The present study was approved by the Institutional Review Board at Taipei Veterans General Hospital, Taipei, Taiwan (IRB no. 2022-01-033CC). The data used in this study were anonymized before analysis. From January 2016 to December 2020, a total of 1,268 patients admitted to the Coronary Care Unit due to AMI were retrospectively screened in this study. The exclusion criteria were as follows: (1) nondiabetic patients, (2) those with a baseline estimated glomerular filtration rate (eGFR) of <30 mL/min/1.73 m², (3) those who experienced failed revascularization or mortality during the index AMI episode. The patients who had been taking SGLT2 inhibitors before or during the index AMI episode, and continued using SGLT2i during the clinics follow-up, were defined as group 1. The patients who had never used SGLT2 inhibitors were defined as group 2. As shown in Figure 1, the study subjects of groups 1 and 2 were matched in terms of age, sex, congestive heart failure (CHF), and chronic kidney disease (CKD) at a 1:2 ratio. The definition of AMI was based on the diagnostic criteria; AMI was diagnosed by a coronary angiogram and confirmed by two expert cardiologists (21). By definition, all of the enrolled patients received successful revascularization and were discharged with stable conditions.

The demographic profiles, cormobidities, laboratory data, and medication profiles of the participants were collected by the review of medical records. After being discharged, all of the participants were scheduled to have a follow-up visit at the clinic 2 weeks later. The subsequent clinic visits were arranged at 1–3-month intervals, according to the discretion of the attending cardiologist. Data of the follow-up period were obtained by a medical record review and confirmed by a telephone interview. The composite of rehospitalization due to acute coronary syndrome (ACS) or sudden cardiac death was defined as an adverse cardiovascular outcome. Meanwhile, the composite of the initiation of chronic dialysis, renal transplant, eGFR decline of ≥40%, or eGFR of <15 mL/min/1.73 m² was defined as an adverse renal outcome. Chronic dialysis was defined as dialysis treatment for more than 90 days (22).

All analyses were performed using SPSS statistical software, version 24.0 (SPSS, Inc., Chicago, IL, USA). Continuous variables were reported as the mean and standard deviation. Categorical variables were reported as frequencies and percentages. The chi-squared test was used to analyze categorical variables. Kaplan–Meier survival curves with the log-rank test were used to compare the occurrence of outcomes. The Cox proportional regression model was used to calculate the hazard ratio (HR) for the occurrence of outcomes.

This retrospective observational study included 198 AMI patients, among whom 66 patients were in group 1, and 132 patients were in group 2. The mean age of the patients was 66.1 ± 12.3 years old in group 1 and 67.7 ± 11.9 years old in group 2. The age, sex, comorbidities, and baseline eGFR were similar between the two groups. Notably, the percentage of patients with ST-elevation myocardial infarction, the thrombolysis in myocardial infarction (TIMI) risk score, and the volume of contrast medium used for revascularization were not significantly different between the two groups (Table 1). The above findings suggest that the two matched groups presented to the hospital with a similar severity of AMI. In group 1, the majority of SGLT2 inhibitor prescription was empagliflozin (78.8%), and 12(18.2%) patients have been using the SGLT2 inhibitors before the indexe AMI hospitalization, with the mean duration of 12.8 ± 9.2 months.

With a mean follow-up period of 23.5 ± 15.7 months (data not shown), 3 (4.5%) patients in group 1 and 22 (16.7%) patients in group 2 experienced rehospitalization due to ACS, while 1 (1.5%) patient in group 1 and 7 (5.3%) patients in group 2 experienced sudden cardiac death. Overall, the adverse cardiovascular outcomes occurred more frequently in group 2 (Table 1). To confirm the higher risk of adverse cardiovascular outcomes in group 2, Kaplan–Meier survival curves were constructed. The results demonstrated that the patients in group 1 had a longer adverse cardiovascular outcome-free survival than the patients in group 2 (Figure 2). The above findings suggest that post-AMI patients treated with SGLT2 inhibitors have a lower risk of adverse cardiovascular outcomes than those who are not taking this class of drug.

Regarding the adverse renal outcomes, there were 4 (6.1%) events in group 1 and 17 (12.9%) events in group 2; however, the difference was not significantly different between the two groups (Table 1). Similarly, the Kaplan–Meier survival curves showed that the adverse renal outcome-free survival was not significantly different between the two groups (Figure 3). The above findings suggest that SGLT2 inhibitors do not change the risk of adverse renal outcomes in post-AMI patients.

Compared with the patients without adverse cardiovascular outcomes, those who experienced adverse cardiovascular outcomes were older (72.3 ± 14.6 years old vs. 68.8 ± 11.5 years old, P = 0.045) and had a lower baseline eGFR (58.0 ± 20.8 vs. 69.3 ± 20.3, P = 0.004). Notably, the percentage of patients using SGLT2 inhibitors was significantly lower in the group of patients with an adverse cardiovascular event (12.1 vs. 37.6%, P = 0.004, Table 2).

To build the multivariate regression model, the candidate predictors were evaluated for their association with adverse cardiovascular outcomes using univariate Cox proportional regression analysis. The predictors with P-values of <0.05 were included in the multivariate Cox proportional regression model. Accordingly, age, baseline eGFR, and the use of SGLT2 inhibitors were included in the multivariate regression model. The results demonstrated that only the baseline eGFR (P = 0.008, 95% CI: 0.944–0.991) and the use of SGLT2 inhibitors (P = 0.039, 95% CI: 0.116–0.947) remained significantly associated with the risk of adverse cardiovascular outcomes (Table 3).

Following AMI, heart failure with adverse remodeling of left ventricle characterized by chamber dilatation and impaired cardiac function is the common outcome (31). In one recently published meta-analysis, including a total of 13 randomized controlled trials that evaluated the effects of SGLT2 inhibitors on cardiac remodeling in patients with T2DM and/or HF, SGLT2 inhibitors improved left ventricular (LV) ejection fraction, LV mass, LV mass index, LV end-systolic volume, LV end-systolic volume index, and E-wave deceleration time significantly (32). There were plausible effects of SGLT2 inhibitors on adversecardiac remodeling. First, myocardial ischemia could impair cardiomyocyte autophagy, which has been shown an essential mechanism that protects against adverse cardiac remodeling (33). Recent experimental studies have indicated that SGLT2 inhibitors might exert cardioprotective effects by stimulating autophagy (34). Second, cardiac mitochondrial dysfunction during ischaemia and reperfusion injury is a critical determinant of post-infarcted cardiac cell death, and is associated with future adverse cardiac remodeling (35). There have been several studies demonstrated that cardiac mitochondrial function could be improved by SGLT2 inhibitors (36–38). Taken together, reversed cardiac remodeling may be a mechanism responsible for the favorable clinical effects of SGLT2 inhibitor on patients with heart failure (39).

The meta-analysis of the four major CVOTs mentioned above showed a 12% reduction of the risk for MACE in the group taking SGLT2 inhibitors (HR, 0.88; 95%CI: 0.82–0.94) (40). In particular, there was a 17% reduction in sudden cardiac deaths and a 12% reduction in myocardial infarctions. In this meta-analysis, the authors also analyzed the effects of SGLT2 inhibitors in those with vs. without cardiovascular disease at baseline. The results showed that the risk reduction of myocardial infarction in the secondary prevention cohort was more significant than that in the primary prevention cohort (HR: 0.86; 95% CI: 0.80–0.93 for secondary prevention; and HR: 0.94; 95% CI: 0.82–1.07 for primary prevention). The other two meta-analysis studies also revealed risk reductions of 18% and 21%, respectively, for cardiovascular mortality in the SGLT2 inhibitor-treated group (41, 42). In accordance with these previous studies, our results showed that the patients taking SGLT2 inhibitors had fewer adverse cardiovascular outcomes, including rehospitalization for ACS and sudden cardiac death. Since the study population consisted of patients with stabilized AMI, our results demonstrate the protective effect of SGLT2 inhibitors on the secondary prevention of AMI, which is in line with a previous report (40).

The meta-analysis of four CVOTs, including 38,723 participants with T2DM, demonstrated a risk reduction of 35% for end-stage renal disease in patients taking SGLT2 inhibitors (43). Nonetheless, another meta-analysis showed that while treatment with SGLT2 inhibitors reduced the risk of major renal outcomes by 46% in patients with macroalbuminuria and atherosclerotic cardiovascular disease, it had no significant effect on renal outcomes in the subgroup with eGFR <60 mL/min/1.73 m² (HR: 0.74, 95% CI: 0.51–1.06) (44). In the current study, we did not observe a significant benefit of SGLT2 inhibitors on renal outcomes in diabetic patients with AMI. This finding could be explained by the small number of cases analyzed or the limited observation period.

In the past two decades, although the mortality due to AMI has improved (45, 46), the long-term cardiovascular mortality and post-AMI heart failure remain significant issues (45, 47). Thus, cardiac protection with early reperfusion to reduce the size of the myocardial infarct and the incidence of post-AMI cardiovascular adverse events are important topics to be studied. To the best of our knowledge, this is the first study showing that SGLT2 inhibitor use reduces the risk of adverse cardiovascular outcomes in T2DM patients with stabilized AMI.

This study has several limitations that must be addressed. First, due to the long period of enrollment, heterogeneity in treatment/stenting strategies may exist and confound the analysis. A second limitation is the single-center design and the relatively small number of included patients. A third limitation is that the physician skills regarding coronary revascularization were not controlled, which may also confound the analysis. Fourth, although a propensity score-adjusted analysis was employed to minimize the selection bias, remaining bias may still affect the analysis. Fifth, the retrospective nature of this study and the low number of events in each group limited further important subgroup analyses, such as an analysis based on the presence of absence of heart failure.

In conclusion, the findings of the present study might suggest that in T2DM patients with stabilized AMI, the use of SGLT2 inhibitors is associated with a lower risk of adverse cardiovascular outcomes, including rehospitalization for ACS and sudden cardiac death. Our study also demonstrates at least partly that the use of SGLT2 inhibitors could provide cardioprotection to T2DM patients with AMI. However, studies with a larger sample size are needed to verify these findings.