Abstract
Background:
Left atrial appendage occlusion (LAAO) is an approved alternative for stroke prevention in atrial fibrillation (AF) for patients with an ‘appropriate rationale’ to avoid long-term oral anticoagulation (OAC). Many patients undergoing LAAO are at high risk of bleeding.
Objectives:
Investigate whether dual antiplatelet therapy (DAPT) is a safe alternative to OAC (direct oral anticoagulation [DOAC] or warfarin) with aspirin after LAAO.
Methods:
Using National Cardiovascular Data Registry LAAO Registry data, patients undergoing WATCHMAN FLX implantation (August 5, 2020-September 30, 2021) were included in 1:1 propensity-matched analyses comparing discharge medication regimens (DAPT, DOAC/aspirin, or warfarin/aspirin). A composite endpoint (death, stroke, major bleeding, and systemic embolism), its components, and device-related thrombus (DRT) between discharge and 45 days were evaluated.
Results:
In 49,968 patients implanted with WATCHMAN FLX during the study period, mean age was 77 years and 40% were women. Post-implant, DOAC/aspirin was prescribed in 24,497 patients, warfarin/aspirin in 3,913 and DAPT in 4,155. DAPT patients had more comorbid conditions than patients receiving OAC/aspirin. After propensity score-matching, 45-day composite endpoint rates were similar among the groups (DAPT 3.44% vs DOAC/aspirin 4.06%, p=0.13; DAPT 3.23% vs warfarin/aspirin 3.08%, p=0.75). Death, stroke, and DRT were also similar; major bleeding was slightly increased in DOAC/aspirin patients (DAPT 2.48%, DOAC/aspirin 3.25%, p=0.04; DAPT 2.25% vs warfarin/aspirin 2.22%, p=0.93).
Conclusions:
In a large registry, DAPT had a similar safety profile compared with current FDA-approved post-implant drug regimens of oral anticoagulation with aspirin following LAAO with WATCHMAN FLX. Shared decision making for nonpharmacologic stroke prevention should include discussion of post-procedure medical therapy options.
Keywords: Left atrial appendage occlusion, anticoagulation, antiplatelet, bleeding, stroke
Condensed abstract
Left atrial appendage occlusion (LAAO) is an approved alternative for stroke prevention in atrial fibrillation (AF). This analysis investigated whether dual antiplatelet therapy (DAPT) is a safe alternative to oral anticoagulation (OAC) with aspirin after LAAO with WATCHMAN FLX. Patients discharged on DAPT had more comorbid conditions than those on OAC/aspirin. After adjustment for these differences, rates of death, stroke, and major bleeding with DAPT were not increased compared to OAC/aspirin at 45 days following LAAO with WATCHMAN FLX. A shared decision-making discussion between patients and clinicians in the selection of post-procedure medications following LAAO is reasonable.
Tweet:
DAPT is a safe alternative to OAC/aspirin immediately after WATCHMAN FLX in select patients: a registry study comparing DAPT to OAC/aspirin post-implant showed no difference in stroke or device-related thrombus. (add link to article) #CVD
Introduction
Patients with atrial fibrillation (AF) have a five-fold greater risk of stroke, with the left atrial appendage as the most common source of embolus.(1–3) While oral anticoagulation (OAC) is proven to reduce stroke compared to aspirin or no antiplatelet, many patients are not good candidates for long-term OAC for a variety of reasons, including a need for additional medications associated with bleeding risk; prior bleeding; or high risk of future bleeding.(4–6) LAAO with WATCHMAN was approved in 2015 with post-procedure medication protocols designed in the 2000s.(7–9) The initial approved post-procedure medication regimen included 45 days of warfarin with aspirin, followed by 4.5 months of dual antiplatelet therapy (DAPT; P2Y12 inhibitor with aspirin) and was designed to reduce device-related thrombus (DRT) and stroke. When WATCHMAN FLX was approved by the FDA, DOACs were included as a post-procedure medication regimen, still along with aspirin. Given the high bleeding risk seen in patients referred for LAAO, interest remained high in less intensive post-procedure anti-thrombotic regimens.(10)
Limited data on the use of DAPT with WATCHMAN 2.5, the initial commercial device, suggested similar outcomes between patients discharged on DAPT compared to OAC/aspirin.(11–13) In addition, DAPT is FDA-approved post-implant for the Amulet LAAO device following the Amulet IDE trial.(13,14) To compare DAPT as a post-procedure medication regimen following WATCHMAN FLX with current FDA-approved strategies (OAC/ aspirin) we examined outcomes of death, stroke, DRT and bleeding in the mandated nationwide LAAO registry.(11–13) Given the current practice of post-implant imaging limited to 45 day follow-up (no longer at 6 months or 12 months in most practices), the timeframe of 45 days was selected for the primary outcome.
Methods
Data Source
The American College of Cardiology (ACC) Foundation’s National Cardiovascular Data Registry (NCDR) for LAAO was initiated in December 2015 after FDA approval of the first WATCHMAN device.(15) Reimbursement for LAAO for patients participating in Medicare was made contingent on participation in the registry. Participating centers use standardized definitions to collect clinical information on consecutive LAAO cases including patient demographics, comorbidities, device implant information, medications, and outcomes. The data captured reflects the real-world setting. The NCDR LAAO Registry has developed and validated a computer-based algorithm to adjudicate clinical endpoints using discrete event data and supporting variables against standard event definitions. (15) Patients receiving WATCHMAN FLX and participating in the NCDR LAAO Registry were included. The LAAO Registry has been approved by Chesapeake Research Review, Inc., an independent IRB, and in accordance with 45 CFR 46.116(d) of the federal regulations, Chesapeake’s IRB has waived the requirement for obtaining informed patient consent for the registry. The IRB has also waived HIPAA Authorization in accordance with 45 CFR 164.512(i)(2) (from https://cvquality.acc.org/docs/default-source/ncdr/Data-Collection/laao-registry-faq-8_17_16.pdf?sfvrsn=ae748dbf_0).
This analysis was supported by Boston Scientific (Marlborough, MA) and utilized data from the NCDR LAAO Registry for patients treated with WATCHMAN FLX between August 5, 2020 to September 30, 2021. During this time period, existing practice driven by clinical trials was for 45-day post-procedure imaging with transition from OAC/aspirin to DAPT at that time point if no DRT or leak >5 mm found. Therefore, this analysis is focused on medical therapy within the first 45 days and this follow-up period is used for the analysis of efficacy and safety of DAPT compared to OAC/aspirin immediately post-implant. Given existing label-directed therapy to transition from OAC/aspirin to DAPT at 45 days, medical treatment following this period would be expected to be similar in all patients (i.e., all patients treated with DAPT after 45 days).
Study Population
Patients were included in the analysis if they had a calculated CHA2DS2-VASc score of ≥2 in men and ≥3 in women; successfully received a WATCHMAN FLX device without any concomitant cardiac or non-cardiac procedures (including, but not limited to, cardiac ablation, transcatheter valve implantation, coronary intervention); and received either DAPT, warfarin/ aspirin, or DOAC/aspirin at discharge.
Study Endpoints
Endpoints were evaluated between the date of hospital discharge and 45-days post-procedure, as well as at 6 months. A composite endpoint (all-cause death, stroke [ischemic and/or hemorrhagic], major bleeding [any bleeding requiring hospitalization, and/or causing a decrease in hemoglobin level > 2g/dL, and/or requiring blood transfusion that was not hemorrhagic stroke (15)], and systemic embolism) was evaluated as well as its components and DRT.
Statistical Analysis
Data was summarized using descriptive statistics for continuous variables and frequency tables or proportions for discrete variables. To adjust for differences in baseline characteristics, 1:1 propensity score-matching using logistic regression was performed for the DAPT vs DOAC/aspirin and, separately, for the DAPT vs warfarin/aspirin comparisons. Pairs of medication groups (either DAPT vs DOAC/aspirin or DAPT vs warfarin/aspirin) were matched using the nearest neighbor method of calculating propensity scores (caliper ≤0.05). Propensity score matching (1:1) was performed between DOAC/aspirin and DAPT patients, leading to 4,155 patients in each group. Propensity score matching (1:1) in warfarin/aspirin and DAPT patients led to 2,663 patients in each group. Following propensity score-matching, there were no differences in baseline characteristics between groups (Table 1). The composite endpoint and its components were evaluated in the propensity-score matched patient populations. Given that TEE is not recommended and was not performed in all patients at 6 months, DRT was not included at this timepoint.
Table 1.
Demographic and clinical characteristics in the propensity-score matched WATCHMAN FLX patient population
| Variable | DAPT (n=4,155) | DOAC/ aspirin (n=4,155) | p-value | DAPT (n=2,663) | Warfarin/ aspirin (n=2,663) | p-value |
|---|---|---|---|---|---|---|
| Age at enrollment (years) | 77.1±7.7 | 77.1±7.4 | 0.99 | 77.0±7.8 | 77.0±7.6 | 0.92 |
| Women | 1741 (41.9) | 1701 (40.9) | 0.37 | 1081 (40.6) | 1082 (40.6) | 0.99 |
| White | 3829 (92.2) | 3842 (92.5) | 0.59 | 2519 (94.6) | 2511 (94.3) | 0.63 |
| Black/African American | 237 (5.7) | 230 (5.5) | 0.74 | 98 (3.7) | 99 (3.7) | 0.94 |
| American Indian/Alaska Native | 6 (0.1) | 7 (0.2) | 0.78 | 3 (0.1) | 7 (0.3) | 0.34 |
| Asian | 54 (1.3) | 42 (1.0 | 0.22 | 27 (1.0) | 21 (0.8) | 0.38 |
| Native Hawaiian/Pacific Islander | 0 (0.0) | 6 (0.1) | 0.03 | 0 (0.0) | 2 (0.1) | 0.50 |
| Hispanic Origin | 119 (2.9) | 118 (2.9) | 0.93 | 73 (2.8) | 48 (1.8) | 0.02 |
| CHA2DS2-VASc risk score | 5.0±1.4 | 5.0±1.4 | 0.83 | 5.0±1.4 | 5.0±1.4 | 0.86 |
| Congestive heart failure | 1708 (41.1) | 1733 (41.7) | 0.58 | 1097 (41.2) | 1104 (41.5) | 0.85 |
| Hypertension | 3865 (93.0) | 3882 (93.4) | 0.46 | 2475 (92.9) | 92.6% (2465/2663) | 0.60 |
| Diabetes mellitus | 1627 (39.2) | 1672 (39.2) | 1.00 | 1033 (38.8) | 1043 (39.2) | 0.78 |
| Stroke | 1006 (24.2) | 1004 (24.2) | 0.96 | 620 (23.3) | 609 (22.9) | 0.72 |
| Transient ischemic attack | 481 (11.6) | 489 (11.8) | 0.79 | 329 (12.4) | 316 (11.9) | 0.58 |
| Thromboembolic event | 685 (16.5) | 675 (16.2) | 0.76 | 400 (15.0) | 417 (15.7) | 0.52 |
| Vascular disease | 2433 (58.6) | 2460 (59.2) | 0.55 | 1506 (56.6) | 1504 (56.5) | 0.96 |
| Left ventricular dysfunction | 435 (10.5) | 446 (10.7) | 0.70 | 278 (10.4) | 289 (10.9) | 0.63 |
| HAS-BLED score | 2.6±1.0 | 2.6±1.0 | 0.6164 | 2.5±1.0 | 2.5±1.0 | 0.98 |
| Uncontrolled hypertension | 1114 (26.8) | 1149 (27.7) | 0.38 | 715 (26.9) | 737 (27.7) | 0.50 |
| Abnormal renal function | 723 (17.4) | 740 (17.8) | 0.62 | 439 (16.5) | 454 (17.1) | 0.58 |
| Abnormal liver function | 161 (3.9) | 140 (3.4) | 0.22 | 98 (3.7) | 90 (3.4) | 0.55 |
| Prior bleeding | 3305 (79.6) | 3328 (80.1) | 0.57 | 1942 (73.0) | 1949 (73.2) | 0.88 |
| Labile INR | 132 (3.2) | 145 (3.5) | 0.42 | 131 (4.9) | 124 (4.7) | 0.65 |
| Alcohol use | 166 (4.0) | 163 (3.9) | 0.87 | 95 (3.6) | 103 (3.9) | 0.56 |
| Antiplatelet medication use | 1994 (48.0) | 1989 (47.9) | 0.91 | 1153 (43.3) | 1158 (43.5) | 0.91 |
| Nonsteroidal anti-inflammatory drug use | 812 (19.6) | 778 (18.7) | 0.34 | 450 (16.9) | 439 (16.5) | 0.69 |
| Increased fall risk | 1652 (39.8) | 1641 (39.6) | 0.81 | 1110 (41.8) | 1117 (42.1) | 0.83 |
| Clinically relevant bleeding event | 3259 (78.5) | 3259 (78.6) | 0.90 | 1800 (67.6) | 1788 (67.2) | 0.76 |
| Cardiac structural intervention | 386 (9.3) | 404 (9.7) | 0.50 | 278 (10.4) | 283 (10.6) | 0.81 |
| Cardiomyopathy | 815 (19.7) | 849 (20.5) | 0.37 | 540 (20.3) | 564 (21.2) | 0.43 |
| Chronic lung disease | 980 (23.6) | 991 (23.9) | 0.79 | 575 (21.6) | 562 (21.1) | 0.68 |
| Coronary artery disease | 2082 (50.2) | 2112 (50.9) | 0.52 | 1286 (48.3) | 1271 (47.7) | 0.68 |
| Sleep apnea | 1095 (26.4) | 1070 (25.8) | 0.52 | 807 (30.3) | 792 (29.8) | 0.65 |
| Paroxysmal AF | 2666 (64.9) | 2722 (65.9) | 0.32 | 1582 (60.1) | 1608 (60.7) | 0.65 |
| Persistent or long-standing persistent AF | 967 (23.5) | 941 (22.8) | 0.42 | 642 (24.4) | 636 (24.0) | 0.75 |
| Permanent AF | 477 (11.6) | 467 (11.3) | 0.67 | 407 (15.5) | 403 (15.2) | 0.81 |
| Prior ablation | 613 (15.4) | 619 (15.5) | 0.85 | 503 (19.6) | 506 (19.6) | 0.98 |
| LVEF | 54.0±10.3 | 53.9±9.9 | 0.90 | 53.8±9.9 | 53.7±10.3 | 0.65 |
| No residual leak | 3911 (95.8) | 3926 (95.9) | 0.85 | 2504 (95.7) | 2533 (96.0) | 0.63 |
| Residual leak 0 – ≤3mm | 151 (3.7) | 146 (3.6) | 0.75 | 100 (3.8) | 92 (3.5) | 0.52 |
| Residual leak 3 – ≤5mm | 18 (0.4) | 20 (0.5) | 0.75 | 12 (0.5) | 14 (0.5) | 0.71 |
| Missing leak information | 72 | 60 | 47 | 24 |
Values are mean ± SD, or n (%). Abbreviations: AF=atrial fibrillation, DAPT = dual antiplatelet therapy, DOAC = direct oral anticoagulant, INR= international normalized ratio, LVEF=left ventricular ejection fraction
For this analysis, the propensity score included the following variables based on clinical relevance and differences between the medication cohorts: age, gender, race/ethnicity, CHA2DS2-VASc and HAS-BLED score components, type of atrial fibrillation, diabetes, fall risk, history of bleeding, chronic lung disease, sleep apnea, cardiomyopathy, coronary artery disease, prior ablation, left ventricular ejection fraction and post-implant device margin residual leak. Statistical analyses were conducted using SAS version 9.4 (SAS Institute, Cary, North Carolina).
Results
Baseline Characteristics
The WATCHMAN FLX patient cohort included patients enrolled and meeting analysis criteria in the NCDR LAAO Registry between August 5, 2020 to September 30, 2021.
Baseline characteristics are described in Supplemental Table 1. The total number of patients implanted during that time was 49,968. Overall, the mean age of enrolled subjects was 77 years old and 40% were women. OAC/aspirin was prescribed in 28,410 patients; DOAC/aspirin was prescribed more often at discharge (n=24,497) than warfarin/aspirin (n=3,913). One in ten patients of the total cohort were prescribed DAPT (n=4,155). Patients on OAC/aspirin or DAPT totaled 32,565 patients (65% of all implants); the remainder received DOAC alone (n=10,976), warfarin alone (n=1,207), or other medication regimens (Supplemental Table 2).
More Black patients received DAPT compared to either DOAC/aspirin or warfarin/aspirin-treated patients. There was no difference by sex. The mean CHA2DS2-VASc and HAS-BLED scores were higher in the DAPT group compared to DOAC/aspirin or warfarin/aspirin groups (Supplemental Table 1).
Compared with patients discharged on DOAC/aspirin and warfarin/aspirin, patients receiving DAPT post-implant were more likely to have a history of vascular disease, antiplatelet or nonsteroidal anti-inflammatory drug use, chronic lung disease, coronary artery disease and paroxysmal AF. DAPT patients also had a higher prevalence of diabetes, prior stroke, left ventricular dysfunction, abnormal renal function, and abnormal liver function compared to DOAC/aspirin. Fall risk, sleep apnea and prior ablation were less common in DAPT compared to DOAC/aspirin patients. The most significant difference between post-implant regimens was a history of clinically relevant bleeding: 78.5% of patients in the DAPT group met this definition compared with only 55.9% in the DOAC/aspirin group (p<0.01) and in the warfarin/aspirin group 55.1% (p<0.01). (Supplemental Table 1)
Comparison of DAPT with DOAC/aspirin
In the unadjusted patient population, the composite endpoint (death, stroke, major bleeding, and systemic embolism) occurred in 3.4% of DAPT and 3.0% of DOAC/aspirin patients between discharge and 45 days (±14 days, p=0.09; Table 2). The rates of death (all-cause and cardiovascular/unknown) and stroke (all and hemorrhagic) were higher in DAPT compared to DOAC/aspirin treated patients. Other unadjusted 45-day event rates are shown in Table 2 and were not significantly different. At 6-months post-discharge (+ 60 days/−30 days), the rates of the composite endpoint, death (all-cause and cardiovascular/unknown), and stroke (all and hemorrhagic) continued to be higher in the DAPT compared to DOAC/aspirin cohort (Table 2).
Table 2.
Outcomes in the unmatched cohorts between discharge, 45 days and 6 months
| DAPT | DOAC/Aspirin | p-value (DAPT vs DOAC/Aspirin) | Warfarin/Aspirin | p-value (DAPT vs Warfarin/Aspirin) | |
|---|---|---|---|---|---|
| Events discharge through 45 days | 4,155 | 24,497 | 3,913 | ||
| Composite endpoint | 143 (3.44) | 724 (2.96) | 0.09 | 107 (2.73) | 0.07 |
| All death | 34 (0.82) | 128 (0.52) | 0.02 | 30 (0.77) | 0.79 |
| Cardiovascular/unknown death | 34 (0.6) | 45 (0.2) | <0.01 | 16 (0.4) | 0.35 |
| All stroke | 13 (0.31) | 41 (0.17) | 0.05 | 9 (0.23) | 0.48 |
| Ischemic stroke | 9 (0.22) | 32 (0.13) | 0.18 | 6 (0.15) | 0.51 |
| Hemorrhagic stroke | 4 (0.10) | 7 (0.03) | 0.06 | 3 (0.08) | >0.99 |
| Systemic embolism | 0 (0) | 1 (0.004) | >0.99 | 1 (0.03) | 0.49 |
| Major bleeding | 103 (2.48) | 574 (2.34) | 0.59 | 77 (1.97) | 0.12 |
| Device-related thrombus out of those with TEE or CT | 6 (0.24) | 33 (0.19) | 0.47 | 9 (0.30) | 0.70 |
| Events discharge through 6 months | 3,575 | 21,590 | 3,727 | ||
| Composite endpoint | 332 (9.3) | 1605 (7.4) | 0.0001 | 289 (8.2) | 0.10 |
| All death | 138 (3.9) | 625 (2.9) | 0.002 | 144 (4.1) | 0.63 |
| Cardiovascular/unknown death | 47 (1.3) | 205 (0.9) | 0.04 | 46 (1.3) | 0.97 |
| All stroke | 37 (1.0) | 146 (0.7) | 0.02 | 30 (0.9) | 0.42 |
| Ischemic stroke | 22 (0.6) | 115 (0.5) | 0.53 | 19 (0.5) | 0.67 |
| Hemorrhagic stroke | 13 (0.4) | 26 (0.1) | 0.0006 | 9 (0.3) | 0.41 |
| Systemic embolism | 3 (0.1) | 7 (0.03) | 0.16 | 3 (0.1) | >0.99 |
| Major bleeding | 181 (5.1) | 967 (4.5) | 0.12 | 152 (4.3) | 0.13 |
Binary rates. Values are n (%); P-value is from Chi-square or Fisher Exact test. The composite endpoint was defined as all-cause death, stroke (ischemic and/or hemorrhagic), major bleeding, and systemic embolism between date of hospital discharge and 45-days post-procedure. Abbreviations: DAPT = dual antiplatelet therapy, DOAC = direct oral anticoagulant, OAC = oral anticoagulant
Patients who received DAPT (n=4,155) as a post-procedure medication regimen were matched to patients who received DOAC/aspirin (n=4,155). Following propensity matching, there was no statistically significant difference in the composite endpoint: 3.4% of patients in the DAPT group and 4.1% of patients in the DOAC/aspirin group experienced a composite endpoint event at 45 days (p=0.13; Table 3). Major bleeding was the main driver of the composite outcome and was statistically lower in the DAPT group compared to DOAC/aspirin (DAPT 2.5% vs DOAC/Aspirin 3.3%, p=0.04; Central Illustration, Table 3). The rates of death and stroke at 45 days post-discharge were low and similar between groups (DAPT 0.3% vs DOAC/aspirin 0.2%, p=0.18; Central Illustration, Table 3). DRT rates at 45 days were low and not different between post-implant regimens of DAPT (0.1%) versus DOAC/aspirin (0.2%, p=0.59). After 6 months of follow-up, no significant differences were found between the matched DAPT and DOAC/aspirin treated patients in clinical outcomes.
Table 3.
Outcomes in the propensity-score matched cohorts between discharge, 45 days and 6 months
| DAPT | DOAC/Aspirin | p-value (DAPT vs DOAC/ Aspirin) | DAPT | Warfarin/Aspirin | p-value(DAPT vs Warfarin/Aspirin) | |
|---|---|---|---|---|---|---|
| Events discharge through 45 days | ||||||
| Composite endpoint | 143 (3.44) | 169 (4.06) | 0.13 | 86 (3.23) | 82 (3.08) | 0.75 |
| All death | 34 (0.82) | 32 (0.77) | 0.80 | 22 (0.83) | 22 (0.83) | >0.99 |
| Cardiovascular/unknown death | 23 (0.6) | 12 (0.3) | 0.06 | 17 (0.6) | 13 (0.5) | 0.46 |
| All stroke | 13 (0.31) | 7 (0.17) | 0.18 | 7 (0.26) | 6 (0.23) | 0.78 |
| Ischemic stroke | 9 (0.22) | 7 (0.17) | 0.62 | 4 (0.15) | 4 (0.15) | >0.99 |
| Hemorrhagic stroke | 4 (0.10) | 0 (0) | 0.12 | 3 (0.11) | 2 (0.08) | >0.99 |
| Systemic embolism | 0 (0) | 0 (0) | NE | 0 (0) | 0 (0) | NE |
| Major bleeding | 103 (2.48) | 135 (3.25) | 0.04 | 60 (2.25) | 59 (2.22) | 0.93 |
| Device-related thrombus out of those with TEE or CT | 6 (0.24) | 8 (0.26) | 0.89 | 4 (0.26) | 4 (0.19) | 0.73 |
| Events discharge through 6 months | ||||||
| Composite endpoint | 332 (9.3) | 326 (8.7) | 0.42 | 192 (8.5) | 218 (9.0) | 0.54 |
| All death | 138 (3.9) | 116 (3.1) | 0.08 | 82 (3.6) | 102 (4.2) | 0.30 |
| Cardiovascular/unknown death | 47 (1.3) | 39 (1.0) | 0.29 | 31 (1.4) | 34 (1.4) | 0.92 |
| All stroke | 37 (1.0) | 24 (0.6) | 0.07 | 24 (1.1) | 21 (0.9) | 0.50 |
| Ischemic stroke | 22 (0.6) | 21 (0.6) | 0.77 | 14 (0.6) | 14 (0.6) | 0.86 |
| Hemorrhagic stroke | 332 (9.3) | 326 (8.7) | 0.42 | 9 (0.4) | 7 (0.3) | 0.52 |
| Systemic embolism | 3 (0.1) | 0 (0) | 0.12 | 1 (0.04) | 2 (0.1) | >0.99 |
| Major bleeding | 181 (5.1) | 215 (5.8) | 0.18 | 102 (4.5) | 116 (4.8) | 0.65 |
Binary rates. Values are n (%); P-value is from Chi-square or Fisher Exact test. The composite endpoint was defined as all-cause death, stroke (ischemic and/or hemorrhagic), major bleeding, and systemic embolism between date of hospital discharge and 45-days post-procedure. NE=not evaluable. Abbreviations: DAPT = dual antiplatelet therapy, DOAC = direct oral anticoagulant, OAC = oral anticoagulant
Central illustration. Study overview and findings.
Propensity-matched (1:1) outcomes at 45 days post-implant in patients discharged on DAPT (blue) vs DOAC/aspirin (green) and warfarin/aspirin (orange) treated with WATCHMAN FLX. Abbreviations: DAPT = dual antiplatelet therapy, DOAC = direct oral anticoagulant
Comparison of DAPT with Warfarin/aspirin
Comparing unadjusted patients treated with DAPT or warfarin/aspirin, no differences were found in the composite endpoint (DAPT 3.4%, warfarin/aspirin 2.7%) between discharge and 45 days (p=0.07; Table 2). Other unadjusted event rates were not significantly different between groups and are shown in Table 2. Unadjusted clinical event rates at 6-months were also similar between DAPT and warfarin/aspirin treated patients.
A total 2,663 DAPT patients were matched with warfarin/aspirin patients (n=2,663). There was no statistically significant difference in the composite endpoint in these patient cohorts (Table 3). In the DAPT group, 3.2% of patients compared to 3.1% of patients on warfarin/aspirin experienced a composite endpoint event (p=0.75). No differences in the components of the endpoint were found (death: DAPT 0.8% vs warfarin/aspirin 0.8%, p>0.99; stroke: 0.3% vs 0.2%, p=0.78; major bleeding: 2.3% vs 2.2%, p=0.93; systemic embolism: 0% vs 0%, P>0.99; Central Illustration, Table 3). There were no differences in DRT (0.2% in both DAPT and warfarin/aspirin patients; p>0.99) at 45 days.
Notably, not all patients had TEE data reported at 45 days: 54% of DAPT, 65% of DOAC/aspirin, and 71% of warfarin/aspirin patients reported TEE results. After 6 months of follow-up, no significant differences in clinical outcomes were found between the matched DAPT and warfarin/aspirin treated patients.
Discussion
In this large, propensity-matched nationwide analysis of patients undergoing LAAO with WATCHMAN FLX for stroke reduction in the setting of AF and elevated bleeding risk, there are several clinically relevant findings. First, there was no increase in the composite outcome of death, stroke, bleeding, and systemic embolism between patients selected to take DAPT post-procedure versus OAC/aspirin at 45 days. Second, rates of DRT at 45-day follow-up (the time when OAC/aspirin is transitioned to DAPT if no DRT or significant peridevice leak is seen) were low, and not different between groups. Third, patients prescribed DAPT compared to DOAC/aspirin had significantly less bleeding, even as patients selected for DAPT had a greater history of prior bleeding. No differences in bleeding between DAPT and warfarin/aspirin were seen.
Prior research on WATCHMAN examined the role of immediate post-implant DAPT in smaller patient cohorts and without propensity matching but provided early clues that DAPT may be a reasonable alternative. Data from the EWOLUTION trial, conducted in Europe between 2013–2015 and monitored by a contract research organization, examined the prior device iteration (WATCHMAN 2.5) among patients deemed unsuitable for anticoagulation.(11) Patients in this prospective registry (n=1,020) were followed for two years, with 60% discharged on DAPT. The rate of DRT was 4.0% and most were found at the initial post-implant imaging period, generally within 90 days. Over half of patients underwent treatment with intensification of medication; there were no strokes, transient ischemic attacks, or systemic emboli in 21 months of follow-up among patients with DRT found on follow-up imaging.
An analysis of patients receiving WATCHMAN 2.5 followed during an earlier time frame in the NCDR LAAO Registry also examined real-world practice patterns of post-implant medications between 2016–2018. Compared to the current NCDR analysis, the DAPT group was a smaller portion of the analysis cohort of 31,994 patients with follow-up was reported to 6 months.(12) Note that 58% were discharged on OAC/aspirin and only 5% on DAPT (1,614 patients). In the unadjusted analysis there was an increase in DRT in the DAPT group compared to OAC (DAPT: 3.3% vs warfarin/aspirin 1.9% vs DOAC/aspirin 1.7%, overall P<0.0001) with no differences in the adjusted analysis. Like EWOLUTION, notably, there was no difference in rates of stroke or transient ischemic attacks between groups.(12).
Data on DAPT following the Amulet (Abbott Vascular, Plymouth, MN) device for LAAO are also available. The Amulet LAAO IDE trial was conducted from 2016–2019, and of those in the Amulet group, 75.7% were discharged on DAPT, compared to the WATCHMAN 2.5 group, of which 82.0% were discharged on warfarin/aspirin.(13) Despite the significant post-procedure medication differences, rates of stroke/systemic embolism (p=0.50) and major bleeding (p=0.32) were not significantly different. Rates of DRT were low in both arms (Amulet 3.3%, WATCHMAN 4.5%). There was a finding of increased procedure-related complications with Amulet (4.5% vs 2.5%, p=0.02), comprised primarily of pericardial effusion requiring intervention. Following this IDE trial, Amulet was approved in 2021 by the FDA with a label for 6 months DAPT post procedure.
There is also emerging data on the safety and efficacy of post-implant regimens of DOAC alone, or half-dose DOAC.(16) The prior NCDR analysis of WATCHMAN 2.5 suggested DOAC alone carried the lowest bleeding rates when compared to OAC/aspirin and DAPT.(12) A prospective, 3-center study used half-dose DOAC with aspirin for the first 45 days post-implant, followed by long-term half-dose DOAC and found lower rates of both DRT and bleeding compared to the standard regimen (DOAC/aspirin followed by DAPT).(16) This is reminiscent of data from ACTIVE W in 2006 showing less bleeding with warfarin compared to DAPT in patients treated medically for stroke reduction in AF who had previously tolerated warfarin.(17)
Our study found a significantly higher risk of bleeding in patients treated with DOAC/aspirin compared to DAPT. The NCDR study mentioned previously showed lowest bleeding risk with a DOAC-only regimen that excluded aspirin.(12,13) A small, recent multicenter study comparing outcomes in 592 patients after LAAO managed with DOAC alone or DAPT, matched by a propensity score, showed a nonsignificant lower rate of bleeding with DOAC compared to DAPT.(18) More information is needed regarding DOAC alone vs DAPT in the early post LAAO period. The available data, including the current study, support DAPT as an acceptable early post-LAAO drug regimen.
Questions also remain about the timing of post-procedure imaging to detect DRT, which were previously based on a period when it was felt that OAC/aspirin could be transitioned to DAPT due to endothelization. Timing varies greatly for post-procedure imaging, from 45-day imaging as recommended in the FDA-approved Instructions for Use, to the 4 months outlined in the CHAMPION-AF protocol. CHAMPION-AF is examining the role of LAAO with WATCHMAN FLX in patients without high bleeding risk, and allowed for DAPT, DOAC/aspirin, or DOAC alone in the post-procedure protocol. Similarly, the CATALYST trial, examining the role of the Amulet device for LAAO in low-risk patients, is utilizing DAPT as the post-implant regimen. Simplifying the post-implant regimen thus opens discussion about optimal timing, or need for, post-implant imaging for rare DRT and peri-device leak. Following completion of these trials, new data will be available on rates of DRT and embolic events in a variety of post-procedure medication regimens and follow-up periods for imaging which may help inform this discussion.
Decision making regarding stroke prevention in AF remains highly preference-sensitive, and thus appropriate for a shared decision-making (SDM) process. SDM is currently mandated for patients being considered for LAAO who are Medicare beneficiaries and must be performed by a non-implanting physician.(19) A SDM process includes transparency of the choices available, including the option of no medication or procedure, the risks and benefits of the available choices with an evidence-based decision aid, and incorporation of informed patient values and goals in final decision.(19,20). The FDA has now approved additional post-LAAO medical therapy options to include DAPT for 6 months following WATCHMAN FLX. Ongoing research, including evaluating additional post-implant regimens such as DOAC alone, will be important for maintaining a well-informed SDM process between patients and their clinicians. (21)
Limitations
The use of a variety of post-procedure medication regimens in the LAAO registry demonstrates individualization of the balance of risk between DRT and stroke vs bleeding despite the FDA-approved indications for OAC in the first 45 days post-procedure at the time of data collection. Although the NCDR LAAO Registry uses auditing and adjudication processes as quality control measures, (1) we cannot rule out the possibility of underreporting and misclassification though the impact should be similar across medication groups. Furthermore, DRT was not adjudicated; imaging data was incomplete at 45 days as described; and 6 month imaging was even more rare (11% of DAPT group, 9% of DOAC/aspirin group and 8% of warfarin/aspirin group). Given limited imaging, DRT at 6 months not reported.
Propensity matching cannot adjust for all differences inherent in the selection process by physicians. In addition, the percentage of patients selected for DAPT post-procedure remained small. This study looked at 45-day and 6-month outcomes given the timing of transition from OAC/aspirin to DAPT at 45 days with the existing label. Studies with longer follow-up may be helpful in determining the optimal post-LAAO drug strategy.
Conclusions
In this analysis of a large, mandated national registry, no differences in either the composite endpoint of death, stroke, bleeding, or systemic embolism, or DRT, were detected between DAPT and DOAC/aspirin or warfarin/aspirin at 45 days following LAAO with WATCHMAN FLX. A significantly lower risk of bleeding was noted when DAPT was prescribed post-LAAO compared with DOAC/aspirin. These data also demonstrate that great variability exists for post-procedure medications following LAAO, illustrating the challenge of balancing the rare risk of DRT-related stroke with bleeding following LAAO.
Supplementary Material
Figure 1. Study flow diagram.
Abbreviations: DAPT = dual antiplatelet therapy, DOAC = direct oral anticoagulant
Clinical Perspectives.
What’s known?
Prior clinical trials have utilized oral anticoagulation (OAC) with aspirin in the 45 days following left atrial appendage occlusion (LAAO) and provide the basis of comparison. Recent clinical trials for the Amulet device utilized dual-antiplatelet therapy (DAPT) in the post-implant period with similar clinical outcomes.
What’s new?
This large registry analysis shows similar clinical outcomes for patients managed with DAPT compared to OAC with aspirin after WATCHMAN FLX. Patients were selected for their post-implant regimen based local physician assessment of their comorbid illnesses and clinical assessment of bleeding risk, and not in a randomized fashion, leading to a limitation of unmeasured confounders.
What’s next?
Given off-label use of direct oral anticoagulation alone, further research is needed to determine the optimal strategy of antithrombotic therapy post LAAO. In addition, patient engagement research is needed to understand patient preferences, as prior literature has revealed distinct goals and values between patients and their clinicians related to anticoagulation for stroke prevention in atrial fibrillation.
Acknowledgements
Funding/Support
This analysis was supported by the Boston Scientific Corporation. The authors thank and Kristine Roy, PhD for writing/editing assistance, Jennifer L. Williams for assistance with design and conduct of the study, and Hong Wang, MS for assistance with statistical analysis (paid employees of Boston Scientific Corporation).
Abbreviations list
- ACC
American College of Cardiology
- LAAO
left atrial appendage occlusion
- NCDR
National Cardiovascular Data Registry
- AF
atrial fibrillation
- OAC
oral anticoagulation
- DAPT
dual antiplatelet therapy
- DOAC
direct oral anticoagulant
Footnotes
Disclosures
The views expressed here represent those of the authors, and do not necessarily represent the official views of the American College of Cardiology Foundation’s National Cardiovascular Data Registry (NCDR) or its associated professional societies identified at CVQuality.ACC.org/NCDR. The authors interpreted the data and had final control over manuscript content. The lead authors (MC, DN) had full access to the analyzed data and attest to the integrity and accuracy of the data. All authors reviewed and approved the manuscript. The authors disclose the following conflicts of interest related to this manuscript. MC: Honoraria from Edwards LifeSciences, Medtronic, Boston Scientific, Occlutech. JCH: Honoraria from Medtronic, Abbott, Boston Scientific, Biotronik, Janssen Pharmaceuticals, Bristol-Myers Squibb, Pfizer, Sanofi, Zoll Medical, iRhythm, Acutus Medical, Galvanize Therapeutics, and Biosense-Webster, research grants from Biotronik and Biosense-Webster, and equity interest in Vektor Medical. JVF: Research funding from the NIH/NHLBI and the American College of Cardiology National Cardiovascular Data Registry and consulting/advisory board fees from Medtronic, Boston Scientific, Biosense Webster, PaceMate, and equity in PaceMate. RWY: Research funding and consulting fees from Abbott Vascular, Boston Scientific, and Medtronic, and research funding from Bard, Cook, and Philips. JPP: Supported by R01AG074185 from the National Institutes of Aging; receives grants for clinical research from Abbott, the American Heart Association, the Association for the Advancement of Medical Instrumentation, Bayer, Boston Scientific, iRhythm, and Philips and serves as a consultant to Abbott, Abbvie, ARCA biopharma, Bayer, Boston Scientific, Bristol Myers Squibb (Myokardia), Element Science, Itamar Medical, LivaNova, Medtronic, Milestone, ElectroPhysiology Frontiers, ReCor, Sanofi, Philips, and Up-to-Date. MJP: Consulting honoraria, speaker’s fees, and proctoring fees from Abbott Vascular and Boston Scientific, consulting honoraria from W.L. Gore, Baylis Medical, Biotronik, and Philips, consulting honoraria and speaker’s fees from Medtronic, consulting honoraria from Biosense Webster and Shockwave, and equity in Indian Wells, Inc. DJA and TC: Full-time employees and stockholders of Boston Scientific. All other authors report no relationships relevant to the contents of this paper to disclose.
References
- 1.Cresti A, García-Fernández MA, Sievert H, et al. Prevalence of extra-appendage thrombosis in non-valvular atrial fibrillation and atrial flutter in patients undergoing cardioversion: a large transoesophageal echo study. EuroIntervention 2019;15(3):e225–30. 10.4244/EIJ-D-19-00128. [DOI] [PubMed] [Google Scholar]
- 2.Ball J, Carrington MJ, McMurray JJV, Stewart S Atrial fibrillation: Profile and burden of an evolving epidemic in the 21st century. Int J Cardiol 2013;167(5):1807–24. 10.1016/j.ijcard.2012.12.093. [DOI] [PubMed] [Google Scholar]
- 3.Hart RG., Pearce LA, Aguilar MI Meta-analysis: Antithrombotic Therapy to Prevent Stroke in Patients Who Have Nonvalvular Atrial Fibrillation. Ann Intern Med 2007;146(12):857–67. 10.7326/0003-4819-146-12-200706190-00007. [DOI] [PubMed] [Google Scholar]
- 4.Cimmino G, Loffredo FS, Gallinoro E, et al. Percutaneous Left Atrial Appendage Occlusion: An Emerging Option in Patients with Atrial Fibrillation at High Risk of Bleeding. Medicina (Mex) 2021;57(5):444. 10.3390/medicina57050444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Daimee UA., Wang Y, Masoudi FA, et al. Indications for Left Atrial Appendage Occlusion in the United States and Associated In-Hospital Outcomes: Results From the NCDR LAAO Registry. Circ Cardiovasc Qual Outcomes 2022;15(8):e008418. 10.1161/CIRCOUTCOMES.121.008418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Calabrò P, Gragnano F, Niccoli G, et al. Antithrombotic Therapy in Patients Undergoing Transcatheter Interventions for Structural Heart Disease. Circulation 2021;144(16):1323–43. 10.1161/CIRCULATIONAHA.121.054305. [DOI] [PubMed] [Google Scholar]
- 7.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 non-inferiority trial. Lancet 2009;374(9689):534–42. 10.1016/S0140-6736(09)61343-X. [DOI] [PubMed] [Google Scholar]
- 8.Holmes DR., Doshi SK, Kar S, et al. Left Atrial Appendage Closure as an Alternative to Warfarin for Stroke Prevention in Atrial Fibrillation: A Patient-Level Meta-Analysis. J Am Coll Cardiol 2015;65(24):2614–23. 10.1016/j.jacc.2015.04.025. [DOI] [PubMed] [Google Scholar]
- 9.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):1–12. 10.1016/j.jacc.2014.04.029. [DOI] [PubMed] [Google Scholar]
- 10.WATCHMAN FLX [Instructions for use]. Marlborough, MA: Boston Scientific Corporation; 2023. [Google Scholar]
- 11.Boersma LVA., Hüseyin I, Kische Stephan, et al. Evaluating Real-World Clinical Outcomes in Atrial Fibrillation Patients Receiving the WATCHMAN Left Atrial Appendage Closure Technology. Circ Arrhythm Electrophysiol 2019;12(4):e006841. 10.1161/CIRCEP.118.006841. [DOI] [PubMed] [Google Scholar]
- 12.Freeman JV., Higgins AY, Wang Y, et al. Antithrombotic Therapy After Left Atrial Appendage Occlusion in Patients With Atrial Fibrillation. J Am Coll Cardiol 2022;79(18):1785–98. 10.1016/j.jacc.2022.02.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lakkireddy D, Thaler D, Ellis CR, et al. AmplatzerTM amuletTM left atrial appendage occluder versus watchmanTM device for stroke prophylaxis (amulet ide): a randomized controlled trial. Circulation n.d;0(0). 10.1161/CIRCULATIONAHA.121.057063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.U.S. Food and Drug Administration AmplatzerTM AmuletTM Left Atrial Appendage Occluder PMA P200049 Approval Letter. Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P200049. Accessed June 7, 2023.
- 15.Freeman JV, Varosy P, Price MJ, et al. The NCDR Left Atrial Appendage Occlusion Registry. J Am Coll Cardiol 2020;75(13):1503–18. 10.1016/j.jacc.2019.12.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Della RDG., Magnocavallo M, Di BL, et al. Half-Dose Direct Oral Anticoagulation Versus Standard Antithrombotic Therapy After Left Atrial Appendage Occlusion. JACC Cardiovasc Interv 2021;14(21):2353–64. 10.1016/j.jcin.2021.07.031. [DOI] [PubMed] [Google Scholar]
- 17.ACTIVE W writing group. 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(9526):1903–12. 10.1016/S0140-6736(06)68845–4. [DOI] [PubMed] [Google Scholar]
- 18.Faroux L, Cruz-González I, Arzamendi D, et al. Short-term direct oral anticoagulation or dual antiplatelet therapy following left atrial appendage closure in patients with relative contraindications to chronic anticoagulation therapy. Int J Cardiol 2021;333:77–82. 10.1016/j.ijcard.2021.02.054. [DOI] [PubMed] [Google Scholar]
- 19.Knoepke CE., Allen LA, Kramer DB, Matlock DD Medicare Mandates for Shared Decision Making in Cardiovascular Device Placement. Circ Cardiovasc Qual Outcomes 2019;12(7):e004899. 10.1161/CIRCOUTCOMES.119.004899. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Barry MJ., Edgman-Levitan S. Shared Decision Making — The Pinnacle of Patient-Centered Care. N Engl J Med 2012;366(9):780–1. 10.1056/NEJMp1109283. [DOI] [PubMed] [Google Scholar]
- 21.Coylewright M, Holmes DR Caution Regarding Government-Mandated Shared Decision Making for Patients With Atrial Fibrillation. Circulation 2017;135(23):2211–3. 10.1161/CIRCULATIONAHA.117.026285. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.