Predictive value of the GRACE discharge score on the long-term out-of-hospital coronary thrombotic events after implantation of drug-eluting stents

Objective: To evaluate the predictive value of GRACE discharge score on the long-term out-of-hospital coronary thrombotic events (CTE) after percutaneous coronary intervention (PCI) with drug-eluting stents. Methods: Present study was a prospective, observational, single center study. 10 724 consecutive patients underwent PCI in Fuwai Hospital between January and December 2013 were included, stents were implanted with conventional method. After PCI, patients were prescribed aspirin 100 mg once daily indefinitely, and either clopidogrel 75 mg once daily or ticagrelor 90 mg twice daily for at least 1 year. A total of 9 782 patients were included in the final analysis after excluding patients who did not undergo successful stent implantation, who were not discharged on dual anti-platelet therapy (DAPT), who only underwent bare-metal stents, who experienced in-hospital major bleeding, stent thrombosis, myocardial infarction (MI) or death,and who lost follow up. Clinical data were collected from all patients. 9 543 patients with complete baseline data were further analyzed for risk stratification and predictive value of GRACE discharge score. CTE was defined as stent thrombosis or spontaneous myocardial infarction. All patients were followed through Fuwai Hospital Follow-up Center, and evaluated either by phone, letter, or clinic visits or at 1, 6, 12 and 24 months after PCI. Risk stratification was performed according to the GRACE discharge score, and the predictive value of the GRACE discharge score was assessed using the receiver operating characteristic (ROC) curve. Results: After 2 years follow-up, there were 95 CTE among the 9 782 patients. The patients were divided into 2 groups according to the presence or absence of CTE: CTE group (95 cases) and no CTE group (9 687 cases). GRACE discharge score was significantly higher in CTE group than no CTE group (82.98±27.58 vs. 75.51±22.46, t=-2.57, P=0.012). According to risk stratification of GRACE discharge score, the patients were divided into low-risk (≤88) group (n=6 902), moderate-risk (89-118) (n=2 988) and high-risk (>118) (n=343) groups. As compared to the low-risk group, CTE risk in moderate- and high-risk groups was 1.59 times (HR 1.59, 95%CI 1.01-2.52, P=0.046) and 3.89 times higher (HR 3.89, 95%CI 1.98-7.65, P<0.001), respectively. Further analysis showed that the GRACE score had predictive value in the total cohort for CTE (area under the receiver operating characteristic (AUROC) 0.576, 95%CI 0.512-0.640, P=0.012) and in the acute coronary syndromes(ACS) subgroup for CTE: (AUROC 0.594, 95%CI 0.509-0.680, P=0.019), but not in the non-ACS subgroup: (AUROC 0.561, 95%CI 0.466-0.657, P=0.187). Conclusion: GRACE discharge score can predict the long-term out-of-hospital CTE in patients undergoing PCI with drug-eluting stents and treated with DAPT, and patients can be stratified into the low-, moderate- and high-risk groups of CTE by the GRACE discharge score.