Article Text

Download PDFPDF

Anti-thrombotic medications for the neurointerventionist: aspirin and clopidogrel
  1. David Fiorella
  1. Department of Neurosurgery, Cerebrovascular Center, Stony Brook University Medical Center, Stony Brook, NY, USA
  1. Correspondence to David Fiorella, Department of Neurological Surgery, Cerebrovascular Center, Stony Brook University Medical Center, Health Sciences Center T-12 080, Stony Brook, NY 11794-8122, USA; dfiorella{at}


The majority of neurological complications encountered during neuroendovascular procedures are a result of thromboembolic phenomena. This has become increasingly evident as techniques have evolved to incorporate an ever-growing volume of temporary and permanently implanted adjunctive devices. By optimizing our anti-thrombotic management strategies we can potentially improve procedural safety.

The appropriate selection of anti-thrombotic regimens for neurointerventional procedures poses a unique set of challenges when compared with endovascular interventions in other vascular territories. Neurological procedures frequently target lesions that have hemorrhaged or have a lethal hemorrhagic potential—thus requiring that the operator maintain a tenuous balance between bleeding and thrombosis. Neurological end-organs are unforgiving, even of small emboli, further magnifying the implications of inadequate therapy. The spectrum of disease processes encountered within the context of routine clinical practice is very heterogeneous and each lesion requires a specific anti-thrombotic strategy. The diseases treated (eg, aneurysms, AVM, acute ischemic stroke, chronic atherosclerotic stenosis) are relatively uncommon in comparison with coronary or peripheral atherosclerotic disease. The field of neurointervention itself is still relatively new and continues to evolve rapidly, with innovation frequently outpacing clinical evidence. Predictably, within this environment, there are no large, controlled trials to guide anti-thrombotic management in most cases. For these reasons, medical decision making is based largely upon an understanding of the pharmacology of the agents used and an extrapolation of the literature from other fields.

In this review we discuss aspirin and clopidogrel, the two anti-platelet agents commonly used in conjunction with endovascular stents and stent-like devices. We will discuss their pharmacology and applications in neuroendovascular therapeutics with a focus on practical solutions to dilemmas that are encountered during the course of daily clinical practice.

  • Artery
  • device
  • drug
  • stent
  • technique

Statistics from

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.


Mechanism of action

Thromboxane A2 (TXA2) is synthesized within activated platelets and released to function as a platelet surface receptor agonist. The binding of TXA2 to platelet surface receptors acts within a positive feedback cycle to amplify regional platelet activation. TXA2 also acts to recruit additional platelets and stimulates local vasoconstriction. The synthesis of TXA2 is mediated by the cyclooxygenase-1 (COX-1) enzyme. Aspirin (acetylsalicylic acid: ASA) blocks the activity of COX-1 by transferring an acetyl group to a serine residue near the active site of the enzyme. The irreversible acetylation of this serine residue prevents the synthesis of TXA2 and this inhibition persists throughout the life of the platelet.1

While the inhibitory effect of aspirin on TXA2 synthesis does reduce the overall level of platelet function, it is important to recognize that TXA2 is only one of a myriad of soluble surface agonists (eg, thrombin, collagen, catecholamines, ADP) and other potential stimuli (such as shear stress) that can function to activate platelets. So, even in the setting of optimal aspirin therapy, platelets retain an ability to become activated and form a thrombus, that is, the platelet inactivation created by aspirin is partial.


Aspirin has a remarkably fast onset of activity, as it is absorbed within the stomach and proximal duodenum and very quickly begins to acetylate the COX-1 enzyme within the platelets of the presystemic portal circulation.2 3 This process begins within 5 min of the oral administration of aspirin with maximal activity measurable at between 30 and 60 min.4 Even at relatively low doses (approximately 100 mg), the entire population of circulating platelets is exposed to a concentration of ASA that is sufficient to completely block TXA2 synthesis.5 Since platelets are anucleate and lack the synthetic machinery to create new COX-1 enzyme, the irreversible inhibition of the enzyme induced by aspirin persists throughout the lifetime of the platelet (7–10 days). Restoration of platelet function is dependent upon the synthesis and release of platelets formed within the bone marrow after the administration of the last dose of the drug.6 Approximately 10% to 15% of platelets are replaced each day.7 Thus, the cessation of aspirin therapy leads to the normalization of platelet function as measured by in vitro assays over a period of 4–7 days after the last dose.8 More importantly, the overall recovery of in vivo platelet function seems to occur more quickly as normal or near normal hemostasis can be achieved when as little as 20% of the circulating platelet pool has been replenished. This is reflected as a normalization of bleeding times observed at between 48 and 72 h after aspirin discontinuation. Immediate reversal of aspirin therapy can only be achieved with a platelet transfusion.


Aspirin demonstrates effective anti-thrombotic activity both clinically and in various bioassays over a wide spectrum of doses.5 The doses of aspirin used clinically to achieve an anti-thrombotic effect have varied widely, ranging from 30 mg to 1500 mg per day. The Anti-Thrombotic Trialists Collaboration performed a meta-analysis of 287 studies including >100 000 patients at ‘high risk’ for vascular events (acute or previous vascular disease or significant risk factors for vascular disease). This study confirmed that aspirin therapy led to a reduction in serious vascular events, all vascular mortality, non-fatal myocardial infarction (MI) and stroke. The beneficial effect of aspirin was most robust at low to moderate doses (75–325 mg) and less evident at the higher doses (>500 mg). Similar results were reported by the Aspirin and Carotid Endarterectomy Trial Collaborators who observed low-dose aspirin (81–325 mg) to be more effective in preventing the primary endpoints of stroke, death and MI when compared with higher doses (625–1300 mg).9

In neuroendovascular therapeutics, aspirin is most commonly used for the avoidance of periprocedural and post-procedural thromboembolic complications associated with extracranial or intracranial stenting for the treatment of either atherosclerosis or aneurysm. The standard dose provided is 325 mg per day, but no dose-finding data are available to support the selection of either a dose or duration of aspirin therapy required for neurovascular applications. A surprising lack of evidence also exists to support a given dose of aspirin for the prophylaxis of coronary stents. In a recent study, 1840 patients were randomized to receive either 81 or 325 mg of aspirin after percutaneous coronary interventions. After 1 year of follow-up, no differences in either death or MI were observed between the high- and low-dose groups.10 The published guidelines from the American Heart Association advocate aspirin doses of 162–325 mg for at least 1 month after bare metal stent implantation and for 3–6 months after drug-eluting stents (DESs)11 with low-dose (75–162 mg) aspirin continued indefinitely thereafter.

Aspirin resistance

Aspirin resistance remains a controversial topic with reported rates of non-responsivity ranging between 5% and 40%.12 A number of different assays are available to monitor the effect of aspirin on platelet activity and the results of these tests may vary widely. Lordkipanidze et al13 reported the results of six different platelet function assays that were applied in a population of 201 patients with stable coronary artery disease (CAD) on aspirin therapy (>80 mg per day). The rate of ‘aspirin resistance’ detected in this population ranged between 4% and 51.9% depending on which assay was used. Correspondingly, the literature in this area is confusing and sometimes contradictory since no generally agreed upon definition or standardized assay for aspirin resistance exists.

However, regardless of the assay used, there is substantial clinical evidence to support the concept that those patients who have a ‘subnormal’ or poor response to aspirin therapy have higher rates of clinical events. Gum et al14 reported that stable cardiovascular patients were at a three times higher risk of death, MI and CVA over an approximately 2-year period if they were aspirin resistant. Aspirin-resistant patients were found to be at a three times greater risk of having CK-MB elevations following non-emergent percutaneous coronary interventions.15 One of the more common assays used to test aspirin responsivity in neuroendovascular therapy is the widely available VerifyNow Aspirin (Accumetrics Inc, San Diego, California, USA) test. In a recent publication, Chen et al16 observed three times higher rates (15.6% vs 5.3%) of cardiovascular death, MI, stroke, TIA and unstable angina requiring hospitalization in aspirin-resistant CAD patients (n=128 (27.4%)) in comparison with aspirin-responsive CAD patients (n=340).

Despite these data, the mechanisms of aspirin resistance and the appropriate interventions to take in patients who are identified as non-responders remain unclear. Alberts et al17 initially reported that aspirin resistance was related to both aspirin dose and preparation, with more than half of patients on a low dose of aspirin (81 mg) and two-thirds taking enteric coated aspirin demonstrating no significant inhibition of platelet function using the PFA-100 assay. While additional studies have also concluded that lower doses of aspirin may be associated with higher levels of aspirin resistance,18 19 the impact of enteric coating has been largely dismissed. Karha et al20 found no difference in the rates of aspirin resistance in any of three assays of platelet function in a group of 50 normal volunteers administered 7-day courses of enteric coated and non-enteric coated aspirin (81 mg dose). For this reason, enteric coated aspirin is favored as a preparation if felt to improve patient compliance. Finally, in patients on chronic aspirin therapy, resistance may arise in previously responsive patients. Pulcinelli et al21 observed that approximately 40% of patients treated with aspirin eventually became resistant with assays of platelet function gradually returning to a pre-aspirin therapy baseline over a period of 24 months.

Aspirin interaction with ibuprofen

Aspirin is frequently administered along with non-steroidal anti-inflammatory drugs (NSAIDs), particularly ibuprofen, and the interaction between these two agents has been relatively well described. While aspirin irreversibly acetylates the COX-1 enzyme, ibuprofen binds reversibly to the same region of the enzyme. Correspondingly, when administered together, ibuprofen will compete with aspirin for the COX-1 binding site, thus limiting the irreversible deactivation of the enzyme by aspirin. Since the half-life of aspirin in the blood is short (20 min), it is quickly cleared from the circulation during this period of competition, resulting in a reduction in the efficacy of aspirin when the two agents are administered together. For this reason, in patients on both drugs, aspirin should be administered at least 30 min prior to the NSAID.

This aspirin–ibuprofen interaction has been validated in several clinical studies. Catella-Lawson et al measured TXB2 levels in patients being administered both aspirin and ibuprofen for six consecutive days. When ASA was administered 2 h before ibuprofen, a 99% level of TXB2 inhibition was achieved. When ibuprofen was administered first, only a 53% level of TXB2 inhibition was achieved.22 In the US physicians health study, subjects administered aspirin for the primary prevention of MI experienced a two–threefold increased risk of MI if they regularly used non-steroidal anti-inflammatory medications for >60 days out of the year.23

When administered alone, NSAIDs, similar to aspirin, can inhibit platelet function through a blockade of COX-1 activity. However as a reversible inhibitor of the enzyme, the duration and magnitude of this inhibition is related to the dose of the agent and its half-life within the circulation. For example, ibuprofen creates a transient inhibition of platelet aggregation within 2 h of administration; however, this effect typically has dissipated within 12 h of a single dose.24

The widespread use of ibuprofen and the potential implications of this interaction warrant a brief discussion of this issue with patients on aspirin therapy.


Mechanism of action

Clopidogrel (Plavix; Sanofi-Aventis), along with ticlopidine (Ticlid) is within the class of theinopyridines. These agents function to inhibit platelet activity by irreversibly blocking the binding of the soluble agonist adenosine diphosphate (ADP) to the platelet surface P2Y12 receptor. After the introduction of clopidogrel, ticlopidine use markedly declined due to its association with drug-induced neutropenia in approximately 2% of treated patients and the superior pharmacokinetics of clopidogrel.25 26 Currently, ticlopidine is typically reserved for patients who are intolerant of clopidogrel.

Clopidogrel is a pro-drug which has no inherent activity in vitro. The activity of clopidogrel requires its hepatic transformation to an active thiol metabolite (R130964) which irreversibly binds to and inhibits the P2Y12 receptor. For this reason, the anti-platelet activity of clopidogrel is not manifest immediately after administration.


The typical daily maintenance dose of clopidogrel is 75 mg. If therapy is initiated at this dose, between 3 and 7 days are required to reach a therapeutic level of platelet inhibition.27 If a bolus is provided, therapeutic levels of platelet inhibition can be achieved much more quickly, sometimes within hours, depending on the dose. The originally advocated loading dose of 300 mg of clopidogrel was based upon the findings of the CREDO study,28 which demonstrated that patients who received a 300 mg clopidogrel loading dose at least 6 h before percutaneous coronary intervention (PCI) had a 38.6% reduction in death, MI or urgent target vessel revascularization at 1 month. Subsequent investigations demonstrated that a higher, 600 mg, clopidogrel loading dose achieved an even faster onset of action (2–4 h to peak effect), lower rates of resistance, lower rates of rates of recurrent and periprocedural ischemic events and no increase in hemorrhagic complications in patients undergoing PCI.29–32 In a subsequent meta-analysis of 10 studies (7 randomized, 3 non-randomized) with 1567 total patients, higher clopidogrel loading doses (>300 mg) were associated with a reduction in cardiac death or non-fatal MI within 1 month in patients undergoing PCI. The number needed to treat with a high loading dose to prevent one cardiac death or MI was only 33. With respect to safety endpoints, high- and low-dose regimens were found to be no different—specifically there was no difference in bleeding (either severe or mild) between groups. The benefits seen in this meta-analysis were most evident in high-risk patients.33 34 When even higher clopidogrel loading doses (900 mg) were evaluated, no additional benefit was realized.35


As with aspirin, the inhibition achieved with clopidogrel is irreversible and the recovery of normal platelet function is governed by the generation and release of unaffected platelets from the bone marrow into the circulation.36 Platelet function returns toward normal between 5 and 7 days after the discontinuation of therapy.36 37 Price et al37 observed that platelet inhibition significantly decreased each day after clopidogrel was discontinued and that 5 days after the last dose, only two of 45 (5%) healthy volunteers demonstrated >40% residual platelet inhibition. The median level of platelet inhibition in the group as a whole after 5 days was 12%. The immediate reversal of clopidogrel activity requires a platelet transfusion.


Similar to aspirin, the measured effect of clopidogrel upon platelets varies dramatically between patients and between assays. Estimates of the incidence of low response or non-response to clopidogrel range between 5% and 30%.38 39 Responsivity to clopidogrel has been demonstrated to be stable over time; however, the existing studies do not extend beyond 30 days.40 Clopidogrel resistance, similar to aspirin resistance, has been associated with adverse clinical outcomes in association with PCI.41 For example, Matetzky et al42 stratified a series of 60 patients undergoing PCI into quartiles based on their responsivity to clopidogrel. Six of 15 (40%) patients in the first quartile (clopidogrel-resistant patients) experienced a recurrent cardiovascular event during 6-month follow-up while only one patient (6.7%) in the second quartile and no patients in the third or fourth quartile experienced recurrent events.

It has also been observed that aspirin-resistant patients may have a high rate of clopidogrel resistance as well. Lev et al43 found in a population of 150 patients that, while 19 (12.7%) were resistant to aspirin and 36 (24%) were resistant to clopidogrel, that nearly half (9 of 19 or 47%) of the aspirin-resistant patients were also resistant to clopidogrel. Gori et al44 also observed that 6% of patients treated with drug eluting stents were resistant to both aspirin and clopidogrel, and that these patients were at very high risk for DES thrombosis and death.43 44

In contrast to aspirin, there is growing evidence that resistance to clopidogrel can be managed in some patients by increasing the administered dose. Gurbel et al45 have demonstrated that when compared with the 300 mg clopidogrel loading dose, the 600 mg dose reduced the incidence of clopidogrel resistance from approximately 30% to 8%. Similarly, increasing the daily maintenance dose can reduce the level of resistance and may overcome resistance in some patients who are non-responsive to lower doses.46 Aleil et al47 observed rates of clopidogrel non-responsiveness of 33.7% and 8.6% of patients of maintenance doses of 75 mg and 150 mg, respectively. Moreover, when resistant patients were changed from the 75 mg dose to the 150 mg dose, 65% (20 of 31) were converted to responders. Importantly, no increased risk of bleeding was observed in patients treated with the higher daily maintenance dose. These observations have led some to advocate the optimization of clopidogrel therapy for patients by individualized dosing based on the results of aggregometry.46 This practice pattern is currently being tested within the context of the Gauging Responsiveness With a VerifyNow Assay: Impact on Thrombosis and Safety (GRAVITAS) trial. In GRAVITAS, patients undergoing PCI will be randomized to receive either standard dose clopidogrel or adjusted clopidogrel doses that have been guided by the VerifyNow Assay. This represents the first large clinical trial that will assess whether ‘optimization’ of clopidogrel therapy on the basis of platelet function testing actually improves clinical outcomes.48

Clopidogrel interaction with proton pump inhibitors?

A series of recent studies have provided evidence of an interaction between omeprazole (Prilosec) and clopidogrel. The basis for this interaction is the potential for some proton pump inhibitors (PPIs) to inhibit the function of cytochrome P450 2C19, one of the hepatic enzymes responsible for the conversion of clopidogrel to its active metabolite (R-130964).49 The OCLA (Omeprazole CLopidogrel Aspirin) study was a randomized, placebo-controlled trial of 140 patients undergoing PCI on asprin and clopidogrel therapy. Patients randomized to receive 20 mg of omeprazole had a mean platelet reactivity index (PRI) of 51.4% while those on placebo had a mean PRI of 39.8% (p=0.0001).50 The PRI is considered to reflect a therapeutic response to clopidogrel if it is <50% and a poor response if >50%. In the Ontario Public Drug Study Program the health records of 13 636 patients with claims for clopidogrel within 3 days of an acute MI were reviewed. While pantoprozole (Protionix) was not associated with any increased risk, other PPIs were associated with a 40% increased risk of recurrent MI (OR 1.4, 1.1–1.8) and on the basis of these data it was estimated that 14% of all readmissions for recurrent MI were associated with this PPI–clopidogrel interaction.51 However, the recently presented COGENT trial, a randomized assessment of the effects of clopidogrel and PPIs on clinical events, showed no evidence of any clinically relevant adverse cardiovascular interaction between the two medications. These findings not only cast doubt on the importance of the interaction, but raise concerns about the overall relevance of platelet function assays to clinical outcomes.53

Dual anti-platelet therapy with aspirin and clopidogrel

Why aspirin and clopidogrel?

Given that aspirin and the theinopyridines inhibit platelet function through different surface receptors, it would be expected that the simultaneous administration of these two agents would provide a synergistic inhibitory effect on platelet function. This synergy has been validated in ex vivo studies which have demonstrated that when administered together, the anti-platelet activities of aspirin and ticlopidine are far greater than either agent administered individually.54 The synergy of dual anti-platelet therapy (DAT) has had a tremendous impact on the periprocedural safety profile of PCI. Schomig et al55 compared regimens of aspirin–warfarin and aspirin–ticlopidine after PCI in a series of 517 patients. Adverse cardiac events at 30 days were reduced from a rate of 6.2% in the aspirin–warfarin group to 1.6% in the dual anti-platelet group. Leon et al56 confirmed these observations, reporting that DAT reduced stent thrombosis (within 30 days of PCI) to a rate of 0.6% in comparison with aspirin alone (3.6%) or aspirin–warfarin (2.4%) in a series of 1653 patients. These large clinical trials from cardiology have formed the foundation for the rationale behind the use of DAT in neurovascular stenting.

How long is DAT needed?

Even with the copious volume of data from the cardiology literature, this issue remains somewhat controversial at this point even within cardiology. Not surprisingly, there is no evidence from clinical trials to specifically guide the duration of DAT required after neurovascular stenting.

Presumably, once completely incorporated into the vessel wall and covered by neointimal and neoendothelial tissue, an implanted stent should become non-thrombogenic. While the existing data are sparse, an evaluation of the evolving histology of implanted coronary stents in humans has demonstrated that early after stenting (<11 days), the surface of the implant is covered only by fibrin, platelets and acute inflammatory cells.57 At 2 weeks, neointima formation is beginning and there is incomplete coverage of the stent surface. At 4 weeks, a thin, complete, neointimal coverage is seen in some cases (‘neointimal repair’). Over the next 8 weeks the neointimal layer thickens and becomes organized into a three-layered structure. At the time of coronary stent placement (or angioplasty) there is complete denudation of the endothelial surface. In the first weeks after stenting, a thin, membranous thrombus covers the vascular and stent surface and essentially functions as the ‘endothelial coverage’. By 2–4 weeks, the smooth muscle cells constituting the surface of the neointima function as the interface between the vessel lumen and the vessel wall. This situation persists, at least to some extent, until complete neoendothelialization occurs, which is not observed until approximately 3 months.58 This rate of neoendothelialization described in human pathology studies is considerably slower than what has been observed in animal studies. As such, inferences about rates of stent incorporation derived from animal models must be interpreted with caution.

Clinical studies examining the thrombogenicity of implanted stents have predictably demonstrated that the risk of stent thrombosis is at its highest during the days immediately after placement and decreases considerably over the ensuing 2–4 weeks. During surveillance of the 30-day periprocedural period, Schomig et al55 observed no stent thrombosis after 4 days and Leon et al56 observed no stent thrombosis after 14 days in their respective trials of different anti-thrombotic regimens after coronary stenting.55 56 Berger et al analyzed periprocedural (30 day) clinical outcomes for a series of 827 consecutive patients who received only 2 weeks of DAT (aspirin and ticlopidine) after coronary stenting and observed no stent thrombosis following the discontinuation of ticlopidine. Unanticipated surgery represents a relatively common scenario in which DAT is prematurely discontinued after coronary stenting. Kaluza et al59 reported ‘catastrophic outcomes’ (7 MIs, 11 major bleeding episodes and 8 deaths) in a series of 40 patients undergoing non-cardiac surgery <6 weeks after coronary stenting. Notably all of these events occurred when surgery was performed within 14 days of coronary stenting. Wilson et al60 reported on a similar cohort of patients undergoing non-cardiac surgery <8 weeks after coronary stenting and observed eight events (MI or death) in 207 subjects. All thrombotic events in this series were observed within 6 weeks of the stenting procedure. These authors recommended that whenever possible, surgeries be deferred for at least 2–6 weeks after coronary stenting to allow completion of the required course of anti-platelet therapy.59 60

On the basis of the available clinical evidence, the American College of Cardiology and the American Heart Association have issued recommendations for anti-platelet therapy following coronary stenting. For bare metal stents, moderate-dose (162–325 mg) aspirin therapy is recommended for at least 1 month, to be followed by low-dose aspirin (75–162 mg per day), indefinitely. Clopidogrel (75 mg per day) is recommended for at least a month, with continued administration preferred for a full year. The authors acknowledge that no reliable data exist upon which to base a recommendation for continued therapy beyond 1 year. For DES, moderate-dose aspirin therapy is recommended for 3–6 months and low-dose aspirin indefinitely thereafter. 75 mg of clopidogrel is recommended for at least 1 year. The optimal duration of clopidogrel therapy beyond 1 year has not been determined and decisions therefore must be individualized with respect to the risk:benefit ratio for each patient.11

It is clearly very difficult to translate these recommendations for coronary stenting to neurovascular applications as profound differences exist with respect to the disease processes treated, the status of the vessels into which the devices are implanted (eg, an atherosclerotic coronary artery versus a normal caliber parent vessel giving rise to an aneurysm), and the metal surface area of the implanted constructs (eg, 6% to 9.5% with a self-expanding Nitinol stent, 14% to 18% with a balloon expandable coronary stent and 30% to 35% with a braided flow diverting stent). However, the same principles likely apply in that thromboembolic events are most likely to occur during the first 2 weeks after implantation with the risks likely decreasing considerably each day after implantation. As a practical matter, for patients with recently implanted neurovascular stents (<14 days) who require surgical procedures, each additional day that the procedure can be deferred (and anti-platelet medication can be maintained) would be expected to considerably reduce the risk of construct thrombosis when DAT is ultimately discontinued or reversed.

What are the risks of DAT?

The institution of DAT for extended periods carries with it a low, but not insignificant risk of hemorrhage which must be considered, particularly in neurovascular patients. The data set which probably most closely reflects the risk of DAT for most neuroendovascular patients is the MATCH trial, since this trial was restricted to patients with ischemic cerebrovascular disease. In MATCH, 7599 patients with recent stroke or TIA (with vascular risk factors) were randomized to receive either clopidogrel alone or clopidogrel with low-dose (75 mg) aspirin. With respect to the primary endpoint, there was no difference in ischemic stroke, MI, vascular death or rehospitalization for acute ischemia; however, the risk of life-threatening bleeding was significantly higher in the DAT group (2.6%) in comparison with the clopidogrel alone group (1.3%) over the 18-month study period.61 In addition to the 1.7% per year risk of life-threatening bleeding, the risk of symptomatic intracranial hemorrhage was approximately 0.7% and the risk of fatal hemorrhage was approximately 0.3%. In the CHARISMA trial, 15 603 patients with cardiovascular disease or multiple vascular risk factors were treated with either aspirin (81–162 mg) alone or aspirin with clopidogrel (75 mg) for a median of 28 months. This trial revealed a very modest but significant reduction in the composite endpoint of MI, stroke and cardiovascular death in patients with established cardiovascular disease. Over the median 28-month treatment period, the risk of severe hemorrhage was 1.7%, the risk of intracranial hemorrhage was 0.33% and the risk of fatal hemorrhage was 0.33% in the patients on DAT.62 A meta-analysis of 13 studies including 87 205 patients treated with anti-platelet agents for secondary stroke prevention revealed a major bleeding risk of 1.8% per year for patients on DAT.63

Thus, in trials of patients with cerebrovascular disease, DAT is associated with a risk of severe bleeding which ranges between 1.5% and 2.0% per year. This represents a significant consideration with respect to the application of devices for the treatment of cerebrovascular disease which may require long-term DAT in otherwise healthy patients. On the contrary, the results of the CHARISMA trial suggest that in patients with established cardiovascular disease, this risk may be acceptable and the therapy may actually hold some benefit.


The optimized administration of anti-platelet medications represents an important component of the comprehensive management of neuroendovascular patients. There are no data specific to percutaneous neuroendovascular interventions, a working knowledge of the pharmacology of the individual agents and the existing cardiology literature must be extrapolated to construct appropriate patient- and procedure-specific treatment regimens.



  • Competing interests None.

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