Aspirin Resistance: An Evaluation of Current Evidence and Measurement Methods

Christopher P. Martin, Pharm.D.; Robert L. Talbert, Pharm.D., FCCP


Pharmacotherapy. 2005;25(7):942-953. 

In This Article

Measurement of Aspirin Resistance

No clinically validated, uniformly accepted method exists to assess the effect of aspirin on platelet aggregation at present. In vitro platelet aggregation studies have traditionally used the Born method, which has been the method of choice to assess platelet function and has been extrapolated to assess aspirin sensitivity in several studies.[22,31,33,36] Another method that has been used extensively is measurement of the urinary TXA2 metabolite 11-dehydro-TXB2 using both radioimmunoassay (RIA) and enzyme immunoassay (EIA). A version of EIA is commercially available under the marketed name of AspirinWorks (Creative Clinical Concepts Inc., Denver, CO). Two automated, point-of-care platelet function assays have also been studied: the Platelet Function Analyzer 100 (PFA-100) (Dade Behring Inc., Deerfield, IL) and the VerifyNow Aspirin Assay (Accumetrics Inc., San Diego, CA) ( Table 3 ).[29,35,39] A recent article offers an informative review of platelet function tests for research and clinical use.[40]

Traditional platelet aggregation studies are performed with platelet-rich plasma prepared from a citrated whole blood sample.[41,42] Platelet aggregation is determined using an optical platelet aggregometer such as the PAP4 (BioData Corp., Horsham, PA). This method relies on the concept that platelet aggregation in a uniform solution of platelet-rich plasma decreases the turbidity of the solution as the platelets cross-link and clump together. The amount of platelet aggregation is directly related to the amount of light that is allowed to be transmitted through the solution. Results are expressed in units of percentage of light transmission on a scale of 0-100%.

Aggregation is induced by the addition of a single platelet agonist to the platelet-rich plasma at standard concentrations; typical agonists used are ADP, arachadonic acid, collagen, epinephrine, and thrombin. Aspirin almost completely inhibits platelet aggregation induced by arachadonic acid and collagen, but aggregation in response to ADP and epinephrine is only partially inhibited.[41,42,43] Platelet aggregation is sensitive to changes in temperature and hydrogen-ion concentration, and must be conducted within hours of sample collection. The technical difficulties and cumbersome nature of the assay require a specialized laboratory setting to perform optical platelet aggregation. Thus, this assay is unsuitable for routine clinical practice.


Rather than measure platelet aggregation directly, another approach to quantify the activity of aspirin has been to measure the levels of the products of COX-1 enzyme action. Reduced levels of these products would be expected as a result of aspirin administration. Thromboxane A2 is unsuitable for measurement since it is a highly unstable compound with a short half-life.[44] Thromboxane A2 is rapidly hydrated to form the more stable TXB2, which is subsequently converted by the liver into two major metabolites: 2,3-dinor TXB2 and 11-dehydro-TXB2.[44,45] Both metabolites, along with TXB2, are excreted unchanged in the urine.[46] Urinary TXB2 and 2,3-dinor TXB2 levels are more reflective of renal TXA2 biosynthesis than of platelet activity, and measurement of serum TXB2 is prone to artifact.[47,48,49] In contrast, 11-dehydro-TXB2 is a stable metabolite of TXB2 that can be measured noninvasively in the urine and can serve as an indirect measure of TXA2 activity in vivo.[50]

Aspirin has demonstrated dose-dependent reduction in urinary levels of 11-dehydro-TXB2.[27,32,51] The normal range of 11-dehydro-TXB2 levels and the expected percentage decrease in response to aspirin have differed greatly between investigations and method of measurement used (EIA vs RIA). AspirinWorks is a standardized EIA for urinary 11-dehydro-TXB2 that was developed using controls from a study discussed above.[35] Use of standardized controls circumvents the variability of the assay among testing laboratories. The assay requires a 10-ml random urine sample; results are expressed as pg•(mg/dl of creatinine)-1 units. Results are placed in categories of cardiovascular event risk based on the quartiles established in the same study for patients who experienced an event despite aspirin therapy.[35]

This assay is advantageous because it is noninvasive and is normalized with standard controls. However, it is based on a retrospective case-control study[35] in which the frequencies of significant risk factors for cardiovascular disease (obesity, coronary artery disease, peripheral vascular disease, hypertension, diabetes, and smoking) were higher in the case group than in controls. As such, the event risk reported in the trial and by the AspirinWorks test may be inflated and should be used cautiously as a basis for therapeutic decision making.

The PFA-100 is a whole blood, point-of-care platelet aggregometer that has been investigated for measurement of platelet dysfunction as a result of disease and in response to antiplatelet agents.[52] The system functions by aspirating a blood sample through a capillary tube at a high shear rate, and through a small slit aperture that is cut into a membrane coated with either collagen-epinephrine (CEPI) or collagen-ADP (CADP) platelet agonists.[52,53] The time for a platelet plug to occlude the slit aperture, reported in seconds as the closure time, is inversely related to platelet activity. A closure time of 193 seconds or less is considered normal; longer than 300 seconds is considered nonclosure.[52,53] Attenuated closure times have been reported when the blood sample is anticoagulated with a citrate concen-tration of less than 3.8%.[54,55]

A diurnal increase in closure time of 30% from 8:30 A.M.-5:00 P.M. has been reported with the PFA-100.[56] Platelet aggregation measured by the PFA-100 has correlated well with optical platelet aggregometry.[57] However, this is in contrast to the low concordance reported when only four patients were determined to be aspirin resistant by both optical platelet aggregation and the PFA-100, compared with 18 and 31 patients by each individual test, respectively.[4]

Measurement of platelet aggregation under the influence of aspirin is measured using a CEPI cartridge; aspirin has little to no effect on CADP cartridge closure time.[52] Aspirin has prolonged CEPI closure time in a dose-dependent fashion after administration of a single dose of intravenous L-aspirin 100 mg.[58] Prolongation of CEPI closure time has been variable in response to both single- and multiple-dose aspirin regimens.[39,58,59,60,61]

In a study of 31 patients with peripheral arterial disease who started receiving aspirin 100 mg/day, 40% of patients were nonresponsive when CEPI closure time was measured with the PFA-100.[60] In a subset of patients enrolled in the second Warfarin-Aspirin Reinfarction Study (WARIS), 35% of patients taking aspirin were aspirin resistant as determined by PFA-100 CEPI closure time measurement.[61] During 4 years of follow-up, no difference was noted in vascular event rates between aspirin responders and nonresponders. In another investigation, patients took aspirin 100 mg/day for secondary stroke prevention.[39] The CEPI closure time was significantly shorter in 18 patients who had a cerebrovascular event within 3 days of the measurement than in 53 who had been event free for 2 years (225 vs 279 sec, p=0.007).

The PFA-100 is a convenient point-of-care device for measurement of platelet aggregation; however, its role in measurement of aspirin resistance has not been satisfactorily explored. The breakpoint for determination of aspirin sensitivity by CEPI closure time is ill defined. In addition, aspirin resistance measured by the PFA-100 has been only weakly correlated with an increased risk of clinical events in one observational cohort study.[39]

Like the PFA-100, the VerifyNow Aspirin Assay (formerly Ultegra Rapid Platelet Function Analyzer) is a whole blood, point-of-care device that measures platelet aggregation using different cartridges for different applications. Unlike the PFA-100, the VerifyNow is designed only to detect platelet dysfunction as the result of exposure to antiplatelet agents, such as aspirin, clopidogrel, and glycoprotein IIb-IIIa inhibitors. When first developed, the VerifyNow used a proprietary platelet agonist consisting of metallic cations and propyl gallate, but it was recently changed to use an arachadonic acid agonist.[62,63] Platelet aggregation detection is based on the agglutination of platelets on fibrinogen-coated beads detected by an optical turbidimetry method.

Platelet inhibition measured by the VerifyNow in response to single-dose aspirin administration has correlated well with measurements by the PFA-100 (r2 = 0.73-0.86)[64] and with traditional optical platelet aggregation (r2 = 0.902).[63,65] Platelet responsiveness is expressed in aspirin response units, with a cutoff for aspirin resistance at 550 or more aspirin response units.[62,63] The reported rate of aspirin resistance using the VerifyNow in a cohort of patients with vascular disease or at high risk of developing vascular disease was 23%.[63]

The impact of aspirin resistance as measured by the VerifyNow on serum markers of myonecrosis (cardiac enzymes) was investigated in 151 patients who were taking aspirin 80-325 mg/day and underwent elective PCI.[29] The frequency of myonecrosis was determined by elevated serum creatine kinase-myocardial band isoenzyme (CK-MB) and troponin I levels after PCI. After baseline CK-MB, troponin I, and aspirin sensitivity were measured, all patients received a loading dose of clopidogrel 300 mg. Troponin I was measured again 12-24 hours after PCI, and CK-MB was measured 6-8 hours afterward. Both troponin I and CK-MB levels were measured serially until peak levels were reached. Of the 151 patients, 29 (19.2%) were determined to be aspirin resistant.

Significantly more patients experienced CK-MB level elevation after PCI in the aspirin-resistant group than in the aspirin-sensitive group (51.7% vs 24.6%, p=0.006). Similarly, more aspirin-resistant patients demonstrated troponin I level elevations (65.5% vs 38.5%, p=0.012). This study linked aspirin resistance measured by the VerifyNow to surrogate markers of myonecrosis but did not address whether the increase in serum CK-MB and troponin I levels translates into an increased frequency of reinfarction or other clinical events. In addition, clopidogrel resistance has been recently reported and cannot be excluded as a possible confounder in this study.[66,67]