Introduction

Current guidelines1,2 recommend 12-month (12-mo) dual antiplatelet therapy (DAPT) consisting of aspirin with a P2Y12 inhibitor after percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI). However, more recent reports3,4 have shown that 3-month (3-mo) duration of DAPT could reduce hemorrhagic risk without increasing the risk of ischemic events. Platelets play a key role in the development of acute coronary syndrome5, and platelet activation and plugging are of significant importance in the development of impaired pre-PCI flow because platelet-mediated release of vasoactive mediators increases platelet-rich thrombi formation5,6. Under the circumstance where blood supply is completely absent, available oxygen in the ischemic zone of the myocardium disappears within seconds. Hence, after a certain duration of complete ischemia, there is no treatment modality that can salvage ischemic myocardium7. However, cardiomyocytes that are exposed to low residual oxygen levels may be able to maintain sufficient adenosine triphosphate to survive for an extended period, even if the amount of adenosine triphosphate is insufficient to allow their contraction7. Hence, we might think that patients with pre-PCI thrombolysis in myocardial infarction (pre-PCI TIMI) flow grade 0/1 (pre-TIMI 0/1) or pre-TIMI 2/38 are in a meaningful different situation. Compared to patients with pre-TIMI 0/1, those with pre-TIMI 2/3 have a lower incidence of cardiogenic shock and improved early and late left ventricular ejection fraction (LVEF) through preservation of flow to the infarct zone, with consequent preservation of myocardial viability9,10. However, comparative clinical outcomes between short-term and standard 12-mo DAPT according to pre-TIMI in patients with AMI has not been reported. Therefore, the authors thought that the reevaluation of safety and efficacy of 3-mo and 12-mo DAPT according to the different pre-TIMI in patients with AMI could provide beneficial information to treat those patients. In this post hoc analysis of the TICO trial (Ticagrelor Monotherapy After 3 Months in the Patients Treated With New Generation Sirolimus-eluting Stent for Acute Coronary Syndrome)4, we compared 1-year clinical outcomes between these 2 different antiplatelet strategies in patients with AMI, after new-generation drug-eluting stent (DES) implantation.

Results

Baseline characteristics

Detailed information on antiplatelet therapy during the study period, causes of non-adherence to the allocated treatment, and medications during the study period are shown in Supplementary materials 1, 2, and 3. Table 1 shows the baseline characteristics of the study population. In both patients with pre-TIMI 0/1 and 2/3, the mean age, the number of males, and the mean value of LVEF were similar between the 3-mo and 12-mo DAPT groups. In patients with pre-PCI TIMI 0/1, the number of patients with a history of prior MI, the mean value of estimated glomerular filtration rate (eGFR), and the prescription rate of angiotensin receptor blockers (ARB) and calcium channel blockers (CCB) were significantly higher in the 3-mo DAPT group than in the 12-mo DAPT group. In contrast, the mean value of serum creatinine, and the prescription rates of beta-blockers and angiotensin converting enzyme inhibitors (ACEI) were higher in the 12-mo DAPT group. In patients with pre-TIMI 2/3, eGFR was higher in the 3-mo DAPT group, and the prescription rate of ticagrelor and beta-blockers was higher in the 12-mo DAPT group. Supplementary material 4 shows the baseline characteristics according to 3-mo or 12-mo DAPT strategies. Supplementary materials 5 and 6 show the baseline characteristics of the as-treated population.

Table 1 Baseline clinical, laboratory, angiographic and procedural characteristics.

Clinical outcomes

Clinical outcomes are summarized in Tables 2, 3, and 4, Supplementary materials 710, and Fig. 1a–j.

Table 2 Clinical outcomes by Kaplan–Meier analysis and Cox-proportional hazard ratio analysis at 1 year.
Table 3 Clinical outcomes between pre-PCI TIMI 0/1 and 2/3 groups according to 3-month or 12-month DAPT strategies.
Table 4 Interaction between pre-PCI TIMI during an index PCI (pre-PCI TIMI 0/1 vs. 2/3) and the duration of DAPT (3-mo DAPT vs. 12-mo DAPT) for clinical outcomes.
Figure 1
figure 1figure 1

Time-to-event curves for NACE (a and b), TIMI major bleeding (c and d), TIMI minor bleeding (e and f), TIMI major or minor bleeding (g and h), and MACCE (i and j) in pre-PCI TIMI flow grade 0/1 (a, c, e, g, and i) and 2/3 groups (b, d, f, h, and j).

Figure 2
figure 2

Subgroup analysis for NACE in pre-TIMI flow grade 0/1 group. NACE net adverse clinical event, Pre-PCI pre-percutaneous coronary intervention, TIMI Thrombolysis In Myocardial Infarction, 3-mo DAPT, ticagrelor monotherapy after 3-month dual antiplatelet therapy, 12-mo DAPT ticagrelor-based 12-month dual antiplatelet therapy, HR hazard ratio, CI confidence interval, BMI body mass index, eGFR estimated glomerular filtration rate, NSTEMI non-ST segment elevation myocardial infarction, STEMI ST segment elevation myocardial infarction.

Net adverse clinical events (NACE)

Table 2 shows clinical outcomes by Kaplan–Meier analysis and Cox-proportional hazard ratio analysis. In patients in both pre-TIMI 0/1 and 2/3 groups, the occurrence of NACE was not significantly different between the 3-mo and 12-mo DAPT groups (adjusted hazard ratio [aHR]: 0.689; 95% confidence interval [CI]: 0.389–1.220; p = 0.201; and aHR: 0.741; 95% CI: 0.432–1.273; p = 0.278, respectively) (Fig. 1a,b). In patients in both as-treated pre-TIMI 0/1 and 2/3 groups, the occurrence of NACE was not significantly different between 3-mo and 12-mo DAPT groups (Supplementary materials 7 and 8). On 3-mo landmark analyses between the 3-mo and 12-mo groups (Supplementary material 9), although the occurrence of NACE was not significantly different between 3-mo and 12-mo DAPT (aHR: 0.358; 95% CI: 0.128–1.003; p = 0.051) in pre-TIMI 0/1 group, it was significantly higher in the 12-mo DAPT group than in 3-mo DAPT group in the pre-TIMI 2/3 group (aHR: 0.422; 95% CI: 0.184–0.965; p = 0.041).

Thrombolysis in myocardial infarction (TIMI) bleedings

In patients with pre-TIMI 0/1, the occurrence of TIMI major, minor, and major or minor bleedings (Table 2, Fig. 1c,e,g) were not significantly different between the 3-mo and 12-mo DAPT groups. These results were repeated in patients in the as-treated group and on the 3-mo landmark analysis. In patients with pre-TIMI 2/3, although the occurrence of TIMI major bleeding was similar between the 3-mo and 12-mo DAPT groups (Table 2 and Fig. 1d), the occurrence of TIMI minor bleeding (aHR: 0.294; 95% CI: 0.108–0.799; p = 0.016) and TIMI major or minor bleeding (aHR: 0.483; 95% CI: 0.271–0.862; p = 0.014) were significantly higher in the 12-mo DAPT group than in the 3-mo DAPT group (Table 2, Fig. 1f,h). These results were repeated in patients in the as-treated group. However, on 3-mo landmark analyses between the 3-mo and 12-mo groups (Supplementary material 9), the occurrence of TIMI major, minor, and major or minor bleeding in patients in the pre-TIMI 2/3 group was significantly higher in the 12-mo DAPT group than in the 3-mo DAPT group (aHR: 0.100; 95% CI: 0.012–0.796; p = 0.030, aHR: 0.103; 95% CI: 0.013–0.817; p = 0.031, and aHR: 0.109; 95% CI: 0.025–0.467; p = 0.003, respectively).

Major adverse cardiac and cerebrovascular events (MACCE)

The occurrence of MACCE was not significantly different between the 3-mo and 12-mo DAPT groups in patients in both pre-TIMI 0/1 (aHR: 0.773; 95% CI: 0.391–1.527; p = 0.458, Table 2, Fig. 1i) and 2/3 groups (aHR: 0.766; 95% CI: 0.362–1.623; p = 0.487, Table 2, Fig. 1j). These results were repeated in patients in the as-treated group and on the 3-mo landmark analysis.

Subgroup analyses

Subgroup analyses for NACE are shown in Figs. 2 and 3. In patients with pre-TIMI 0/1 (Fig. 2) and old age (≥ 65 years, HR: 0.37; 0.16–0.89; p = 0.026), female (HR: 0.22; 0.06–0.81; p = 0.022), those with single-vessel disease (HR: 0.30; 0.10–0.92; p = 0.035); and in patients in the pre-TIMI 2/3 group (Fig. 3) with single-vessel disease (HR: 0.36; 0.14–0.92; p = 0.033), 3-mo DAPT showed better outcomes over 12-mo DAPT in this study.

Figure 3
figure 3

Subgroup analysis for NACE in pre-TIMI flow grade 2/3 group. NACE net adverse clinical event, Pre-PCI pre-percutaneous coronary intervention, TIMI Thrombolysis In Myocardial Infarction, 3-mo DAPT, ticagrelor monotherapy after 3-month dual antiplatelet therapy, 12-mo DAPT ticagrelor-based 12-month dual antiplatelet therapy, HR hazard ratio, CI confidence interval, BMI body mass index, eGFR estimated glomerular filtration rate, NSTEMI non-ST segment elevation myocardial infarction, STEMI ST segment elevation myocardial infarction.

Figure 4
figure 4

Flowchart. TICO Ticagrelor Monotherapy After 3 Months in the Patients Treated With New Generation Sirolimus-eluting Stent for Acute Coronary Syndrome, Pre-PCI pre-percutaneous coronary intervention, TIMI Thrombolysis In Myocardial Infarction, DAPT dual antiplatelet therapy.

Independent predictors for NACE

In Supplementary material 11, after multivariate analysis of patients with pre-TIMI 0/1, age, prior MI, eGFR, and diameter of deployed stents were independent predictors for NACE. In patients with pre-TIMI 2/3, diabetes mellitus was an independent predictor of NACE in this study.

Clinical outcomes between pre-TIMI 0/1 and 2/3 groups according to 3-month or 12-month DAPT strategies

In Table 3, in patients with 3-mo DAPT, the occurrence of NACE, TIMI major, minor, and major or minor bleeding was similar between the pre-TIMI 0/1 group and the pre-TIMI 2/3 group after adjustment. However, in patients with 12-mo DAPT, the occurrence of TIMI major or minor bleeding was significantly higher in the pre-TIMI 2/3 group than in the pre-TIMI 0/1 group (aHR: 0.514; 95% CI: 0.299–0.884; p = 0.016).

ST-segment elevation versus non-ST-segment elevation myocardial infarction (STEMI vs. NSTEMI)

The comparison of clinical outcomes between STEMI and NSTEMI is summarized in Supplementary material 12. After adjustment, in patients with both pre-TIMI 0/1 and 2/3, the occurrence of NACE, TIMI bleedings (major, minor, and major or minor), and MACCE were not significantly different between STEMI and NSTEMI. Supplementary material 13 shows univariate analysis for NACE according to the pre-TIMI in comparing STEMI and NSTEMI.

Interaction between pre-TIMI during an index PCI and the duration of DAPT

Table 4 shows the interaction between pre-TIMI during an index PCI (pre-TIMI 0/1 vs. 2/3) and the duration of DAPT (3-mo or 12-mo DAPT). There were no significant interactions between the different pre-PCI TIMI during an index PCI and the duration of DAPT after adjustment.

Discussion

The TIMI flow grade is a traditional method for assessing coronary blood flow11. Previous studies have shown that various pro-thrombotic markers including platelet count, reactivity, and mean platelet volume were associated with patency of the infarct-related artery in patients with STEMI, before primary PCI12,13. Moreover, in the CADILLAC (Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications) and HORIZONS-AMI (Harmonizing Outcomes with RevasculariZatiON and Stents in Acute Myocardial Infarction) Trials, pre-TIMI 3 was an important independent predictor of 1-year survival14. The main findings of this study were as follows: (1) in patients with pre-TIMI 0/1, the occurrence of NACE, TIMI bleedings (major, minor, and major or minor), and MACCE (all-cause death, cardiac death, MI, ST, and stroke) were not significantly different between the 3-mo and 12-mo DAPT groups. (2) In patients with pre-TIMI 2/3, the occurrence of TIMI minor bleeding and major or minor bleeding were significantly higher in the 12-mo DAPT group than in the 3-mo DAPT group. Moreover, on 3-mo landmark analyses between 3-mo and 12-mo, the occurrence of TIMI major bleeding was significantly higher in the 12-mo DAPT group than in the 3-mo DAPT group. However, the occurrence of MACCE was similar between the 3-mo and 12-mo DAPT groups. (3) The occurrence of NACE, TIMI bleeding, and MACCE were not significantly different between STEMI and NSTEMI.

The present data indicate that pre-TIMI 0/1 is present in 54.8% (STEMI, 68.5%, vs. NSTEMI, 31.5%) and pre-TIMI 2/3 is present in 45.2% (STEMI, 31.1%, vs. NSTEMI, 68.9%). The ratio STEMI/NSTEMI is at the complete opposite between the pre-TIMI 0/1 and 2/3 groups. However, these results are comparable with Bailleul et al. study15. Although the proportions of STEMI and NSTEMI in patients with pre-TIMI 0/1 or 2/3 were different, these proportions were not significantly different between the 3-mo DAPT or 12-mo DAPT groups (Table 1). Especially, in Table 3, STEMI was included in the multivariate analysis as a significant variable with other variables. The occurrence of TIMI major or minor bleeding was significantly in the pre-TIMI 2/3 group than in the pre-TIMI 0/1 group, similar to the results in Table 2. In addition, as shown in Supplementary material 10, in patients with both pre-TIMI 0/1 and 2/3, the occurrence of NACE, TIMI bleeding (major, minor, and major or minor), and MACCE were similar between STEMI and NSTEMI.

A ruptured, eroded, or protruding calcified atherosclerotic plaque could trigger local thrombosis, which is a critical step in the pathogenesis of AMI16. To date, it remains unclear why some plaques lead to STEMI with poor pre-TIMI but others do not17. Compared to pre-TIMI 0/1, which has prolonged ischemia and late reperfusion, can impair endothelial function, and cause myocardial tissue edema, pre-TIMI 2/3 would have shorter ischemic time and less myocardial damage17. In a state of endothelial dysfunction, disruption of the balance between anti-thrombosis and pro-thrombosis can lead to increased platelet aggregation18. More recently, Bauer et al.19 reported that, after adjustment, definite stent thrombosis (ST) occurred only in patients with pre-TIMI 0/1 in their ATLANTIC (Administration of Ticagrelor in the Cath Lab or in the Ambulance for New ST Elevation Myocardial Infarction to Open the Coronary Artery) sub-study. Moreover, they showed that prehospital administration of ticagrelor was less effective in patients with pre-TIMI 0/1 than those with pre-TIMI 2/3 (0.3% vs. 1.3%, p < 0.05). Hence, it could be assumed that on treatment with 3-mo or 12-mo DAPT, the major clinical outcomes could be influenced by pre-TIMI. However, the effect of pre-TIMI on bleeding and cardiovascular events after ticagrelor-based 3-mo or 12-mo DAPT in patients with AMI has not been reported. Thus, this paper may be considered as the first report focused on this perspective.

In our study, compared to the patients with pre-TIMI 0/1, the occurrences of TIMI minor bleeding and major or minor bleeding were significantly higher in the 12-mo DAPT group than in the 3-mo DAPT group, in patients with pre-TIMI 2/3. In addition, on 3-mo landmark analyses between 3-mo and 12-mo, the occurrence of NACE and TIMI major, minor, and major and minor bleeding were also higher in the 12-mo DAPT group than in the 3-mo DAPT group. Because of the absence of previous reports, it could be difficult to provide comparative results between our and previous reports. However, based on our results, it could be considered that the beneficial effects of 3-mo DAPT over 12-mo DAPT in reducing bleeding events are mainly determined by pre-TIMI 2/3 rather than by pre-TIMI 0/1. However, in patients with 12-mo DAPT, the mean age, the number of hypertensive and diabetic patients, and patients with prior history of PCI were significantly higher in the pre-TIMI 2/3 group than in the pre-TIMI 0/1 group. In this study, to adjust the diverse variables, multivariate analysis was performed. But, it could be speculated that these baseline characteristics may play an important role in explaining this higher TIMI major or minor bleeding. Despite the possible benefit of DAPT in reducing ischemic events of infarction, it may be considered that no compound can enter an ischemic no-flow area of myocardium, especially if the culprit coronary artery in totally occlude7. Hence, in patients with pre-TIMI 0/1, the occurrence of NACE, TIMI bleeding, and MACCE would not be significantly different between the 3-mo or 12-mo DAPT groups. In contrast, patients with pre-TIMI 2/3 treated with 12-mo DAPT showed a higher incidence of bleeding tendency than those with 3-mo DAPT without showing increased incidences of ischemic events. According to the subgroup analysis (Figs. 2 and 3), in both pre-TIMI 0/1 and 2/3, and in patients with single-vessel disease, 3-mo DAPT may be preferred over 12-mo DAPT to reduce NACE in this study.

In the FAST-MI (French Registry of Acute ST-Elevation or Non-ST-Elevation Myocardial infarction) study, after 2010, there was no further mortality gain was founded in patients with STEMI with reperfusion therapy or in patients with NSTEMI, regardless of performing PCI20. Moreover, there are some debates regarding the long-term prognosis between STEMI and NSTEMI21,22,23, the occurrence of NACE, TIMI bleeding, and MACCE were not significantly different between these two groups both in pre-TIMI 0/1 and 2/3 groups (Supplementary material 12) in our study and our results were consistent with those of Montalescot’s findings23.

Based on our results, considering pre-TIMI may be helpful to understand more accurately the comparative outcomes between short and standard 12-mo DAPT in patients with AMI receiving new-generation DES.

This study has some limitations. First, although the TICO trial was an open-label trial, this trial was not a placebo-controlled study. Therefore, drug adherence was not monitored. Second, the lower-than-expected rate of adverse events in the main TICO trial could be a limitation in this study. Therefore, caution regarding the interpretation of our study results is needed. Third, because the ultrathin bioresorbable polymer sirolimuls-eluting stent (Orsiro; BIOTRONIK, Buelach, Switzerland) was deployed in the whole study population in this study, diverse results that would reflect many other kinds of new-generation DES that are currently in use were not included in this study. Fourth, because platelet function tests (e.g., P2Y12 reaction unit) and aspirin reaction unit, mean platelet volume, and myocardial blush grade were not monitored before, during, and after PCI in the TICO trial, we could not provide this information. Fifth, because the use of intravascular ultrasound, optical coherence tomography, and fractional flow reserve (FFR) were not mandatory tools for treatment in the main TICO trial, we could not provide results according to the use of these imaging and functional testing tools for the lesions. Finally, even if pre-TIMI is easy and inexpensive, it could be a suboptimal, incomplete measure of myocardial perfusion. Moreover, although the TIMI flow grade is often used to evaluate blood flow during acute coronary occlusion and/or reperfusion, as a more instructional reference and more relevant indicators, such as FFR, should be considered to give a more accurate evaluation.

In conclusion, our results suggest that the higher bleeding tendency in 12-mo DAPT compared with 3-mo DAPT was more obvious in patients with pre-TIMI 2/3 than in those with pre-TIMI 0/1. However, more studies are warranted to confirm these results.

Methods

Study design

A total of 3056 participants from the TICO randomized clinical trial (ClinicalTrials.gov Identifier: NCT02494895; First registration: 10/07/2015)4,24 were evaluated in this study. The TICO trial was an investigator-initiated, multicenter, randomized, unblinded trial conducted at 38 centers in South Korea. Briefly, after PCI, patients were randomly assigned in a 1:1 ratio to receive ticagrelor monotherapy after 3-mo DAPT or ticagrelor-based 12-mo DAPT. A web-response permuted-block randomization was used, and the allocation sequence was computer generated by an external programmer. If patients were not taking aspirin or ticagrelor at the time of PCI, loading doses of aspirin (300 mg) and ticagrelor (180 mg) were administered. A single aspirin dose (100 mg per day) and 2 ticagrelor (180 mg per day) were maintained. After 3-mo of DAPT, aspirin use was continued in patients who were randomized to receive 12-mo ticagrelor-based DAPT group. The concomitant use of other antiplatelet agents or anticoagulants was not allowed. Other medical treatments were left to physician discretion4. The main aim of the TICO trial was to estimate the validity of ticagrelor monotherapy after 3-mo of DAPT compared to 12-mo of aspirin and ticagrelor DAPT, with respect to bleeding outcomes and MACCE, in patients with acute coronary syndrome (ACS) who received the Orsiro® (BIOTRONIK, Buelach, Switzerland) stent.

Study population

Key exclusion criteria included increased risk of bleeding due to prior hemorrhagic stroke, traumatic brain injury or brain surgery within the past 6 months, internal bleeding within the past 6 weeks, need of oral anticoagulation therapy, and anemia (hemoglobin ≤ 8 g/dL)4. The full inclusion and exclusion criteria are listed in Supplementary material 14. Additionally, patients with unstable angina (n = 926, 30.3%) and those with post-PCI TIMI flow grade < 3 (n = 47, 1.5%) were excluded. During a 12-mo follow-up period, 17 patients in the pre-TIMI 0/1 group (3-mo DAPT, n = 9; 12-mo DAPT, n = 8) and 12 patients in the pre-TIMI 2/3 group (3-mo DAPT, n = 7; 12-mo DAPT, n = 5) were lost to follow-up. Participants who withdrew the consent (pre-TIMI 0/1 group, n = 14 [3-mo DAPT, n = 6; 12-mo DAPT, n = 8]; pre-TIMI 2/3 group, n = 9 [3-mo DAPT, n = 6; 12-mo DAPT, n = 3]) or those who died (pre-TIMI 0/1 group, n = 15 [3-mo DAPT, n = 7; 12-mo DAPT, n = 8]; pre-TIMI 2/3 group, n = 16 [3-mo DAPT, n = 6; 12-mo DAPT, n = 10]) were also excluded. Hence, a total of 2083 AMI patients were finally included. The patients were classified into pre-PCI TIMI flow grade 0/1 (pre-TIMI 0/1, n = 1143, 54.9%) and pre-TIMI 2/3 (n = 940, 45.1%) groups. Thereafter, the pre-TIMI 0/1 group was further divided into the ticagrelor monotherapy after 3-mo DAPT (3-mo DAPT, n = 582, 50.9%) group and the ticagrelor-based 12-mo DAPT (12-mo DAPT, n = 561, 49.1%) group. The pre-TIMI 2/3 group was also divided into the 3-mo (n = 475, 50.5%) and 12-mo (n = 465, 49.5%) DAPT groups (Fig. 4). The study protocol was approved by the Institutional Review Board (IRB) of each participating center and the Kangwon National University IRB (No. KNUH-2021-02-014), and was conducted in compliance with the ethical standards of the Declaration of Helsinki. Informed written consent was obtained from all patients prior to their inclusion in the study.

PCI procedure and medical treatment

Diagnostic coronary angiography and PCI were performed using standard techniques25. If the patient was not taking aspirin or ticagrelor at the time of PCI, a loading dose of aspirin (300 mg) and ticagrelor (180 mg) were administered before PCI. Thereafter, 100 mg of aspirin per day and 90 mg of ticagrelor twice per day were prescribed as daily maintenance therapy4. After 3-mo DAPT consisting of aspirin and ticagrelor, aspirin was discontinued in the ticagrelor monotherapy group and continued in the 12-mo DAPT group4.

Study endpoints and definitions

The primary outcome was the occurrence of a NACE, defined as a composite of TIMI major bleeding and MACCE within 12 months of index PCI. The second outcome was the occurrence of TIMI major, minor, and major or minor bleeding and the occurrence of individual components of MACCE, defined as all-cause death, cardiac death (CD), myocardial infarction (MI), target vessel revascularization (TVR), ST, and stroke. Major bleeding was defined according to the TIMI criteria: intracranial bleeding, hemorrhage with a hemoglobin decrease of at least 5 g/dL, or fatal bleeding that caused death within 7 days4,26. Definitions of CD, MI, TVR, ST, and stroke have already been published4. In case of NSTEMI, culprit vessel was evaluated by coronary angiographic findings, 12-lead electrocardiogram, two-dimensional echocardiogram, and noninvasive stress test27,28. A successful PCI was defined as a residual stenosis of < 30% and TIMI flow grade 3 for the infarct-related artery after the procedure. All baseline and procedural angiographic images including TIMI flow grade of the enrolled patients were centrally collected, and quantitative and qualitative analyses were independently performed in the central angiographic core laboratory (Cardiovascular Research Institute, Severance Cardiovascular Hospital, Seoul, South Korea). Moreover, the PRECISE-DAPT (Predicting Bleeding Complications in Patients Undergoing Stent Implantation and Subsequent Dual Antiplatelet Therapy) score was assessed using an online calculator (http://www.precisedaptscore.com) with 5 variables (age, creatine clearance, hemoglobin, white blood cell count, and previous spontaneous bleeding)29. Adverse events were centrally collected, and any document that could lead to unblinding of treatment assignment was obliterated before submission to the clinical event committee. Outcomes were categorized according to predefined criteria by an independent clinical event committee blinded to the treatment assignments and primary results of the trial4.

Statistical analysis

Primary analyses of this study were performed in an intention-to-treat manner. Pre-specified 3-month landmark analyses were performed. Post-hoc analyses were performed for the as-treated population regarding the actual treatments received. Categorical data were reported as numbers and percentages, and they were compared using the chi-square test or Fisher’s exact test, as appropriate. Continuous variables were expressed as mean ± standard deviation, and were compared using the Student’s t-test. Various clinical outcomes were estimated using the Kaplan–Meier method, and intergroup differences were compared using the log-rank test. To determine meaningful variables, all variables with p < 0.1 and known conventional risk factors for poor outcomes in the AMI population were considered potential confounding factors and were included in the univariate analysis (Supplementary materials 15 and 16). Variables with p < 0.05 were included in the multivariate analysis model. For all analyses, a two-sided p value < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software version 20 (IBM, Armonk, NY, USA).