JACC: CARDIOVASCULAR INTERVENTIONS
VOL. 12, NO. 11, 2019
ª 2019 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION PUBLISHED BY ELSEVIER
Propensity-Matched Comparison of Oral Anticoagulation Versus Antiplatelet Therapy After Left Atrial Appendage Closure With WATCHMAN Lars Søndergaard, MD, DMSC,a Yam-Hong Wong, MBBS,a,b Vivek Y. Reddy, MD,c Lucas V.A. Boersma, MD, PHD,d Martin W. Bergmann, MD, PHD,e Shephal Doshi, MD,f Saibal Kar, MD,g Horst Sievert, MD,h Scott Wehrenberg, MS,i Kenneth Stein, MD,i David R. Holmes, JR, MDj
ABSTRACT OBJECTIVES In this propensity-matched analysis of post–left atrial appendage closure antithrombotic therapy, the safety and effectiveness of oral anticoagulation (OAC) and antiplatelet therapy (APT) were compared. BACKGROUND Left atrial appendage closure with the WATCHMAN device is an alternative to OAC in patients with nonvalvular atrial fibrillation, who are at high bleeding risk. Initial trials included 45 days of post-implantation OAC, but registry data suggest that APT may suffice. METHODS Patients from the PROTECT-AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients With Atrial Fibrillation), PREVAIL (Watchman LAA Closure Device in Patients With Atrial Fibrillation Versus Long Term Warfarin Therapy), CAP (Continued Access to PROTECT-AF), CAP2 (Continued Access to PREVAIL), ASAP (ASA Plavix Feasibility Study With Watchman Left Atrial Appendage Closure Technology), and EWOLUTION (Registry on WATCHMAN Outcomes in Real-Life Utilization) trials receiving either OAC or APT post-implantation were matched and compared for nonprocedural bleeding and stroke or systemic thromboembolism over 6 months following implantation. Each patient on APT was matched with 2 patients on OAC, with propensity scores derived from age, sex, congestive heart failure, hypertension, diabetes, prior transient ischemic attack or stroke, peripheral vascular disease, left ventricular ejection fraction, renal impairment, and different atrial fibrillation subtypes. RESULTS The cohort on OAC (n ¼ 1,018; 95% receiving warfarin and 5% receiving nonwarfarin OAC) was prescribed 45-day OAC post-implantation (92% also received single APT), followed by 6-month single or dual APT. The cohort on APT (n ¼ 509; 91% receiving dual APT and 9% receiving single APT) received APT for variable durations. Six-month freedom from nonprocedural major bleeding was similar (OAC, 95.7%; APT, 95.5%; p ¼ 0.775) despite more early bleeds with OAC. Freedom from thromboembolism beyond 7 days was similar between groups (OAC, 98.8%; APT, 99.4%; p ¼ 0.089). However, device-related thrombosis was more frequent with APT (OAC, 1.4%; APT, 3.1%; p ¼ 0.018). CONCLUSIONS After left atrial appendage closure with the WATCHMAN, although device-related thrombosis was more common with APT, both APT and OAC strategies resulted in similar safety and efficacy endpoints. (J Am Coll Cardiol Intv 2019;12:1055–63) © 2019 by the American College of Cardiology Foundation.
From the aRigshospitalet, University of Copenhagen, Copenhagen, Denmark; bMedicine and Geriatric Department, Tuen Mun Hospital, Hong Kong, China; cIcahn School of Medicine at Mount Sinai, New York, New York; dSt. Antonius Hospital, Nieuwegein, the Netherlands; eCardiologicum Hamburg, Hamburg, Germany; fProvidence St. John’s Health Center, Santa Monica, California; g
Cedars-Sinai Medical Center, Los Angeles, California; hCardioVascular Center Frankfurt, Frankfurt am Main, Germany; iBoston
Scientific, Minneapolis, Minnesota; and the jMayo Clinic, Rochester, Minnesota. Dr. Søndergaard has received consulting fees and institutional research grants from Boston Scientific. Dr. Reddy has received consulting fees and research grants from Boston Scientific, Coherex, SentreHEART, and St. Jude Medical; has conflicts with other companies not related to this manuscript; and his wife is employed by Boston Scientific but has no connection with Mount Sinai Hospital and receives no direct financial benefit from any LAAC procedures performed at Mount Sinai. Dr. Boersma has received personal fees from Boston Scientific and Medtronic outside the submitted work. Dr. Bergmann has received personal fees from Boston Scientific, St. Jude Medical, Biosense
ISSN 1936-8798/$36.00
https://doi.org/10.1016/j.jcin.2019.04.004
1056
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
P
atients with nonvalvular atrial fibril-
recommended to continue OAC in case of incom-
lation are at increased risk for throm-
plete
boembolic complications, and oral
Because LAAC in other countries is used mainly in
anticoagulation (OAC) is recommended if
patients with strong contraindications to OAC, the
this risk is considered high as usually
most common post-procedural therapy is APT rather
assessed by the CHA2DS2-VASc score. How-
than OAC. Although OAC may be a better therapy to
ever, OAC by its very nature confers an
protect against device-related thrombosis (DRT)
ongoing bleeding risk, and patients and
during endothelialization in the initial phase after
closure
physicians need to carefully weigh these
LAAC, it may associate with a higher bleeding rate
OAC = oral anticoagulation
risks against the benefits in deciding whom
than APT. However, the relative merits of these 2
SAPT = single APT
to
Because
alternative treatment strategies remain uncertain.
thrombus formation in most cases is located
Accordingly, the aim of the present study was to
in the left atrial appendage (1), an alterna-
assess the safety and effectiveness of limited early
tive treatment is to exclude the appendage
OAC (45 days) versus APT post-LAAC with the
from the blood circulation. One option is left atrial
WATCHMAN device using propensity score match-
appendage closure (LAAC) by implantation of a de-
ing of patients from the available prospective trials
vice in the proximal part of the appendage. The 2
and registries.
ABBREVIATIONS AND ACRONYMS APT = antiplatelet therapy DAPT = dual-antiplatelet therapy
DRT = device-related thrombosis
LAAC = left atrial appendage
TEE = transesophageal echocardiography
treat
with
long-term
OAC.
closure
with
a
peridevice
leak
>5
mm.
landmark trials, PROTECT-AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients
With
Atrial
Fibrillation)
and
METHODS
PREVAIL
(Watchman LAA Closure Device in Patients With
The study included all patients who underwent LAAC
Atrial Fibrillation Versus Long Term Warfarin Ther-
with the WATCHMAN device in the randomized
apy), compared this concept using the WATCHMAN
PROTECT-AF and PREVAIL trials (2–5) as well as the
device (Boston Scientific, Marlborough, Massachu-
CAP (Continued Access to PROTECT-AF), CAP2
setts) against warfarin in patients who were other-
(Continued Access to PREVAIL), ASAP (ASA Plavix
wise candidates for long-term OAC (2–5). These
Feasibility
trials have led to U.S. Food and Drug Administration
Appendage Closure Technology), and EWOLUTION
approval of the WATCHMAN in this patient cohort.
(Registry on WATCHMAN Outcomes in Real-Life Uti-
In contrast to the United States, in all other coun-
lization) registries (6–8). Table 1 summarizes the trials
tries, LAAC is reserved mainly for patients who are
and registries.
Study
With
Watchman
Left
Atrial
The immediate post-procedural antithrombotic
deemed unsuitable for OAC (6).
therapy was OAC with warfarin (on top of aspirin) for
SEE PAGE 1064
at least 45 days in PROTECT-AF, PREVAIL, CAP, and was
CAP2, followed by dual APT (DAPT) after documen-
continued for at least 45 days after LAAC, followed
tation of satisfactory closure (peridevice leak <5 mm)
by transesophageal echocardiography (TEE). OAC
of the LAA. In ASAP, 6 months of DAPT was followed
was replaced by anti-platelet therapy (APT) after
by lifelong single APT (SAPT). In EWOLUTION, DAPT
echocardiographic
effective
was used in most patients (60%), while other patients
closure of the appendage, whereas patients were
received OAC, direct OAC, or SAPT for variable
In
PROTECT-AF
and
PREVAIL,
documentation
OAC
of
Webster, and Johnson & Johnson outside the submitted work. Dr. Doshi has received consulting fees and research grants from Boston Scientific, St. Jude Medical, Coherex, and SentreHEART; and is the national principal investigator of CAP2 (Continued Access to PREVAIL). Dr. Kar has received research grants from Boston Scientific, St. Jude Medical, and Abbott Vascular; is a member of the advisory board for LAAC; is the national principal investigator of CAP (Continued Access to PROTECT-AF) and CAP2; and has served as a proctor for Boston Scientific. Dr. Kar also has received honorarium from Abbott Vascular, Boston Scientific, Medtronic, and W.L. Gore. Dr. Seivert has received personal fees from Abbott, Aptus, Atrium, Biosense Webster, Boston Scientific, Carag, Cardiac Dimensions, CardioKinetix, CardioMEMS, Cardiox, Celonova, CGuard, Coherex, Comed, Contego, Covidien, CSI, CVRx, ev3, FlowCardia, Gardia, Gore, GTIMD Medical, Guided Delivery Systems, Hemoteq, InSeal Medical, InspireMD, Kona Medical, Lumen Biomedical, Lifetech, Lutonix, Maya Medical, Medtronic, Occlutech, pfm Medical, Recor, Trireme, Trivascular, Valtech, Vascular Dynamics, Venus Medical, Veryan, and Vessix outside the submitted work; and holds stock options in Cardiokinetix, Access Closure, Coherex, and SMT outside the submitted work. Mr. Wehrenberg is an employee and a shareholder at Boston Scientific. Dr. Stein is an employee and a shareholder at Boston Scientific. Dr. Holmes, Jr. and the Mayo Clinic have financial interests in technology related to this research. That technology has been licensed to Boston Scientific. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Manuscript received December 21, 2018; revised manuscript received March 6, 2019, accepted April 3, 2019.
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
T A B L E 1 Trials and Registries Included in the Present Study
Study
Purpose
Sample Size (Number of Patients Included in Analysis)*
Design
Antithrombotic Therapy Immediately Post-LAAC
PROTECT-AF Determine noninferiority of WATCHMAN to warfarin in patients with NVAF
Multicenter, prospective RCT
463 (408)
45 days of warfarin plus aspirin, then DAPT up to 6 months, then aspirin for life
PREVAIL
Assess safety and efficacy of WATCHMAN compared with warfarin, even with new operators
Multicenter, prospective RCT
269 (252)
45 days of warfarin plus aspirin, then DAPT up to 6 months, then aspirin for life
CAP
Assess long-term safety and effectiveness following the PROTECT-AF trial
Multicenter, prospective registry
566 (534)
See PROTECT-AF trial
CAP2
Assess long-term safety and effectiveness following the PREVAIL trial
Multicenter, prospective registry
578 (545)
See PREVAIL trial
ASAP
Assess safety and efficacy of WATCHMAN in patients with NVAF contraindicated for warfarin
Multicenter, prospective registry
150 (142)
6 months of clopidogrel or ticlopidine plus lifelong aspirin
Multicenter, prospective registry
1,025 (931)
EWOLUTION Assess real-world safety and efficacy of WATCHMAN
Warfarin 16%, DOAC 11%, DAPT 60%, SAPT 7%, no antithrombotic therapy 6%
*Sample size includes the total number in each study, enrolled or randomized to WATCHMAN, while patients who did not undergo successful WATCHMAN implantation were excluded from the present analysis. ASAP ¼ ASA Plavix Feasibility Study With Watchman Left Atrial Appendage Closure Technology; CAP ¼ Continued Access to PROTECT-AF; CAP2 ¼ Continued Access to PREVAIL; DAPT ¼ dual-antiplatelet therapy; DOAC ¼ direct oral anticoagulation; EWOLUTION ¼ Registry on WATCHMAN Outcomes in Real-Life Utilization; LAAC ¼ left atrial appendage closure; NVAF ¼ Nonvalvular atrial fibrillation; PREVAIL ¼ Watchman LAA Closure Device in Patients With Atrial Fibrillation Versus Long Term Warfarin Therapy; PROTECT-AF ¼ Watchman Left Atrial Appendage System for Embolic Protection in Patients With Atrial Fibrillation; RCT ¼ randomized controlled trial; SAPT ¼ singleantiplatelet therapy.
durations at the implanting physician’s discretion.
F I G U R E 1 Antithrombotic Regimens Before Matching
Patients in PROTECT-AF, PREVAIL, CAP, CAP2, and ASAP studies were followed up with TEE at 45 days, while in EWOLUTION, TEE was performed per local practice. Patients in these studies were divided into OAC and APT groups according to the choice of immediate post-procedural antithrombotic regimen (Figure 1). Baseline characteristics for the OAC and APT groups were compared using standardized differences instead of p values (9). Next, patients were propensity-matched on the following baseline characteristics: age, sex, presence of congestive heart failure, hypertension, diabetes, history of transient ischemic attack or stroke, peripheral vascular disease, different ranges of left ventricular ejection fraction (<30%, 30% to 50%, or >50%), renal impairment, and different types of AF (persistent, paroxysmal, or permanent). Propensity scores were obtained for all subjects using a logistic regression
model
of
post-implantation
medical
therapy (OAC vs. APT) on the aforementioned baseline characteristics. Missing data were imputed using bootstrap samples from the subjects within the same group with known data. OAC subjects were matched
Choice of immediate post-procedural antithrombotic regimen. DAPT ¼ dual-
2:1 to APT subjects using a greedy nearest neighbor
antiplatelet therapy; DOAC ¼ direct oral anticoagulation.
algorithm. Baseline risk factors were considered
1057
1058
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
T A B L E 2 Baseline Demographics Before Propensity Score Matching
Age at enrollment (yrs) Female CHF Hypertension
Overall (N ¼ 2,812)
APT (n ¼ 808)
OAC (n ¼ 2,004)
73.5 8.6 (33.0–95.0) (2,755)
74.5 8.2 (47.0–94.0) (794)
73.1 8.7 (33.0–95.0) (1,961)
0.1576
36.5 (1,005/2,755)
39.7 (316/795)
35.2 (873/2,144)
0.0928
Standardized Difference
27.9 (770/2,755)
33.2 (264/795)
25.8 (506/2,144)
0.1638
89.3 (2,460/2,755)
89.9 (714/795)
89.0 (1,746/2,144)
0.0290
Diabetes
29.4 (810/2,755)
32.5 (258/795)
28.1 (552/2,144)
0.0946
History of TIA or stroke
32.1 (884/2,755)
38.5 (306/795)
29.5 (578/2,144)
0.1922
Vascular disease
46.1 (1,271/2,755)
45.2 (359/795)
46.5 (912/2,144)
0.0259
LVEF <30% 30%–50% >50%
1.0 (25/2,545) 23.4 (596/2,545) 75.6 (1,924/2,545)
2.2 (14/795) 21.5 (139/795) 76.4 (495/795)
0.6 (11/2,144) 24.1 (457/2,144) 75.3 (1429/2,144)
0.1363 0.0630 0.0248
AF type Persistent Paroxysmal Permanent
25.4 (701/2,755) 47.3 (1,303/2,755) 27.3 (751/2,755)
20.7 (164/794) 42.1 (334/794) 37.3 (296/794)
27.4 (537/1,961) 49.4 (969/1,961) 23.2 (455/1,961)
0.1580 0.1479 0.3102
Values are mean SD (range) (n) or % (n/N). AF ¼ atrial fibrillation; APT ¼ antiplatelet therapy; CHF ¼ congestive heart failure; LVEF ¼ left ventricular ejection fraction; OAC ¼ oral anticoagulation; TIA ¼ transient ischemic attack.
balanced across the OAC and APT groups in the
stroke as defined by the Bleeding Academic Research
matched sample if the absolute difference of the
Consortium (10) up to 6 months post-procedure and
standardized difference was #0.2.
device-related thrombus beyond 7 days after the
The efficacy endpoint was freedom from ischemic
procedure and up to 6 months.
stroke or systemic embolism beyond 7 days from the procedure and up to 6 months. Strokes within 7 days
STATISTICAL ANALYSIS. The efficacy and safety
were excluded, to avoid any confounding effect of
endpoints, as well as freedom from DRT, were
periprocedural strokes, which likely reflect a different
compared across the propensity-matched groups us-
mechanism unrelated to the post-implantation drug
ing a long-rank test stratified by propensity score
regimen. Safety endpoints included non-procedure-
quintile. All analyses were performed using SAS
related
version 9.4 (SAS Institute, Cary, North Carolina).
major
bleeding,
including
hemorrhagic
T A B L E 3 Baseline Demographics After Propensity Score Matching
Overall (N ¼ 1,527)
Age at enrollment (yrs)
APT (n ¼ 509)
OAC (n ¼ 1,018)
73.8 8.4 (39.0–95.0) (1,527) 73.8 8.5 (47.0–94.0) (509) 73.7 8.4 (39.0–95.0) (1,018)
Standardized Difference
0.0072
Female
37.4 (571/1,527)
38.1 (194/509)
37.0 (377/1,018)
0.0223
CHF
28.3 (432/1,527)
29.1 (148/509)
27.9 (284/1,018)
0.0261
89.8 (1,371/1,527)
89.2 (454/509)
90.1 (917/1,018)
0.0290
Diabetes
31.2 (476/1,527)
30.8 (157/509)
31.3 (319/1,018)
0.0106
History of TIA or stroke
32.9 (503/1,527)
32.2 (164/509)
33.3 (339/1,018)
0.0230
Vascular diseases
Hypertension
45.7 (698/1,527)
45.6 (232/509)
45.8 (466/1,018)
0.0039
LVEF <30% 30%–50% >50%
0.6 (9/1,527) 23.4 (358/1,527) 76.0 (1,160/1,527)
0.8 (4/509) 24.0 (122/509) 75.2 (383/509)
0.5 (5/1,018) 23.2 (236/1,018) 76.3 (777/1,018)
0.0185 0.0252 0.0232
AF type Persistent Paroxysmal Permanent
23.1 (353/1,527) 50.9 (777/1,527) 26.0 (397/1,527)
23.8 (121/509) 50.1 (255/509) 26.1 (133/509)
22.8 (232/1,018) 51.3 (522/1,018) 25.9 (264/1,018)
0.0045 0.0072 0.0223
Values are mean SD (range) (n) or % (n/N). Abbreviations as in Table 2.
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
C ENTR AL I LL U STRA T I O N Efficacy Endpoint: Ischemic Stroke or Systemic Embolism Beyond 7 Days
Ischemic Stroke Systemic Embolism Beyond 7 Days Propensity Score Matching by First Nine Variables
Event Free Probability
1.00
0.90
0.80
0.70
509
499
491
483
476
471
468
APT
1018
1012
1007
1000
996
987
982
OAC
0
1
2
5
6
3 4 Time (Months)
Treatment
APT
OAC
Log-Rank p-value = 0.3240 Søndergaard, L. et al. J Am Coll Cardiol Intv. 2019;12(11):1055–63.
No significant difference was observed. APT ¼ antiplatelet therapy; OAC ¼ oral anticoagulation.
RESULTS
(interquartile range: 68 to 80 years), and 62% of the patients were male. In the OAC group, 95% were on
There were 2,812 patients who successfully under-
vitamin K antagonists and 5% on direct OAC.
went WATCHMAN implantation in the aforemen-
Furthermore, 92% of the patients received SAPT in
tioned trials and registries. Baseline demographics of
addition to OAC. In the APT group, 91% were treated
the unmatched groups receiving post-procedural
with DAPT and 9% with SAPT.
OAC or APT are given in Table 2. Patients in the
EFFICACY ENDPOINT. There was no significant dif-
APT group were generally older, with higher rates of
ference in freedom from ischemic stroke or systemic
congestive heart failure, previous stroke, permanent
embolism beyond 7 days and up to 6 months between
rather than paroxysmal or persistent atrial fibrilla-
the 2 groups (OAC, 98.8%; APT, 99.4%; hazard ratio:
tion, low left ventricular ejection fraction, and renal
0.82; 95% confidence interval: 0.45 to 1.50; p ¼ 0.3240)
impairment.
(Central Illustration).
After propensity score matching, a total of 1,527
So as not to miss the possibility of a different result,
patients (1,018 patients in the OAC group and 509
total ischemic stroke or systemic embolism including
patients in the APT group) were included in the ana-
the initial 7 days after LAAC was also analyzed. Again,
lyses (Table 3). The median age was 75 years
there was no significant difference between 2 groups
1059
1060
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
(OAC, 98.3%; APT, 99.4%; p ¼ 0.0887) (Figure 2,
F I G U R E 2 Efficacy Endpoint: Ischemic Stroke or Systemic Embolism
Table 4).
(Including Initial 7 Days)
SAFETY ENDPOINTS. Similarly, accumulated 6-month
freedom from hemorrhagic stroke or other non-
Ischemic Stroke Systemic Embolism Propensity Score Matching by First Nine Variables
procedural major bleeding was similar between the
Event Free Probability
1.00
OAC and APT groups (95.7% and 95.5%, respectively; hazard ratio: 1.00; 95% confidence interval: 0.65 to 1.52; p ¼ 0.775), despite a larger number of early
0.90
bleeds with OAC (Figure 3). Freedom from hemorrhagic stroke or nonprocedural major bleeding at 45 days was 98.2% in the APT group and 98.0% in the
0.80
OAC group (p ¼ 0.456). Details of the bleeding events 509
499
491
483
476
471
468
APT
0.70 1018
1007
1002
995
991
983
978
OAC
1
2
3
4
5
6
0
are presented in Tables 5 and 6. Device-related thrombus was significantly more frequent in the APT group than in the OAC group
Time (Months)
Treatment
APT
(OAC, 1.4%; APT, 3.1%; p ¼ 0.014; hazard ratio not
OAC
constant from 0 to 6 months and therefore not pre-
Log-Rank P-value = 0.0887
sented) (Figure 4). When patients on post-procedural SAPT were excluded, this comparison still resulted
No significant difference was observed. APT ¼ antiplatelet therapy; OAC ¼ oral
in significant difference (OAC, 1.1%; APT, 3.3%;
anticoagulation.
p ¼ 0.0048) (Figure 5). The same conclusion was reached if patients (from EWOLUTION) who did not undergo follow-up TEE (13%) were excluded from the analysis (OAC, 1.3%; APT, 2.9%; p ¼ 0.0419)
T A B L E 4 Thromboembolism With and Without Events Occurring Within 7 Days
(Figure 6, Table 7).
6-Month Survival (%) Event
OAC
APT
p Value
Ischemic stroke or SE
98.3
99.4
0.0887
Ischemic stroke or SE >7 days
98.8
99.4
0.3240
In the OAC group, 21.4% of subjects with DRT (3 of 14) experienced ischemic stroke or systemic embolism beyond 7 days and up to 6 months, compared with 0% (0 of 16) in the APT matched group. The difference was not statistically significant (p ¼ 0.09).
SE ¼ systemic embolism; other abbreviations as in Table 2.
There were no hemorrhagic strokes in subjects experiencing DRT. F I G U R E 3 Safety Endpoint: Accumulated Freedom From Hemorrhagic Stroke or
Nonprocedural Major Bleeding
The major findings of this study are that although
Hemorrhagic Stroke or Non-procedure related Major Bleeding Propensity Score Matching by First Nine Variables
antithrombotic therapy with APT as an immediate therapy after LAAC with the WATCHMAN device is
1.00 Event Free Probability
DISCUSSION
associated with a slightly higher rate of devicerelated thrombus, thromboembolic event rates are 0.90
similar compared with immediate post-implantation OAC, and bleeding complications are the same for the 2 post-procedural regimens.
0.80
APT and OAC seem to be comparably effective 509
497
483
473
465
457
453
APT
antithrombotic options in the immediate period after
0.70 1017
998
986
976
970
961
952
OAC
LAAC with the WATCHMAN device. This supports
1
2
3
4
5
6
0
Treatment
APT
the recent change in instruction for use in Conformité Européenne mark countries, where both
Time (Months
OAC
Log-Rank P-value = 0.7749
therapeutic regimens are included. How this will translate into clinical practice may vary among institutions and on a patient-by-patient basis. A prag-
No significant difference was observed. APT ¼ antiplatelet therapy; OAC ¼ oral anticoagulation.
matic approach could be to continue OAC until efficient closure has been confirmed in patients with no contraindications, whereas patients with strong
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
T A B L E 5 Nonprocedural Major Bleeding Breakdown at 45 Days
APT Number of Events (% of All Events)
Adverse Event Classification
OAC
Number of Patients Experiencing Events (% of All Patients)
Number of Events (% of All Events)
Number of Patients Experiencing Events (% of All Patients)
12 (1.2)
Gastrointestinal bleeding
2 (20.0)
2 (0.4)
14 (56.0)
Hematuria
0 (0.0)
0 (0.0)
1 (4.0)
1 (0.1)
Other causes of blood transfusion
7 (70.0)
6 (1.2)
8 (32.0)
7 (0.7)
Stroke (hemorrhagic)
1 (10.0)
1 (0.2)
1 (4.0)
1 (0.1)
Subdural hematoma
0 (0.0)
0 (0.0)
1 (4.0)
1 (0.1)
10 (100.0)
9 (1.8)
25 (100.0)
21 (2.1)
Number of Events (% of All Events)
Number of Patients Experiencing Events (% of All Patients)
Number of Events (% of All Events)
Number of Patients Experiencing Events (% of All Patients)
Epistaxis
0 (0.0)
0 (0.0)
5 (9.1)
2 (0.2)
Gastrointestinal bleeding
5 (19.2)
5 (1.0)
34 (61.8)
28 (2.8)
Total Values are n (%). Abbreviations as in Table 2.
T A B L E 6 Nonprocedural Major Bleeding Breakdown at 6 Months
APT
Adverse Event Classification
Hematuria Other causes of blood transfusion
OAC
0 (0.0)
0 (0.0)
1 (1.8)
1 (0.1)
17 (65.4)
13 (2.6)
12 (21.8)
10 (1.0) 2 (0.2)
Stroke (hemorrhagic)
1 (3.8)
1 (0.2)
2 (3.6)
Subdural hematoma
3 (11.5)
3 (0.6)
1 (1.8)
1 (0.1)
26 (100.0)
22 (4.3)
55 (100.0)
43 (4.2)
Total Values are n (%). Abbreviations as in Table 2.
F I G U R E 4 Freedom From Device-Related Thrombosis
Device Thrombus Propensity Score Matching by First Nine Variables
Event Free Probability
1.00
0.90
0.80
509
499
480
472
465
459
457
APT
0.70 1018
1015
1002
996
993
984
979
OAC
1
2
3
4
5
6
0
Time (Months) Treatment
APT
OAC
Log-Rank P-value = 0.0148
Significantly more device-related thrombosis was seen in the antiplatelet therapy (APT) group. OAC ¼ oral anticoagulation.
1061
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
Nonwarfarin OACs were used in only 5% of the
F I G U R E 5 Freedom From Device-Related Thrombosis, With
patients in the OAC group. This reflects that
Single-Antiplatelet Therapy Patients Excluded
PROTECT-AF and PREVAIL were conducted before the widespread introduction of those agents. At pre-
Device Thrombus Propensity Score Matching by First Nine Variables
sent, only nonrandomized retrospective data exist on
1.00
the outcomes using direct OAC as post-procedural
Event Free Probability
therapy (11), although this will be tested in the upcoming PINNACLE FLX (Investigational Device Eval-
0.90
uation of the WATCHMAN FLX LAA Closure Technology) trial (NCT02702271). Similarly, outcomes in the population deemed unsuitable for OAC and
0.80
treated with short-course DAPT post-implantation are 0.70
461
451
433
425
421
415
413
APT
922
919
910
906
903
894
889
OAC
0
1
2
3
4
5
6
APT
being studied in the ongoing randomized ASAP-TOO (Assessment of the WATCHMAN Device in Patients Unsuitable for Oral Anticoagulation) trial.
Time (Months) Treatment
Although DRT was seen more frequently in the APT OAC
group, the overall incidence was low (3.1%). Importantly, it is still controversial whether DRT is directly
Log-Rank P-value = 0.0048
associated with increased risk for stroke, transient Significantly more device-related thrombosis was seen in the antiplatelet
ischemic attack, or systemic embolization. In a pre-
therapy (APT) group. OAC ¼ oral anticoagulation.
vious analysis of PROTECT-AF, PREVAIL, CAP, and CAP2, DRT was associated with higher rate of stroke
contraindications for OAC may be on APT immediately after LAAC. Because the majority of the patients in the APT group were on dual therapy, the present study does not provide robust evidence for the outcome of SAPT during
the
immediate
post-procedural
period.
Comparative studies are warranted before one can specifically comment on the efficacy as well as safety of SAPT in particular, after LAAC.
and systemic embolism (12). A similar finding was noted in a retrospective analysis (n ¼ 469) (13). But in EWOLUTION, this was not the case. And in this focused analysis comparing short-term OAC and APT, the elevated DRT rate with APT also did not translate into significantly more clinical events, though it should be noted that the overall sample size was still small, despite the fact that this is the most comprehensive analysis to this date. STUDY LIMITATIONS. There are challenges in using
F I G U R E 6 Freedom From Device-Related Thrombosis, With Patients Who
Did Not Undergo Transesophageal Echocardiography Excluded
a DRT endpoint. First, it may be difficult to diagnose DRT, particularly on transthoracic echocardiography. Second, there is no standardized definition of
Device Thrombus Propensity Score Matching by First Nine Variables
DRT. For instance, EWOLUTION identified DRT through site reporting, on-site monitoring, and
1.00
sponsor classification of events, while the clinical Event Free Probability
1062
events committee adjudicated these events for 0.90
other studies. Similarly, as discussed earlier, the availability, frequency, and timing of TEE to identify DRT were not
0.80
necessarily uniform among trials. In particular, trans0.70
415
414
401
397
393
390
388
APT
esophageal
830
827
817
813
811
804
800
OAC
available in 87% of the patients in EWOLUTION, per
0
1
2
3
4
5
6
APT
assessments
were
local practice, with an average time to first TEE of 77 days and an average time to last TEE of 225 days. In
Time (Months) Treatment
echocardiographic
OAC
Log -Rank P-value = 0.0419
contrast, in PROTECT-AF and PREVAIL, TEE was performed regularly in all patients at 45 days, 6 months, and 12 months after WATCHMAN implantation.
Significantly more device-related thrombosis was seen in the antiplatelet therapy (APT) group. OAC ¼ oral anticoagulation.
Furthermore, in this study only LAAC cases using the WATCHMAN device were evaluated. Whether the results can be generalized to all other LAAC devices
Søndergaard et al.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 11, 2019 JUNE 10, 2019:1055–63
Anticoagulation Versus Antiplatelet Therapy After LAAC
T A B L E 7 Device Thrombus Rates After Excluding Subsets of Subjects*
6-Month Device Thrombus Rate Subset
OAC
APT
Odds Ratio for APT vs. OAC (95% CI)
p Value
All subjects (N ¼ 1,018 OAC and N ¼ 509 APT)
1.4
3.1
2.3 (1.12–4.76)
0.0148
Excluding subjects on single APT (n ¼ 922 OAC and n ¼ 462 APT)
1.1
3.3
3.1 (1.4–6.9)
0.0048
Excluding EWOLUTION subjects without FU TEE (n ¼ 830 OAC and n ¼ 415 APT)
1.3
2.9
2.2 (1.0–5.1)
0.0419
*All subjects who are propensity matched to subjects in the subset of interest are also excluded. FU ¼ follow-up; TEE ¼ transesophageal echocardiography; other abbreviations as in Tables 1 and 2.
remains uncertain. This study also had limitations
PERSPECTIVES
inherent to all retrospective and not pre-defined analyses. Last, risk factors such as renal function and alcohol consumption were not included in the propensity score, as these were not captured in all 6 trials and registries.
higher DRT rate with APT, both APT and OAC therapies resulted in similar safety and efficacy endpoints. This study suggests that APT might be a reasonable post-LAAC antithrombotic alternative in patients with high bleeding risk. CORRESPONDENCE:
Blegdamsvej
rate of DRT, thromboembolic event rates were similar compared with immediate post-implantation OAC, and bleeding complications were the same for the 2 post-procedural regimens. APT might be a reasonable post-LAAC antithrombotic therapy in patients at high bleeding risk. WHAT IS NEXT? Further studies are warranted to clarify
Dr.
Lars
Søndergaard, Rigshospitalet, Section 2011, University of Copenhagen,
suffice.
antithrombotic therapy with APT was associated with a higher
After LAAC with the WATCHMAN device, despite a
FOR
post-implantation OAC, but registries suggested that APT may
WHAT IS NEW? This study demonstrates that although
CONCLUSIONS
ADDRESS
WHAT IS KNOWN? Initial trials of LAAC included 45 days of
9,
2100
Copenhagen,
efficacy and safety of DAPT versus SAPT and to address controversies concerning the association between DRT and clinical outcomes of embolization.
Denmark. E-mail:
[email protected].
REFERENCES 1. Blackshear JL, Odell JA. Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation. Ann Thorac Surg 1996;61:755–9. 2. Holmes DR, Reddy VY, Turi ZG, et al. Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised noninferiority trial. Lancet 2009;374:534–42. 3. Reddy VY, Doshi SK, Sievert H, et al. Percutaneous left atrial appendage closure for stroke prophylaxis in patients with atrial fibrillation: 2.3year follow-up of the PROTECT AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients with Atrial Fibrillation) trial. Circulation 2013;127:720–9. 4. Reddy VY, Sievert H, Halperin J, et al. Percutaneous left atrial appendage closure vs warfarin for atrial fibrillation: a randomized clinical trial. JAMA 2014;312:1988–98. 5. Holmes DR, Kar S, Price MJ, et al. Prospective randomized evaluation of the Watchman left atrial appendage closure device in patients with atrial fibrillation versus long-term warfarin therapy: the PREVAIL trial. J Am Coll Cardiol 2014;64:1–12.
6. Boersma LV, Schmidt B, Betts TR, et al. Implant success and safety of left atrial appendage closure with the WATCHMAN device: peri-procedural outcomes from the EWOLUTION registry. Eur Heart J 2016;37:2465–74. 7. Reddy VY, Holmes D, Doshi SK, et al. Safety of percutaneous left atrial appendage closure: results from the WATCHMAN Left Atrial Appendage System for Embolic Protection in Patients with AF (PROTECT AF) clinical trial and the Continued Access Registry. Circulation 2011;123:417–24. 8. Reddy VY, Möbius-Winkler S, Miller MA, et al. Left atrial appendage closure with the Watchman device in patients with a contraindication for oral anticoagulation: the ASAP study (ASA Plavix Feasibility Study With Watchman Left Atrial Appendage Closure Technology). J Am Coll Cardiol 2013;61:2551–6. 9. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083–107. 10. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical
trials—a consensus report from the Bleeding Academic Research Consortium. Circulation 2011;123: 2736–47. 11. Enomoto Y, Gadiyaram VK, Gianni C, et al. Use of non-warfarin oral anticoagulants instead of warfarin during left atrial appendage closure with the Watchman device. Heart Rhythm 2017;14: 19–24. 12. Dukkipati SR, Kar S, Holmes DR, et al. Devicerelated thrombus after left atrial appendage closure—incidence, predictors, and outcomes. Circulation 2018;138:874–85. 13. Fauchier L, Cinaud A, Brigadeau F, et al. Device-related thrombosis after percutaneous left atrial appendage occlusion for atrial fibrillation. 1528–36.
J
Am
Coll
Cardiol
2018;71:
KEY WORDS antiplatelet therapy, antithrombotic therapy, atrial fibrillation, direct oral anticoagulation, left atrial appendage closure, oral anticoagulation, stroke
1063