Review
Old and new anticoagulants
U. Harbrecht
Institute of Experimental Haematology and Transfusion Medicine and Rheinische Friedrich Wilhelms Universität, Bonn,
Germany
Keywords
Vitamin K antagonists, heparins, direct oral
thrombin inhibitors, factor Xa inhibitors,
clinical trials
Summary
Vitamin-K-antagonists (VKA) and heparins
have been complementary anticoagulants for
prevention and treatment of thrombosis for
almost 70 years. In contrast to heparins, VKA
have not been modified pharmacologically,
however treatment surveillance has improved
by introducing INR and self-monitoring/man-
agement. Disclosure of the molecular basis of
interaction with VKORC1, the target enzyme
of VKA, has helped to better understand cou-
marin sensitivity and resistance. New oral
anticoagulants have now been approved and
stimulated expectations in patients and phys-
icians to get rid of the burdening frequent
controls of VKA without loss of efficacy and
safety.
This review will summarize the development
and profile of the new substances. Main dif-
ference compared to VKA is their direct mode
of action against one clotting factor which is
factor IIa in dabigatran and factor Xa in rivar-
oxaban and other “xabanes” currently under
intensive investigation. Half lifes of the new
anticoagulants are much shorter than that of
the mainly used coumarins (phenprocoumon,
warfarin), making “anticoagulation bridging”
unnecessary before surgery. Therapeutic
width of direct thrombin inhibitors and factor
Xa inhibitors is broader and they are given at
fixed doses. Clinical studies in thrombopro-
phylaxis, thromboembolism and atrial fibril-
lation indicate at least non-inferiority or even
superior efficacy compared with enoxaparin
Correspondence to:
Ursula Harbrecht, MD
Institute of Experimental Haematology and Transfusion
Medicine, University of Bonn
Sigmund-Freud-Str. 25, 53105 Bonn, Germany
Tel. +49/(0)228/287 67 28, Fax +49/(0)228/287 60 87
E-mail: ursula.harbrecht@ukb.uni-bonn.de
and VKA at comparable safety outcomes. Limi-
tations of the new substances may arise from
gastrointestinal side effects, mode of metab-
olism and route of elimination. Specific anti-
dots are not available for none of them.
Undoubtedly, the new oral anticoagulants are
very promising. But, although thousands of
study patients already have been treated, there
are questions to be answered such as treat-
ment adherence in absence of monitoring,
safety and efficacy in risk patients, dosage ad-
justment and interactions with other drugs, be-
fore conclusions can be drawn towards their
potential to replace VKA.
Schlüsselwörter
Vitamin-K-Antagonisten, Heparine, direkte
orale Thrombininhibitoren, Faktor-Xa-Inhibito-
ren, klinische Studien
Zusammenfassung
Vitamin-K-Antagonisten (VKA) und Heparine
waren fast 70 Jahre lang komplementäre Anti-
koagulanzien für die Prävention und Therapie
von Thrombosen. Im Gegensatz zu den Hepari-
nen wurden die VKA pharmakologisch nicht
verändert, aber die Möglichkeiten der Thera-
pieüberwachung verbesserten sich mit Einfüh-
rung des INR-Bestimmung und des Selbstmoni-
torings. Die Entdeckung der molekularen
Grundlagen des Vitamin-K-Epoxid-Reduktase-
Komplexes (VKORC1), dem Schlüsselenzym der
Interaktion mit VKA, hat das Verständnis für
Phänomene wie Cumarinsensitivität und -re-
sistenz entscheidend erweitert. Neue Substan-
zen wurden mittlerweile entwickelt und zuge-
lassen. Sie haben sowohl bei Patienten als auch
Ärzten große Erwartungen geweckt, in naher
Alte und neue Antikoagulanzien
Hämostaseologie 2011; 31: 21–27
doi:10.5482/ha-1149
© Schattauer 2011 21
Zukunft die häufigen und belastenden INR-
Kontrollen nicht mehr zu benötigen.
Diese Übersichtsarbeit fasst die Entwicklung
und das Profil der neuen Substanzen zusam-
men. Ihr wesentlicher Unterschied im Ver-
gleich zu den VKA besteht darin, dass sie di-
rekt gegen einen einzelnen Gerinnungsfaktor
gerichtet sind, Faktor IIa (Thrombin) im Falle
von Dabitgatran und Faktor Xa bei Rivaroxa-
ban und anderen „Xabanen“, die derzeit in-
vensiv untersucht werden. Die Halbwertzei-
ten der neuen Antikoagulanzien sind wesent-
lich kürzer als die der überwiegend verwende-
ten VKA (Phenprocoumon, Warfarin), womit
ein „Antikoagulations-Bridging“ z. B. präope-
rativ unnötig würde. Die therapeutische Brei-
te von Thrombin- und Faktor-Xa-Inhibitoren
ist größer und sie werden in fixer Dosierung
appliziert. Klinische Studien aus den Berei-
chen Thromboseprophylaxe, Therapie von
Thromboembolien und Vorhoffflimmern wei-
sen zumindest Nicht-Unterlegenheit, wenn
nicht sogar bessere Effektivität im Vergleich
zu Enoxaparin und VKA nach, bei vergleichba-
ren Sicherheitsendpunkten. Limitierungen der
neuen Substanzen könnten sich auf Grund
gastrointestinaler Nebenwirkungen, Art der
Metabolisierung und dem Eliminierung erge-
ben. Ein spezifisches Antidot steht für keine
der Substanzen zur Verfügung.
Unzweifelhaft sind die neuen oralen Anti-
koagulanzien sehr vielversprechend. Dennoch
bleiben offene Fragen, obwohl schon tausen-
de von Patienten unter Studienbedingungen
behandelt wurden, z. B. die nach der Therapie-
adhärenz bei nicht mehr erfolgendem Moni-
toring, Effektivität und Sicherheit bei Risiko-
patienten sowie Dosisanpassung und Inter-
aktion mit anderen Medikamenten. Erst nach
deren Beantwortung wird eine abschließende
Bewertung möglich sein, ob die neuen Sub-
stanzen geeignet sind, die Vitamin-K-Antago-
nisten zu ersetzen.
Hämostaseologie 1/2011
22 U. Harbrecht: Old and new anticoagulants
In 2010, an estimated number of one mil-
lion patients (according to patient net-
works) were treated with oral anticoagu-
lants in Germany due to
● atrial fibrillation,
● deep vein thrombosis,
● pulmonary embolism,
● heart valve prosthesis,
● few other indications.
Another estimated one million patients
with atrial fibrillation is supposed to bene-
fit from antithrombotic therapy. For many
decades vitamin K antagonists (VKA) have
been the only choice for long-term treat-
ment (1, 2). After more than 30 years of re-
search new substances such as direct
thrombin inhibitors and direct factor Xa
inhibitors have been designed and more
and more of them are approved. The prin-
ciple expectation of patients and physicians
is focussed on long term treatment with
these substances instead of VKA which
hopefully would relief them from the
burden of frequent INR-controls and fear
of bleeding due to the narrow therapeutic
window of VKA.
This short review will summarize the
development and profile of the new oral
anticoagulants and compare them with the
old agents in terms of pharmacology,
mechanism of action, efficacy and safety.
An overview of the relevant clinical trials
will be given together with a careful projec-
tion of what we might expect in future.
Old anticoagulants
Vitamin K antagonists and heparins are
used complementary for antithrombotic
treatment. Unfractionated heparins with
their short half-life, quick onset and paren-
teral mode of administration are safe and
well controllable in thromboprophylaxis
and/or acute treatment of thrombosis. But
need of frequent applications in the s.c.
route (twice to three times daily) and
requirement of monitoring of intravenous
therapeutic doses due to considerable
intra- and interindividual variability makes
therapy unpredictable and inappropriate
for long-term use. Heparins inhibit all ser-
ine proteases to various extents with pre-
dominant inhibition of thrombin (factor
Hämostaseologie 1/2011
IIa) and factor Xa requiring antithrombin
as cofactor.
Further development of unfractionated
heparin (UFH) by shortening the glycos-
aminoglycan chain resulted in low-molec-
ular-weight heparin (LMWH) with higher
bioavalability and therefore better pre-
dictability of effect and without need of
routine monitoring. Longer half-life of
LMWH (2–4 hours after single s.c. injec-
tion) allow once daily injection for throm-
boprophylaxis or twice daily for treatment
of thrombosis (3).
Long-term anticoagulation is still done
by VKA, which are therapeutically used
coumarins, acting indirectly by inhibition
of carboxylation of the vitamin K depend-
ent factors II, VII, IX and X and proteins C,
S and Z, thereby abolishing their potential
to be activated on phospholipid surfaces
after calcium-dependent binding. Car-
boxylation does not take place because vit-
amin K is needed in this reaction and VKA
inhibit regeneration of vitamin K by inter-
fering with vitamin K epoxide reductase
(VKOR).
The gene was discovered in 2004 and en-
codes several isoforms, collectively termed
VKOR complex 1 (VKORC1) (4, 5). Since
introduction of dicumarol in 1941 in the
US and phenprocoumon (Marcumar®)
1954 in Germany these compounds have
not been altered pharmacologically. How-
ever, much has been done to improve treat-
ment surveillance. Increasing knowledge of
mechanism of VKA-action as well as in-
sights into interaction with foods and other
drugs rendered them more predictable. In-
troduction of INR and self-measurement/
monitoring and more careful and individ-
ually tailored loading doses have further in-
creased safety (6, 7). Recognition of
VKORC1 as the target of VKA and sub-
sequent disclosure of mutations and poly-
morphisms in VKORC1 as well as in
CYP2C9 have contributed substantially to
understand phenomena of coumarin sen-
sitivity and resistance (8, 9).
Common characteristics of
new anticoagulants
The new anticoagulants are synthetically
produced small molecules that act against
one clotting factor. In contrast to heparins
and fondaparinux their effect is indepen-
dent of antithrombin. They are binding se-
lectively and with high affinity to the active
site of the enzyme, thereby competitively
inhibiting the turnover of the natural sub-
strates. Reversible binding to the enzymes
is thought to be important for lowering the
bleeding risk. Whether factor IIa or factor
Xa inhibitors are better remains to be inves-
tigated. Prinicple of factor Xa inhibition is
reduction of thrombin burst in the propa-
gation phase of coagulation, an approach
with a straightforward effect. Inhibition of
thrombin itself, which is the key enzyme in
the coagulation cascade with many other
functions such as platelet activation, inter-
action with fibrinolysis, inflammatory pro-
cesses, and cell proliferation is less predict-
able as it may affect all properties of throm-
bin, however, this pleotropic effect may
provide advantages, too (10).
Direct thrombin inhibitors
Direct thrombin inhibitors (DTI) selec-
tively block the activity of the protease
thrombin (soluble and fibrin-bound) re-
sulting in inhibition of fibrin generation
and thrombus formation. DTI were in-
itially only available for parenteral use, later
followed by orally applicable substances.
Development was accompanied by severe
draw backs, the most surprising was with-
drawal of the first oral DTI ximelagatran
(Exanta®) in 2006 due to suspected liver
toxicity after it had been introduced on the
markets in 2004 (11). Lepirudin (Reflud-
an®) and argatroban (Argatra®) have been
implemented already some years ago as al-
ternative anticoagulants in heparin-in-
duced thrombocytopenia type II. Desiru-
din (Revasc®) is approved for thrombopro-
phylaxis in hip and knee replacement sur-
gery and bivalirudin (Angiox®) for percut-
aneous coronary interventions.
Dabigatran (Pradaxa®)
Dabigatran is a nonpeptide small molecule
acting as reversible inhibitor of thrombin.
In contrast to factor Xa inhibitors dabigat-
ran is a very polar, permanently charged
hydrophilic molecule, which itself has no
© Schattauer 2011
 U. Harbrecht: Old and new anticoagulants 23

bioavailability after oral administration.
Hence, a double prodrug was generated,
dabigatran etexilate (Pradaxa®), by ma-
sking the polar amidinium moiety as a car-
bamate ester and by turning the carboxy-
late into an ester group which allows bio-
availability of 6.5%. After resorption of the
prodrug, both polar groups of dabigatran
are restored quickly in plasma and liver by
esterase-mediated hydrolysis leading to
rapid onset of action with peak plasma
concentrations after 1 to 2 hours after in-
gestion. Terminal half-life is 14–17 hours.
Galenism of the drug was constructed with
tartaric acid allowing for solubility and re-
sorption in an acidic milieu. This however
accounts for gastrointestinal side effects,
mainly dyspepsia, and leads to reduction of
approximately 30% of bioavailability of da-
bigatran when combined with proton
pump inhibitors. Dabigatran is not meta-
bolized by cytochrome P450, reveales low
interactions with other drugs and no inter-
actions with foods. It is excreted renally, ap-
proximately 85% as active drug (씰Tab. 1)
(12, 13).
Factor Xa inhibitors
Factor Xa inhibitors can be “indirect” (e. g.
fondaparinux, idraparinux, idrabiotapari-
nux) or “direct” (e. g. rivaroxaban, apixaban,
edoxaban) depending on the requirement of
antithrombin to exert their antithrombotic
effect. They reduce thrombin burst during
propagation of coagulation. Whereas fonda-
parinux was rapidly integrated into clinical
use and the “xabanes” have started to follow,
idraparinux as a derivate of fondaparinux
with longer half-life (80 hours) was investi-
gated for treatment of deep vein thrombosis
and pulmonary embolism with a once
weekly subcutaneous administration, how-
ever it revealed increased bleeding compli-
cations and is no longer followed (14).
mediated conformational change of anti-
thrombin results in reversible binding to
factor Xa with 700-fold higher affinity
compared to heparins. Its administration is
once daily subcutaneously, half-life is
14–17 hours and it is exctreted renally (15).
Fondaparinux is approved since 2002,
meanwhile with a broad spectrum of indi-
cations comparable to that of enoxaparin,
including thromboprophylaxis, treatment
of venous thrombosis, pulmonary embol-
ism and unstable angina pectoris (16).
However, like all heparins, it is dependent
on the presence and functionality of anti-
thrombin. Nevertheless, the favourable
profile and convincing outcomes in clinical
trials in comparison to low-molecular-
weight heparins, mainly enoxaparin, has
stimulated a lot of work on factor Xa as
promising target for antithrombitic agents.
Rivaroxaban (Xarelto®)
Rivaroxaban is the first and most advanced
representative of the direct oral factor Xa in-
hibitors with 10 000-fold higher selectivity
for factor Xa than for other serine proteases.
It is an oxazolidininon derivate which is also
used in some antibiotics. Inhibition of factor
Xa is concentration-dependent and affects
prothrombinase complex-bound as well as
clot-associated factor Xa. Pharmacokinetic
and pharmacodynamics have been shown to
Tab. 1 Pharmacological characteristics of the direct thrombin inhibitor dabigatran and the oral fac-
tor Xa inhibitors rivaroxaban, apixaban and edoxaban
Dabigatran Rivaroxaban
(Pradaxa®) (Xarelto®)
molecular mass 628/472* 436
[g/mol]
protein binding [%] 34–35 92–95
bioavailability [%] 6.5 80–100
reduction by PPI 30% no
T(max) [h] 1–2 2–4
be predictable across 5 to 80 mg daily inges-
tion. Range of mean plasma concentration
was 12 to 201 ng/ml after daily intake of 5 to
20 mg of rivaroxaban. Peak plasma concen-
trations are achieved after 2 to 4 hours. Bioa-
vailability is 80–100% and terminal half-life
of 7–11 hours is between that of LMWH and
fondaparinux. Two third of the drug are
metabolised, one third is excreted un-
changed. Two third are eliminated renally,
one third through faeces. Rivaroxaban is
metabolised in the liver involving cytoch-
rome P450 (CYP3A4, CYP2J2) (17, 18). Of
the coagulation assays it affects prothrombin
time more than aPTT and it can be moni-
tored by anti-factor-Xa assays if desired (19).
Apixaban
Pharmacokinetics of apixaban have been
studied in healthy human males after ad-
ministration of a single 20 mg oral dose of
radiolabeled [14C]apixaban (20) and in
other species including rabbits (21). Bioa-
valability in humans is more than 50% and
peak plasma concentrations are reached in
3 to 4 hours (22). Other properties are sum-
marized in 씰Table 1.
Edoxaban (DU-176b)
Pharmacokinetics and pharmacodynamics
of edoxaban have been investigated in ani-
Apixaban Edoxaban
(DU-176b)
460 548
ca. 87 ?
> 50 45–50
no unlikely
3–4 1–2

Fondaparinux (Arixtra®) terminal half-life [h] 14–17 7–11 10–14 9–11

metabolism ca. 20% ca. 66% (liver) ? ?
Fondarinux was the first substance of the CYP450 dependent no ca. 32% CYP3A4, CYP2J2 CYP3A4
factor Xa inhibiting group, synthetically
excretion urine ca. 85% ca. 66% ca. 25% ca. 35%
generated, further condensating the low- (active) (ca. 33% active) (ca. 22% active) (ca. 24% active)
molecular heparin structure to the essential
pentasaccharid motif, responsible for bind- faeces ca. 6% ca. 33% ca. 56% ca. 62%
ing to antithrombin. The fondaparinux- *prodrug/drug

© Schattauer 2011 Hämostaseologie 1/2011

24 U. Harbrecht: Old and new anticoagulants

mal studies (23) and in a clinical safety
study in healthy volunteers. Peak plasma
concentrations are observed after 1–2
hours and terminal half-life is 5–6 hours.
Oral factor Xa inhibitors
Many other substances are currently inves-
tigated in clinical trials such as PRT-054021
(betrixaban), eribaxaban, LY 517717,
TAK-442, YM150.
Clinical trials
Clinical trials are listed for the most ad-
vanced new oral anticogulants with various
indications (씰Tab. 2) and results are pres-
ented briefly. Additionally, further pro-
grams are under way concerning other in-
dications including medically ill patients,
paediatric cases, haemodialysis, elective
percuteanous interventions.
Thromboprophylaxis in hip and
knee arthroplasty
Dabigatran
Dabigatran was compared with enoxaparin
in three phase III trials (RE-NOVATE, RE-
MODEL, RE-MOBILIZE) including more
than 8000 patients (25, 26, 27). Patients
were randomly assigned to receive 220 mg
or 150 mg dabigatran once daily, starting 1
to 4 hours with a half-dose after surgery, or
enoxaparin (40 mg once daily or 30 mg
bid), beginning the evening before surgery.
A pooled analysis with 6200 evaluable pa-
tients revealed comparable efficacy and
bleeding outcomes. The composite out-
comes of major VTE (proximal deep vein
thrombosis and/or pulmonary embolism)
and VTE-related mortality occurred in
3.3% of the enoxaparin group versus 3.0%
of the dabigatran 220 mg and 3.8% of the
dabigatran 150 mg group (proof of non-in-
feriority). Major bleeding occurred in 1.4
% of the enoxaparin group versus 1.4% in
the 220 mg and 1.1% in the 150 mg dabig-
atran group (28).
Direct factor Xa inhibitors
Rivaroxaban was investigated in knee and
hip arthroplasty in the RECORD 1–4 pro-
gram (29–32). Patients were randomly as-
signed to 10 mg rivaroxaban once daily be-
ginning 6–8 hours after surgery or enox-
aparin (40 mg once daily or 30 mg bid),
starting the evening before surgery in the
40 mg studies or beginning 12–24 hours
after surgery in 2 x 30 mg study (RECORD
4). A pooled analysis of the four studies
(12 729 patients) revealed better efficacy of
rivaroxaban, but higher, though not signifi-
cantly different, bleeding rates. Composite
endpoint of symptomatic VTE and all-
cause mortality in the day 12 ± 2 active
treatment pool occurred in 29/6183 pa-
tients receiving rivaroxaban (0.5%) versus
60/6200 patients receiving enoxaparin
(1.0%; p = 0.001). Major bleeding was ob-
served in 21 (0.3%) versus 13 (0.2%) pa-
tients, p = 0.23, major plus non-major
clinically relevant bleeding in 176 (2.8%)
versus 152 (2.5%) and any bleeding in 409
(6.6%) versus 384 (6.2%) patients, p =
0.38, respectively (33).
Huisman and co-workers compared En-
oxaparin versus dabigatran and rivar-
oxaban by pooling 6 phase III trials (18 405
participants) and came to the conclusion
that enoxaparin and dabigatran show simi-
lar rates of efficacy and bleeding, whereas
enoxaparin was less effective than rivar-
oxaban but had a lower risk of bleeding
(34).
Apixaban was assessed in knee replace-
ment surgery in the ADVANCE-2 study
with 3057 patients randomly allocated to
oral apixaban 2.5 mg twice daily starting 12
to 24 hours after wound closure or
enoxaparin 40 mg once daily starting 12
hours before surgery. In the recently pub-
lished analysis, 1973 patients were select-
able for primary outcome (composite of
asymptomatic and symptomatic DVT,
non-fatal PE and all-cause mortality)
which occurred in 147/976 (15%) patients
receiving apixaban versus 243/997 (24%)

Tab. 2 Overview of clinical trials with new oral anticoagulants for prevention and treatment of thromboembolism

substance indication
DVT-prophylaxis DVT/PE treatment ACS atrial fibrillation
Dabigatran phase III phase III phase II phase III
(Boehringer) RE-NOVATE* (25) RE-COVER* (39) RE-DEEM RE-LY* (48)
RE-MODEL* (26) RE-MEDY (40) completed completed (45)
RE-MOBILIZE* (27) RE-SONATE (41) ongoing
Rivaroxaban phase III phase III phase III phase III
(Bayer) RECORD 1–3* (29–31) EINSTEIN-DVT (42) completed ATLAS ACS TIMI 51 (46) ROCKET-AF (49)
RECORD 4* (32) EINSTEIN-PE (42) recruiting completed
EINSTEIN-Extension (42) completed
Apixaban phase III phase II phase III phase III
(Pfizer/ ADVANCE-2* (35) BOTTICELLI* (43) APPRAISE-2 (47) ARISTOTLE (51) ongoing, not recruiting
Bristol-Myers Sqibb) ADOPT ongoing (37) completed completed AVERROES (50) ongoing, not recruiting
Edoxaban phase III phase III Ø phase III
(DU-176b) STARS E-3 (38) HOKUSAI-VTE (44) ENGAGE AF TIMI 48 (52)
(Daiichi Sankyo) completed recruiting recruiting
DVT: deep vein thrombosis; PE: pulmonary embolism; ACS: acute coronary syndrome; * results of study published

Hämostaseologie 1/2011 © Schattauer 2011


enoxaparin (p < 0.0001). Major or clini-
cally relevant bleeding was observed in 53
(4%) of 1501 patients receiving apixaban
and 72 (5%) of 1508 patients treated with
enoxaparin. The investigators conclude
from the study that apixaban 2.5 mg twice
daily, starting on the morning after surgery,
offers a more effective alternative to enox-
aparin without increased bleeding (35). In
a meta-analysis of patients with total knee
arthroplasty it is concluded, that apixaban
is non-inferior to enoxaparin when used
for the same duration, but with consider-
able advantage regarding safety profile
(36). In the phase III ADOPT study apixa-
ban (2.5 mg bid) is investigated for preven-
tion of thrombosis-related events in pa-
tients with acute medical illness during and
after hospitalisation in comparison to
enoxaparin and placebo (37).
Edoxaban 30 mg once daily was com-
pared in the STARS E-3 trial with enoxapa-
rin 20mg twice daily for VTE prevention in
knee arthroplasty (38). An interim analysis
revealed DVT in 7.4% of edoxaban treated
patients versus 13.9% of patients receiving
enoxaparin. Major and clinically relevant
bleedings were not significantly different.
Treatment of deep vein thrombosis
and pulmonary embolism
Several study programs have been initiated
and are awaiting completion, analysis/in-
terpretation and publication (씰Tab.1).
The comparators are therapeutic doses of
UFH/LMWH/fondaparinux and VKA.
Dabigatran
Dabigatran has already shown non-in-
feriority in treatment of DVT and PE com-
pared to warfarin. In the RE-COVER study
2564 patients with DVT (~70%), PE
(~20%) or both (~10%) were randomly as-
signed after initial treatment with thera-
peutic LMWH, UFH or fondaparinux for 9
days to receive 150 mg dabigatran twice
daily or warfarin (INR 2.0–3.0) and fol-
lowed for 6 months. Primary endpoints
(recurrence of thromboembolism and/or
death) were observed in 30/1274 (2.4%)
patients treated with dabigatran versus
27/1265 (2.1%) under warfarin. Major
© Schattauer 2011
bleeding episodes occurred in 1.6% versus
1.9% in dabigatran and warfarin, respect-
ively. However, discontinuation of the
study drug due to adverse events, among
these diarrhoea and dyspepsia (signifi-
cant), was more frequent in dabigatran
(9.0%) than in warfarin (6.8%, p = 0.05),
(39). RE-MEDY (40) will look at long term
(18 months) secundary prophylaxis after
symptomatical VTE in comparison to war-
farin in patients with successful initial
treatment for 3 to 6 months. RE-SONATE
(41) is a placebo-controlled trial in patients
with recurrent VTE who have completed 6
to 18 months of treatment.
Direct factor Xa inhibitors
Rivaroxaban is currently investigated in
the EINSTEIN-DVT and EINSTEIN-PE
program at a dosage of 15 mg twice daily
for proof of non-inferiority compared to
standard treatment with LMWH/VKA. In
the EINSTEIN-extension-study, patients of
the DVT- and PE-program will be further
followed after 6 or 12 months of treatment
in a placebo controlled arm receiving 20 mg
once daily rivaroxaban (42).
Apixaban was investigated in a dose-
finding phase II study (BOTTICELLI-
DVT) at dosages of 5 mg twice-daily, 10 mg
twice-daily or 20 mg once-daily compared
to LMWH followed by VKA (43).
Edoxaban is currently under investi-
gation in the phase III HOKUSAI-VTE
study comparing LMWH/edoxaban tosyl-
ate (DU-176b) with LMWH/warfarin. Ed-
oxaban is given once daily 60 mg (44).
Acute coronary syndrome
Dose-finding studies in patients with acute
coronary syndrome (ACS) have not shown
uniform risk reduction. In case of equal or
superior efficacy by trend they tended to
have increased bleeding risks in com-
bination with single or dual anti-platelet
therapy. RE-DEEM (45) evaluated 4 doses
of dabigatran twice daily and placebo in ad-
dition to dual antiplaelet therapy. In the
placebo controlled ATLAS-TIMI 51 study
rivaroxaban is assessed with 2.5 mg and 5
mg twice daily additionally to aspirin alone
or aspirin and clopidogrel (46). The AP-
U. Harbrecht: Old and new anticoagulants 25
PRAISE-2 study is a placebo-controlled
study, too, investigating apixaban 5 mg
twice daily aganist placebo additionally to
standard care for ACS (47).
Atrial fibrillation
Atrial fibrillation (AF) is the predominant
indication for long-term anticoagulation.
Though it is no causal therapy, antithrom-
botic treatment represents basic care for
stroke prevention. Greatest expectations
with the new substances are focussed on
long-term anticoagulation with hope that
they finally may replace VKA. According to
AF-guidelines, indication for anticoagu-
lation in atrial fibrillation follows the
CHADS2 score or the more recently pro-
posed CHA2DS2-VASc score usually recom-
mended when ≥ 2 points are given. Among
the new anticoagulants dabigatran is so far
the most intensively studied substance in at-
rial fibrillation. In the RE-LY study 18113
patients had received dabigatran 110 mg or
150 mg twice daily in comparison to ad-
justed-dose warfarin (INR 2.0–3.0). Mean
CHADS2 score was 2.1. Primary outcome
was stroke or systemic embolism and was
observed after a median follow up period of
2 years in 1.69% of patients treated with war-
farin compared to 1.53% in the group that
received 110 mg dabigatran (p<0.001 for
non-inferiority), and 1.11% in the 150 mg
dabigatran group (p < 0.001 for superior-
ity). Major bleeding complications were not
significantly different in warfarin compared
to higher dose of dabigatran (3.11 vs 3.36%
per year) but were significantly lower in the
patients that received 110 mg dabigatran
(2.71%, p = 0.003). The rate of haem-
orrhagic stroke per year was lower with both
dabigatran doses (110 mg bid 0.12%; 150 mg
bid 0.10%) as compared to warfarin (0.38%;
p < 0.001). However, gastrointestinal bleeds
were 50% higher (p < 0.001) in the 150 mg
dabigatran group compared to warfarin and
this was related to patients >75 years and
with reduced renal function. Gastrointesti-
nal side effects, mainly dyspepsia, were more
than twice as frequently as with dabigatran
(11.8% and 11.3%) compared to warfarin
(5.8%) and accounted for 21% of discon-
tinuation of the study drug in the dabigatran
group compared to 17% in warfarin (48).
Hämostaseologie 1/2011
26 U. Harbrecht: Old and new anticoagulants
More than 14 000 patients have been rando-
mised in the ROCKET AF study to 20 mg
once daily rivaroxaban or dose-adjusted
warfarin (INR 2.0 –3.0). Follow up is
planned until 405 primary outcomes are ob-
served (49). In the ARISTOTLE study 18 206
patients with AF have been randomised to
apixaban 5 mg twice daily with warfarin as
comparator. At least 448 primary efficacy
outcomes will be awaited (50). The AVER-
ROES trial investigates apixaban 5 mg twice
daily compared to acetylsalicylic acid
(81–324 mg) in AF patients with at least one
risk factor for stroke and who have failed or
are unsuitable for VKA therapy (51).
Conclusions, perspectives
After decades of constant and nearly un-
changed options for antithrombotic treat-
ment new competitive and innovative sub-
stances are rushing on the markets backed
with very promising results of historically
large clinical trials. Most data at present are
available for the direct thrombin inhibitor
dabigatran and the factor Xa inhibitor ri-
varoxaban, being catched up by other fac-
tor Xa inhibitors as apixaban and edox-
aban. Although already thousands of pa-
tients have been treated under study con-
ditions, the new substances have to be
further followed carefully and intensively
in post-approval studies when a broad
spectrum of patients will be treated in
order to clarify open questions such as
● treatment adherence in absence of
monitoring,
● safety aspects in patients with renal/he-
patic impairment and multimorbidity,
● side effects (e. g. gastrointestinal) and
drug interactions,
● monitoring of risk patients,
● interpretation of coagulation tests (e. g.
before surgery).
Answering these questions will be one pre-
requisite before conclusions can be drawn
towards the potential of the new substances
to replace vitamin K antagonists. If they in
deed turn out to be superior and/or safer
than VKA in routine use with no need of
drug monitoring at regular intervals would
implicate a great step in the direction of
substantially improved medical care.
Hämostaseologie 1/2011
20. Raghavan N, Frost CE, Yu Z et al. Apixaban metab-
References olism and pharmacokinetics after oral adminis-
tration to humans. Drug Metab Dispos 2009: 37:
1. Link KP. The discovery of dicumarol and its sequels.
74–81.
Circulation 1959; 19: 97–107.
2. Ansell J, Hirsh , Hylek E et al. Pharmacology and
21. Zhang D, He K, Raghavan N et al. Comparative me-
tabolism of 14C-labeled apixaban in mice, rats, rab-
management of the vitamin K antagonists. Chest
bits dogs, and humans. Drug Metab Dispos 2009;
2008; 133: 160S-198S.
37: 1738–1748.
3. Weitz JI. Low-molecular-weight heparins. New
22. Garcia D, Libby E, Crowther MA. The new oral anti-
Engl J Med 1997; 337: 688–98.
4. Rost S, Fregin A, Ivascevicius V et al. Mutations in
coagulants. Blood 2010; 115: 15–20.
23. Furugohri T, Isobe K, Honda Y et al. DU-176b, a po-
VKORC1 cause warfarin resistance and multiple
tent and orally active factor Xa inhibitor: in vitro
coagulation factor deficiency type 2. Nature 2004;
and in vivo pharmacological profiles. J Thromb
427: 537–541.
Haemost 2008; 6: 1542–1549.
5. Li T, Chang CY, Jin DY et al. Identification of the
gene for vitamin K epoxide reductase. Nature 2004;
24. Ogata K, Mendell-Harary J, Tachibana M et al.
Clinical safety, tolerability, pharmacokinetics and
427: 541–544.
pharmacodynamics of the novel factor Xa inhibitor
6. Poller L. INR-calibration of the therapeutic range.
edoxaban in healthy volunteers. J Clin Pharm 2010;
Adv Exp Med Biol 1987; 214: 95–112.
50: 743–753.
7. Heneghan C, Alonso-Coello P, Grarcia-Alamino
25. Eriksson BI, Dahl OE, Rosencher N et al. Dabig-
JM. Self-monitoring of oral anticoagulation: a sys-
atran etexilate versus enoxaparin for prevention of
tematic review and meta-analysis. Lancet 2006; 367:
venous thromboembolism after total hip replace-
404–411.
8. Sconce EA, Khahn TI, Wynne HA et al. The impact
ment: a randomised, double- blind, non-inferiority
trial. Lancet 2007; 370: 949–956.
of CYP2C9 and VKOR1 genetic polymorphism and
patient characteristics upon warfarin dose require-
26. Eriksson BI, Dahl OE, Rosencher N et al. Oral da-
bigatran etexilate vs. subcutaneous enoxaparin for
ments: proposal for a new dosing regimen. Blood
the prevention of venous thromboembolism after
2005; 106: 2329–2333.
total knee replacement: The RE-MODEL random-
9. Oldenburg J, Bevans CG, Fregin A et al. Current
ized trial. J Thromb Haemost 2007; 5: 2178–2185.
pharmacogenetic developments in oral anticoagu-
27. RE-MOBILIZE Writing Committee, Ginsberg JS,
lation therapy: the influence of variant VKORC1
Davidson BL, Comp PC et al. Oral thrombin in-
and CYP2C9 alleles. Thromb Haemost 2007; 98:
hibitor dabigatran etexilate vs North American en-
570–578.
oxaparin regimen for prevention of venous throm-
10. Ansell J. Factor Xa or thrombin: is factor Xa a better
target? J Thromb Haemost 2007; Suppl 1: 60–64.
boembolism after knee arthroplasty surgery. J
Arthroplasty 2009; 24: 1–9.
11. Keisu M, Andersson TB. Drug induced liver injury
28. Friedmann RJ, Dahl OE, Rosencher N et al. Dabig-
in humans: the case of ximelagatran. Handb Exp
atran versus enoxaparin for prevention of venous
Pharmacol 2010; 196: 407–418.
thromboembolism after hip or knee arthroplasty: a
12. European Medicines Agency (2008) CHMP assess-
ment report for Pradaxa, INN dabigatran etexilate.
pooled analysis of three trials. Thromb Res 2010
126: 175–182.
Procedure No. EMEA/H/C/829. Doc.Ref.:
29. Eriksson BI, Borris LC, Friedmann RJ et al. Rivar-
EMEA/174363/2008 (www.ema.europa.eu).
oxaban versus enoxaparin for thromboprophylaxis
13. Eisert WG, Hauel N, Stangier W et al. Dabigatran:
after hip arthroplasty. N Engl J Med 2008; 358:
An oral novel potent reversible nonpeptide in-
hibitor of thrombin. Arterioscler Thromb Vasc Biol
2765–2775.
30. Kakkar AK, Brenner B, Dahl OE et al. Extended du-
2010; 30: 1885–1889.
ration rivaroxaban versus short-term enoxaparin
14. Prandoni P, Tormene D, Perlati M et al. . Idrapari-
fort he prevention of venous thromboembolism
nux: review of its clinical efficacy and safety for pre-
after total hip arthroplasty: a double-blind, rando-
vention and treatment of thromboembolic dis-
mised controlled trial. Lancet 2008; 372: 31–39.
orders. Expert Opin Investig Drugs 2008; 17:
31. Lassen MR, Ageno W, Borris LC et al. Rivroxaban
773–777.
versus enoxaparin for thromboprophylaxis after
15. Bauer KA. Fondaparinux sodium: a selective in-
total knee arthroplasty. N Engl J Med 2008; 358:
hibitor of factor Xa. Am J Health Syst Pharm 2001;
2776–2786.
58 Suppl 2: S14–S17.
32. Turpie AG, Lassen MR, Davidson BL et al. Rivarox-
16. Turpie AG, Bauer KA, Eriksson BI, Lassen MR. Fon-
aban versus enoxaparin for thromboprophylaxis
daparinux vs enoxaparin fort he prevention of ve-
after total knee arthroplasty (RECORD4): a rando-
nous thromboembolism in major orthopedic sur-
mised trial. Lancet 2009; 373: 1673–1680.
gery: a meta-analysis of 4 randomized double-blind
33. Turpie AG, Lassen MR, Eriksson BI et al. Rivar-
studies. Arch Int Med 2002; 162: 1833–1840.
oxaban for the prevention of venous thromboem-
17. European Medicines Agency (2008) CHMP assess-
bolism after hip or knee arthroplasty. Pooled analy-
ment report for Xarelto, INN rivaroxaban. Pro-
sis of four studies. Thromb Haemost 2010; 105.
cedure No. EMEA/H/C/000944. Doc.Ref.:
DOI: 10.1160/TH10-09-0601.
EMEA/543519/2008 (www.ema.europa.eu).
18. Perzborn E, Roehrig S, Straub A et al. Rivaroxaban:
34. Huisman MV, Quinlan DJ, Dahl OE, Schulman S.
Enoxaparin versus dabigatran and rivaroxaban for
A new oral factor Xa inhibitor. Arterioscler Thromb
thromboprophylaxis after hip or knee arthroplasty.
Vasc Biol 2010; 30: 376–381.
Circ Cardiovac Qual Outcomes 2010; 3: 652–660.
19. Lindhoff-Last E, Samama MM, Ortel TL et al. Assay
35. Lassen MR, Raskob GE, Gallus A et al. ADVANCE-2
for measuring rivaroxaban: their suitability and
limitations. Ther Drug Monit 2010; 32: 673–679.
investigators: Apixaban versus enoxaparin for
thromboprophylaxis after knee replacement (AD-
© Schattauer 2011

VANCE-2): a randomized double-blind trial. Lan-
cet 2010; 375: 807–815.
36. Huang J, Cao Y, Liao C et al. Apixaban versus enox-
aparin in patients with total knee arthroplasy. A
meta-analysis of randomised trials. Thromb
Haemost 2010; 105: DOI 10.1160/TH10-08-0552.
37. Website: http//clinicaltrials.gov/ct2/NCT00457002
38. Website: http//clinicaltrials.gov/ct2/NCT01181102
39. Schulman S, Kearon C, Kakkar AK et al. Dabigatran
versus warfarin in the tratment of acute venous
thromboembolism. N Engl J Med 2009; 361:
2342–2352.
40. http//clinicaltrials.gov/ct2/NCT00329238
41. http//clinicaltrials.gov/ct2/NCT00558259
42. http//clinicaltrials.gov/ct2/NTC00440193;
NCT00439777; NCT00439725
43. Botticelli investigators, Writing Committee; Buller
H, Deitchman D, Prins M, Segers A. Efficacy and
safety of the oral direct factor Xa inhibitor apixaban
© Schattauer 2011
for symptomatic deep vein thrombosis. Botticelli
DVT dose-ranging study. J Thromb Haemost 2008;
6: 1313–1318.
44. http//clinicaltrials.gov/ct2/NCT00986154
45. http//clinicaltrials.gov/ct2/NCT00621855
46. http//clinicaltrials.gov/ct2/NCT00809965
47. http//clinicaltrials.gov/ct2/NCT00831441
48. Connolly SJ, Ezekowitz MD, Yusuf S et al. Dabigat-
ran versus warfarin in patients with atrial fibril-
lation. N Engl J Med 2009; 361: 1139–1151.
49. ROCKET-AF Study Investigators: Rivaroxaban-
once daily, oral direct factor Xa inhibition com-
pared with vitamin K antagonism for prevention of
stroke and Embolism Trial in atrial Fibrillation:
rationale and design of the ROCKET AF study. Am
Heart J 2010; 159: 340–347.
50. Lopes RD, Alexander JH, Al-Khatib SM et al. Apixa-
ban for reduction in stroke and other thromboem-
bolic events in atrial fibrillation (ARISTOTLE)
U. Harbrecht: Old and new anticoagulants 27
trial: design and rationale. Am Heart J 2010; 159:
331–339.
51. Eikelboom JW, O’Donnell M, Yusuf S et al.
Rationale and design of AVERROES: apixaban ver-
sus acetylsalicylic acid to prevent stroke in atrial fi-
brillation patients who have failed or are unsuitable
for vitamin K antagonist treatment. Am Heart J
2010; 159: 348–353.
52. Ruff CT, Guigliano RP, Antman EM et al. Evalu-
ation of the novel factor Xa inhibitor edoxaban
compared with warfarin in patients with atrial fi-
brillation: design and rationale for the effective
anticoagulation with factor Xa next generation in
atrial fibrillation-thrombolysis in myocardial in-
farction study 48 (ENGAGE AF-TIMI 48). Am
Heart J 2010; 160: 635–641.
Hämostaseologie 1/2011