Triple vs Dual Antithrombotic Therapy in Patients with Atrial Fibrillation and Coronary Artery Disease

Accepted Manuscript Triple Versus Dual Antithrombotic Therapy in Patients With Atrial Fibrillation and Coronary Artery Disease Renato D. Lopes, MD, MHS, PhD, Meena Rao, MD, DaJuanicia N. Simon, MS, Laine Thomas, PhD, Jack Ansell, MD, Gregg C. Fonarow, MD, Bernard J. Gersh, MB, ChB, DPhil, Alan S. Go, MD, Elaine M. Hylek, MD, MPH, Peter Kowey, MD, Jonathan P. Piccini, MD, MHS, Daniel E. Singer, MD, Paul Chang, MD, Eric D. Peterson, MD, MPH, Kenneth W. Mahaffey, MD PII:

S0002-9343(16)30025-0

DOI:

10.1016/j.amjmed.2015.12.026

Reference:

AJM 13329

To appear in:

The American Journal of Medicine

Received Date: 16 November 2015 Revised Date:

14 December 2015

Accepted Date: 15 December 2015

Please cite this article as: Lopes RD, Rao M, Simon DN, Thomas L, Ansell J, Fonarow GC, Gersh BJ, Go AS, Hylek EM, Kowey P, Piccini JP, Singer DE, Chang P, Peterson ED, Mahaffey KW, Triple Versus Dual Antithrombotic Therapy in Patients With Atrial Fibrillation and Coronary Artery Disease, The American Journal of Medicine (2016), doi: 10.1016/j.amjmed.2015.12.026. 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 1

Triple Versus Dual Antithrombotic Therapy in Patients With Atrial Fibrillation and Coronary Artery Disease

RI PT

Renato D. Lopes, MD, MHS, PhD,a Meena Rao, MD,a DaJuanicia N. Simon, MS, a Laine Thomas, PhD,a Jack Ansell, MD,b Gregg C. Fonarow, MD,c Bernard J. Gersh, MB, ChB, DPhil,d Alan S. Go, MD,e Elaine M. Hylek, MD, MPH,f Peter Kowey, MD,g Jonathan P. Piccini, MD, MHS,a Daniel E. Singer, MD,h Paul Chang, MD,i Eric D. Peterson, MD, MPH,a Kenneth W.

a

SC

Mahaffey, MD j

Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC; bHofstra

M AN U

North Shore/LIJ School of Medicine, New York, NY; cUCLA School of Medicine, Los Angeles, CA; dMayo Clinic College of Medicine, Rochester, MN; eKaiser Permanente, Oakland, CA; f

Boston University School of Medicine, Boston, MA; gLankenau Institute for Medical Research,

Wynnewood, PA; hHarvard Medical School and Massachusetts General Hospital, Boston, MA; i

Janssen Pharmaceuticals, Raritan, NJ; jStanford University School of Medicine, Stanford, CA.

TE D

Running title: Triple Versus Dual Antithrombotic Therapy

Address for correspondence: Dr. Renato D. Lopes, Duke Clinical Research Institute, Box

EP

3850, 2400 Pratt Street, Room 0311, Terrace Level, Durham, NC 27705; E-mail: [email protected]; phone: 919-668-8241; fax: 919-668-7056.

AC C

Funding: The ORBIT-AF registry is sponsored by Janssen Scientific Affairs, Raritan, NJ. The

authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper, and its final contents. Conflict of Interest: R.D.L. reports consulting fees and research grants from Bristol-Myers

Squibb, research grants from GlaxoSmithKline, and consulting fees from Boehringer-Ingelheim, Bayer, and Pfizer. J.A. reports consultant/advisory board fees from Bristol Myers Squibb, Pfizer, Janssen Pharmaceuticals, Daiichi Sankyo, Boehringer-Ingelheim, and Alere. G.C.F. reports consultant/advisory board fees from Ortho McNeil. B.J.G. reports DSMB/advisory board fees

ACCEPTED MANUSCRIPT 2

from Medtronic, Baxter Healthcare Corporation, InspireMD, Cardiovascular Research Foundation, PPD Development, LP, Boston Scientific, and St. Jude. E.M.H. reports honoraria support from Boehringer-Ingelheim and Bayer; and consultant/advisory board fees from Johnson & Johnson, Boehringer-Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Pfizer, and Ortho-

RI PT

McNeil-Janssen. P.K. reports consultant/advisory board fees from Boehringer-Ingelheim, Bristol Myers Squibb, Johnson & Johnson, Portola, Merck, Sanofi, and Daiichi Sankyo. J.P.P. reports research grant support from ARCA Biopharma, GE Healthcare, Johnson & Johnson, and

ResMed, and consulting fees from Forest Laboratories, Johnson & Johnson, Medtronic and

SC

Spectranetics. D.E.S. reports research grant support from Johnson & Johnson and Bristol-Myers Squibb, and consultant/advisory board fees from Bayer HealthCare, Boehringer-Ingelheim,

M AN U

Bristol-Myers Squibb, Daiichi Sankyo, Johnson & Johnson, and Pfizer. P.C. reports employment with Janssen Pharmaceuticals. E.D.P. reports research grant support from Eli Lilly & Company, Janssen Pharmaceuticals, and the American Heart Association; consultant/advisory board support from Boehringer-Ingelheim, Bristol-Myers Squibb, Janssen Pharmaceuticals, Pfizer, and Genentech. K.W.M.’s financial disclosures prior to August 1, 2013, can be viewed at https://www.dcri.org/about-us/conflict-of-interest/Mahaffey-COI_2011-2013.pdf; disclosures

TE D

after August 1, 2013, can be viewed at http://med.stanford.edu/profiles/kenneth-mahaffey. None of the other authors report disclosures.

Authorship: All authors had access to the data and played a role in writing this manuscript. Article type: Clinical Research Study

AC C

EP

Word count: 4836 (total); 4 tables, 5 figures

ACCEPTED MANUSCRIPT 3

ABSTRACT Background: The role of triple antithrombotic therapy versus dual antithrombotic therapy in patients with both atrial fibrillation and coronary artery disease remains unclear. This study

RI PT

explores the differences in treatment practices and outcomes between triple antithrombotic therapy and dual antithrombotic therapy in patients with atrial fibrillation and coronary artery disease.

SC

Methods: Using the ORBIT-AF registry (n=10,135), we analyzed outcomes in patients with coronary artery disease (n=1,827) according to treatment with triple antithrombotic therapy

M AN U

(defined as concurrent therapy with an oral anticoagulant, a thienopyridine, and aspirin) or dual antithrombotic therapy (comprising either an oral anticoagulant and 1 antiplatelet agent [OAC plus AA] or 2 antiplatelet drugs and no anticoagulant [DAP]).

Results: The use of triple antithrombotic therapy, OAC plus AA, and DAP at baseline was 8.5%

TE D

(n=155), 80.4% (n=1,468), and 11.2% (n=204), respectively. Among patients treated with OAC plus AA, aspirin was the most common antiplatelet agent used (90%), followed by clopidogrel (10%) and prasugrel (0.1%). The use of triple antithrombotic therapy was not affected by either

EP

patient risk of stroke or bleeding. Patients treated with triple antithrombotic therapy at baseline were hospitalized for all causes (including cardiovascular) more often than patients on OAC plus

AC C

AA (adjusted HR 1.75, 95%CI: 1.35-2.26, P<.0001) or DAP (HR 1.82, 95%CI: 1.25-2.65, P=.0018). Rates of major bleeding or a combined cardiovascular outcome were not significantly different by treatment group. Conclusions: Choice of antithrombotic therapy in patients with atrial fibrillation and coronary artery disease was not affected by patient stroke or bleeding risks. Triple antithrombotic therapytreated patients were more likely to be hospitalized for all causes than those on OAC plus AA or on DAP.

ACCEPTED MANUSCRIPT 4

Keywords: Anticoagulants • Antiplatelets • Combined therapy • Atrial fibrillation • Coronary

AC C

EP

TE D

M AN U

SC

RI PT

artery disease.

ACCEPTED MANUSCRIPT 5

Atrial fibrillation is a common cardiac arrhythmia that increases risk of stroke by 5- to 7-fold.1-3 Use of anticoagulation in those with atrial fibrillation decreases the risk of stroke in patients with nonvalvular atrial fibrillation and is therefore recommended in atrial fibrillation patients with at

RI PT

least 1 risk factor for stroke.4-6 Warfarin has been shown to be superior in reducing embolic events in patients with atrial fibrillation as compared with dual antiplatelet therapy with aspirin and clopidogrel.7 Despite that, antiplatelet therapy continues to be used instead of oral

SC

anticoagulants in patients with atrial fibrillation, particularly in the elderly and other high-risk patients.8-10 Patients with atrial fibrillation and coronary artery disease have a potential indication

M AN U

for antiplatelet therapy for secondary prevention; however, combining anticoagulation and aspirin therapy increases rates of major bleeding.11-14 Given the uncertainty of which therapy to choose and for how long, guidelines recommend balancing the risk between stent thrombosis, recurrent myocardial infarction, stroke, embolism, and bleeding for each patient to determine

TE D

treatment choice and duration.

Using data from the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) we examined use and associated outcomes of dual antiplatelet therapy

EP

in addition to anticoagulation (i.e., triple antithrombotic therapy) and dual antithrombotic therapy

AC C

among patients with both atrial fibrillation and coronary artery disease.

METHODS

ORBIT-AF is an observational, multicenter, prospective registry of outpatients with atrial fibrillation. The rationale and study design of the ORBIT-AF registry have been previously described.15 The patients enrolled in the registry had atrial fibrillation documented by electrocardiography and were over 18 years of age. Enrollees needed to be able to provide

ACCEPTED MANUSCRIPT 6

informed consent and be available for follow-up every 6 months out to a minimum of 2 years. Participants were excluded if they had reversible causes of atrial fibrillation or a life expectancy less than 6 months. The ORBIT-AF registry was approved by the Duke University Institutional

RI PT

Review Board (IRB) and local site IRBs as appropriate per local requirements.

Overall, 10,135 patients were enrolled in the ORBIT-AF registry from 2010 through 2011. From the registry, this analysis was limited to patients who were on triple antithrombotic

SC

therapy or dual antithrombotic therapy and had a history of coronary artery disease (n = 1,827 [18.0%], from 154 different sites). For the purpose of this analysis, coronary artery disease at

M AN U

study entry was defined as a history of coronary artery disease, history of coronary artery bypass grafting, percutaneous coronary intervention, or drug-eluting stent implantation. Triple antithrombotic therapy was defined as concurrent therapy with an oral anticoagulant, a thienopyridine, and aspirin. Dual antithrombotic therapy comprised either an oral anticoagulant

Statistical Analysis

TE D

and 1 antiplatelet agent (OAC plus AA) or 2 antiplatelet drugs and no oral anticoagulant (DAP).

EP

Baseline characteristics were stratified by type of therapy and are presented as percentages for categorical variables and medians (interquartile range) for continuous variables. Risk scores for

AC C

stroke and bleeding risk were calculated using the CHADS2, CHA2DS2-VASc, and ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) scores.16-18 The clinical outcomes of interest were all-cause hospitalization, myocardial infarction, revascularization (percutaneous coronary intervention or coronary artery bypass grafting), stroke/non-central nervous system embolism or transient ischemic attack, and International Society of Thrombosis and Haemostasis major bleeding within 1 year of enrollment in the ORBIT-AF registry. Outcomes were modeled

ACCEPTED MANUSCRIPT 7

using Cox proportional hazards modeling accounting for clustering of patients within sites by adding a random effect for site (multivariable Cox frailty model). The results are presented as hazard ratios and 95% confidence intervals. All P values were 2-sided, and P < .05 was

RI PT

considered statistically significant. Adjustment of outcomes was done based on covariates previously established to be associated with the given outcome and components of the

ATRIA,CHA2DS2 VASc, and HAS-BLED scores19 (see also Supplementary Appendix).

SC

Briefly, backward selection with a stay criteria of P < .05 was used to identify predictors of each outcome. Missing data were handled with multiple imputation, and imputed values were

M AN U

obtained by Markov chain Monte Carlo or regression methods. Standard errors and results reflect the combined analysis over 5 imputed data. For all models, continuous variables were evaluated for non-linearity with the outcome and when non-linear fit with linear splines. Additionally, in patients with 6-month and/or 12-month follow-up, the percentage of patients who persisted on

TE D

triple antithrombotic therapy was measured. The rates of previous acute myocardial infarction and percutaneous coronary intervention were collected in the case report form, which included dates of the most recent events at baseline. All analyses were performed using SAS version 9.3

AC C

RESULTS

EP

(SAS Institute, Cary, NC).

From the overall ORBIT-AF study population, 36% (n=3,645) had coronary artery disease. Of those, 49.9% (n=1,818) were on mono antithrombotic therapy or no therapy. From the final study population of 1,827 patients, 155 patients (8.5%) were on triple antithrombotic therapy, 1,468 (80.4%) received OAC plus AA, and 204 (11.2%) were on DAP (Figure 1). We also noted

ACCEPTED MANUSCRIPT 8

that roughly 20% of patients with atrial fibrillation who did not have coronary artery disease were on triple antithrombotic therapy or dual antithrombotic therapy.

RI PT

Baseline Characteristics

Baseline characteristics are shown in Table 1. Patients who were treated with triple

antithrombotic therapy were younger and had a higher creatinine clearance. DAP was used more

SC

commonly in patients with a prior history of gastrointestinal bleed, and triple antithrombotic therapy was not used in patients with history of hemorrhagic stroke. DAP and triple

OAC plus AA.

Antithrombotic Medications

M AN U

antithrombotic therapy were more often to have had prior myocardial infarction and stents than

TE D

The most frequently occurring triple antithrombotic therapy combination was warfarin, aspirin, and clopidogrel. Among patients on OAC plus AA, aspirin was the most common antiplatelet used (90%), followed by clopidogrel (10%) and prasugrel (0.1%) (Table 2). For DAP, aspirin

EP

was used in 100% of the patients, clopidogrel in 96%, and prasugrel in 3.9%. Of the patients receiving triple antithrombotic therapy, OAC plus AA, or DAP at baseline and with follow-up

AC C

data at 6 months and/or 12 months, 38.5%, 75.8%, and 65.3% remained on these treatments at 1 year, respectively.

Risk Scores and Antithrombotic Therapy Any of the antithrombotic treatments were significantly altered by patients risk for systemic emboli or stroke (as assessed by CHADS2 or CHA2DS2 VASc risk scores) (Figure 2,

ACCEPTED MANUSCRIPT 9

Supplementary Figure 1). Patients at the highest risk of bleeding (ATRIA score ≥5) were the least likely to use triple therapy (Figure 3).

RI PT

Clinical Outcomes

In this population, there were low rates of patients with acute myocardial infarction and

percutaneous coronary intervention in the year prior to enrollment and during study follow-up;

SC

98 patients had had a myocardial infarction in the past year (5.4%). Of the 92 patients who had had a drug-eluting stent placed in the past year, 41% were on triple antithrombotic therapy

M AN U

(n=38), 33% on OAC plus AA (n=28), and 28% on DAP (n=26). Unadjusted rates of myocardial infarction during 1-year follow-up were 1.0 events/100 patient years for patients on OAC plus AA, 2.8 events/100 patient years for DAP, and 4.2 events/100 patient years for triple antithrombotic therapy.

TE D

Primary adjusted outcomes at 1 year compared between patients on triple antithrombotic therapy at baseline and patients on OAC plus AA or DAP before and after adjustment are shown in Figures 4 and 5. Patients treated with triple antithrombotic therapy at baseline were

EP

hospitalized for all causes (including cardiovascular) more often than patients on OAC plus AA (adjusted HR 1.75, 95% CI: 1.35-2.26, P<.0001) or DAP (HR 1.82, 95% CI: 1.25-2.65,

AC C

P=.0018) (Figures 4 and 5; Tables 3 and 4). No differences in stroke, bleeding or death were seen among the 3 different antithrombotic therapy strategies.

DISCUSSION

Patients with atrial fibrillation and coronary artery disease have potential indications for both anticoagulation and antiplatelet therapy. The ideal combination of treatments in these patients

ACCEPTED MANUSCRIPT 10

remains unclear and thus it is important to understand how the treatments are being used and the associated outcomes of these various strategies. Our study looked at patients with atrial fibrillation and coronary artery disease enrolled in the ORBIT-AF registry and examined

RI PT

outcomes between patients on dual antithrombotic therapy (either OAC plus AA or DAP) and triple antithrombotic therapy. Overall, we saw that use of antithrombotic therapy was not

strongly influenced by patients’ risk of stroke; if anything, use of more aggressive therapies

SC

tended to decline with increasing risk. Similarly, there were no significant differences in use of these regimens as a function of bleeding risk. Finally, although our results were observational,

M AN U

we found that adjusted risks of stroke and bleeding were similar by treatment regimen, but hospitalization for all causes (including cardiovascular) was higher in triple antithrombotic therapy-treated patients than in others.

Although anticoagulation is an effective therapy to decrease the risk of stroke in patients

TE D

with nonvalvular atrial fibrillation, there exists a paradox of underuse, as patients with the highest risk of stroke have decreased anticoagulation use.4 Decreasing use of anticoagulation in patients with an increasing CHADS2 score has been described in high-risk atrial fibrillation

EP

patients with acute coronary syndromes20,21 and after percutaneous coronary intervention.21,22 This paradox was also seen in our study population with atrial fibrillation and coronary artery

AC C

disease for both dual antithrombotic therapy and triple antithrombotic therapy. With increasing CHADS2 and CHA2DS2 VASc scores, the rates of anticoagulation with all 3 treatment strategies initially increased, before decreasing with the highest risk patients. We found, however, that the rates of aggressive anticoagulation use decreased among those at high risk for bleeding, and those at highest risk for bleeding (with ATRIA score ≥ 5) had the lowest use of triple antithrombotic therapy (Figure 4). These findings suggest that while the risk scores are a useful

ACCEPTED MANUSCRIPT 11

guide for providers to identify patients at risk for stroke or bleeding, there is more complexity to what providers use in practice to guide clinical decisions, highlighting the need for a better understanding of antithrombotic treatment strategies.23-25

RI PT

Triple antithrombotic therapy has been associated with a 4-fold increase in fatal and nonfatal bleeding without additional survival benefit when compared to anticoagulation with aspirin therapy in patients post-acute myocardial infarction.13,26 In a large Danish registry

SC

including patients with a previous diagnosis of atrial fibrillation who were admitted with acute myocardial infarction, the use of oral anticoagulation with the antiplatelet agent clopidogrel had

M AN U

similar efficacy and safety outcomes when compared to triple antithrombotic therapy.22 In our patient population with low rates of acute myocardial infarction prior to enrollment and during follow-up, there was a trend toward higher risks for bleeding events; however, this did not reach statistical significance. This finding may reflect the low event rates overall and wide confidence

TE D

intervals or some degree of residual confounding in this observational data. The WOEST (What Is the Optimal Antiplatelet & Anticoagulant Therapy in Patients With Oral Anticoagulation and Coronary Stenting) trial studied dual antithrombotic therapy with

EP

clopidogrel and oral anticoagulation compared to triple antithrombotic therapy in patients who underwent percutaneous coronary intervention and found that the patients receiving dual

AC C

antithrombotic therapy had similar outcomes with less major bleeding compared with the patients receiving triple antithrombotic therapy.27 Although there are several limitations to this study (e.g., open label with just 573 patients), it suggests that the bleeding risk associated with aspirin therapy may be underestimated.28 In fact, low-dose aspirin therapy was shown to have similar risk of major bleeding to apixaban in patients with atrial fibrillation and CHADS2 score ≥1 without the benefit in stroke prevention.29

ACCEPTED MANUSCRIPT 12

The most common anticoagulant used in our patient population was warfarin, and the most common antiplatelet agent used together with aspirin was clopidogrel. In the era of novel oral anticoagulants and more potent antiplatelet agents, a “less is more” approach may provide

RI PT

better net clinical benefit. However, data on the outcomes and guidance on management of these agents used in combination in this high-risk population are limited.5 The choice of antiplatelet agent also adds complexity to the antithrombotic therapy strategy. In our study, the majority of

SC

triple antithrombotic therapy-treated patients used warfarin, clopidogrel, and aspirin, as

suggested by current guidelines. Sarafoff et al. showed that among 377 patients who underwent

M AN U

drug-eluting stent implantation and had an indication for oral anticoagulation, prasugrel was used in 5.6% of the time instead of clopidogrel, and TIMI major and minor bleeding occurred significantly more often in the prasugrel group compared with the clopidogrel group, with no significant difference regarding the combined ischemic secondary endpoint.30 In our study,

TE D

prasugrel was used in 3.9% of the patients on triple antithrombotic therapy. Thus, caution should be taken with more potent antiplatelet agents such as prasugrel or ticagrelor for patients who need triple antithrombotic therapy, and more data on different antithrombotic therapy

EP

combinations as well as different durations are needed in patients with atrial fibrillation and coronary artery disease. Studies evaluating the efficacy and safety of triple therapy with non-

AC C

vitamin K antagonist oral anticoagulants for patients with atrial fibrillation and acute coronary syndromes or percutaneous coronary intervention are ongoing and will improve the knowledge in this important field.

Study Limitations

ACCEPTED MANUSCRIPT 13

This study has several limitations. It describes observational data from a prospective national registry, and the treatment groups were not randomly assigned. There were also low rates of previous and incident ischemic events (stroke or myocardial infarction) and bleeding events,

RI PT

which limit the ability to detect meaningful differences in these outcomes. There was little

information about the use of novel oral anticoagulants in patients with both coronary artery

disease and atrial fibrillation since these agents were just starting to be used in clinical practice

SC

when our study was performed. Thus, temporal bias needs to be recognized. Finally, use of

clopidogrel in the OAC plus AA group was low; therefore, a comparison of anticoagulation with

M AN U

clopidogrel versus aspirin therapy could not be made.

Conclusion

Choice of antithrombotic therapy in patients with atrial fibrillation and coronary artery disease

TE D

was not affected by patient stroke or bleeding risks. Patients treated with triple antithrombotic therapy were more likely to be hospitalized for all causes (including cardiovascular) than those on OAC plus AA or on DAP. Since information about non-vitamin K antagonist oral

EP

anticoagulants in this setting is limited, it will be important to look at the additional benefit of the

AC C

newer anticoagulants as well as the more powerful antiplatelet inhibitors.

ACCEPTED MANUSCRIPT 14

References 1.

Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: A major contributor to stroke in the elderly: The Framingham Study. Arch Intern Med 1987;147:1561–1564. Wolf PA, Dawber TR, Thomas HE Jr, et al. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke. Neurology 1978;28:973–977.

3.

Flegel K, Shipley M, Rose G. Risk of stroke in non-rheumatic atrial fibrillation. Lancet

SC

1987;329:526–529. 4.

RI PT

2.

Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke

5.

M AN U

in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857–867. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart

6.

TE D

Rhythm Society. J Am Coll Cardiol 2014;64:e1–e76.

Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the

7.

EP

management of atrial fibrillation. Eur Heart J 2012;33:2719–2747. ACTIVE Writing Group of the ACTIVE Investigators, Connolly S, Pogue J, et al.

AC C

Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 2006;367:1903–1912. 8.

van Walraven C, Hart RG, Connolly S, et al. Effect of age on stroke prevention therapy in patients with atrial fibrillation: the atrial fibrillation investigators. Stroke 2009;40:1410– 1416.

ACCEPTED MANUSCRIPT 15

9.

Mant J, Hobbs FD, Fletcher K, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet

10.

RI PT

2007;370:493–503.

Lee S, Shafe AC, Cowie MR. UK stroke incidence, mortality and cardiovascular risk

management 1999-2008: time-trend analysis from the General Practice Research Database.

11.

SC

BMJ Open 2011;1:e000269.

Steinberg BA, Kim S, Piccini JP, et al. Use and associated risks of concomitant aspirin

M AN U

therapy with oral anticoagulation in patients with atrial fibrillation: insights from the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) Registry. Circulation 2013;128:721–728. 12.

Alexander JH, Lopes RD, Thomas L, et al. Apixaban vs. warfarin with concomitant aspirin

2014;35:224–232. 13.

TE D

in patients with atrial fibrillation: insights from the ARISTOTLE trial. Eur Heart J

Hansen ML, Sørensen R, Clausen MT, et al. Risk of bleeding with single, dual, or triple

EP

therapy with warfarin, aspirin, and clopidogrel in patients with atrial fibrillation. Arch Intern Med 2010;170:1433–1441. Dans AL, Connolly SJ, Wallentin L, et al. Concomitant use of antiplatelet therapy with

AC C

14.

dabigatran or warfarin in the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY) trial. Circulation 2013;127:634–640. 15.

Piccini JP, Fraulo ES, Ansell JE, et al. Outcomes registry for better informed treatment of atrial fibrillation: rationale and design of ORBIT-AF. Am Heart J 2011;162:606–612.

ACCEPTED MANUSCRIPT 16

16.

Fang MC, Go AS, Chang Y, et al. A new risk scheme to predict warfarin-associated hemorrhage: The ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) Study. J Am Coll Cardiol 2011;58:395–401. Lip GY, Nieuwlaat R, Pisters R, et al. Refining clinical risk stratification for predicting

RI PT

17.

stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest 2010;137:263–272.

Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes

SC

18.

for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA

19.

M AN U

2001;285:2864–2870.

Steinberg BA, Kim S, Piccini JP, et al. Use and associated risks of concomitant aspirin therapy with oral anticoagulation in patients with atrial fibrillation: insights from the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF)

20.

TE D

Registry. Circulation 2013;128:721–728.

Lopes RD, Starr A, Pieper CF, et al. Warfarin use and outcomes in patients with atrial fibrillation complicating acute coronary syndromes. Am J Med 2010;123:134–140. Lopes RD, Elliott LE, White HD, et al. Antithrombotic therapy and outcomes of patients

EP

21.

with atrial fibrillation following primary percutaneous coronary intervention: results from

22.

AC C

the APEX-AMI trial. Eur Hear J 2009;30:2019–2028. Lamberts M, Gislason GH, Olesen JB, et al. Oral anticoagulation and antiplatelets in atrial fibrillation patients after myocardial infarction and coronary intervention. J Am Coll Cardiol 2013;62:981–989.

ACCEPTED MANUSCRIPT 17

23.

Lopes RD, Al-Khatib SM, Wallentin L, et al. Efficacy and safety of apixaban compared with warfarin according to patient risk of stroke and of bleeding in atrial fibrillation: a secondary analysis of a randomized controlled trial. Lancet 2012;380:1749–1758. Hess CN, Broderick S, Piccini JP, et al. Antithrombotic therapy for atrial fibrillation and

RI PT

24.

coronary artery disease in older patients. Am Heart J 2012;164:607–615. 25.

Lopes RD, Li L, Granger CB, et al. Atrial fibrillation and acute myocardial infarction:

26.

SC

antithrombotic therapy and outcomes. Am J Med 2012;125:897–905.

Fosbol EL, Wang TY, Li S, et al. Warfarin use among older atrial fibrillation patients with

M AN U

non-ST-segment elevation myocardial infarction managed with coronary stenting and dual antiplatelet therapy. Am Heart J 2013;166:864–870. 27.

Dewilde WJ, Oirbans T, Verheugt FW, et al. Use of clopidogrel with or without aspirin in patients taking oral anticoagulant therapy and undergoing percutaneous coronary

28.

TE D

intervention: an open-label, randomised, controlled trial. Lancet 2013;381:1107–1115. Freedman SB, Gersh BJ, Lip GYH. Misperceptions of aspirin efficacy and safety may perpetuate anticoagulant underutilization in atrial fibrillation. Eur Heart J 2015;36:653–

29.

EP

656.

Connolly SJ, Eikelbloom J, Joyner C, et al. Apixaban in patients with atrial fibrillation. N

30.

AC C

Engl J Med 2011;364:806–817. Sarafoff N, Martischnig A, Wealer J, et al. Triple therapy with aspirin, prasugrel, and vitamin K antagonists in patients with drug-eluting stent implantation and an indication for oral anticoagulation. J Am Coll Cardiol 2013;61:2060–2066.

ACCEPTED MANUSCRIPT 18

Figure Legends Figure 1. Flow diagram of study population. CAD = coronary artery disease; OAC + AA = dual

antithrombotic therapy with 2 antiplatelet agents.

RI PT

antithrombotic therapy with an oral anticoagulant plus 1 antiplatelet agent; DAP = dual

Figure 2. Antithrombotic strategy by CHA2DS2 VASc score. OAC + AA = dual antithrombotic therapy with an oral anticoagulant plus 1 antiplatelet agent; DAP = dual antithrombotic therapy

SC

with 2 antiplatelet agents; TT = triple antithrombotic therapy (oral anticoagulation plus dual antiplatelet therapy).

M AN U

Figure 3. Antithrombotic strategy by ATRIA bleeding score. ATRIA = Anticoagulation and Risk Factors in Atrial Fibrillation; OAC + AA = dual antithrombotic therapy with an oral anticoagulant plus 1 antiplatelet agent; DAP = dual antithrombotic therapy with 2 antiplatelet agents; TT = triple antithrombotic therapy (oral anticoagulation plus dual antiplatelet therapy).

TE D

Figure 4. 1-year outcomes. CNS = central nervous system; OAC + AA = dual antithrombotic therapy with an oral anticoagulant plus 1 antiplatelet agent; DAP = dual antithrombotic therapy with 2 antiplatelet agents; MI = myocardial infarction; TIA = transient ischemic attack; TT =

EP

triple antithrombotic therapy (oral anticoagulation plus dual antiplatelet therapy). Figure 5. Cumulative incidence plot showing all-cause hospitalization at 1 year. DAP = dual

AC C

antithrombotic therapy with 2 antiplatelet agents; OAC + AA = dual antithrombotic therapy with an oral anticoagulant plus 1 antiplatelet agent; TT = triple antithrombotic therapy (oral anticoagulation plus dual antiplatelet therapy).

ACCEPTED MANUSCRIPT 1

Table 1. Baseline characteristics stratified by antithrombotic therapy DAP

TT

(n=204)

(n=155)

Age, y

76 (70–81)

79 (70–84)

73 (67– 79)

Female

27%

36%

93% 3% 1.2% 2% 0.30%

90% 6% 1.50% 2% 0%

Hypertension

90%

87%

Hyperlipidemia

89%

89%

90%

Hemodialysis

2%

2%

2%

Diabetes

39%

34%

43%

Mechanical valve

4%

1%

1%

Heart failure Myocardial infarction in past year

49%

46%

46%

3%

12%

14%

Drug eluting stent in past year

2%

13%

25%

Previous gastrointestinal bleed

9%

19%

8%

Previous cerebrovascular event

21%

19%

21%

Non-hemorrhagic stroke

11%

8%

9%

Hemorrhagic stroke

1%

2%

0%

67 (49-90)

58 (42-81)

74 (56-92)

EP

AC C

Creatinine clearance

27%

94% 3% 0% 2.60% 0.70%

SC

M AN U

TE D

Race White Black Other* Hispanic ethnicity Missing

RI PT

OAC plus AA (n=1,468)

Characteristic

92%

Values are presented as % or median (interquartile range). OAC plus AA = dual antithrombotic therapy with an oral anticoagulant plus an antiplatelet agent; DAP = dual antithrombotic therapy with 2 antiplatelet agents; TT = triple antithrombotic therapy (oral anticoagulant plus dual antiplatelet therapy). *Other – Asian, American Indian/Alaska Native, or Native Hawaiian/Pacific Islander

ACCEPTED MANUSCRIPT 2

Table 2. Antithrombotic medication use DAP (n=196)

TT (n=152)

Aspirin

90%

100%

100%

Clopidogrel

10%

96%

99%

Prasugrel

0.1%

3.9%

Ticagrelor

0

0

Warfarin

94%

0

Dabigatran

6%

0

SC

Antithrombotic Medication

RI PT

OAC plus AA (n=1,347)

1.3% 0

99%

0.7%

AC C

EP

TE D

M AN U

OAC plus AA = dual antithrombotic therapy with an oral anticoagulant plus an antiplatelet agent; DAP = dual antithrombotic therapy with 2 antiplatelet agents; TT = triple antithrombotic therapy (oral anticoagulant plus dual antiplatelet therapy).

ACCEPTED MANUSCRIPT

3

Outcome

Triple therapy: no. of events (no. of events OAC plus AA: no. of per 100 patient- events (no. of events years) per 100 patient-years) (n=149) (n=1431)

RI PT

Table 3. Cox regression model – association of triple therapy use at baseline vs. OAC plus AA and 1-year outcomes (n=1,580)*

Unadjusted HR (95% CI)

Adjusted HR (95% CI)

Adjusted p value

6 (4.14)

74 (5.41)

0.77 (0.33-1.78)

0.77 (0.31-1.92)

.5747

Cardiovascular death

4 (2.76)

37 (2.71)

1.03 (0.36-2.94)

1.01 (0.33-3.11)

.9921

Stroke/TIA/MI/revascularization

12 (8.67)

68 (5.08)

1.72 (0.93-3.19)

1.68 (0.88-3.18)

.1139

1st all-cause hospitalization

79 (79.29)

525 (47.84)

1.62 (1.26-2.07)

1.75 (1.35-2.26)

<.0001

1st bleeding-cause hospitalization 11 (7.89)

70 (5.24)

1.56 (0.82-2.97)

1.90 (0.97-3.71)

.0603

1st cardiovascular hospitalization

55 (48.64)

316 (26.23)

1.78 (1.32-2.39)

1.74 (1.28-2.38)

.0004

1st major bleed

8 (5.68)

78 (5.85)

1.03 (0.49-2.16)

1.19 (0.55-2.53)

.6602

M AN U

SC

All-cause death

AC C

EP

TE D

OAC plus AA = dual antithrombotic therapy with an oral anticoagulant plus an antiplatelet agent; HR = hazard ratio; TIA = transient ischemic attack; MI = myocardial infarction. *The number of observations for all the outcomes was (N=1,580) except cardiovascular death (N=1,572).

ACCEPTED MANUSCRIPT

4

Table 4. Cox regression model – association of triple therapy use at baseline vs. dual antithrombotic therapy with 2

RI PT

antiplatelet agents and 1-year outcomes (n=347)*

6 (4.14)

17 (9.34)

Cardiovascular death

4 (2.76)

10 (5.54)

Stroke/TIA/MI/revascularization

12 (8.67)

12 (6.74)

1st all-cause hospitalization

79 (79.29)

81 (56.41) 7 (3.91)

1st cardiovascular hospitalization

55 (48.64)

44 (27.35)

1st major bleed

8 (5.68)

0.42 (0.16-1.07)

TE D

1st bleeding-cause hospitalization 11 (7.89)

SC

All-cause death

9 (5.06)

Adjusted HR (95% CI)

.2727

0.45 (0.14-1.48)

0.21 (0.03-1.70)

.1440

1.29 (0.58-2.86)

1.74 (0.65-4.64)

.2713

1.42 (1.03-1.97)

1.82 (1.25-2.65)

.0018

2.04 (0.78-5.34)

2.24 (0.62-8.19)

.2208

1.77 (1.19-2.63)

2.29 (1.40-3.73)

.0009

1.13 (0.43-2.97)

1.55 (0.44-5.47)

.4974

EP

DAP = dual antithrombotic therapy with 2 antiplatelet agents; HR = hazard ratio; TIA = transient ischemic attack; MI = myocardial infarction. *The number of observations for all the outcomes was (N=347) except cardiovascular death (N=344).

AC C

Adjusted p value

0.34 (0.05-2.36)

M AN U

Outcome

Triple therapy: no. of events (no. of events DAP: no. of events (no. per 100 patientof events per 100 years) patient-years) Unadjusted HR (95% (n=149) (n=198) CI)

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Research Highlights •

The use of triple antithrombotic therapy in patients with atrial fibrillation and coronary artery disease was 8.5%. Choice of triple or dual therapy in patients was not affected by patient stroke or bleeding risks.

•

Patients treated with triple therapy were more likely to be hospitalized than those on dual therapy.

AC C

EP

TE D

M AN U

SC

RI PT

•

ACCEPTED MANUSCRIPT

SUPPLEMENTARY APPENDIX Variables in the adjusted models

RI PT

All-cause death adjusted for: Level of education, rhythm control, cognitive impairment/dementia, hyperlipidemia, eGFR ≤ 80, EGFR > 80 LAD type, cancer, diastolic blood pressure truncated at 70, intraventricular conduction, frailty, HEIGHT, heart rate, hematocrit, diabetes, smoking, systolic blood pressure ≤ 120, COPD, BMI truncated at 30, sex, CHF, age, functional status

SC

Cardiovascular death adjusted for: Diabetes, anemia, diastolic blood pressure, systolic blood pressure ≤ 125, hyperlipidemia, AV Node/HIS bundle ablation, COPD, significant valvular disease, height, sex, LVEF type, functional status, eGFR, CHF, age

M AN U

Stroke, non-CNS embolism, TIA, MI or revascularization adjusted for: Diabetes, prior MI, prior cardioversions, history of CAD, peripheral vascular disease, prior antiarrhythmic drug use, EHRA score, insurance status, prior PCI, age, eGFR, LVEF type, dialysis, AF type, AV Node/HIS bundle ablation, weight, hypertension, age, history of stroke/TIA 1st hospitalization (all-cause) adjusted for: Age ≤ 70, age > 70, BMI, weight, osteoporosis, height, PCI, cancer, OSA, anemia, frailty, insurance status, history of CAD, PI/Site specialty, prior antiarrhythmic drug use, peripheral vascular disease, functional status, heart rate > 68, diabetes, hematocrit, eGFR ≤ 80, COPD, diastolic blood pressure truncated at 70, EHRA score, CHF

EP

TE D

1st bleeding hospitalization adjusted for: BMI, history of CAD, prior valve replacement/repair, COPD, history of PCI, rhythm control, peripheral vascular disease, functional status, eGFR, LVEF type, diastolic blood pressure, AF type, diabetes, hematocrit, prior antiarrhythmic drug use, LAD type, age ≤ 75, age > 75, heart rate > 70, PI/Site specialty, EHRA score, CHF

AC C

1st cardiovascular hospitalization adjusted for: BMI, history of CAD, prior valve replacement/repair, COPD, history of PCI, rhythm control, peripheral vascular disease, functional status, eGFR, LVEF type, diastolic blood pressure, AF type, diabetes, hematocrit, prior antiarrhythmic drug use, LAD type, age ≤ 75, age > 75, heart rate > 70, PI/Site specialty, EHRA score, CHF Major bleeding adjusted for: COPD, OSA, LAD type, cancer, functional status, level of education, intraventricular conduction, rhythm control, smoking status, significant valvular disease, eGFR, insurance status, history of GI bleed, anemia, hematocrit

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Supplementary Figure 1. Antithrombotic strategy by CHADS2 score. OAC + AA = dual antithrombotic therapy with an oral anticoagulant plus 1 antiplatelet agent; DAP = dual antithrombotic therapy with 2 antiplatelet agents; TT = triple antithrombotic therapy (oral anticoagulation plus dual antiplatelet therapy).