IJCA-23979; No of Pages 6

International Journal of Cardiology xxx (2016) xxxxxx

Contents lists available at ScienceDirect

International Journal of Cardiology

journal homepage: www.elsevier.com/locate/ijcard

Major bleeding with vitamin K antagonists or direct oral anticoagulants

in real-life
Cecilia Becattini, MD, PhD a,, Laura Franco, MD a, Jan Beyer-Westendorf, MD b, Luca Masotti, MD c,
Cinzia Nitti, MD d, Simone Vanni, MD e, Giorgia Manina, MD f, Sergio Cattinelli, MD g, Roberto Cappelli, MD h,
Rodolfo Sbrojavacca, MD i, Fulvio Pomero, MD j, Sandra Marten, MD b, Giancarlo Agnelli, MD a
a

Internal and Cardiovascular Medicine-Stroke Unit, University of Perugia, Italy

Center for Vascular Medicine, University Hospital Carl Gustav Carus, Technische Universitt Dresden, Dresden, Germany
c
Internal Medicine, Santa Maria Nuova Hospital, Firenze, Italy
d
Emergency Medicine, Ospedali Riuniti Umberto I - Lancisi- Salesi, Ancona, Italy
e
Department of Emergency Medicine, Careggi University Hospital, Firenze, Italy
f
Emergency Medicine, Ospedale Maggiore, Lodi, Italy
g
Emergency Medicine, Ospedale Cattinara, Trieste, Italy
h
Internal Medicine, University of Siena, Italy
i
Emergency Medicine, Santa Maria della Misericordia Hospital, Udine, Italy
j
Internal Medicine, Santa Croce Hospital, Cuneo, Italy
b

a r t i c l e

i n f o

Article history:
Received 16 August 2016
Accepted 6 November 2016
Available online xxxx
Keywords:
Anticoagulants
Major bleeding
Intracranial hemorrhage
Dabigatran
Rivaroxaban
Apixaban

a b s t r a c t
Background: Limited data are available on major bleeding (MB) occurring during treatment with vitamin K
(VKAs) or direct oral anticoagulants (DOACs) outside clinical trials.
Methods: Patients hospitalized for MB while on treatment with VKAs or DOACs were included in a multicenter
study to compare clinical presentation, management and outcome of bleeding. The primary study outcome
was death at 30 days.
Results: Between September 2013 and September 2015, 806 patients were included in the study, 76% on VKAs
and 24% on DOACs. MB was an intracranial hemorrhage in 51% and 21% patients on VKAs or DOACs, respectively
(Odds Ratio [OR] 3.79; 95% condence interval [CI] 2.595.54) a gastrointestinal bleeding in 46% and 25% patients
on DOACs and VKAs, respectively (OR 2.62; 95% CI 1.873.68). Death at 30 days occurred in 130 patients (16%),
18% and 9% of VKA and DOAC patients (HR 1.95; 95% CI 1.193.22, p = 0.008). The rate of death at 30 days was
similar in VKA and DOAC patients with intracranial hemorrhage (26% and 24%; HR 1.05, 95% CI 0.542.02) and
gastrointestinal bleeding (11% and 7%; HR 1.46, 95% CI 0.573.74) and higher in VKA than DOAC patients with
other MBs (10% and 3%; HR 3.42, 95% CI 0.7815.03).
Conclusions: Admission for ICH is less frequent for DOAC patients compared with VKA patients. Admission for
gastrointestinal MB is more frequent for DOAC as compared to VKA patients. Mortality seems lower in patients
with MBs while on DOACs than VKAs but this nding varies across different types of MBs.
2016 Published by Elsevier Ireland Ltd.

1. Background
Bleeding is the most feared complication of oral anticoagulant treatment. During treatment with vitamin K antagonists (VKA), the incidence of major bleeding (MB) and fatal bleeding is estimated to be 2
5% and 0.51% per year [1,2]. Direct oral anticoagulants (DOACs) were
evaluated in clinical trials and then introduced in clinical practice for
the prevention of stroke and systemic embolism in atrial brillation
and for the prophylaxis and treatment of venous thromboembolism.
Corresponding author.
E-mail address: cecilia.becattini@unipg.it (C. Becattini).

In randomized clinical trials, DOACs were non-inferior to conventional

anticoagulant treatment in terms of efcacy and, in patients with atrial
brillation, safer than VKAs in terms of intracranial hemorrhage [36].
Meta-analyses of randomized clinical trials in patients with atrial brillation showed a lower incidence of fatal bleeding with DOACs than with
VKAs [7]. Preliminary real-world evidences from large cohorts are consistent with the results of clinical trials [814]. However, limited information is available on the clinical management and outcome of DOACassociated MB outside clinical trials [1517].
The aim of this study was to compare clinical presentation, management and outcome of MB in patients on treatment with VKAs or DOACs
in real-life.

http://dx.doi.org/10.1016/j.ijcard.2016.11.117
0167-5273/ 2016 Published by Elsevier Ireland Ltd.

Please cite this article as: C. Becattini, et al., Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life, Int J Cardiol (2016),
http://dx.doi.org/10.1016/j.ijcard.2016.11.117

C. Becattini et al. / International Journal of Cardiology xxx (2016) xxxxxx

2. Methods

3. Results

2.1. Study design

This is a prospective study obtained by joining an Italian (9 Italian hospitals) and a
German cohort.
2.2. Patients
Consecutive patients admitted for MB while on treatment with VKAs or DOACs
(regardless of the clinical indication for anticoagulation) were included in the Italian cohort. Patients were excluded in case of refusal of consent or incomplete data on clinical
management or outcome of index MB. In patients with low GCS or severe neurological impairment with speech or understanding disorders, the informed consent was obtained by
interview with the family. Patients included in the Dresden NOAC registry [9] who experienced a MB during follow-up were included in the German cohort provided that data on
clinical management and outcome were available.
MB was dened according to the criteria of the International Society on Thrombosis
and Hemostasis as clinically overt bleeding which was fatal or associated with any of the
following: i) a fall in hemoglobin level of 2 g/dl or more or documented transfusion of
at least 2 units of packed red blood cells, ii) involvement of a critical anatomical site
(intracranial, spinal, ocular, pericardial, articular, intramuscular with compartment syndrome, retroperitoneal) [18]. Fatal bleeding was adjudicated according to the assessment
of the attending physician.
Both the management of VKA- and DOAC-associated MB were at discretion of the attending physicians.
The study protocol was approved by local Ethical Committees and all patients gave informed consent before inclusion in the study.
2.3. Study outcomes
The primary outcome of the study was death at 30 days from major bleeding occurring during the hospital stay.
2.4. Data collection
The following data were collected: indication for anticoagulant therapy, patients
features (age, gender and comorbidities), concomitant medications, site of bleeding,
symptoms at admission, hemodynamic parameters and relevant laboratory results
(INR, hemoglobin count, renal function, etc.), medical (red blood cell, plasma units or hemostatic agents) or interventional treatment (surgery, endoscopy or endovascular treatment), and length of hospital stay.
At diagnosis of bleeding demographics (age, gender), hemodynamics, type of MB, indication for anticoagulation, comorbidities or conditions associated with MB (systemic hypertension, diabetes, history of renal or liver failure, malignancy, chronic heart failure, vascular
disease, previous stroke or bleeding, trauma), concomitant medications (antiplatelets,
NSAIDs) and laboratory data (INR, hemoglobin, renal function, etc.) were collected.
At discharge clinical status (death/survival), management strategies for bleeding (surgery, medical therapies, procedures) and duration of hospitalization were collected.
Shock was dened as systolic blood pressure lower than 90 mm Hg or a pressure drop
of 40 mm Hg or over for at least 15 min not due to a cause different from bleeding.
Abnormal kidney function was dened as chronic dialysis, or the presence of creatinine
clearance lower than 30 ml/min. Abnormal liver function as chronic hepatic disease
(as cirrhosis) or biochemical evidence of signicant hepatic derangement (bilirubin over
2-fold upper limit of normal, in association with aspartate aminotransferase/alanine
aminotransferase/alkaline phosphatase over 3-fold upper limit normal).
Chronic heart failure and vascular diseases were dened according to the CHADS2 and
CHA2DS2-VASc criteria [1921], as it follows:

Between September 2013 and September 2015, 1019 patients with

bleeding were evaluated for inclusion in the study, 806 were adjudicated as having a MB and were denitively included in the analysis. The
ow-diagram of the study is reported in Fig. 1.
At time of MB, 615 patients were on VKAs (76%) and 191 (24%) on
DOACs. The indication for anticoagulant treatment was atrial brillation
in 602 patients (75%), treatment of venous thromboembolism in 93
(11%) and valvular or other heart diseases in 111 (14%).
The baseline characteristics of DOAC and VKA patients at time of MB
are shown in Table 1. Diabetes, chronic heart failure and history of previous bleeding were more common in DOAC patients, while renal or
liver function impairment was more common in VKA patients.
3.1. Clinical presentation
The majority of MBs, 554 events, occurred spontaneously; 252 MBs
(31%) were associated with trauma or medical procedures: 40 on
DOAC patients (21%) and 212 on VKA patients (34%) (OR 1.99; 95% CI
1.352.92; p b 0.001). At presentation, similar proportions of DOAC or
VKA patients had hypotension or shock (17% and 18%, respectively); a
higher proportion of DOAC patients had a fall in hemoglobin of at least
2 g/dl (68% vs 49%, respectively; OR 2.24, 95% CI 1.593.15; p b 0.001).
The distribution of bleeding sites across DOAC and VKA patients is
reported in Table 2. Within the study population, admission for intracranial hemorrhage was less common in DOAC compared with VKA
patients (21% versus 51%; p b 0.001); gastrointestinal bleeding was
more common in DOAC as compared to VKA patients (46% vs 25%;
p b 0.001). Concerning intracranial hemorrhage, 182 patients had a parenchymal (46% in DOAC and 52% in VKA patients), 45 a subarachnoid
(19% vs 12%) and 125 a subdural localization (32% vs 36%). The localization of intracranial hemorrhage was intraventricular in one VKA and
unknown in one DOAC patient. Intracranial hemorrhage was not associated with trauma in 165 VKA patients (53%) and in 18 DOACs patients
(44%).
3.2. Clinical management
The mean length of hospital stay was 11 days in DOAC patients and
12 days in VKA patients.

- Chronic heart failure by the presence of signs and symptoms of either right or left ventricular failure or both, conrmed by non-invasive or invasive measurements demonstrating objective evidence of cardiac dysfunction.
- Vascular disease as the presence of peripheral vascular disease (any the following: intermittent claudication, previous surgery or percutaneous intervention on the abdominal aorta or the lower extremity vessels, vascular abdominal or thoracic
surgery, arterial and venous thrombosis) or coronary artery disease (prior myocardial
infarction, angina, percutaneous coronary interventions or coronary artery by-pass
surgery).

2.5. Statistical analysis

Frequency data are presented as proportions with 95% condence intervals (CI). Continuous data are shown as means standard deviations (SD). Student's t-test test and the
2 test were used for comparisons of continuous and categorical variables, respectively.
The Cox proportional hazards model was used to assess the risk of death during the
hospital stay and within 30 days from index MB in VKA and DOAC patients. Separate analyses were run for the overall study population and the subgroups of patients with intracranial hemorrhage and those with gastrointestinal hemorrhage. All the analyses were
adjusted for clinical features with different prevalence in VKA and DOAC patients.
Study analyses were performed by using SPSS 21.0.

Fig. 1. Study ow diagram.

Please cite this article as: C. Becattini, et al., Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life, Int J Cardiol (2016),
http://dx.doi.org/10.1016/j.ijcard.2016.11.117

C. Becattini et al. / International Journal of Cardiology xxx (2016) xxxxxx

Table 1
Patients' features associated with MB in the overall study population and in patients on
DOAC or VKA.
Patients
DOAC
VKA
OR
(N = 806) (N = 191) (N = 615) 95% CI
Age in years,
mean SD
Range
Male gender, n (%)
Indication for
anticoagulant
treatment
AF
VTE
Valvular disease
Other
Systemic arterial
hypertension, n (%)
Diabetes, n (%)

79 10

79 9

78 11

3699
448 (56)

3699
108 (56)

3794
340 (55)

602 (75)
93 (11)
87 (10)
24 (3)
591 (73)

154 (80)
33 (18)

3 (2)
142 (74)

448 (73)
60 (10)
87 (14)
21 (3)
449 (73)

184 (23)

60 (31)

124 (20)

Previous bleeding, n
(%)
Previous ICH, n (%)

174 (22)

80 (42)

94 (15)

27 (3)

4 (2)

23 (4)

Renal failure, n (%)

226 (28)

31 (16)

195 (32)

Liver failure, n (%)

40 (5)

3 (2)

37 (6)

Alcohol, n (%)

43 (5)

18 (9)

25 (4)

Previous stroke, n (%)

161 (20)

47 (25)

114 (18)

Malignancy, n (%)

92 (11)

29 (15)

63 (10)

49 (26)

168 (27)

Cardio-vascular
217 (27)
disease, n (%)
Chronic heart failure, 207 (26)
n (%)
Antiplatelet use, n (%) 77 (10)

60 (31)

147 (24)

13 (7)

64 (10)

NSAIDs use, n (%)

8 (4)

25 (4)

33 (4)

107 (56)

255 (41)

1.79
1.292.49

b0.001

13 (7)

74 (12)

0.53
0.291.00

0.042

Fibrinogen
Prothrombin
complex
concentrates
Activated factor VII

1
268 (33)

1
35 (18)

0
233 (38)

0.37
0.250.55

b0.001

9 (1)

8 (1)

ns

Vitamin K

483 (60)

12 (6)

471 (77)

ns

Tranexamic acid

18 (2)

12 (6)

6 (1)

0.001

Endoscopy

231 (29)

88 (46)

143 (23)

b0.001

Angiography

49 (6)

5 (3)

44 (7)

ns

Other procedures

39 (5)

17 (9)

22 (4)

b0.001

Surgery

138 (17)

23 (12)

115 (19)

0.40
0.053,21
0.02
0.010.04
6.80
2.5218.39
2.82
2.013.96
0.35
0.140.89
2.63
1.375.07
0.59
0.370.96

b0.001
b0.001
b0.001
0.022
0.003
0.033

0.013
0.04
0.067
0.061
ns
0.038
ns
ns

The management of MBs is reported in Table 3. At admission, anticoagulant treatment was withdrawn in 89% of patients (97% and 62% of
VKA and DOAC patients, respectively; OR 21.76, 95% CI 12.3938.23,
p b 0.001).
Table 2
Sites of MB.
MBs

TOT
DOAC
VKA
OR
(N = 806) (N = 191) (N = 615) 95% CI

Intracranial, n (%)

354 (44)

41 (21)

313 (51)

b0.001

Gastrointestinal, n
(%)
Soft/muscle, n (%)

239 (30)

88 (46)

151 (25)

80 (10)

11 (6)

69 (11)

Retroperitoneal, n
(%)
Genito-urinary, n
(%)
Pleural/pericardial/
peritoneal, n (%)
Articular, n (%)

33 (4)

2 (1)

31 (5)

27 (3)

15 (8)

12 (2)

21 (3)

5 (3)

16 (3)

18 (2)

9 (5)

9 (1)

Upper airways, n (%) 15 (2)

9 (5)

6 (1)

Ocular, n (%)
Spinal, n (%)
Other, n (%)

9 (5)
0

0
5 (1)

9 (1)
5 (1)
5 (1)

Patients
DOAC
Warfarin
OR
(N = 806) (N = 191) (N = 615) 95% CI
Transfusion 2units 362 (45)
units of red blood
cells
Fresh frozen plasma 87 (11)

ns

AF: atrial brillation; VTE: venous thromboembolism; ICH: intracranial hemorrhage;

NSAIDs: nonsteroidal anti-inammatory drugs.

0.26
0.180.39
2.62
1.873.68
0.48
0.250.93
0.20
0.050.84
4.28
1.979.32
1.01
0.362.78
3.33
1.308.51
5.02
1.7614.29

Table 3
Management of major bleedings.

1.05
ns
0.761.46

1.07
0.741.55
1.81
1.262.61
3.99
2.785.74
0.55
0.191.61
0.42
0.270.63
0.25
0.080.82
2.45
1.314.61
1.43
0.972.11
1.57
0.982.52
0.92
0.631.33
1.46
1.022.08
0.63
0.341.17
1.03
0.462.33

b0.001
0.027
0.015
b0.001
ns
0.008
0.001
b0.001
ns

Transfusion of at least two units of red blood cells was given more
commonly in DOAC than in VKA patients. Overall, fresh frozen plasma
and prothrombin complex concentrates were used in 11% and 33% of
patients, respectively, both less commonly in DOAC than in VKA patients. More specically, prothrombin complex concentrates were
used in 160 and 14 patients with intracranial hemorrhage on VKAs
and DOACs, respectively (51 vs 34%, OR 2.02, 95% CI 1.023.99, p =
0.044), and in 73 and 21 patients with extracranial bleedings on VKA
and DOAC, respectively (24 vs 14%, OR 1.96, 95% CI 1.153.33, p =
0.013). Fibrinogen, activated factor VII and tranexamic acid were rarely
used in both patient groups (Table 3).
For the management of MBs, DOAC patients more frequently required endoscopy or other procedures (arthrocentesis, pleural drainage,
nasal packing, and other), while VKA patients more frequently required
angiography or surgery (Table 3). No patient received hemodialysis. In
91 patients, MB was managed only by transfusion of red blood cells.
3.3. Clinical outcome
During the hospital stay 134 patients died. Death at 30 days occurred
in 130 patients (16%), 112 VKA and 18 DOAC patients (18% and 9%, HR
1.95; 95% CI 1.193.22, p = 0.008) (Fig. 2). The increased risk for 30-day
mortality in VKA compared to DOAC patients was conrmed after
adjusting for differences on baseline clinical features (diabetes, previous
bleeding, renal and renal failure, chronic heart failure; HR 1.67, 95% CI
1.002.79, p = 0.05) and after exclusion of patients on anticoagulant
treatment for valvular diseases (HR 2.07, 95% CI 1.253.43, p = 0.04)
(Table 4).
Death at 30 days occurred in 26% and 24% of the VKA and DOAC patients with intracranial hemorrhage (81 and 10 patients, HR 1.05, 95% CI
0.542.02), respectively. The similar risk for 30-day mortality in VKA
compared to DOAC patients after intracranial hemorrhage was conrmed at adjusted analysis (HR 0.92, 95% CI 0.471.81). Among patients
with gastrointestinal bleedings, death at 30 days occurred in 11% and 7%
of the VKA and DOAC patients, respectively (16 and 6 patients, HR 1.46,
95% CI 0.573.74; adjusted HR 1.76, 95% CI 0.654.76). Concerning other
MBs, death at 30 days occurred in 10% and 3% of the VKA and DOAC patients (15 and 2 patients, HR 3.42, 95% CI 0.7815.03; adjusted HR 1.71,
95% CI 0.319.47), respectively (Table 4).
When the analysis was limited to MBs not associated with trauma,
death at 30 days occurred in 19% and 9% of VKA and DOAC patients,

Please cite this article as: C. Becattini, et al., Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life, Int J Cardiol (2016),
http://dx.doi.org/10.1016/j.ijcard.2016.11.117

C. Becattini et al. / International Journal of Cardiology xxx (2016) xxxxxx

complex concentrates were independent predictors for death at

30 days (eTable 1).
4. Discussion

Fig. 2. Kaplan Meier curves for cumulative risk of death at 30 days in DOAC and VKA
patients.

respectively (78 and 14 patients; HR 2.13, 95% CI 1.203.76, p = 0.009;

adjusted HR 1.84, 95% CI 1.013.32, p = 0.045); 30-day mortality in
non-traumatic intracranial hemorrhage was 39% and 31% in DOAC and
VKA patients, respectively (7 and 51 patients; HR 1.05, 95% CI 0.54
2.03; adjusted HR 1.08, 95% CI 0.542.13) (Table 4).
Bleeding was the main cause of death (13 DOAC and 72 VKA patients, corresponding to 7% and 12%), followed by infection (1 DOAC
and 18 VKA patients, corresponding to 0.5% vs 3%). Four patients died
due to ischemic stroke (all VKA patients), 4 due to acute coronary syndrome (two patients in each treatment group) and 2 due to pulmonary
embolism (both VKA patients).
Among fatal bleedings, 68 were intracranial and 17 extracranial; the
case fatality rate of intracranial hemorrhage was 24% and 19% with
DOACs as compared with VKAs (OR 1.42, 95% CI 0.663.06) and that
of extracranial hemorrhage was 5% and 2% with VKAs as compared
with DOACs (OR 2.38, 95% CI 0.678.42).
At univariate analysis, an association with death at 30 days was
found for increasing age (HR 1.05, 95% CI 1.031.08, p b 0.001), intracranial hemorrhage (HR 2.9, 95% CI 1.984.24, p b 0.001), shock at admission (HR 2.12, 95% CI 1.273.54, p 0.004), use of prothrombin complex
concentrates (HR 1.94, 95% CI 1.372.74, p b 0.001), fall in hemoglobin
of 2 g/L or higher (HR 0.38, 95% CI 0.260.56, p b 0.001), and need for
transfusion of 2 units of red blood cells or more (HR 0.42, 95% CI 0.28
0.62, p b 0.001) while treatment with DOACs was associated with reduced death at 30 days.
At multivariate analysis, treatment with DOACs was no longer associated with reduced death at 30 days; increasing age, intracranial hemorrhage, and shock at admission but not treatment with prothrombin

Our real-life study in patients admitted to the hospital for MBs

shows that intracranial hemorrhage is more common in patients treated with VKAs and gastrointestinal bleeding in patients treated with
DOACs. As a remarkable nding, the risk for death at 30 days seems
lower in DOAC as compared with VKA patients.
The nding of less common intracranial hemorrhage with DOACs is
consistent with the results of phase III studies with all DOACs in patients
with atrial brillation [2228] and with those of phase IV studies with
dabigatran and rivaroxaban in this clinical setting [6,10,12,29,30]. The
biological plausibility of this favorable association seems to be based
on lower suppression of thrombin generation with DOACs compared
with warfarin, reduced penetration of the blood brain barrier and probably with tissue factordependent mechanisms [31,32]. No denitive
difference was observed in the incidence of intracranial bleeding in
DOAC and VKA patients in trials on the treatment of VTE. However,
these studies were not sized to show such difference.
In our study, MB was more likely to manifest as gastrointestinal MBs
in DOAC patients as compared to VKA patients. This nding was previously suggested in phase III studies in the prevention of stroke and systemic embolism in patients with atrial brillation [2225,27] but not
consistently conrmed in phase IV studies [6,10,12,33]. Our ndings
on gastrointestinal bleeding, together with the increased likelihood for
upper airways and genito-urinary MBs, could suggest an increase in
mucosal bleeding with DOACs as compared with VKAs more than a specic effect on gastrointestinal tract.
Consistently with the results from phase III studies and metaanalyses, we found a lower 30-day mortality in patients with MB
while on treatment with DOACs as compared with those on VKAs; this
difference was conrmed in the adjusted analyses and seems to be
mainly accounted for by non-intracranial MBs. In previous studies, conicting results were reported on mortality associated with intracranial
hemorrhage in patients treated with DOAC or VKA [7,3437]. In our
study, in patients with intracranial hemorrhage, similar risks for death
and similar case-fatality rates were observed in patients on treatment
with DOACs or VKAs. The lower mortality observed in our study in patients with extracranial MB on DOACs as compared with VKAs is consistent with the results of a recent systematic review and meta-analysis of
20 randomized controlled trials on risk of fatal bleeding while on DOACs
or VKAs [7].
Different management strategies for MBs were used in DOAC patients as compared with VKA patients. In particular, prothrombin complex concentrates were more frequently given to VKA as compared with
DOAC patients while blood transfusions and endoscopy or other procedures were more commonly used in DOAC patients. These differences
are probably related to the different types of MBs observed in DOAC
and VKA patients. Concerning the lower proportion of patients withdrawn for anticoagulant treatment in DOAC as compared to VKA

Table 4
Death at 30 days.
Death rates (%)

Overall study population

Patient with ICHa
Patients with gastrointestinal bleeding
Patients with other major bleeding
Exclusion of patients on treatment for valvular diseases
Non traumatic major bleeding
Non traumatic ICHa
a
b

VKA

DOAC

18
26
7
10
19
19
31

9
24
11
3
9
9
39

HR (95% CI)

aHRb (95% CI)

1.78 (1.082.93)
1.05 (0.542.02)
1.46 (0.573.74)
3.42 (0.7815.03)
2.07 (1.253.43)
2.13 (1.203.76)
1.05 (0.542.03)

1.67 (1.002.79)
0.92 (0.471.81)
1.76 (0.654.76)
1.71 (0.319.47)

1.84 (1.013.32)
1.08 (0.542.13)

Intracranial hemorrhage.
Adjusted analysis for differences on baseline clinical features (diabetes, previous bleeding, renal and liver failure, chronic heart failure).

Please cite this article as: C. Becattini, et al., Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life, Int J Cardiol (2016),
http://dx.doi.org/10.1016/j.ijcard.2016.11.117

C. Becattini et al. / International Journal of Cardiology xxx (2016) xxxxxx

patients, this could reect a lower prevalence of MB perceived as immediately life-threatening by the attending physician. It is conceivable that
the increased use of prothrombin complex concentrates in VKA patients
was driven by an increased severity of bleeding, as it was suggested in a
previous case series [37]. However, it should be considered that no consensus exists on the potential role of prothrombin complex concentrates in the treatment of DOACs-associated MB [38]. According to
current available evidences, mortality in patients with intracranial hemorrhage while on VKA treatment is still quite high despite warfarin reversal [3941]. Whether the availability in clinical practice of specic
DOAC antidotes will change management and clinical outcome of
DOAC-associated MBs remains to be dened [42].
Our study has some limitations. This is not a management study, and
thus all comparisons between DOACs and VKAs should be seen with
caution as all therapeutic decisions were left to the discretion of attending physicians. This cohort study included patients at time of MB and
thus it does not allow the assessment of predictors of MBs or a direct
comparison of the safety of DOACs and VKAs. In particular, as about
one third of the study patients were included in one tertiary hospital
with a Neurosurgery Department, a Stroke Unit and a large Intensive
Care Unit, the proportion of patients with ICH is considerable.
Patients on anticoagulant treatment for valvular diseases were only
represented in the VKA group and this could have accounted for a selection bias. However, the results obtained after exclusion of these patients
(n = 87) were consistent with those obtained in the overall study population. As further limitations, laboratory tests at admission (as coagulation tests, hemoglobin values or platelet count) and information on
anticoagulation resumption were not systematically reported. However, our study also has some strengths. We evaluated one of the largest
cohort of patients with MBs occurring on oral anticoagulant treatment
in real-life, with a considerable number of patients on treatment with
DOACs [15,17,4345]. This allowed the comparison of DOAC and VKA
MBs on clinically relevant issues. In this setting the multicenter design
should also be seen as a strength of the study.
In conclusion, admission for ICH is less frequent for DOAC patients
compared with VKA patients. Admission for gastrointestinal MB is
more frequent for DOAC as compared to VKA patients. Mortality
seems lower in patients with MBs while on DOACs than VKAs but this
nding varies across different types of MBs.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.ijcard.2016.11.117.
Disclosures
Cecilia Becattini reports lectures fees from Boehringer Ingelheim and
Bayer HealthCare.
Jan. Beyer-Westendorf has received honoraria and research support
from Bayer HealthCare, Boehringer Ingelheim, Bristol-Myers Squibb,
and Pzer.
Sandra Marten has received honoraria from Bayer HealthCare.
Laura Franco, Luca Masotti, Cinzia Nitti, Simone Vanni, Giorgia
Manina, Sergio Cattinelli, Roberto Cappelli, Rodolfo Sbrojavacca and
Fulvio Pomero have nothing to disclose.
Giancarlo Agnelli reports lectures fees from Boehringer Ingelheim,
from Sano, from Bayer HealthCare, and from Daiichi-Sankyo.
Contributions of authors
Cecilia Becattini contributed to the interpretation of data, drafting
and critical revision of the manuscript as well as supervision of all statistical analyses and is the guarantor of the paper, taking responsibility for
the integrity of the work as a whole, from inception to published article;
Laura Franco contributed to the conception and design of the study, to
the interpretation of data, drafting and critical revision of the manuscript; Jan Beyer-Westendorf, Luca Masotti, Cinzia Nitti, Simone Vanni,
Giorgia Manina, Sergio Cattinelli, Roberto Cappelli, Rodolfo Sbrojavacca,

Fulvio Pomero, Sandra Marten contributed to the interpretation of data,

drafting and critical revision of the manuscript; Giancarlo Agnelli contributed to the interpretation of data, drafting and critical revision of
the manuscript.
Conict of interest
The authors report no relationships that could be construed as a conict of interest.
References
[1] A. Rubboli, C. Becattini, F.W. Verheugt, Incidence, clinical impact and risk of bleeding
during oral anticoagulation therapy, World J. Cardiol. 3 (2011) 351358.
[2] S. Schulman, R.J. Beyth, C. Kearon, M.N. Levine, Hemorrhagic complications of anticoagulant and thrombolytic treatment: American college of chest physicians
evidence-based clinical practice guidelines (8th edition), Chest 133 (2008)
257S298S.
[3] R. Providencia, E.L. Grove, S. Husted, S. Barra, S. Boveda, J. Morais, A meta-analysis of
phase III randomized controlled trials with novel oral anticoagulants in atrial brillation: comparisons between direct thrombin inhibitors vs factor Xa inhibitors and
different dosing regimens, Thromb. Res. 134 (2014) 12531264.
[4] L.A. Castellucci, C. Cameron, G. Le Gal, M.A. Rodger, D. Coyle, P.S. Wells, T. Clifford, E.
Gandara, G. Wells, M. Carrier, Clinical and safety outcomes associated with treatment of acute venous thromboembolism: a systematic review and meta-analysis,
JAMA 312 (2014) 11221135.
[5] E.S. Eerenberg, S. Middeldorp, M. Levi, A.W. Lensing, H.R. Bller, Clinical impact and
course of major bleeding with rivaroxaban and vitamin K antagonists, J. Thromb.
Haemost. 13 (2015) 15901596.
[6] S.S. Adam, J.R. McDufe, T.L. Ortel, J.W. Williams Jr., Comparative effectiveness of
warfarin and new oral anticoagulants for the management of atrial brillation and
venous thromboembolism: a systematic review, Ann. Intern. Med. 157 (2012)
796807.
[7] J. Skaistis, T. Tagami, Risk of fatal bleeding in episodes of major bleeding with new
oral anticoagulants and vitamin K antagonists: a systematic review and meta-analysis, PLoS One 10 (2015), e0137444.
[8] M.R. Southworth, M.E. Reichman, E.F. Unger, Dabigatran and postmarketing reports
of bleeding, N. Engl. J. Med. 368 (2013) 12721274.
[9] J. Beyer-Westendorf, K. Frster, S. Pannach, F. Ebertz, V. Gelbricht, C. Thieme, F.
Michalski, C. Khler, S. Werth, K. Sahin, L. Tittl, U. Hnsel, N. Weiss, Rates, management, and outcome of rivaroxaban bleeding in daily care: results from the Dresden
NOAC registry, Blood 124 (2014) 955962.
[10] T.B. Larsen, L.H. Rasmussen, F. Skjth, K.M. Due, T. Callrus, M. Rosenzweig, G.Y. Lip,
Efcacy and safety of dabigatran etexilate and warfarin in real-world patients with
atrial brillation: a prospective nationwide cohort study, J. Am. Coll. Cardiol. 61
(2013) 22642273.
[11] G.V. Fontaine, K.D. Mathews, S.C. Woller, S.M. Stevens, J.F. Lloyd, R.S. Evans, Major
bleeding with dabigatran and rivaroxaban in patients with atrial brillation: a
real-world setting, Clin. Appl. Thromb. Hemost. 20 (2014) 665672.
[12] Food and Drug Administration, FDA Drug Safety Communication: FDA study of
Medicare patients nds risks lower for stroke and death but higher for gastrointestinal bleeding with Pradaxa (dabigatran) compared to warfarin, 2014. 2014 (Available at: http://www.fda.gov/Drugs/DrugSafety/ucm396470.htm. Accessed
December 19).
[13] D.J. Graham, M.E. Reichman, M. Wernecke, R. Zhang, M.R. Southworth, M. Levenson,
T.C. Sheu, K. Mott, M.R. Goulding, M. Houstoun, T.E. MaCurdy, C. Worrall, J.A.
Kelman, Cardiovascular, bleeding, and mortality risks in elderly Medicare patients
treated with dabigatran or warfarin for non-valvular atrial brillation, Circulation
131 (2015) 157164.
[14] Food and Drug Administration, FDA Drug Safety Communication: Update on the risk
for serious bleeding events with the anticoagulant Pradaxa, 2012. Available at:
http://www.fda.gov/Drugs/DrugSafety/ucm326580.htm.
[15] R. Berger, S.D. Salhanick, M. Chase, M. Ganetsky, Hemorrhagic complications in
emergency department patients who are receiving dabigatran compared with warfarin, Ann. Emerg. Med. 61 (2013) 475479.
[16] S. Kaatz, M. Crowther, Reversal of target-specic oral anticoagulants, J. Thromb.
Thrombolysis 36 (2013) 195202.
[17] J. Bouget, E. Oger, Emergency admissions for major haemorrhage associated with direct oral anticoagulants, Thromb. Res. 136 (2015) 11901194.
[18] S. Schulman, C. Kearon, Subcommittee on Control of Anticoagulation of the Scientic
and Standardization Committee of the International Society on Thrombosis and
Haemostasis, Denition of major bleeding in clinical investigations of
antihemostatic medicinal products in non-surgical patients, J. Thromb. Haemost. 3
(2005) 692694.
[19] European Heart Rhythm Association, European Association for Cardio-Thoracic Surgery, A.J. Camm, P. Kirchhof, G.Y. Lip, U. Schotten, I. Savelieva, S. Ernst, I.C. Van
Gelder, N. Al-Attar, G. Hindricks, B. Prendergast, H. Heidbuchel, O. Aleri, A. Angelini,
D. Atar, P. Colonna, R. De Caterina, J. De Sutter, A. Goette, B. Gorenek, M. Heldal, S.H.
Hohloser, P. Kolh, J.Y. Le Heuzey, P. Ponikowski, F.H. Rutten, ESC Committee for Practice Guidelines, Guidelines for the management of atrial brillation: the task force
for the management of atrial brillation of the European Society of Cardiology
(ESC), Europace 12 (2010) 13601420.

Please cite this article as: C. Becattini, et al., Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life, Int J Cardiol (2016),
http://dx.doi.org/10.1016/j.ijcard.2016.11.117

C. Becattini et al. / International Journal of Cardiology xxx (2016) xxxxxx

[20] G.Y. Lip, R. Nieuwlaat, R. Pisters, D.A. Lane, H.J. Crijns, Rening clinical risk stratication for predicting stroke and thromboembolism in atrial brillation using a novel
risk factor-based approach: the Euro heart survey on atrial brillation, Chest 137
(2010) 263272.
[21] B.F. Gage, A.D. Waterman, W. Shannon, M. Boechler, M.W. Rich, M.J. Radford, Validation of clinical classication schemes for predicting stroke: results from the national
registry of atrial brillation, JAMA 285 (2001) 28642870.
[22] S.J. Connolly, M.D. Ezekowitz, S. Yusuf, J. Eikelboom, J. Oldgren, A. Parekh, J. Pogue,
P.A. Reilly, E. Themeles, J. Varrone, S. Wang, M. Alings, D. Xavier, J. Zhu, R. Diaz,
B.S. Lewis, H. Darius, H.C. Diener, C.D. Joyner, L. Wallentin, RE-LY Steering Committee and Investigators, Dabigatran versus warfarin in patients with atrial brillation,
N. Engl. J. Med. 361 (2009) 11391151.
[23] M.R. Patel, K.W. Mahaffey, J. Garg, G. Pan, D.E. Singer, W. Hacke, G. Breithardt, J.L.
Halperin, G.J. Hankey, J.P. Piccini, R.C. Becker, C.C. Nessel, J.F. Paolini, S.D.
Berkowitz, K.A. Fox, R.M. Califf, R.O.C.K.E.T.A.F. Investigators, Rivaroxaban versus
warfarin in nonvalvular atrial brillation, N. Engl. J. Med. 365 (2011) 883891.
[24] C.B. Granger, J.H. Alexander, J.J. McMurray, R.D. Lopes, E.M. Hylek, M. Hanna, H.R. AlKhalidi, J. Ansell, D. Atar, A. Avezum, M.C. Bahit, R. Diaz, J.D. Easton, J.A. Ezekowitz, G.
Flaker, D. Garcia, M. Geraldes, B.J. Gersh, S. Golitsyn, S. Goto, A.G. Hermosillo, S.H.
Hohnloser, J. Horowitz, P. Mohan, P. Jansky, B.S. Lewis, J.L. Lopez-Sendon, P. Pais,
A. Parkhomenko, F.W. Verheugt, J. Zhu, L. Wallentin, ARISTOTLE Committees and
Investigators, Apixaban versus warfarin in patients with atrial brillation, N. Engl.
J. Med. 365 (2011) 981992.
[25] R.P. Giugliano, C.T. Ruff, E. Braunwald, S.A. Murphy, S.D. Wiviott, J.L. Halperin, A.L.
Waldo, M.D. Ezekowitz, J.I. Weitz, J. pinar, W. Ruzyllo, M. Ruda, Y. Koretsune, J.
Betcher, M. Shi, L.T. Grip, S.P. Patel, I. Patel, J.J. Hanyok, M. Mercuri, E.M. Antman,
ENGAGE AF-TIMI 48 Investigators, Edoxaban versus warfarin in patients with atrial
brillation, N. Engl. J. Med. 369 (2013) 20932104.
[26] D. Caldeira, M. Barra, F.J. Pinto, J.J. Ferreira, J. Costa, Intracranial hemorrhage risk
with the new oral anticoagulants: a systematic review and meta-analysis, J. Neurol.
262 (2015) 516522.
[27] C.T. Ruff, R.P. Giugliano, E. Braunwald, E.B. Hoffman, N. Deenadayalu, M.D.
Ezekowitz, A.J. Camm, J.I. Weitz, B.S. Lewis, A. Parkhomenko, T. Yamashita, E.M.
Antman, Comparison of the efcacy and safety of new oral anticoagulants with warfarin in patients with atrial brillation: a meta-analysis of randomised trials, Lancet
383 (2014) 955962.
[28] C. Chai-Adisaksopha, M. Crowther, T. Isayama, W. Lim, The impact of bleeding complications in patients receiving target-specic oral anticoagulants: a systematic review and meta-analysis, Blood 124 (2014) 24502458.
[29] A.J. Camm, P. Amarenco, S. Haas, S. Hess, P. Kirchhof, S. Kuhls, M. van Eickels, A.G.
Turpie, X.A.N.T.U.S. Investigators, XANTUS: a realworld, prospective, observational
study of patients treated with rivaroxaban for stroke prevention in atrial brillation,
Eur. Heart J. 37 (2016) 11451153.
[30] S. Tamayo, W. Frank Peacock, M. Patel, N. Sicignano, K.P. Hopf, L.E. Fields, T. Sarich, S.
Wu, D. Yannicelli, Z. Yuan, Characterizing major bleeding in patients with
nonvalvular atrial brillation: a pharmacovigilance study of 27 467 patients taking
rivaroxaban, Clin. Cardiol. 38 (2015) 6368.
[31] B. Dale, J.W. Eikelboom, J.I. Weitz, E. Young, J.S. Paikin, M. Coppens, R.P. Whitlock, S.J.
Connolly, J.S. Ginsberg, J. Hirsh, Dabigatran attenuates thrombin generation to a
lesser extent than warfarin: could this explain their differential effects on intracranial hemorrhage and myocardial infarction? J. Thromb. Thrombolysis 35 (2013)
295301.
[32] S. Haas, Rivaroxaban an oral, direct factor Xa inhibitor lessons from a broad clinical study programme, Eur. J. Haematol. 82 (2009) 339349.

[33] N.S. Abraham, S. Singh, G.C. Alexander, H. Heien, L.R. Haas, W. Crown, N.D. Shah,
Comparative risk of gastrointestinal bleeding with dabigatran, rivaroxaban, and
warfarin: population based cohort study, BMJ 350 (2015) h1857.
[34] R.G. Hart, H.C. Diener, S. Yang, S.J. Connolly, L. Wallentin, P.A. Reilly, M.D. Ezekowitz,
S. Yusuf, Intracranial hemorrhage in atrial brillation patients during
anticoagulation with warfarin or dabigatran: the RE-LY trial, Stroke 43 (2012)
15111517.
[35] G.J. Hankey, S.R. Stevens, J.P. Piccini, Y. Lokhnygina, K.W. Mahaffey, J.L. Halperin, M.R.
Patel, G. Breithardt, D.E. Singer, R.C. Becker, S.D. Berkowitz, J.F. Paolini, C.C. Nessel,
W. Hacke, K.A. Fox, R.M. Califf, ROCKET AF Steering Committee and Investigators, Intracranial hemorrhage among patients with atrial brillation anticoagulated with
warfarin or rivaroxaban: the rivaroxaban once daily, oral, direct factor Xa inhibition
compared with vitamin K antagonism for prevention of stroke and embolism trial in
atrial brillation, Stroke 45 (2014) 13041312.
[36] C. Held, E.M. Hylek, J.H. Alexander, M. Hanna, R.D. Lopes, D.M. Wojdyla, L. Thomas,
H. Al-Khalidi, M. Alings, D. Xavier, J. Ansell, S. Goto, W. Ruzyllo, M. Rosenqvist,
F.W. Verheugt, J. Zhu, C.B. Granger, L. Wallentin, Clinical outcomes and management
associated with major bleeding in patients with atrial brillation treated with
apixaban or warfarin: insights from the ARISTOTLE trial, Eur. Heart J. 36 (2015)
12641272.
[37] C. Beynon, O.W. Sakowitz, D. Strzinger, B. Orakcioglu, A. Radbruch, A. Potzy, A.W.
Unterberg, Intracranial haemorrhage in patients treated with direct oral anticoagulants, Thromb. Res. 136 (2015) 560565.
[38] G. Dickneite, M. Hoffman, Reversing the new oral anticoagulants with prothrombin
complex concentrates (PCCs): what is the evidence? Thromb. Haemost. 111 (2014)
189198.
[39] A.R. Parry-Jones, M. Di Napoli, J.N. Goldstein, F.H. Schreuder, S. Tetri, T. Tatlisumak, B.
Yan, K.M. van Nieuwenhuizen, N. Dequatre-Ponchelle, M. Lee-Archer, S. Horstmann,
D. Wilson, F. Pomero, L. Masotti, C. Lerpiniere, D.A. Godoy, A.S. Cohen, R. Houben, R.
Al-Shahi Salman, P. Pennati, L. Fenoglio, D. Werring, R. Veltkamp, E. Wood, H.M.
Dewey, C. Cordonnier, C.J. Klijn, F. Meligeni, S.M. Davis, J. Huhtakangas, J. Staals, J.
Rosand, A. Meretoja, Reversal strategies for vitamin K antagonists in acute intracerebral hemorrhage, Ann. Neurol. 78 (1) (2015) 5462.
[40] D. Dowlatshahi, K.S. Butcher, N. Asdaghi, S. Nahirniak, M.L. Bernbaum, A. Giulivi, J.K.
Wasserman, M.C. Poon, S.B. Coutts, Canadian PCC Registry (CanPro) Investigators,
Poor prognosis in warfarin-associated intracranial hemorrhage despite
anticoagulation reversal, Stroke 43 (2012) 18121817.
[41] M. Alonso de Leciana, N. Huertas, J.A. Egido, A. Muriel, A. Garca, G. Ruiz-Ares, E.
Dez-Tejedor, B. Fuentes, Questionable reversal of anticoagulation in the therapeutic
management of cerebral haemorrhage associated with vitamin K antagonists,
Thromb. Haemost. 110 (6) (2013) 11451151.
[42] J.H. Levy, W. Ageno, N.C. Chan, M. Crowther, P. Verhamme, J.I. Weitz, Subcommittee
on Control of Anticoagulation, When and how to use antidotes for the reversal of direct oral anticoagulants: guidance from the SSC of the ISTH, J. Thromb. Haemost. 14
(2016) 623627.
[43] P. Harper, L. Young, E. Merriman, Bleeding risk with dabigatran in the frail elderly, N.
Engl. J. Med. 366 (2012) 864866.
[44] B. Ross, M.A. Miller, K. Ditch, Clinical experience of life-threatening dabigatranrelated bleeding at a large, tertiary care, academic medical center: a case series, J.
Med. Toxicol. 10 (2014) 223228.
[45] L. Masotti, G. Lorenzini, C. Seravalle, G. Panigada, G. Landini, R. Cappelli, S. Schulman,
Management of new oral anticoagulants related life threatening or major bleedings
in real life: a brief report, J. Thromb. Thrombolysis 39 (2015) 427433.

Please cite this article as: C. Becattini, et al., Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life, Int J Cardiol (2016),
http://dx.doi.org/10.1016/j.ijcard.2016.11.117