Article Text

Download PDFPDF

Original research
Relationship between low response to clopidogrel and periprocedural ischemic events with coil embolization for intracranial aneurysms
  1. Takumi Asai1,
  2. Shigeru Miyachi1,2,
  3. Takashi Izumi1,
  4. Noriaki Matsubara1,
  5. Kenichi Haraguchi1,3,
  6. Takashi Yamanouchi1,
  7. Keisuke Ota1,
  8. Kazunori Shintai1,
  9. Hayato Tajima1,
  10. Toshihiko Wakabayashi1
  1. 1Department of Neurosurgery, Nagoya University Graduate School of Medicine, Nagoya, Japan
  2. 2Department of Neurosurgery and Endovascular Neurosurgery, Osaka Medical College, Takatsuki, Japan
  3. 3Department of Neurosurgery, Toyohashi Municipal Hospital, Toyohashi, Japan
  1. Correspondence to Dr S Miyachi, Department of Neurosurgery, Osaka Medical College, 2-7 Daigaku-cho, Takatsuki City, Osaka 569-8686, Japan; neu137{at}


Objectives Low response to antiplatelet drugs is one of the risk factors for ischemic events. We examined the influence of low response to clopidogrel on symptomatic ischemic events and new ischemic MRI lesions with endovascular intracranial aneurysmal coil embolization.

Materials and methods Between August 2010 and July 2013, 189 procedures in 181 consecutive patients who underwent endovascular coiling and received clopidogrel before treatment were investigated retrospectively. Platelet aggregation activity was examined by VerifyNow analysis. Low response to clopidogrel was defined as P2Y12 reaction units ≥230 in this study. Symptomatic ischemic complications within 30 days and postoperative new ischemic lesions on MRI–diffusion weighted imaging were evaluated.

Results 66 of 189 (34.9%) cases were low responders to clopidogrel. Ischemic complications occurred in 2 of 66 (3.0%) low responders compared with 6 of 123 (4.9%) responders (p=0.72). A new high intensity spot larger than 5 mm was significantly more frequent in low responders (26 of 66; 39.4%) than in responders (26 of 121; 21.2%; p=0.01). On multivariate analysis, independent risk factors for larger new ischemic lesions were low response to clopidogrel, smokers, posterior location, and aneurysms with a larger neck.

Conclusions Low response to clopidogrel had little effect on clinical outcome although it increased asymptomatic large ischemic lesions in this cohort.

  • Drug
  • Embolic
  • Coil
  • Aneurysm
  • MRI
View Full Text

Statistics from


In recent years, neuroendovascular treatments have been widely used with a variety of improved devices, as well as enhanced angiographic systems. Coil embolization for intracranial aneurysms is a representative example of such a treatment. Although endovascular embolization has shown improved outcomes over the past decade, perioperative thromboembolic complications cannot be overlooked.1 One of the causes of thromboembolic events is injury of the intimal membrane of the artery by the devices. Another cause is metallic materials, such as stents or coils, deployed in the arterial lumen.2–4 To prevent such complications, antiplatelet drugs, mainly clopidogrel and aspirin, have been used empirically, and previous studies have reported the effectiveness of antiplatelet therapy.5–7

Recently, resistance to antiplatelet drugs has developed,8 ,9 and many articles have noted that resistance to antiplatelet drugs is one of the risk factors for ischemic events, mainly in percutaneous coronary interventions (PCI).10 ,11 However, only a few case series have discussed this subject in neurointervention.12 ,13

In this study, we measured the effectiveness of antiplatelet drugs before endovascular operation using the VerifyNow system (Accumetrics, San Diego, California, USA). The relationship between resistance to clopidogrel and symptomatic ischemic events or new ischemic lesions on postoperative MRI was investigated.

Materials and methods

Patient population

Between August 2010 and July 2013, 189 procedures in 181 consecutive patients, who received clopidogrel orally for at least 5 days before treatment and whose results of VerifyNow P2Y12 testing were available, were included in this single center retrospective study. Patient background characteristics, comorbidities, medications, and results of blood chemistry studies were investigated retrospectively through their clinical records. Data were analyzed by each procedure, not by individual.

VerifyNow analysis

Platelet function was measured with the P2Y12 assay, and the aspirin assay if the patient received aspirin, just before the procedure. By measuring the rate and extent of changes in light transmittance caused by platelets aggregating in whole blood samples, the VerifyNow system determines the level of platelet function in response to agonists (ADP/arachidonic acid). Blood samples with inhibited platelets produce a low level of light transmittance.10 ,14 The result of the platelet function assay is expressed as P2Y12 reaction units (PRUs) for clopidogrel and aspirin reaction units (ARUs) for aspirin. A low response to clopidogrel was defined as PRU ≥230 in this study. This was referred from the Gauging Responsiveness with A VerifyNow P2Y12 assay—Impact on Thrombosis And Safety (GRAVITAS) study.10

Endovascular treatment

Antiplatelet drugs were started usually 7 days, and at least 5 days, before the procedures. Our standard antiplatelet therapy was dual medications, commonly clopidogrel and aspirin, but in some cases single or triple antiplatelet medications were used depending on the risk of ischemic events or allergy to these medicines. After inserting the sheath, heparin was injected intravenously, with a target activated clotting time of 250–350 s. After the procedure, antiplatelet drugs were continued.

In this study, 189 endovascular saccular coil embolizations for intracranial aneurysms were included. Cases of parent artery occlusion or the acute phase of a ruptured aneurysm were excluded. Stent assisted coil embolization was used in 60 of 189 (32%) cases. In our institution, argatroban (60 mg daily) was continuously infused intravenously for 24 h after the procedure when the stent was placed.

MRI study

Brain MRI was checked approximately 4–5 days after the procedure in almost all patients. MRI studies were performed on 1.5 T and 3.0 T systems. A new ischemic lesion was defined as a high intensity signal area on diffusion weighted imaging (DWI) and a low intensity signal area on apparent diffusion coefficient maps. When a new ischemic lesion was present, its maximum diameter was measured.


A clinical outcome was defined as a symptomatic ischemic event that occurred within 30 days postoperatively and continued over 24 h, and outcomes were compared between the two groups (responders and low responders). New ischemic lesions were evaluated by MRI–DWI, and new ischemic lesions larger than 5 mm in diameter were also investigated.

Statistical analysis

Categorical variables are expressed as counts (percentages) and evaluated with the χ2 test or Fisher's exact test. Continuous variables are expressed as means±SD and evaluated with two sided unpaired t tests or Mann–Whitney U tests. Multiple logistic regression analysis was performed to identify independent correlates of events. Significance was defined as p<0.05. Analyses were performed with IBM SPSS Statistics V.20 (IBM, Chicago, Illinois, USA).


For all 189 cases, mean PRU was 189.9±80.4 (range 5–370, median 195), and 66 (34.9%) were low responders (PRU ≥230).

Baseline demographic and clinical characteristics are shown in table 1. The percentage of male patients, current smokers, and mean hemoglobin level were significantly different between the two groups. Although low responders included more patients with a past history of hypertension or peripheral artery disease than responders, the difference was not significant.

Table 1

Baseline patient characteristics

In all cases, dual antiplatelet therapy was administered in 158 (83.6%) cases, single in 25 (13.2%), and triple in 6 (3.2%). The combinations of antiplatelet drugs were not different between the two groups (p=0.12).

Results of the VerifyNow analysis, characteristics of the endovascular procedures, and MRI–DWI studies are shown in table 2. Mean PRU of low responders was higher than that in responders (277.1±33.4 vs 143.2±55.3). In contrast, mean ARU and number of cases of aspirin resistance were not significantly different. There were no significant differences between the two groups regarding profiles of aneurysms, characteristics of procedures, or MRI studies.

Table 2

Results of VerifyNow analysis and procedure characteristics

Two symptomatic ischemic events (3.0%) occurred in low responders compared with 6 (4.9%) in responders (p=0.72). The rate of new ischemic lesions on MRI–DWI was 66.7% in low responders and 57.0% in responders (p=0.26). However, the rate of new ischemic lesions larger than 5 mm in diameter was significantly higher in low responders than in responders (39.4% vs 21.2%, p=0.01) (table 3). The number of new MRI–DWI spots were more frequently found in low responders than in responders, but the difference was not significant (4.4±8.5 vs 6.2±10.5, p=0.09).

Table 3

Ischemic events within 30 days and results of MRI–DWI

In the logistic regression model (table 4), posterior location (OR 17.5, 95% CI 1.99 to 154, p=0.01) and operation time (1 h longer, OR 2.34, 95% CI 1.11 to 4.93, p=0.03) were independent risk factors for symptomatic ischemic events. On the other hand, low response to clopidogrel was not a risk factor. Low response to clopidogrel (OR 3.35, 95% CI 1.53 to 7.32, p=0.002), current smoker (OR 3.29, 95% CI 1.40 to 7.74, p=0.006), posterior location (OR 2.59, 95% CI 1.10 to 6.14, p=0.03) and neck size (1 mm longer, OR 1.35, 95% CI 1.15 to 1.58, p<0.001) were independent risk factors for postoperative new large ischemic lesions on MRI–DWI. Aspirin resistance was not related to any ischemic event.

Table 4

Results of multiple logistic regression analysis


This study has demonstrated the relationship between low response to clopidogrel and the incidence of large new ischemic lesions on MRI–DWI. However, the incidence of symptomatic ischemic events or any new ischemic lesions was not significantly different between responders and low responders.

Despite the evolution of the devices and techniques used for endovascular aneurysmal coil embolization and improvement in the results, ischemic complications are a serious issue. Thus use of antiplatelet drugs has been recommended to prevent perioperative thromboembolic events.5–7 In spite of the use of antiplatelet drugs, thromboembolic events continue to occur. One systematic review on endovascular treatment of unruptured aneurysms1 reported that the incidence of symptomatic ischemic complications was 6.9%.

Over the past decade, response variability to clopidogrel has been reported in some publications. Gurbel et al8 showed interindividual variability in the platelet inhibitory response to clopidogrel in patients undergoing elective PCI. Matetzky et al9 reported that up to 25% of ST segment elevation myocardial infarction patients undergoing primary PCI with stenting were resistant to clopidogrel and might be at increased risk of recurrent cardiovascular events. The mechanism of low response to clopidogrel is multifactorial, including poor compliance, variable absorption of the prodrug, clearance of active metabolite, and drug interactions.15

Because of the variability in the response to antiplatelet drugs, measuring the degree of platelet function is helpful for endovascular treatment. However, laboratory tests such as light transmission aggregometry are traditionally utilized for evaluation and as an historical gold standard, but their use is limited to specialized laboratories. VerifyNow is one of the point of care testing devices that measures platelet inhibition. The VERIfy Thrombosis risk ASsessment (VERITAS) study16 showed that VerifyNow is a reliable, fast, and sensitive device for monitoring platelet inhibition. However, Gaglia et al17 demonstrated that the VerifyNow system showed only moderate agreement with light transmission aggregometry, the gold standard method.

Some studies showed a relationship between low response to clopidogrel and ischemic events after PCI. The GRAVITAS trial10 demonstrated that achievement of PRU <208 at 12–24 h after PCI or during follow-up was associated with a lower risk of cardiovascular events at 60 days (HR 0.23, p=0.047). The ARMYDA-PROVE (Antiplatelet therapy for Reduction of MYocardial Damage during Angioplasty-Platelet Reactivity for Outcome Validation Effort) study11 also found a higher risk of ischemic events in patients with PRU ≥239 than with PRU <239 (AUC 0.68, 95% CI 0.61 to 0.76, p<0.0001). Meanwhile, some trials have failed to show any relationship.18 ,19

Only a few studies have dealt with the relationship between resistance to antiplatelet agents and ischemic events in neurointervention. Kang et al12 measured PRU with VerifyNow in 189 consecutive patients who underwent elective coil embolization. In their study, patients were stratified into four quartiles according to PRU, and thromboembolic events increased, from 4.3% in the first quartile (PRU <240), to 2.2% in the second quartile (PRU 240–284), to 6.5% in the third quartile (PRU 285–332), and to 17.0% in the fourth quartile (PRU >332). Fifi et al13 showed that resistance to clopidogrel was an independent risk factor for thromboembolic complications in endovascular procedures with stent implantation.

In the present study, the relationship between low response to clopidogrel and symptomatic ischemic complications was not significant. Eight symptomatic ischemic events (4.2%) occurred, and mean PRU values were lower in patients with events than in those without events (153.2 vs 191.6, p=0.25). Posterior location and longer procedure time were independent risk factors for ischemic events. These factors affected ischemic events more strongly than variability of antiplatelet inhibition. The reasons for ischemic complications were perforator infarction (2 cases), in-stent thrombosis and distal embolization (3 cases), and thromboembolism from wedged guiding catheter in the vertebral artery (3 cases).

However, new MRI–DWI spots larger than 5 mm were detected more frequently in low responders; therefore, there might be a difference in the strength of platelet aggregation that depends on the response to clopidogrel and reduces the size of thrombi in the arterial lumen. This suggests that response to clopidogrel contributes to reducing the area of cerebral infarction and decreasing potential neurological symptoms.

Recently, the pharmacodynamic mechanism of clopidogrel has been thought to be an important cause of low response. Clopidogrel is an inactive prodrug that requires several biotransformation steps by cytochrome P450 isozymes.20 A loss of function polymorphism in CYP2C19 has been reported in several studies21 to be associated with a higher level of ADP induced platelet aggregation. In particular, these low functional polymorphisms of CYPC19 are more frequent in Asian populations (13–23%) than in Caucasian populations (1–6%).22 Nakata et al23 reported that 60% of Japanese patients had a CYP2C19 loss of function genotype and higher platelet activity than patients who had none of these genotypes.

There were some study limitations. Firstly, low response to clopidogrel was defined as PRU ≥230, but there is still no consensus on the cut-off PRU value. Secondly, four patterns of combinations of antiplatelet drugs were included in this study, so the results between response to clopidogrel and thromboembolic events might be diluted. Thirdly, this was a retrospective study. Twelve of 189 cases (6.3%) received additional medication (cilostazol or intra/post-procedural administration of intravenous ozagrel sodium or argatrovan). However, this may not have affected the results of the study.


No significant difference in clinically symptomatic ischemic event rates were seen between responders and low responders. Low response was defined by a VerifyNow PRU value of 230 in this cohort. However, low response was associated with an increased frequency of larger asymptomatic ischemic lesions detected by MRI–DWI after the procedures. In summary, low response to clopidogrel had little effect on clinical outcome, although it may increase ischemic lesions.


View Abstract


  • Contributors Concept and design of the study: TA, SM, and TI. Supervision of the study: TW. Preparing the paper: TA, SM, and NM. Data collection and analysis: TA, TI, NM, KH, TY, KO, KS, and HT.

  • Competing interests None declared.

  • Ethics approval The study was approved by the institutional review board of the Nagoya University Graduate School of Medicine.

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

  • Data sharing statement There are no additional unpublished data in this article.

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.