Article Text


Original research
Antiplatelet therapy and the risk of ischemic and hemorrhagic complications associated with Pipeline embolization of cerebral aneurysms: a systematic review and pooled analysis
  1. Hamidreza Saber1,
  2. Riyad Y Kherallah2,
  3. Mohamad O Hadied2,
  4. Shaghayegh Kazemlou3,
  5. Parthasarathi Chamiraju4,
  6. Sandra Narayanan1,4
  1. 1 Department of Neurology, Wayne State University School of Medicine, Detroit, Michigan, USA
  2. 2 Department of Medicine, Wayne State University School of Medicine, Detroit, Michigan, USA
  3. 3 Danaher Labs, San Jose, California, USA
  4. 4 Department of Neurosurgery, Wayne State University School of Medicine, Detroit, Michigan, USA
  1. Correspondence to Dr Hamidreza Saber, University Health Center, Detroit, MI 48201, USA; hsaber{at}


Background Pipeline embolization devices (PEDs) are increasingly used in the treatment of cerebral aneurysms. Yet, major ischemic or hemorrhagic complications after PED treatment associated with antiplatelet regimens are not well-established.

Objective To investigate the risk of ischemic and hemorrhagic complications associated with common antiplatelet regimens following PED treatment, and to examine whether platelet function testing (PFT) is associated with a lower risk of these complications.

Methods We searched Medline, Embase, and Cochrane from 2009 to 2017. Twenty-nine studies were included that had reported a uniform antiplatelet regimen protocol and had provided data on major ischemic and hemorrhagic complications following PED treatment. Random-effect meta-analysis was used to pool overall ischemic and hemorrhagic event rates across studies. The rate of these complications with respect to the antithrombotic regimen and PFT was assessed by χ2 proportional tests.

Results Overall, 2002 patients (age 55.9 years, 76% female) were included. A low-dose acetylsalicylic acid (ASA) regimen before and after PED treatment was associated with a higher rate of late ischemic complications than with high-dose ASA therapy (2.62 (95% CI 1.46 to 4.69) and 2.56 (1.41 to 4.64), respectively). Duration of post-procedure clopidogrel therapy <6 months was associated with greater rates of ischemic complications (1.56, 95% CI 1.11 to 2.20) than a clopidogrel regimen of ≥6 months. Performing PFT before PED treatment was not associated with the risk of ischemic complications (1.27, 95% CI 0.77 to 2.10).

Conclusion High-dose ASA therapy and clopidogrel treatment for at least 6 months were associated with a reduced incidence of ischemic events, without affecting the risk of hemorrhagic events.

  • aneurysm
  • stroke
  • flow diverter
View Full Text

Statistics from


Cerebral aneurysm treatment using flow diversion with a Pipeline embolization device (PED) is being employed for an increasing range of aneurysm sizes and shapes as an alternative to endovascular coiling or open surgical clipping.1 The PED works by providing a scaffold for endothelialization of the parent artery, diverting blood flow away from the aneurysmal sac and leading to clotting of the aneurysm. It is composed of 30–35% cobalt chromium/platinum tungsten, which can activate platelets and act as a nidus for thromboembolic complications,2 therefore requiring concomitant antiplatelet therapy. These treatments, on the other hand, pose a potential risk for hemorrhagic complications.3 Current evidence about optimal antiplatelet regimens is drawn from observational neurointerventional studies2–30 and randomized interventional cardiology trials and may not be fully generalizable to the cerebrovascular system. Although most interventionalists use a combination of acetylsalicylic acid (ASA) and clopidogrel before and after PED treatment, there is no standardized dose and duration.24

In addition, an important consideration relates to genetic polymorphisms in cytochrome P450 enzymes leading to varying patient responses to clopidogrel,24 with reported rates of clopidogrel resistance ranging between 16% and 50%. Whether platelet function testing (PFT) should be used before PED treatment to identify hyporesponders and adjust the antiplatelet regimen is contentious.31

In this study, we aimed to conduct a pooled analysis using a systematic methodology to investigate the optimal dose and duration of main antiplatelet agents (ASA and clopidogrel) for minimizing ischemic and hemorrhagic complications in patients undergoing PED treatment for cerebral aneurysms. We also assessed the role of PFT for improving outcomes among patients who undergo PED placement.


Study selection and eligibility criteria

A systematic review and meta-analysis was carried out in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A systematic electronic search was performed on Medline (PubMed), Embase, and Cochrane central registry of controlled trials to identify studies reporting complications associated with PEDs from January 2009 until June 2017. Two authors (HS, RYK) independently reviewed the literature, and screened the articles yielded by our search strategy. The following keywords were used in combination or individually: ‘flow diversion’, ‘Pipeline,’ and ‘Pipeline embolization device’. Exclusion criteria included case reports, case series with fewer than 10 patients, review articles, guidelines, technical notes, mortality and morbidity reports, articles in languages other than English, use of multiple flow diverters besides PED, unspecified antiplatelet duration, lack of uniform antiplatelet regimen protocol, and lack of follow-up data on complications. Both ruptured and unruptured aneurysms were included. Complications associated with technical failure during PED implementation were not included in the analysis. Cochrane’s risk of bias tool was used to assess the individual risk of bias of each study.

Outcome measures

Primary outcomes included PED treatment complications, categorized as ischemic and hemorrhagic events. Each complication was stratified as periprocedural (within ≤3 days), or late (>3 days). Ischemic events were defined as any symptomatic or asymptomatic complete or partial in-stent thrombosis, thromboembolic events, and cerebral infarcts associated with PED treatment. Hemorrhagic events were defined as subarachnoid or intracerebral hemorrhage. Asymptomatic perforator occlusions and complications caused by technical error were not included. Complications definitively related to non-compliance with the antiplatelet regimen were also excluded.

Exposure variables

The main exposure variables were antiplatelet regimen, categorized by type (ASA high dose vs low dose and clopidogrel) and length of treatment (<6 months vs ≥6 months), and the use of PFT (VerifyNow, Multiplate, Youden Index, Accumetrics). Low-dose ASA was defined as ≤150 mg and high-dose ASA as >150 mg. Patients with alternative medications—for example, prasugrel, ticagrelor, ticlopidine or abciximab, were not included in this analysis.

We examined the association between the antiplatelet regimen (before/after procedure) and PFT with the total rates of each of the studied complications. A sensitivity analysis was conducted with respect to outcomes, restricting our analysis to patients who received clopidogrel for <6 months and ≥6 months.

Statistical analysis

Random-effect meta-analysis was performed to obtain the overall estimate of ischemic and hemorrhagic complications across studies. This was calculated as the number of participants developing a complication as a proportion of the total number of participants undergoing Pipeline embolization in the category of interest. Standard errors and confidence intervals for a single proportion were also derived. Subsequently, meta-analysis of proportions was performed to estimate the pooled rates of ischemic and hemorrhagic complications. To compare the risk of outcomes in each category of antiplatelet therapy, percentages of each event (yes vs no) were computed and their differences were assessed by χ2 proportional tests. Event rate and 95% CI for each outcome were assessed according to the exposure variables. All analyses were implemented in R statistical software. P-values were two-tailed and P-values <0.05 were considered statistically significant.


A total of 1852 references were identified through three electronic database searches, of which 29 studies met our inclusion criteria (online supplemental figure 1). The study characteristics of these references are summarized in the online supplemental table 1.

Overall, our baseline studies comprised 2002 patients (with 2243 aneurysms, aged 55.9 years, 75.8% female, 249 posterior circulation aneurysms) for analysis of the antiplatelet regimen, and 1297 patients (with 1377 aneurysms, aged 57.4 years, 76.5% female) for analysis of PFT in relation to complications. Complications and follow-up length for each study are summarized in table 1. Periprocedural and late complication rates were available for 1297 patients who underwent PFT.

Table 1

Complications and follow-up length across eligible studies

Ischemic complication

Overall, the ischemic complication rate was 7% (95% CI 6% to 9%) (supplemental figure 2a figure 1). A low-dose ASA regimen before and after PED treatment was associated with a higher rate of late ischemic complications than high-dose ASA therapy (2.62 (95% CI 1.46 to 4.69) and 2.56 (95% CI 1.41 to 4.64), respectively) (table 2). The ASA dose before PED treatment was not associated with a risk of periprocedural (0.62, 95% CI 0.27 to 1.41) or total (1.18, 95% CI 0.83 to 1.67) ischemic complications. Similarly, the ASA dose after PED treatment was not associated with a risk of periprocedural (0.51, 95% CI 0.22 to 1.16) or total (1.07, 95% CI 0.75 to 1.51) ischemic complications. Duration of postprocedural high-dose ASA therapy (<6 months vs ≥6 months) was also not associated with ischemic complications (1.19, 95% CI 0.58 to 2.43). However, duration of postprocedure clopidogrel therapy (<6 months) was associated with significantly higher rates of ischemic complications (1.56, 95% CI 1.11 to 2.20) than in those who received clopidogrel for ≥6 months.

Figure 1

Proportion meta-analysis of ischemic complications associated with Pipeline embolization devices.

Table 2

Association of antiplatelet regimens with ischemic and hemorrhagic complications after treatment with a Pipeline embolization device

Performing PFT before PED treatment was not associated with a risk of ischemic complications (1.27, 95% CI 0.77 to 2.10) (table 3).

Table 3

Association of platelet function testing (PFT) with ischemic and hemorrhagic complications after treatment with a Pipeline embolization device

Hemorrhagic complication

Overall, the hemorrhagic complication rate was 5% (95% CI 4% to 6%) (supplemental figure 2bfigure 2). The ASA dose before PED treatment was not associated with a risk of periprocedural (0.35, 95% CI 0.11 to 1.09) or late (1.70, 95% CI 0.76 to 3.83) hemorrhagic complications (table 2). The ASA dose after PED treatment was similarly not associated with a risk of periprocedural (0.77, 95% CI 0.31 to 1.88) or late (1.81, 95% CI 0.79 to 4.17) hemorrhagic complications. Duration of postprocedural high-dose ASA therapy (<6 months vs ≥6 months), and duration of clopidogrel therapy (<6 months vs ≥6 months) were not associated with rates of hemorrhagic complications (1.25, 95% CI 0.45 to 3.51 and 0.93, 95% CI 0.55 to 1.58, respectively).

Figure 2

Proportion meta-analysis of hemorrhagic complications associated with Pipeline embolization devices.

Performing PFT before PED treatment was not associated with risk of total hemorrhagic complications (1.08, 0.57–2.05) (table 3).


Our meta-analysis of 2002 patients with 2243 aneurysms showed the following: (1) higher rates of late ischemic complications associated with low-dose ASA therapy than with high-dose ASA before PED treatment; (2) higher rates of late ischemic complications associated with low-dose ASA therapy than with high-dose ASA after PED treatment; (3) higher rates of ischemic complications found with clopidogrel therapy for <6 months than with clopidogrel therapy for ≥6 months; (4) no increased risk of hemorrhagic complications in patients treated with high-dose ASA before or after PED treatment compared with low-dose ASA, or continuing clopidogrel for ≥6 months versus <6 months; and (5) lack of association between hemorrhagic/ischemic complications, and the duration of high-dose ASA therapy (<6 months or ≥6 months). Performing PFTs before PED treatment did not change ischemic or hemorrhagic complication rates.

Embolization with PED is an effective treatment for cerebral aneurysms, but ischemic and hemorrhagic complications are associated with the use of this device. Antiplatelet therapy with ASA and clopidogrel are commonly used to reduce these complications, but dose and duration after the PED procedure vary substantially among different practitioners.32To date, no randomized controlled trial has been conducted to assess the role of antiplatelet regimens on ischemic or hemorrhagic outcomes following PED treatment. Our findings from a well-selected large sample of studies provide stronger evidence about which antiplatelet regimens are associated with the best outcomes in this patient population.

In our analysis, the total rate of ischemic complications was 7%, which is higher than the 4.7% ischemic events rate reported by Kallmes et al from retrospective evaluation of 906 aneurysms in the International Retrospective Study of the Pipeline Embolization Device (IntrePED) study, but lower than the study by Adeeb et al (n=402, ischemic rate: 9%).33 Of note, the IntrePED study was not included in our meta-analyses because it lacked a uniform antiplatelet regimen. The etiology of ischemic events includes stent wall thrombus, parent vessel occlusion, and thromboembolic events. Our analysis showed that patients treated with low-dose ASA or with clopidogrel for <6 months had significantly more total ischemic complications than patients receiving high-dose ASA or clopidogrel for ≥6 months. These findings appear to be in line with a previous meta-analysis by Skukalek et al, whose results also supported an antiplatelet regimen of high-dose ASA and clopidogrel for ≥6 months.34 Our analysis encompasses a larger patient population (2002 vs 1110) from a larger number of studies (29 vs 19) that were performed more recently, thus better reflecting current procedural expertise and complication rates. Another difference is that in our analysis, we found no association between hemorrhagic complications and high-dose ASA therapy or clopidogrel therapy for ≥6 months, whereas Skukalek et al reported lower hemorrhagic complications with high-dose ASA or longer clopidogrel therapy. However, comparison of the two studies is difficult because Skukalek et al stratified complications into transient and permanent, whereas we chose to combine them into one outcome. These results suggest that practitioners should be mindful of the risk of ischemic complications in patients with suboptimal dosing or duration of antiplatelet agents.

Hemorrhagic complications occurred at a rate of 5%, slightly higher than the 3% rate reported by the IntrePed trial.33 Our analysis showed no significant difference between current antiplatelet regimens for the risk of hemorrhagic complications. There was no significant difference between low-dose and high-dose ASA, either before or after the procedure, or clopidogrel maintained for ≥6 months compared with <6 months. These results suggest that high-dose ASA and clopidogrel durations of ≥6 months can be used without increasing the risk of hemorrhagic complications.

Genetic polymorphisms in cytochrome P450 enzymes are responsible for variable patient response to clopidogrel (hyper- or hyporesponsiveness). Whether PFT improves outcomes, however, is controversial. Daou et al conducted a retrospective study of 238 patients in which preprocedural PFT was associated with ischemic and hemorrhagic complications when P2Y12 reactivity units were outside an optimal range of 60–240.35 Adeeb et al retrospectively reviewed 402 patients treated with PED and found 28.8% to be clopidogrel non-responders, with a significantly higher rate of thromboembolic complications than clopidogrel responders.24 However, Brinjikji et al 36 analyzed the IntrePed registry and showed PFT to be associated with higher rates of hemorrhagic events. Our meta-analysis showed no significant correlation between PFT and complication rates. Whether alternative interventions improve outcomes in patients with abnormal preprocedural PFT warrants further investigation.


Most of the studies were observational retrospective or prospective cohorts without randomization. Most published studies did not report complication rates stratified by aneurysm rupture status, location, aneurysm size/type, or number of PEDs used. Therefore, we were unable to control for these factors in our pooled analysis. In addition, follow-up of patients differed between studies. However, we followed a prespecified protocol, and included studies that reported antiplatelet regimen, complication types, and follow-up periods. Some included in their protocol a comparison between prior and follow-up brain imaging that detected asymptomatic strokes which would have been missed by other study protocols. Our results represent a wide spectrum of aneurysms and may not be applicable to subsets of patients with specific aneurysm characteristics. We did not have information about which patients had changes to their antiplatelet regimens based on results from platelet response testing.


Our findings help better characterize the risk of ischemic and hemorrhagic complications with respect to the antiplatelet regimen and support the benefit and safety of high-dose ASA therapy and continuation of clopidogrel for ≥6 months to minimize major ischemic or hemorrhagic complications following PED treatment.


View Abstract


  • Contributors HS, RYK and SN participated in the conception and design of the study. HS, RYK and SK analyzed and interpreted the data. HS carried out the statistical analysis and wrote the article. SN, PC, MOH and HS revised the draft paper for intellectual content.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.