Introduction

Thromboembolic complications are common and cause morbidity after neuroendovascular treatment for cerebral aneurysms.1 Antiplatelet therapy is necessary to prevent these complications, even with simple coiling.2 With recent advancements in neuroendovascular technology, the use of intra-arterial devices as endovascular treatment for cerebral aneurysms has gradually increased.3 Intra-arterial devices have a potential risk of thromboembolic complications,4 5 which is higher in patients undergoing stent-assisted coiling (SAC) or flow diversion (FD) than in those undergoing simple coiling6–8; therefore, SAC/FD necessitates intensive antiplatelet therapy consisting of two agents (dual antiplatelet therapy, DAPT).

Currently, the standard periprocedural antiplatelet therapy for SAC/FD is DAPT with aspirin and clopidogrel for at least 6 months, followed by monotherapy for up to 12 months after the procedure.9 10 Although aggressive antiplatelet therapy is effective in preventing ischemic complications, it can lead to bleeding complications. A meta-analysis of percutaneous coronary intervention (PCI) showed that prolonged DAPT was associated with increased bleeding complications without reducing ischemic complications.11 A meta-analysis comprising five randomized control trials for second-generation, drug-eluting stents showed that DAPT within 3 months was associated with fewer bleeding events and similar ischemic events compared with prolonged DAPT.12 Short-DAPT and de-escalation strategies are now becoming mainstream antiplatelet therapies for PCI.

However, the optimal regimen of antiplatelet therapy for SAC/FD remains unknown because current regimens do not have high-quality evidence to guide them.

Therefore, the optimal duration of antiplatelet therapy for SAC/FD, which can prevent both ischemic events associated with inadequate antiplatelet therapy and bleeding events associated with prolonged DAPT, is required.

In this retrospective, large multicenter cohort study, we evaluated the relationship between the duration of antiplatelet therapy and the occurrence of thromboembolic or bleeding events during long-term observation in patients with cerebral aneurysm treated with SAC/FD.

MethodsStudy population

The APT-ANswer study (antiplatelet therapy in endovascular treatment for intracranial aneurysms study; clinical trial registration: unique identifier: UMIN000044122, URL: https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000050384) is a retrospective, multicenter cohort study that enrolled consecutive patients who received SAC/FD for cerebral aneurysms in seven Japanese academic institutions between January 2010 and December 2020. The inclusion criteria were patients aged 20 years or older with cerebral aneurysms treated with SAC/FD. Patients who did not receive sufficient preoperative antiplatelet therapy (eg, emergency treatment for symptomatic aneurysm, emergency treatment for ruptured aneurysm less than 2 weeks after onset, and bleeding tendency due to comorbidities) were excluded. The study size was derived from the number of patients who underwent SAC/FD after 2010 when vascular remodeling stents were first approved in Japan and who completed at least 1 year's follow-up.

The institutional review boards approved the protocol in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan (No 2021–013). The requirement for written informed consent from patients was waived in this study because we used clinical information obtained in routine clinical practice.

Demographic and baseline clinical information, as well as postprocedural clinical data, were retrospectively extracted from medical records by each attending physician in participating centers and recorded on a web-based electronic data capture system. The following information was collected: patient demographics, aneurysm characteristics, details of the procedure, the regimen of periprocedural antiplatelet therapy, and the results of platelet function tests and complications. The present study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Complications and outcomes

The clinical events that occurred between the day of the procedure and the last follow-up in December 2021 were measured.

Thromboembolic events were identified by both clinical and radiological findings and included the following: target vessel-related thromboembolic events (defined as thrombus formation identified on digital subtraction angiography or thrombotic occlusion of the target vessel), ischemic stroke, transient ischemic attack related to the territory of the target vessel (major stroke was defined as National Institutes of Health Stroke Scale score >4 points or neurological symptoms lasting more than 7 days, and the others were defined as minor stroke), and ipsilateral ocular ischemia (central retinal artery occlusion or amaurosis fugax). Asymptomatic ischemic lesions were defined as newly developed high-intensity areas on magnetic resonance diffusion-weighted images obtained every few months or half yearly as routine follow-up examinations during the observation period. Data on ischemic stroke or transient ischemic attack unrelated to the territory of the target vessel and ischemic diseases other than cerebrovascular disease were also extracted.

Bleeding events were identified by clinical or laboratory findings and included the following events: procedure-unrelated major bleeding events according to the International Society on Thrombosis and Haemostasis criteria13 (fatal bleeding, and/or symptomatic bleeding in a critical area or organ, and/or bleeding causing a fall in hemoglobin level of ≥2 g/dL, or leading to transfusion of two or more units of whole blood or red cells), clinically relevant non-major bleeding requiring additional medical evaluation of clinical findings, laboratory data or radiographic findings, and procedure-related bleeding complications. Procedure-related bleeding complications were defined as those directly related to the procedure itself, such as vessel perforation or puncture-site hematoma requiring intervention by a medical professional. Details of deaths that occurred during the observation period, regardless of cause, were also extracted.

The primary outcome was the time from the day of the procedure to the development of a composite of (1) target vessel-related thromboembolic events, (2) procedure-unrelated major bleeding events, or (3) death. The secondary outcomes were target vessel-related thromboembolic events, procedure-unrelated major bleeding events, and death. The baseline dates were defined as the day of the procedure (day 0). The data of patients who were lost to follow-up or terminated the observation without any events were censored.

Statistical analysis

Patient characteristics were summarized using medians and interquartile ranges for continuous variables, whereas counts and proportions were used for categorical variables. We performed a quality check of the dataset and recoded entry errors as missing data; therefore, we did not have any missing values for this analysis. Kaplan–Meier curves were plotted for the overall number of patients, patients who underwent SAC, and those who underwent FD to confirm the association between each clinical outcome and the number of days since the start of the procedure. Using the date at which monotherapy was initiated as the landmark time point, the cumulative non-occurrence rate of each outcome was shown by Kaplan–Meier curves for the two groups (<90 days and >91 days of DAPT; the reason for this cut-off value is described in the Results section) and compared using the log-rank test.

Comparisons of the characteristics of the two groups were performed using the Mann–Whitney U test for continuous variables and the Χ2 test for categorical variables. A two-sided value of P <0.05 was considered significant. All statistical analyses were performed using R 4.2.2 (the R Foundation for Statistical Computing).

ResultsBaseline demographic characteristics

In total, 657 patients were registered in the database. Of these, 25 were confirmed to be ineligible by the data manager. The remaining 632 patients were included in the analysis (figure 1). The median patient age was 64 years, and 77.5% of the patients were female. The total number of treated aneurysms was 661, and 25 patients underwent simultaneous treatment for multiple aneurysms. Most aneurysms (416/661 aneurysm) were located in the internal carotid artery. Among the 632 aneurysms that were the primary target, the median aneurysm size was 8.0 mm, 75% were asymptomatic, and 87.2% were saccular (table 1). The details of previous treatments for the 55 patients with recurrent aneurysms were as follows: simple coiling, 37; SAC, 10; FD, 5; and clipping, 3.

Figure 1

Study flow charts. EDC, electronic data capture; FD, flow diverter; SAC, stent-assisted coiling.

Table 1

Patients' characteristics and procedural information

Treatment characteristics

In total, 423 (66.9%) patients were treated with SAC, and 209 (33.1%) were treated with FD. Single stents were used in most of the procedures (91.1%), and multiple stents were used in 8.9% of the procedures (table 1). Details of the multiple stenting technique are as follows: T or Y configuration, 9; telescope or overlapping, 44; and different targets, 3. The various types of stents used were Enterprise/Enterprise 2 (Johnson and Johnson, Miami, Florida, USA) in 118, Neuroform EZ/Atlas (Stryker, Kalamazoo, Michigan, USA) in 181, LVIS or LVIS Jr (Microvention, Tustin, California, USA) in 161, FRED (Microvention, Tustin, California, USA) in 64, Pipeline Flex (Medtronic, Dublin, Ireland) in 116, and Pipeline Shield (Medtronic, Dublin, Ireland), a more recent generation FD with surface modification, in 61 patients.

The majority of the patients (98.3%) received perioperative antiplatelet therapy as a combination of clopidogrel and aspirin, and the remaining received other combinations (table 1). The usual maintenance dose of antiplatelet agents in Japan is aspirin 100 mg/day, clopidogrel 75 mg/day, cilostazol 200 mg/day, and prasugrel 3.75 mg/day. In some patients, including hyporesponders, loading doses of the antiplatelets were administered (aspirin 200 mg in 37 patients, clopidogrel 300 mg in 15 patients, prasugrel 20 mg in 36 patients). Cilostazol 200 mg/day was administered temporarily as add-on management for clopidogrel hyporesponders, and in four patients it was used in the initial combination of DAPT.

Almost all patients underwent platelet function testing, and management for abnormal responders (the definition was dependent on the discretion of each institution or physician) was performed in 133 (21%) patients before the procedure (escalation for hyporesponders such as dosing up, changing to triple therapy or switching to prasugrel: 115 patients; de-escalation for hyper-responders such as dosing down or omitting: 18 patients) and in 41 (6.5%) after the procedure. The median duration of DAPT including triple therapy was 169 (IQR 100-215) days, and the duration of total antiplatelet therapy including monotherapy was 440 (IQR 306–695) days. The durations were longer in FD patients (DAPT for 198 days, 478 days in total) than in SAC patients (DAPT for 125 days, 413 days in total).

Clinical outcomes

The median duration of the observation period was 646 (386–1098) days. During the observation period, primary outcome most frequently occurred within 30 days after the procedure in 63 patients (10.0%; 42 of the 63 patients were receiving DAPT). Details of the events are presented in online supplemental table 1. Cumulative event-free survival rates at 30 days, 1 year, and 2 years after the procedure were 93.3% (91.4 to 95.3%), 91.5% (89.3 to 93.7%), and 89.5% (87.0 to 92.0%), respectively (figure 2A).

Figure 2

Kaplan–Meier curves demonstrating the rate of event-free survival in total patients. Primary outcome (A), target vessel-related thromboembolic events (B), procedure-unrelated major bleeding events (C), and death (D). FD, flow diverter; SAC, stent-assisted coiling.

Cumulative thromboembolic or bleeding event-free survival rates showed a similar trend (target vessel-related thromboembolic events, 95.1%, 93.9%, and 93%, figure 2B; procedure-unrelated major bleeding events, 98.3%, 97.4%, and 97.1%, respectively, figure 2C), but not death (99.5%, 99.5%, and 98.8%, respectively; figure 2D). These results were not significantly different between patients who underwent SAC and FD, or between multiple stent and single stent subgroups (online supplemental figure 1). The distribution of modified Rankin Scale scores of 0 or 1 at the last observation was 573 (90.7%) patients.

Optimal duration of DAPT

Procedure-unrelated major and minor bleeding events were observed continuously during the DAPT period for at least 1 year after the procedure (online supplemental figure 2). Shortening the DAPT duration was considered necessary to reduce bleeding events, but there was concern about an increase in thromboembolic events associated with early switching to monotherapy. To elucidate the optimal duration of DAPT, the primary outcome after switching to monotherapy was analyzed using the log-rank test in a population limited to patients who were able to complete any DAPT period without any events.

The log-relative hazard of thromboembolic events after switching to monotherapy decreased with the duration of DAPT, reaching zero at approximately 100 days of DAPT (not significant, online supplemental figure 3). From this result and a previous study on PCI,12 we dichotomized these patients depending on the duration of DAPT: >91 days defined as the long-DAPT group and <90 days defined as the short-DAPT group.

The cumulative event-free survival rate did not differ between the short- and long-DAPT groups (log-rank test, P=0.84), and also for target vessel-related thromboembolic events (P=0.28), procedure-unrelated major bleeding events (P=0.29), and death (P=0.16) (figure 3). In the subgroup analyses between patients who underwent FD and SAC, the cumulative event-free survival rate did not differ between the groups for primary outcome (P=0.18 for FD, P=0.53 for SAC), thromboembolic events (P=0.7 for FD, P=0.34 for SAC), and bleeding events (P=1 for FD, P=0.24 for SAC), except for death in the FD group (P=0.03, higher in the short-DAPT group).

Figure 3

Comparison of the rate of event-free survival after switching to monotherapy. The analysis was performed in a population limited to patients who switched from DAPT to monotherapy, without major clinical events. All (A), SAC (B), and FD (C) patients. DAPT, dual antiplatelet therapy; FD, flow diverter; SAC, stent-assisted coiling.

In univariate analyses, differences in the duration of DAPT were seen in baseline characteristics such as sex, body mass index, aneurysm size, type of stent, and oral anticoagulants (online supplemental table 2). Multivariable Cox regression analysis was not performed because the number of events was too small to be analyzed with guaranteed reproducibility.

Discussion

This large multicenter cohort study evaluated the relationship between the duration of antiplatelet therapy after SAC/FD and clinical outcomes. The median duration of DAPT (mostly in combination with clopidogrel 75 mg and aspirin 100 mg) was 5 months, and subsequent monotherapy was continued for 1 year after the procedure. During the observational period, both thromboembolic and bleeding events mainly occurred within 30 days after the procedure and at approximately 1.5% per year thereafter, and the cumulative events-free survival rate was not different between the long- and short-DAPT groups. Although the duration was managed at the physician’s discretion according to the risk of each patient or procedure, these findings suggest that a DAPT duration of <90 days after SAC/FD may be sufficient to prevent major clinical events.

Aneurysm obliteration requires endothelialization along the stent to be isolated from the systemic circulation. SAC/FD carry the risk of thromboembolic complications until this process is completed because platelets activated by the metal surface lead to in situ thrombosis; therefore, more potent antiplatelet therapy is necessary for patients undergoing FD. The process of aneurysm healing is different between FD and SAC. SAC stents have substantially lower metal coverage than FD stents; hence, SAC needs intrasaccular coiling to promotes rapid intrasaccular thrombosis and subsequent neo-endothelialization of the aneurysm orifice. FD reduces flow inside the aneurysm by covering the aneurysm orifice with a higher metal coverage mesh stent, which acts as a scaffolding for neo-endothelialization of the aneurysm orifice. Complete aneurysm occlusion largely relies on adequate endothelialization in FD; this slow healing process is another reason for the need for longer antiplatelet therapy after FD. In this study, the duration of DAPT and subsequent monotherapy were longer for FD than for SAC, which may be a possible confounder in this study.

P2Y12 receptor inhibitors, such as clopidogrel, play a central role in the prevention of ischemic events in patients with atherosclerotic vascular diseases.14 They reduce major cardiovascular events in patients with the acute coronary syndrome (ACS) who undergo PCI.15 For patients with cerebral aneurysms who underwent SAC/FD, the duration of postprocedural DAPT16 or discontinuation of clopidogrel17 was associated with ischemic complications.

However, such potent P2Y12 receptor inhibitors also carry the risk of bleeding complications.18 19 Prolonged DAPT may be more hazardous in patients with cerebral aneurysms because they are non-arteriosclerotic. Since antiplatelet therapy for PCI has been changed to a de-escalation strategy, antiplatelet therapy for SAC/FD may also change.

Another important point is that platelet reactivity to antiplatelet agents varies among individuals,20 and the relationship between perioperative ischemic or bleeding complications and platelet reactivity has been recognized. Delgado-Almandoz reported an association between P2Y12 reaction units and periprocedural complications in patients who underwent FD and showed that the optimal range of last recorded P2Y12 reaction unit (PRU) values is between 60 and 240.21 In a meta-analysis of FD, hyper-responders were associated with a greater incidence of bleeding events with an increased absolute risk of 12%, and hyporesponders were associated with a greater incidence of thromboembolic events with an absolute risk of 15%.22 Platelet function tests were widely used before the procedure,23 but they are rarely used for long-term monitoring.

In our study, 93.5% of patients underwent a preprocedural platelet function test, and 21.3% of patients received preprocedural management of abnormal responders based on the results. Among the 133 patients who underwent preprocedural management, 115 (85.7%) underwent escalation management as hyporesponders. On the other hand, although only 41 (6.5%) patients received postprocedural management, a majority of them, 26(63.4%), underwent de-escalation management as hyper-responders. Long-term observation of platelet function is also important to detect delayed conversion to hyper-responders.24 A recent meta-analysis showed racial differences in the ischemia–bleeding trade-off during antiplatelet therapy for PCI.25 Especially for East Asians, long-term DAPT showed no benefit in reducing ischemic events, but increased bleeding events. We should take care of platelet function in antiplatelet therapy before the procedure and also in the long term.

Recently developed surface-modification techniques, such as phosphatidylcholine coating,26 glycan-based polymer coating,27and fibrin-based nano-coating,28 have been developed to reduce the thrombogenicity of FD. Although it is not yet a common procedure and indications should be restricted, these surface modifications may reduce the need for antiplatelet therapy. Clinical studies on surface-medicated FD implantation under single antiplatelet therapy with potent P2Y12 receptor inhibitors, such as prasugrel29 or ticagrelor,30 have begun to demonstrate its efficacy and safety. These results suggest that the duration of P2Y12 receptor inhibitor administration may be more important than the duration of DAPT, but further investigation is necessary.

Limitations

This study had a few limitations. First, it had a retrospective design, and the selection of endovascular procedures and antiplatelet therapy depended on each physician’s discretion, based on the characteristics of the aneurysm and the individual patient’s potential risk of both thrombosis and bleeding. Second, most patients underwent preprocedural platelet function tests, and subsequent drug management was performed in 21.3% of the patients with abnormal responses. Appropriate management can influence clinical outcomes. Third, comparisons between the long- and short-DAPT groups were limited to patients who survived and completed the DAPT period without events because of inevitable confounding and selection biases, such as a longer duration of DAPT for patients with thromboembolic events and a shorter duration for patients with major bleeding. The impact on these excluded patients with events during DAPT is unknown. Fourth, multivariable Cox regression analysis was not performed to adjust for confounding factors in the comparison between the groups, but the small number of events in this study precluded analysis with guaranteed reproducibility. Fifth, the only available recent generation FD with surface modification in Japan is Pipeline Shield (Medtronic, Dublin, Ireland), which was used in only 30% of patients in the FD group. Intrasaccular flow disruptors such as the Woven EndoBridge (Microvention, Tustin, California, USA), a new alternative device minimizing parent vessel metal coverage and the risk of thromboembolic events, were not included in this study. The results might have been different if more patients with these newer devices had been included.

Finally, we systematically registered patients with cerebral aneurysms who underwent SAC/FD at the seven centers, and this study is the largest study to date. However, the sample size was not sufficiently large for deductive conclusions to be drawn, and there was no racial diversity. Prospective randomized studies are warranted to determine the optimal duration of DAPT. This study provides a statistical basis for further studies.

Conclusions

This large multicenter cohort study evaluated the duration of antiplatelet therapy in patients with cerebral aneurysms treated with SAC/FD and long-term clinical outcomes. Thromboembolic events rarely occurred more than 30 days after SAC or FD, but bleeding events were observed throughout the DAPT period.

The incidence of thromboembolic events after switching to monotherapy was similar for the >91 days and <90 days DAPT groups in a univariate analysis. Although a lack of multivariate analysis and possible confounding and selection biases are major limitations, our study suggests that the duration of DAPT could be shortened to 90 days or less.

Further prospective studies such as randomized controlled trials of short versus long DAPT are warranted to determine the optimal duration of antiplatelet therapy.

The benefit of antiplatelet therapy to reduce thromboembolic events must be balanced with the corresponding risk of bleeding complications.