The prospective randomised trial of the device therapy compared with warfarin (PROTECT AF trial) commenced in 2005 across sites in Europe and the USA. Patients with nonvalvular (i.e., excluding patients with rheumatic valvular disease) AF with a CHADS₂ score of 1 or more were randomised in a 2:1 ratio to WATCHMAN implant or continuation of warfarin (target INR 2.0–3.0). In patients receiving a device implant, warfarin therapy was discontinued after a 45-day follow-up transesophageal echocardiogram (TEE) if satisfactory criteria for LAA closure were demonstrated (residual peri-device flow <5 mm). The primary endpoints were of (a) efficacy—a composite occurrence of all-cause stroke (ischemic and hemorrhagic), systemic embolism, cardiovascular, and unexplained death; and (b) safety—a composite occurrence of excessive bleeding and procedure-related complications. A total of 707 patients were enrolled with a follow-up duration of 5 years. Clinical characteristics of the PROTECT AF subjects are shown in Table 10.1.

After a mean follow-up of 18 ± 10 months (1065 patient years), the WATCHMAN intervention group met non-inferiority criteria for the primary efficacy endpoint. The primary efficacy event rate was 3.0 per 100 patient years (95 % CI 1.9–4.5) in the WATCHMAN group and 4.9 per 100 patient years (95 % CI 2.8–7.1) in the warfarin group. The primary safety event rate was however significantly higher in the intervention group at 2 years, 10.2 % (95 % CI 7.4–13.0) than the warfarin control group 6.8 % (95 % CI 3.0–10.6). Expanded details of results are shown in Table 10.2.

Subsequent follow-up data on the cohorts was published for 2.3 years and 4.0 years [3, 4]. The primary efficacy event rate at 4 years of follow-up for the WATCHMAN group was 2.3 per 100 patient years (95 % CI 1.7–3.2) and 3.8 per 100 patient years (95 % CI 2.5–4.9) in the warfarin group. This suggested a 40 % relative risk reduction for all-cause stroke, systemic embolism, cardiovascular, and unexplained death in the WATCHMAN group with superiority over warfarin therapy demonstrated.

The safety event rate proved to be a significant hurdle in pursuing Food and Drug Administration (FDA) approval in the USA. However, regulatory approvals were followed in other regions of the world including Europe and Australasia, with commercialisation of the WATCHMAN device by Atritech Inc in 2009. Atritech Inc and the WATCHMAN device technology was subsequently acquired by Boston Scientific (Natick, Massachusetts) in 2011.

In the USA, the FDA permitted a Continued Access Program (CAP) nonrandomised registry for a subset of PROTECT AF investigators to continue implanting the WATCHMAN device according to study protocol to gain further safety and efficacy data on the device. The rate of procedure or device-related safety events within 7 days of the procedure declined from 7.7 % in PROTECT AF to 3.7 % in the CAP registry [5].

A further randomised control trial was still mandated by the FDA with a similar protocol, but requiring a minimum of 25 % enrolment by new operators to re-examine the safety issues. The PREVAIL study included a further 407 patients with a mean CHADS₂ score of 2.6 and was commenced in 2011. The required safety endpoint was reached with a 7-day safety event rate of 2.2 % (95 % upper CI 2.61) in the WATCHMAN group [6]. The primary efficacy composite endpoint, however, did not meet prespecified criteria for non-inferiority at the 18-month follow-up. The event rate was 1.07 per 100 patient years (95 % CI 0.57–1.88) in the WATCHMAN group and 0.7 per 100 patient years (95 % CI 0.1–5.1) in the warfarin group. Comparative event rates for warfarin control groups were significantly higher in other anticoagulation trials: PROTECT AF 1.6 % [1], RELY 1.7 % [7], ROCKET AF 2.2 % [8], and ARISTOTLE 1.6 % [9]. FDA approval of the WATCHMAN device in the USA was subsequently granted in March 2015 for use in patients with nonvalvular AF at high stroke risk who are suitable for warfarin, but who have an appropriate rationale to seek a non-pharmacologic alternative.

Additional data on the safety and efficacy of WATCHMAN LAA device closure on warfarin-ineligible patients came from the ASAP Study published in 2013 [10]. One hundred and fifty patients with contraindications to warfarin therapy were included in the prospective nonrandomised study. Following device implantation, patients were administered 6 months of clopidogrel or ticlopidine antiplatelet therapy in addition to lifelong low-dose aspirin. The primary efficacy (all-cause stroke, systemic embolism, cardiovascular, and all-cause death) event rate was 4.6 per 100 patient years. The ischemic stroke annual event rate was 1.7 %. Comparison was made with the expected annual ischemic stroke rate predicted by CHADS₂ score (mean score 2.8) while taking aspirin of 7.3 %, suggesting that WATCHMAN device therapy conferred a 77 % reduction in ischemic stroke rate.

Technical success of the implant procedure has a demonstrated learning curve over the WATCHMAN clinical studies. Implant success rates were 91 % in PROTECT AF, 94.7 % in ASAP, 95 % in CAP registry, and 95.1 % in PREVAIL studies.

The WATCHMAN device was designed and patented as a filter to prevent harmful-sized thrombi from exiting the LAA in patients with nonvalvular AF. Subsequent animal studies and post-mortem analysis have confirmed that full endothelialisation of the device atrial surface occurs generally over a 6-month period. The device is a self-expanding nitinol frame with a 160 μm permeable PET (polyethylene terephthalate) membrane cap. Ten active fixation anchors positioned at the distal third of the nitinol frame (composed of ten struts) help to achieve fixation and stability in the LAA tissue. The device is designed to be oversized for the LAA ostium so that radial force imparts stability and apposition with the LAA walls (Fig. 10.2). The device is available in 21, 24, 27, 30, and 33 mm diameters to accommodate individual variations in anatomy (Fig. 10.3). The device length shortens as it is deployed from the delivery catheter. The fully constrained device within the delivery catheter measures a similar length to the deployed maximum device diameter. The device is classified as magnetic resonance conditional according to the American Society for Testing and Materials . A patient with a WATCHMAN device can be safely scanned with magnetic resonance imaging immediately after implant with a static magnetic field of 3-T or less.