• Real-World Watchman LAAO Patients at Higher Risk of Death vs. Pivotal Trials’ Populations – Registry Study

    ‘Excess mortality’ may be related to increased burden of noncardiovascular death

    Atrial fibrillation (AF) patients in real-world clinical practice who underwent left atrial appendage occlusion (LAAO) with the Watchman device were at higher risk of dying than patients in the two pivotal trials used for its U.S. Food and Drug Administration (FDA) approval.

    However, the real-world cohort experienced fewer periprocedural complications and had a lower risk of ischemic stroke and a similar risk of hemorrhagic stroke.

    The raised mortality risk for real-world patients in the 425 days after implant may be related to an increased burden of noncardiovascular death, “but this warrants further study,” said the authors, led by Daniel J. Friedman, MD, from the Yale University School of Medicine and Duke University School of Medicine.

    The findings were published online Monday ahead of the May 9 issue of JACC: Cardiovascular Interventions.

    The two trials used for FDA approval of Watchman were PROTECT-AF and PREVAIL. Both trials randomized warfarin-eligible patients to receive Watchman LAAO or long-term warfarin therapy, but PREVAIL enrolled patients at higher risk of ischemic stroke and mandated the use of newer operators to scrutinize periprocedural safety.

    PROTECT-AF demonstrated that LAAO was noninferior to warfarin with respect to all-cause stroke, cardiovascular or unexplained death and systemic embolism, but – despite showing similar rates against the same endpoints – PREVAIL was unable to prove noninferiority due to a “lower than expected event rate” in the warfarin arm, the researchers said.

    “After extensive Food and Drug Administration (FDA) scrutiny, the Watchman LAAO device was approved for patients who could tolerate short-term warfarin but had a compelling reason to avoid long-term anticoagulation,” they added.

    Watchman LAAO has since been rapidly adopted by an increasing number of operators in “populations that are likely to carry a significant burden of coexisting illnesses,” the researchers said, stressing that this “requires scrutiny, particularly considering the periprocedural risks observed in the clinical trials”.

    The current propensity-matched analyses therefore, set out to compare LAAO in real-world clinical practice – based on the National Cardiovascular Data Registry’s (NCDR) LAAO Registry – with pivotal trial usage of both the device regimen and those randomized to receive warfarin. It matched up to three registry patients to each trial patient, and used Cox proportional hazards and Fine-Gray models.


    A total of 1,904 registry patients were matched to 667 trial LAAO patients, and 1,010 registry patients were matched to 348 warfarin patients.

    Compared with trial patients, registry patients were older (mean age 76.3 years vs. 72 years in PROTECT-AF and 74.3 years in PREVAIL; P < 0.0001), had higher CHA2DS2-VASc score (registry: 4.8 ± 1.5; PROTECT-AF: 3.5 ± 1.5; and PREVAIL: 4 ± 1.2; P < 0.0001), and were more likely to be women (43% vs. 30%).

    “Most comorbidities were more common among eligible registry patients, including diabetes, congestive heart failure, prior ischemic stroke, and peripheral vascular disease,” said the researchers. “A transient ischemic attack or stroke (driven by a history of transient ischemic attacks) was more common among trial patients.”

    LAAO trial patients had more pericardial effusion requiring intervention than registry patients (3.8% vs. 0.6%; P < 0.001), as well as higher periprocedural ischemic stroke rates (0.9% vs. 0.2%; P = 0.005), and more failed device implantation (7.5% vs. 3.6%; P < 0.001).

    The 425-day risk of ischemic stroke in the LAAO trial patients was also higher than in registry patients (2.7% vs. 1.21%; hazard ratio [HR]: 1.951 [95% confidence interval (CI): 1.05-3.62]; P = 0.03). Warfarin trial patients had comparable rates of ischemic stroke to registry patients (1.15% vs. 1.29%; HR: 0.728 [95% CI: 0.24-2.20]; P = 0.57).

    Hemorrhagic stroke, meanwhile, was similar for LAAO trial and registry patients (P = 0.88), but was greater in warfarin trial patients (1.44% vs. 0.2% in registry; HR: 5.871 [95% CI: 1.179-29.245]; P = 0.03).

    Nevertheless, death rates were lower in LAAO trial patients than registry patients (2.92% vs. 6.23%; HR: 0.477 [95% CI: 0.29-0.79]; P = 0.004).

    The researchers said the difference was attributable to noncardiovascular deaths. In a competing-risk assessment for cardiovascular/unknown cause of death vs. noncardiovascular death at 425 days, the researchers reported that 2.15% of registry deaths were cardiovascular/unknown compared to 1.2% of the matched LAAO trial patients (P = 0.06).

    Overall mortality was similar (P = 0.44) among warfarin trial (4.48%) and registry (5.86%) patients.

    “These results suggest improving safety and efficacy of Watchman LAAO in clinical practice and highlight the value of a national registry in understanding the adoption of technology and assessing procedural outcomes in contemporary practice,” they concluded.

    The researchers stressed that sponsors/funders – including Watchman’s manufacturer Boston Scientific – had no role in the design and conduct of the study. Several of the authors disclosed affiliations with the company.

    Uptick for LAAO

    In an accompanying editorial, Stephen A. Tang, MD, and Shephal K. Doshi, MD, from the Pacific Heart Institute, Santa Monica, California, cited “exciting additions” in the last 2 years to the clinical armamentarium for LAAO, including the Watchman FLX device with implant success rates of 98.8%, and the Amulet device (Abbott) with similar rates of 98.9%.

    “Propelled by the pandemic with limitation on hospital beds, there has been renewed interest in facilitating same day discharge by eliminating the need for transesophageal echo and general anesthesia and instead using intracardiac echocardiography guidance for LAAO,” they noted.

    “Commercial release of 4-dimensional intracardiac echocardiography may provide enabling technology for less experienced operators. As we modify device selection choice and implant technique, the NCDR will remain vital in providing real world assessment of our progress with some limitations.”


    Friedman DJ, Du C, Wang Y, et al. Patient-Level Analysis of Watchman Left Atrial Appendage Occlusion in Practice Versus Clinical Trials. JACC: Cardiovasc Interv 2022;15:950-961.

    Tang SA, Doshi SK. Left Atrial Appendage Occlusion: Practice Makes Perfect? JACC: Cardiovasc Interv 2022;15:962-964.

    Image Credit: Damian – stock.adobe.com

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Review our Privacy Policy for more details