Prognostic Impact of Sodium-Glucose Cotransporter 2 Inhibitors in Patients with Type 2 Diabetes Mellitus and Coronary Ischemia: A Retrospective Cohort Study
Article information
Abstract
Background
Patients with type 2 diabetes mellitus (T2DM) and coronary ischemia face an exceptionally elevated risk, and the achievement of complete revascularization (CR) within this population could be challenging.
Methods
Patients with T2DM and coronary ischemia based on coronary angiography and retrospective angiographic fractional flow reserve analysis between 2014 and 2016 were included. The impact of the extent of revascularization on the improvement of endpoint events by sodium-glucose cotransporter 2 (SGLT2) inhibitors was analyzed. The primary study endpoint was major adverse cardiac events (MACE), while all-cause mortality served as secondary endpoints. Kaplan-Meier analysis and Cox proportional hazards regression model were adopted to assess the association between SGLT2 inhibitors and endpoint incidence.
Results
A total of 671 patients were identified. Among them, 206 (30.7%) were prescribed with SGLT2 inhibitors, while 484 (72.1%) achieved CR after the operation. During a mean 36-month follow-up, 100 MACE and 89 all-cause mortality were recorded. SGLT2 inhibitor users demonstrated lower rates of MACE (8.3% vs. 17.8%, P=0.002) and all-cause mortality (6.3% vs. 16.3%, P<0.001) compared to non-users. After adjusting for confounding factors in multivariable Cox analysis, the association between SGLT2 inhibitors and reduced MACE incidence remained consistent both in the CR and incomplete revascularization subgroups (hazard ratio [HR], 0.498; 95% confidence interval [CI], 0.246 to 0.938; P=0.040; and HR, 0.341; 95% CI, 0.123 to 0.805; P=0.023, respectively).
Conclusion
SGLT2 inhibitors were found to be associated with a reduced risk of 3-year MACE and all-cause mortality in patients with T2DM and coronary ischemia, regardless of extent of revascularization.
Highlights
• SGLT2 inhibitors reduce 3-year MACE and all-cause mortality in T2DM with coronary ischemia.
• The benefit persists irrespective of complete or incomplete revascularization.
• Risk reduction is primarily driven by decreased CV mortality and HF hospitalization.
• SGLT2 inhibitors improve prognosis in high-risk patients with incomplete revascularization.
INTRODUCTION
Coronary ischemia caused by coronary artery disease (CAD) is a prevalent condition among individuals with type 2 diabetes mellitus (T2DM) and plays a substantial role in mortality and morbidity [1-3]. Although revascularization has been acknowledged as the key factor in determining the prognosis of patients with coronary ischemia, those with T2DM frequently exhibit advanced age, multiple comorbidities, heart failure (HF), and impaired renal function, predisposing this population to an increased risk associated with procedures. Furthermore, coronary artery lesions in patients with T2DM are characterized by multivessel involvement, diffuse distribution, calcification, and suboptimal peripheral vascular conditions, posing significant challenges to achieve complete revascularization (CR) [2,4-6]. Previous studies on percutaneous coronary intervention (PCI) revascularization have largely found that one-third to one-half of the overall population fail to achieve CR, and those with incomplete revascularization (ICR) often have a higher risk of cardiovascular adverse events and death [7]. However, beyond the well-known medications that are effective for CAD patients, such as anti-platelet agents, statins, and beta-blockers, there has been limited research on medical therapy specifically targeting the high-risk population with ICR.
Sodium-glucose cotransporter 2 (SGLT2) inhibitors, a novel class of hypoglycemic agents, have been widely prescribed to patients with T2DM or HF due to their unique cardioprotective benefits [8-10]. Multiple randomized controlled trials (RCTs) have affirmed the role of SGLT2 inhibitors in mitigating cardiovascular risk independent of glucose control [11-14], leading to their recommendation for use in patients with T2DM and atherosclerotic cardiovascular disease (ASCVD) in the 2025 American Diabetes Association and 2023 European Society of Cardiology guidelines [15,16]. However, in the real world, the effectiveness of SGLT2 inhibitors in patients with T2DM and coronary ischemia, especially those at residual risk due to ICR, remains unclear. This study aims to examine the 3-year clinical outcomes stratified by revascularization status and the use of SGLT2 inhibitors in patients with T2DM and coronary ischemia with either CR or ICR.
METHODS
Study population
In this retrospective observational cohort study, patients with T2DM who received coronary angiography (CAG) at Queen Mary Hospital between January 1, 2014 and December 31, 2016 were enrolled. The T2DM population was defined as those who met a T2DM diagnosis and were using hypoglycemic medications. A diagnosis of T2DM was established by fulfilling at least one of the following criteria: glycosylated hemoglobin (HbA1c) level of 6.5% or higher; random plasma glucose level of 200 mg/dL (11.1 mmol/L) or higher; fasting blood glucose (FBG) level of 126 mg/dL (7.0 mmol/L) or higher; and a 2-hour glucose level in venous plasma during an oral glucose tolerance test of 200 mg/dL (11.1 mmol/L) or higher [17]. Patients were excluded if the stenosis of each major coronary artery <50% (n=167), the angiographic fractional flow reserve (FFR) value cannot be obtained (n=111), coronary artery bypass graft (CABG) revascularization was performed after CAG (n=15). Patients who have received SGLT2 inhibitors but did not fulfil the criteria of SGLT2 inhibitor user were also excluded (n=101). Demographic information (i.e., age, sex) and clinical data including diagnosis, medical history, laboratory findings or CAG and revascularization procedures were retrived from the electronic patient record system. Baseline data were collected within 1 week prior to CAG.
Coronary ischemia and revascularization
The determination of coronary ischemia was based on CAG results and retrospective coronary angiography-derived fractional flow reserve (caFFR) analysis. The caFFR measurements were conducted by two experienced cardiologists independently, utilizing the advanced FlashAngio system’s software (FlashAngio, Rainmed Ltd., Suzhou, China). To ensure impartiality and objectivity, the practitioners were blinded to the patients’ initial data and subsequent clinical outcomes throughout the process, and the specific measurement principles and operational procedures can refer to our previous study [18]. The presence of coronary ischemia was defined as the patient having at least one major coronary artery with a caFFR value ≤0.8 [19,20], or a diameter stenosis of 90% or more [21,22] (including chronic total occlusion lesions) after CAG. For cases that caFFR measurement values could not be obtained due to the following reasons: suboptimal angiographic image quality, significant vascular overlap or distortion of the target vessel, and poor contrast opacification were deemed inadequate for contour detection.
The definition of CR adhered to the ischemic/functional CR criteria, entailing the successful revascularization of all major coronary artery lesions or segments exhibiting evidence of ischemia on CAG or caFFR, irrespective of their anatomical severity. In contrast, ICR was defined as the persistence of at least one major coronary artery lesion or segment with ischemic evidence following revascularization procedures [7]. In this study, regardless of whether the patient underwent interventional therapy, ICR was considered if the caFFR value of any major coronary artery branch was ≤0.8 based on the final angiographic image analysis before the end of the procedure. The methods utilized for revascularization encompassed bare metal stents, drug-eluting stents, bioabsorbable stents, and drug-coated balloons.
SGLT2 inhibitor user
SGLT2 inhibitor users were defined as patients who received SGLT2 inhibitor therapy for ≥90 consecutive days during follow-up, irrespective of prior exposure [23]. SGLT2 inhibitor non-users comprised those with no history of SGLT2 inhibitor use, or former users who discontinued therapy ≥6 months before enrollment and remained off-treatment throughout follow-up. Patients initiating SGLT2 inhibitors during follow-up but failing to complete ≥90 days of therapy were excluded.
Follow-up and clinical endpoints
Follow-up started from the date after CAG. For patients who needed to achieve CR in several times, regardless of whether the final revascularization was successful or not, the last angiography was taken as the starting point for follow-up observation [24,25]. The primary endpoint of this study was the occurrence of major adverse cardiac events (MACE), a composite measure that encompassed cardiovascular (CV) mortality, non-fatal myocardial infarction (MI), and HF hospitalization. The secondary endpoint was defined as all-cause mortality. CV mortality was a broad term that encompassed deaths resulting from a diverse range of CV conditions, including but not limited to acute MI, sudden cardiac death, HF, stroke, CV procedures, CV hemorrhage, and other related CV disorders. The diagnosis of MI was established in the presence of clinical symptoms suggestive of myocardial ischemia, coupled with significant dynamic changes in cardiac biomarkers, reflecting myocardial injury. HF hospitalization was defined as an episode requiring a minimum of 24 hours of inpatient care, during which the patient’s HF medication regimen remained unchanged, with the primary objective being the management and alleviation of HF symptoms. All endpoint events occurring in patients during the follow-up period were recorded, but if the same patient experienced multiple MACEs, only the first occurrence of the event was counted when tallying the composite MACEs. The follow-up continued until occurrence of study outcome, death, or end of data collection, whichever came first.
Statistical analysis
The data obtained from this study were meticulously analyzed using the latest versions of the SPSS version 27 (IBM Co., Armonk, NY, USA) and GraphPad Prism version 10 (GraphPad Software Inc., San Diego, CA, USA). To ensure accuracy and precision, numerical variables that adhered to a normal distribution were presented as the mean±standard deviation, whereas those that deviated from normality were aptly represented by the interquartile range. Categorical variables, on the other hand, were expressed as percentages. Intergroup comparisons of numerical variables that follow a normal distribution are performed using the independent samples t-test, and the Mann-Whitney U test was used if they did not follow a normal distribution. For categorical variables, the chi-square test and Fisher’s exact test were employed to determine differences. Kaplan- Meier analysis was applied to estimate the survival rates free from endpoint events, with the log-rank test used to assess differences in survival curves. The relationship between SGLT2 inhibitor and clinical outcomes was evaluated via Cox proportional hazards regression analysis. Univariate analyses were conducted to investigate the correlation between various variables and MACE. Subsequently, multivariable regression models were constructed based on the variables from the univariate regression models. The hazard ratio (HR) with a 95% confidence interval (CI) was computed to quantify the strength of associations. All statistical tests were two-sided, with a P value threshold of less than 0.05 for determining statistical significance.
Ethics approval and consent to participate
This study was carried out in accordance with the Helsinki Declaration and was approved by the ethics committee of the West Cluster Hospital Authority of Hong Kong (UW 19-575). Considering the retrospective study design, the need for informed consent was waived.
RESULTS
Baseline characteristics
A total of 671 patients with T2DM who confirmed to have coronary ischemia by angiographic FFR were enrolled (Fig. 1). The mean age of the overall study population was 69 years, with 490 (73.0%) were male. In the entire cohort, 206 (30.7%) patients fulfilled our definition of SGLT2 inhibitor users. The median duration from the start of medication to CAG for patients in the SGLT2 inhibitor group was 5.4 months, with an average medication duration of 17 months throughout the follow-up period. The SGLT2 inhibitors users were younger and had a higher proportion of males and smokers. Apart from a higher proportion of patients with history of stroke/transient ischemic attack (TIA) in the SGLT2 inhibitors users, there were no differences in other medical histories between the two groups. FBG and HbA1c levers were higher in the SGLT2 inhibitor users compared to the non-users at baseline. SGLT2 inhibitor users had significantly higher baseline estimated glomerular filtration rate than non-users. CR was achieved in 484 (72.1%) individuals in the entire cohort, there is no difference in the proportion of CR between SGLT2 inhibitors users and non-users. No statistically significant differences were detected in other baseline parameters between the two cohorts (Table 1).
Flowchart of patients enrolled. CAG, coronary angiography; T2DM, type 2 diabetes mellitus; FFR, fractional flow reserve; SGLT2i, sodium-glucose cotransporter 2 inhibitor; CABG, coronary artery bypass graft; CR, complete revascularization; ICR, incomplete revascularization.
Baseline characteristics of the study population
Coronary artery lesion and revascularization status of the total population are shown in Supplementary Table 1. A total of 1,043 vessels exhibited functional ischemia as defined by angiographic FFR. The left anterior descending branch (LAD) was the most frequently affected vessel, accounting for 41.0% of cases, followed by the left circumflex artery (LCX) at 32.5% and the right coronary artery at 26.5%. Among the 226 (21.7%) non-revascularized vessels, the LCX had the highest non-revascularization rate at 34.5%, while the LAD had the lowest rate at 11.7%, as demonstrated in Supplementary Fig. 1. The most common reason for not achieving revascularization was a joint decision made by the physician and the patient, accounting for 47.8% of cases. Intervention failure was observed in 16.4% of cases, and 35.8% of cases involved coronary artery lesions initially unidentified as having functional ischemia, only recognized through retrospective angiographic FFR measurements (Supplementary Fig. 1).
Clinical outcome
Over an average follow-up period of 36 months, a total of 100 MACE and 89 all-cause mortality cases were recorded. The use of SGLT2 inhibitors was associated with a reduced risk of MACE (8.3% vs. 17.8%; HR, 0.433; 95% CI, 0.249 to 0.711; P=0.002) and all-cause mortality (6.3% vs. 16.3%; HR, 0.360; 95% CI, 0.191 to 0.625; P<0.001) in patients with T2DM and coronary ischemia. Further observation revealed that the difference in MACE in the overall population was driven by CV mortality (1.5% vs. 5.8%; HR, 0.244; 95% CI, 0.058 to 0.690; P=0.020) and HF hospitalization (5.3% vs. 11.8%; HR, 0.434; 95% CI, 0.215 to 0.795; P=0.011). However, SGLT2 inhibitors failed to reduce the incidence of non-fatal MI within 3 years (Fig. 2). Kaplan-Meier survival curves also showed similar trend as the main analysis (log-rank P value were all <0.01) (Fig. 3). In the analysis stratified by revascularization status, SGLT2 inhibitors was able to reduce the risk of MACE and all-cause mortality in both the CR and ICR subgroups (P for interaction=0.804 and 0.730, respectively) (Fig. 2).
The impact of sodium-glucose cotransporter 2 inhibitors (SGLT2is) on outcomes in patients with type 2 diabetes mellitus and coronary ischemia. HR, hazard ratio; CI, confidence interval; MACE, major adverse cardiac events; CR, complete revascularization; ICR, incomplete revascularization; CV, cardiovascular; MI, myocardial infarction; HF, heart failure.
(A) Kaplan-Meier survival curves of major adverse cardiac events (MACE) in patients with type 2 diabetes mellitus (T2DM) and coronary ischemia according to sodium-glucose cotransporter 2 inhibitors (SGLT2is). (B) Kaplan-Meier survival curves of all-cause mortality in patients with T2DM and coronary ischemia according to SGLT2is. (C) Kaplan-Meier survival curves of heart failure (HF) hospitalization in patients with T2DM and coronary ischemia according to SGLT2is.
Association between SGLT2 inhibitors and clinical outcomes
Univariate Cox regression analysis demonstrated that in addition to SGLT2 inhibitors, age, history of MI, hyperlipidemia, HF, stroke/TIA, chronic coronary syndrome and multivessel disease (MVD) were also significant predictor associated with MACE among the patients with T2DM and coronary ischemia (Table 2). Multivariate Cox regression analysis showed that the association between SGLT2 inhibitors and lower risk of MACE in patients with T2DM and coronary ischemia remained robust even after adjusting for additional confounding risk factors (HR, 0.442; 95% CI, 0.252 to 0.736; P=0.003) (Table 3). In both CR and ICR groups, SGLT2 inhibitors were independently associated with MACE, with HR of 0.500 (95% CI, 0.245 to 0.947; P=0.042) and 0.351 (95% CI, 0.124 to 0.844; P=0.030), respectively, after adjustment for confounding factors.
Univariate analysis for MACE
Multivariable analysis for MACE
Characteristics, prognosis, and subgroup analysis of the ICR population
In the entire cohort, a total of 187 individuals (27.9%) had ICR due to various reasons (Supplementary Table 2). Patients in the ICR group were more likely to have history of MI and stroke/TIA, as well as more frequent usage of statins, beta-blockers, and nitrates. A higher percentage of patients in the ICR group exhibited two- or three-vessel disease compared to those in the CR group. The ICR population was associated with a higher risk of MACE (22.5% vs. 12.0%; HR, 1.960; 95% CI, 1.310 to 2.908; P<0.001), all-cause mortality (18.7% vs. 11.2%; HR, 1.744; 95% CI, 1.130 to 2.655; P=0.010), non-fatal MI (7.5% vs. 2.1%; HR, 3.683; 95% CI, 1.648 to 8.543; P=0.002), and HF hospitalization (15.5% vs. 7.6%; HR, 2.113; 95% CI, 1.290 to 3.428; P=0.003) (Supplementary Table 3). In subgroup analysis in the ICR cohort, there was a consistent benefit of SGLT2 inhibitors in MACE across all subgroups (all of P for interaction >0.05) (Supplementary Fig. 2).
DISCUSSION
In this retrospective cohort study, we focused on the high-risk population of patients with T2DM complicated by coronary ischemia. This study is first to confirmed that the use of SGLT2 inhibitors was independently associated with a reduced risk of 3-year MACE and all-cause mortality, irrespective of the extent of revascularization. For patients with T2DM and coronary ischemia, it is recommended to administer SGLT2 inhibitors early to improve long-term prognosis.
Previous landmark cardiovascular outcomes trials have established the cardioprotective benefits of SGLT2 inhibitors in patients with T2DM and ASCVD, demonstrating reductions in HF hospitalization and CV mortality independent of glycemic control [11-14]. Compared to ASCVD patients, patients with CAD who have definite coronary functional ischemia have a higher risk of CV events and mortality. However, there are currently limited studies on SGLT2 inhibitors in populations with T2DM and CAD/coronary ischemia. Effect of empagliflozin on left ventricular mass in patients with type 2 diabetes mellitus and coronary artery disease (EMPA-HEART CardioLink-6), a RCT with a small sample size, enrolled 97 patients diagnosed with T2DM and definitive CAD. Magnetic resonance imaging demonstrated that, after a period of 6 months, empagliflozin was significantly associated with a decrease in the left ventricle mass indexed to body surface area [26]. Another RCT study [27] enrolled patients hospitalized for MI with a risk of HF, who were treated with empagliflozin and placebo in a 1:1 ratio. With a median follow-up time of 17.9 months. It was found that, compared to placebo, empagliflozin treatment did not lead to a significant reduction in the risk of first hospitalization for HF or death from any cause. Our study evaluated outcomes in patients with T2DM and functional coronary ischemia, a population characterized by complex coronary lesions and heightened vulnerability to adverse events. Despite higher baseline HbA1c and FBG levels among SGLT2 inhibitor users, our findings revealed a significant reduction in 3-year MACE (8.3% vs. 17.8%) and all-cause mortality (6.3% vs. 16.3%), underscoring the therapeutic potential of these agents beyond glucose-lowering effects. These results align with recent guideline recommendations [15,16] but extend their applicability to patients with residual ischemia, irrespective of revascularization completeness.
A key finding of this study is that the benefits of SGLT2 inhibitors persisted regardless of revascularization status, with similar hazard reductions in both CR and ICR subgroups. While CR remains a cornerstone in managing coronary ischemia, associated with lower residual plaque burden and improved long-term outcomes [7], achieving CR is often hindered by complex lesion morphology and comorbidities [4-6]. A meta-analysis of 35 studies, encompassing 89,883 patients who underwent revascularization, found that ICR was present in 56% of patients after PCI [28]. Previous randomized trials, including complete revascularization with multivessel PCI for myocardial infarction (COMPLETE), comparison between FFR guided revascularization versus conventional strategy in acute STEMI patients with MVD (Compare-Acute), and The Third Danish Study of Optimal Acute Treatment of Patients with ST-segment Elevation Myocardial Infarction: Primary PCI in Multivessel Disease (DANAMI 3-PRIMULTI), have found that the proportion of functional ICR ranges from 33.9% to 66.7% [29-31]. Despite focusing solely on the main coronary branches in this study, the ICR rate remained notably elevated at 27.9% due to the complexity of coronary artery lesions in patients with T2DM and the untreated ischemic lesions identified using angiographic FFR. It is noteworthy that despite SGLT2 inhibitors being able to improve the prognosis of individuals with ICR, CR should not be overlooked. It reduces ischemic burden and enhances event-free survival, as evidenced by the higher MACE rates in the ICR cohort (22.5% vs. 12.0%). Consistent with our findings, the Synergy between PCI with TAXUS and Cardiac Surgery (SYNTAX) trial observed higher rates of MACE at 3 years in patients with ICR after PCI compared to those without ICR (33.5% vs. 23.8%, P<0.001) [32]. Analyses of Non-ST elevation acute coronary syndrome (NSTE-ACS) populations have also found better clinical outcomes with CR than ICR [28,33-35]. The dual strategy of pursuing CR where feasible and initiating SGLT2 inhibitors early offers a pragmatic approach to mitigating residual risk in this high-risk population. Future studies should explore whether combining CR with SGLT2 inhibitors yields synergistic benefits, particularly in patients with MVD and advanced diabetes-related complications.
The cardioprotective mechanisms of SGLT2 inhibitors likely contribute to their observed benefits in coronary ischemia. Beyond modest glycemic improvements, these agents promote natriuresis, reduce blood pressure, and induce weight loss, collectively alleviating hemodynamic stress on the myocardium [36,37]. Emerging evidence suggests direct cardiac benefits, including enhanced myocardial energetics through ketone utilization, attenuation of oxidative stress, and suppression of inflammatory pathways such as nucleotide binding domain and leucine-rich repeat containing family pyrin domain containing 3 (NLRP3) inflammasome activation [38-40]. Additionally, SGLT2 inhibitors mitigate endothelial dysfunction and myocardial fibrosis, processes implicated in the progression of ischemic injury [41,42]. In patients with T2DM, whose coronary microvascular dysfunction and diffuse atherosclerosis exacerbate ischemia, these pleiotropic effects may improve myocardial perfusion and stabilize vulnerable plaques. The present study found that the effect of SGLT2 inhibitors on improving prognosis is primarily driven by reducing CV mortality and hospitalization for HF, rather than by lowering the risk of nonfatal MI. Additionally, the extent of revascularization did not affect the efficacy of SGLT2 inhibitors. These findings suggest that the mechanism by which SGLT2 inhibitors improve CV event endpoints is independent of the severity of coronary ischemia, but instead acts through more complex energy metabolism and comprehensive vascular effects.
The findings of this study hold significant implications for clinical practice in managing high-risk patients with T2DM and coronary ischemia. The consistent association of SGLT2 inhibitors with reduced 3-year MACE and all-cause mortality, irrespective of revascularization completeness, underscores their therapeutic value in this population. Clinicians should prioritize early initiation of SGLT2 inhibitors in patients with T2DM and documented coronary ischemia, even when anatomical or procedural challenges limit the achievement of CR. This is particularly relevant for individuals with residual ischemia due to ICR, who face elevated CV risk yet derive comparable benefits from SGLT2 inhibitors as those with CR. Furthermore, while CR remains a cornerstone for optimizing outcomes in coronary ischemia, the high prevalence of ICR (27.9%) in this cohort highlights the need for adjunctive pharmacological strategies to mitigate residual risk. Current guidelines recommend SGLT2 inhibitors for T2DM patients with established ASCVD. This study extends these recommendations to patients with functional coronary ischemia, including those with ICR, thereby broadening the target population for these agents. In summary, SGLT2 inhibitors represent a critical therapeutic tool for improving outcomes in T2DM patients with coronary ischemia, regardless of revascularization status. Their integration into multidisciplinary management protocols could substantially reduce the residual CV risk in this vulnerable population.
Our study has some limitations. First, this study only included empagliflozin and dapagliflozin, the SGLT2 inhibitors approved for clinical use in Hong Kong at that time. Further validation is needed to determine whether other types of SGLT2 inhibitors have the same effect. Additionally, our non-user group included a small number of patients who had previously used SGLT2 inhibitors (n=19, 4.1%). Even though the 6-month washout period we set exceeded pharmacological requirements, the lack of events in this subgroup made the comparison between patients who had never used SGLT2 inhibitors and those who had used them before inadequate. Second, this study only observed the impact of revascularization in the main coronary arteries and did not analyze the stenosis and revascularization of side branches. Therefore, there may be some cases which had branches ischemia but without PCI and was classified into the CR group. However, considering that coronary side branches have a relatively minor impact on prognosis and that side branch involvement is more common in T2DM patients, this study only evaluated the presence of ischemia based on the condition of the main coronary arteries. Third, angiographic images were collected in a retrospective manner, which could influence the feasibility and reliability of angiographic FFR analysis. Forth, the revascularization methods in this study only included PCI, the patients who underwent CABG were excluded. As an important method of revascularization, it is unknown whether SGLT2 inhibitors have the same effect on patients after CABG. Fifth, the primary reason for ICR was joint physician-patient decisions (47.8% of cases). This introduces potential selection bias, as the choice to defer revascularization may reflect unmeasured confounders such as frailty, comorbidities, or socioeconomic factors that influenced both clinical decisions and long-term outcomes. While we adjusted for established prognostic variables, residual confounding cannot be excluded. Nevertheless, the consistent benefit of SGLT2 inhibitors across both CR and ICR subgroups (P interaction >0.7) suggests that the observed treatment effects are robust to this limitation. Sixth, while our study demonstrated significant benefits of SGLT2 inhibitors in the ICR subgroup (HR, 0.341; P=0.023), we acknowledge that the relatively small number of MACE events in this subgroup limits the statistical power for nuanced analyses. This constraint necessitates caution in interpreting subgroup interactions and underscores the need for larger prospective studies to validate these findings. Nevertheless, the magnitude of risk reduction and consistent directionality across all endpoints suggest clinical relevance, particularly given the high unmet need in this understudied population.
In conclusion, the use of SGLT2 inhibitors is associated with a lower 3-year MACE and all-cause mortality in patients with T2DM combined with coronary ischemia, irrespective of the extent of revascularization. The findings of this study offer a new treatment regimen for high-risk patients with T2DM complicated by coronary ischemia, especially those with ICR.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2025.0200.
Coronary artery lesion and revascularization status of the population
Baseline characteristics of the study population based on the extent of revascularization
Primary and secondary outcomes by the extent of revascularization
(A) The distribution of coronary arteries without revascularization. (B) The reasons for coronary arteries not undergoing revascularization. LAD, left anterior descending branch; LCX, left circumflex coronary artery; RCA, right coronary artery; caFFR, coronary angiography-derived fractional flow reserve.
Subgroup analysis for the primary endpoint events in the incomplete revascularization cohort. SGLT2i, sodium-glucose cotransporter 2 inhibitor; HR, hazard ratio; CI, confidence interval; MI, myocardial infarction; HF, heart failure.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: H.X., Y.M.H., K.H.Y.
Acquisition, analysis, or interpretation of data: H.X., R.G., Q. R., J.H., J.Z., W.G., G.C., C.K.L.L.
Drafting the work or revising: all authors.
Final approval of the manuscript: all authors.
FUNDING
This work was supported by the National Natural Science Foundation of China (No. 82270400), the Natural Science Foundation of Guangdong Province (No. 2023A1515010731), and the Sanming Project of Medicine in Shenzhen (No. SZSM202411021).
ACKNOWLEDGMENTS
None