DISCUSSION
Optimal antithrombotic therapy after LAAC with the Watchman occluder remains uncertain. This study aimed to investigate the efficacy and safety of a 3-month dual antiplatelet therapy (DAPT) after LAAC, as well as the risk factors that could affect the net clinical benefit and bleeding events. The results strongly suggest that in patients who underwent LAAC with the Watchman occluder, 3-month DAPT is associated with fewer bleeding events.
A suitable bridging antithrombotic protocol is essential to prevent thromboembolic complications during the endothelialization period of LAAC devices. The strategy of ACT adopted in the PROTECT-AF and PREVAIL trials 9, 10 has the most solid scientific evidence for effectiveness, but it limited to patients eligible for OAC. In contrast to clinical trials, the bridging protocols are very heterogeneous in everyday clinical practice 8. The most commonly used strategy, although it is only based on observational evidence, is DAPT directly after LAAC for a limited period (6 weeks to 6 months), followed by antiplatelet monotherapy, and the regimen prescribed in the PROTECT-AF and PREVAIL trials was less adopted8.
In the present post hoc study of 524 patients who underwent successful LAAC with the Watchman occluder and with 12 months of follow-up, 3-month DAPT regimen was compared with the regimen of 45-day ACT plus aspirin followed by 4.5-month DACT (totally 6 months) after LAAC. Despite the fact that the investigated population in the DAPT group was a high-risk patient group (mean age, 78.4±7.2; CHA2DS2-VASc, 4.8±1.5; HAS-BLED, 3.3±0.7), no significant difference was found in primary efficacy and safety outcomes and net clinical benefit events between the two type of antithrombotic regimens. The less aggressive bridging antithrombotic regimen with DAPT did not reduce the net clinical benefits after 12 months. Age ≥75 years was independently associated with an increased risk of decreased clinical net benefits while bridging antithrombotic therapy was not a significant risk factor after multivariable analysis. Those findings confirm the results of the EWOLUTION registry and another propensity-matched study based on several Watchman registries, which demonstrated similar efficacy and safety of both ACT and DAPT regimens11, 12.
Before device endothelialization, DRT and ischemic stroke are dreaded complications. On the other hand, bleeding events, due to anticoagulation, play a substantial role after LAAC. Of note, the rate of the overall bleeding events was significantly higher in the ACT group than in the DAPT group, although no differences in MB during follow-up were observed. More patients under the ACT treatment suffered from CRNMB). We found that the ACT regimen was an independent risk factor for bleeding events. Despite differences between studies, whether about the characteristics or definitions of outcomes, prior history of bleeding as well as elderly people >75 years were identified as independent risk factors for bleeding events 7. In the present study, there was only a trend toward an association between these two factors with bleeding events. This is likely due to the low numbers of patients. Consistent with the EWOLUTION trial, in both treatment arms, all fatal and most of the bleeding events occurred during the most aggressive phase of antithrombotic therapy12, 14.
Less aggressive antithrombotic postprocedural LAAC therapies would be appealing, potentially causing fewer bleeding complications, which, to this date, is the most common and challenging complication after LAAC25. The current Watchman instructions for use recommend an antithrombotic regimen tailored to each patient’s individual stroke and bleeding risk, allowing either a 3-months DAPT or an OAC. Several studies investigated various antithrombotic schemes following LAAC. The real-world ASAP registry evaluated a 6-months DAPT in 150 patients undergoing LAAC with the Watchman and reported a low annual rate of ischemic stroke of 1.7% 26. Another all-comer single-center registry documented the same rate of ischemic stroke (1.7%) with a short-term DAPT for 6 weeks following LAAC with Amplatzer and Watchman devices in 298 patients 7. Furthermore, it revealed a low annual major bleeding rate of 3.9%. Two registries with 110 and 76 patients investigated a SAPT after LAAC with Amplatzer devices 27, 28. They reported a slightly higher stroke rate of 2.3% and 4.0%, as well as a very low bleeding rate of 2.1% and 1.3%. Lastly, the 3-months results of the EWOLUTION registry found no relevant differences in DAPT, VKA, NOAC, SAPT, or no therapy with regard to effectiveness and safety. Interestingly, NOAC therapy had the lowest event rate, numerically 14.
The annual rates of ischemic stroke and MB in the present study were similar for both treatment arms while slightly higher than the aforementioned registries. This could partially be explained by higher risks for ischemic stroke and MBs, as depicted by the relatively high CHA2DS2-VASc (4.8±1.5) and HAS-BLED scores (3.3±0.7), and a shorter follow-up duration with most events occurring in the early phase after device implantation. In the ACT group, MBs, including three fatal events, were observed during early follow-up and occurred predominantly in patients with a high HAS-BLED score and prior bleeding events. This observation was also made in the PROTECT and PREVAIL studies, which used the ACT regimen and documented a relatively high estimated annual bleeding rate of 10.5% 9, 10. In the long term, at five years, the overall annual risk of MB was reduced to 3.1%29.
The rate of major periprocedural complications (3.4%) in the present study was in line with other studies (PREVAIL study: 4.2%9, CAP registry: 4.1% 30, EWOLUTION registry: 2.7% 31). TEE at follow-up revealed a low rate of major peri-device leaks for both groups, which is comparable to that reported in the EWOLUTION registry (0.7%) 31. In the ACT group, one case of TIA was documented in a patient with a major peri-device leak. An association of peri-device leaks and the occurrence of thromboembolic events has not been found yet 32. The rate of DRT did not differ between the treatment arms. In contrast to this finding, the propensity-matched analysis of ACT and DAPT observed more DRT in the DAPT group 11. All patients with major peri-device leaks and DRT were switched to OAC. In the ACT group, two ischemic events were associated with DRT during follow-up. Nevertheless, neither peri-device leak nor DRT was associated with clinical adverse events during follow-up in the DAPT group. This is likely a chance finding due to the small sample size of the study, low event rate, and different frequency for TEE follow-up. The impact of DRT on ischemic stroke is well documented in several studies32-34.
All-cause mortality in the DAPT group was considerable and higher than in the ACT group. This is probably due to the slightly higher age, rates of coronary heart disease and a significantly poorer left ventricular function in the DAPT group compared with the ACT group. Nonetheless, the DAPT subgroup of the EWOLUTION registry also reported a death rate of 10%, which reflects the elderly, fragile, and multimorbid LAAC patient population in European countries. Individual patient’s aspects like comorbidities, quality of life, and residual life expectancy should be taken into account when considering LAAC in octogenarians.