Comparative Data of Procedural and Midterm Outcomes in Patients Who Underwent Percutaneous Left Atrial Appendage Closure Between the WATCHMAN FLX and WATCHMAN 2.5 Devices　― Insight From the OCEAN-LAAC Registry ―

Abstract

Background: Limited data are available regarding clinical outcomes after percutaneous left atrial appendage closure using WATCHMAN FLX (WM-FLX) and WATCHMAN-2.5 (WM2.5) devices in Asian patients.

Methods and Results: Data of 1,464 consecutive patients (WM-FLX, n=909; WM2.5, n=555) were extracted from a Japanese multicenter registry, and clinical data were compared between the 2 groups. No in-hospital deaths, periprocedural stroke, or device embolization occurred. Procedural success was significantly higher in the WM-FLX than WM2.5 group (95.8% vs. 91.9%; P=0.002) owing to the lower incidence of periprocedural pericardial effusion (0.55% vs. 1.8%; P=0.021). No significant differences in all-cause death, postprocedural stroke, and device-related thrombus were observed between the 2 groups. However, the cumulative bleeding rate at 1 year was substantially lower in the WM-FLX group (7.8% vs. 16.4%; P<0.001). Landmark analysis of bleeding events highlighted lower bleeding rates in the WM-FLX than WM2.5 group within the first 6 months (6.4% vs. 14.8%; P<0.001), with comparable bleeding rates over the 6- to 12-month period (1.5% vs. 3.2%, respectively; P=0.065).

Conclusions: This study demonstrated higher early safety and lower 1-year bleeding rates in the WM-FLX than WM2.5 group. The lower bleeding events with WM-FLX are likely due to multiple factors other than purely difference in devices, such as postprocedural drug regimen.

Left atrial appendage (LAA) closure (LAAC) has been developed as an alternative therapy to oral anticoagulant (OAC) therapy in patients with non-valvular atrial fibrillation (NVAF). The predate WATCHMAN 2.5 (WM2.5) device has been clinically available in Japan based on the results of SALUTE (A Study to Evaluate the Safety and Effectiveness of the LAAC Therapy Using BSJ003 W for Patients With Nonvalvular Atrial Fibrillation at Increased Risk of ThromboEmbolism in Japanese Medical Environment).^1,2 Subsequently, the next-generation WATCHMAN FLX (WM-FLX) system was approved, providing improvements over WM2.5, such as adaptability to various anatomies, a coating to reduce device-related thrombosis (DRT), and a “FLX ball” structure in the distal part of the device to prevent LAA injury. The recent PINNACLE FLX (Protection Against Embolism for Nonvalvular AF Patients: Investigational Device Evaluation of the Watchman FLX LAA Closure Technology) trial demonstrated the remarkable safety and efficacy of LAAC and a significant decrease in the rate of major adverse events, which was only 0.5%.³ The latest guidelines indicate that LAAC is a Class 2a indication for patients with NVAF with elevated thromboembolic risk and a contraindication to long-term OACs because of a non-reversible cause.⁴ Compared with Western populations, the East Asian population is prone to developing bleeding complications,⁵ suggesting that LAAC can be a more beneficial option. Although US real-world practice reported short-term procedural safety benefits of WM-FLX compared with WM2.5 after LAAC,⁶ data on the clinical advantages of WM-FLX over WM2.5, particularly in Asian patients of Japanese ethnicity, are scarce. Therefore, the aim of this study was to elucidate the periprocedural and mid-term clinical outcomes after LAAC for the WM-FLX compared with WM2.5 using large-scale Japanese multicenter registry data.

Methods

Study Design and Patient Population

The Optimized Catheter Valvular Intervention (OCEAN)-LAAC registry, which is registered with University Hospital Medical Information Network (UMIN) Clinical Trials Registry (ID: UMIN000038498), is an ongoing prospective investigator-initiated multicenter observational registry of patients with NVAF undergoing percutaneous LAAC in Japan. This registry involves 20 structural heart (SH) and electrophysiology (EP) specialist facilities in Japan. Each facility was classified based on the area of expertise of the responsible physician.

The data used in the present study are for patients who underwent LAAC between September 2019 and December 2022. During this period, the WM2.5 device was used until May 2021; after that, the WM-FLX device was clinically approved as a new-generation device. Periprocedural and early to mid-term clinical outcomes, including bleeding events, were compared between the WM-FLX and WM2.5 groups.

The ethics committees of all participating centers approved the study protocol, and all procedures were conducted in accordance with the Declaration of Helsinki. All patients provided informed consent for the intervention and prospective follow-up before being enrolled in the registry. Data were collected prospectively using a dedicated web-based database (Canon Medical Systems).

The indication for LAAC was defined as NVAF requiring OACs and a high bleeding risk that met at least 1 of the following criteria: (1) a HAS-BLED score ≥3; (2) multiple histories of traumas due to a fall; (3) a history of diffuse cerebral amyloid angiopathy; (4) taking both an OAC and dual antiplatelet therapy (DAPT) long term; and (5) a history of major bleeding of Bleeding Academic Research Consortium (BARC) Type 3 or more.⁷ The exclusion criteria for this analysis were: (1) thrombus in the atrium or appendage; (2) prohibitions against either transesophageal echocardiography (TEE) or catheter insertion because of anatomical reasons; and (3) contraindications for OACs or antiplatelet agents.

Outcome Measures

Device, technical, and procedural successes were compared between the 2 groups. Device success was defined as precise deployment and correct positioning of the device. Technical success was defined as the complete closure of the LAA without any device-related complications and no residual leaks greater than 5 mm on color Doppler according to TEE findings. Procedural success encompassed technical success without any procedure-related complications.⁸ For in-hospital data, adverse events, such as all-cause death, surgical conversion, pericardial effusion, device embolization, any periprocedural stroke, any bleeding, and other procedure-related complications (e.g., access-related complications, TEE-related complications, acute kidney injury, and iatrogenic atrial septal defect [ASD] requiring closure) were recorded. The need for iatrogenic ASD closure does not affect the determination of success. For follow-up data, all-cause death; pericardial effusion; device embolization; DRT; postprocedural stroke, including transient ischemic attack (TIA); systemic embolization; and any bleeding events were recorded. Pericardial effusion was subdivided into cardiac and non-cardiac tamponade. Bleeding events were categorized using the BARC criteria, into overall bleeding (BARC Type 2 or more), BARC Type 3 bleeding, and BARC Type 5 bleeding. Clinical follow-up was generally scheduled for hospital discharge and at 45–90 days, 180 days, and 1 year after LAAC.

Postprocedural Antithrombotic Therapy

According to the instructions for use and previous literature,^3,9 patients were prescribed an OAC in concomitant with single antiplatelet therapy (SAPT), namely low-dose aspirin (81–100 mg), after the procedure. If sufficient sealing of the LAA with a leak ≤5 mm was seen on Day 45 follow-up imaging, it was recommended that the patient converted to a DAPT regimen comprising clopidogrel (75 mg) and low-dose aspirin until the 6-month follow-up visit, followed by continued low-dose aspirin indefinitely. However, the specific postoperative antithrombotic therapy regimen was ultimately determined by the attending physician at each facility. The drug regimens were evaluated in the device success cohort. We defined the use of ≥2 antithrombotic drug regimens as a classical regimen, whereas a single-drug regimen was referred to as a de-escalated regimen.

Statistical Analysis

Continuous variables with a normal distribution are presented as the mean±SD. Categorical variables are expressed as numbers and percentages. Continuous variables were compared using Student’s t-test or the Mann-Whitney U test, depending on their distribution. Categorical variables were compared using the Chi-squared test or Fisher’s exact test. Univariate analyses were performed to evaluate associations between variables, using appropriate statistical methods based on data distribution and characteristics. A Kaplan-Meier estimate was used with a log-rank test to detect differences in time-to-event variables. Univariate and multivariate regression analyses were performed to examine associations with bleeding events among all participant cohorts, incorporating age, sex, and baseline clinical characteristics with a univariate P<0.05. Two-sided P<0.05 was considered statistically significant. Statistical analyses were performed using JMP (version 15.2.1; SAS Institute, Cary, NC, USA) and R (version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria).

Results

Study Population

In this study, the WM-FLX device was used in 909 patients and the WM2.5 device was used in 505 patients. The baseline characteristics of the study population are presented in Table 1. Overall, the mean age was 77.1±7.6 years, 68% of patients were male, the mean CHA₂DS₂-VASc score was 4.9±1.5, and the mean HAS-BLED score was 3.1±1.0. The WM-FLX group included a lower number of patients with a history of major bleeding, coronary artery disease, and previous open heart surgery, and had lower hemoglobin (all P<0.05). The WM2.5 group included a lower proportion of older patients, patients with a history of stroke, residual disabilities, and previous LAA thrombosis, and patients who had received catheter ablation (all P<0.05). The components of the CHA₂DS₂-VASc and HAS-BLED scores are presented in Supplementary Table 1. There was no significant difference in the CHA₂DS₂-VASc score between the 2 groups (P=0.15), but the HAS-BLED score was significantly higher in the WM2.5 group (P<0.001).

Table 1.

Baseline Characteristics

	Overall (n=1,464)	WM-FLX (n=909)	WM2.5 (n=555)	P value
Age (years)	77.1±7.6	77.6±7.2	76.4±8.2	0.003
Male sex	990 (68)	609 (67)	381 (69)	0.51
BMI (kg/m2)	23.3±3.9	23.2±3.9	23.5±3.9	0.11
mRS 0–2	1,348 (92)	826 (91)	522 (94)	0.027
Clinical frailty scale ≥4	464 (32)	292 (32)	172 (31)	0.65
CHA2DS2-VASc score	4.9±1.5	4.9±1.6	4.8±1.5	0.15
HAS-BLED score	3.1±1.0	3.0±1.1	3.2±1.0	<0.001
SBP (mmHg)	126±19	126±20	126±18	0.73
DBP (mmHg)	72±13	73±13	71±13	0.035
Heart rate (beats/min)	73±14	72±14	73±15	0.53
Rhythm				0.41
Sinus rhythm	468 (32)	298 (33)	170 (31)
AF/AFl/AT	894 (61)	545 (60)	349 (63)
Pacemaker rhythm	99 (6.8)	63 (6.9)	36 (6.5)
Type of AF				0.024
Paroxysmal	595 (40.6)	375 (41.3)	220 (39.6)
Persistent	285 (19.5)	194 (21.3)	91 (16.4)
Long standing	527 (36)	303 (33.3)	224 (40.4)
History of thrombus formation in LAA	69 (4.7)	53 (5.8)	16 (2.9)	0.010
Previous catheter ablation	353 (24)	245 (27)	108 (19)	0.001
Previous surgical ablation	9 (0.6)	3 (0.3)	6 (1.1)	0.075
Congestive heart failure	718 (49)	426 (47)	292 (53)	0.033
NYHA Class III or IV	78 (11)	62 (15)	16 (5)	<0.001
Coronary artery disease	651 (44.4)	374 (41.1)	277 (49.9)	0.001
Previous any open heart surgery	152 (10.4)	45 (8.3)	77 (14)	<0.001
Previous CABG	80 (5.5)	40 (4.4)	40 (7.2)	0.022
Previous TAVI	65 (4.4)	35 (3.9)	30 (5.4)	0.16
Cerebrovascular accident
TIA	65 (4.4)	35 (3.9)	30 (5.4)	0.16
Ischemic stroke	549 (37.5)	365 (40)	184 (33.2)	0.007
Hemorrhagic stroke	169 (12)	104 (11)	65 (12)	0.99
Thromboembolic event	364 (25)	208 (23)	156 (28)	0.025
History of major bleeding	851 (58)	502 (55)	349 (63)	0.004
Peripheral vascular disease	186 (13)	107 (12)	79 (14)	0.16
Carotid stenosis	73 (5.0)	45 (5.0)	28 (5.1)	0.94
Renal failureA	1,098 (75)	678 (75)	420 (76)	0.64
Hemodialysis	172 (12)	108 (12)	64 (12)	0.84
COPD	73 (5.0)	55 (6.1)	18 (3.2)	0.017
Hepatic disease	50 (3.4)	34 (3.7)	16 (2.9)	0.46
History of malignancy	272 (19)	162 (18)	110 (20)	0.34
LVEF (%)	59±11	59±11	58±12	0.052
Hemoglobin (g/dL)	12.3±2.1	12.4±2.0	12.1±2.2	0.015
Platelet count (×1,000/μL)	19±6.5	19±6.3	19±6.6	0.66
Albumin (g/dL)	3.9±0.5	3.8±0.5	3.9±0.4	0.52

Unless indicated otherwise, values are presented as the mean±SD or n (%). ^ARenal failure was defined as estimated glomerular filtration rate <60 mL/min/1.73 m². AF, atrial fibrillation; AFl, atrial flutter; AT, atrial tachycardia; BMI, body mass index; CABG, coronary artery bypass grafting; COPD, chronic obstructive pulmonary disease; DBP, diastolic blood pressure; LAA, left atrial appendage; LVEF, left ventricular ejection fraction; mRS, modified Rankin scale; NYHA, New York Heart Association; PCI, percutaneous coronary intervention; SBP, systolic blood pressure; TAVI, transcatheter aortic valve implantation; TIA, transient ischemic attack; WM2.5, WATCHMAN 2.5; WM-FLX, WATCHMAN FLX.

Procedural Characteristics

The procedural characteristics and intraprocedural TEE findings are summarized in Table 2. The anesthesia and procedural times were significantly shorter and contrast dosage was lower in the WM-FLX group (all P<0.05). Single-curve sheaths were more commonly used in the WM2.5 group, whereas double-curve sheaths were used in approximately 90% of patients in the WM-FLX group. The incidence of device exchange was significantly lower in the WM-FLX than WM2.5 group (14% vs. 18%; P=0.022). The diameters of the LAA orifice, measured at each echocardiographic angle, and the maximum diameter were shallower in the WM-FLX group. LAA depth was significantly lower in the WM-FLX group. In the WM-FLX group, 20-, 24-, 27-, 31-, and 35-mm devices were used in 2.1%, 10.8%, 25.2%, 30.3%, and 29.4% of the patients, respectively. In the WM2.5 group, 21-, 24-, 27-, 30-, and 33-mm devices were used in 3.2%, 11.4%, 23.8%, 24.9%, and 33.3% of the patients, respectively. Peridevice leakages were comparable between the 2 groups, with none exceeding 5 mm.

Table 2.

Procedural Characteristics, Prrocedural Outcomes and Adverse Events

	Overall (n=1,464)	WM-FLX (n=909)	WM2.5 (n=555)	P value
Procedural characteristics
Anesthesia
General	1,462 (99.9)	908 (99.9)	554 (99.8)
Local sedation	2 (0.1)	1 (0.1)	1 (0.2)
Intraprocedural TEE guidance	1,464 (100)	909 (100)	505 (100)	NA
Anesthesia time (min)	105±40	101±37	110±43	<0.001
Procedural time (min)	58±33	55±30	63±36	<0.001
Contrast use dosage (mL)	58±39	52±39	68±36	<0.001
Fluoroscopic duration (min)	15±11	15±11	15±12	0.23
Mean LA pressure (mmHg)	13±4.1	13±4.1	13±4.3	0.245
Access sheath at deployment				<0.001
Single curve	286 (20)	102 (11)	184 (34)
Double curve	1,160 (80)	797 (89)	363 (66)
No. sheaths used				0.68
1	1,402 (96)	867 (96)	535 (97)
2	50 (3.4)	34 (3.8)	16 (2.9)
3	5 (0.3)	3 (0.3)	2 (0.4)
Incidence of device exchange	221 (15)	122 (14)	99 (18)	0.022
No. devices used
1	1,232 (85)	780 (86)	452 (82)
2	169 (12)	100 (11)	69 (13)
3	42 (2.9)	16 (1.8)	26 (4.7)
≥4	10 (0.7)	6 (0.7)	4 (0.7)
Peridevice leak				0.12
None	1,316 (89.9)	825 (90.8)	491 (88.5)
≤3 mm	73 (5)	37 (4.1)	36 (6.5)
3–5 mm	11 (0.8)	5 (0.6)	6 (1.1)
≥5 mm	0 (0)	0 (0)	0 (0)
Prrocedural outcomes and adverse events
Device success	1,424 (97.3)	888 (97.7)	536 (96.6)	0.21
Technical success	1,421 (97.1)	887 (97.6)	534 (96.2)	0.13
Procedural success	1,381 (94.3)	871 (95.8)	510 (91.9)	0.0016
All-cause death	0	0	0	NA
Device embolization	0	0	0	NA
Any stroke	0	0	0	NA
Surgical conversion	4 (0.27)	1 (0.11)	3 (0.54)	0.13
Pericardial effusion	15 (1.02)	5 (0.55)	10 (1.8)	0.021
Cardiac tamponade	9 (0.61)	2 (0.22)	7 (1.26)	0.013
Non-cardiac tamponade	6 (0.41)	3 (0.33)	3 (0.54)	0.54
All bleeding	33 (2.25)	11 (1.21)	22 (3.96)	0.0006
Major bleeding	13 (0.89)	3 (0.33)	10 (1.8)	0.0036
Minor bleeding	20 (1.37)	8 (0.88)	12 (2.16)	0.040
Access site-related complications	11 (0.75)	6 (0.66)	5 (0.90)	0.61
TEE-associated complications	4 (0.27)	2 (0.22)	2 (0.36)	0.612
Acute kidney injury	5 (0.34)	0 (0)	5 (0.9)	0.0041
ASD requiring closure	2 (0.14)	1 (0.11)	1 (0.18)	0.72
Others	2 (0.14)	2 (0.22)	0	0.27

Unless indicated otherwise, values are presented as the mean±SD or n (%). ASD, atrial septal defect; LA, left atrial; TEE, transesophageal echocardiography. Other abbreviations as in Table 1.

Procedural Outcomes and Adverse Events

Procedural and clinical outcomes are also presented in Table 2. The procedural flow chart, including detailed numbers and reasons for failing to achieve procedural success, is shown in the Supplementary Figure. The device and technical success rates were comparable between the WM-FLX and WM2.5 groups (97.7% vs. 96.6% [P=0.21] and 97.6% vs. 96.2% [P=0.13], respectively). The procedural success rate was significantly higher in the WM-FLX than WM2.5 group (95.8% vs. 91.9%; P=0.002). There were no instances of all-cause death, stroke, or device embolization. Among the other adverse events, pericardial effusion and acute kidney injury occurred more frequently in the WM2.5 group (all P<0.05). Sixteen SH facilities treated 81.7% (743/909) of the WM-FLX group and 80.2% (445/555) of the WM2.5 group (P=0.460). The remaining 20% of patients were treated in 4 EP facilities. There were no significant differences between the SH and EP facilities in rates of device (97.2% vs. 97.5%, respectively), technical (96.2% vs. 97.5%, respectively), or procedural success (94.2% vs. 94.9%, respectively; all P>0.05).

In the WM-FLX group, device success was not achieved in 21 patients. In the WM2.5 group, device success was not achieved in 16 patients, predominantly because of the decision to abandon implantation. However, in the WM2.5 group, 3 patients required surgical conversion. Each patient had device-related cardiac tamponade, which led to emergency surgical LAA resection because hemostasis could not be achieved despite successful pericardiocentesis.

In the device success cohort, 1 patient in the WM-FLX group had pericardial effusion. Although the device was implanted successfully, pericardiocentesis was difficult, finally necessitating subxiphoid minithoracotomy for drainage and hemostasis. In the WM2.5 group, 2 patients had pericardial effusion. In 1 of these 2 patients, hemostasis was achieved through device placement and pericardial effusion was managed conservatively. Conversely, the second patient needed pericardiocentesis after device implantation.

In the WM-FLX group, the reasons for not attaining procedural success were access site injuries (n=6), pericardial effusions (n=4), TEE-related complications (n=4), any bleeding (n=3), and other (n=1). In the WM2.5 group, the reasons for not attaining procedural success were access site injuries (n=5), pericardial effusions (n=5), TEE-related complications (n=2), any bleeding (n=8), and others (n=5).

Postprocedural Antithrombotic Therapy

Changes in the drug regimen in each group are shown in Figure 1 and Supplementary Table 2. At baseline, 94.8% of patients in the WM-FLX group and 94.4% of patients in the WM2.5 group were prescribed an OAC. Patients in the WM-FLX group were most likely to be prescribed a direct oral anticoagulant (DOAC) as an OAC, whereas patients in the WM2.5 group were more likely to be prescribed a vitamin K antagonist as an OAC. A similar trend was observed in the choice of medication at the time of discharge. In terms of the antithrombotic regimens at discharge, the classical regimen was more often used in the WM2.5 group (50.9% vs. 75.9%), whereas the de-escalated regimen was more often used in the WM-FLX group (48.1% vs. 22.6%). At 3 months, the prevalence of SAPT (36.8% vs. 20.6%) or OAC alone (23.0% vs. 9.7%) was higher in the WM-FLX group, whereas the prevalence of DAPT (25.5% vs. 55.5 %) or OAC plus SAPT (10.4% vs. 11.9%) was higher in the WM2.5 group. At 1 year, 84.1% of patients in the WM-FLX group and 89.7% in the WM2.5 group had discontinued OAC therapy.

Figure 1.

Changes in antithrombotic drug regimens. Antithrombotic drugs administered preoperatively, at discharge, 3 months after left atrial appendage closure (LAAC), and 1 year after LAAC are shown for the WATCHMAN FLX (WM-FLX) and WATCHMAN 2.5 (WM2.5) groups. DAPT, dual antiplatelet therapy; OAC, oral anticoagulant; SAPT, single antiplatelet therapy.

Clinical Outcomes

To investigate postimplant performance, adverse events were compared in the device success cohort during the follow-up period (Table 3; Figures 2,3). Specifically, comparisons were made between 888 patients in the WM-FLX group and 536 patients in the WM2.5 group. The mean follow-up period was 307±138 days in the WM-FLX group and 583±323 days in the WM2.5 group. The incidence of all-cause mortality 1 year after LAAC was 5.1% in both groups (P=0.53). Furthermore, there were no significant differences in the cumulative incidence of cardiovascular death, non-cardiovascular death, the occurrence of DRT, postprocedural stroke, TIA, and pericardial effusion between the 2 groups (all P>0.05). Notably, bleeding complications differed significantly between the 2 groups. The overall cumulative bleeding rate at 1 year was substantially lower in the WM-FLX than WM2.5 group (7.8% vs. 16.4%; P<0.001). There were markedly fewer bleeding events in the WM-FLX than WM2.5 group within the early 6-month period after LAAC (6.4% vs. 14.8%; P<0.001). Landmark analysis revealed a trend towards a lower incidence of bleeding in the WM-FLX group from 6 to 12 months after LAAC, although the difference did not reach statistical significance (1.5% vs. 3.2%; P=0.065). BARC Type 3 bleeding was also significantly lower in the WM-FLX group (P<0.001). The frequency of BARC Type 5 bleeding was similar in the WM-FLX and WM2.5 groups (0.4% vs. 0.4%; P=0.76). Figure 4 shows the time distribution of all bleeding events and the postoperative antithrombotic therapy regimens. Bleeding was more common in the period of multidrug therapy within 6 months, whereas in the second half of the study period, relatively more bleeding occurred in patients on single agents or not receiving any antithrombotic therapy.

Table 3.

Adverse Event Rates at the 1-Year Mark for the Device Success Cohort

	WM-FLX (n=888)	WM2.5 (n=536)	P value
All-cause death	5.1 (3.4–6.8)	5.1 (3.2–7.0)	0.53
Cardiovascular death	1.8 (0.8–2.8)	2.1 (0.9–3.3)	0.80
Non-cardiovascular death	3.4 (2.0–4.7)	3.1 (1.6–4.6)	0.32
Pericardial effusion	0.8 (0.2–1.4)	1.9 (0.7–3.0)	0.073
Cardiac tamponade	0.3 (0–0.7)	1.1 (0.2–2.0)	0.075
Noncardiac tamponade	0.5 (0–0.9)	0.7 (0–1.5)	0.48
Stroke	3.5 (2.1–4.9)	2.0 (0.8–3.2)	0.18
Ischemic stroke	2.6 (1.4–3.9)	1.6 (0.5–2.7)	0.25
Hemorrhagic stroke	0.9 (0.4–1.5)	0.6 (0–1.2)	0.75
TIA	0.6 (0–1.2)	0 (0–0)	0.82
Systemic embolization	0.4 (0–1.0)	0.4 (0–1.0)	0.86
DRT	2.6 (1.4–3.8)	2.8 (1.3–4.2)	0.32
Overall bleeding	7.8 (5.9–9.7)	16.4 (13.1–19.5)	<0.0001
BARC Type 3 bleeding	3.4 (2.2–4.7)	9.0 (6.5–11.4)	<0.0001
BARC Type 5 bleeding	0.4 (0–0.8)	0.4 (0–0.9)	0.76

Data show 1-year event rates, estimated by Kaplan-Meier analysis, for each group, along with 95% confidence intervals in parentheses. P values are based on the log-rank test. BARC, Bleeding Academic Research Consortium; DRT, device-related thrombus. Other abbreviations as in Table 1.

Figure 2.

Kaplan-Meier curves for postoperative adverse events at 1 year in the WATCHMAN FLX (WM-FLX) and WATCHMAN 2.5 (WM2.5) groups, showing freedom from (A) all-cause death (P=0.53), (B) cardiovascular (CV) death (P=0.80), (C) ischemic stroke (P=0.25), and (D) all device-related thrombosis (DRT; P=0.32).

Figure 3.

Kaplan-Meier curves for postoperative bleeding events in the WATCHMAN FLX (WM-FLX) and WATCHMAN 2.5 (WM2.5) groups. (A,B) Freedom from overall bleeding (A) and Bleeding Academic Research Consortium (BARC) Type 3 bleeding (B) at the 1-year mark (P<0.001 for both). (C) Landmark analysis of the freedom from overall bleeding at 6 months after left atrial appendage closure. Within the early period, overall bleeding occurred more frequently in the WM2.5 than WM-FLX group (P<0.001), whereas overall bleeding rates were similar between the 2 groups in the second 6-month period after the procedure (P=0.065).

Figure 4.

Time distribution of all bleeding events and antithrombotic drug regimens at the bleeding event up to 1 year. We have categorized all bleeding events occurring within the first year postoperatively into 3 time periods: within 3 months, between 3 and 6 months, and from 6 months to 1 year after left atrial appendage closure. The figure shows the number of bleeding events within each period and provides a detailed breakdown of the anticoagulation therapy being administered at the time of each bleeding event. DAPT, dual antiplatelet therapy; OAC, oral anticoagulant; SAPT, single antiplatelet therapy.

Predictors of Bleeding Events

Univariate and multivariate regression analyses were performed to identify the factors that contribute to overall bleeding events after LAAC (Table 4). Univariate analysis identified the absence of previous catheter ablation, congestive heart failure, a history of major bleeding, malignancy, anemia, and hypoalbuminemia as predictors. Further multivariate analysis revealed that the absence of previous catheter ablation, a history of major bleeding, and baseline anemia were independent predictive factors for late bleeding events after LAAC.

Table 4.

Univariate and Multivariate Regression Analysis for Predicting Overall Bleeding Events

	Univariate analysis			Multivariate analysis
	OR	95% CI	P value	OR	95% CI	P value
Age	1.00	0.98–1.02	0.907	0.99	0.97–1. 02	0.24
Male sex	1.21	0.85–1.74	0.301	1.34	0.93–1.96	0.12
BMI	0.98	0.93–1.02	0.363	–
mRS 0–2	0.99	0.56–1.89	0.979
Hypertension	1.01	0.67–1.57	0.952
Diabetes	0.96	0.68–1.35	0.831
Clinical frailty scale ≥4	1.15	0.81–1.61	0.441
AF rhythm at procedure	1.04	0.75–1.47	0.803
History of thrombus formation in LAA	0.48	0.14–1.18	0.117
Previous catheter ablation	0.59	0.38–0.90	0.013	0.62	0.39–0.95	0.028
Congestive heart failure	1.52	1.10–2.12	0.012	1.38	0.98–1.96	0.063
Ischemic heart disease	1.07	0.77–1.49	0.674
Previous any open heart surgery	1.49	0.91–2.36	0.112
Ischemic stroke	0.60	0.42–0.86	0.0048	0.73	0.50–1.05	0.095
Hemorrhagic stroke	0.69	0.38–1.19	0.194
History of major bleeding	2.17	1.52–3.12	<0.0001	1.98	1.37–2.90	0.0002
History of malignancy	1.74	1.19–2.51	0.0050	1.00	0.67–1.49	NA
Renal failureA	1.32	0.90–2.01	0.164
Hemodialysis	0.99	0.57–1.61	0.969
Hepatic disease	1.31	0.53–2.79	0.526
Hemoglobin (<12 g/dL)	1.88	1.35–2.62	0.0002	1.65	1.15–2.37	0.0068
Platelet count (<150×1,000/μL)	1.34	0.93–1.90	0.113
Albumin (<3.5 g/dL)	1.82	1.23–2.65	0.0033	1.49	0.98–2.23	0.061
CHA2DS2-VASc score ≥5	0.77	0.56–1.07	0.121
HAS-BLED score ≥3	1.24	0.86–1.84	0.252

^ARenal failure was defined as estimated glomerular filtration rate <60 mL/min/1.73 m². CI, confidence interval; OR, odds ratio. Other abbreviations as in Table 1.

Discussion

We investigated the periprocedural and mid-term outcomes in patients who underwent LAAC using the newer-generation WM-FLX and classic WM2.5 devices from a large-scale Japanese cohort. The main findings of our study are outlined below.

First, periprocedural safety was significantly improved with the use of the WM-FLX compared with WM2.5 device, although the device and technical success rates were similar in both groups. This improvement in periprocedural safety was due primarily to a reduction in periprocedural pericardial effusion.

Second, the mid-term outcomes of all-cause death, postprocedural stroke, and occurrence of DRT were comparable between the 2 devices. However, the incidence of late bleeding events was significantly lower in the WM-FLX group. Notably, bleeding events more commonly developed in the WM2.5 group within the early 6 months following LAAC. During this period, a classical antithrombotic regimen was more commonly used in the WM2.5 group. When combined with the similar frequency of other adverse events, including stroke and DRT, the recent de-escalated drug regimen may be a beneficial choice after LAAC.

Safety Outcomes

When developing a stroke prevention plan for patients with NVAF, balancing the risks, including thromboembolism without OAC treatment, bleeding risks from long-term OAC therapy, potential complications from LAAC procedures, and device-related adverse events, including DRT, is crucial. Given that this treatment is preventive, ensuring the safety of the procedure must be given the highest priority. Although the rate of complications has been reported to decrease with accumulating clinical experience,¹⁰ there are concerns about injuries to the LAA leading to cardiac tamponade with the WM2.5.^11,12 Because of the atraumatic design of the WM-FLX device, the incidence of pericardial effusion decreased significantly. No pericardial effusions requiring intervention occurred before discharge or up to 7 days after the procedure in the PINNACLE FLX trial.³ A large retrospective registry indicated that there were fewer pericardial effusions requiring intervention with the WM-FLX than with the WM2.5 (0.42% vs. 1.23%).⁶ In the present study, the frequency of periprocedural pericardial effusion was as high as 1.8% (10/555) in the WM2.5 group, with 3 patients requiring emergency thoracotomies. However, this decreased to 0.44% (4/909) in the WM-FLX group, with only 1 patient requiring subxiphoid minithoracotomy for drainage. Because the anesthesia time, procedural time, and contrast amount were significantly reduced among the WM-FLX group, this could be attributed to the operator’s learning curve. However, it is reasonable to consider that the benefits of the device, as evidenced in previous studies,^3,6 have also significantly improved safety. The WM-FLX device has also provided improved echocardiographic visibility. In combination with enhanced safe maneuverability within the LAA, it greatly enhances the feasibility of contrast-free LAAC. In fact, the efficacy and safety of “zero-contrast” LAAC have been demonstrated by previous Japanese data.¹³

Efficacy Outcomes

Studies have demonstrated that the implant success rate using the WM2.5 device ranges from 90.9% to 99.5%,^9,14–16 with the success rate increasing over time.¹⁷ Initial Japanese findings were revealed in SALUTE, which was a prospective multicenter single-arm clinical trial involving 10 facilities and 42 participants assessing the safety and efficacy of the WM2.5 device in patients with NVAF. In that trial, the procedural outcome was excellent, with 100% device success, without perioperative all-cause mortality, stroke, or major bleeding.^1,2 Despite the accumulation of clinical experience, the WM-FLX group continued to experience device implant failures at a rate of 2.3%, which underscores the challenges in achieving device success. Notably, the LAA orifice dimensions in our cohort were larger than those reported in Western countries. In the present WM-FLX group, the 31-mm device was the most frequently used, accounting for 30.3% of cases, closely followed by the 35-mm device, which was used in 29.4% of patients. In contrast, Kar et al reported that the 27-mm device was the most frequently in their study, used in 31.1% of patients, with the 35-mm device used less frequently (7.8%).³ Although such differences in size and potential internal structural complexity may impair procedural success, further studies are warranted to identify the reason for device implant failure. The distribution of patients treated by SH and EP facilities did not differ in the WM-FLX and WM2.5 groups. However, most patients were treated in SH facilities. Patient selection was shifted to more elderly and high-risk stroke subsets in the SH facilities. Conversely, patients with a history of catheter ablation and bleeding were more prevalent in the EP facilities (data not shown). This unique patient selection bias may affect the treatment efficacy based on the respective areas of expertise. However, the clinical outcomes appeared to be similar in the SH and EP facilities.

Midterm Outcomes and Postoperative Antithrombotic Therapy

Postoperative antithrombotic therapy remains a major topic of debate, seeking the best balance between bleeding and thromboembolic events. In the WM-FLX group, a significant reduction in bleeding events was observed, particularly in the early 6-month period.

In a previous prospective trial conducted in the US, a classical regimen was prescribed after LAAC.^3,9 In the era of the WM2.5 device, the classical regimen was predominantly used. Our previous OCEAN-LAAC data demonstrated that the prevalence rate of major bleeding events during the 45-day follow-up was significantly higher among patients aged >80 years (6.9%).¹⁸ Considering the extremely high incidence of bleeding events within 45 days, a classical aggressive drug regimen may be harmful for elderly Japanese patients. This is consistent with reports indicating that bleeding is more frequently observed during the initial 6-month period when intensive antithrombotic therapy is required.¹⁹

In a US observational study,¹² only 12.2% adhered to this trial-based classical antithrombotic regimen. Notably, monotherapy with OAC was associated with a trend towards fewer adverse events than the combination of OAC with aspirin. Results from network analysis suggest that monotherapy with DOAC offers the most favorable balance between bleeding and thrombotic events.²⁰ These results led to a trend towards de-escalated OAC monotherapy as a postoperative regimen in the WM-FLX era. In addition, the Japanese single-center data demonstrated that various reduced-dose antithrombotic modified regimens were not associated with any increase in adverse events after LAAC.²¹ Based on these findings, reducing the dose of postoperative antithrombotic therapy should be considered based on an individual patient’s background. Although modifications to the antithrombotic regimen following LAAC are not recommended, optimal management concerning the postprocedural drug regimen should be established through further investigations.

In the present study, the WM 2.5 group included patients at higher bleeding risk. This is thought to be because LAAC was conducted with patients selected for their higher risk of bleeding. The differences in patient backgrounds, combined with the incorporation of recent evidence into postoperative management, are believed to have led to a reduction in bleeding in the WM-FLX group. Univariate and multivariate analyses identified the absence of prior ablation as a potential factor in reducing overall bleeding events. This could be attributed to confounding, because the WM-FLX group had more previous catheter ablations. In addition, there could be some selection bias, where those at high bleeding risk and intolerant to short-term OACs after ablation are typically precluded as candidates for catheter ablation. This should be considered when interpreting the results.

The incidence of DRT is one of the most important clinical concerns after LAAC. Although the WM-FLX was expected to reduce the occurrence of DRT due to the reduced exposure of the metallic hub, the incidence of DRT was similar in the 2 groups. Interpreting this result is too difficult because the duration of clinical follow-up in the present study differed significantly between the WM-FLX and WM2.5 groups. De-escalated regimens were more often used in the WM-FLX than WM2.5 group. In this context, evaluating the performance between devices is challenging. Patient data from further large-scale studies are needed to assess the previously investigated factors with potentially greater specific impacts on DRT, such as a history of LAA thrombosis or ischemic stroke and the presence of spontaneous echo contrast.²² In the midst of this chaos of postoperative antithrombotic therapy, there is also a growing interest in the potential of DRT reduction through the thromboresistant polymer coating of the next-generation WATCHMAN FLX pro device.²³

Study Limitations

This study has several limitations. First, this was a multicenter registry, and the determination of endpoints was left to the discretion of each participating facility. The possibility of underreporting adverse events cannot be denied. Second, the comparison between WM2.5 and WM-FLX occurred during a period when WM2.5 was in the nascent stages of introduction, whereas WM-FLX was already in a more advanced phase of dissemination for LAAC in Japan. Although this study evaluated clinical outcomes up to 1-year after LAAC, the observation period for the WM2.5 cohort was longer, including more data from chronic phase cases. Third, the limited number of DRTs detected in this registry restricts the accurate determination of the incidence, mechanism, and predictive factors after LAAC. The dynamic change in the drug regimen between the WM-FLX and WM2.5 groups was considered an unadjusted factor. Concerning DRT, more detailed information needs to be obtained through further investigations, including more patient data. Finally, detailed data on the dosage and administration of medications were lacking, meaning that it remains unclear whether the dosages were appropriate or inappropriately reduced, that is, the intended half-dose DOAC regimen.²⁴ Choosing an optimal drug regimen after LAAC is beyond the scope of the present study. Further clinical investigations are required to define the optimal medical management of patients after LAAC.

Conclusions

This study demonstrated that LAAC using the WM-FLX device provided superior safety, mainly because of reduced pericardial effusion and consistent efficacy in terms of all-cause death, stroke, and incidence of DRT compared with the previous WM2.5 device. The lower 1-year bleeding rate in the WM-FLX group may be related to multiple factors, such as the current de-escalated postprocedural drug regimen in Japanese real-world practice for patients who have undergone LAAC.

Acknowledgments

The authors thank the investigators and institutions that participated in the OCEAN-LAAC registry.

Sources of Funding

The OCEAN-LAAC registry, which is part of OCEAN-SHD registry, is supported by Edwards Lifesciences, Medtronic, Boston Scientific, Abbott Medical, and Daiichi-Sankyo Company.

Disclosures

M.N., M.Y., M.A., D.H., H.U., and S.K. are clinical proctors for Boston Scientific. M.Y. has received the lecture fees from Daiichi-Sankyo and Boston Scientific. M. Saji has received lecture fees from Daiichi-Sankyo and Abbott Medical Japan. K.H. is a member of Circulation Journal’s Editorial Team. The remaining authors have no relationships relevant to the contents of this paper to disclose.

IRB Information

The study protocol was reviewed and approved by the institutional review board of the lead facility, Keio University Hospital Institutional Review Board (Reference no. 20190194). In addition, approvals were obtained from the ethics committees of all other participating centers.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-24-0062

References

• 1.   Aonuma K, Yamasaki H, Nakamura M, Ootomo T, Takayama M, Ando K, et al. Percutaneous WATCHMAN left atrial appendage closure for Japanese patients with nonvalvular atrial fibrillation at increased risk of thromboembolism: First results from the SALUTE trial. Circ J 2018; 82: 2946–2953.
• 2.   Aonuma K, Yamasaki H, Nakamura M, Matsumoto T, Takayama M, Ando K, et al. Efficacy and safety of left atrial appendage closure with WATCHMAN in Japanese nonvalvular atrial fibrillation patients: Final 2-year follow-up outcome data from the SALUTE trial. Circ J 2020; 84: 1237–1243.
• 3.   Kar S, Doshi SK, Sadhu A, Horton R, Osorio J, Ellis C, et al. Primary outcome evaluation of a next-generation left atrial appendage closure device: Results from the PINNACLE FLX trial. Circulation 2021; 143: 1754–1762.
• 4.   Joglar JA, Chung MK, Armbruster AL, Benjamin EJ, Chyou JY, Cronin EM, et al. 2023 ACC/AHA/ACCP/HRS guideline for the diagnosis and management of atrial fibrillation: A report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2024; 149: e1–e156.
• 5.   Kim HK, Tantry US, Smith SC Jr, Jeong MH, Park SJ, Kim MH, et al. The East Asian paradox: An updated position statement on the challenges to the current antithrombotic strategy in patients with cardiovascular disease. Thromb Haemost 2021; 121: 422–432.
• 6.   Price MJ, Friedman DJ, Du C, Wang Y, Lin Z, Curtis JP, et al. Comparative safety of transcatheter LAAO with the first-generation watchman and next-generation watchman FLX devices. JACC Cardiovasc Interv 2022; 15: 2115–2123.
• 7.   Mehran R, Rao SV, Bhatt DL, Gibson CM, Caixeta A, Eikelboom J, et al. Standardized bleeding definitions for cardiovascular clinical trials: A consensus report from the Bleeding Academic Research Consortium. Circulation 2011; 123: 2736–2747.
• 8.   Tzikas A, Holmes DR Jr, Gafoor S, Ruiz CE, Blomström-Lundqvist C, Diener HC, et al. Percutaneous left atrial appendage occlusion: The Munich consensus document on definitions, endpoints, and data collection requirements for clinical studies. Europace 2017; 19: 4–15.
• 9.   Holmes DR Jr, Kar S, Price MJ, Whisenant B, Sievert H, Doshi SK, 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.
• 10.   Reddy VY, Gibson DN, Kar S, O’Neill W, Doshi SK, Horton RP, et al. Post-approval U.S. experience with left atrial appendage closure for stroke PREVENTION in atrial fibrillation. J Am Coll Cardiol 2017; 69: 253–261.
• 11.   Munir MB, Khan MZ, Darden D, Pasupula DK, Balla S, Han FT, et al. Pericardial effusion requiring intervention in patients undergoing percutaneous left atrial appendage occlusion: Prevalence, predictors, and associated in-hospital adverse events from 17,700 procedures in the United States. Heart Rhythm 2021; 18: 1508–1515.
• 12.   Price MJ, Valderrábano M, Zimmerman S, Friedman DJ, Kar S, Curtis JP, et al. Periprocedural pericardial effusion complicating transcatheter left atrial appendage occlusion: A report from the NCDR LAAO registry. Circ Cardiovasc Interv 2022; 15: e011718.
• 13.   Fukushima T, Fukunaga M, Isotani A, Nakamura M, Yamamoto K, Ishizu K, et al. Zero-contrast left atrial appendage closure in patients with chronic kidney disease. Circ J 2023; 88: 170–174.
• 14.   Reddy VY, Holmes D, Doshi SK, Neuzil P, Kar S. 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: 417–424.
• 15.   Reddy VY, Sievert H, Halperin J, Doshi SK, Buchbinder M, Neuzil P, et al. Percutaneous left atrial appendage closure vs warfarin for atrial fibrillation: A randomized clinical trial. JAMA 2014; 312: 1988–1998.
• 16.   Boersma LVA, Schmidt B, Betts TR, Sievert H, Tamburino C, Teiger E, 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: 2465–2474.
• 17.   Freeman JV, Varosy P, Price MJ, Slotwiner D, Kusumoto FM, Rammohan C, et al. The NCDR left atrial appendage occlusion registry. J Am Coll Cardiol 2020; 75: 1503–1518.
• 18.   Asami M, Naganuma T, Ohno Y, Tani T, Okamatsu H, Mizutani K, et al. Initial Japanese multicenter experience and age-related outcomes following left atrial appendage closure: The OCEAN-LAAC registry. JACC Asia 2023; 3: 272–284.
• 19.   Price MJ, Reddy VY, Valderrábano M, Halperin JL, Gibson DN, Gordon N, et al. Bleeding outcomes after left atrial appendage closure compared with long-term warfarin: A pooled, patient-level analysis of the WATCHMAN randomized trial experience. JACC Cardiovasc Interv 2015; 8: 1925–1932.
• 20.   Carvalho PEP, Gewehr DM, Miyawaki IA, Nogueira A, Félix N, Garot P, et al. Comparison of initial antithrombotic regimens after left atrial appendage occlusion: A systematic review and network meta-analysis. J Am Coll Cardiol 2023; 82: 1765–1773.
• 21.   Ryuzaki S, Kondo Y, Nakano M, Nakano M, Kajiyama T, Ito R, et al. Antithrombotic regimen after percutaneous left atrial appendage closure: A real-world study. Circ J 2023; 87: 1820–1827.
• 22.   Sedaghat A, Vij V, Al-Kassou B, Gloekler S, Galea R, Fürholz M, et al. Device-related thrombus after left atrial appendage closure: Data on thrombus characteristics, treatment strategies, and clinical outcomes from the EUROC-DRT-registry. Circ Cardiovasc Interv 2021; 14: e010195.
• 23.   Saliba WI, Kawai K, Sato Y, Kopesky E, Cheng Q, Ghosh SKB, et al. Enhanced thromboresistance and endothelialization of a novel fluoropolymer-coated left atrial appendage closure device. JACC Clin Electrophysiol 2023; 9: 1555–1567.
• 24.   Della Rocca DG, Magnocavallo M, Di Biase L, Mohanty S, Trivedi C, Tarantino N, et al. Half-dose direct oral anticoagulation versus standard antithrombotic therapy after left atrial appendage occlusion. JACC Cardiovasc Interv 2021; 14: 2353–2364.