The optimal threshold of high post-treatment platelet reactivity could be defined by a point-of-care VerifyNow P2Y12 assay

We read with great interest the study by Price et al., which verifies that high post-treatment platelet reactivity (HPPR) measured with a point-of-care VerifyNow assay (Accumetrics Inc., San Diego, CA, USA) is associated with post-discharge events after percutaneous coronary intervention (PCI) with drug-eluting stent (DES), including stent thrombosis. To the best of our knowledge, this is the first study to identify a threshold of HPPR of VerifyNow based on the clinical outcomes. Recently, a number of studies have demonstrated that clopidogrel non-responsiveness proven in the laboratory testing, i.e. HPPR, has been associated with an increased risk for cardiovascular events. Light transmittance aggregometry (LTA) is the gold standard test to determine the clopidogrel responsiveness. However, the abundant demands of LTA make it difficult to utilize in daily practice. VerifyNow was developed as a point-of-care test and showed a significant correlation with ADP-induced LTA (r 1⁄4 0.64 2 0.73). In previous studies using LTA, platelet aggregation of .50% induced by 5 mM ADP or of .70% induced by 10 mM ADP has been suggested an absolute threshold of HPPR for predicting the ischaemic outcomes. In the study of Price et al., the optimal cut-off for the combined endpoint was a post-treatment reactivity of 235 PRU (P2Y12 reactivity unit) (area under curve [AUC] 0.711, 95% confidence interval [CI] 0.529–0.893, P 1⁄4 0.03). Because this study did not show the association between the HPPRs by a point-of-care test and ADP-induced LTA, we estimated the relation using our data. Three hundred consecutive patients undergoing PCI with DES implantation at our hospital were enrolled between October 2007 and March 2008. We performed 5 mM ADP-induced LTA and VerifyNow using the same blood sampling via the arterial sheath. LTA was performed in all patients according to standard protocols. Both PRU (r 1⁄4 0.641, P , 0.001) and percentage platelet inhibition (r 1⁄4 0.679, P , 0.001) measured by VerifyNow had significant correlations with the results of 5 mM ADP-induced platelet aggregation. By the receiver-operating characteristics curve analysis, the optimal cut-off for predicting HPPR on LTA (5 mM ADP-induced platelet aggregation .50%) was PRU 239 (AUC 0.794, 95% CI 0.736–0.851, P , 0.001). The PRU value 239 showed a sensitivity of 83.6% and a specificity of 68.3%, and was similar to the threshold of high reactivity value (PRU 235), suggested by Price et al. The percentage platelet inhibition of 20 was the optimal cut-off for predicting HPPR on LTA (AUC 0.841, 95% CI 0.790–0.891, P , 0.001), which showed a sensitivity of 76.2% and a specificity of 83.6%. A VerifyNow assay has been used widely in the daily practice instead of LTA. However, its usefulness for predicting adverse cardiovascular events has been still undetermined. On the basis of our data analysis, we could ascertain that high platelet reactivity on VerifyNow (PRU 235 suggested by Price et al.) is significantly correlated with HPPR on ADP-induced LTA. It might suggest a substitutability of VerifyNow in terms of assessment of clopidogrel responsiveness and practical implication for risk stratification. Funding to pay the Open Access publication charges for this article was provided by J.-Y. Hwang, the senior of the Cardiology Division.