Abstract
Aspirin resistance is often defined as failure to demonstrate an inhibitory effect of aspirin on standard platelet function assays. Although many studies have been published on this phenomenon, the performance of individual assays to detect actual aspirin effects is poorly defined. The aim of this study is to define the sensitivity and potential limitations of five different assays of platelet function on detecting aspirin effects. A prospective cohort of 45 healthy volunteers consented to taking 80 mg of non-coated aspirin for 8–10 days. Participants were selected if they were non-smokers, did not present any significant medical condition and were excluded if they had ingested any potential drug / supplement known to effect platelet function in the past 10 days or were found to have abnormal blood counts. Samples were collected immediately prior to the first aspirin dose and repeated after 8 to 10 days of aspirin therapy. The study population consisted of 16 males and 29 females, 19 to 59 years of age without acquired cardiovascular risk factors and a mean BMI of 24,1 (range: 18,8–32,5). Study drug intake was carefully monitored by patient questionnaire and unused pill count at the second study visit. Partial non compliance was identified in 4 participants (1–2 missed doses) although no missed doses occurred in the last 3 days of the study. Mean delay between the last aspirin dose and the post study sample was 221 minutes (range 60 to 720 min, 5 patients with delays over 6 hours). Test characteristics and performances in detecting aspirin intake in our population are resumed in the table below.
Conclusion: Although aspirin clearly has an effect in all the assays tested, the broad distribution of values with LTA-ADP, platelet count drop, TEG-MA(AA) and urinary dTxB2 does not allow accurate discrimination of aspirin effects in healthy individuals. Considering their adequate distinction between subjects on and off aspirin, we suggest that LTA-AA and the VerifyNow assay should be preferred in studies evaluating aspirin effect and potential resistance to this drug. Findings such as ours should serve to better define the limitations of laboratory studies of platelet function for the assessment of aspirin effects.
. | LTA-AA 1,6 mM (%) . | LTA-ADP 10 umol (%) . | VerifyNow (ARU) . | Platelet count drop (%) . | TEG-MAAA(%) . | dTxB2 (pg/ml) . |
---|---|---|---|---|---|---|
* 95% confidence intervals; **paired t-test, ***as suggested by Gum et al, ¤as suggested by manufacturer, 2best fit on ROC curve | ||||||
Prior to ASA | 61,2 (33,9;88,4)* | 74,2 (35,1;113,3) | 634,5 (564,8;704,3) | 74,9 (35,3;114,4) | 100,6 (27,2;174) | 240,8 (0;821) |
After ASA | 2,4 (0;7,9) | 40,4 (0;85,8) | 407,9 (355,7;460,1) | 41,5 (8,8;74,3) | 53,2 (−44,4;150,8) | 115 (0;407,9) |
p value** | <10e-29 | <10e-13 | <10e-32 | <10e-10 | <10e-7 | <10e-6 |
Proposed cut-off values | < 20%*** | < 70%*** | <550 ARU¤ | <55%2 | < 90%2 | < 602 |
Sensitivity | 100 % | 84,4 % | 100 % | 82,2 % | 82,9 % | 62,2% |
Specificity | 95,6 % | 77,8 % | 95,6 % | 86,7 % | 75,8 % | 82,2% |
. | LTA-AA 1,6 mM (%) . | LTA-ADP 10 umol (%) . | VerifyNow (ARU) . | Platelet count drop (%) . | TEG-MAAA(%) . | dTxB2 (pg/ml) . |
---|---|---|---|---|---|---|
* 95% confidence intervals; **paired t-test, ***as suggested by Gum et al, ¤as suggested by manufacturer, 2best fit on ROC curve | ||||||
Prior to ASA | 61,2 (33,9;88,4)* | 74,2 (35,1;113,3) | 634,5 (564,8;704,3) | 74,9 (35,3;114,4) | 100,6 (27,2;174) | 240,8 (0;821) |
After ASA | 2,4 (0;7,9) | 40,4 (0;85,8) | 407,9 (355,7;460,1) | 41,5 (8,8;74,3) | 53,2 (−44,4;150,8) | 115 (0;407,9) |
p value** | <10e-29 | <10e-13 | <10e-32 | <10e-10 | <10e-7 | <10e-6 |
Proposed cut-off values | < 20%*** | < 70%*** | <550 ARU¤ | <55%2 | < 90%2 | < 602 |
Sensitivity | 100 % | 84,4 % | 100 % | 82,2 % | 82,9 % | 62,2% |
Specificity | 95,6 % | 77,8 % | 95,6 % | 86,7 % | 75,8 % | 82,2% |
Author notes
Disclosure:Research Funding: Funding provided by “Fond de Recherche en Sante du Quebec”.