ESC Congress FA eBook 2016 - 11

Your FREE access to
ESC Congress content
all year long
www.escardio.org/365

Lower Bleeding Risks Associated
With Apixaban and Dabigatran
Compared With Warfarin
Written by Maria Vinall

Gregory Y. H. Lip, MD, University of Birmingham,
Birmingham, United Kingdom, reported result from an
electronic health records (EHR) study showing that patients
with non-valvular atrial fibrillation (NVAF) initially treated
with apixaban, or dabigatran have a lower risk of bleeding
compared with patients initiated on warfarin therapy.
NVAF increases the risk of stroke 5-fold and although
warfarin, can reduce the relative risk of stroke in these
patients by as much as two-thirds [Connolly AJ et al.
Circulation. 2008] it doubles the risk of bleeding compared to non-anticoagulated patients [Fang MC. J Interv
Card Electrophysiol. 2009]. Non-vitamin K antagonist oral
anticoagulants (NOACs) offer efficacy, safety and relative
convenience compared with warfarin, with a more predictable pharmacokinetic profile [Stangier J, Clemens A.
Clin Appl Thromb Hemost. 2009].
The primary objective of this study was to use EHR to
evaluate bleeding risk among patients with NVAF initiating therapy with warfarin (n = 47,215) or standard doses
of apixaban (5 mg BID; n = 5003), dabigatran (150 mg BID;
n = 3886), or rivaroxaban (20 mg QD; n = 7326). Patients
were followed until the earliest of OAC switch, bleeding
event, last encounter, or until 6-months post initiation of
index therapy. The primary study endpoint was presence
of any bleeding during the follow-up period. Bleeding
was defined using ICD-9-CM codes indicative of a major
or clinically relevant non-major bleed in an in- or outpatient setting. Cox proportional hazard analysis was
used to calculate the bleeding hazard ratios (HR) for
OAC therapies versus warfarin adjusting for age, sex,
region, stroke risk as measured by CHADS2 score and
baseline co-morbidities.
The mean follow-up period was 180 days for all but the
apixaban group where it was 153 days. Patients treated
with warfarin tended to be older (mean age about 74
years vs ≤ 69 years for the NOACs) and have a slightly
higher CHADS2 score (1.75 vs 1.31 to 1.45 for the NOACs)
and Charlson Comorbidities Index (1.50 vs 1.08 to 1.16 in
the NOAC groups). More warfarin patients had a history
of bleeding (10.2% vs 7.0% to 8.3% in NOAC groups).
After adjustment, there was a significantly lower risk of
bleeding in patients who initiated therapy with apixaban
(HR, 0.811; 95% CI, 0.732 to 0.899; P <.0001) or dabigatran (HR, 0.698; 95% CI, 0.621 to 0.785; P < .0001) compared with those who initiated therapy with warfarin.
No statistical difference was observed between rivaroxaban

ESC CONGRESS 2016
IN REVIEW

FOCUS ON ARRHYTHMIAS

and warfarin (HR, 0.985; 95% C!, 0.911 to 1.066; P = .707. The
results were consistent among the subgroup of patients
aged > 75 years. Overall, demographic risk factors
associated with an increase in bleeding risk with anticoagulation were as expected in this patient population.
In conclusion, this observational study identified an
association between lower bleeding with apixaban and
dabigatran compared with warfarin after adjustment for
differences in patient characteristics.

Risk of Intracranial Bleeds Lower
With Rivaroxaban Versus Warfarin
Written by Maria Vinall

Leif Friberg, MD, Karolinska Institute, Stockholm,
Sweden, reported that although the rate of major bleeding was similar in Swedish patients with atrial fibrillation
(AF) treated with rivaroxaban and warfarin, rivaroxaban
use was associated with fewer intracranial bleeds.
The large degree of variability in therapeutic range
(TTR) seen in the RE-LY, ROCKET AF, and ARISTOTLE
studies suggests that local standards of care can affect
the benefits of anticoagulant treatment. In Sweden, TTR
in routine clinical practice is 73.4%. The aim of this study
was to compare the incidence of bleeding with rivaroxaban (n = 6049) and warfarin (n = 40,855) in adult patients
identified in the National Swedish Patient Registry
between October 2012 and December 2014 as having AF
not previously treated with oral anticoagulants. Major
bleeding outcomes were defined as hospitalization or
death with a bleeding diagnosis as the principal or first
secondary diagnosis.
Mean follow-up was shorter for rivaroxaban than
for warfarin (304 days for rivaroxaban and 442 days
for warfarin) due to the gradual introduction of rivaroxaban during the inclusion period. The validity of the
diagnoses of bleeding was determined using a random
sample of the medical records of 761 patients with AF
(80% with and 20% without a diagnosis of bleeding). The
sensitivity and specificity for detection of fatal bleeding
or bleeding with hospitalization were 99.5% and 94.0%,
respectively. In this study the intention-to-treat (ITT)
population was analyzed based on the first prescribed
drug but without any data on treatment compliance.
For the on-treatment group the analysis was based on
records indicating possession of drug sufficient to cover
≥ 80% of days at risk.
At baseline patients in the rivaroxaban group were significantly older, had significantly higher CHA2DS2-VASc
and HASBLED scores, and were more educated with
significantly higher incomes (all P < .001).

Official Peer-Reviewed Highlights From ESC Congress 2016

http://www.escardio.org/365

Table of Contents for the Digital Edition of ESC Congress FA eBook 2016