Statistics from Altmetric.com
Over the past decade there has been a growing use of intracranial stents for the treatment of both ischemic and hemorrhagic cerebrovascular disease, including stents to assist in the remodeling of the neck of aneurysms as well as the use of flow diverting devices for aneurysm treatment. With this increase in stent usage has come a growing need for the neurointerventional (NI) community to understand the pharmacology of medications used for modifying platelet function, as well as the testing methodologies available. Platelet function testing in NI procedures remains controversial. While pre-procedural antiplatelet assays might lead to a reduced rate of thromboembolic complications, little evidence exists to support this as a standard of care practice. Despite the routine use of dual antiplatelet therapy (DAT) with aspirin and a P2Y12 receptor antagonist (such as clopidogrel, prasugrel, or ticagrelor) in most neuroembolization procedures necessitating intraluminal reconstruction devices, thromboembolic complications are still encountered.1–3 Moreover, DAT carries the risk of hemorrhagic complications, with intracerebral hemorrhage (ICH) being the most potentially devastating.4 ,5
Light transmission aggregometry (LTA) is the gold standard to test for platelet reactivity, but it is usually expensive and may not be easily obtainable at many centers. This has led to the development of point-of-care assays, such as the VerifyNow (Accumetrics, San Diego, California, USA), which correlates strongly with LTA and can reliably measure the degree of P2Y12 receptor inhibition.6–9 VerifyNow results are reported in P2Y12 reaction units (PRUs), with a lower PRU value corresponding to a higher level of P2Y12 receptor inhibition and, presumably, a lower probability of platelet aggregation, and a higher PRU value corresponding to a lower level of P2Y12 receptor inhibition and, hence, a higher chance of platelet activation and aggregation.
While aspirin resistance is perhaps less common, clopidogrel resistance may be more challenging as it is reported in the monitoring cohort to be as high as 30–35% and is usually due in part to genetic variation, which is reported to increase thromboembolic complications even with escalating dosing of clopidogrel.3 ,4 ,10 ,11 Patients who have CYP2C19 allelic variants are highly likely to exhibit clopidogrel resistance. The cardiology literature, representing many large multicenter randomized controlled trials, did not show an overall clinical outcome benefit of antiplatelet therapy modification based on pre-procedural assays.12 ,13 However, there is evidence in the cardiology literature to suggest that clopidogrel resistance leads to a higher level of thrombotic complications. The validity of extrapolating the cardiology literature to the NI population is questionable, especially since the underlying vessel in the setting of cerebral aneurysm treatment with concomitant stent placement may not be as diseased with atherosclerotic plaque and precedent angioplasty as in the coronary circulation. Furthermore, NI procedures using stents and flow diverters have been associated with perioperative ICH, and some studies indicate that P2Y12 receptor over-inhibition may play a role.4 ,5 ,14 Many NI procedures necessitate patient treatment with mono antiplatelet therapy or DAT for variable periods of time, with or without pre-procedural loading doses. The purpose of this consensus paper is to develop recommendations for the use of platelet function testing in this unique patient population.
Recommendations were developed based on the existing literature, a robust discussion regarding the interpretation of the literature, and the collective experience of the members of the writing group (table 1). Experts from academic institutions in North America from the specialties of neurosurgery, neurology, and interventional neuroradiology were recruited based on their expertise related to each scenario discussed. A computerized search of the MEDLINE database (PubMed) from 1 January 1991 to 31 November 2013 was performed using the search terms ‘antiplatelet’, ‘treatment’, ‘VerifyNow’, ‘LTA’, ‘PRUs’, ‘endovascular’, ‘neuro-endovascular’, and ‘interventional radiology’ with the purpose of identifying published articles on the utility of antiplatelet function tests in NI procedures. All relevant English language articles published during this period were taken into consideration while writing this consensus paper. The literature review consisted mostly of case series and non-randomized single-center studies.
Mechanisms of intraprocedural thrombosis
There are four major platelet function factors which influence potential complications in NI procedures: adherence,15 activation and secretion,16 aggregation,17 and interaction with coagulation factors.18 When a foreign body (eg, aneurysm coil, stent, catheter, or wire) is introduced into the intravascular space, a process of thrombosis and inflammation is initiated.19 Leukocytes are summoned to the foreign body and express tissue factor on exposed monocytes20 that can then activate platelet aggregation. Additionally, released inflammatory mediators can activate platelets, allowing for thrombus formation.19 Furthermore, a number of plasma proteins, including fibrinogen, bind to foreign bodies within minutes, initiating a platelet-induced thrombus cascade.21 ,22
How much platelet aggregation and thrombosis occurs on neuroendovascular foreign bodies is determined by the composition of the implants/tools, surface charge, endothelial damage, as well as sheer stress.23 Similarly, when detachable coils are detached via a positive charge, negatively charged blood products including platelets and red blood cells may be attracted to this site and this process may induce significant occlusion of aneurysms during coiling.23 Likewise, some authors have theorized that high radial force stents, such as balloon-expandable stents, induce significant endothelial injury and more platelet aggregation and thrombus formation than would be seen with less traumatic low radial force nitinol self-expanding stents.23 While thought provoking, the validity of this theory and its long-term significance remain unknown.
Pharmacology of antiplatelet agents
Acetylsalicylic acid (aspirin, ASA)
Mechanism of action
Aspirin is absorbed by the gastrointestinal tract and hydrolyzed to salicylic acid.24 It irreversibly acetylates cyclo-oxygenase COX-1 and COX-2, thereby blocking the conversion of arachidonic acid to prostaglandins and eventually thromboxane A2.25 This equates to a prolonged bleeding time secondary to platelet aggregation being blocked, rather than adhesion.15–17 Platelets lack the ability to regenerate COX, meaning that aspirin is a suicide inhibitor as it lasts for the life of the platelet, which is about 7–10 days.25 Only low doses of aspirin are needed to inhibit COX-1 in healthy subjects,26 whereas higher doses are needed to inhibit COX-2, with the latter providing analgesia and reduction in inflammation.25
The bioavailability of aspirin is approximately 50%,25 with activity noted as early as 5 min after ingestion and peak plasma levels obtained relatively quickly at about 30–60 min.23 This rapid maximal effect has been attributed to acetylation of COX in the portal circulation.27 Enteric-coated preparations have lower bioavailability and a longer time to peak plasma levels of 3–4 h.27 Interestingly, although aspirin has a short half-life of 15–20 min, given that its acetylation is irreversible, its effect is long-acting.25 However, COX activity recovers at about 10% per day, given new platelet synthesis, and it has been shown that as little as 20% of platelets with normal activity can result in normal hemostasis.28
Platelet prostaglandin synthesis is nearly completely inhibited by 100 mg oral aspirin taken once or by 30 mg taken daily for 7–10 days.27 ,28 Doses below 100 mg result in dose-dependent effects on thromboxane A2 production, and repeated daily doses have a cumulative effect.28 To determine the best dose in preventing any serious vascular event (including myocardial infarction and stroke), a meta-analysis of 287 randomized trials, equating to 135 000 patients, found that doses of 75–150 mg resulted in a 32% reduction in vascular events, doses of 160–325 mg a 26% reduction, and higher doses of 500–1500 mg a 19% reduction, whereas doses <75 mg produced only a 13% reduction.29 Overall, doses <75 mg or >325 mg are not as effective, and today most practitioners agree with either an 81 mg or 325 mg dose with no loading needed.
Given the implications for blockade of thrombus formation in the prevention of stroke, aspirin resistance is a substantial problem. Outside non-adherence, the natural incidence of aspirin resistance is between 5% and 40% at doses of 325 mg.23 ,30 ,31 In those who are resistant, higher doses can help to overcome the resistance; for example, in one study 56% resistance was found at 81 mg and only 28% at 325 mg.32 Resistance is even higher in those taking enteric-coated aspirin,32 and may develop with long-term therapy such as over several years.33 In the interventional cardiology population, aspirin resistance leads to 300% higher rates of death, myocardial infarction, and stroke, observations that might have implications for the neuroendovascular patient population.23 Biochemical mechanisms of resistance include drug–drug interactions, as seen with proton pump inhibitors which are commonly prescribed for chronic aspirin therapy to reduce the incidence of ulcers.34 The resistance mechanism is secondary to increased gastric acid which ionizes aspirin, making it less readily absorbable, but the consequence of this is unknown.34 Other mechanisms such as anion efflux pumps, esterase-mediated metabolism, competitive inhibition by other non-steroidal anti-inflammatory drugs (NSAIDs), COX polymorphisms, high platelet turnover, COX regeneration, and tachyphylaxis have also been implicated.35
After stopping aspirin, normal hemostasis usually returns in about 4 days36 although it may take up to 5–7 days. Restoration of platelet activity can be obtained more immediately with a transfusion of one pack (5 units) of platelets and/or administration of desmopressin (ddAVP), 0.3 μg/kg body weight intravenously or intranasally once, with practically a normalization in the platelet functional assay. The effect of desmopressin is seen within 30 min and lasts up to 4 h.37
Thienopyridines (P2Y12 inhibitors): ticlopidine (Ticlid), clopidogrel (Plavix), prasugrel (Effient), ticagrelor (Brilique, Brilinta), and cangrelor
Mechanism of action
Four oral antagonists have been developed in rapid sequence—ticlopidine, clopidogrel, prasugrel, and ticagrelor—with each agent having a more reliable and rapid onset of action than the other. Ticlopidine and clopidogrel are oral pro-drugs that are metabolized by hepatic cytochrome 450 enzymes to active metabolites that irreversibly prevent ADP from binding to the P2Y12 platelet receptor.38 ,39 In doing so, platelet aggregation is inhibited because the glycoprotein GP IIb/IIIa cannot be induced to its high affinity state, which prevents fibrin cross-links resulting in a prolonged bleeding time. Ticlopidine was the first-generation thienopyridine but, given its side effect profile of bone marrow suppression, the second-generation clopidogrel has become the current thienopyridine of choice in the neuroendovascular suite, with ticlopidine reserved for patients intolerant of clopidogrel.23 In the Clopidogrel Aspirin Stent International Cooperative Study (CLASSICS), clopidogrel was safer, had a faster mechanism of action, and was equally efficacious in preventing thrombotic complications after coronary stenting.40 This has been confirmed in other randomized trials.41
Newer agents include the third-generation prasugrel and fourth-generation ticagrelor, which have different mechanisms of action to inhibit the P2Y12 receptor. Prasugrel is an orally active thienopyridine that is not affected by clopidogrel resistance seen with some cytochrome P450 polymorphisms.42 Prasugrel has more potent antiplatelet effects, a lower incidence of interpatient variability in antiplatelet response, and a reduced time to onset of antiplatelet activity compared with clopidogrel.43 Prasugrel is superior to clopidogrel in preventing ischemic events in patients undergoing coronary interventions, but has a higher bleeding risk.43 Ticagrelor is an orally active P2Y12 blocker that does not require conversion to an active metabolite and inhibits platelet aggregation more completely than clopidogrel or prasugrel.44 ,45 The PLATO trial demonstrated that ticagrelor is superior to clopidogrel when treating patients with acute coronary syndrome with similar overall bleeding rates.46 However, this has been challenged, with others showing higher bleeding rates with ticagrelor and advising continued use of clopidogrel.47 Both of these newer agents have been tested in the neuroendovascular suite in observational studies; some groups have reported similar bleeding rates with prasugrel and ticagrelor compared with clopidogrel48–50 while others have reported an increased bleeding risk with prasugrel.5 ,14 ,51 Lastly, cangrelor45 is a potent intravenous P2Y12 inhibitor which, in a randomized trial (CHAMPION-PHOENIX), significantly reduced the rate of ischemic events during percutaneous coronary interventions (PCIs), including stent thrombosis, with no significant increase in severe bleeding compared with clopidogrel.52 However, it is not currently available in the USA or Europe for clinical use.
Pharmacokinetics and dosing
The bioavailability of clopidogrel is approximately 50% with a half-life of 6 h.38 Following oral administration, clopidogrel is rapidly absorbed in the intestine and activated in the liver.53 It is an irreversible inhibitor and thus, like aspirin, its effects last the lifespan of platelets at 7–10 days.53 It takes 12–24 h to have significant platelet inhibition (25–30%) if a standard daily dose of 75 mg is used, which equates to about 3–7 days (average 5 days) to achieve maximal steady state with 50–60% platelet inhibition.54 ,55 A loading dose of 300–600 mg results in maximal steady state within 2–6 h which lasts up to 48 h.54 ,55 The loading dose needs to be given at least 6 h prior to the endovascular procedure to have benefit.23 Lastly, it has several drug–drug interactions due to shared metabolism by the cytochrome P450 system.53
Prasugrel is a pro-drug that is metabolized to one active metabolite with rapid absorption.56 Time to peak plasma concentration is approximately 30 min with a half-life of 4–7 h, and maximal steady state is achieved in 3–5 days with 70–80% platelet inhibition.56 It has a more rapid and more potent platelet aggregation inhibition than clopidogrel and also has no significant drug–drug interactions.56 A typical loading dose is 60 mg, and within 1 h patients will have 50% platelet inhibition, which is much faster than clopidogrel.57 A normal daily dose is 10 mg and results in more platelet inhibition than a clopidogrel dose of 75 mg or even double dosing.57 Patients can be transitioned immediately from clopidogrel to prasugrel without an interruption of antiplatelet effects.57 Given that this is an irreversible inhibitor, it takes 7–10 days for platelet function to return to normal.
Ticagrelor (Brilinta, Brilique)
Ticagrelor is the first oral P2Y12 antagonist that is a reversible allosteric binder, with a bioavailability of 36%.46 It is absorbed quickly and reaches its peak concentration after about 1.5 h with a half-life of 7–9 h.46 Given its pharmacokinetics, it has to be taken twice a day. The loading dose is 180 mg and the maintenance dose is 90 mg twice daily.58 Of note, patients can be transitioned immediately from clopidogrel to ticagrelor without an interruption of antiplatelet effects, and do not require a load.46 It is recommended that, if taken with aspirin, the aspirin dose should not exceed 100 mg.
The pharmacodynamic response to clopidogrel is quite variable, with 20–40% of patients being classified as non-responders, poor responders, or resistant.59 More specifically, investigators have reported resistance in 31% within 24 h to 5 days of dosing, but this effect decreases to 15–20% at 30 days after a 300 mg load and then 75 mg daily,60 ,61 and remains in 18–19% of patients after 2 years.62 Clopidogrel resistance does not develop with time and a durable antiplatelet effect is maintained.63 The mechanism of resistance is thought to be secondary to variable absorption,64 as well as variable activation in the liver secondary to cytochrome P450 polymorphisms.59 ,65 Resistance to prasugrel is rare, and this drug is a recommended alternative for individuals with clopidogrel resistance.59 In the rare setting of prasugrel resistance, which may be secondary to hepatic metabolic polymorphisms or interference from antiretroviral drugs, ticagrelor is an attractive alternative.66 ,67 Lastly, as expected, a poor response to clopidogrel correlates with significant clinical effects in the cardiovascular arena leading to higher death, stent stenosis, myocardial infarction, and stroke.62
After stopping P2Y12 inhibitors, platelet function usually takes about 7–10 days to return to normal (lifespan of a platelet). For immediate restoration of platelet activity, a transfusion of two pooled packs of platelets (∼10 units) are required because the longer half-life of the inhibitors will consume the first pooled pack.68 Also, desmopressin results in partial reversal69 and is employed by many centers. Additionally, some cardiothoracic and orthopedic surgeons use aprotinin or tranexamic acid for reversal.70
Cyclic AMP inhibitors: dipyridamole (Persantine), dipyridamole + aspirin (Aggrenox), cilostazol (Pletal)
Mechanism of action
Dipyridamole and cilostazol inhibit the phosphodiesterase enzymes that break down cAMP, thereby increasing cAMP levels that block the platelet response to ADP and prevent platelet activation.71 ,72 In addition, dipyridamole blocks thromboxane synthase and the thromboxane receptor preventing thromboxane A2 formation,71 which inhibits platelet aggregation. Lastly, dipyridamole increases plasma adenosine levels73 and potentiates nitric oxide signaling through cyclic GMP,74 which inhibits platelet aggregation.
Pharmacokinetics and dosing
Following oral administration of dipyridamole, the time to peak plasma concentration is 1–2 h (mean 75 min),75 with a complex metabolic breakdown yielding a half-life of 10–13 h,76 with some reports up to 24 h.75 Given its poor bioavailability, an extended release form is typically prescribed and, in terms of stroke prevention, it has been formulated with aspirin as Aggrenox (25 mg aspirin + 200 mg dipyridamole) given twice daily, as determined by the European Stroke Prevention Study.77 ,78 Cilostazol is dosed at 100 mg twice a day and has peak concentrations at about 3 h, a half-life of 11 h, and achieves steady state in 4 days.72 In patients resistant to clopidogrel undergoing carotid stenting, adding cilostazol for triple therapy instead of increasing the clopidogrel dose resulted in a decrease in ischemic lesions and no increase in bleeding.79
Dipyridamole resistance has not been reported but, in its usually prescribed form as Aggrenox, patients can be expected to have resistance to the aspirin component as discussed above.
There are no clear reversal guidelines, but centers practicing NI procedures commonly administer two pooled packs of platelets (∼10 units) and one dose of desmopresssin (0.3 μg/kg).
Given the importance of antiplatelet agents in the endovascular suite, several new groups of drugs have been developed or are under development. Thrombin is the most potent platelet activator and thus blockade of this interaction would be paramount. Protease-activated receptor-1 (PAR-1; thrombin receptor) antagonists such as vorapaxar and atopaxar block thrombin-mediated platelet activation but do not interfere with thrombin-mediated cleavage of fibrinogen,80 meaning they are pure antiplatelet agents. They have shown mixed results in PCIs.81 ,82
Platelet function evaluation
The ability to assess the efficacy of an individual's platelet function is valuable for patient management to evaluate appropriate antiplatelet effects and also when emergency reversal is required for a hemorrhagic complication or need for emergency open surgery. Additionally, as described above, many individuals have either inherent resistance to some agents or are poor responders, meaning they are unprotected during procedures for which antiplatelets are recommended (eg, stenting). Below we discuss the range of assays available including platelet counting, platelet aggregometry, bleeding time, platelet function assays, and point-of-care assays.
This is the first-line test for platelet function which continues to be used by surgeons and interventionalists alike, with a goal count of ≥80–100 000/μL being the recommended cut-off to proceed with an intervention. The gold standard is a manual count via phase contrast microscopy,83 but the majority of centers employ automated cell counter methods, optical counting methods, or flow cytometry.84 However, the absolute platelet count does indicate functionality of the platelets. Platelet function can only be measured by formal functional assays.
Platelet function by platelet count ratio method
An indirect measurement of platelet function can be employed with available automated cell counters by stimulating platelets within anticoagulated whole blood samples with agonists such as ADP or epinephrine, inducing aggregation. A ratio between the platelet count in the control sample and the activated sample is then calculated, which correlates well with platelet aggregometry.85 This technique has been commercialized as the ICHOR Point of Care Hematology Counter for the acute setting.86
LTA was developed in the 1960s and has become the gold standard to test platelet function.87 In this assay, blood is centrifuged slowly to obtain platelet-rich plasma that is continuously stirred while a light source is shone through the sample. Platelet agonists such as ADP are then added to induce aggregation which results in increased light transmission, given a decrease in the turbidity of the platelet-rich plasma. A series of platelet agonists are employed through a range of concentrations to assess shape changes and aggregation responses. The test is labor-intensive and requires technical expertise, and may not be easily obtainable at all centers.88 Additionally, this test underestimates the degree of GP IIb/IIIa inhibitors,89 is not sensitive enough to test aggregates of <100 platelets (micro-aggregates), and is not reliable at detecting pre-existing aggregates, which is important in patients with hyperfunctioning of platelets.90 Resistance to various antiplatelet agents is monitored by selection of the appropriate platelet aggregation agonist. For example, aspirin therapy is monitored with ADP, arachidonic acid, or collagen,23 clopidogrel is monitored by platelet responses to ADP, and IIb/IIIa inhibitors are measured by the platelet response to ADP and thrombin receptor agonists.23
Whole blood aggregometry, flow cytometry, and laser platelet aggregometry
Because of the limitations of classic aggregometry, whole blood aggregometry was developed. In this assay, whole blood is stirred between two platinum electrodes which then become platelet-covered with aggregates after the addition of agonists, thereby changing the impedance in the circuit.91 This method is superior to measuring antiplatelet therapy with classic aggregometry, but is still insensitive to micro-aggregates.85 Nevertheless, the method is in widespread use in Europe.92 In order to assess small micro-aggregates, flow cytometry can be employed with high accuracy, but the equipment is expensive and requires a specialized operator. Employing a laser to detect platelet aggregates as small as a few platelets via light scattering has gained attention because of its accuracy,93 but its adoption has been limited.
This was the first test for platelet function and studies natural hemostasis using a simple technique used at the bedside. It requires a skin incision, often on the forearm, to a depth to disrupt capillaries. It has been used to predict surgical bleeding as it is a good test of platelet function.94 It does not require any special equipment but it has been criticized as invasive, insensitive, irreproducible, and with a subjective endpoint.95
Platelet function assays
In an effort to mimic in vivo physiology including platelet adhesion, activation, and aggregation, several platelet function assays have been developed. Many of them seek to mimic vessel wall damage and thus base their assays on shear-induced platelet activation.96 For example, the clot signature analyzer measures the ability of flowing non-anticoagulated blood to form hemostatic plugs via fibrin formation in a tube of punched holes.97 The thrombotic status analyzer test employs hemodynamic forces on blood drawn through a capillary tube to induce platelet activation and capillary tube occlusion.98 The more well-known platelet function analyzer (PFA) exposes platelets in whole blood to high shear stress within a capillary tube coated with collagen and ADP/epinephrine and monitors the drop in flow rate as the platelets form a hemostatic plug.99 The test is used as a screening tool to assess platelet abnormalities and is simple and rapid in execution, but is limited by the absolute platelet count, blood count, and von Willebrand factor levels.100 Thromboelastography (TEG) is another method of testing the efficiency of blood coagulation. It is mainly used in surgery and anesthesiology and has been established as a sensitive test for hemostatic function assessment in several clinical settings.101 Although few centers are capable of performing it, TEG can assess platelet function, clot strength, and fibrinolysis by triggering clot formation followed by computerized coagulation analysis.102 The TEG trace is analyzed for the reaction time, which represents the rate of initial fibrin formation, and for the maximal amplitude, which correlates with the absolute strength of the clot.103 The use of TEG in NI procedures is underutilized and rarely described in the literature.104
Point-of-care platelet function assays (ICHOR and VerifyNow)
The ICHOR Point of Care Hematology Counter86 employs a ratio to measure platelet functionality pre- and post-activation. The VerifyNow Assay, previously known as the Ultegra Rapid Platelet Function Assay, was developed based on the premise that, in order to achieve significant clinical efficacy with antiplatelet agents, ≥80% of platelet receptors need to be blocked.105 Thus, monitoring of receptor blockade is crucial to ensure optimal dosing for patient outcomes. This system allows for rapid measurements of aspirin,95 clopidogrel,95 and IIb/IIIa inhibitor function106 at the bedside using a whole blood sample. The assay is based upon fibrinogen-coated bead agglutination in response to the proportion of available GP IIb/IIIa receptors. For aspirin, the activator is arachidonic acid, for P2Y12 inhibitors it is ADP, and for IIb/IIIa inhibitors it is thrombin receptor activating peptides, all with results that are available in minutes. In 2014 the US Food and Drug Administration (FDA) recalled the VerifyNow IIb/IIIa test due to concerns in the reporting of erroneous platelet aggregation unit results.
Specific neurointerventional procedures
Elective coil embolization of cerebral aneurysms
Elective coil embolization of unruptured cerebral aneurysms carries a very low overall complication rate of 3.4–6.1%.107 Antiplatelet premedication with aspirin and clopidogrel has been proposed for reducing the thromboembolic risk and is used in some centers (Class C evidence).23 ,107 ,108 The hesitancy in applying DAT in elective coil embolization due to concerns regarding hemorrhagic complications was eased after our experience with DAT during stent-assisted coiling showed a lower risk of thromboembolic events with a relatively low risk of hemorrhage related to antiplatelet therapy.109 Responsiveness to aspirin and/or clopidogrel is quite variable, and although resistance has been reported to be associated with thromboembolic events in the setting of NI procedures, a general consensus on a definition for aspirin and clopidogrel resistance is still missing.110 Even though recent large cardiology studies failed to show clear clinical benefit to increasing platelet inhibition in resistant patients,111 ,112 the available data to answer this question in the NI population are still scant. A recent report on a cohort undergoing stenting procedures showed a non-significant decrease in thromboembolic events in patients whose clopidogrel dose was tailored to the VerifyNow assay.113 Moreover, the marked variability in individual response to routine clopidogrel premedication108 based on VerifyNow testing makes clopidogrel a less attractive antiplatelet in NI procedures. This inconsistency in both the clopidogrel dose-response effect and the VerifyNow test definitions/interpretation has prompted some neurointerventionalists to pursue more potent antiplatelet regimens derived from the cardiac literature (such as prasugrel and ticagrelor).49 ,50 However, the literature to support this practice is generally lacking (Class C evidence).
Stent-assisted coil embolization of cerebral aneurysms
The fundamental goal of stent-assisted coil embolization of cerebral aneurysms is to keep the cerebral blood flow hemodynamically stable after coil embolization. Even though self-expandable stents are used during coil embolization of wide-necked cerebral aneurysms in order to maintain a normal cerebral blood flow, its use poses a higher risk for thromboembolic complications. This is reflected in the World Federation of Interventional and Therapeutic Neuroradiology (WFITN) recommendation of pre- and post-treatment administration of DAT for a variable duration depending on the stent model used (Class C evidence).114 The routine standard of care monitoring of the response to antiplatelet agents in the setting of stent-assisted coil embolization, as in any other cerebrovascular disease, has not been proven, although clearly resistance to antiplatelet medications or a supratherapeutic effect can lead to complications.115 Moreover, the response to aspirin and clopidogrel may differ according to the method of measurement, which necessitates a standardized method of measurement of the resistance level such as the VerifyNow test.116 With a relatively low risk of hemorrhagic complications, DAT in the setting of stent-assisted coiling has a lower risk of thromboembolic events.109 Accordingly, in the setting of antiplatelet resistance, it is reasonable to increase the loading dose (eg, increase the loading dose of clopidogrel from 300 mg to 600 mg) and/or the maintenance dose (eg, increase the maintenance dose of clopidogrel from 75 mg/day to 150 mg/day),117 although this may be associated with a higher bleeding risk.118 Other options to overcome antiplatelet resistance include avoiding smoking,119 better control of other medical comorbidities,116 and avoiding administration of NSAIDs, competitors of COX inhibition, with aspirin. Finally, based on the cardiology literature, newer antiplatelet agents (such as prasugrel, ticagrelor, and cangrelor) and triple antiplatelet therapy with the administration of cilostazol are believed to be superior to conventional regimens in preventing stent thrombosis.120–122 The applicability of these new regimens in the NI population is yet to be proven.
Balloon-assisted coil embolization of cerebral aneurysms
Balloon-assisted coil embolization (BACE) was first described by Moret in 1997 as a safe technique for embolization of wide-necked aneurysms.123 Antiplatelet therapy in BACE remains an area of debate. Even though the WFITN did recommend post-treatment aspirin, pre-treatment loading with antiplatelet was not emphasized114 as the efficacy of premedication with an antiplatelet agent in the setting of BACE has not been established; the use of an antiplatelet loading regimen for BACE is still mostly center-dependent. A significant reduction in the rate of thromboembolic complications by clopidogrel premedication of patients undergoing BACE has previously been reported.124 In this report the use of clopidogrel alone or combined with aspirin was the only pre-procedural predictor of a significant difference in local thrombus formation (p=0.01) or symptomatic thromboembolic complication (p=0.04). The use of antiplatelet agents or antiplatelet function testing prior to BACE is still not supported and requires more investigation in a randomized prospective trial (Class C evidence).
Flow diverters such as the Pipeline embolization device (PED; Covidien/ev3, Irvine, California, USA), Silk flow diverter (SFD; Balt Extrusion, Montmorency, France), Flow Re-Direction Endoluminal Device (FRED; Microvention, Tustin, California, USA) and the Surpass flow diverter (StrykerNeurovascular, Fremont, California, USA) have been developed to treat wide-necked aneurysms (neck >4 mm) with unfavorable dome/neck ratios (<1.5).125 Flow diversion requires deployment of the stent within the lumen of the parent artery, allowing the artery to endothelialize along the stent to isolate the aneurysm from the circulation. The PED is currently the most studied and the only FDA-approved flow diverter in the USA. PED deployment is typically performed under DAT with aspirin and a P2Y12 receptor antagonist because of the potential risk of thromboembolic events, estimated to be 0–14%,5 ,14 ,126–140 due to either in situ stent thrombosis or distal embolization. PED deployment also carries a risk of hemorrhagic complications potentially due to DAT, estimated to be 0–11%,5 ,14 ,126–140 with parenchymal ICH being the most serious complication.
In most studies of the PED patients were treated with a daily regimen of aspirin 81–325 mg and clopidogrel 75 mg. Patients are usually pretreated for several days or loaded with aspirin 325–650 mg and clopidogrel 300–600 mg h prior to the procedure. DAT is continued for at least 6 months after the procedure in most studies. Aspirin is typically continued indefinitely while clopidogrel may be stopped, depending on angiographic and clinical results. In studies addressing the use of platelet aggregometry, point-of-care platelet function testing such as VerifyNow used P2Y12 inhibition of 20–40% as a minimum degree of pre-procedure inhibition required.51 ,135 ,137 However, since August 2012, VerifyNow assays report the PRU value only. The Assessment of Dual Anti-Platelet Therapy with Drug Eluting Stents (ADAPT-DES) registry proposed that the optimal clopidogrel therapeutic window to minimize both ischemic and hemorrhagic complications after coronary artery stenting is 95–207 PRU.141 Most recently, a few retrospective reports have suggested that a therapeutic window of 60–240 PRU could reasonably be used in the setting of flow diversion with PED to reduce both thromboembolic events in the ‘clopidogrel hyporesponsive population’ and hemorrhagic complications in the ‘clopidogrel hyper-responsive population’.5 ,14 However, a PRU therapeutic window is yet to be validated in randomized controlled studies.
Neurointerventional complications of antiplatelet agents
Antiplatelet agents prevent the formation of platelet-rich white thrombi, especially during intimal wall disruption. Given the widespread use of mono, dual, or triple antiplatelet therapy to help prevent thromboembolic complications in NI procedures, an assessment of actual antiplatelet complications might be warranted. The question remains whether the reduction in thromboembolic complications outweighs the risks of hemorrhagic complications. Furthermore, the duration of treatment with dual or triple antiplatelet therapy in patients who have undergone NI procedures is not well-defined and thus the long-term hemorrhagic outcomes are not known. Here we review the current data, with several outcome measures extrapolated from the interventional cardiology literature. DAT is used for at least 1 month in the majority of cases and some practitioners continue beyond 1 month because of concerns for late thromboembolic events while others discontinue DAT earlier given the concern for intracranial hemorrhage.
The Management of Atherothrombosis with Clopidogrel in High-risk patients (MATCH) trial found that adding aspirin to clopidogrel in high-risk patients with recent ischemic stroke had no significant impact on ischemic outcome, but life-threatening intracranial hemorrhage was higher in the DAT group.142 This was confirmed in a subsequent trial,143 but a meta-analysis demonstrated that aspirin and clopidogrel DAT only had an increased rate of intracranial hemorrhage when compared with clopidogrel but not with aspirin monotherapy.144
Similarly, the Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management, and Avoidance (CHARISMA) trial which compared aspirin and clopidogrel versus aspirin monotherapy found a non-significant increase in intracranial hemorrhage in the DAT group.52 Conversely, the Fast Assessment of Stroke and Transient Ischemic Attack to Prevent Early Recurrence (FASTER) trial found a higher risk of intracranial hemorrhage with aspirin and clopidogrel that was not offset by the decreased risk of ischemic stroke.145 These studies assessed patients with ischemic stroke and not specifically patients with stents. Although a potential complication of stenting is thromboembolic events, the direct applicability of these studies remains unclear.
Congruently, the Japanese Registry of Neuroendovascular Therapy (JR-NET) is a nationwide survey from January 2005 to December 2009 that assessed periprocedural adverse events.146 They found that, as antiplatelet therapy became standard in the later years, there was a significantly decreased rate of ischemic complications (4.2–2.1%) but a significantly increased rate of intracranial hemorrhagic complications (2.1–5.3%), as well as a significantly increased rate in death or severe disability (1.5–2.1%).94 More recently, a study of 110 patients treated with neuroendovascular angioplasty and/or stent placement maintained on DAT with aspirin and clopidogrel did not find a significantly increased bleeding rate at 3 months, with only one intracranial hemorrhage in the cohort.147 In terms of carotid artery stenting, DAT with clopidogrel plus aspirin in a series of 139 patients only found one intracranial hemorrhage at 30 days and an ischemic stroke or death rate of 4.3%.148 Similarly, a randomized controlled trial of 47 patients with carotid artery stenosis (>70%) showed that clopidogrel and aspirin significantly reduced the incidence of adverse neurological outcomes at 1 month without an additional increase in bleeding.149 The Stenting of Symptomatic Atherosclerotic Lesions in the Vertebral or Intracranial Arteries (SSYLVIA) trial150 recommends aspirin (minimum 100 mg) for 1 year and clopidogrel (75 mg) for at least 4 weeks after stenting, but given concerns for late stenting, the SAMMPRIS (Stenting vs Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis) trial recommends DAT for 3 months post-procedure.151 However, only limited large trial data beyond 1 month for DAT following neuroendovascular procedures are available. Caution is currently advised, as large trials in patients with a history of ischemic stroke maintained on DAT have a higher rate of intracranial hemorrhage. These trials include the CURE (Clopidogrel in Unstable angina to prevent Recurrent Events),152 CREDO (Clopidogrel for the Reduction of Events During Observation),153 and COMMIT (Clopidogrel and Metoprolol in Myocardial Infarction)154 trials, in addition to the already mentioned CHARISMA and FASTER trials. The MATCH trial showed that the incidence of primary intracranial hemorrhage between the two treatment groups is the same for the first 3 months, but increases in the DAT group after 3.5 months.
In summary, the use of DAT following a neuroendovascular procedure may decrease the risk of thromboembolic complications, but this may be at the expense of increasing the risk of hemorrhagic complications. More controlled studies specific to the NI patient population are needed to clarify this important issue.
Limitations of current testing
One major limitation in the evaluation of platelet function in patients treated with antiplatelet medication is the lack of consensus on the optimal method of measuring platelet reactivity. The clinical tests in current use for measurement of platelet function all have some limitations. The more established test or ‘gold standard’ is LTA,155 but it has major disadvantages including cost, slow assay time, and poor reproducibility, with the need for sample preparation and a skilled technician. Because the performance of the test (including the concentration of ADP used and the platelet concentration) is not standardized, there is no universal metric for the detection of high platelet reactivity.156 Flow cytometry can be used to measure various markers of platelet activity such as vasodilator-stimulated phosphoprotein phophorylation. Again, this is labor-intensive and not standardized. Ease of use is a major advantage of the newer tests that are now more widely used. There are point-of-care devices that can be used at the patient's bedside. Two of the more commonly used assays are the PFA-100 analyzer and the VerifyNow system. The VerifyNow assay is a simple rapid method that has been shown to correlate (r∼0.7) with LTA measurements.157 It is commonly used in the cardiology setting where much of the published data originate. In the neurovascular literature, this is also the most commonly used test.2–5 ,10 ,14 ,50 ,51 ,108 ,116 ,158 ,159
Data for thresholds on antiplatelet testing
Because there is no standard definition for high on treatment platelet reactivity, neurovascular specialists have relied on data from cardiovascular patients. However, even in this literature, there is again no clear cut-off value associated with adverse clinical events. Bonello et al156 addressed these concerns and described a range of values in the literature over which platelet activity appears to be safe for PCI. Initial studies using LTA used a change in platelet aggregation of ≤10% as a definition of antiplatelet medication resistance.9 Several studies have been published regarding the level of platelet activity measured by the VerifyNow P2Y12 assay and risk of thromboembolic complications in the PCI patient population.13 ,160 ,161 This relationship has been confirmed in numerous studies using other types of platelet activity measurements. The cut-off value for subsequent thrombotic events using the VerifyNow assay has been evaluated using receiver operating characteristic curve analysis. A PRU cut-off value of around 230 appears to predict thrombotic events of all types in the cardiovascular literature.13 ,160 ,161 ADAPT-DES is a recently published registry of patients treated with drug eluting cardiac stents aimed to determine the relationship of PRU and clinical outcomes in a study of 8583 patients in the USA. It suggests a target window of PRU 95–208 in order to minimize the risk of both ischemic and hemorrhagic complications.141 Interestingly, although PRU levels correlated with stent thrombosis, there was no independent correlation with 1-year mortality in this cardiovascular patient population. The applicability of these data to the neurovascular patient population is yet to be proven.
NI patients and cardiovascular patients differ in demographics and risk factors. The clinically significant risk of cardiovascular stenting is largely stent thrombosis, as well as potential risks associated with DAT such as gastrointestinal tract hemorrhage, intracranial hemorrhage, and epistaxis. In neurovascular patients, hemorrhagic as well as thromboembolic complications are serious concerns. While these are important differences, lessons can be learned from the similarities in the data. The rates of platelet activity in neurovascular patients on antiplatelet medications have been found to be similar to those in cardiovascular patients.2 ,3 ,10 ,108 ,116 ,158 ,159 The neurovascular data linking platelet activity and complications are currently limited but are increasing in volume. One report of 48 PED procedures found that a PRU value of >240 predicted thromboembolic events while a PRU value of <60 predicted hemorrhagic events in the follow-up period.5 ,14 In addition, the same PRU cut-off values predicted thromboembolic and hemorrhagic events in 100 patients undergoing different types of elective endovascular treatment of cerebral aneurysms.108 Similarly, an association between a hyper-response of ≥72% platelet inhibition using the VerifyNow assay and hemorrhagic complications has been reported.4 Notably, the ADAPT-DES trial found an association between lower PRU values and clinically relevant bleeding in patients treated with a cardiac stent.141
Increasing platelet inhibition
Various studies have demonstrated that resistant patients can have platelet inhibition increased in a number of ways. The first is increased medication compliance. Second, increased doses of aspirin or clopidogrel may reduce resistance in some patients. In addition, switching to a different antiplatelet agent can reduce platelet activity. In patients resistant to clopidogrel, double dosing has been shown to produce therapeutic platelet activity in some patients.3 ,13 ,162 However, this reduction in platelet activity has not decreased thrombotic complications. The two randomized clinical trials which studied this are both in the cardiology literature. In GRAVITAS, resistant patients were randomized to receive either standard or double dose clopidogrel after PCI and no difference was seen between the two groups in the number of deaths or the development of myocardial infarction or stent thrombosis.13 However, patients were given double dose clopidogrel and not targeted to a specific PRU level. In the ARCTIC trial, patients undergoing PCI were given additional clopidogrel loading or prasugrel. Again there was no difference in outcomes. However, PRU values <230 were not achieved in 15% of patients on follow-up testing. There are limited data on modification of treatment in neurovascular patients. One study found a non-significant decrease in thromboembolic complications with double dose for patients with ≤20% platelet inhibition.3 Nevertheless, if changes to the dose or type of P2Y12 receptor antagonist administered are made in order to increase the degree of platelet inhibition prior to a NI procedure, obtaining follow-up platelet reactivity testing could be considered to minimize the risk of trading thromboembolic complications for hemorrhagic complications.
Decreasing platelet inhibition
There are no reliable data regarding hyper-responders to clopidogrel and reduction of clopidogrel dosing to achieve more platelet activity. In one study, reduction of clopidogrel dosing to every other day, every third day, twice weekly, or once weekly to achieve a PRU value in the optimal range was described.14 However, the correlation between such clopidogrel dose reductions to achieve a target PRU value and reduction in hemorrhagic complications is yet to be proven.
We have summarized the currently available antiplatelet medications available, their mechanisms of action and typical dosages. We have also summarized the currently available testing methodology for platelet inhibition and reviewed the current literature in this area. Both the cardiology and NI literature indicate that subtherapeutic platelet inhibition may lead to thrombotic complications and supratherapeutic inhibition may lead to a higher hemorrhage complication rate. There are a few studies associating the results of platelet function testing to thromboembolic and hemorrhagic complications in the NI field.
However, at present there are insufficient data to recommend routine platelet function testing prior to NI procedures. Individual centers may choose to use testing for their local practice, and we encourage those centers to collect and publish their data. Having a larger body of literature specific to NI patients will allow us to better understand the role of antiplatelet testing in the management of our patients.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.
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.