You are on page 1of 9

Archives of Cardiovascular Disease (2014) 107, 406—414

Available online at

ScienceDirect
www.sciencedirect.com

REVIEW

Effective diagnosis and treatment of


pulmonary embolism: Improving patient
outcomes
Des changements dans le diagnostic et la thérapeutique pour
améliorer le pronostic de l’embolie pulmonaire

Guy Meyer a,b,∗

a
Division of respiratory and intensive care, hôpital européen Georges-Pompidou, 20, rue
Leblanc, 75015 Paris, France
b
Université Paris Descartes, Sorbonne Paris Cité, 75006 Paris, France

Received 4 April 2014; received in revised form 2 May 2014; accepted 5 May 2014
Available online 9 July 2014

KEYWORDS Summary Pulmonary embolism can be life threatening and difficult to diagnose as signs
Anticoagulants; and symptoms are not specific. European guidelines recommend stratification of pulmonary
Diagnostic embolism by risk of early mortality. Patients with suspected pulmonary embolism should be
techniques; assessed for clinical probability of pulmonary embolism using a validated risk score. A low
Pulmonary embolism or intermediate clinical probability plus a negative high-sensitivity D-dimer test excludes
pulmonary embolism. Anticoagulation is indicated in patients with a positive multidetector
computed tomography or high-probability lung scan. An important part of the management of
patients with pulmonary embolism has traditionally been anticoagulant treatment with par-
enteral heparins and oral vitamin K antagonists. Although effective, this dual-drug approach is
associated with limitations. Direct oral anticoagulants that may overcome some of these prob-
lems have been tested in phase III clinical trials for the treatment of venous thromboembolism.
Of these, rivaroxaban and apixaban have demonstrated non-inferiority to standard therapy
when given as single-drug approaches for venous thromboembolism treatment, and provided

Abbreviations: CI, Confidence interval; CrCl, Creatinine clearance; CT, Computed tomography; DVT, Deep vein thrombosis; GARFIELD,
Global Anticoagulant Registry in the FIELD; HR, Hazard ratio; LMWH, Low-molecular-weight heparin; OAC, Oral anticoagulant; PE, Pul-
monary embolism; RIETE, Registro Informatizado de Pacientes con Enfermedad TromboEmbólica; VKA, Vitamin K antagonist; VTE, Venous
thromboembolism.
∗ Correspondence to: Division of respiratory and intensive care, hôpital européen Georges-Pompidou, 20, rue Leblanc, 75015 Paris, France.

E-mail address: guy.meyer@egp.aphp.fr

http://dx.doi.org/10.1016/j.acvd.2014.05.006
1875-2136/© 2014 Elsevier Masson SAS. All rights reserved.
Improving patient outcomes in pulmonary embolism 407

significant reductions in major bleeding rates. Dabigatran and edoxaban were non-inferior to
standard therapy when given as part of a dual-drug approach after initial parenteral antico-
agulation, and reduced clinically relevant bleeding rates. There may be a benefit to extended
anticoagulation with direct oral anticoagulants for the prevention of recurrent venous throm-
boembolism. Registry studies will provide more information on the use of these agents in
real-world populations. Accurate diagnosis and risk stratification of patients with pulmonary
embolism, together with simplified anticoagulation therapy, is likely to improve outcomes.
© 2014 Elsevier Masson SAS. All rights reserved.

MOTS CLÉS Résumé L’embolie pulmonaire est une pathologie potentiellement létale et est difficile à
Traitement diagnostiquer. Les signes et les symptômes de l’embolie pulmonaire ne sont pas spécifiques.
anticoagulant ; Les recommandations européennes recommandent de stratifier les malades selon leur risque
Diagnostic ; de mortalité. La probabilité clinique de l’embolie pulmonaire doit être évaluée à l’aide d’un
Embolie pulmonaire score devant toute suspicion clinique d’embolie pulmonaire. Une probabilité clinique faible ou
intermédiaire associée à un taux de D-dimère normal élimine le diagnostic. Le diagnostic est
affirmé par l’angioscanner thoracique ou la scintigraphie pulmonaire. Le traitement initial asso-
cie traditionnellement une héparinothérapie parentérale et un antagoniste de la vitamine K.
Ce traitement est efficace mais comporte certains inconvénients. Des inhibiteurs oraux directs
de la coagulation, qui pourraient limiter ces inconvénients, ont été testés dans de grands essais
de phase III. Parmi eux, le rivaroxaban et l’apixaban administrés dès l’inclusion ont démontré
leur non-infériorité par rapport au traitement standard en termes d’efficacité tout en obtenant
une réduction des hémorragies majeures. Le dabigatran et l’edoxaban sont également non
inférieurs au traitement standard après un traitement parentéral initial et sont également asso-
ciés à une réduction des saignements cliniquement significatifs. La prolongation du traitement
au-delà de six mois en cas de thrombose non provoquée pourrait être associée à une réduc-
tion des récidives. Des études de registres devraient apporter des informations sur l’utilisation
de ces nouvelles molécules en situation de soin courant. Un diagnostic rigoureux, associé à
une stratification du risque et à un traitement anticoagulant simplifié devraient permettre une
amélioration du pronostic de l’embolie pulmonaire.
© 2014 Elsevier Masson SAS. Tous droits réservés.

Introduction is associated with some limitations, including the need to


co-administer the parenteral agent and VKA concurrently
Pulmonary embolism (PE) is a relatively common disease, for several days at the start of treatment, and the sub-
with an incidence ranging from 60 to 112 per 100,000 inhabi- sequent need for regular coagulation monitoring and dose
tants of the United States [1], and is the third most common adjustments during VKA monotherapy. The recently devel-
cause of death among patients with cardiovascular diseases oped direct oral anticoagulants circumvent some of these
[2]. Patients are at particular risk in the acute stage of the limitations, and several have completed large phase III clin-
disease, with 30-day mortality rates in excess of 15% for PE ical trials in the treatment of acute VTE. In Europe, only
associated with shock and/or hypotension [3]. PE is diffi- rivaroxaban is currently approved for the treatment and sec-
cult to diagnose because of the wide range of presentations ondary prevention of deep vein thrombosis (DVT) and PE. In
of the disease. Among those patients who die of PE, 94% the USA, rivaroxaban and, more recently, dabigatran have
do so before diagnosis [4]. The mainstay of treatment for been approved for VTE treatment.
most patients with PE is anticoagulation, and the risk of This review covers the diagnosis and treatment of PE,
death is much reduced in optimally anticoagulated patients. focusing on data for direct oral anticoagulants.
The recurrence rate of venous thromboembolism (VTE) after
stopping anticoagulant treatment varies with the cause of
the disease and is much higher in patients with unprovoked Clinical presentations of PE
VTE than in patients with VTE provoked by a major tran-
sient risk factor [5]. Duration of anticoagulant treatment is Most patients with suspected PE present with some degree
tailored to the risks of VTE recurrence and bleeding for each of chest pain and dyspnoea, a frequent cause of refer-
individual patient [3,6]. ral to the emergency department. These symptoms are
Treatment of PE has predominantly involved the use non-specific and can be confused with other differential
of low-molecular-weight heparins (LMWHs), unfractionated diagnoses, such as acute coronary syndromes, exacerbation
heparin or fondaparinux in combination with vitamin K of chronic obstructive pulmonary disease or pneumonia [7].
antagonists (VKAs) [3,6]. However, this dual-drug approach The most specific symptoms of PE—haemoptysis and calf
408 G. Meyer

pain—are encountered in only up to 10% and 42% of patients or embolectomy, particularly when other tests cannot be
with PE, respectively [8,9]. In the most severe cases of performed (Fig. 1) [3].
PE, patients may present with shock and/or haemodynamic
instability [3]. The lack of specific symptoms and the pres-
ence of underlying disease in a significant proportion of Confirmatory diagnosis in suspected
patients with PE probably explain the significant diagnostic non-high-risk PE
delay observed in some cases.
If a diagnosis of non-high-risk PE is suspected, the first step
is to assess the clinical probability of PE using a standard-
ized clinical prediction rule (Fig. 1) [3]. According to the
Risk stratification for suspected PE result, patients can be divided into three (low, interme-
diate or high probability of PE) or two (‘PE unlikely’ and
Risk stratification is a relatively new concept in the field ‘PE likely’) groups, each with different probabilities of PE.
of PE. The size of the emboli and degree of vessel occlu- A negative result from a high-sensitivity D-dimer test in
sion do not accurately describe the risk of death of a patients with either low—intermediate probability or ‘PE
patient with suspected PE, and the terms ‘massive’ and unlikely’ safely excluded PE in about 30% of outpatients pre-
‘sub-massive’ are misleading [3]. Therefore, other methods senting to seven Dutch hospitals with suspected PE [15]. In
are required to assess mortality risk and inform manage- patients with either a high clinical probability or an elevated
ment decisions. Accordingly, the guidelines of the European D-dimer plasma concentration, three options for confirma-
Society of Cardiology categorize patients presenting with tory diagnosis are available: CT pulmonary angiography,
suspected PE by their predicted risk of early mortality. ventilation—perfusion lung scanning or venous compression
Patients are divided into high-risk or non-high-risk (which is ultrasound [3].
further divided into intermediate- and low-risk) categories Nowadays, multidetector spiral CT pulmonary angiog-
based on the presence of shock, myocardial injury and right raphy is considered the standard option for confirming
ventricular dysfunction detected by echocardiography and non-high-risk PE (Fig. 1). Using multidetector spiral CT pul-
cardiac biomarkers [3]. The risk category to which a patient monary angiography, the rate of recurrent VTE within 3
is assigned informs the approach, whether it be emergency months after a negative examination has been reported to
thrombolysis or embolectomy (high-risk: > 15% early mortal- be as low as 0.3% with or without additional ultrasound,
ity risk) or further confirmatory diagnostic steps and, if PE confirming a high sensitivity [16]. Compared with single-
is confirmed, management with anticoagulants (non-high- detector CT pulmonary angiography and the first generation
risk: < 1—15% early mortality risk) [3]. Risk stratification also of multidetector devices, the recent 64-detector systems
allows for the selection of patients who may be suitable for detect a higher rate of isolated subsegmental PEs, the clin-
outpatient anticoagulant treatment [10]. ical significance of which has been questioned [17].
The Pulmonary Embolism Severity Index is a clinical tool Ventilation—perfusion lung scanning has been used to
designed to assess the risk of death in patients with PE. diagnose PE for many years. A normal ventilation—perfusion
According to this, patients with PE can be divided into five lung scan virtually eliminates the diagnosis of PE, but about
groups with different outcomes. The 30-day risk of death in 50% of examinations are non-diagnostic and do not by them-
patients belonging to the low-risk categories I and II is usu- selves allow exclusion or confirmation of the diagnosis of PE
ally < 3%, whereas that in patients belonging to the high-risk [18]. On the other hand, a high-probability scan is strongly
category V varies between 10% and 25% [11,12]. A simpli- correlated with a diagnosis of PE, but further tests may be
fied version of the rule, based on six variables that all carry considered in patients with a low clinical probability risk
the same weight (one of which is a composite of two origi- score [3].
nal variables), has been shown to have the same sensitivity Compression ultrasound allows diagnosis of DVT in
and specificity as the original [13]. Right ventricular dilata- patients with clinically suspected PE, and the finding of a
tion on echocardiography or computed tomography (CT) proximal DVT in a patient with clinically suspected PE is
pulmonary angiography, and right ventricular dysfunction or accepted as a surrogate for the diagnosis of PE and does not
injury detected by brain natriuretic peptide and troponin need to be confirmed by an examination of the chest [3].
testing, allow further risk stratification of clinically stable However, a normal proximal compression ultrasound does
patients with PE [14]. not exclude the diagnosis of PE [19], and only 10% of exami-
nations are positive in patients with suspected PE [20]. Thus,
compression ultrasound cannot be considered an efficient
diagnostic method for most patients with suspected PE.
Confirmatory diagnosis in suspected
high-risk PE
Current treatment options for PE
In patients with suspected high-risk PE, CT pulmonary
angiography is the preferred technique to confirm the diag- In the relatively few patients with high-risk PE (presence
nosis (Fig. 1) [3]. Echocardiography is an alternative if CT of shock and/or hypotension), thrombolytic therapy is rec-
pulmonary angiography is unavailable or the patient is too ommended. In cases where thrombolytic therapy fails or
unstable. In haemodynamically unstable patients, acute pul- is contraindicated, catheter-assisted thrombus removal or
monary hypertension and right ventricular overload seen on surgical embolectomy provides a back-up option [3,6]. Such
echocardiography may justify a decision for thrombolysis patients should also receive immediate anticoagulation with
Improving patient outcomes in pulmonary embolism 409

Figure 1. Recommended diagnostic pathway for patients with suspected PE—European Society of Cardiology guidelines 2008 [3]. Notes:
CT is defined as not immediately available if the patient’s critical condition allows only bedside diagnostic tests. Confirmation of DVT via
compression venous ultrasound may also assist with clinical decisions. CT is considered diagnostic of PE if the most proximal thrombus is
at least segmental. If single-detector CT is negative, a negative proximal lower limb venous ultrasound is required to safely exclude PE.
If MDCT is negative in patients with a high clinical probability of PE, further investigations may be considered. CT: computed tomography;
CTPA: computed tomography pulmonary angiography; DVT: deep vein thrombosis; echoCG: echocardiography; MDCT: multidetector spiral
computed tomography; PE: pulmonary embolism; RV: right ventricular.

unfractionated heparin with haemodynamic and respira- until the international normalized ratio reaches between
tory support as required. Anticoagulation is the mainstay 2.0 and 3.0 [3,6]. Owing to their slow onset of action,
treatment option for haemodynamically stable patients with VKAs must initially be used in conjunction with the faster-
non-high-risk PE and should be started as soon as the diag- acting parenteral anticoagulants. During treatment with a
nosis is suspected, at least in patients with a high clinical VKA, patients require frequent coagulation monitoring to
probability of PE and no contraindication to anticoagulant maintain treatment within a therapeutic range (interna-
therapy [3,6]. tional normalized ratio 2.0—3.0). The initial dosage varies
greatly between patients and, because of drug and food
Traditional anticoagulants for the treatment interactions, the subsequent dosage may vary according to
of PE changes in associated treatment and diet. Even under con-
trolled conditions, the time in the therapeutic range usually
The traditional parenteral anticoagulants employed in does not exceed 65% [21—24]. VKAs represent one of the
the initial treatment of non-high-risk PE include LMWH, most frequent causes of referral to emergency departments
intravenous or subcutaneous unfractionated heparin or fon- because of adverse drug reactions, especially in the elderly
daparinux. Except for patients with VTE and cancer, LMWH [25]. These limitations underscore the need for simpler
is normally given for the first 5—7 days of treatment and is anticoagulant drugs for the initial and long-term treatment
overlapped with and followed by treatment with an oral VKA of PE.
410 G. Meyer

Direct oral thrombin and Factor Xa inhibitors to standard therapy for efficacy and led to a 46% relative
for the treatment of PE risk reduction in major bleeding (Fig. 3) [33]. Rivaroxa-
ban was associated with a significant improvement in net
A series of novel anticoagulant drugs have recently been clinical benefit (defined as the composite incidence of recur-
developed. These drugs are direct inhibitors of Factor IIa rent VTE and major bleeding) in fragile patients (age > 75
(thrombin) or Factor Xa, are not subject to food interac- years, CrCl < 50 mL/min or weight ≤ 50 kg; HR 0.51, 95% CI
tions and have minimal drug interactions compared with 0.34—0.77), primarily as a result of a 63% relative risk reduc-
VKAs. They can be administered orally at a fixed dosage tion in major bleeding in this group compared with standard
without the need for routine coagulation monitoring [26]. therapy [33].
Among these drugs, rivaroxaban, dabigatran, apixaban and AMPLIFY (Apixaban for the Initial Management of Pul-
edoxaban have been tested for the initial and/or secondary monary Embolism and Deep-Vein Thrombosis as First-line
prevention of VTE in large phase III studies [21—24,27—30]. Therapy) was a randomized, double-blind study comparing
In 2012, rivaroxaban became the first direct oral antico- apixaban (10 mg twice daily for 7 days followed by 5 mg
agulant to be approved for the treatment of DVT and PE twice daily for 6 months) with conventional therapy (subcu-
and the prevention of recurrent VTE [31,32]. In contrast to taneous enoxaparin followed by warfarin) in 5395 patients
the combination of LMWH and VKA, rivaroxaban provides a with acute VTE [24]. Reasons for ineligibility included con-
fixed-dose, single-drug approach to the initial and contin- traindications to standard anticoagulant therapy, cancer,
ued treatment of PE in patients who are haemodynamically provoked thrombosis without a persistent risk factor for
stable and do not require thrombolysis or embolectomy. recurrence, if < 6 months of treatment was planned, treat-
ment with > 165 mg aspirin daily or CrCl < 25 mL/min. No
more than two doses of once-daily LMWH, fondaparinux or
VKA; three doses of twice-daily LMWH; or > 36 hours of con-
Phase III studies of the direct oral tinuous intravenous heparin were permitted. Apixaban was
anticoagulants in the treatment of acute non-inferior to standard therapy for the primary efficacy
VTE outcome of recurrent symptomatic VTE or VTE-related death
(P < 0.001 for non-inferiority), with a 69% relative risk reduc-
The EINSTEIN PE study was an open-label, randomized, tion in major bleeding (Fig. 3) [24]. Outcomes in subgroups,
non-inferiority study that compared oral rivaroxaban with including elderly patients and those weighing > 100 kg were
standard therapy in patients with acute, symptomatic PE consistent with the overall population.
with or without DVT [23]. Among other criteria, patients RE-COVER was a double-blind, randomized trial that com-
were excluded if they had haemodynamic instability, cre- pared 6 months of treatment with dabigatran (fixed dose
atinine clearance (CrCl) < 30 mL/min, clinically significant of 150 mg twice daily) with dose-adjusted warfarin ther-
liver disease, had received therapeutic parenteral anti- apy in patients with proximal DVT or PE [21]. Patients were
coagulation for > 48 hours, or more than a single dose of excluded for reasons of haemodynamic instability, symptoms
a VKA before randomization. Rivaroxaban was given at a lasting > 14 days, recent unstable cardiovascular disease,
dose of 15 mg twice daily for the first 3 weeks followed high bleeding risk, clinically significant liver disease,
by 20 mg once daily. Standard therapy was subcutaneous CrCl < 30 mL/min, or a requirement for aspirin > 100 mg/day.
enoxaparin 1.0 mg/kg twice daily followed by either war- All patients were initially given an approved parenteral
farin or acenocoumarol for 3, 6 or 12 months. A total anticoagulant (generally unfractionated heparin or LMWH).
of 2419 patients were assigned to rivaroxaban and 2413 Dabigatran was non-inferior to standard therapy for the pre-
to standard therapy. Recurrent VTE occurred in 2.1% of vention of recurrent VTE (P < 0.001 for non-inferiority), with
patients who were given rivaroxaban compared with 1.8% a significant reduction in clinically relevant bleeding but
who received standard therapy (P = 0.003 for non-inferiority; not major bleeding (Fig. 3) [21]. Outcomes were consistent
Fig. 2). Major or non-major clinically relevant bleeding regardless of variations in demographic factors. The RE-
occurred in 10.3% and 11.4% of patients, respectively (haz- COVER II study had essentially the same design as RE-COVER
ard ratio [HR] 0.90, 95% confidence interval [CI] 0.76—1.07; and the results, recently published, support the outcomes
P = 0.23; Fig. 2). Major bleeding occurred in 1.1% and of RE-COVER [27].
2.2% of patients, respectively (HR 0.49, 95% CI 0.31—0.79; Edoxaban was compared with warfarin in the Hokusai-
P = 0.003), representing a 51% relative risk reduction (Fig. 2) VTE study, a randomized, double-blind, non-inferiority study
[23]. Subgroup analysis demonstrated that rates of recur- of 8292 patients with VTE [30]. Patients were excluded if
rent VTE and bleeding were similar in both treatment groups they had contraindications to standard anticoagulant treat-
regardless of age, weight, sex, renal function or extent of ment, had received > 48 hours of heparin treatment or more
PE. than one dose of VKA, had cancer with anticipated long-
The EINSTEIN DVT study enrolled patients with acute, term LMWH treatment, were receiving aspirin > 100 mg/day
symptomatic proximal DVT without symptomatic PE [22]. or dual antiplatelet therapy, or had CrCl < 30 mL/min. As
General exclusion criteria were similar to those applied with dabigatran, treatment with edoxaban (60 mg once daily
in EINSTEIN PE. As in EINSTEIN PE, rivaroxaban was non- or 30 mg once daily in patients with CrCl 30—50 mL/min
inferior to standard therapy (P < 0.001 for non-inferiority) or weight ≤ 60 kg or who were receiving treatment with
[22]. Major or non-major clinically relevant bleeding and potent P-glycoprotein inhibitors) was initiated after initial
major bleeding alone occurred with similar incidences in parenteral anticoagulation [30]. Edoxaban was non-inferior
both treatment groups [22]. In a pooled analysis of EIN- to standard therapy for the prevention of symptomatic
STEIN DVT and EINSTEIN PE, rivaroxaban was non-inferior recurrent VTE (P < 0.001 for non-inferiority) [30]. Clinically
Improving patient outcomes in pulmonary embolism 411

Figure 2. Primary efficacy, safety and net clinical benefit outcomes in patients with PE with or without DVT in the phase III EINSTEIN
PE study of single-drug rivaroxaban compared with standard therapy (enoxaparin/VKA). Efficacy endpoints (recurrent VTE and net clinical
benefit) were evaluated in the intention-to-treat population. Bleeding outcomes were evaluated in the safety population. Net clinical benefit
was defined as the composite incidence of recurrent VTE and major bleeding. Clinically relevant bleeding was defined as the composite
of major and non-major clinically relevant bleeding. CI: confidence interval; DVT: deep vein thrombosis; PE: pulmonary embolism; VKA:
vitamin K antagonist; VTE: venous thromboembolism.

Figure 3. Published, phase III clinical studies of direct oral anticoagulants for the treatment of acute venous thromboembolism
[21,24,30,33]. a 12 months after the start of treatment. CI: confidence interval; OAC: oral anticoagulant; VTE: venous thromboembolism.
412 G. Meyer

relevant bleeding was significantly reduced with edoxaban, increase in major plus non-major clinically relevant bleed-
but major bleeding was not (Fig. 3) [30]. Efficacy with ing with rivaroxaban and dabigatran, although not apixaban,
edoxaban was also confirmed in a subset of patients with but without a significant increase in major bleeding for any
severe PE (N-terminal prohormone of brain natriuretic pep- agent. The net increase in major bleeding did not exceed
tide level ≥ 500 pg/mL) [30]. 1%, whereas the net decrease in recurrent VTE (plus or
minus VTE-related or all-cause mortality, depending on the
study) varied from 5.2% to 7.8% [22,28,29]. Only dabiga-
tran has been compared with warfarin for the secondary
Secondary prevention of VTE with direct prevention of VTE [28]. The two treatments were equally
oral anticoagulants efficacious for the prevention of recurrent VTE (P = 0.01
for non-inferiority) and the incidence of major bleeding
Patients with unprovoked PE (i.e. occurring in the absence was similar (HR 0.52, 95% CI 0.27—1.02). Dabigatran was
of a major risk factor such as trauma, surgery, immobiliza- associated with an increase in the rate of acute coronary
tion or cancer) have a high risk of recurrent VTE after the syndromes (0.9% vs 0.2%; P = 0.02) [28].
end of anticoagulant treatment [5]. Approximately 5—10%
of these patients will have a recurrent event during the first
year after a 3- or 6-month course of anticoagulant ther- Discussion
apy, and about 30% of them will have a recurrence within
5 years [34]. Dabigatran, rivaroxaban and apixaban have Previous results in patients with VTE explain why, in phase
been investigated for the prevention of these recurrent III trials, dabigatran and edoxaban were started only after a
events. Dabigatran and rivaroxaban were administered using standard course of parenteral treatment with LMWH, and
a single-dose regimen (150 mg twice daily for dabigatran and why rivaroxaban and apixaban were started immediately
20 mg once daily for rivaroxaban), whereas apixaban was after randomization, but at a higher dosage during the acute
given at two different doses of 2.5 mg and 5 mg twice daily treatment phase. These treatment schedules were selected
[22,28,29]. The three drugs achieved a relative risk reduc- to optimize the efficacy of the direct oral anticoagulant regi-
tion of recurrent VTE compared with placebo of 64—92% mens during the initial period of treatment where the risk
(Fig. 4). This was obtained at the cost of a significant of recurrent VTE is high. Previous experience with ximela-

Figure 4. Published phase III clinical studies of direct oral anticoagulants for the prolonged treatment of acute VTE [22,28,29]. a P = 0.11.
b P = 1.0. bid: twice daily; CI: confidence interval; OAC: oral anticoagulant; od: once daily; VTE: venous thromboembolism.
Improving patient outcomes in pulmonary embolism 413

gatran and idraparinux, which were the first drugs designed Acknowledgements
to challenge the usual combination of heparin or fonda-
parinux and warfarin, suggested a higher risk of recurrent The author would like to acknowledge Stephen Purver, who
VTE with the new compounds during the first month of provided editorial support with funding from Bayer Health-
therapy [35,36]. These results, and those of other stud- Care Pharmaceuticals and Janssen Scientific Affairs, LLC.
ies, highlight the need for a highly effective anticoagulation
regimen early in the course of VTE.
The results of clinical studies of direct oral anticoagu- References
lants were obtained in selected patients who were relatively
young and had few associated diseases, but at least for [1] Wiener RS, Schwartz LM, Woloshin S. Time trends in pulmonary
rivaroxaban [22,23] and edoxaban [30], the results also embolism in the United States: evidence of overdiagnosis. Arch
apply for important subgroups of PE patients. Additionally, Intern Med 2011;171:831—7.
the VKA time in therapeutic range was potentially higher in [2] Goldhaber SZ, Bounameaux H. Pulmonary embolism and deep
these studies than in real life, and so VKAs can be consid- vein thrombosis. Lancet 2012;379:1835—46.
ered to have been given optimally in these trials. Therefore, [3] Torbicki A, Perrier A, Konstantinides S, et al. Guidelines on
the observed benefits of the direct oral anticoagulants may the diagnosis and management of acute pulmonary embolism:
the task force for the diagnosis and management of acute pul-
be more pronounced in routine clinical practice than in the
monary embolism of the European Society of Cardiology (ESC).
study settings. Several registries and real-world studies are Eur Heart J 2008;29:2276—315.
underway or planned to investigate the epidemiology and [4] Dalen JE. Pulmonary embolism: what have we learned since
treatment outcomes of VTE encountered in daily clinical Virchow? Natural history, pathophysiology, and diagnosis. Chest
practice, including in groups of patients who would nor- 2002;122:1440—56.
mally be excluded from clinical trials. Established registries [5] Baglin T, Luddington R, Brown K, Baglin C. Incidence of
include the Registro Informatizado de Pacientes con Enfer- recurrent venous thromboembolism in relation to clinical and
medad TromboEmbólica (RIETE), which has recruited more thrombophilic risk factors: prospective cohort study. Lancet
than 39,000 patients since its inception in 2001 and has, 2003;362:523—6.
among other important data, provided predictive variables [6] Kearon C, Akl EA, Comerota AJ, et al. Antithrombotic
therapy for VTE disease: antithrombotic therapy and pre-
for major bleeding events in patients who present with acute
vention of thrombosis, 9th ed. American College of Chest
VTE [37]. Physicians evidence-based clinical practice guidelines. Chest
More recently, two registries have been created to look 2012;141:e419S—94S.
specifically at patterns of treatment and outcomes in acute [7] Laack TA, Goyal DG. Pulmonary embolism: an unsuspected
and long-term management of VTE with direct oral anticoag- killer. Emerg Med Clin North Am 2004;22:961—83.
ulants. The Dresden Registry aims to recruit 2000 unselected [8] Stein PD, Beemath A, Matta F, et al. Clinical characteristics of
adult patients in Germany (http://www.noac-register.de/), patients with acute pulmonary embolism: data from PIOPED II.
and the Global Anticoagulant Registry in the FIELD Am J Med 2007;120:871—9.
(GARFIELD) VTE registry will include patients in more than [9] Robert-Ebadi H, Le Gal G, Carrier M, et al. Differences in clin-
20 countries (http://www.tri-london.ac.uk/garfield-vte). ical presentation of pulmonary embolism in women and men.
J Thromb Haemost 2010;8:693—8.
[10] Aujesky D, Roy PM, Verschuren F, et al. Outpatient versus inpa-
tient treatment for patients with acute pulmonary embolism:
Conclusions an international, open-label, randomised, non-inferiority trial.
Lancet 2011;378:41—8.
[11] Aujesky D, Obrosky DS, Stone RA, et al. Derivation and vali-
Accurate diagnosis and risk stratification of patients with
dation of a prognostic model for pulmonary embolism. Am J
PE, together with the simplified treatment that the direct
Respir Crit Care Med 2005;172:1041—6.
oral anticoagulants can provide, are likely to improve [12] Aujesky D, Roy PM, Le Manach CP, et al. Validation of a
patient outcomes and reduce mortality associated with this model to predict adverse outcomes in patients with pulmonary
disease. embolism. Eur Heart J 2006;27:476—81.
[13] Jiménez D, Aujesky D, Moores L, et al. Simplification of the pul-
monary embolism severity index for prognostication in patients
with acute symptomatic pulmonary embolism. Arch Intern Med
Disclosure of interest 2010;170:1383—9.
[14] Sanchez O, Trinquart L, Colombet I, et al. Prognostic value of
G.M. clinical trials: as an investigator for Bayer Health- right ventricular dysfunction in patients with haemodynam-
Care Pharmaceuticals, Daiichi Sankyo, Leo Pharma and ically stable pulmonary embolism: a systematic review. Eur
Sanofi-Aventis. One-off interventions: advisory activity Heart J 2008;29:1569—77.
for Bayer HealthCare Pharmaceuticals, Leo Pharma and [15] Douma RA, Mos IC, Erkens PM, et al. Performance of 4 clinical
Pfizer. Conferences: invitations as a contributor for decision rules in the diagnostic management of acute pul-
Bayer HealthCare Pharmaceuticals, Boehringer-Ingelheim, monary embolism: a prospective cohort study. Ann Intern Med
2011;154:709—18.
Leo Pharma, and Sanofi-Aventis. Conferences: invita-
[16] Righini M, Le Gal G, Aujesky D, et al. Diagnosis of pulmonary
tions to attend from Bayer HealthCare Pharmaceuticals, embolism by multidetector CT alone or combined with venous
Boehringer-Ingelheim and Leo Pharma. Research grants or ultrasonography of the leg: a randomised non-inferiority trial.
support to the author’s institution from Bayer HealthCare Lancet 2008;371:1343—52.
Pharmaceuticals, Boehringer-Ingelheim, Leo Pharma and [17] Stein PD, Goodman LR, Hull RD, Dalen JE, Matta F. Diag-
Sanofi-Aventis. nosis and management of isolated subsegmental pulmonary
414 G. Meyer

embolism: review and assessment of the options. Clin Appl [29] Agnelli G, Buller HR, Cohen A, et al. Apixaban for extended
Thromb Hemost 2012;18:20—6. treatment of venous thromboembolism. N Engl J Med
[18] The PIOPED Investigators. Value of the ventilation/perfusion 2013;368:699—708.
scan in acute pulmonary embolism. Results of the Prospective [30] The Hokusai-VTE Investigators. Edoxaban versus warfarin for
Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA the treatment of symptomatic venous thromboembolism. N
1990;263:2753—9. Engl J Med 2013;369:1406—15.
[19] Righini M, Le Gal G, Perrier A, Bounameaux H. The challenge of [31] Bayer Pharma AG. Xarelto® (rivaroxaban) summary of
diagnosing pulmonary embolism in elderly patients: influence product characteristics. Published online on August 2013,
of age on commonly used diagnostic tests and strategies. J Am vol. 15; 2013. Available at: http://www.ema.europa.eu/
Geriatr Soc 2005;53:1039—45. docs/en GB/document library/EPAR - Product Information/
[20] Righini M, Le Gal G, Aujesky D, et al. Complete venous ultra- human/000944/WC500057108.pdf
sound in outpatients with suspected pulmonary embolism. J [32] Janssen Pharmaceuticals Inc. Xarelto® (rivaroxaban) prescrib-
Thromb Haemost 2009;7:406—12. ing information. Published online on January 2014; 2014.
[21] Schulman S, Kearon C, Kakkar AK, et al. Dabigatran versus war- Available at: http://www.accessdata.fda.gov/drugsatfda
farin in the treatment of acute venous thromboembolism. N docs/label/2014/022406s007lbl.pdf
Engl J Med 2009;361:2342—52. [33] Prins MH, Lensing AWA, Bauersachs R, et al. Oral rivaroxa-
[22] The EINSTEIN Investigators. Oral rivaroxaban for symptomatic ban versus standard therapy for the treatment of symptomatic
venous thromboembolism. N Engl J Med 2010;363:2499—510. venous thromboembolism: a pooled analysis of the EINSTEIN-
[23] The EINSTEIN—PE Investigators. Oral rivaroxaban for the treat- DVT and PE randomized studies. Thromb J 2013;11:21.
ment of symptomatic pulmonary embolism. N Engl J Med [34] Prandoni P, Noventa F, Ghirarduzzi A, et al. The risk of
2012;366:1287—97. recurrent venous thromboembolism after discontinuing antico-
[24] Agnelli G, Buller HR, Cohen A, et al. Oral apixaban for the agulation in patients with acute proximal deep vein thrombosis
treatment of acute venous thromboembolism. N Engl J Med or pulmonary embolism. A prospective cohort study in 1,626
2013;369:799—808. patients. Haematologica 2007;92:199—205.
[25] Budnitz DS, Lovegrove MC, Shehab N, Richards CL. Emergency [35] Fiessinger JN, Huisman MV, Davidson BL, et al. Ximelaga-
hospitalizations for adverse drug events in older Americans. N tran vs low-molecular-weight heparin and warfarin for the
Engl J Med 2011;365:2002—12. treatment of deep vein thrombosis: a randomized trial. JAMA
[26] Eikelboom JW, Weitz JI. New anticoagulants. Circulation 2005;293:681—9.
2010;121:1523—32. [36] van Gogh Investigators, Büller HR, Cohen AT, et al. Idraparinux
[27] Schulman S, Kakkar AK, Goldhaber SZ, et al. Treatment of versus standard therapy for venous thromboembolic disease. N
acute venous thromboembolism with dabigatran or warfarin Engl J Med 2007;357:1094—104.
and pooled analysis. Circulation 2014;129:764—72. [37] Nieto JA, Solano R, Trapero Iglesias N, et al. Validation of
[28] Schulman S, Kearon C, Kakkar AK, et al. Extended use of dabi- a score for predicting fatal bleeding in patients receiving
gatran, warfarin, or placebo in venous thromboembolism. N anticoagulation for venous thromboembolism. Thromb Res
Engl J Med 2013;368:709—18. 2013;132:175—9.