| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2005;36:1001.)
© 2005 American Heart Association, Inc.
Original Contributions |
From the Oxford Haemophilia Centre & Thrombosis Unit (P.H., K.B., C.F.), Churchill Hospital, Oxford, UK; and the Stroke Prevention Research Unit (H.S., L.S., P.R.), Department of Clinical Neurology, University of Oxford, UK.
Correspondence to Professor Peter M. Rothwell, Stroke Prevention Research Unit, Department of Clinical Neurology, Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK. E-mail peter.rothwell{at}clneuro.ox.ac.uk
| Abstract |
|---|
|
|
|---|
Methods Platelet function was assessed by all 3 tests in 100 patients receiving low-dose ASA therapy after transient ischemic attack (TIA) or ischemic stroke.
Results The incidence of ASA nonresponsiveness was 17% by the RPFA and 22% by the PFA-100, compared with only 5% by LTA (ie, as defined with both arachidonic acid and ADP). Agreement between the RPFA and the PFA-100 and arachidonic acid induced LTA was poor (
=0.16, 95% CI, 0.08 to 0.39, P=0.11; and
=0.09 0.12 to 0.30, P=0.32, respectively). Agreement between the 2 POC tests was also poor (
=0.14, 0.08 to 0.36, P=0.15). Only 2% of patients were aspirin nonresponders by all 3 tests.
Conclusions The prevalence of apparent ASA nonresponsiveness was higher with both the POC tests than with LTA. However, agreement between the tests was poor and very few patients were ASA nonresponsive by all 3 tests. Aspirin nonresponsiveness is therefore highly test-specific and large prospective studies will be required to determine the prognostic value of each of the separate tests.
Key Words: aspirin platelets
| Introduction |
|---|
|
|
|---|
20% after ischemic stroke and acute coronary syndrome.1 However, the antiplatelet properties of ASA are not uniform between individuals and recurrent events in some patients may be caused by "ASA resistance" or ASA nonresponsiveness.210 The reported incidence of ASA nonresponsiveness varies widely (between 5% and 60%), partly because there is no accepted standard definition based on either clinical or laboratory criteria. Recently it has been proposed that the term "ASA resistance" should only be used as a description of the failure of ASA to inhibit thromboxane A2 production, irrespective of a nonspecific test of platelet function.9 There is now some evidence that ASA nonresponsive individuals as detected by platelet function tests may be at increased risk of ischemic vascular events.11,12 Although it could therefore be argued that the response to ASA should be monitored, the platelet function tests that has been shown possibly to be of prognostic value (light transmission aggregometry [LTA]) is time-consuming and difficult and cannot realistically be performed on large numbers of patients in routine practice. However, 2 simpler "point-of-care" (POC) tests of platelet function are now available, the PFA-100 and the Ultegra-RPFA-VerifyNow ASA test (RPFA),13,14 which could offer the possibility of the rapid and reliable identification of ASA nonresponsive patients, without the requirement of a specialized laboratory. Although some studies have suggested that these tests can detect ASA nonresponders and could be therefore clinically informative, there have been few validation studies and/or direct comparisons of these tests with LTA. We therefore compared LTA with both the PFA-100 and the RPFA in 100 patients with transient ischemic attack or stroke receiving daily low-dose ASA treatment.
| Materials and Methods |
|---|
|
|
|---|
Blood Sampling and Processing
3x2.6 mL of blood was anticoagulated with one-tenth volume 3.2% buffered trisodium citrate within Vacutainer tubes (Becton Dickinson). An additional 1.8 mL of blood was taken into the special citrated Vacutainer tube for RPFA analysis (Accumetrics). All assays were performed within 2 hours of sampling
Platelet Aggregation
Platelet-rich plasma was prepared by centrifugation at 250g for 10 minutes. The platelet-rich plasma was removed and then platelet-poor plasma prepared by further centrifugation at 2000g for 20 minutes. Aggregation was performed using a Biodata-PAP-4 aggregometer (Alpha Laboratories) within 300 µL minicuvettes stirred at 900 rpm at 37°C. The 100% line was set using platelet-poor plasma and a 0% baseline established with platelet-rich plasma (adjusted to 20x109/L) before addition of 1 of 2 different agonistsarachidonic acid and ADP (final concentrations of 1 mg/mL and 10 µmol/L, respectively). The percent aggregation after 10 minutes was recorded. An aspirin response was defined as <20% aggregation with 1 mg/mL arachidonic acid and <70% aggregation with 10 µmol/L ADP in a similar fashion as reported by Gum et al.3
PFA-100
The PFA-100 (Dade-Behring) simulates high shear platelet function within test cartridges.1720 Blood is aspirated under constant vacuum from the sample reservoir through a capillary and a microscopic aperture (147 µm) cut into a membrane. The membrane is coated with collagen/epinephrine (CEPI) or collagen/ADP (CADP). Platelet adhesion, activation, and aggregation result in formation of a platelet plug within the aperture. Platelet function is thus measured as a function of the time (closure time [CT]) it takes to occlude the aperture. When normal individuals ingest varying dosages (75 to 1000 mg) of ASA there is a dose-dependent prolongation of the CT on the CEPI cartridge but not on the CADP cartridge.21,22 The coefficient of variation of CTs has been reported as 10%.17
For analysis, 0.8 mL of the mixed whole blood was pipetted into the sample reservoirs of 1 CADP (batches 559555 and 569634) and 1 CEPI cartridge (batches 559634 and 579519) (prewarmed to room temperature) and then loaded into the PFA-100. In-house normal ranges are 55 to 112 seconds for CADP and 79 to 164 seconds for CEPI. Full blood counts were performed on the blood samples. An ASA response was therefore defined using the normal range cutoff of 164 seconds on the CEPI cartridge. ASA nonresponders were defined with a CEPI CT
164 seconds and ASA responders with a CEPI CT >164 seconds.
RPFA
RPFA (Accumetrics Inc) is a turbidimetric-based optical detection system that measures platelet-induced aggregation.21 This device was originally developed as a POC testing instrument to provide a simple and rapid functional means of monitoring anti-Gp IIb/IIIa therapy.2328 The test has been adapted to measure the effect of ASA within a modified cartridge (VerifyNow aspirin). The disposable cartridge contains fibrinogen-coated beads and a platelet activator (metallic cations and propyl gallate) to stimulate the COX-1 pathway and activate platelets.29 The instrument simply measures changes in light transmission automatically and thus the rate of aggregation. Precision testing using VerifyNow ASA level 1 (n=3x20) and level 2 wet quality controls (n=3x20) gave coefficients of variation of 3.2%, 3.5%, and 5.4%, and 4.3%, 2.6%, and 4.5%, respectively (Accumetrics VerifyNow ASA product information sheet).
Blood tubes were mixed and placed into RPFA cartridges (lot numbers WD0021 and W17507 preloaded into the RPFA instrument; Accumetrics). Results are expressed as aspirin reaction units (ARU) and there is an assigned cutoff of 550 ARU. Therefore, control samples or ASA-nonresponsive individuals would be expected to give values >550 ARU. ASA responders give values <550 ARU.
Statistical Analysis
All statistical analysis was performed using SPSS (version 10.0) and Analyze-it. Agreement between the different tests was determined by kappa statistics and 95% confidence intervals (CIs) were calculated. Kappa values of <0.20 are taken to indicate poor agreement, 0.21 to 0.40 indicate fair agreement, 0.41 to 0.60 indicate moderate agreement, 0.61 to 0.80 indicate good agreement, and >0.81 indicate very good agreement.
| Results |
|---|
|
|
|---|
|
Three normal subjects (aged 40, 44, and 51 years) were tested by all 3 platelet function tests either before (n=3) and 2 hours after ingestion of 300 mg of ASA (n=2) or after addition and incubation for 30 minutes with 100 µmol/L ASA at 37°C in vitro (n=1) (Table 2). One hundred patients were analyzed by PFA-100 and RPFA; 99 of 100 and 98 of 100 were additionally analyzed by LTA using arachidonic acid and ADP, respectively. Results are given in Table 3.
|
|
Arachidonic acidinduced LTA identified 12 of 99 (12%) patients who were ASA nonresponders. ADP-induced LTA identified 14 of 98 (14%) patients who were ASA nonresponders. Five of 98 (5%) patients were classified as fully ASA nonresponsive because they were detected by both tests, and 16 of 98 (16%) patients were classified as partially nonresponsive because they were only detected by one of the aggregation agonists.
The RPFA defined 17 of 100 (17%) patients as ASA nonresponders and the PFA-100 CEPI cartridge defined 22 of 100 (22%) patients with a CEPI CT <164 seconds. Analysis of the CADP CTs demonstrated that the median CT (74 seconds; 95% CI, 66 to 89) of the ASA nonresponders (as defined by a CT <164 seconds on the CEPI cartridge) was significantly lower than the median CT (86 seconds; 95% CI, 82 to 96) of the ASA-responsive group (P=0.009, MannWhitney test). However, comparison of von Willebrand factor (VWF) levels between nonresponders and responders as defined by the CEPI CT revealed a trend to higher levels in the nonresponders (Median VWF of 163 IU/dL, 95% CI, 129 to 187 compared with 149 IU/dL, 95% CI, 133 to 163, P=0.19). Two of the 5 fully ASA nonresponders by LTA were also simultaneously ASA-resistant by the PFA-100 and RPFA.
Figure 1 shows a scattergram comparing platelet aggregation by LTA to arachidonic acid with RPFA in 99 patients. The tests agree in 78 samples (79%), with both tests indicating ASA responsiveness in 74 and both indicating nonresponsiveness in 4. However, the tests gave discordant results in 21 patients (Table 3) and overall agreement was not significantly greater than chance (
=0.16, 95% CI, 0.08 to 0.39, P=0.11).
|
Figure 2 shows the scattergram comparing the PFA-100 CEPI CTs with arachidonic acidinduced LTA. Of the 99 patient samples tested by both, 74 were concordant, with 70 ASA responders and 4 nonresponders. Twenty-five samples gave discordant results (Table 3), with 17 patients categorized as ASA nonresponders by the CEPI CT alone and 8 by LTA alone. Overall agreement between the PFA-100 and arachidonic acid aggregation was poor (
=0.09, 95% CI, 0.12 to 0.30, P=0.32).
|
Figure 3 shows the comparison of the RPFA with the PFA-100 CEPI. Of the 100 samples tested, 73 were concordant, with 67 ASA responders and 6 nonresponders. Twenty-seven samples were discordant, with 16 categorized as ASA nonresponders by the PFA-100 CEPI alone and 11 by the RPFA alone. Overall agreement between the 2 POC tests was poor (
=0.14, 0.08 to 0.36, P=0.15).
|
Among those patients who were using another antiplatelet agent in addition to ASA, 4 of the 6 patients receiving clopidogrel and both patients receiving dipyrimadole were ASA-responsive by all 3 tests. The 3 patients who were receiving 150 mg of ASA were also responsive according to all 3 tests.
| Discussion |
|---|
|
|
|---|
We found that the prevalence of ASA nonresponsiveness was 12% and 14% by LTA with arachidonic acid and ADP, respectively. Only 5% of patients were nonresponsive by both tests. The prevalence of ASA nonresponders by either the RPFA (17%) or the PFA-100 CEPI cartridge (22%) were higher than with LTA. Interestingly, only 2% of patients were aspirin nonresponders by all 3 tests, and overall agreements between each of the tests were poor. Using 3 normal volunteers, we showed that each test was able to detect the expected influence of ASA on platelet function using the established cutoffs, which is consistent with previous studies20,21,31 but the use of specific cutoffs to define aspirin nonresponsiveness could in theory have reduced agreement in the clinical study. However, expression of the test results as continuous measurements (Figure 1 to 3![]()
) did not suggest that agreement had been underestimated.
Previous studies of the RPFA identified ASA nonresponsiveness in 23% of 422 patients with coronary artery disease,32 and 19.2% of 151 patients scheduled for nonurgent percutaneous coronary intervention were classified as ASA nonresponders.33 However, very few data are available comparing the test with LTA or PFA-100.31
The RPFA cartridge contains fibrinogen-coated beads and a platelet activator (metallic cations and propyl gallate) to stimulate the COX-1 pathway and activate platelets. The test should therefore theoretically produce similar results to those obtained by arachidonic acid LTA. Yet the incidence of ASA nonresponsiveness was higher by the RPFA test than with either arachidonic acid alone (17.0% versus 12.0% including 13 false-positives and 8 false-negatives) or both agonists (17.0% versus 5%). Previous data comparing propyl-gallate and other agonists by platelet aggregometry demonstrated that this agonist detects a lower number of nonresponders in volunteers receiving either 400 mg or 100 mg of ASA.29 Because the majority of patients in this study were receiving low-dose ASA, this may explain the discrepancy.
One previous study of the PFA-100 reported that 37% of stroke patients using low-dose ASA were nonresponsive.34 This rate is higher than the 22% prevalence using the CEPI CT in our study but the sample sizes in both studies were relatively small. The PFA-100 has been shown to be more sensitive than LTA at detecting ASA nonresponsiveness,3,11 and our data confirm this. In theory, because the PFA-100 is a high-shear system, it may be more physiologically relevant. It is interesting therefore that although CADP CTs were also significantly lower in the ASA nonresponsive group than the responders, their VWF levels were not significantly higher. However, this may be because of our relatively small sample size because higher VWF levels have previously been reported in patients who are ASA nonresponders.35 Both platelet hyperfunctional response(s) and VWF levels could both contribute to the occurrence of normal CTs in ASA nonresponders.3436
Our study does have potential shortcomings. It could be argued that we should have derived more control data, rather than relying on previous studies. It could also be argued that we should have determined the reproducibility of the different tests in our hands rather than relying on previous reports of coefficients of variation. However, longitudinal studies will be necessary to determine intra-individual reproducibility over longer periods of time.
In conclusion, we found that the prevalence of apparent ASA nonresponsiveness was higher with both the POC tests than with LTA, that agreement between the tests was poor, and that very few patients were ASA-nonresponsive by all 3 tests. Aspirin nonresponsiveness is therefore highly test-specific and large prospective studies will be required to determine the prognostic value, if any, of each of the separate tests.
| Acknowledgments |
|---|
Received January 8, 2005; accepted January 21, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. S. Karon, A. Wockenfus, R. Scott, S. J. Hartman, J. P. McConnell, P. J. Santrach, and A. S. Jaffe Aspirin Responsiveness in Healthy Volunteers Measured with Multiple Assay Platforms Clin. Chem., June 1, 2008; 54(6): 1060 - 1065. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Y. Gasparyan, T. Watson, and G. Y.H. Lip The Role of Aspirin in Cardiovascular Prevention: Implications of Aspirin Resistance J. Am. Coll. Cardiol., May 13, 2008; 51(19): 1829 - 1843. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Lordkipanidze, C. Pharand, E. Schampaert, J. Turgeon, D. A. Palisaitis, and J. G. Diodati A comparison of six major platelet function tests to determine the prevalence of aspirin resistance in patients with stable coronary artery disease Eur. Heart J., July 2, 2007; 28(14): 1702 - 1708. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Cattaneo Laboratory detection of 'aspirin resistance': what test should we use (if any)? Eur. Heart J., July 2, 2007; 28(14): 1673 - 1675. [Full Text] [PDF] |
||||
![]() |
P. A. Gurbel, K. P. Bliden, J. DiChiara, J. Newcomer, W. Weng, N. K. Neerchal, T. Gesheff, S. K. Chaganti, A. Etherington, and U. S. Tantry Evaluation of Dose-Related Effects of Aspirin on Platelet Function: Results From the Aspirin-Induced Platelet Effect (ASPECT) Study Circulation, June 26, 2007; 115(25): 3156 - 3164. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. D. Michelson, A. L. Frelinger, and M. I. Furman Resistance to antiplatelet drugs Eur. Heart J. Suppl., October 1, 2006; 8(suppl_G): G53 - G58. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Patrono, L. A. Garcia Rodriguez, R. Landolfi, and C. Baigent Low-dose aspirin for the prevention of atherothrombosis. N. Engl. J. Med., December 1, 2005; 353(22): 2373 - 2383. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2005 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |