrictive
than at the start of the study and required that patients
be extremely immobile to participate in the trial. Overall,
extended duration enoxaparin reduced the rate of venous
thrombo embolism at 28 days from 4·0% in the placebo
group to 2·5% in the enoxaparin group (absolute risk
diff erence –1·53, 95% CI –2·54 to –0·52). Major
haemorrhage at 30 days was more frequent in patients
receiving extended duration enoxaparin than in those
receiving placebo. In the IMPROVE registry89 of
15 156 medical patients admitted to hospital, 45% of the
184 patients who developed venous thromboembolism
had hospital events after discharge rather than in
hospital. Independent risk factors for venous thrombo-
embolism were previous venous thromboembolism,
known thrombophilia, cancer, age older than 60 years,
leg paralysis, immobilisation for at least 1 week, or
admission to an intensive-care or coronary-care unit.
The biggest diffi culty in the specialty of in-hospital
prophylaxis of venous thromboembolism is underuse of
available prophylactic anticoagulant drugs. In a review90
of almost 200 000 charts of US medical patients at
moderate-risk or high-risk of venous thromboembolism
who were admitted to hosiptal, appropriate prophylaxis
for venous thromboembolism was ordered in only 34%.
In a separate cohort study of patients admitted to hospital
with deep vein thrombosis from 183 US institutions, the
2609 medical patients had more concomitant pulmonary
embolism than did the 1953 non-medical patients with
deep vein thrombosis (22% vs 16%).91 Paradoxically,
patients on the medical service had received prophylaxis
for venous thromboembolism far less frequently than
had non-medical patients (25% vs 54%). Thus, patients
on the medical service are susceptible to so-called double
trouble because they more often have prophylaxis
omitted, but when they do develop venous thrombo-
embolism, it is often more extensive with more frequent
concomitant pulmonary embolism compared with non-
medical patients who develop deep vein thrombosis.92
Failure to prevent venous thromboembolism happens
worldwide. In ENDORSE, a cross-sectional study,
68 183 patients were enrolled from 358 hospitals in
32 countries across six continents. Of these patients, 52%
were at moderate to high risk of developing venous
thromboembolism. Although rates of prophylaxis were
low, surgical patients more often received guideline
recommended prophylaxis than did medical patients
(58% vs 40%).93 Of the 9257 US patients from 81 hospitals
enrolled in ENDORSE, wide variation was noted in
prophylaxis practices for venous thromboembolism. The
top quartile of hospitals implemented prophylaxis in 74%
of at-risk patients, whereas the bottom quartile imple-
mented prophylaxis in only 40%. Compared with the
lowest quartile, more hospitals in the best performing
quartile had residency training programmes (43% vs 5%),
a larger number of beds (277 vs 140), and had formulated
and implemented individualised hospital-wide prophylaxis
protocols for venous thrombo embolism (76% vs 40%).94
In Switzerland, prophylaxis was not provided to 40% of
257 patients with cancer admitted to hospital before the
onset of an acute venous thromboembolic event.95
Regardless of the specifi c prophylaxis strategy selected
for venous thromboembolism, institutional and profes-
sional culture seems to be changing. Failure to institute
venous thromboembolism prophylaxis in at-risk patients
will no longer be tolerated. Panel 3 lists catalysts for
improved implementation of venous thromboembol ism
prophylaxis. However, even when pharmacological
prophyl axis is ordered for patients admitted to hospital,
these orders are not necessarily followed. In one study,96
patient refusal was the most common reason for non-
adherence to injectable anticoagulant medication for
Diverse approaches are available to improve clinical
eff ectiveness of prophylaxis for venous thrombo embol-
ism in patients admitted to hospital. Computerised
decision support97 with electronic alerts can be a catalyst
to the responsible physician to order prophylaxis and,
in a randomised controlled trial,98 has reduced the rate
of symptomatic venous thrombo embolism by more
than 40%. Multiscreen alerts might be more eff ective
than single-screen alerts.99 Such electronic alert systems
maintain their eff ectiveness over time.100 For hospitals
without the resources to establish and maintain
computerised systems, hospital staff can screen for at-
risk patients not being given prophylaxis and can alert
the responsible physician with a telephone call or
page.101 Eradication of most hospital-acquired venous
thrombo embolism is within our reach. By combination
of educational eff orts with behaviour modifying
techniques, implementation of proven prevention
strategies can be maximised.102
Panel 3: Catalysts for improved implementation of
prophylaxis for venous thromboembolism
• Evidence from clinical trials
• Expanded educational outreach to clinicians and the public
• Initiatives for quality improvement, including
individualised hospital protocols for prophylaxis
• Electronic reminders to clinicians whose patients admitted
to hospital are at high risk, but not given prophylaxis
• Peer pressure
• Oversights in hospital administration
• Government audits and inspection
• Patient and family inquiries
• Advocacy of non-profi t organisations
• Financial penalty or withholding of a fi nancial bonus
imposed by the government or private insurer
• Malpractice litigation
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1
SZG and HB conceptualised and wrote the paper.
Confl icts of interest
SZG has received research grants from Boehringer Ingelheim,
Bristol-Myers Squibb, Eisai, EKOS, Johnson & Johnson, and
Sanofi -Aventis; fees for consultancy from Baxter, Boehringer Ingelheim,
Bristol-Myers Squibb, Daiichi Sankyo, Eisai, EKOS, Medscape, Merck,
Pfi zer, Portola, and Sanofi -Aventis. HB has received research grants
from Bayer Schering Pharmaceutical AG, honoraria from Bayer
Schering Pharmaceutical, Daiichi Sankyo, GlaxoSmithKline, Pfi zer,
Sanofi -Aventis, and Servier, and fees for serving on advisory boards for
Bayer Schering Pharmaceutical, Boehringer Ingelheim, Canyon
Pharmaceuticals, Daiichi Sankyo, GlaxoSmithKline, Pfi zer, and
We thank Kathryn Mikkelsen for providing editorial support in the
preparation of this Seminar.
1 Kucher N. Clinical practice. Deep-vein thrombosis of the upper
extremities. N Engl J Med 2011; 364: 861–69.
2 Naess IA, Christiansen SC, Romundstad P, Cannegieter SC,
Rosendaal FR, Hammerstrøm J. Incidence and mortality of venous
thrombosis: a population-based study. J Thromb Haemost 2007;
3 Nordström M, Lindblad B, Bergqvist D, Kjellström T. A prospective
study of the incidence of deep-vein thrombosis within a defi ned
urban population. J Intern Med 1992; 232: 155–60.
4 Heit JA. The epidemiology of venous thromboembolism in the
community. Arteriosclerosis Thromb Vasc Biol 2008; 28: 370–72.
5 Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism:
clinical outcomes in the International Cooperative Pulmonary
Embolism Registry (ICOPER). Lancet 1999; 353: 1386–89.
6 Spencer FA, Goldberg RJ, Lessard D, et al. Factors associated with
adverse outcomes in outpatients presenting with pulmonary
embolism: the Worcester Venous Thromboembolism Study.
Circ Cardiovasc Qual Outcomes 2010; 3: 390–94.
7 Laporte S, Mismetti P, Décousus H, et al. Clinical predictors for
fatal pulmonary embolism in 15,520 patients with venous
thromboembolism: fi ndings from the Registro Informatizado de la
Enfermedad TromboEmbolica venosa (RIETE) Registry. Circulation
2008; 117: 1711–16.
8 Pollack CV, Schreiber D, Goldhaber SZ, et al. Clinical
characteristics, management, and outcomes of patients diagnosed
with acute pulmonary embolism in the emergency department:
initial report of EMPEROR (Multicenter Emergency Medicine
Pulmonary Embolism in the Real World Registry). J Am Coll Cardiol
2011; 57: 700–06.
9 Ng AC, Chung T, Yong AS, et al. Long-term cardiovascular and
noncardiovascular mortality of 1023 patients with confi rmed acute
pulmonary embolism. Circ Cardiovasc Qual Outcomes 2011; 4: 122–28.
10 Piazza G, Goldhaber SZ. Chronic thromboembolic pulmonary
hypertension. N Engl J Med 2011; 364: 351–60.
11 Kahn SR. The post-thrombotic syndrome.
Hematology Am Soc Hematol Educ Prog 2010; 2010: 216–20.
12 Kahn SR, Shrier I, Julian JA, et al. Determinants and time course of
the postthrombotic syndrome after acute deep venous thrombosis.
Ann Intern Med 2008; 149: 698–707.
13 Klok FA, Zondag W, van Kralingen KW, et al. Patient outcomes after
acute pulmonary embolism. A pooled survival analysis of diff erent
adverse events. Am J Respir Crit Care Med 2010; 181: 501–06.
Piazza G, Goldhaber SZ. Venous thromboembolism and
atherothrombosis: an integrated approach. Circulation 2010;
15 Sim DS, Jeong MH, Kang JC. Current management of acute
myocardial infarction: experience from the Korea Acute Myocardial
Infarction Registry. J Cardiol 2010; 56: 1–7.
16 Folsom AR, Lutsey PL, Astor BC, Cushman M. C-reactive protein
and venous thromboembolism. A prospective investigation in the
ARIC cohort. Thromb Haemost 2009; 102: 615–19.
17 Chandra D, Parisini E, Mozaff arian D. Meta-analysis: travel and risk
for venous thromboembolism. Ann Intern Med 2009; 151: 180–90.
18 Severinsen MT, Kristensen SR, Johnsen SP, et al. Smoking and
venous thromboembolism: a Danish follow-up study.
J Thromb Haemost 2009; 7: 1297–303.
19 Ageno W, Dentali F, Grandi AM. New evidence on the potential role
of the metabolic syndrome as a risk factor for venous
thromboembolism. J Thromb Haemost 2009; 7: 736–38.
20 Baccarelli A, Martinelli I, Pegoraro V, et al. Living near major traffi c
roads and risk of deep vein thrombosis. Circulation 2009; 119: 3118–24.
21 Blanco-Molina A, Trujillo-Santos J, Tirado R, et al. Venous
thromboembolism in women using hormonal contraceptives.
Findings from the RIETE Registry. Thromb Haemost 2009;
22 Lee AY. Thrombosis in cancer: an update on prevention, treatment,
and survival benefi ts of anticoagulants.
Hematology Am Soc Hematol Educ Prog 2010; 2010: 144–49.
23 Piazza G, Goldhaber SZ. Pulmonary embolism in heart failure.
Circulation 2008; 118: 1598–601.
24 Dalen JE. Should patients with venous thromboembolism be
screened for thrombophilia? Am J Med 2008; 121: 458–63.
25 Corral J, Roldan V, Vicente V. Deep venous thrombosis or
pulmonary embolism and factor V Leiden: enigma or paradox.
Haematologica 2010; 95: 863–66.
26 Kyrle PA, Rosendaal FR, Eichinger S. Risk assessment for recurrent
venous thrombosis. Lancet 2010; 376: 2032–39.
27 Roy PM, Meyer G, Vielle B, et al. Appropriateness of diagnostic
management and outcomes of suspected pulmonary embolism.
Ann Intern Med 2006; 144: 157–64.
28 Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple
clinical model to categorize patients probability of pulmonary
embolism: increasing the models utility with the SimpliRED
D-dimer. Thromb Haemost 2000; 83: 416–20.
29 Le Gal G, Righini M, Roy PM, et al. Prediction of pulmonary
embolism in the emergency department: the revised Geneva score.
Ann Intern Med 2006; 144: 165–71.
30 Wells PS, Anderson DR, Bormanis J, et al. Value of assessment of
pretest probability of deep-vein thrombosis in clinical management.
Lancet 1997; 350: 1795–98.
31 van Belle A, Büller HR, Huisman MV, et al. Eff ectiveness of
managing suspected pulmonary embolism using an algorithm
combining clinical probability, D-dimer testing, and computed
tomography. JAMA 2006; 295: 172–79.
32 Ceriani E, Combescure C, Le Gal G, et al. Clinical prediction rules
for pulmonary embolism: a systematic review and meta-analysis.
J Thromb Haemost 2010; 8: 957–70.
33 Righini M, Perrier A, De Moerloose P, Bounameaux H. D-Dimer
for venous thromboembolism diagnosis: 20 years later.
J Thromb Haemost 2008; 6: 1059–71.
34 Carrier M, Righini M, Djurabi RK, et al. VIDAS D-dimer in
combination with clinical pre-test probability to rule out pulmonary
embolism. A systematic review of management outcome studies.
Thromb Haemost 2009; 101: 886–92.
35 de Monyé W, Huisman MV, Pattynama PM. Observer dependency
of the SimpliRed D-dimer assay in 81 consecutive patients with
suspected pulmonary embolism. Thromb Res 1999; 96: 293–98.
36 Righini M, Le Gal G, Perrier A, Bounameaux H. The challenge of
diagnosing pulmonary embolism in elderly patients: infl uence of
age on commonly used diagnostic tests and strategies.
J Am Geriatr Soc 2005; 53: 1039–45.
37 Douma RA, le Gal G, Söhne M, et al. Potential of an age adjusted
D-dimer cut-off value to improve the exclusion of pulmonary
embolism in older patients: a retrospective analysis of three large
cohorts. BMJ 2010; 340: c1475.
38 Johnson SA, Stevens SM, Woller SC, et al. Risk of deep vein
thrombosis following a single negative whole-leg compression
ultrasound: a systematic review and meta-analysis. JAMA 2010;
39 Righini M, Paris S, Le Gal G, et al. Clinical relevance of distal deep
vein thrombosis. Review of literature data. Thromb Haemost 2006;
40 Ten Wolde M, Hagen PJ, Macgillavry MR, et al. Non-invasive
diagnostic work-up of patients with clinically suspected pulmonary
embolism; results of a management study. J Thromb Haemost 2004;
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1 11
41 Kearon C, Kahn SR, Agnelli G, et al. Antithrombotic therapy for
venous thromboembolic disease: American College of Chest
Physicians Evidence-Based Clinical Practice Guidelines
(8th edition). Chest 2008; 133 (6 suppl): 454S–545S.
42 Perrier A, Howarth N, Didier D, et al. Performance of helical
computed tomography in unselected outpatients with suspected
pulmonary embolism. Ann Intern Med 2001; 135: 88–97.
43 Anderson DR, Kovacs MJ, Dennie C, et al. Use of spiral computed
tomography contrast angiography and ultrasonography to exclude
the diagnosis of pulmonary embolism in the emergency
department. J Emerg Med 2005; 29: 399–404.
44 Musset D, Parent F, Meyer G, et al. Diagnostic strategy for patients
with suspected pulmonary embolism: a prospective multicentre
outcome study. Lancet 2002; 360: 1914–20.
45 Ghaye B, Szapiro D, Mastora I, et al. Peripheral pulmonary arteries:
how far in the lung does multi-detector row spiral CT allow
analysis? Radiology 2001; 219: 629–36.
46 Perrier A, Roy PM, Sanchez O, et al. Multidetector-row computed
tomography in suspected pulmonary embolism. N Engl J Med 2005;
47 Righini M, Le Gal G, Aujesky D, et al. Diagnosis of pulmonary
embolism by multidetector CT alone or combined with venous
ultrasonography of the leg: a randomised non-inferiority trial.
Lancet 2008; 371: 1343–52.
48 Moores LK, Jackson WL Jr, Shorr AF, Jackson JL. Meta-analysis:
outcomes in patients with suspected pulmonary embolism
managed with computed tomographic pulmonary angiography.
Ann Intern Med 2004; 141: 866–74.
49 van Beek EJ, Brouwerst EM, Song B, Stein PD, Oudkerk M.
Clinical validity of a normal pulmonary angiogram in patients with
suspected pulmonary embolism—a critical review. Clin Radiol
2001; 56: 838–42.
50 Stein PD, Sostman HD, Bounameaux H, et al. Challenges in the
diagnosis of acute pulmonary embolism. Am J Med 2008;
51 Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed
tomography for acute pulmonary embolism. N Engl J Med 2006;
52 Stein PD, Chenevert TL, Fowler SE, et al. Gadolinium-enhanced
magnetic resonance angiography for pulmonary embolism:
a multicenter prospective study (PIOPED III). Ann Intern Med 2010;
53 Roy PM, Colombet I, Durieux P, et al. Systematic review and
meta-analysis of strategies for the diagnosis of suspected
pulmonary embolism. BMJ 2005; 331: 259.
54 Torbicki A, Perrier A, Konstantinides S, et al. Guidelines on the
diagnosis and management of acute pulmonary embolism: the Task
Force for the Diagnosis and Management of Acute Pulmonary
Embolism of the European Society of Cardiology (ESC). Eur Heart J
2008; 29: 2276–315.
55 Aujesky D, Obrosky DS, Stone RA, et al. Derivation and validation
of a prognostic model for pulmonary embolism.
Am J Respir Crit Care Med 2005; 172: 1041–46.
56 Jiménez D, Aujesky D, Moores L, et al. Simplifi cation of the
pulmonary embolism severity index for prognostication in patients
with acute symptomatic pulmonary embolism. Arch Intern Med
2010; 170: 1383–89.
57 Kucher N, Rossi E, De Rosa M, Goldhaber SZ. Massive pulmonary
embolism. Circulation 2006; 113: 577–82.
58 Aujesky D, Roy PM, Verschuren F, et al. Outpatient versus inpatient
treatment for patients with acute pulmonary embolism: an
international, open-label, randomised, non-inferiority trial.
Lancet 2011; 378: 41–48.
59 Agnelli G, Becattini C. Acute pulmonary embolism. N Engl J Med
2010; 363: 266–74.
60 Schulman S, Kearon C, Kakkar AK, et al. Dabigatran versus
warfarin in the treatment of acute venous thromboembolism.
N Engl J Med 2009; 361: 2342–52.
61 Bauersachs R, Berkowitz SD, Brenner B, et al. Oral rivaroxaban for
symptomatic venous thromboembolism. N Engl J Med 2010;
62 Bounameaux H, de Moerloose P. Is laboratory monitoring of
low-molecular-weight heparin therapy necessary? No.
J Thromb Haemost 2004; 2: 551–54.
63 Gould MK, Dembitzer AD, Doyle RL, Hastie TJ, Garber AM.
Low-molecular-weight heparins compared with unfractionated
heparin for treatment of acute deep venous thrombosis. A
meta-analysis of randomized, controlled trials. Ann Intern Med
1999; 130: 800–09.
64 Simonneau G, Sors H, Charbonnier B, et al. A comparison of
low-molecular-weight heparin with unfractionated heparin for acute
pulmonary embolism. The THÉSÉE Study Group. Tinzaparine ou
Heparine Standard: Evaluations dans l’Embolie Pulmonaire.
N Engl J Med 1997; 337: 663–69.
65 Büller HR, Davidson BL, Decousus H, et al. Fondaparinux or
enoxaparin for the initial treatment of symptomatic deep venous
thrombosis: a randomized trial. Ann Intern Med 2004; 140: 867–73.
66 Büller HR, Davidson BL, Decousus H, et al. Subcutaneous
fondaparinux versus intravenous unfractionated heparin in the
initial treatment of pulmonary embolism. N Engl J Med 2003;
67 Halkin H, Lubetsky A. Warfarin dose requirement and CYP2C9
polymorphisms. Lancet 1999; 353: 1972–73.
68 Schwarz UI, Ritchie MD, Bradford Y, et al. Genetic determinants of
response to warfarin during initial anticoagulation. N Engl J Med
2008; 358: 999–1008.
69 Gage BF, Yan Y, Milligan PE, et al. Clinical classifi cation schemes
for predicting hemorrhage: results from the National Registry of
Atrial Fibrillation (NRAF). Am Heart J 2006; 151: 713–19.
70 Ruíz-Giménez N, Suárez C, González R, et al. Predictive variables
for major bleeding events in patients presenting with documented
acute venous thromboembolism. Findings from the RIETE
Registry. Thromb Haemost 2008; 100: 26–31.
71 Garcia-Alamino JM, Ward AM, Alonso-Coello P, et al.
Self-monitoring and self-management of oral anticoagulation.
Cochrane Database Syst Rev 2010: 4: CD003839.
72 Bloomfi eld HE, Krause A, Greer N, et al. Meta-analysis: eff ect of
patient self-testing and self-management of long-term
anticoagulation on major clinical outcomes. Ann Intern Med 2011;
73 Matchar DB, Jacobson A, Dolor R, et al. Eff ect of home testing of
international normalized ratio on clinical events. N Engl J Med 2010;
Warkentin TE, Greinacher A, Koster A, Lincoff AM. Treatment and
prevention of heparin-induced thrombocytopenia: American
College of Chest Physicians Evidence-Based Clinical Practice
Guidelines (8th edition). Chest 2008; 133: 340S–80S.
75 Lobo JL, Jiménez D, Orue MT, et al. Recurrent venous
thromboembolism during coumarin therapy. Data from the
computerised registry of patients with venous thromboembolism.
Br J Haematol 2007; 138: 400–03.
76 Le Gal G, Kovacs MJ, Carrier M, et al. Validation of a diagnostic
approach to exclude recurrent venous thromboembolism.
J Thromb Haemost 2009; 7: 752–59.
77 Palareti G, Cosmi B, Legnani C, et al. D-dimer testing to determine
the duration of anticoagulation therapy. N Engl J Med 2006;
78 Prandoni P, Lensing AW, Prins MH, et al. Residual venous
thrombosis as a predictive factor of recurrent venous
thromboembolism. Ann Intern Med 2002; 137: 955–60.
79 Bounameaux H, Righini M. Thrombosis: duration of
anticoagulation after VTE: guided by ultrasound? Nat Rev Cardiol
2009; 6: 499–500.
80 Mavrakanas T, Bounameaux H. The potential role of new oral
anticoagulants in the prevention and treatment of
thromboembolism. Pharmacol Ther 2011; 130: 46–58.
81 Bounameaux H, Reber G. New oral antithrombotics: a need for
laboratory monitoring. Against. J Thromb Haemost 2010;
82 Stein PD, Matta F, Hull RD. Increasing use of vena cava fi lters for
prevention of pulmonary embolism. Am J Med 2011; 124: 655–61.
83 Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous
thromboembolism: American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines (8th edition). Chest
2008; 133: 381S–453S.
84 Sweetland S, Green J, Liu B, et al. Duration and magnitude of the
postoperative risk of venous thromboembolism in middle aged
women: prospective cohort study. BMJ 2009; 339: b4583.
Seminar Download full-text
www.thelancet.com Published online April 10, 2012 DOI:10.1016/S0140-6736(11)61904-1
85 American College of Cardiology. Rivaroxaban compares favorably
with enoxaparin in preventing venous thromboembolism in acutely
ill patients without showing a net clinical benefi t. April 5, 2011.
News-Releases/2011/04/MAGELLAN.aspx (accessed Jan 13, 2012).
86 Spencer FA, Lessard D, Emery C, Reed G, Goldberg RJ. Venous
thromboembolism in the outpatient setting. Arch Intern Med 2007;
87 Hull RD, Schellong SM, Tapson VF, et al. Extended-duration
venous thromboembolism prophylaxis in acutely ill medical
patients with recently reduced mobility: a randomized trial.
Ann Intern Med 2010; 153: 8–18.
88 Kent DM, Lindenauer PK. Aggregating and disaggregating patients
in clinical trials and their subgroup analyses. Ann Intern Med 2010;
89 Spyropoulos AC, Anderson FA Jr, Fitzgerald G, et al. Predictive and
associative models to identify hospitalized medical patients at risk
for venous thromboembolism. Chest 2011; 140: 706–14.
90 Amin A, Stemkowski S, Lin J, Yang G. Thromboprophylaxis rates
in US medical centers: success or failure? J Thromb Haemost 2007;
91 Goldhaber SZ, Tapson VF. A prospective registry of 5,451 patients
with ultrasound-confi rmed deep vein thrombosis. Am J Cardiol
2004; 93: 259–62.
92 Piazza G, Seddighzadeh A, Goldhaber SZ. Double trouble for
2609 hospitalized medical patients who developed deep vein
thrombosis: prophylaxis omitted more often and pulmonary
embolism more frequent. Chest 2007; 132: 554–61.
93 Cohen AT, Tapson VF, Bergmann JF, et al. Venous
thromboembolism risk and prophylaxis in the acute hospital care
setting (ENDORSE study): a multinational cross-sectional study.
Lancet 2008; 371: 387–94.
94 Anderson FA Jr, Goldhaber SZ, Tapson VF, et al. Improving
practices in US hospitals to prevent venous thromboembolism:
lessons from ENDORSE. Am J Med 2010; 123: 1099–106.
95 Kucher N, Spirk D, Baumgartner I, et al. Lack of prophylaxis before
the onset of acute venous thromboembolism among hospitalized
cancer patients: the SWIss Venous ThromboEmbolism Registry
(SWIVTER). Ann Oncol 2010; 21: 931–35.
96 Fanikos J, Stevens LA, Labreche M, et al. Adherence to
pharmacological thromboprophylaxis orders in hospitalized
patients. Am J Med 2010; 123: 536–41.
97 Piazza G, Goldhaber SZ. Computerized decision support for the
cardiovascular clinician: applications for venous thromboembolism
prevention and beyond. Circulation 2009; 120: 1133–37.
98 Kucher N, Koo S, Quiroz R, et al. Electronic alerts to prevent venous
thromboembolism among hospitalized patients. N Engl J Med 2005;
99 Fiumara K, Piovella C, Hurwitz S, et al. Multi-screen electronic
alerts to augment venous thromboembolism prophylaxis.
Thromb Haemost 2010; 103: 312–17.
100 Lecumberri R, Marqués M, Diaz-Navarlaz MT, et al. Maintained
eff ectiveness of an electronic alert system to prevent venous
thromboembolism among hospitalized patients. Thromb Haemost
2008; 100: 699–704.
101 Piazza G, Rosenbaum EJ, Pendergast W, et al. Physician alerts to
prevent symptomatic venous thromboembolism in hospitalized
patients. Circulation 2009; 119: 2196–201.
102 Goldhaber SZ. Eradication of hospital-acquired venous
thromboembolism. Thromb Haemost 2010; 104: 1089–92.