Sex-specific Long-term Outcomes of Watchman Left Atrial Appendage Closure for Stroke Prevention in Atrial Fibrillation

Abstract

Background: A recent analysis of a large registry showed differences in periprocedural outcomes of the Watchman left atrial appendage closure device in males compared with females. The objective of our study was to investigate the 5-year event rate in males and females enrolled in the Watchman device premarket clinical studies submitted for US Food and Drug Administration review.

Methods: We conducted a patient-level meta-analysis of 2256 patients from 4 studies: the PROTECT AF (Embolic Protection in Patients with Atrial Fibrillation) and PREVAIL (Prospective Randomized Evaluation of the Watchman Left Atrial Appendage Closure Device in Patients with Atrial Fibrillation vs Long-Term Warfarin Therapy) randomized controlled trials and their continued-access registries-CAP1 (Continued Access to PROTECT AF) and CAP2 (Continued Access to PREVAIL). The outcomes evaluated were ischemic stroke (IS), IS/systemic embolism, hemorrhagic stroke (HS), and all-cause mortality. Mixed-effects Cox regression models and statistical testing for treatment-by-sex interaction were used to compare left atrial appendage closure vs warfarin in males and females. Hazard ratios adjusted (aHRs) for CHADS₂ scores were generated using the same model with CHADS₂ score as a covariate. Time-to-event end points were evaluated using the Kaplan-Meier method and log-rank test.

Results: For Watchman vs warfarin in the 2 randomized controlled trials, there was no significant interaction between sex and treatment for IS, IS/systemic embolism, HS, and all-cause mortality (P > .05); both males and females in the Watchman group had a lower aHR for HS than that in the warfarin group, which was statistically significant for males (aHR, 0.163; 95% CI, 0.045-0.593). In addition, there were no differences in outcomes between females and males treated with the Watchman device when pooling all studies.

Conclusions: These data suggest that sex does not significantly affect the long-term safety and effectiveness of the Watchman device in patients with nonvalvular atrial fibrillation; however, further studies are needed.

Keywords: Watchman; atrial fibrillation; females; left atrial appendage closure; meta-analysis.

Graphical abstract

Figure 1

CHADS₂score distribution in males (blue) and females (red) forall 4 studies. CAP1, Continued Access to PROTECT AF; CAP2, Continued Access to PREVAIL; PREVAIL, Prospective Randomized Evaluation of the Watchman Left Atrial Appendage Closure Device in Patients With Atrial Fibrillation Versus Long-Term Warfarin Therapy; PROTECT AF, Embolic Protection in Patients with Atrial Fibrillation.

Central Illustration

Forest plots for sex-differences analyses for each end point in individual and combined 2 RCTs. PREVAIL, Prospective Randomized Evaluation of the Watchman Left Atrial Appendage Closure Device in Patients With Atrial Fibrillation Versus Long-Term Warfarin Therapy; PROTECT AF, Embolic Protection in Patients with Atrial Fibrillation; RCT, randomized controlled trial.

Figure 2

Kaplan-Meier curves for 3 clinical outcomes in all 4 studies (N = 2256). For each end point, the curves, number of patients at risk, and event-free probabilities are provided and compared using the log-rank test. The P values are not adjusted for multiplicity. The dashed lines correspond to 1874 patients treated with the Watchman device in all 4 pooled trials (device group), 1228 males (blue dashed line) and 646 females (red dashed line). The solid lines correspond to 382 patients treated with warfarin (control group) in the 2 randomized controlled trials (RCTs), 274 males (blue, solid line) and 108 females (red, solid line). (A) The log-rank test results for ischemic stroke/systemic embolism: device group in 4 pooled studies, male vs female: P = .0598; control group in 2 RCTs, male vs female: P = .2163. P values of the device vs. control are .1732 in males and .5306 in females. (B) The log-rank test results for hemorrhagic stroke: device group in all 4 trials, male vs female: P = .8084; control group in 2 RCTs, male vs female: P = .6866. P values of the device vs control are <.0001 for males and .0327 for females. (C) The log-rank test results for all-cause mortality: dvice group in all 4 trials, male vs female: P = .6793; control group in 2 RCTs, male vs female: P = .8976. P values of the device vs control are .1763 in the male group and .5770 in the female group.

Figure 3

Sex differences in Watchman patients only for each end point from pooled analyses. Adjusted hazard ratios (aHRs) (adjusted for CHADS₂ score; hazard ratio = female/male) and 95% CIs via mixed-effects Cox model for the 2 combined randomized controlled trial (RCTs) and for the 2 RCTs combined with the corresponding registries (all 4 studies). HS, hemorrhagic stroke; IS, ischemic stroke; SE, systemic embolism.