From the Korea AMI Registry (KAMIR) [14], a total of 23,391 AMI patients aged ≥ 30 years at diabetes onset, who underwent successful DES implantation from November 2005 to June 2015, were evaluated. KAMIR [14] is a prospective, observational, and on-line registry with a multicenter cohort study in South Korea, established in November 2005. Details of the registry can be found at the KAMIR website (http://​www.​kamir.​or.​kr). In this study, we tried to confine T2DM patients to diabetes cases. Therefore, we defined T2DM based on a previous study [15] which also included patients from the KAMIR. In our study, patients with incomplete laboratory results such as unidentified results of blood hemoglobin (Hb) A1c and blood glucose (n = 8432, 36.1%), patients lost to follow-up (n = 1069, 4.6%), patients treated with first-generation DES (n = 1928, 8.2%), and patients treated with uncertain doses of statins (n = 350, 1.5%) were excluded. Thus, 11,612 AMI patients who received newer-generation DES were included. The types of newer-generation DES used are listed in Table 1. The patients were classified as statin users (n = 9893, 85.2%) and statin non-users (n = 1719, 14.8%). Thereafter, statin users were further divided into high-intensity (n = 2984, 30.2%) and low-moderate-intensity statin users (n = 6909, 69.8%). Finally, those in these two groups (A and B, respectively) were further classified as patients with normoglycemia (group A1 [n = 806, 27.0%] and B1 [n = 1815, 26.3%]), prediabetes (group A2 [n = 935, 31.3%] and B2 [n = 2145, 31.0%]), and T2DM (group A3 [n = 1243, 41.7%] and B3 [n = 2949, 42.7%]) (Fig. 1, Table 1, and Additional file 1: 1). Additionally, over time, patients enrolled later may have benefited from innovative therapies that may have impacted prognosis. To assess how much the results are influenced by this point, we stratified patients into two groups before and after October 2012 according to the enrolled date of individual patient (Additional file 1: 2, 3, 4, and 5). Because a European Society of Cardiology guideline for management of AMI [16] was published in October 2012, and many treatment strategies could be changed according to the newly published guidelines, October 2012 became the cutoff point for our classification. The study protocol was approved by the Institutional Review Board of each participating center, and it was conducted in compliance with the ethical standards of the Declaration of Helsinki 1975. Informed consent was obtained from all patients prior to their inclusion in the study, and we followed up all enrolled patients through face-to-face interviews, phone calls, and chart reviews. All 11,612 patients completed a 2-year clinical follow-up, and all clinical events were evaluated by an independent event adjudication committee. The processes of event adjudication have been described previously by the KAMIR investigators [14].

Diagnostic coronary angiography and PCI were performed using standard techniques [17]. All patients received loading doses of aspirin (200–300 mg) and other antiplatelet agents such as clopidogrel (300–600 mg), ticagrelor (180 mg), or prasugrel (60 mg), before PCI was performed. It was recommended that the duration of dual antiplatelet therapy (DAPT; a combination of aspirin 100 mg/day with clopidogrel 75 mg/day, ticagrelor 90 mg twice daily, or prasugrel 5–10 mg/day) should be for at least 1 year after the index PCI. Based on previous reports [18, 19], triple antiplatelet therapy (TAPT; cilostazol [100 mg twice daily] combined with DAPT) was determined by the individual operator’s discretion. In this study, the patients who received atorvastatin, rosuvastatin, simvastatin, pitavastatin, pravastatin, and fluvastatin were included (Table 1), and the kind and dose of statins to be used was left at the physicians’ discretion.

In this study, as mentioned [10], because moderate-dose statin treatment is more commonly administered due to lower bodyweights in Asian population, atorvastatin (≥ 40 mg), rosuvastatin (≥ 20 mg), simvastatin (≥ 40 mg), pitavastatin (≥ 4 mg), and pravastatin (≥ 40 mg) were considered as high-intensity statins, while others were considered as low-moderate-intensity statins [20], compared with current guideline [21]. Glycemic status was determined by the clinical practice recommendations of the American Diabetes Association [22]. T2DM was defined as either known T2DM for which patients received medical treatment (insulin or antidiabetics) or newly diagnosed T2DM defined as an HbA1c level ≥ 6.5%, a fasting plasma glucose (FPG) level ≥ 126 mg/dL (7 mmol/L), and/or random plasma glucose (RPG) level ≥ 200 mg/dL (11.1 mmol/L), during the index hospitalization or according to their medical history. Prediabetes was defined as an HbA1c level of 5.7%–6.4% and an FPG of 100–125 mg/dL (5.6–6.9 mmol/L). Moreover, in the case of discrepancies between HbA1c and FPG or RPG levels, we made HbA1c level a priority [12]. AMI was defined according to the current guidelines [5‐8]. Estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease (MDRD) study equation [23]. The major outcome was the occurrence of MACE defined as all-cause death, Re-MI, or any repeat coronary revascularization. All-cause death was classified as CD or non-CD. Any repeat revascularization comprised of target lesion revascularization (TLR), target vessel revascularization (TVR), and non-TVR. The definitions of Re-MI, TLR, TVR, and non-TVR have been published previously [24].

The normality test was conducted using the Kolmogorov–Smirnov test. Categorical data were reported as numbers and percentages, and they were compared using the chi-square or Fisher’s exact test, as appropriate. For continuous variables, differences among the three groups are evaluated using an analysis of variance or the Jonckheere-Terpstra test, while a post-hoc analysis was performed using the Hochberg test or Dunnett T3 test. The data were expressed as mean ± standard deviation. To determine meaningful variables, all variables with p < 0.001 were included in the univariate analysis (Additional file 1: 6). After univariate analysis, variables with p < 0.001 and known conventional risk factors of poor outcomes in the AMI population were considered potential confounding factors, and were entered into the multivariate analysis [25]. Various clinical outcomes were estimated using the Kaplan–Meier method, and intergroup differences were compared using the log-rank test. For all analyses, a two-sided p value < 0.05 was considered statistically significant. All statistical analyses were performed using the SPSS software version 20 (IBM, Armonk, NY, USA).

Table 1, Additional file 1: 1, 7, 8, and 9 show the baseline characteristics of the study population. Both in high-intensity (group A) and in low-moderate-intensity (group B) statin users, the number of men, single-vessel disease, and the prescription rates of ticagrelor and angiotensin-converting enzyme inhibitors (ACEIs) were the highest in normoglycemia groups (group A1 and B1). The number of current smokers and peak creatine kinase-MB level and the levels of total and low-density lipoprotein (LDL) cholesterols were the highest in prediabetes groups (group A2 and B2). The mean age and the number of patents with hypertension, dyslipidemia, previous history of PCI and cerebrovascular accidents; levels of N-terminal pro-brain natriuretic peptide, serum creatinine, and triglyceride; the prescription rates of cilostazole and calcium channel blockers; the number of cases with right coronary artery (RCA) as infarc-related artery (IRA) and treated vessel, multivessel disease, and deployed stents, were the highest in T2DM group (group A3 and B3).

In both high-intensity and low-moderate intensity statin users, the comparisons of clinical outcomes among the three glycemic groups during the 2-year follow-up period are presented in Tables 2, 3, and Fig. 2. In high-intensity statin users, the cumulative incidences of MACE (adjusted hazard ratio [aHR]: 2.187; 95% confidence interval [CI]: 1.341–3.569; p = 0.002) and any repeat revascularization (aHR 3.009; 95% CI 1.342–6.745; p = 0.006) were higher in group A2 (prediabetes) than in group A1 (normoglycemia). Similarly, the cumulative incidences of MACE (aHR 2.368; 95% CI 1.480–3.788; p = 0.001) and any repeat revascularization (aHR 3.619; 95% CI 1.659–7.898; p = 0.001) were significantly higher in group A3 (T2DM) than in group A1. However, the cumulative incidences of MACE, all-cause death, CD, Re-MI, and any repeat revascularization were similar between groups A2 and A3 (Table 2). In low-moderate-intensity statin users, the cumulative incidences of MACE, all-cause death, CD, Re-MI, and any repeat revascularization were not significantly different between groups B1 (normoglycemia) and B2 (prediabetes) as well as between groups B2 and B3 (T2DM). However, the cumulative incidences of MACE (aHR 1.285; 95% CI 1.014–1.629; p = 0.038) and all-cause death (aHR 1.784; 95% CI 1.156–2.751; p = 0.009) were significantly higher in group B3 than in group B1. In normoglycemia groups (Table 3), the cumulative incidences of MACE (aHR 1.903; 95% CI 1.203–3.010; p = 0.006) and any repeat revascularization (aHR 3.248; 95% CI 1.539–6.854; p = 0.002) were significantly lower in high-intensity than in low-moderate-intensity statin users. However, both in the prediabetes (groups A2 and B2) and T2DM (group A3 and B3) groups, the cumulative incidences of MACE, all-cause death, CD, Re-MI, and any repeat revascularization were similar between high-intensity (group A2 vs. B2) and low-moderate-intensity statin users (group A3 vs. B3). Furthermore, in the total study population, there were no significant differences in major clinical outcomes between high-intensity (A1 + A2 + A3) and low-moderate-intensity stain users (B1 + B2 + B3). In high-intensity statin users, in both before and after October 2012 groups (Additional file 1: 2 and 3), the cumulative incidences of MACE (aHR 2.635; p = 0.003 and aHR 1.845; p = 0.048, respectively) and any repeat revascularization (aHR 4.162; p = 0.002 and aHR 2.845; p = 0.044, respectively) were higher in group A2 than in group A1. The cumulative incidences of MACE (aHR 2.896; p = 0.002 and aHR 2.146; p = 0.033, respectively) and any repeat revascularization (aHR 4.666; p = 0.001and aHR 3.241; p = 0.040, respectively) were significantly higher in group A3 than in group A1. In low-moderate-intensity statin users, in before October 2012 group (Additional file 1: 2), the cumulative incidence of all-cause death was significantly higher in group B3 than in group B1 (aHR 1.621; 95% CI 1.102–2.614; p = 0.044). In after October 2012 group (Additional file 1: 3), the cumulative incidences of MACE (aHR 1.429; 95% CI 1.001–1.998; p = 0.043), all-cause death (aHR 2.940; 95% CI 1.388–6.225; p = 0.005), CD (aHR 3.319; 95% CI 1.235–8.919; p = 0.017) were significantly higher in group B3 than in group B1. Moreover, the cumulative incidence of CD (aHR 2.757; 95% CI 1.038–7.327; p = 0.042) was significantly higher in group B3 than in group B2. In before October 2012 group (Additional file 1: 4), in normoglycemia groups, the cumulative incidences of any repeat revascularization (aHR 3.025; 95% CI 1.045–8.760; p = 0.041) was significantly lower in high-intensity than in low-moderate-intensity statin users In after October 2012 group (Additional file 1: 5), in both normoglycemia groups and total study population, the cumulative incidences of MACE (aHR 2.002; p = 0.042 and aHR 1.533; p = 0.004, respectively) and any repeat revascularization (aHR 3.308; p = 0.028 and aHR 1.587; p = 0.038, respectively) were significantly lower in high-intensity than in low-moderate-intensity statin users. In the comparison of major clinical outcomes between statin users and non-users (Additional file 1: 10), statin non-users showed higher cumulative incidences of MACE, all-cause death, and CD in all three glycemic statuses. Additionally, in the T2DM group, the cumulative incidence of any repeat revascularization (aHR 1.637; 95% CI 1.171–32.290; p = 0.004) was significantly higher in statin non-users than in statin users. Moreover, in the total study population, the cumulative incidences of MACE, all-cause death, CD, and any repeat revascularization were significantly higher in statin non-users than in statin users.

Independent predictors for MACE in high-intensity statin users and in low-moderate-intensity statin users at 2 years are listed in Additional file 1: 11 and 12. In both high-intensity and low-moderate-intensity statin users, decreased left ventricular ejection fraction (LVEF) (< 40%), cardiogenic shock, and decreased eGFR (< 60 mL/min/1.73m²) were found to be significantly common independent predictors for MACE.