Author + information
- Received December 22, 2004
- Revision received July 11, 2005
- Accepted July 19, 2005
- Published online November 15, 2005.
- Paul A. Gurbel, MD, FACC⁎ (, )
- Kevin P. Bliden, BS,
- Waiel Samara, MD,
- Jason A. Yoho, MD,
- Kevin Hayes, MD,
- Mulugeta Z. Fissha, MD and
- Udaya S. Tantry, PhD
- ↵⁎Reprint requests and correspondence:
Dr. Paul A. Gurbel, Sinai Center for Thrombosis Research, Hoffberger Building, Suite 56, 2401 West Belvedere Avenue, Baltimore, Maryland 21215.
Objectives We investigated whether patients who suffered subacute stent thrombosis (SAT) have higher post-treatment reactivity than those who do not encounter stent thrombosis.
Background High post-treatment platelet reactivity has been reported after coronary stenting after clopidogrel therapy and may be an important factor in the occurrence of SAT.
Methods We identified patients with SAT treated at two tertiary care centers over a 1.5-year period. Light transmittance aggregation induced by adenosine diphosphate (ADP) and arachidonic acid, total and activated glycoprotein (GP) IIb/IIIa after stimulation with ADP, and vasodilator-stimulated phosphoprotein phosphorylation levels to measure P2Y12receptor inhibition were determined (n = 20) and compared with an age-matched group of patients without SAT (n = 100). High post-treatment platelet reactivity was defined as >75th percentile ADP-induced aggregation in the group without SAT.
Results The SAT patients had higher mean platelet reactivity than those without SAT by all measurements (p < 0.05): 49 ± 4% versus 33 ± 2% for 5 μmol/l ADP-induced aggregation and 65 ± 3% versus 51 ± 2% for 20 μmol/l ADP-induced aggregation (p < 0.001), 69 ± 5% versus 46 ± 9% for P2Y12reactivity ratio (p = 0.03), and 138 ± 19 mean fluorescence intensity (MFI) versus 42 ± 4 MFI for stimulated GP IIb/IIIa expression (p < 0.001). Of patients with SAT, 60% had high platelet reactivity.
Conclusions High post-treatment platelet reactivity and incomplete P2Y12receptor inhibition are risk factors for SAT. Measures to uniformly determine platelet reactivity after coronary stenting and treatment strategies to improve P2Y12receptor inhibition in patients with high post-treatment platelet reactivity should be further investigated.
Clopidogrel and aspirin are the major antiplatelet drugs of choice to prevent stent thrombosis (SAT) (1). Reports indicate that acute and subacute thrombosis are significant clinical problems despite therapy with thienopyridines and aspirin (2–4). We have recently demonstrated that a significant percentage of patients after stent implantation had high post-treatment platelet reactivity (5). Furthermore, we also demonstrated that pretreatment platelet reactivity strongly affected the final platelet reactivity after clopidogrel therapy (5,6). Recent studies have suggested that non-responsiveness to clopidogrel and high platelet reactivity might be risk factors for post-stent ischemic events and stent thrombosis (7–9). These findings suggest that an insufficient anti-platelet effect might occur in a percentage of patients undergoing stenting treated with clopidogrel, and those patients with high reactivity despite clopidogrel therapy are at the greatest risk of ischemic events.
We hypothesized that patients who have suffered stent thrombosis have higher post-treatment platelet reactivity than those who do not encounter stent thrombosis. To investigate this hypothesis, we measured platelet reactivity by multiple methods in patients receiving clopidogrel therapy who had experienced stent thrombosis and compared this group with a group of age-matched patients who had undergone stent implantation without the occurrence of thrombosis.
Patients and blood samples
This study was approved by the investigational review boards of our respective hospitals. We identified all of the cases of stent thrombosis (n = 30) by searching the medical records of patients who underwent coronary stenting in the last 1.5 years at the Sinai Hospital of Baltimore and Union Memorial Hospital, Baltimore, Maryland. Among the 30 SAT patients, 20 patients underwent analyses of platelet reactivity. Stent thrombosis was defined by the sudden onset of coronary artery occlusion in a stented vessel resulting in hospitalization and judged by the treating interventionalist as due to thrombosis.
We compared the platelet reactivity in patients with SAT (n = 20) to patients without SAT (n = 100). In patients without SAT, platelet studies were performed 5 to 14 days after procedure and these patients were enrolled consecutively. In patients with SAT, the average time from the occurrence of stent thrombosis to the initial blood drawn for laboratory evaluation was 218 ± 204 days. In-hospital patients identified with SAT had blood drawn at day five after the procedure. In patients with SAT already receiving a maintenance dose of clopidogrel (75 mg q.d.), blood samples were evaluated on the day of arrival at our center. Those patients (n = 2) not receiving clopidogrel at the time of the study were reloaded with 300 mg, maintained on 75 mg/day, and returned for blood sampling five days later.
Blood was drawn with a 21-gauge needle and placed into Vacutainer blood collecting tubes (Becton Dickinson, Franklin Lakes, New Jersey) containing 3.8% trisodium citrate after discarding the first 2 to 3 ml of free-flowing blood. The Vacutainer tube was filled to capacity and gently inverted three to five times to ensure complete mixing of the anticoagulant. No patients in the study had been receiving glycoprotein (GP) IIb/IIIa inhibitors or anticoagulants within 96 h of blood sampling. All patients were receiving aspirin (81 to 325 mg q.d.) except two patients in the SAT group.
Platelet reactivity measurements
Platelet aggregation was determined by conventional light transmittance aggregometry (model 490, Chronolog Aggregometer with AggregoLink software; Havertown, Pennsylvania) in response to 5 and 20 μmol/l ADP and 1 mmol/l arachidonic acid with standard methods as previously described (10).
GP IIb/IIIa receptors
The surface expression of platelet receptors was determined by whole blood flow cytometry with a multicolor analysis method (Immunocytometry Systems, Cytometry Source Book, BD Biosciences, San Diego, California) with the following monoclonal antibodies: fluorescein isothiocyanate (FITC)-conjugated PAC-1 (recognizes the active GP IIb/IIIa receptor) and R-phycoerythrin (R-PE) conjugated CD41a (recognizes the total GP IIb/IIIa receptor population). Antibodies were obtained from BD Biosciences. The blood-citrate mixture was stimulated with 5 μmol/l ADP for 2 min. Saturating concentrations of respective antibodies were added to unstimulated and simulated blood, and the tubes were incubated at room temperature for 20 min in the dark. The labeled samples were fixed by the addition of 1% buffered paraformaldehyde and stored at 4°C for at least 2 h. The labeled samples were analyzed by a Becton Dickinson FACScan flow cytometer, set up to measure fluorescence light scatter that was calibrated daily with fluorescence beads for the multicolor flow cytometer setup (CaliBRITE 3, BD Biosciences). After setting the gate around platelets, FL1(FITC)/FL2(R-PE) compensations were adjusted. All the variables were collected with four-decade logarithmic amplification. The data were collected in list mode and then analyzed with CELL Quest Software (Becton Dickinson Biosciences). Total and activated GP IIb/IIIa receptor levels were expressed as log mean fluorescence intensity (MFI) (11). Measurement of the PAC-1 binding to the activated GP IIb/IIIa receptor on platelets is an established method to determine the effect of ADP on the P2Y12receptor. Activated GP IIb/IIIa receptor binding to fibrinogen is the final step during platelet aggregation in response to ADP. Therefore, the measurement of activated GP IIb/IIIa receptor expression directly indicates the effectiveness of ADP binding to P2Y12receptors (12).
The measurement of vasodilator-stimulated phosphoprotein (VASP) phosphorylation levels is a marker of P2Y12receptor reactivity and, thus, clopidogrel-induced inhibition (9,13–16). Prostaglandin E1(PGE1) increases VASP phosphorylation levels by stimulation of adenylate cyclase. Binding of ADP to P2Y12leads to Gi-coupled inhibition of adenylate cyclase. Therefore, the addition of ADP to PGE1-stimulated platelets reduces PGE1-induced VASP phosphorylation levels. If P2Y12receptors were successfully inhibited by clopidogrel, addition of ADP will not reduce the PGE1-stimulated VASP phosphorylation levels. This principle is used in the present method where phosphorylated VASP levels were quantified with labeled monoclonal antibodies by flow cytometry with the Platelet VASP-FCM kit (Biocytex Inc., Marseille, France). The P2Y12reactivity ratio is calculated after measuring the VASP phosphorylation levels after stimulation with PGE1(MFI PGE1) and also PGE1+ ADP (MFI PGE1+ ADP). The P2Y12reactivity ratio = ([MFI PGE1] − [MFI PGE1+ ADP]/[MFI PGE1]) × 100%. Thus, as the ratio falls, there is clopidogrel-induced inhibition of the P2Y12receptor.
High post-treatment platelet reactivity was defined as >75th percentile for 5 and 20 μmol/l ADP-induced aggregation as measured in the group without SAT. This definition is stricter than our previous definition, where the highest tertile of aggregation was used to define high platelet reactivity (5). High post-treatment platelet reactivity was also defined as >75th percentile of VASP and stimulated active GP IIb/IIIa expression values. Aspirin resistance was defined as >20% change in light transmittance after stimulation with arachidonic acid (16).
Comparisons were made between groups by unpaired ttests for continuous variables and by Fisher exact test for categorical variables (statistical software by StatSoft Inc., Tulsa, Oklahoma). Regression analysis with calculation of the Pearson correlation coefficient was used to correlate platelet aggregation with the other markers measured (statistical software by StatSoft Inc.). The Wilks-Shapiro test was used to assess conformity with a normal distribution. On the basis of the normal distribution of data, the mean ± SD and mean ± SE were used, and p < 0.05 was considered significant.
A total of 5,355 interventional coronary procedures were performed at the two hospitals over an 18-month period, and 30 patients (0.6%) were identified as having SAT. Among these patients, there were 3 deaths, 5 patients could not be contacted, 2 could not participate because of clopidogrel allergy (skin rash), and the remaining 20 patients agreed to participate in the study. Among these 20 patients, 18 were receiving a 75-mg maintenance dose of clopidogrel and 2 patients were re-treated with clopidogrel. Of 20 SAT patients, 18 were on aspirin therapy (81 to 325 mg/day). The mean time to SAT (time from the day of procedure to the development of SAT) was 23 ± 16 days.
The patient demographics and procedural characteristics of both groups are shown in Tables 1 and 2,⇓⇓respectively. Among the interventions performed in patients without SAT, 2 patients were admitted with myocardial infarction and 11 patients had unstable angina. The remainder of the patients had stable angina, and the procedures were performed electively. Twelve of the procedures resulting in SAT were performed emergently (myocardial infarction in seven patients and unstable angina in five patients). The remainder had stable angina. The ages of both groups were the same. Patients with SAT had a non-significantly greater incidence of family history of coronary artery disease. Hematological data did not differ between groups (data not shown). The ejection fraction was lower, and the total lesion length was greater in the SAT group. Drug-eluting stents were more commonly used in patients without SAT.
Platelet aggregation in response to 5 and 20 μmol/l ADP was higher in the group with SAT as compared with the group without SAT (Figs. 1Aand 1B, p < 0.05 for both 5 and 20 μmol/l ADP-induced aggregation). Patients with SAT had greater active GP IIb/IIIa expression and a higher P2Y12reactivity ratio, whereas total GP IIb/IIIa expression was not significantly different between groups (Table 3).
The estimated prevalence of high post-treatment platelet reactivity was 65% and 60%, as measured by 5 and 20 μmol/l ADP-induced aggregation in SAT patients, respectively (Figs. 1A and 1B). The r value comparing 5 μmol/l ADP-induced aggregation to 20 μmol/l ADP-induced aggregation was 0.93. The P2Y12reactivity ratio by VASP assay correlated with 20 μmol/l ADP-induced aggregation (r = 0.57, p = 0.019). The correlation of stimulated active GP IIb/IIIa expression with 20 μmol/l ADP-induced aggregation was weaker (r = 0.29, p = 0.23). Fifty percent of patients with SAT met the definition of high post-treatment platelet reactivity on the basis of VASP values, whereas 40% of the patients met the definition with respect to activated GP IIb/IIIa expression. All patients were responsive to aspirin, except one in the SAT group (54% arachidonic acid-induced aggregation). Aggregation by arachidonic acid was 4 ± 2% in the non-SAT group versus 3 ± 2% (p = NS).
With established measurements of platelet reactivity, we demonstrated that post-treatment platelet reactivity is higher in patients who had suffered SAT as compared with those without SAT, despite the presence of clopidogrel therapy in both groups. We found higher mean platelet aggregation induced by two concentrations of ADP as well as higher expression of active GP IIb/IIIa after stimulation by ADP in SAT patients. The VASP assay is a direct measure of P2Y12reactivity and, therefore, directly assesses the intrinsic functional response of the receptor. Our results strongly suggest that the P2Y 12receptor is not adequately inhibited by clopidogrel in a large percentage of patients who had experienced SAT. This assay might be useful in dose titration studies to tailor clopidogrel therapy in patients before discharge. The demonstration of a higher P2Y12reactivity ratio in patients with SAT strongly suggests lower receptor occupancy by clopidogrel, thus implying insufficient active metabolite generation. All of the data strongly support high platelet reactivity as a risk factor for the development of SAT.
Subacute stent thrombosis is a serious complication with incidences ranging from 0.4% to 3% in high-risk patients (2–4). A recent study reported the incidence to be only 0.4% in 500 patients treated with a sirolimus-eluting stent (3). The 0.6% incidence of SAT in our study is in agreement with these data. Many factors are likely to affect the development of an acute thrombotic episode in a recently stented vessel (2). Most of the patients with SAT in our study had their index procedure performed emergently and had long lesions. These risk factors are also supported by other studies (2,4).
Our results are concordant with Muller et al. (7), who prospectively examined 105 patients undergoing elective percutaneous coronary intervention and found that two patients subsequently developed stent thrombosis, both of which were non-responders to clopidogrel. The latter investigators also reported high platelet aggregation in three other patients not enrolled in the study who had experienced SAT. These findings were supported in a prospective study by Matetzky et al. (8), who found that those patients undergoing stenting with ST-segment elevation myocardial infarction in the lowest quartile of clopidogrel responsiveness were at increased risk for cardiovascular events after discharge.
We have previously defined clopidogrel resistance as <10% absolute change in aggregation compared with baseline (5). The current study, however, was not prospective with respect to the SAT group and, therefore, a baseline measurement of platelet reactivity was not recorded. Given the overall low event rate of SAT, the number of patients required to assess the relation of clopidogrel resistance to SAT in a prospective investigation was prohibitively large. Therefore, the incidence of clopidogrel resistance in SAT patients cannot be determined in the present study; however, the higher platelet reactivity measured in SAT patients with multiple techniques strongly suggests that platelets were inadequately inhibited by clopidogrel in these patients.
Although there is no accepted definition of aspirin resistance, various measurements of platelet function in patients receiving aspirin have been correlated to a greater risk of cardiovascular events. The prevalence of aspirin resistance has been reported to be between 5% and 45% (17–20). Of interest in our study, platelet reactivity to 1.0 mmol/l arachidonic acid was extremely low. Our results are discordant with observations by other investigators (17,20). Gum et al. (17) estimated aspirin resistance by using a combination of responsiveness to both ADP and a slightly higher dose of arachidonic acid (1.6 mmol/l). In their study, mean aggregation in response to arachidonic acid was 11.4 ± 10.3%, as compared with approximately 4% in our study. The lower arachidonic acid concentration in our study may in part explain why we found only 1 of 118 met the criteria for resistance as compared with 5.5% in their study. The effect of aspirin to block cyclooxygenase activity is likely much more uniform, whereas the effect on platelet aggregation as measured by agonists other than arachidonic acid is influenced by many factors. Therefore, the incidence of aspirin resistance (lack of inhibition of cyclooxygenase) is probably rare, as reflected in the present study.
Although we believe that high post-stent platelet reactivity is a risk factor for SAT, at this time, significant prospective data are lacking to recommend routine screening on all patients undergoing stenting. Larger-scale investigations are needed to support our findings. Finally, high post-treatment platelet reactivity might be due to incomplete inhibition of the P2Y12receptor by the current dosing of clopidogrel. We have demonstrated that a 600-mg clopidogrel loading dose results in superior early platelet inhibition as compared with a 300-mg dose (21); however, there are no data available regarding the effect of a higher maintenance dose and platelet reactivity. Future therapies with new P2Y12inhibitors promise less variability in post-treatment reactivity (22).
Platelet studies were performed at different times in non-SAT and SAT patients. Non-SAT patients were enrolled and studied prospectively, whereas patients with SAT were identified retrospectively and subsequently studied. The analyses of platelet reactivity conducted at different intervals from the index procedure might affect the results. However, we have previously demonstrated that clopidogrel responsiveness is lowest and platelet reactivity is highest early after stenting (5). Therefore, this fact would only strengthen our hypothesis, because at a later date, we would expect the non-SAT patients to have even higher clopidogrel responsiveness.
In conclusion, the current study strongly suggests that high platelet reactivity and incomplete inhibition of P2Y12receptor are risk factors for SAT. Measures to uniformly determine platelet reactivity after coronary stenting and treatment strategies to improve platelet inhibition in patients with high post-treatment platelet reactivity should be further investigated.
This study was supported by a grant from Astra Zeneca LP, Wilmington, Delaware.
- Abbreviations and Acronyms
- adenosine diphosphate
- mean fluorescence intensity
- prostaglandin E1
- stent thrombosis
- vasodilator-stimulated phosphoprotein
- Received December 22, 2004.
- Revision received July 11, 2005.
- Accepted July 19, 2005.
- American College of Cardiology Foundation
- Gurbel P.A.,
- O’Connor C.M.,
- Cummings C.C.,
- et al.
- Cutlip D.E.,
- Baim D.S.,
- Ho K.K.,
- et al.
- Gurbel P.A.,
- Bliden K.P.,
- Hiatt B.L.,
- O’Connor C.M.
- Matetzky S.,
- Shenkman B.,
- Guetta V.,
- et al.
- Samara W.M.,
- Bliden K.P.,
- Tantry U.S.,
- Gurbel P.A.
- Schwarz U.R.,
- Geiger J.,
- Walter U.,
- Eigenthaler M.
- Geiger J.,
- Brich J.,
- Honig-Liedl P.,
- et al.
- Gum P.A.,
- Kottke-Marchant K.,
- Welsh P.A.,
- White J.,
- Topol E.J.
- Eikelboom J.W.,
- Hirsh J.,
- Weitz J.I.,
- Johnston M.,
- Yi Q.,
- Yusuf S.
- Zimmermann N.,
- Wenk A.,
- Kim U.,
- et al.
- Gurbel P.A.,
- Bliden K.P.,
- Hayes K.M.,
- Yoho J.A.,
- Herzog W.R.,
- Tantry U.S.
- Wallentin L.,
- Jernberg T.,
- Leese P.T.,
- et al.