Patients with intracranial bleeding and atrial fibrillation treated with left atrial appendage occlusion: Results from the Amplatzer Cardiac Plug registry

    Patients with intracranial bleeding and atrial fibrillation treated with left atrial appendage occlusion: results from the Amplatzer Cardiac Plug registry Apostolos Tzikas, Xavier Freixa, Laura Llull, Sameer Gafoor, Samera Shakir, Heyder Omran, George Giannakoulas, Sergio Berti, Gennaro Santoro, Joelle Kefer, Adel Aminian, Steffen Gloekler, Ulf Landmesser, Jens Erik Nielsen-Kudsk, Ignacio Cruz-Gonzalez, Prapa Kanagaratnam, Fabian Nietlispach, Reda Ibrahim, Horst Sievert, Wolfgang Schillinger, Jai-Wun Park, Bernhard Meier, Haralampos Karvounis PII: DOI: Reference:

S0167-5273(16)33705-6 doi:10.1016/j.ijcard.2017.02.042 IJCA 24568

To appear in:

International Journal of Cardiology

Received date: Revised date: Accepted date:

8 November 2016 24 January 2017 10 February 2017

Please cite this article as: Tzikas Apostolos, Freixa Xavier, Llull Laura, Gafoor Sameer, Shakir Samera, Omran Heyder, Giannakoulas George, Berti Sergio, Santoro Gennaro, Kefer Joelle, Aminian Adel, Gloekler Steffen, Landmesser Ulf, Nielsen-Kudsk Jens Erik, Cruz-Gonzalez Ignacio, Kanagaratnam Prapa, Nietlispach Fabian, Ibrahim Reda, Sievert Horst, Schillinger Wolfgang, Park Jai-Wun, Meier Bernhard, Karvounis Haralampos, Patients with intracranial bleeding and atrial fibrillation treated with left atrial appendage occlusion: results from the Amplatzer Cardiac Plug registry, International Journal of Cardiology (2017), doi:10.1016/j.ijcard.2017.02.042

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Patients with intracranial bleeding and atrial fibrillation treated with left atrial appendage occlusion: results from the

PT

Amplatzer Cardiac Plug registry

Apostolos Tzikas1, MD, PhD, Xavier Freixa2, MD, PhD, Laura Llull2, MD, PhD,

SC RI

Sameer Gafoor3, MD, Samera Shakir4, MD, Heyder Omran5, MD, George Giannakoulas1, MD, PhD, Sergio Berti6, MD, Gennaro Santoro7, MD, Joelle Kefer8, MD, Adel Aminian9, MD, Steffen Gloekler4, MD, Ulf Landmesser10, MD,
 Jens Erik

NU

Nielsen-Kudsk11, MD, Ignacio Cruz-Gonzalez12, MD, Prapa Kanagaratnam13, MD,

MA

Fabian Nietlispach4,10, MD, Reda Ibrahim14, MD, Horst Sievert3, MD, Wolfgang Schillinger15, MD, Jai-Wun Park16, MD, Bernhard Meier4, MD, Haralampos

ED

Karvounis1, MD, PhD

AHEPA University Hospital, Thessaloniki, Greece.

2

Hospital Clinic of University of Barcelona, Barcelona, Spain.

3

CardioVascular Center Frankfurt, Frankfurt, Germany.

4

University Hospital of Bern, Bern, Switzerland.

5

University Hospital of Bonn, Bonn, Germany.

6

Fondazione Toscana Gabriele Monasterio, Massa, Italy.

7

Ospedale Careggi di Firenze, Florence, Italy.

8

St-Luc University Hospital, Brussels, Belgium.

9

Centre Hospitalier Universitaire de Charleroi, Charleroi, Belgium.

AC

CE

PT

1

10

University Hospital of Zurich, Zurich, Switzerland.

11

Aarhus University Hospital, Skejby, Denmark.

12

University Hospital of Salamanca, Salamanca, Spain.

13

Imperial College Healthcare NHS Trust, London, United Kingdom.

14

Montreal Heart Institute, Montreal, Canada.

15

Universitätsmedizin Göttingen, Göttingen, Germany.

1

ACCEPTED MANUSCRIPT 16

Coburg Hospital, Coburg, Germany.

PT

Total word count: 3635

SC RI

Brief title: Patients with AF and previous intracranial bleeding treated with LAAO

NU

Conflict of interest:

A.T., X.F., Sa.Ga., H.O., S.B., G.S., J.K., A.A., U.L., J.N.K., I.C., P.K., F.N., R.I.,

MA

H.S., W.S., J-W.P. and B.M. are consultants for SJM. L.L., S.S., G.G., St. Gl. and H.K. have no conflict of interest to declare.

ED

Address of correspondence:

PT

Apostolos Tzikas, MD, PhD, AHEPA University Hospital, Department of Cardiology, Asklipiou 10, 57001 Thessaloniki, Greece. E-mail: [email protected]. Tel:

AC

CE

+302310994335, Fax: +302310994336

2

ACCEPTED MANUSCRIPT ABSTRACT BACKGROUND In patients with non-valvular atrial fibrillation (NVAF),

PT

intracranial bleeding (ICB) constitutes a very challenging situation in which the rate of both ischemic and hemorrhagic events is increased. In these patients,

SC RI

left atrial appendage occlusion (LAAO) might represent a very valid alternative.

OBJECTIVES To investigate the procedural safety and long-term outcome of

NU

patients undergoing LAAO therapy due to previous ICB.

MA

METHODS Data from the Amplatzer Cardiac Plug multicenter registry on 1047 consecutive patients were analyzed. Patients with previous ICB as

ED

indication for LAAO were compared to patients with other indications. RESULTS A total of 198 patients (18.9%) with previous ICB were identified.

PT

The CHA2DS2-VASc score was similar (4.5±1.5 vs. 4.4±1.6, p=0.687) and the HAS-BLED score was higher in patients with previous ICB compared to those

CE

without (3.5±1.1 vs. 3.1±1.2, p<0.001). No significant differences in peri-

AC

procedural major adverse events were observed (2.5 vs 5.4%, p=0.1). Patients with previous ICB were more frequently on single acetylsalicylic acid therapy after LAAO (42.4% vs. 28.3%; p<0.001). With an average follow-up of 1.3 years, the observed annual stroke/TIA rate (procedure and follow-up) for patients with previous ICB was 1.4% (75% relative risk reduction). The observed annual major bleeding rate (procedure and follow-up) for patients with previous ICB was 0.7% (89% relative risk reduction). CONCLUSIONS In patients with NVAF and previous ICB, LAAO seemed to be a safe procedure and was associated with a significant reduction in stroke/TIA and a remarkably low frequency of major bleeding during follow-up.

3

ACCEPTED MANUSCRIPT Word count: 242

PT

KEYWORDS

SC RI

LAA closure, stroke, prevention, device

ABBREVIATIONS LIST ACP = Amplatzer Cardiac Plug

NU

ASA = acetylsalicylic acid

AF = atrial fibrillation

ED

ICB = intracranial bleeding

MA

ACT = activated clotting time

LAA = left atrial appendage

PT

LAAO = left atrial appendage occlusion NVAF = non-valvular atrial fibrillation

CE

OAC = oral anticoagulation

AC

TEE = trans-esophageal echocardiography TIA = transient ischemic attack VKA = vitamin K antagonist

4

ACCEPTED MANUSCRIPT TEXT INTRODUCTION

non-valvular

atrial

fibrillation

(NVAF)

and

PT

Left atrial appendage occlusion (LAAO) may be considered in patients with contraindication

to

oral

SC RI

anticoagulation (OAC) therapy (1,2). In most patients, contraindication to OAC is mainly related to the risk or clinical consequences of any repeated bleeding. Intracranial bleeding (ICB), one of the most feared medical conditions, is

NU

associated with a high recurrence rate and devastating clinical consequences

MA

in case of recurrence (3). Moreover, ICB constitutes a very challenging situation in which the rate of ischemic events is also increased, requiring more

ED

intensive measures to prevent ischemic strokes (4). LAAO has emerged as a valuable alternative for these patients as both ischemic and bleeding risks

PT

seem to be reduced (1,5,6). However, several questions arise in such a highrisk cohort of patients, as the safety and efficacy of the procedure or the

CE

tolerance to different antithrombotic drugs after LAAO remain unclear. The

AC

objective of the present study was to investigate the procedural safety and long-term outcomes of patients undergoing LAAO therapy due to previous ICB.

METHODS Data from the Amplatzer Cardiac Plug (ACP) multicenter registry on 1047 patients with NVAF were analyzed. This registry prospectively included all consecutive patients who underwent LAA closure with the ACP between December 2008 and November 2013 in 22 centers, starting from their first

5

ACCEPTED MANUSCRIPT patient, without roll-in subjects. In this series, the most common indication for LAAO was previous major bleeding. Details regarding the registry have been

PT

previously published (1). Patients with previous ICB as indication for LAAO were identified and compared to patients with other indications. Procedural

SC RI

success was defined as successful implantation of the ACP in the LAA. The definition of major adverse events was: Acute (0-7 days) occurrence of death, stroke (ischemic or hemorrhagic), systemic embolism, and procedure or

NU

device related complications requiring major cardiovascular or endovascular

MA

intervention. Report of adverse events during follow-up included death (cardiovascular or non-cardiovascular), stroke, transient ischemic attack (TIA),

ED

systemic embolism, and major bleeding. All centers provided a summary for every reported major adverse event. Antithrombotic medication was recorded

PT

at the admission date, and at last follow-up visit. During the procedure, patients received heparin with a target activated clotting time (ACT) of >250

CE

sec. The recommendation by the device manufacturer following LAAO was to

AC

prescribe acetylsalicylic acid (ASA) 80-100 mg and clopidogrel 75 mg daily for 1-3 months and then only ASA 80-100 mg for at least another 3 months. However, the choice and the duration of antithrombotic therapy were individualized depending on the patient history, indication for LAAO, and physician preference. Device efficacy to prevent stroke, TIA, and systemic embolism was tested by comparing the actual event rate at follow-up with the predicted event rate by the CHA2DS2-VASc score. In a similar way, bleeding reduction was assessed by comparing the actual major bleeding events to the rate predicted by the HAS-BLED score. (7,8)

6

ACCEPTED MANUSCRIPT

STATISTICAL ANALYSIS

PT

Continuous variables are shown as means (SD) and categorical variables are shown as frequencies and percentages. In case of skewed distribution,

SC RI

variables are presented as medians (inter-quartile range). Continuous variables were tested by using the independent samples t-test and categorical variables by using the Fischer’s exact test. A two-sided p value < 0.05 was

NU

considered statistically significant. All statistical analyses were performed with

ED

MA

SPSS 20.0 software (SPSS Inc., Chicago, IL, USA).

RESULTS

PT

A total of 198 patients (18.9%) with previous ICB were identified (Table 1). Patients were younger (73.7 ± 7.9 vs. 75.2 ± 8.5 years), more commonly male

CE

(69.7 vs 60.1%, p=0.012), had a higher rate of previous stroke (63.6 vs

AC

32.7%, p<0.001), and less congestive heart failure (16.2% vs 28.5%, p<0.001) compared to those without previous ICB. The CHA2DS2-VASc score was similar between groups (4.5 ± 1.5 vs. 4.4 ± 1.6, p=0.687) and the HASBLED score was higher in patients with previous ICB (3.5 ± 1.1 vs 3.1 ± 1.2, p<0.001). The predicted annual stroke risk was similar according to the CHA2DS2-VASc score (5.8 ± 2.8% vs. 5.6 ± 2.8%, p=0.480) whereas the annual major bleeding risk according to the HAS-BLED score was higher in patients with previous ICB (6.4 ± 3.9% vs. 5.1 ± 3.7%, p<0.001).

7

ACCEPTED MANUSCRIPT Procedural success was achieved in 97.3% (1019/1047 patients) (Table 2). No significant differences in peri-procedural major safety events among

PT

groups were observed (2.5 vs. 5.4%, p=0.1). One patient died due to periprocedural ICB; this patient did not have a history of previous ICB and

SC RI

was treated with LAAO due to recurrent stroke while being on vitamin K antagonist (VKA).

NU

The antithrombotic medication at baseline, after LAAO and at last follow-up is

MA

shown in Table 3. Patients with previous ICB were receiving more frequently ASA and less frequently VKA at baseline. In addition, patients with previous

ED

ICB were more frequently on single ASA therapy after LAAO (74.5% vs.

PT

61.3%; p<0.001).

The average follow-up was 16.2 ± 12.0 months and was complete for 1001

CE

out of 1019 (98.2%) successfully implanted patients. The observed annual

AC

stroke/TIA rate (procedure and follow-up) for patients with previous ICB was 1.4% (4.3% absolute reduction, 75% relative reduction according to the CHA2DS2-VASc score) and 2.5% for patients without previous ICB (3.1% absolute reduction, 55% relative reduction), (Figure 1, panel A). The observed annual major bleeding rate (procedure and follow-up) for patients with previous ICB was 0.7% (5.7% absolute reduction, 89% relative reduction according to the HAS-BLED score) and 2.4% for those without (2.7% absolute reduction, 47% relative reduction), (Figure 1, panel B). In fact, only two major bleeding events were observed in patients with previous ICB: one periprocedural femoral hematoma requiring transfusion and one small

8

ACCEPTED MANUSCRIPT cerebral bleeding that was discovered one day after discharge and had no

PT

clinical sequelae.

A trans-esophageal echocardiogram (TEE) at follow-up was available in

SC RI

632/1001 (63%) of successfully implanted patients [64% (122/190) of patients with previous ICB and 63% (510/811) of those without, p=0.747] and was performed at a median of 7 (interquartile range 3-11) months after LAAO. The

NU

rate of device thrombosis and peri-device leaks did not vary significantly

MA

among groups (device thrombosis in previous ICB patients vs others: 1.7% vs 5.2%; p=0.139 / peri-device leaks in previous ICB patients vs others: 3.3% vs

DISCUSSION

PT

ED

1.4%; p=0.237).

CE

The main findings of this study are the following: 1) LAAO is a safe procedure

AC

with similar procedural outcomes in patients with and without previous ICB; 2) LAAO is associated with a reduction of cerebrovascular and a remarkable reduction of hemorrhagic events after comparing the observed and expected rates according to the CHA2DS2-VASc score and HAS-BLED score, respectively; 3) Despite the less intensive antithrombotic therapy after LAAO in patients with previous ICB, no increase in the frequency of device thrombosis or peri-device leaks was observed.

The study evaluates the role of LAAO in patients with previous ICB. Although no LAA anatomical differences among patients with and without previous ICB

9

ACCEPTED MANUSCRIPT were anticipated, the potential risk of repeated ICB during the procedure and especially after heparinization (ACT>250sec) deserved specific analysis.

PT

Despite heparinization, patients with previous ICB had similar procedural success and complications compared to those without (Table 2). In fact, the

SC RI

single procedural ICB occurred in a patient without previous ICB, highlighting the feasibility and safety of the procedure in this specific setting. Similarly, Saw et al. (9) showed favorable procedural outcomes in a small cohort of

NU

patients with previous ICB treated with either Watchman, ACP, or Amulet

MA

devices.

ED

The 75% reduction of stroke/TIA at follow up compared to the hypothetical incidence without LAAO or OAC was in accordance with previous studies

PT

(1,6,10) but the 89% reduction of major bleeding was even higher than anticipated. A possible explanation for this is the lack of OAC after LAAO and

CE

the short duration of dual antiplatelet therapy. Patients with previous ICB were

AC

on less intensive therapy than patients without as shown in Table 3. Single antiplatelet therapy with aspirin (41%) followed by low weight molecular heparin (12%) and dual antiplatelet therapy with ASA and clopidogrel (11%) were the most common antithrombotic therapies post-LAAO.

Recurrent ICB is one of the most feared complications by physicians as it is associated with a very high rate of mortality and disability (3,11). Patients with a history of ICB are at high risk for a recurrent event and commonly suffer from other major vascular disease (12), particularly those with lobar ICB (13,14) and during the first year after the event, thus provoking uncertainty

10

ACCEPTED MANUSCRIPT about the re-initiation of antithrombotic drugs (15). The strongest risk factors associated to ICB recurrence are: 1) lobar location of the initial ICB; 2) older

PT

age; (3) the presence and number of microbleeds on brain magnetic resonance imaging and; 4) on-going OAC therapy (16). Although antiplatelet

SC RI

agents increase the risk of spontaneous ICB, they have a substantially lower bleeding risk than OAC with a VKA (17). Therefore, in the post acute setting, once the ICB is resolved, physicians need to decide if antithrombotic therapy

NU

should be resumed despite the increased risk of repeated bleeding. As for

MA

other hemorrhagic conditions like subdural hematoma and subarachnoid hemorrhage, there is limited information on the risk of recurrent bleeding and

ED

embolic events. However, the risk seems to be low after aneurysmal subarachnoid bleeding once the aneurysm is correctly secured (18). A recent

PT

meta-analysis addressing the re-initiation of antithrombotic therapy after spontaneous subdural hematoma, paradoxically estimated the risk of

CE

recurrence at 22% in patients not restarting anticoagulation and at 11% in

AC

those restarting anticoagulants (19). All events occurred within the first month after the initial surgical evacuation. However, the analysis included only 64 patients from 3 studies so no firm conclusions can be made apart from the actual lack of data in this important field. Noteworthy, based on our findings, LAAO should be considered as a promising alternative to chronic OAC in patients with previous history of ICB. From our point of view, an individualized selection of the antithrombotic therapy after LAAO based on the ICB source and the risk of recurrence with different drug combinations is probably the most reasonable option. In our series, although single ASA was the most common therapy after LAAO, no specific data on the type of ICB were

11

ACCEPTED MANUSCRIPT available and therefore no specific analysis comparing antithrombotic therapy and the mechanism of bleeding was possible. Nevertheless, in a recent

PT

publication investigating the severity of cerebrovascular events after LAAO in

SC RI

the ACP Registry, most of the events were described as non-disabling. (20)

LAAO also showed efficacy in terms of stroke and hemorrhagic prevention at follow-up. Regardless of the less intensive antithrombotic therapy after LAAO,

(75% vs. 55%) and major bleeding (89% vs. 47%) than patients

MA

events

NU

patients with previous ICB showed a higher reduction of cerebrovascular

without ICB, when comparing the observed and expected rates according to

ED

the CHA2DS2-VASc and HASBLED scores, respectively (Figure 1). In addition, no differences in device thrombosis or leaks despite the

PT

aforementioned differences in the antithrombotic treatment were observed among groups. This finding is of special relevance as device thrombosis is

CE

one of the main concerns of physicians when prescribing mild antithrombotic

AC

regimens. Similarly, Saw et al. observed no device thrombosis in patients treated with less intensive antithrombotic therapies after LAAO for ICB. (7) Again, an individualized selection of antithrombotic drugs post LAAO will be the best option after balancing the risk of repeated bleeding and the potential risk of device thrombosis.

The present study has several limitations that should be acknowledged. This is a non-randomized, retrospective, observational study, which included many centers. The major limitation for estimating the overall value of LAAO is the lack of a randomized control group. Another important limitation of our study is

12

ACCEPTED MANUSCRIPT the lack of information on the type of ICB and the type or combination of antithrombotic treatments received by the patient when the ICB that made the

PT

patient eligible for LAAO occurred. These data would have been useful to stratify the risk of bleeding recurrence in our series and to specifically select

SC RI

the most suitable antithrombotic treatment after the procedure. Also, TEE follow-up was not available for all patients. The clinical and TEE results were self-reported and there was no independent adjudication. However, all the

ED

MA

as much homogeneity as possible.

NU

important events were discussed within the study group members to achieve

CONCLUSIONS

PT

Patients with previous ICB as an indication for LAAO had similar procedural outcome compared to those without. Despite the less intensive antithrombotic

CE

therapy post-LAAO, patients with previous ICB had a significant reduction in

AC

stroke/TIA, a remarkably low frequency of major bleeding and no increase in the rate of device thrombosis or peri-device leaks at follow-up.

PERSPECTIVES COMPETENCY IN MEDICAL KNOWLEDGE: Intracranial bleeding (ICB), one of the most serious complications of oral anticoagulation therapy, is associated with a high recurrence rate and devastating clinical consequences. For patients with non-valvular atrial fibrillation needing anticoagulation for

13

ACCEPTED MANUSCRIPT stroke prevention, left atrial appendage occlusion (LAAO) is a valuable

PT

alternative, as both ischemic and bleeding risks seem to be reduced.

COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In patients

SC RI

with previous ICB, LAAO with the Amplatzer Cardiac Plug is associated with similar procedural risks as compared to patients without previous ICB. Moreover, patients with previous ICB have a significant reduction in

MA

NU

stroke/TIA and a remarkably low frequency of major bleeding at follow-up.

TRANSLATIONAL OUTLOOK: Further studies are needed to define whether

ED

LAAO is exceptionally safe and effective for patients with previous ICB,

AC

CE

PT

particularly in the era of new oral anticoagulant agents.

14

ACCEPTED MANUSCRIPT REFERENCES 1.

Tzikas A, Shakir S, Gafoor S, Omran H, Berti S, Santoro G, et al. Left

PT

atrial appendage occlusion for stroke prevention in atrial fibrillation:

EuroIntervention 2016;11:1170-9. 2.

SC RI

multicentre experience with the AMPLATZER Cardiac Plug.

Camm AJ, Lip GY, De Caterina R, Savelieva I, Atar D, Hohnloser SH et al. 2012 focused update of the ESC Guidelines for the management

NU

of atrial fibrillation: an update of the 2010 ESC Guidelines for the

MA

management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J

3.

ED

2012;33:2719-47.

Hayden DT, Hannon N, Callaly E, Ní Chróinín D, Horgan G, Kyne L et

PT

al. Rates and Determinants of 5-Year Outcomes After Atrial FibrillationRelated Stroke: A Population Study. Stroke 2015;46:3488-93. Lerario MP, Gialdini G, Lapidus DM, Shaw MM, Navi BB, Merkler AE et

CE

4.

AC

al. Risk of Ischemic Stroke after Intracranial Hemorrhage in Patients with Atrial Fibrillation. PloS one 2015;10:e0145579.

5.

Freixa X, Martin-Yuste V. Percutaneous Left Atrial Appendage Occlusion. Rev Esp Cardiol 2013;66:919-22.

6.

Lopez Minguez JR, Asensio JM, Gragera JE, Costa M, González IC, de Carlos FG et al. Two-year clinical outcome from the Iberian registry patients after left atrial appendage closure. Heart 2015;101:877-83.

7.

Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial

15

ACCEPTED MANUSCRIPT fibrillation using a novel risk factor-based approach: the euroheart survey on atrial fibrillation. Chest. 2010;137:263-72. Lip GY, Frison L, Halperin JL, Lane DA. Comparative validationof a

PT

8.

novel risk score for predicting bleeding risk in anticoagulated patients

SC RI

with atrial fibrillation: the HAS-BLED (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol Concomitantly) score. J Am Coll Cardiol.

Fahmy P, Spencer R, Tsang M, Gooderham P, Saw J. Left Atrial

MA

9.

NU

2011;57:173-80.

Appendage Closure for Atrial Fibrillation Is Safe and Effective After

2016;32:349-54.

Meerkin D, Butnaru A, Dratva D, Bertrand OF, Tzivoni D. Early safety

PT

10.

ED

Intracranial or Intraocular Hemorrhage. The Can J Cardiol

of the Amplatzer Cardiac Plug for left atrial appendage occlusion. Int J

Gladstone DJ, Bui E, Fang J, Laupacis A, Lindsay MP, Tu JV et al.

AC

11.

CE

Cardiol 2013;168:3920-5.

Potentially preventable strokes in high-risk patients with atrial fibrillation who are not adequately anticoagulated. Stroke 2009;40:235-40.

12.

Arima H, Tzourio C, Butcher K, Anderson C, Bousser MG, Lees KR et al. Prior events predict cerebrovascular and coronary outcomes in the PROGRESS trial. Stroke 2006;37:1497-502.

13.

Hanger HC, Wilkinson TJ, Fayez-Iskander N, Sainsbury R. The risk of recurrent stroke after intracerebral haemorrhage. J Neurol Neurosurg Psychiatry 2007;78:836-40.

16

ACCEPTED MANUSCRIPT 14.

Weimar C, Benemann J, Terborg C, Walter U, Weber R, Diener HC; German Stroke Study Collaboration. Recurrent stroke after lobar and

PT

deep intracerebral hemorrhage: a hospital-based cohort study. Cerebrovasc Dis 2011;32:283-8.

Poon MT, Fonville AF, Al-Shahi Salman R. Long-term prognosis after

SC RI

15.

intracerebral haemorrhage: systematic review and meta-analysis. J Neurol Neurosurg Psychiatry 2014;85:660-7.

Hemphill JC 3rd, Greenberg SM, Anderson CS, Becker K, Bendok BR,

NU

16.

MA

Cushman M et al. Guidelines for the Management of Spontaneous Intracerebral Hemorrhage: A Guideline for Healthcare Professionals

ED

From the American Heart Association/American Stroke Association. Stroke 2015;46:2032-60.

Ambrosi P, Daumas A, Villani P, Giorgi R. Meta-analysis of major

PT

17.

bleeding events on aspirin vs vitamin K antagonists in randomized

18.

Tarlov N, Norbash AM, Nguyen TN. The safety of anticoagulation in

19.

AC

CE

trials. Int J Cardiol 2016. doi: 10.1016/ j.ijcard.2016.12.055

patients with intracranial aneurysms. J Neurointerv Surg 2013;5:405-9. Chari A, Clemente Morgado T, Rigamonti D. Recommencement of anticoagulation in chronic subdural haematoma: a systematic review and meta-analysis. Br J of Neurosurg 2014;28:2-7.

20.

Freixa X, Llull L, Gafoor S, Cruz-Gonzalez I, Shakir S, Omran H et al. Characterization of Cerebrovascular Events After Left Atrial Appendage Occlusion. Am J Cardiol. 2016;118:1836-1841.

17

ACCEPTED MANUSCRIPT FIGURE TITLES AND LEGENDS Figure 1. Reduction in stroke and bleeding

PT

Legend: Annual risk reduction in stroke and bleeding after left atrial appendage closure with the Amplatzer Cardiac Plug. Patients with previous

SC RI

intracranial bleeding (IC Bleeding) are compared with patients without IC

AC

CE

PT

ED

MA

NU

Bleeding.

18

ACCEPTED MANUSCRIPT

ICB

Other

1047

198

849

Baseline characteristics Age, years

74.9 ± 8.4

73.7 ± 7.9

75.2 ± 8.5

0.004

Male

648 (62)

138 (70)

510 (60)

0.012

permanent

594 (57)

121 (61)

473 (56)

NS

paroxysmal/persistent

453 (43)

77 (39)

376 (44)

NS

Congestive heart failure

274 (26)

32 (16)

242 (29)

<0.001

Arterial hypertension

909 (87)

168 (85)

741 (87)

NS

Diabetes mellitus

317 (30)

54 (27)

263 (31)

NS

404 (39)

126 (64)

278 (33)

<0.001

87 (8)

13 (7)

74 (9)

NS

367 (36)

53 (27)

314 (37)

0.015

164 (16)

23 (12)

141 (17)

NS

228 (22)

27 (14)

201 (24)

0.004

56 (5)

12 (6)

44 (5)

NS

2.8 ± 1.3

3.0 ± 1.3

2.7 ± 1.3

0.014

CHA2DS2-VASc score

4.5 ± 1.6

4.5 ± 1.5

4.4 ± 1.6

NS

Predicted annual risk of

5.7 ± 2.8

5.8 ± 2.8

5.6 ± 2.8

NS

HASB-LED score

3.1 ± 1.2

3.5 ± 1.1

3.1 ± 1.2

<0.001

Predicted annual risk of major

5.4 ± 3.8

6.4 ± 3.9

5.1 ± 3.7

<0.001

1001

188

813

16.2 ± 12.0

18.4 ± 12.0

15.7 ± 11.9

1349

287

1062

Number of patients

MA

NU

Atrial fibrillation

Previous stroke/TIA

PCI

PT

Coronary artery disease

AC

CE

Peripheral embolization Risk scores CHADS2 score

ED

Carotid disease

myocardial infarction

PT

Total

SC RI

Table 1. Baseline patient characteristics. P value

thromboembolism, %

bleeding, % Follow-up Number of patients FU, months FU, total years

0.006

19

ACCEPTED MANUSCRIPT Variables are expressed as n ± SD or n (%). INR=international normalized ratio; PCI=percutaneous coronary intervention; TIA=transient ischemic attack. Table 2. Procedural success and major adverse events. 1047

198

1019 (97)

Other

PT

ICB

P value

849

826 (97)

Total

51 (4.9)

SC RI

Number of patients

Total

Death

8 (0.8)

0 (0.0)

8 (0.9)

NS

Stroke

9 (0.9)

1 (0.5)

8 (0.9)

NS

0 (0.0)

0 (0.0)

0 (0.0)

NS

1 (0.1)

0 (0.0)

1 (0.1)

NS

12 (1.2)

1 (0.5)

11 (1.3)

NS

13 (1.3)

1 (0.5)

12 (1.4)

NS

Device embolization requiring surgery

1 (0.1)

0 (0.0)

1 (0.1)

NS

Device embolization snared

7 (0.7)

2 (1.0)

5 (0.6)

NS

Need for surgery*

0 (0.0)

0 (0.0)

0 (0.0)

NS

Procedural success

MA

NU

Major adverse events

Systemic embolism

ED

Myocardial infarction Cardiac tamponade

CE

PT

Major bleeding

193 (98)

NS

5 (2.5)

46 (5.4)

NS

AC

Variables are expressed as n ± SD or n (%). * Apart from device embolization.

20

ACCEPTED MANUSCRIPT Table 3. Antithrombotic medication at baseline and follow-up Baseline Total

ICB

Follow-up

Other

P

Total

ICB

Other

1047

198

849

641

123

518

(61.2)

(62.1)

(61.0)

232

33

199

(22.2)

(16.7)

(23.4)

255

24

231

(24.4)

(12.1)

NOAC

31 (3.0)

4 (2.0)

LMWH

168

1001

Antithrombotic

845

163

682

(84.4)

(86.7)

(83.9)

242

30

212

(24.2)

(16.0)

(26.1)

<0.001

30 (3.0)

3 (1.6)

27 (3.3)

NS

27 (3.2)

NS

14 (1.4)

0 (0.0)

14 (1.7)

NS

40

128

NS

2 (0.2)

0 (0.0)

2 (0.5)

NS

(16.0)

(20.2)

(15.1)

86 (8.2)

24

62 (7.3)

0.031

60 (6.0)

14 (7.6)

46 (5.7)

NS

0 (0.0)

15 (1.8)

NS

19 (1.9)

1 (0.5)

18 (2.2)

NS

324

84

240

<0.001

638

140

498

<0.001

(31.0)

(42.4)

(28.3)

(63.7)

(74.5)

(61.3)

164

22

142

189

20

169

clopidogrel

(15.7)

(11.1)

(16.7)

(18.9)

(10.6)

(20.8)

ASA + VKA

65 (6.2)

9 (4.5)

56 (6.6)

NS

10 (1.0)

3 (1.6)

7 (0.9)

NS

ASA + NOAC

12 (1.1)

1 (0.5)

11 (1.3)

NS

6 (0.6)

0 (0.0)

6 (0.7)

NS

ASA + LMWH

57 (5.4)

6 (3.0)

51 (6.0)

NS

2 (0.2)

0 (0.0)

2 (0.2)

NS

Clopidogrel

39 (3.7)

10 (5.1)

29 (3.4)

NS

51 (5.1)

10 (5.3)

41 (5.0)

NS

Clopidogrel +

5 (0.5)

0 (0.0)

5 (0.6)

NS

0 (0.0)

0 (0.0)

0 (0.0)

NS

1 (0.1)

0 (0.0)

1 (0.1)

NS

0 (0.0)

0 (0.0)

0 (0.0)

NS

No treatment

NS

NU

0.046

MA

VKA

NS

0.004

(27.2)

ED

Clopidogrel

PT

ASA

813

CE

medication

188

SC RI

N of patients

value

PT

value

P

(12.1)

Unknown

details ASA

AC

Therapy

15 (1.4)

ASA +

NS

0.001

VKA Clopidogrel +

21

ACCEPTED MANUSCRIPT NOAC 167

14 (7.1)

(16.0)

153

<0.001

16 (1.6)

0 (0.0)

16 (2.0)

NS

0 (0.0)

10 (1.2)

NS

2 (0.2)

NS

(18.9)

NOAC

15 (1.3)

3 (1.5)

12 (1.4)

NS

10 (1.0)

LMWH

76 (7.3)

24

52 (6.1)

0.005

2 (0.2)

Triple therapy

20 (1.9)

1 (0.5)

19 (2.2)

NS

Unknown

15 (1.4)

0 (0.0)

15 (1.8)

NS

0 (0.0)

SC RI

(12.1)

PT

VKA

2 (0.2)

0 (0.0)

2 (0.2)

NS

19 (1.9)

1 (0.5)

18 (2.2)

NS

NU

Variables are presented as n (%). ASA=acetylsalicylic acid; ICB=intracranial bleeding; LMWH=low molecular weight heparin; NOAC=non-vitamin K dependent oral anticoagulant; VKA=oral anticoagulation with vitamin K antagonist.

AC

CE

PT

ED

MA

Figure 1

22