Individualised Assessment of Response to Clopidogrel in Patients Presenting with Acute Coronary Syndromes: A Role for Short Thrombelastography?



This article is corrected by:

  1. Errata: Erratum Volume 28, Issue 4, 254, Article first published online: 5 July 2010

Dr James Cotton, M.D., F.R.C.P., Heart and Lung Centre, New Cross Hospital, Wolverhampton WV10 OQP.
Tel: 0044 (0)1902 694200;
Fax: 0044 (0)1902 695646;


Introduction: There is considerable interindividual variation in response to the antiplatelet agent clopidogrel. Hyporesponse predicts negative outcomes in patients presenting with a variety of ischemic cardiac conditions and following intracoronary stent placement. Many tests of clopidogrel activity are time consuming and complex. Short thromboelastography (s-TEG) allows rapid measurement of platelet clopidogrel response. Aims: We initiated this study to investigate the utility of s-TEG in assessing the response to clopidogrel in patients presenting with acute coronary syndromes (ACS) and to compare these results with established clopidogrel monitoring techniques. Methods: Patients admitted with unstable angina (UA) or Non ST elevation myocardial infarction (NSTEMI) undergoing coronary angiography were recruited. After routine loading with clopidogrel, all patients were tested with s-TEG and Accumetrics Verify-Now rapid platelet function analyzer (VN-RPFA). We used the modified TEG technique of measuring area under the curve at 15 min (AUC15), which allows a rapid estimation of antiplatelet response. Vasodilator-stimulated phosphoprotein phosphorylation (VASP) was also tested in a subgroup of patients. Clinical follow-up was obtained at 1 year. s-TEG results were correlated with VN-RPFA and VASP findings. Results: A total of 49 patients (33 male, mean age 63) were recruited and tested with s-TEG and VN-RPFA and a total of 39 patients were also assessed with VASP. s-TEG readings correlated well with VN-RPFA (r2= 0.54, P < 0.0001) and VASP (r2= 0.26, P= 0.001). Conclusion: s-TEG provides timely results which compare to current tests of clopidogrel activity. This technique can also be used to measure a variety of other clotting parameters and as such could develop into a valuable near patient test for the interventional cardiologist.


The addition of a thienopyridine, and in particular clopidogrel, to aspirin has led to a marked improvement in clinical outcome for patients with a variety of cardiovascular conditions [1,2]. An increasing number of clinical trials have demonstrated an association between interindividual variability of response to clopidogrel and adverse outcomes for patients undergoing stent placement for stable angina pectoris [3], unstable angina (UA), [4,5] and ST segment elevation myocardial infarction [6] (STEMI). This field has been brought into sharper focus by the evidence that the placement of drug eluting stents (DES) may be associated with an elevated and prolonged risk of stent thrombosis. Specifically, a number of trials have demonstrated an increased stent thrombosis risk in patients with high residual platelet activity despite clopidogrel therapy [7–11].

This increasing body of evidence that individual response to clopidogrel can determine outcome raises a clinical dilemma: should we test all of our patients for clopidogrel hyporesponsiveness, or continue without testing? One of the reasons that interventional cardiologists have not routinely assessed response to clopidogrel is that there has not been an easily accessible test. Initial studies utilized the accepted gold standard tests for platelet reactivity, light transmittance aggregometry (LTA), [12,13] and vasodilator-stimulated phosphoprotein phosphorylation (VASP) assessment using flow cytometry [14–16]. These techniques are time consuming, require significant technical skill, and are expensive. As such they are largely the preserve of academic research institutions rather than clinical cardiology centers. Given the acute nature of the presentation of many patients with acute coronary syndromes (ACS), it is inevitable that interest has increased in near-patient assays. In response to this issue, a number of such systems have now been developed. Table 1 details the current commonly used tests of antiplatelet activity and describes some of the advantages and disadvantages of each system. In the current study we have compared modified thrombelastography with the Accumetrics Verify-Now rapid platelet function analyzer (RPFA) and flow cytometric estimation of VASP. The Verify-Now system has been shown to predict adverse clinical outcomes in patients undergoing stent placement for both stable angina [17] and UA [5].

Table 1.  A summary of the advantages and disadvantages of some of the currently available tests of platelet function
  1. VASP, vasodilator-stimulated phosphoprotein; TXB2, thromboxane B2; PFA, platelet function analyzer; TEG, thromboelastograph.

  2. aPlatelet Function Analyzer-100 (Siemens Healthcare Diagnostics, Inc., Deerfield, IL, USA).

  3. bVerify-Now (Accumetrics, San Diego, CA, USA).

  4. cImpact cone and plate(let) s(DiaMed, Cressier, Switzerland).

  5. dPlateletworks (Helena Laboratories, Beaumont, TX, USA).

  6. eTEG Platelet Mapping system (Haemoscope, Niles, IL, USA), fMultiplate analyser (Dynabyte GmbH, Munich, Germany).

Optical platelet aggregometryHistorical gold standard. Large amount of data supporting use.Time consuming; requires complex sample preparation and operator expertise; expensive. Suited to research institutions.
Bleeding timePhysiologically relevant test; inexpensive; in vivoNonspecificity; insensitive; high interoperator variability; time consuming; frequent scar formation
VASPDirectly dependent on clopidogrel's target, P2Y12; low sample volume; whole blood assay. Good clinical data.Requires sample preparation and operator expertise; expensive. Suited to research institutions.
Urinary 11-dehydro-TXB2Directly dependent on aspirin's target, COX-1Dependent on renal function; indirect measure of antiplatelet effect; uncertainty over biologic variability and reproducibility
PFA-100 assayaPoint of care; whole blood assay; ease of use; low sample volume; physiologically relevant high shear systemDependence on platelet-independent factors e.g., von Willebrand Factor and hematocrit; no instrument adjustment. Cannot measure thienopyridine effect. Limited clinical data.
Verify-Now P2Y12 devicebPoint of care; rapid; ease of use; whole blood assay; small sample volume. Impressive clinical data.No instrument adjustment
Serum TXB2Directly dependent on aspirin's target, COX-1Not platelet specific; indirect measure of antiplatelet effect.
Impact® Cone-and-plate(let) analyzercPoint of care; ease of use; rapid; whole blood assay; physiologically relevant high shear; high sensitivityNot widely available; Limited published platelet data
Platelet works™dPoint of care; rapid; ease of use; whole blood assay; no sample preparationNot well studied; does not measure overall primary hemostasis. Limited clinical data. Results highly dependent on time between sample collection and testing.
TEG platelet mapping systemePoint of care; rapid; whole blood assay; good technique for measuring clot strength. Can measure more than platelet activity alone.Not exclusive platelet dependent. Substantial manual pipetting required in testing process. Increasing clinical data.
Multiplate® analyzerfPoint of care; rapid; cost effective; ease of use; whole blood assay; small sample volumeLimited published platelet data as yet

Many cardiothoracic departments already use the thrombelastograph (TEG®) Hemostasis Analyser (Haemoscope Corporation, IL, USA) to monitor whole blood clotting parameters following cardiothoracic and other surgical procedures [18–20]. Standard thrombelastography has been further modified to allow assessment of the contribution of the P2Y12 receptor by the addition of adenosine diphosphate (ADP) to one of the wells, and the effect of thienopyridine therapy can then be estimated by comparing the ADP-stimulated TEG curve with the unmodified (kaolin activated) curve. This system is marketed by the manufacturer and termed “platelet mapping”. This group has further modified this technique, by calculating the area under the curve at 15 min (AUC 15) of the ADP trace. We have previously shown that this modified platelet TEG (short-TEG [s-TEG]) can rapidly identify changes in platelet activity relative to loading with aspirin and clopidogrel [21–25]. In this study, we validate the s-TEG AUC 15 parameter and percentage of clotting inhibition (%CIn) in patients admitted with ACS by compairingi with Verify-Now and VASP results.


Study Population

Patients admitted to the Heart and Lung Centre in Wolverhampton for coronary angiography with a history of an ACS for a 6-month period between January and June 2008 were considered for inclusion. All patients had either suggestive ECG changes (ST segment depression, T wave changes) and/or a positive Troponin I level. Patients with concurrent inflammatory disorders, a contraindication to dual antiplatelet therapy or a life expectancy below 18 months were excluded from the trial. The study protocol was approved by the local research ethics committee and conducted in line with the declaration of Helsinki. All patients gave written informed consent prior to inclusion.

All patients received aspirin and clopidgrel loading as per contemporary UK practice (at least 300 mg clopidogrel loading dose if >12 prior to angiography followed by 75 mg daily as maintenance, or 600 mg loading if <12 h prior to angiography with 75 mg daily maintenance).

The primary clinical endpoint was adverse clinical events at 12 months (death, MI, repeat revascularization, stroke, or unplanned cardiovascular hospitalization).

Assessment of Clopidogrel Response

Modified Platelet TEG

Blood was taken from the arterial angiography sheath, before the administration of unfractionated heparin. All patients who were on low molecular weight heparin treatment had this therapy withheld on the day of the procedure to limit any effect on the TEG trace. Five milliliter of blood was discarded before the study sample was drawn into a 6 mL Lithium Heparin Vacutainer®. Samples were analyzed using the TEG® platelet mapping kit (Haemoscope Corp). TEG is a bedside test that provides an overall assessment of hemostatic function providing a graphic representation of the speed of whole blood clot formation and clot strength. In order to detect the effects of clopidogrel acting via the ADP P2Y12 pathway we utilized a s-TEG method using three channels. This system measures the physical properties of a forming clot by means of an oscillating blood filled cup in which a pin is suspended. The torque of the rotating cup is transmitted to the pin (and from there transduced) thus allowing measurement of the platelet/fibrin clot parameters. We calculated the rapid 15 min area under the curve (AUC15) in response to ADP as previously described, using a specially developed software programme, National Instrument Labview 7.0 (Areafinder 2:1). This measurement is demonstrated in Figure 1, showing that AUC 15 is a function both of clot strength (amplitude, y-) and speed of clot formation (time, x-axis). We have also calculated the %CIn (as previously described [26]). This variable gives information about the contribution of both platelet activity and fibrin activity on both the strength and speed of clot formation. With maximal platelet activation due to the addition of thrombin in channel 1 of the TEG and the effect of ADP on the platelet response measured by the addition of ADP in the presence of fibrin stimulation to channel 3 of the TEG, %CIn is calculated by the formula:

Figure 1.

A typical sTEG (short thrombelastography) trace, demonstrating the measurement of the area under the curve (AUC) 15 as a function of time and amplitude. MA = maximum amplitude.

Verify-Now RPFA

After discarding 5 mL, arterial sheath blood was drawn into vacutainer® tubes containing sodium citrate 0.109 M and analyzed using the Verify-Now system (Accumetrics, San Diego, CA, USA) as per the manufacturers' instructions. This system is a turbimetry-based device that measures platelet induced activation in whole blood using an optical transmittance in a blood filled chamber containing fibrinogen coated beads. The P2Y12 assay contains ADP as a platelet agonist and prostaglandin E1 as a suppressor of intracellular free calcium, in order to minimize the nonspecific platelet activation effect of ADP binding on P2Y1 receptors. Results are expressed as P2Y12 reaction units (PRU).

VASP Measurement

The VASP phosphorylation assay was performed within 12 h of sampling using a commercially available method following the manufacturers instructions (Platelet VASP kit, Biocytex, Marseille, France). The analysis was conducted using a FACScalibur Flow cytometer (Beckton Dickinson, NJ, USA). The platelet reactivity index was calculated (PRI, %) from the corrected mean fluorescence intensity (cMFI) after incubation of the whole blood platelets with either prostaglandin E1 alone or prostaglandin E1 + ADP using the following equation:


Data Collection and Follow-Up

All data were collected prospectively and entered into a dedicated database. Clinical follow-up was obtained by contacting all patients at 12 months, and source documents of potential events were obtained.


Continuous variables are described as mean ± standard deviation. Groups were compared using Students t-test or ANOVA as appropriate. Correlations were performed using Pearson's product moment method. Statistical calculations were performed using MINITAB version 15 (MINITAB Inc, PA, USA).


Forty-nine patients admitted with UA/Non ST elevation myocardial infarction (NSTEMI) to the Heart and Lung Centre in Wolverhampton were enrolled between Jan 2008 and June 2008 (mean age 63, range 39–87), baseline characteristic are recorded in Table 2. Twenty-seven (55%) patients were treated with percutaneous coronary intervention (PCI) during the index admission, and 11 (22%) coronary artery bypass graft (CABG) surgery. The remaining 11 (22%) received medical therapy. Of the PCI patients 100% received at least one stent.

Table 2.  Baseline clinical characteristics of subjects in study. Values are n, mean ± SD, or (percentage)
 All (N = 49)
  1. MI, myocardial infarction; CABG, coronary artery bypass graft; TIA, transient ischemic attack; CAD, coronary artery disease; PCI, percutaneous coronary intervention; ECG, electrocardiogram; ACE, angiotensin converting enzyme.

Age (year, mean ± SD)63 ± 11
Sex n (%)
 Male33 (67)
 Female16 (33)
Risk factors n (%)
 Body mass index (kg/m2)28 ± 4
 Hypertension30 (61)
 Diabetes mellitus12 (24)
 Ex-smoker14 (29)
 Current-smoker15 (31)
 Hyperlipidemia29 (59)
 Previous MI14 (29)
 Previous CABG3 (6)
 Previous stroke/TIA0 (0)
 Familial history of CAD27 (55)
Troponin n (%)
 Troponin positive on admission47 (96)
ECG changes on admission n (%)
 ST/T wave changes29 (59)
Clopidogrel dosage (Mean ± SD)
 Days of clopidogrel prior to angiography4.4 ± 2.2
 Total culmative dose prior to angiography623 ± 195
PCI patients: GPIIb/GPIIIa n (%)16/27 (59%)
Patient therapy n (%)
 Medicine11 (22)
 CABG11 (22)
 Single vessel PCI22 (45)
 Two vessel PCI5 (10)
 Three vessel PCI0 (0)
Drug therapy on discharge n (%)
 Aspirin49 (100)
 Clopidogrel46 (94)
 Statin48 (98)
 Beta-blocker40 (82)
 ACE inhibitor35 (71)
 Proton pump inhibitor19 (39)

All patients were investigated with s-TEG and VN P2Y12, with successful readings obtained for analysis in 47 and 49 patients, respectively. VASP analysis was completed in a subset of 39 patients. The 1 year follow-up rate was 100%.

For all subjects the mean s-TEG AUC15 result was 742.1 ± 295.6 mm/min, mean VN PRU was 235.3 ± 105.1 and mean VASP PRI 49.3 ± 20.6%. These three variables were normally distributed.

S-TEG AUC15 correlated with VN PRU (Figure 2A) r2= 0.54 (correlation coefficient, r = 0.74), P < 0.0001. There was also a weaker, but significant correlation seen with VASP PRI r2= 0.26(correlation coefficient, r = 0.51), P= 0.001 (Figure 2B). Previous studies have estimated that a clinically relevant clopidogrel resistance cut off point VN PRU of 235 or 240 [5,7]. The platelet and fibrin-dependent %CIn also correlated with the VN PRU r2= 0.50 (correlation coefficient, r = 0.71), P < 0.0001 (Figure 2C) and VASP PRI, r2= 0.28 (correlation coefficient, r = 0.53), P= 0.0006.

Figure 2.

(A) Correlation between Verify-Now P2Y12 reactivity units (VN PRU) and thrombelastography area under the curve at 15 min (TEG AUC 15). (B) Vasodilator stimulated phosphoprotein phosphorylation platelet reactivity index (VASP PRI) and TEG AUC 15. (C) VN PRU and percentage of clotting inhibition .

In our study, individuals with a PRU ≤ 240 had a mean AUC15 of 493 ± 238, n = 19. Patients with PRU ≥ 240 had a mean AUC 15 of 911 ± 195, n = 29. The difference between groups was highly statistically significant (P < 0.0001, Figure 3). This analysis was equally powerful for a PRU level of 235.

Figure 3.

Difference in thrombelastography area under the curve at 15 min (AUC 15) results for patients with Verify-Now (VN) determined normal response (<240) and hyporesponse (>240).

During the follow-up period there were 5 (10%) adverse clinical events (1 MI, 1 repeat revascularization and 3 unplanned cardiovascular admissions). Three of the adverse clinical events occurred in the PCI treated patients, the remaining two in patients treated medically. Both s-TEG AUC 15, using a cutoff of 800 mm/min and Verify-Now PRU using a cutoff of 240 predicted adverse outcomes at 1 year (P < 0.02, Figure 4). But these results should be treated with great caution given the small sample size and low event rate in this trial (Figure 4).

Figure 4.

Event-free survival for patients with adequate clopidogrel response as determined by (A) short thrombelastography area under the curve at 15 min (sTEG AUC 15), and (B) Verify-Now P2Y12 reactivity units (VN PRU).


Pharmacodynamic studies have shown a marked interindividual response to clopidogrel. Current data demonstrate that a poor response to clopidogrel is associated with negative clinical outcomes in patients with stable and unstable coronary disease using a variety of testing methods including LTA [3,4,6,10,12,25–27], VASP [8,9,16], and the near patient tests Verify-Now [5,7,28] as well as full TEG platelet mapping [26]. A variety of factors will determine the requirement for, and the success of, near patient clopidogrel response testing in the future.

Several new compounds have now been, or are being, developed which may either replace clopidogrel or be a useful alternative therapy in patients with a poor response to clopidogrel. The competitive p2Y12 inhibitor ticagrelor is attracting a great deal of interest following the recent presentation of the PLATO trial at the European Society of Cardiology meeting in Barcelona. This agent appears to perform more effectively than clopidogrel, without excess of bleeding complications. It is a twice daily medication, however, and licensing is still being sought. Prasugrel, a third generation thienopyridine, now licensed for use in Europe, reduces the rate of ischemic events in patients presenting with ACS following stent placement [29]. This agent appears to limit problems with interindividual variation [30], however the reduction in clinical events comes at the expense of an increase in serious bleeding episodes. Bearing this in mind, an attractive strategy might be—to routinely measure individual responses to clopidogrel and target patients with clear-cut evidence of hyporesponsiveness or “resistance” and to treat the clopidogrel hyporesponders with prasugrel. To support this notion, a good response to clopidogrel as measured by Verify-Now, has been shown to have a negative predictive accuracy of 96% for recurrent events in ACS patients [5].

A second therapeutic option could be to increase the dose of clopidogrel in patients in whom a near patient test has shown poor response [31]. This hypothesis is currently being tested in the Gauging Responsiveness With a Verify-Now Assay-Impact on Thrombosis and Safety (GRAVITAS) trial (; identifier; NCT 00645918) an international randomized trial 2800 patients with stable angina or non-STEMI being treated with PCI using DES. Patients with high residual platelet activity 12–24 h following loading will be randomized to standard maintenance therapy or an increased 150 mg dose for a 6-month period. Similarly the Dual AntiPlatelet With Drug Eluting Stents (DANTE) trial (; identifier; NCT 0038794) will randomize approximately 450 patients with ACS treated with DES and high residual platelet activity to either standard clopidogrel (75 mg) maintenance or 150 mg daily. These studies will clarify the utility of near patient tests in these groups significantly.

The relatively high frequency of reduced clopidogrel response and the association between poor response and adverse clinical outcomes now almost mandates that this heterogeneity is taken into account. Assuming that thienopyridine responsiveness testing will become commonplace, the test to gain widest acceptance need to be rapid, reproducible, cost-effective, and available to the interventional cardiologist at the point of patient contact. The development of s-TEG by measuring AUC at 15 min allows the technique to be used in a time-effective manner, and gives results which appear to be clinically applicable. Many cardiothoracic departments already have access to a thrombelastography, as unmodified thrombelastography has been shown to limit the need for transfusion following cardiac and hepatic surgery. Whilst, in this study we have focused on platelet mapping, a variety of other parameters can be measured simultaneously using this system, and unmodified thrombelastography clot strength has been shown to predict ischemic complications following PCI [12].

Study Limitations

This prospective observational study was designed to assess whether s-TEG platelet mapping for clopidogrel activity yielded results that correlated with the increasingly established tests, Verify-Now and VASP. The numbers studied are small and so caution should be used in applying the clinical endpoint results. VASP results were not available in all patients owing to the logistical difficulties of using this test in day-to-day clinical practise and so the correlation between s-TEG and VASP is probably weaker than that might have been seen if larger numbers had been studied. These results are provocative however and further large scale trials of this technique are warranted.

s-TEG is a simple, time-responsive technique that allows platelet response to clopidogrel to be monitored. Many cardiac centers already possess the hardware and the expertise required to use this technique. This study shows that results obtained with this method correlate with Verify-Now and VASP measurements, and in the studied group also identified a cohort of patients with adverse clinical events at 1 year.

Conflict of Interest

The authors declare no conflict of interest.