Poor Association Between Stent Thrombosis, Platelet Function Tests

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Platelet function testing in patients on clopidogrel therapy shows only a weak link between reactivity and stent thrombosis. This drawback could prove to be a serious obstacle to the adoption of such testing in clinical practice, researchers conclude in a study published online July 18, 2011, ahead of print in the American Heart Journal.

Christoph Varenhorst, MD, PhD, of Uppsala University (Uppsala, Sweden), and colleagues retrospectively identified a small group of patients from the Swedish Coronary Angiography and Angioplasty Registry (SCAAR) who had angiographically confirmed stent thrombosis or MI between May 2005 and December 2007 while on dual antiplatelet therapy and within 6 months of PCI. Cases were matched with controls from the registry without stent thrombosis or MI.

P2Y12 inhibition was then measured by VerifyNow P2Y12 (Accumetrics, San Diego, CA) and vasodilator-stimulated phosphoprotein (VASP) assay in all subjects.

‘Optimal’ Cutoff Results in Poor Accuracy

Compared with controls, patients with stent thrombosis and MI had higher rates of congestive heart failure and previous MI. Additionally, those with stent thrombosis had more stents implanted and had more complex lesions treated than controls.

Stent thrombosis occurred early (0-30 days) in 40 patients and late (> 30 days) in 8 patients. The median time from PCI to stent thrombosis was 5 days (interquartile range [IQR], 3.0-11.5 days), and the majority of patients (60.4%) presented with STEMI at the time of stent thrombosis. In those with MI, the median time from PCI to the event was 64 days (IQR, 17.0-118 days).

Assessments of on-treatment platelet reactivity varied by test. On VerifyNow, P2Y12 reactivity unit (PRU) measurements were higher in patients with angiographically confirmed stent thrombosis than in controls. However, platelet reactivity index (PRI) values on the VASP assay did not differ between patients with stent thrombosis and matched controls (table 1).

Table 1. Platelet Reactivity in Patients With and Without Stent Thrombosis

In patients with MI, both VerifyNow and VASP values were lower than in matched controls (table 2).

Table 2. Platelet Reactivity in Patients With and Without MI

For VerifyNow, the optimal cutoff level was identified as a PRU of at least 222, providing 70.2% sensitivity and 67.3% specificity. For the VASP assay, the optimal cutoff level was identified as 24% inhibition, resulting in a sensitivity of 64% and a specificity of 74%.

No Clear Support for Platelet Testing

According to the investigators, this study calculated the diagnostic accuracy of platelet function tests in the largest group of patients with angiographically confirmed stent thrombosis to date. But even so, the resulting low sensitivity and specificity do not justify the use of such tests in clinical decision-making.

While it is surprising that the VASP assay did not distinguish between patients with or without stent thrombosis while VerifyNow did, they say, this discrepancy may be related to VASP assay insensitivity to low levels of P2Y12 receptor blockade.

Regarding patients with MI, the study authors say the results “support the observation from previous studies that a greater incidence of major adverse cardiovascular events in patients with high on-clopidogrel platelet reactivity is driven mainly by early ischemic events related to coronary stenting such as [stent thrombosis] or periprocedural MI, but not by a later need for target vessel revascularization or death.”

While two ongoing studies—the ARCTIC and DANTE trials—are testing whether a strategy that involves monitoring clopidogrel response in order to guide therapy can improve outcomes after PCI, Dr. Varenhorst and colleagues say no evidence currently exists to support that theory. Their results add more evidence suggesting that platelet function testing is not a strong predictor of risk, they stress.

However, the authors acknowledge some limitations to their study, including:

• Small sample size
• Varying time intervals between events and platelet testing
• Some patients were already on clopidogrel, whereas others received loading dose 16 to 24 hours before platelet function testing
• Controls were matched for age, gender, and index PCI site but not their indication for PCI (stable or unstable coronary artery disease) or MI history

Deepak L. Bhatt, MD, MPH, of Brigham and Women's Hospital (Boston, MA), said in an e-mail communication with TCTMD that the study is interesting.

“[It] shows that point-of-care platelet function assays are moderately useful to predict risk of stent thrombosis at a population level but add relatively little value on individual decision making, due to the broad overlap in platelet reactivity between patients with and without stent thrombosis. The ability to predict spontaneous MI also seems quite limited," he noted.

 

---

Source:
Varenhorst C, Koul S, Erlinge D, et al. Relationship between clopidogrel-induced platelet P2Y12 inhibition and stent thrombosis or myocardial infarction after percutaneous coronary intervention: A case-control study. Am Heart J. 2011;Epub ahead of print.

 

 

Related Stories:

• One-Month Platelet Reactivity After PCI Key Risk Factor for Future Events
• Only Some Platelet Function Tests Predict Outcome in PCI Patients
• POPULAR: Some Platelet Function Tests Predict Clinical Outcomes, Some Do Not