Impact of Peridevice Leak on 5-Year Outcomes After Left Atrial Appendage Closure

Srinivas R. Dukkipati, MD; David R. Holmes, JR, MD; Shephal K. Doshi, MD; Saibal Kar, MD; Sheldon M. Singh, MD; Douglas Gibson, MD; Matthew J. Price, MD; Andrea Natale, MD; Moussa Mansour, MD; Horst Sievert, MD; Vicki M. Houle, PHD; Dominic J. Allocco, MD; Vivek Y. Reddy, MD

Disclosures

J Am Coll Cardiol. 2022;80(5):469-483.

Abstract and Introduction

Abstract

Background: In the U.S. Food and Drug Administration (FDA) clinical trials of left atrial appendage (LAA) closure, a postimplantation peridevice leak (PDL) of ≤5 mm (PDL≤5) was accepted as sufficient LAA "closure." However, the clinical consequences of these PDLs on subsequent thromboembolism are poorly characterized.

Objectives: We sought to assess the impact of PDL≤5 on clinical outcomes after implantation of the Watchman device.

Methods: Using combined data from the FDA studies PROTECT-AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients With Atrial Fibrillation), PREVAIL (Evaluation of the Watchman Left Atrial Appendage Closure Device in Patients With Atrial Fibrillation vs Long Term Warfarin Therapy), and CAP2 (Continued Access to PREVAIL), we assessed patients with successful device implantation for PDL by means of protocol-mandated transesophageal echocardiograms (TEEs) at 45 days and 1 year. Five-year outcomes were assessed as a function of the absence or presence of PDL≤5.

Results: The cohort included 1,054 patients: mean age 74 ± 8.3 years, 65% male, and CHA₂DS₂-VASc 4.1 ± 1.4. TEE imaging at 45 days revealed 634 patients (60.2%) without and 404 (38.3%) with PDL≤5, and 1-year TEE revealed 704 patients (71.6%) without and 272 (27.7%) with PDL≤5. The presence of PDL≤5 at 1 year, but not at 45 days, was associated with an increased 5-year risk of ischemic stroke or systemic embolism (adjusted HR: 1.94; 95% CI: 1.15–3.29; P = 0.014), largely driven by an increase in nondisabling stroke (HR: 1.97; 95% CI: 1.03–3.78; P = 0.04), while disabling or fatal stroke rates were similar (HR: 0.69; 95% CI: 0.19–2.46; P = 0.56). PDL≤5 was not associated with an increased risk of cardiovascular or unexplained death (HR: 1.20; P = 0.45) or all-cause death (HR: 0.87; P = 0.42).

Conclusions: PDL≤5 at 1 year after percutaneous LAA closure with the Watchman device are associated with increased thromboembolism, driven by increased nondisabling stroke, but similar mortality. (Watchman Left Atrial Appendage System for Embolic Protection in Patients With Atrial Fibrillation [PROTECT-AF; NCT00129545]; Evaluation of the Watchman Left Atrial Appendage Closure Device in Patients With Atrial Fibrillation vs Long Term Warfarin Therapy [PREVAIL; NCT01182441]; Continued Access to PREVAIL [CAP2; NCT01760291])

Introduction

Percutaneous left atrial appendage (LAA) closure with the Watchman device has been shown to be a suitable alternative to warfarin for patients with nonvalvular atrial fibrillation (AF) who cannot tolerate long-term oral anticoagulation.^[1–7] Compared with warfarin, the Watchman device (Boston Scientific) provides similar overall stroke prevention with significantly lower rates of hemorrhagic stroke, major bleeding, and mortality.^[1]

Though complete sealing of the LAA is desirable, incomplete occlusion can occur in up to 32% of implants.^[8] This is in large part due to the anatomic mismatch that occurs between the device and the LAA ostium. The LAA morphology is highly variable, and the ostium is elliptical with variable lengths and diameters,^[9,10] whereas the Watchman device is circular and available in several fixed sizes. Although the Watchman 2.5 device has 10 individually articulating splines that allow some adaptation to the elliptical morphology, the result may be suboptimal sealing, resulting in a peridevice leak (PDL) that has the potential to allow blood to stagnate, form thrombus, and subsequently embolize. In the U.S. Food and Drug Administration (FDA) clinical trials of LAA closure, a PDL ≤5 mm (PDL≤5) was considered to be acceptable both at the time of implantation and at the 45-day transesophageal echocardiography (TEE) and allowed discontinuation of warfarin in favor of dual antiplatelet therapy per study protocols. Furthermore, in an initial analysis of PROTECT-AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients With Atrial Fibrillation), these small PDLs were not associated with ischemic stroke or systemic embolism in follow-up; however, the numbers of analyzed patients and overall event rates were small, decreasing the confidence of this finding.^[8] In contrast, the surgical literature shows that incomplete surgical ligation of the LAA is associated with stroke and systemic embolism even with PDL≤5.^[11]

Herein, using combined data from the PROTECT-AF, PREVAIL (Evaluation of the Watchman Left Atrial Appendage LAA Closure Device in Patients With Atrial Fibrillation vs Long Term Warfarin Therapy), and CAP2 (Continued Access to PREVAIL) studies, we assessed the significance of persistent PDL≤5 on clinical outcomes—including ischemic stroke and systemic embolism—in a large cohort of patients followed to 5 years.

1 2 3 4 5

Next Section

Table 1. Patient Demographic and Procedural Characteristics by Peridevice Leak (PDL) Status at 1 Year
	No PDL (n = 704)	PDL≤5 mm (n = 272)	P Value
Demographics
Age, y	73.8 ± 8.0	73.5 ± 8.6	0.627
Female/male	247 (35.1)/457 (64.9)	81 (29.8)/191 (70.2)	0.116
LVEF, %	56.3 ± 9.6	57.4 ± 9.7	0.106
AF subtype
Paroxysmal	371 (52.7)	111 (40.8)	0.001
Persistent	169 (24.0)	76 (27.9)	0.204
Permanent	131 (18.6)	77 (28.3)	0.001
Paced	27 (3.8)	5 (1.8)	0.116
Coronary artery disease	309 (43.9)	103 (37.9)	0.088
Previous myocardial infarction	103 (14.6)	33 (12.1)	0.312
CHA₂DS₂-VASc score	4.0 ± 1.4	3.9 ± 1.4	0.170
Congestive heart failure	188 (26.7)	57 (21.0)	0.063
Hypertension	642 (91.2)	247 (90.8)	0.850
Diabetes mellitus	209 (29.7)	95 (34.9)	0.113
History of TIA/stroke	174 (24.7)	71 (26.1)	0.654
History of ischemic stroke	95 (13.5)	49 (18.0)	0.074
Peripheral vascular disease	66 (9.4)	22 (8.1)	0.529
HAS-BLED score	1.9 ± 0.9	2.0 ± 1.0	0.074
Procedural characteristics
Maximum LAA ostium diameter, mm	20.9.0 ± 4.6	21.9 ± 5.5	0.006
Maximum LAA length, mm	29.1 ± 7.4	30.6 ± 6.7	0.004
Final device size, mm	25.7 ± 3.3	25.9 ± 3.2	0.456
Multiple lobes	303 (43.0)	133 (48.9)	0.099
Device recaptures	0.87 ± 1.2	1.19 ± 1.6	0.003
Full recaptures	0.41 ± 0.7	0.61 ± 1.0	0.002
Partial recaptures	0.46 ± 0.9	0.57 ± 1.0	0.105

Values are mean ± SD or n (%).
AF = atrial fibrillation; LAA = left atrial appendage; LVEF = left ventricular ejection fraction; TIA = transient ischemic attack.

Table 2. Relative Risk of the Clinical Endpoints for Patients With PDL≤5
Outcome	No PDL	PDL≤5	Unadjusted		Adjusted
Outcome	No PDL	PDL≤5	HR (95% CI)	P Value	HR (95% CI)	P Value
45-day TEE	634	404
Ischemic stroke/SE	38 (6.6)	34 (9.2)	1.42 (0.89–2.25)	0.141	1.36 (0.85–2.16)	0.196
CV/unexplained death	57 (10.3)	32 (8.8)	0.88 (0.57–1.35)	0.547	0.85 (0.54–1.32)	0.466
All-cause death	129 (21.5)	72 (18.5)	0.87 (0.65–1.16)	0.350	0.87 (0.64–1.17)	0.343
1-year TEE	704	272
Ischemic stroke/SE	33 (5.1)	24 (9.5)	1.92 (1.14–3.25)	0.015	1.94 (1.15–3.29)	0.014
CV/unexplained death	54 (8.6)	26 (10.1)	1.23 (0.77–1.97)	0.382	1.26 (0.78–2.03)	0.342
All-cause death	124 (18.3)	43 (16.2)	0.89 (0.36–1.26)	0.515	0.98 (0.69–1.40)	0.930

Values are n or n (%), unless otherwise indicated. HR adjustment was performed with the use of a Cox hazard regression model.
CV = cardiovascular; PDL = peridevice leak; PDL≤5 = peridevice leak ≤5 mm; SE = systemic embolism; TEE = transesophageal echocardiography.