Assessment of Dual AntiPlatelet Therapy with Drug-Eluting Stents - ADAPT-DES
ADAPT-DES is a large international registry of patients undergoing percutaneous coronary intervention (PCI), which examined the relationship between high on-treatment platelet reactivity and stent thrombosis.
High on-treatment platelet reactivity would be associated with an increased risk of stent thrombosis.
- Successful PCI with noninvestigational drug-eluting stent
- No complications
Number of enrollees: 8,575
Duration of follow-up: 30 days
- Stent thrombosis (definite or probable at 30 days)
All patients received dual antiplatelet therapy with aspirin and a thienopyridine. Use of glycoprotein IIb/IIIa inhibitors was permitted, but a washout period was required after their use.
A total of 8,575 patients were enrolled in the ADAPT-DES registry from 11 sites in the United States and Germany. Platelet reactivity was tested using the Accumetrics VerifyNow Aspirin assay for aspirin, VerifyNow P2Y12 for thienopyridines, and VerifyNow IIb/IIIa assay for glycoprotein IIb/IIIa inhibitor use.
Thirty-day results: Definite or probable stent thrombosis at 30 days occurred in 39 (0.5%) of the patients. There was no difference in aspirin reaction units (ARUs) (425.6 vs. 419.2, p = 0.46) and IIb/IIIa platelet activity units (PAUs) (188.2 vs. 192.7, p = 0.6) between the groups with stent thrombosis and those without; the proportion of patients with ARU ≥550 was also similar (7.7% vs. 5.6%, p = 0.57). In contrast, P2Y12 PRUs were significantly higher in the group with stent thrombosis (249.4 vs. 187.6, p = 0.0001). PRU >208 (74.4% vs. 42.6%, p = 0.0002), PRU ≥230 (64.1% vs. 34.9%, p = 0.0003), and PRU inhibition ≤11% (51.3% vs. 19.9%, p < 0.0001) were all significantly higher in patients with stent thrombosis.
On multivariate modeling, P2Y12 PRU >208 was associated with a threefold increase in the hazard for stent thrombosis, with the highest attributable risk for stent thrombosis (49.6%). Similarly, PRU ≥230 was associated with a 2.75-fold increase in the hazard for stent thrombosis, with a high attributable risk for stent thrombosis (40.8%). PRU >208 had a higher sensitivity (74.4%), but a lower specificity (57.4%) for stent thrombosis, while PRU ≥230 had a lower sensitivity (64.1%), but a higher specificity (65.1%) for stent thrombosis; both tests had a high positive predictive value (0.8%) and negative predictive value (99.8%).
There appeared to be significant effect modification by presenting indication for PCI, such that both the hazard and attributable risk for stent thrombosis were significantly elevated in patients with acute coronary syndrome (ACS), but not in those without ACS.
One-year results: The overall rate of stent thrombosis definite + probable was 0.84% (70 patients), of which 57.1% of the events occurred within 30 days. PRU >208 (65.2% vs. 42.5%, p = 0.0002), PRU ≥230 (53.6% vs. 34.9%, p = 0.001), and PRU inhibition ≤11% (44.9% vs. 19.9%, p < 0.0001) were all significantly higher in patients with stent thrombosis at 1 year. When stratified by PRU >208 versus not, stent thrombosis at 1 year was significantly higher (1.3% vs. 0.5%, hazard ratio 2.54, 95% confidence interval 1.55-4.16, p = 0.0001). Similarly, MI (3.9% vs. 2.7%, p = 0.002) and mortality (2.4% vs. 1.5%, p = 0.002) were significantly higher in patients with PRU >208. Paradoxically, major bleeding was lower (5.6% vs. 6.7%, p = 0.04). Excluding mortality, these results were similar for all other outcomes on multivariate adjustment. Aspirin ARUs continued to be similar between patients with and without stent thrombosis.
The results of this large registry suggest that absolute and relative levels of platelet inhibition to adenosine diphosphate (ADP) antagonists, as assessed by the VerifyNow P2Y12 test, are powerful independent predictors of stent thrombosis within 30 days and 1 year of PCI, with a significant proportion of events directly attributable to thienopyridine hypo-responsiveness. This is mainly true for patients presenting with ACS. A PRU threshold of >208 was associated with a higher attributable risk and a higher sensitivity, whereas a PRU threshold of ≥230 was associated with a higher specificity. In contrast, there was no correlation between aspirin and glycoprotein IIb/IIIa inhibitor hypo-responsiveness (as assessed by the studied assays) and stent thrombosis. Another interesting finding from this trial is that although patients with clopidogrel hypo-responsiveness have higher MI and stent thrombosis risk, they have a lower bleeding risk.
This is one of the largest studies linking high on-treatment platelet reactivity with adverse clinical outcomes. Unfortunately, trials conducted so far such as GRAVITAS and TRIGGER-PCI have not shown an improvement in clinical outcomes with an improvement in platelet reactivity. Given the low incidence of stent thrombosis (0.5-1%), cost-effectiveness analyses of routine platelet function testing in all-comers undergoing PCI are also necessary. The inverse correlation between bleeding and MI/stent thrombosis noted in this study confirms findings from other studies on this topic.
Presented by Dr. Gregg Stone at the Transcatheter Cardiovascular Therapeutics meeting (TCT 2012), Miami, FL, October 25, 2012 (1-year results).
Presented by Dr. Gregg Stone at the Transcatheter Cardiovascular Therapeutics meeting (TCT 2011), San Francisco, CA, November 9, 2011 (30-day results).
Keywords: Acute Coronary Syndrome, Platelet Aggregation Inhibitors, Platelet Function Tests, Drug-Eluting Stents, Ticlopidine, Pyridines, Germany, Percutaneous Coronary Intervention, Thienopyridines, Thrombosis, United States
< Back to Listings