ABSTRACT
-
Background/Aims
Patients with concurrent type 2 diabetes mellitus (T2DM) and metabolic dysfunction-associated steatotic liver disease (MASLD) face elevated cardiovascular risks. However, optimal oral antidiabetic drug (OAD) selection for this population remains unclear.
-
Methods
Using the Korean National Health Information Database, we conducted a target trial emulation comparing cardiovascular outcomes among patients with T2DM and MASLD (defined by fatty liver index ≥30) who initiated sodium-glucose cotransporter 2 (SGLT2) inhibitors, thiazolidinediones, dipeptidyl peptidase-4 (DPP-4) inhibitors, or sulfonylureas with metformin. The primary outcome was major adverse cardiovascular events (MACE), including cardiovascular mortality, nonfatal myocardial infarction, and nonfatal stroke.
-
Results
Among 71,071 patients (331,726 person-years), SGLT2 inhibitor users experienced a significantly lower MACE risk compared to sulfonylurea users (adjusted subdistribution hazard ratio [aSHR], 0.44; 95% confidence interval [CI], 0.31–0.62). SGLT2 inhibitors also demonstrated a lower MACE risk compared to thiazolidinediones (aSHR, 0.61; 95% CI, 0.39–0.96) and DPP-4 inhibitors (aSHR, 0.59; 95% CI, 0.42–0.83). Cardiovascular mortality risk was notably reduced with SGLT2 inhibitors compared to sulfonylureas (aSHR, 0.13; 95% CI, 0.03–0.50), thiazolidinediones (aSHR, 0.19; 95% CI, 0.04–0.86), and DPP-4 inhibitors (aSHR, 0.22; 95% CI, 0.06–0.84). Mediation analysis revealed that MASLD regression accounted for 8.7% of the total cardiovascular benefit when comparing SGLT2 inhibitors to sulfonylureas.
-
Conclusions
In patients with concurrent T2DM and MASLD, SGLT2 inhibitors demonstrated better cardiovascular outcomes compared to other OADs. These findings suggest that SGLT2 inhibitors may be the preferred OAD choice for cardiovascular risk reduction in this high-risk population.
-
Keywords: Non-alcoholic fatty liver disease; Diabetes mellitus; Hypoglycemic agents; Cardiovascular diseases; Cerebrovascular disorders
Study Highlights
• This nationwide target trial emulation demonstrated that sodium-glucose cotransporter 2 (SGLT2) inhibitors significantly lowered the risk of major adverse cardiovascular events in patients with type 2 diabetes mellitus and metabolic dysfunction-associated steatotic liver disease.
• Compared with sulfonylureas, thiazolidinediones, and dipeptidyl peptidase-4 inhibitors, SGLT2 inhibitors showed consistent and robust cardiovascular benefits, particularly in reducing cardiovascular mortality.
• Mediation analysis suggested that MASLD regression partially contributed to these cardiovascular advantages, highlighting both hepatic and extra-hepatic mechanisms of benefit.
• These findings support preferential use of SGLT2 inhibitors as the optimal oral antidiabetic therapy in this high-risk population.
Graphical Abstract
INTRODUCTION
Cardiovascular disease remains the leading cause of mortality worldwide, accounting for approximately 32% of all global deaths [
1]. The concept of metabolic dysfunction-associated steatotic liver disease (MASLD) has recently evolved from nonalcoholic fatty liver disease, specifically recognizing its intrinsic relationship with metabolic dysfunction and cardiovascular risk [
2]. This paradigm shift reflects the understanding that MASLD, affecting approximately 38% of the global population with increasing prevalence, represents not just a liver disease but a systemic metabolic disorder with significant cardiovascular implications [
3,
4].
Type 2 diabetes mellitus (T2DM) frequently coexists with MASLD, with 65% of patients with T2DM having concurrent MASLD, and both conditions additively amplify cardiovascular risk through shared pathophysiological pathways including insulin resistance, systemic inflammation, and atherogenic dyslipidemia [
3,
5-
7]. Several classes of oral antidiabetic drugs (OADs) have demonstrated varying levels of beneficial effects beyond glycemic control, with sodium-glucose cotransporter 2 (SGLT2) inhibitors and thiazolidinediones showing significant improvements in cardiovascular health as well as MASLD regression [
8-
10]. In particular, accumulating evidence from clinical studies suggests that SGLT2 inhibitors may exert beneficial effects on hepatic steatosis and metabolic dysfunction [
11-
13]. Dipeptidyl peptidase-4 (DPP-4) inhibitors, while commonly used as a second-line agent in combination with metformin for T2DM, have shown limited benefits in MASLD regression compared to SGLT2 inhibitors and thiazolidinediones, and lack robust evidence for cardiovascular outcome improvement [
10,
14].
The coexistence of T2DM and MASLD significantly increases the incidence of cardiovascular events compared to either condition alone [
15]. This synergistic increase in cardiovascular risk underscores the urgent need for targeted therapeutic strategies in this population. Currently, there are no specific recommendations for OADs tailored to patients with concurrent T2DM and MASLD, despite their high prevalence and elevated cardiovascular risk. To address this gap in evidence, we utilized a nationwide cohort from the Korean National Health Information Database, encompassing over 43 million adults, to compare the effectiveness of SGLT2 inhibitors, thiazolidinediones, and DPP-4 inhibitors in preventing cardiovascular events among patients with T2DM and MASLD. Our aim was to provide robust, real-world evidence to guide the optimization of treatment strategies for this high-risk group.
MATERIALS AND METHODS
Data source
The National Health Insurance Service (NHIS) provides mandatory universal health coverage and biennial health screenings in South Korea. The NHIS database includes clinical records (e.g., diagnoses, medications, procedures), health examination results (e.g., anthropometric measurements, lifestyle factors, blood tests), and mortality data linked with Statistics Korea [
10]. This database’s reliability has been validated through numerous epidemiological studies [
10]. The institutional review board of Seoul Metropolitan Government Seoul National University Boramae Medical Center approved this study (IRB No.: 07-2023-31).
We designed this study to emulate a target trial comparing OADs in patients with concurrent T2DM and MASLD, following Hernán and Robins’s framework (
Supplementary Table 1) [
16]. We included participants aged ≥18 years with hepatic steatosis (fatty liver index [FLI] ≥30) [
7,
17]. Since OADs other than metformin are rarely prescribed as monotherapy, we included T2DM participants who began treatment with SGLT2 inhibitors, thiazolidinediones, DPP-4 inhibitors, or sulfonylureas as part of a dual therapy regimen with metformin [
18]. This active comparator design also helps ensure comparable treatment groups and reduces unmeasured confounding by selecting patients at similar stages of diabetes progression [
19]. Participants were required to maintain at least 80% adherence over 90 consecutive days between November 2014 and March 2017 [
20]. The index date was set 90 days after starting the additional OADs to minimize immortal time bias [
16].
Participants were excluded if they had significant alcohol consumption (≥210 g/week for males and ≥140 g/week for females) [
2]. To ensure a new-user design that enables evaluation of treatment initiation effects and minimizes confounding from prior treatments, we excluded individuals who had used any OADs between January 2012 and October 2014 [
21]. To enhance comparability and ensure similar baseline glycemic control among participants, individuals who had used ≥3 OAD classes prior to the index date were excluded. To focus solely on the effects of OADs, individuals with a history of using injectable antidiabetic medications for ≥90 days prior to the index date were excluded. Participants with liver diseases other than MASLD (viral/autoimmune/toxic hepatitis, Wilson’s disease, hemochromatosis, primary biliary cholangitis, or Budd-Chiari syndrome) were excluded. Finally, individuals with a history of myocardial infarction or ischemic stroke before the index date were excluded.
Patients were categorized into four groups based on the class of OADs they predominantly used for 90 consecutive days: SGLT2 inhibitors, thiazolidinediones, DPP-4 inhibitors, or sulfonylureas. The inclusion and exclusion criteria are depicted in
Figure 1. Summarized study design and detailed definitions can be found in Supplement (
Supplementary Fig. 1 and
Supplementary Table 2).
The main outcome was major adverse cardiovascular events (MACE), which included cardiovascular-related mortality, nonfatal myocardial infarction, or nonfatal stroke [
22]. Additionally, each component was analyzed separately. Myocardial infarction was defined by hospitalization with the International Classification of Diseases 10th Revision (ICD-10) codes I21/I22, stroke by hospitalization with ICD-10 codes I63/I64 accompanied by brain CT/MRI, and cardiovascular-related mortality by death with ICD-10 “I” codes.
Inverse probability of treatment weighting (IPTW) was utilized to balance baseline characteristics across OAD classes. The weights were calculated using baseline variables: continuous factors such as age, waist circumference, body mass index (BMI), Charlson comorbidity index, laboratory examination results (total cholesterol, triglycerides, glucose, aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transferase, high-density lipoprotein [HDL] cholesterol, low-density lipoprotein [LDL] cholesterol, and creatinine), and calendar time (months since November 2014); and categorical factors such as sex, alcohol consumption (non-drinkers or light drinkers [defined as <210 g/week for males, <140 g/week for females]), smoking status (non-smokers, ex-smokers, or current smokers), moderateto-vigorous physical activity (MVPA) frequency (0, 1-2, 3-4, or ≥5 times/week), hypertension, family history (hypertension, stroke, and heart disease), and use of antithrombotic or lipid-lowering medications [
10].
For group comparisons, we applied multinomial logistic regression for multiple groups and logistic regression with baseline covariate adjustment for pairwise analyses, with weight truncation at the 99.9th percentile. Cumulative incidences were compared using Gray’s test. The Fine-Gray competing risk model was employed to calculate adjusted subdistribution hazard ratios (aSHRs) with 95% confidence intervals (95% CIs), calculated using robust standard error estimation for MACE and its components [
19]. Competing events were non-cardiovascular mortality and liver transplantation for MACE, and all-cause mortality and liver transplantation for myocardial infarction and stroke. Proportional hazards assumption was verified using Gramsch-Therneau testing [
23].
Sensitivity analyses were conducted using different drug adherence thresholds (80% or 100%) and exposure durations (90, 180, or 365 consecutive days). As additional sensitivity analyses, patients with glucagon-like peptide-1 (GLP-1) receptor agonist exposure were included (≥90 consecutive days and ≥1 day), with GLP-1 receptor agonist use incorporated as an additional covariate. Subgroup analyses were performed by age, sex, waist circumference, BMI, alcohol intake, smoking status, MVPA, hypertension, use of lipid-lowering and antithrombotic agents, level of triglycerides, HDL-cholesterol, LDL-cholesterol, and FLI score. The mediating effects of MASLD regression (FLI reduction to <30) were evaluated using Aalen’s additive hazards models as a time-varying mediator [
24]. Analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA) and R version 4.0.3 (The R Foundation, Vienna, Austria) with two-sided
P<0.05 considered significant.
RESULTS
Baseline characteristics of study population
Among 118,437 adults who initiated second-line OAD with metformin between November 2014 and March 2017, 71,071 individuals with concurrent T2DM and MASLD remained after applying exclusion criteria. During 331,726 person-years (median follow-up, 4.79 years), we identified 4,365 SGLT2 inhibitor, 2,033 thiazolidinedione, 51,348 DPP-4 inhibitor, and 13,325 sulfonylurea users (
Fig. 1 and
Table 1). IPTW balanced baseline characteristics among groups (
Supplementary Table 3 and
Supplementary Fig. 2), with weight distribution shown in
Supplementary Figure 3.
MACE occurred in 1,699 patients, with varying frequencies across treatment groups: 55 cases in SGLT2 inhibitor users, 46 in thiazolidinedione users, 1,152 in DPP-4 inhibitor users, and 446 in sulfonylurea users. The corresponding incidence rates per 100,000 person-years were 278.8, 474.2, 482.1, and 704.2, respectively. Compared to sulfonylurea users, both SGLT2 inhibitor users (aSHR, 0.44; 95% CI, 0.31–0.62) and DPP-4 inhibitor users (aSHR, 0.80; 95% CI, 0.72–0.90) showed a significantly lower MACE risk (
Table 2 and
Fig. 2A). Notably, among the OAD classes, SGLT2 inhibitors were associated with a lower MACE risk, showing a significantly reduced risk when compared to both thiazolidinediones (aSHR, 0.61; 95% CI, 0.39–0.96) and DPP-4 inhibitors (aSHR, 0.59; 95% CI, 0.42–0.96) in pairwise analyses (
Table 2 and
Fig. 2A).
Cardiovascular death occurred in 258 patients (3 SGLT2 inhibitor, 7 thiazolidinedione, 165 DPP-4 inhibitor, and 83 sulfonylurea users; incidence rates: 15.1, 71.5, 68.5, and 129.4 per 100,000 person-years, respectively). SGLT2 inhibitors showed lower risk compared with sulfonylureas (aSHR, 0.13; 95% CI, 0.03–0.50), thiazolidinediones (aSHR, 0.19; 95% CI, 0.04–0.86), and DPP-4 inhibitors (aSHR, 0.22; 95% CI, 0.06–0.84) (
Fig. 2B and
Supplementary Table 4).
Myocardial infarction occurred in 564 patients (21 SGLT2 inhibitor, 15 thiazolidinedione, 409 DPP-4 inhibitor, and 119 sulfonylurea users; incidence rates: 106.2, 153.6, 170.1, and 185.9, respectively). SGLT2 inhibitor use suggested a protective trend against myocardial infarction compared to sulfonylureas (aSHR, 0.61; 95% CI, 0.36–1.03), which did not reach statistical significance (
Fig. 2C and
Supplementary Table 5).
Among 986 stroke events (32 SGLT2 inhibitor, 30 thiazolidinedione, 647 DPP-4 inhibitor, and 277 sulfonylurea users; incidence rates: 161.8, 307.7, 269.4, and 434.3, respectively), SGLT2 inhibitors demonstrated a lower risk compared with sulfonylureas (aSHR, 0.41; 95% CI, 0.25–0.66), thiazolidinediones (aSHR, 0.53; 95% CI, 0.29–0.97), and DPP-4 inhibitors (aSHR, 0.59; 95% CI, 0.37–0.95) (
Fig. 2D and
Supplementary Table 6).
Sensitivity analyses demonstrated that the cardiovascular benefits of SGLT2 inhibitors were largely maintained across different drug exposure periods and adherence rates, particularly for MACE and stroke outcomes, although the magnitude and statistical significance of the effects varied (
Supplementary Tables 7 to
26 and
Supplementary Figs. 4 to
7). Regarding GLP-1 receptor agonist exposure, only 10 patients had prolonged GLP-1 receptor agonist use (≥90 consecutive days), all of whom were excluded due to prior use of three or more OAD classes. Subsequent sensitivity analyses including patients with any GLP-1 receptor agonist exposure (≥1 day) included a total of 22 additional patients (4 SGLT2 inhibitor users, 1 thiazolidinedione user, 8 DPP-4 inhibitor users, and 9 sulfonylurea users); neither additional MACE nor component events occurred among these patients, and hazard ratios were consistent with those observed in the primary analysis.
Subgroup analyses showed SGLT2 inhibitors’ cardiovascular benefits remained consistent across most subgroups (
Supplementary Figs. 8 to
11). Specifically, SGLT2 inhibitors consistently demonstrated a lower MACE risk com-pared to other OADs regardless of sex and baseline FLI category (FLI ≥60 and 30–59) (all
P for interaction >0.05). The cardiovascular benefits of SGLT2 inhibitors were more pronounced in patients with higher HDL-cholesterol levels (
P for interaction=0.03) and in non-drinkers compared to light drinkers (
P for interaction=0.017).
In causal mediation analysis, MASLD regression (FLI <30) partially mediated the cardiovascular benefits of SGLT2 inhibitors, accounting for 8.7% of the total effect compared with sulfonylureas (rate difference −1.17; 95% CI, −1.54 to −0.75), 6.8% vs. DPP-4 inhibitors (−0.57; 95% CI, −0.82 to −0.34), and 2.4% vs. thiazolidinediones (−0.19; 95% CI, −0.60 to −0.02), suggesting MASLD improvement as one of several protective mechanisms (
Table 3 and
Supplementary Fig. 12).
DISCUSSION
In this nationwide cohort study of patients with concurrent T2DM and MASLD, we found that SGLT2 inhibitors were associated with significantly lower risks of cardiovascular events compared to other OAD classes. Specifically, SGLT2 inhibitor users demonstrated a 56% lower risk of MACE compared to sulfonylurea users, a 39% lower risk compared to thiazolidinedione users, and a 41% lower risk compared to DPP-4 inhibitor users. The protective effects were particularly pronounced for cardiovascular mortality, showing risk reductions of 87%, 81%, and 78% compared to sulfonylureas, thiazolidinediones, and DPP-4 inhibitors, respectively. These findings were consistent across various sensitivity analyses and subgroups, suggesting the robustness of the superior cardiovascular benefits of SGLT2 inhibitors over other OADs in this high-risk population.
Our results align with and extend previous findings from several landmark randomized controlled trials. The EMPAREG OUTCOME, CANVAS, and DECLARE-TIMI trials demonstrated the cardiovascular benefits of SGLT2 inhibitors in patients with T2DM [
8,
25,
26]. However, these trials did not specifically focus on patients with concurrent MASLD and T2DM. Interestingly, a comprehensive meta-analysis of major cardiovascular outcome trials showed that SGLT2 inhibitors reduced MACE and cardiovascular death only in patients with established atherosclerotic cardiovascular disease but not in those without [
27]. Our study addresses this gap by providing real-world evidence. Our findings that SGLT2 inhibitors significantly reduce cardiovascular risk in patients with concurrent T2DM and MASLD suggest that MASLD may confer a level of cardiovascular risk comparable to that of established atherosclerotic cardiovascular disease.
Current clinical practice guidelines for diabetes management do not provide specific recommendations for OAD preferences in concurrent T2DM and MASLD, despite the high prevalence and significant cardiovascular risk of this population [
28]. Our study addresses this evidence gap by providing comprehensive head-to-head comparisons of all commonly used OADs. Notably, the magnitude of the cardiovascular benefits of SGLT2 inhibitors surpassed those observed with both thiazolidinediones and DPP-4 inhibitors. These findings are particularly noteworthy given that thiazolidinediones have been associated with increased risks of congestive heart failure and acute myocardial infarction, while DPP-4 inhibitors have shown neutral effects on cardiovascular outcomes [
14,
29,
30]. We focused on oral medications because they typically show better adherence and cost-effectiveness compared to injectable agents such as GLP-1 receptor agonists, making them a more practical first-line treatment option in real-world settings [
31,
32].
The observed cardiovascular benefits of SGLT2 inhibitors in patients with T2DM and MASLD may be attributed to several mechanisms. First, SGLT2 inhibitors have been shown to improve various cardiometabolic risk factors, including glycemic control, body weight, and blood pressure [
33]. Second, these agents have demonstrated beneficial effects on cardiac function, including reduced left ventricular mass and improved diastolic function [
34]. Third, SGLT2 inhibitors may exert direct cardioprotective effects through mechanisms such as improved myocardial energetics, reduced oxidative stress, and attenuated inflammation [
35]. Fourth, at the vascular level, SGLT2 inhibitors improve endothelial function by increasing nitric oxide availability, alleviating microvascular dysfunction, and decreasing arterial stiffness and vascular resistance, leading to better cardiac and cerebral perfusion [
35-
37]. Additionally, SGLT2 inhibitors reduce uric acid levels, which may contribute to their cardiovascular protection by decreasing systemic inflammation and improving endothelial function [
35]. SGLT2 inhibitors also have beneficial metabolic effects by reducing visceral adiposity and improving insulin sensitivity, potentially mitigating cardiovascular risk in patients with MASLD [
38]. Finally, SGLT2 inhibitors have been shown to reduce the accumulation of epicardial adipose tissue and improve overall adipose tissue function, which may contribute to their beneficial effects on lowering MACE [
39].
Among MACE components, SGLT2 inhibitors showed significant reduction in cardiovascular mortality, but their effect on myocardial infarction did not reach statistical significance. Although the limited statistical power could be addressed by extending the follow-up duration or participant numbers, several potential mechanisms may explain these discrepant findings. First, SGLT2 inhibitors reduce inflammation and oxidative stress in cardiac and cerebral vessels, which may help prevent subsequent atherothrombotic events [
35,
36,
40]. Second, SGLT2 inhibitors optimize cardiac energy metabolism by increasing ketone body production and utilization, providing an efficient fuel source for the failing heart [
41,
42]. Third, they demonstrate significant natriuretic and osmotic diuretic effects, leading to reduced preload and afterload, which can help protect the heart after myocardial infarction by decreasing cardiac workload [
43]. Fourth, SGLT2 inhibitors attenuate cardiac fibrosis and improve diastolic function [
44]. Finally, SGLT2 inhibitors provide protection against adverse cardiac remodeling and progression of heart failure [
45]. Altogether, these multifaceted cardioprotective mechanisms suggest that SGLT2 inhibitors may be more effective in preventing recurrent myocardial infarction and subsequent cardiovascular death.
Interestingly, our mediation analysis revealed that MASLD regression accounted for 8.7% of the total cardiovascular benefit when comparing SGLT2 inhibitors to sulfonylureas. This suggests that while improvements in liver health may contribute to the cardiovascular benefits of SGLT2 inhibitors, other direct cardiovascular protective mechanisms likely play more substantial roles. Observational and experimental studies have suggested that SGLT2 inhibition may influence hepatic metabolic and inflammatory processes, supporting biological plausibility for potential liver-directed effects beyond glycemic control [
46-
48]. In addition, emerging pathological evidence suggests that SGLT2 is expressed not only in the kidney but also in extrarenal tissues, including the liver, heart, and brain [
49]. Future studies are needed to elucidate the full spectrum of mechanisms underlying the cardiovascular benefits of SGLT2 inhibitors in this population, although the relationship between MASLD regression and cardiovascular outcomes deserves special attention.
Our study has several strengths. First, we utilized a large, nationwide cohort with comprehensive health information, allowing for robust analyses and generalizability of findings. Second, we employed rigorous statistical methods, including IPTW and competing risk models, to minimize confounding and bias. Third, we implemented a rigorous active-comparator and new-user design to bolster the validity of our conclusions, restricting enrollment to individuals on dual OAD therapy and excluding those on injectable agents or triple (or more) antidiabetic drug regimens, thereby addressing potential biases inherent in the retrospective nature of our study [
19,
21]. Fourth, the robustness of our results was further validated through extensive sensitivity analyses across various drug exposure durations, adherence thresholds, and patient subgroups. Finally, our mediation analysis provided valuable insights into the mechanisms through which SGLT2 inhibitors confer cardiovascular risk reduction, quantifying the contribution of MASLD regression to these benefits.
However, our study also has limitations. First, although we emulated a target trial using an active-comparator, new-user design and extensive covariate adjustment, this study remains observational; therefore, residual confounding from unmeasured factors, such as dietary patterns, medication-taking behaviors, or other health-related behaviors not fully captured in claims data, cannot be completely ruled out. Second, the study population was limited to Korean adults, potentially limiting generalizability to other ethnicities. As MASLD pathophysiology, genetic background, and responses to antidiabetic therapies may vary across populations, further studies in diverse ethnic populations would help establish the generalizability of our findings. Third, we used the FLI as a surrogate marker for MASLD rather than liver biopsy, which may introduce potential misclassification bias by failing to capture all cases of MASLD. However, the FLI has been extensively validated and widely adopted in population-based epidemiological studies, demonstrating robust correlations with both imaging-based and histological assessment of hepatic steatosis [
10]. Fourth, we did not have information on medication dosages or the specific SGLT2 inhibitors used, which could potentially influence outcomes. Fifth, we did not assess the impact of injectable antidiabetic medications such as insulin or GLP-1 receptor agonists. However, sensitivity analyses incorporating patients with GLP-1 receptor agonist exposure yielded results consistent with the primary findings, suggesting that exclusion of GLP-1 receptor agonist users did not materially bias the results. Finally, glycated hemoglobin data were not available in our database, limiting our ability to fully adjust for glycemic control. To mitigate glycemic heterogeneity, we employed two methodological approaches: utilizing fasting glucose levels for adjustment and selecting patients exclusively on dual OAD therapy while excluding those on triple therapy or insulin, thereby ensuring a more consistent glycemic profile. The Korean Diabetes Association guidelines recommend dual OAD therapy for patients with HbA1c levels between 7.5% and 9.0%, which aligns with the glycemic control range of our study population [
28].
In conclusion, our findings provide compelling evidence that SGLT2 inhibitors are related to better cardiovascular outcomes than other OADs in patients with T2DM and MASLD. These data indicate that SGLT2 inhibitors may be the best OAD option for lowering cardiovascular risk in this high-risk population. Future research, including randomized controlled trials specifically targeting patients with T2DM and MASLD, is needed to confirm these findings, better understand the mechanisms underlying the cardiovascular benefits of SGLT2 inhibitors in this population, and identify patient subgroups who may benefit the most from SGLT2 inhibitors. Longer-term studies are also required to determine the stability of these cardiovascular advantages and their impact on non-cardiac clinical outcomes. The strong data supporting the preferential use of SGLT2 inhibitors in patients with concurrent T2DM and MASLD, particularly for cardiovascular risk reduction, is a big step forward in our understanding of effective treatment regimens for this high-risk population.
FOOTNOTES
-
Access to Data and Data Analysis
HJ, YK, WL, and WK had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
-
Authors’ contributions
Heejoon Jang: Conceptualization, Methodology, Analysis, Data curation, Writing - Original draft, Visualization, Funding acquisition. Yeonjin Kim: Analysis, Data curation, Writing - Review & Editing, Visualization. Yoo Kyoung Lim: Data curation, Writing - Review & Editing. Dong Hyeon Lee: Data curation, Writing - Review & Editing. Sae Kyung Joo: Data curation, Writing - Review & Editing. Bo Kyung Koo: Data curation, Writing - Review & Editing. Gi-Ae Kim: Writing - Review & Editing. Woojoo Lee: Methodology, Analysis, Data curation, Writing - Review & Editing. Stefano Romeo: Writing - Review & Editing. Won Kim: Conceptualization, Methodology, Analysis, Data curation, Writing - Review & Editing, Supervision, Funding acquisition.
-
Acknowledgements
This research was supported by grants from the National Research Foundation (NRF) of Korea (2021R1A2C2005820, RS-2021-NR056442, RS-2022-NR067269, RS-2023-00223831, RS-2024-00440883, and RS-2025-25458964), Basic Science Research Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education (RS-2024-00408240), the Research Supporting Program of the Korean Association for the Study of the Liver (KASL2024-01), and a multidisciplinary research grant-in-aid from the Seoul Metropolitan Government Seoul National University (SMG-SNU) Boramae Medical Center (04-2024-0028).
-
Conflicts of Interest
The authors have no conflicts to disclose.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 1.
Study design. OAD, oral antidiabetic drug. aAll individuals who had used OADs between January 2012 and September 2014 were excluded from analysis to ensure that all participants were using OADs for the first time during the inclusion period.
Supplementary Figure 2.
Absolute standardized mean differences in baseline characteristics before and after being adjusted by the inverse probability of treatment weighting method. ALT, alanine aminotransferase; AST, aspartate aminotransferase; DPP-4, dipeptidyl peptidase-4; γGT, gamma-glutamyl transferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MVPA, moderate-to-vigorous physical activity; SGLT2, sodium-glucose cotransporter 2. Light alcohol intake was defined as <210 g/week for males and <140 g/week for females.
Supplementary Figure 3.
Weights calculated by inverse probability of treatment weighting method: (A) weights before truncation and (B) weight truncation at the 99.9th percentile.
Supplementary Figure 4.
Major adverse cardiovascular events (MACE) according to oral antidiabetic drug classes in various durations of drug use: (A) throughout 90 consecutive days; (B) 80% of 180 consecutive days; (C) throughout 180 consecutive days; (D) 80% of 365 consecutive days; (E) throughout 365 consecutive days. DPP-4, dipeptidyl peptidase-4; SGLT2, sodium-glucose cotransporter 2.
Supplementary Figure 5.
Cardiovascular-related mortality according to oral antidiabetic drug classes in various durations of drug use: (A) throughout 90 consecutive days; (B) 80% of 180 consecutive days; (C) throughout 180 consecutive days; (D) 80% of 365 consecutive days; (E) throughout 365 consecutive days. DPP-4, dipeptidyl peptidase-4; SGLT2, sodium-glucose cotransporter 2.
Supplementary Figure 6.
Myocardial infarction according to oral antidiabetic drug classes in various durations of drug use: (A) throughout 90 consecutive days; (B) 80% of 180 consecutive days; (C) throughout 180 consecutive days; (D) 80% of 365 consecutive days; (E) throughout 365 consecutive days. DPP-4, dipeptidyl peptidase-4; SGLT2, sodium-glucose cotransporter 2.
Supplementary Figure 7.
Stroke according to oral antidiabetic drug classes in various durations of drug use: (A) throughout 90 consecutive days; (B) 80% of 180 consecutive days; (C) throughout 180 consecutive days; (D) 80% of 365 consecutive days; (E) throughout 365 consecutive days. DPP-4, dipeptidyl peptidase-4; MACE, major adverse cardiovascular events; SGLT2, sodium-glucose cotransporter 2.
Supplementary Figure 8.
Major adverse cardiovascular events according to oral antidiabetic drug classes in various subgroups. aSHR, adjusted subdistribution hazard ratio; CI, confidence interval; DPP-4, dipeptidyl peptidase-4; FLI, fatty liver index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MVPA, moderate-to-vigorous physical activity; SGLT2, sodium-glucose cotransporter 2. Light drinkers are defined as those with an alcohol intake of less than 210 g/week for males and less than 140 g/week for females.
Supplementary Figure 9.
Cardiovascular-related mortality according to oral antidiabetic drug classes in various subgroups. aSHR, adjusted subdistribution hazard ratio; CI, confidence interval; DPP-4, dipeptidyl peptidase-4; FLI, fatty liver index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MVPA, moderate-to-vigorous physical activity; NA, not applicable; SGLT2, sodium-glucose cotransporter 2. Light drinkers are defined as those with an alcohol intake of less than 210 g/week for males and less than 140 g/week for females.
Supplementary Figure 10.
Myocardial infarction according to oral antidiabetic drug classes in various subgroups. aSHR, adjusted subdistribution hazard ratio; CI, confidence interval; DPP-4, dipeptidyl peptidase-4; FLI, fatty liver index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MVPA, moderate-to-vigorous physical activity; NA, not applicable; SGLT2, sodium-glucose cotransporter 2. Light drinkers are defined as those with an alcohol intake of less than 210 g/week for males and less than 140 g/week for females.
Supplementary Figure 11.
Stroke according to oral antidiabetic drug classes in various subgroups. aSHR, adjusted subdistribution hazard ratio; CI, confidence interval; DPP-4, dipeptidyl peptidase-4; FLI, fatty liver index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MVPA, moderate-to-vigorous physical activity; SGLT2, sodium-glucose cotransporter 2. Light drinkers are defined as those with an alcohol intake of less than 210 g/week for males and less than 140 g/week for females.
Supplementary Figure 12.
Cumulative mediation effects of MASLD regression on the association between major cardiovascular events and oral diabetic drug classes: (A) SGLT2 inhibitors versus thiazolidinediones, (B) SGLT2 inhibitors versus DPP-4 inhibitors, (C) SGLT2 inhibitors versus sulfonylureas. DPP-4, dipeptidyl peptidase-4; MASLD, metabolic dysfunction-associated steatotic liver disease; SGLT2, sodium-glucose cotransporter 2.
Figure 1.A study attrition diagram for inclusion of study population. DPP-4, dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1; SGLT2, sodium-glucose cotransporter 2; T2DM, type 2 diabetes mellitus.
Figure 2.Cumulative incidence rate according to oral antidiabetic drug classes: (A) Major adverse cardiovascular events (MACE), (B) Cardiovascular-related mortality, (C) Myocardial infarction, (D) Stroke. DPP-4, dipeptidyl peptidase-4; SGLT2, sodium-glucose cotransporter 2.
Table 1.Unadjusted baseline characteristics
Table 1.
|
Characteristic |
SGLT2 inhibitors (n=4,365) |
Thiazolidinediones (n=2,033) |
DPP–4 inhibitors (n=51,348) |
Sulfonylureas (n=13,325) |
|
Age, year |
50 (43–57) |
54 (46–62) |
54 (46–62) |
55 (47–64) |
|
Sex |
|
|
|
|
|
Female |
1,853 (42.5) |
648 (31.9) |
18,151 (35.3) |
4,750 (35.6) |
|
Male |
2,512 (57.5) |
1,385 (68.1) |
33,197 (64.7) |
8,575 (64.4) |
|
Waist circumference, cm |
92 (86–99) |
90 (85–96) |
90 (85–95) |
90 (85–95) |
|
Body mass index*, kg/m2
|
28.6 (26.3–31.5) |
27.1 (25.2–29.7) |
26.9 (25–29.2) |
26.8 (25–29.2) |
|
Alcohol intake |
|
|
|
|
|
No intake |
2,017 (46.2) |
986 (48.5) |
27,324 (53.2) |
7,415 (55.6) |
|
Light intake†
|
2,348 (53.8) |
1,047 (51.5) |
24,024 (46.8) |
5,910 (44.4) |
|
Smoking status |
|
|
|
|
|
Nonsmoker |
2,319 (53.1) |
934 (45.9) |
24,932 (48.6) |
6,569 (49.3) |
|
Former smoker |
851 (19.5) |
463 (22.8) |
11,265 (21.9) |
2,707 (20.3) |
|
Current smoker |
1,195 (27.4) |
636 (31.3) |
15,151 (29.5) |
4,049 (30.4) |
|
MVPA |
|
|
|
|
|
0 times/week |
2,017 (46.2) |
1,012 (49.8) |
25,709 (50.1) |
7,081 (53.1) |
|
1–2 times/week |
1,337 (30.6) |
579 (28.5) |
14,428 (28.1) |
3,395 (25.5) |
|
3–4 times/week |
664 (15.2) |
262 (12.9) |
6,964 (13.6) |
1,745 (13.1) |
|
≥5 times/week |
347 (7.9) |
180 (8.9) |
4,247 (8.3) |
1,104 (8.3) |
|
Hypertension |
|
|
|
|
|
No |
1,637 (37.5) |
747 (36.7) |
19,843 (38.6) |
4,723 (35.4) |
|
Yes |
2,728 (62.5) |
1,286 (63.3) |
31,505 (61.4) |
8,602 (64.6) |
|
Family history of hypertension |
|
|
|
|
|
No |
3,420 (78.4) |
1,677 (82.5) |
42,462 (82.7) |
11,159 (83.7) |
|
Yes |
945 (21.6) |
356 (17.5) |
8,886 (17.3) |
2,166 (16.3) |
|
Family history of stroke |
|
|
|
|
|
No |
4,026 (92.2) |
1,891 (93) |
47,633 (92.8) |
12,506 (93.9) |
|
Yes |
339 (7.8) |
142 (7.0) |
3,715 (7.2) |
819 (6.1) |
|
Family history of heart disease |
|
|
|
|
|
No |
4,146 (95.0) |
1,947 (95.8) |
49,168 (95.8) |
12,835 (96.3) |
|
Yes |
219 (5.0) |
86 (4.2) |
2,180 (4.2) |
490 (3.7) |
|
Comorbidity index |
2 (1–3) |
2 (1–3) |
2 (1–3) |
2 (1–3) |
|
Hyperlipidemia agent users |
|
|
|
|
|
Nonuser |
1,445 (33.1) |
716 (35.2) |
21,420 (41.7) |
6,585 (49.4) |
|
User |
2,920 (66.9) |
1,317 (64.8) |
29,928 (58.3) |
6,740 (50.6) |
|
Antithrombotic agent users |
|
|
|
|
|
Nonuser |
3,598 (82.4) |
1,618 (79.6) |
42,216 (82.2) |
10,828 (81.3) |
|
User |
767 (17.6) |
415 (20.4) |
9,132 (17.8) |
2,497 (18.7) |
|
Laboratory examination results |
|
|
|
|
|
Total cholesterol, mg/dL |
210 (180–241) |
209 (180–239) |
210 (182–241) |
211 (183–242) |
|
Triglycerides, mg/dL |
182 (132–262) |
184 (133–261) |
186 (134–266) |
189 (136–271) |
|
Glucose, mg/dL |
140 (119–178) |
139 (119–175) |
144 (121–184) |
146 (119–198) |
|
AST, U/L |
30 (22–44) |
29 (22–40) |
28 (22–40) |
28 (21–39) |
|
ALT, U/L |
39 (26–63) |
35 (24–55) |
35 (24–55) |
33 (23–51) |
|
γGT, U/L |
50 (33–80) |
50 (33–80) |
50 (33–80) |
50 (33–80) |
|
HDL-cholesterol, mg/dL |
46 (40–54) |
46 (40–55) |
46 (40–54) |
46 (40–54) |
|
LDL-cholesterol, mg/dL |
121 (94–148) |
119 (93–147) |
121 (94–149) |
121 (95–148) |
|
Creatinine, mg/dL |
0.8 (0.7–1) |
0.9 (0.8–1) |
0.9 (0.7–1) |
0.9 (0.7–1) |
|
Calendar months until entry‡
|
19 (13–25) |
15 (7–21) |
17 (10–22) |
15 (7–20) |
Table 2.Major adverse cardiovascular events according to oral antidiabetic drug class
Table 2.
|
Variable |
SGLT2 inhibitors |
Thiazolidinediones |
DPP-4 inhibitors |
Sulfonylureas |
|
Patients, No. |
4,365 |
2,033 |
51,348 |
13,325 |
|
Events |
55 |
46 |
1,152 |
446 |
|
PYs |
19,724 |
9,701 |
238,967 |
63,333 |
|
Incidence per 100,000 PYs |
278.8 |
474.2 |
482.1 |
704.2 |
|
Adjusted subdistribution hazard ratio (95% CI)*
|
|
|
|
|
|
Versus sulfonylureas |
0.44 (0.31–0.62) |
0.77 (0.56–1.05) |
0.80 (0.72–0.90) |
NA |
|
Versus DPP-4 inhibitors |
0.59 (0.42–0.83) |
0.97 (0.72–1.32) |
NA |
NA |
|
Versus thiazolidinediones |
0.61 (0.39–0.96) |
NA |
NA |
NA |
Table 3.Mediating effects of MASLD regression on the association between oral antidiabetic drugs and major adverse cardiovascular events
Table 3.
|
Mediators |
Total effect
|
Direct effect
|
Indirect (mediated) effect
|
|
Rate difference |
Rate difference |
Proportion |
Rate difference |
Proportion |
|
MASLD regression (FLI <30) |
|
|
|
|
|
|
SGLT2 inhibitors vs. thiazolidinediones |
−7.85 (−17.85 to −1.8) |
−7.66 (−13.81 to −3.47) |
97.6% |
−0.19 (−0.6 to −0.02) |
2.4% |
|
SGLT2 inhibitors vs. DPP-4 inhibitors |
−8.41 (−11.87 to −3.71) |
−7.84 (−15.98 to −10.17) |
93.2% |
−0.57 (−0.82 to −0.34) |
6.8% |
|
SGLT2 inhibitors vs. sulfonylureas |
−13.46 (−18.84 to −10.27) |
−12.29 (−17.37 to −8.7) |
91.3% |
−1.17 (−1.54 to −0.75) |
8.7% |
Abbreviations
adjusted subdistribution hazard ratio
International Classification of Diseases 10th Revision
inverse probability of treatment weighting
major adverse cardiovascular events
metabolic dysfunction-associated steatotic liver disease
moderate-to-vigorous physical activity
National Health Insurance Service
sodium-glucose cotransporter 2
REFERENCES
- 1. Mensah GA, Fuster V, Murray CJL, Roth GA. Global burden of cardiovascular diseases and risks, 1990-2022. J Am Coll Cardiol 2023;82:2350-2473.
- 2. Rinella ME, Lazarus JV, Ratziu V, Francque SM, Sanyal AJ, Kanwal F, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. Hepatology 2023;78:1966-1986.
- 3. Targher G, Byrne CD, Tilg H. MASLD: a systemic metabolic disorder with cardiovascular and malignant complications. Gut 2024;73:691-702.
- 4. Younossi ZM, Kalligeros M, Henry L. Epidemiology of metabolic dysfunction-associated steatotic liver disease. Clin Mol Hepatol 2025;31:S32-S50.
- 5. Younossi ZM, Golabi P, Price JK, Owrangi S, Gundu-Rao N, Satchi R, et al. The global epidemiology of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis among patients with type 2 diabetes. Clin Gastroenterol Hepatol 2024;22:1999-2010.e8.
- 6. Barrera F, Uribe J, Olvares N, Huerta P, Cabrera D, Romero-Gómez M, et al. The Janus of a disease: diabetes and metabolic dysfunction-associated fatty liver disease. Ann Hepatol 2024;29:101501.
- 7. Han E, Han KD, Lee YH, Kim KS, Hong S, Park JH, et al. Fatty liver & diabetes statistics in Korea: nationwide data 2009 to 2017. Diabetes Metab J 2023;47:347-355.
- 8. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 2015;373:2117-2128.
- 9. Dormandy JA, Charbonnel B, Eckland DJ, Erdmann E, Massi-Benedetti M, Moules IK, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet 2005;366:1279-1289.
- 10. Jang H, Kim Y, Lee DH, Joo SK, Koo BK, Lim S, et al. Outcomes of various classes of oral antidiabetic drugs on nonalcoholic fatty liver disease. JAMA Intern Med 2024;184:375-383.
- 11. Kawaguchi T, Murotani K, Kajiyama H, Obara H, Yamaguchi H, Toyofuku Y, et al. Effects of luseogliflozin on suspected MASLD in patients with diabetes: a pooled meta-analysis of phase III clinical trials. J Gastroenterol 2024;59:836-848.
- 12. Kawaguchi T, Fujishima Y, Wakasugi D, Io F, Sato Y, Uchida S, et al. Effects of SGLT2 inhibitors on the onset of esophageal varices and extrahepatic cancer in type 2 diabetic patients with suspected MASLD: a nationwide database study in Japan. J Gastroenterol 2024;59:1120-1132.
- 13. Wu JY, Hsu WH, Kuo CC, Tsai YW, Liu TH, Huang PY, et al. A retrospective analysis of combination therapy with GLP-1 receptor agonists and SGLT2 inhibitors versus SGLT2 inhibitor monotherapy in patients with MASLD. Nat Commun 2025;16:7459.
- 14. Rosenstock J, Perkovic V, Johansen OE, Cooper ME, Kahn SE, Marx N, et al. Effect of linagliptin vs placebo on major cardiovascular events in adults with type 2 diabetes and high cardiovascular and renal risk: the CARMELINA randomized clinical trial. JAMA 2019;321:69-79.
- 15. Kim KS, Hong S, Han K, Park CY. Association of non-alcoholic fatty liver disease with cardiovascular disease and all cause death in patients with type 2 diabetes mellitus: nationwide population based study. BMJ 2024;384:e076388.
- 16. Hernán MA, Sauer BC, Hernández-Díaz S, Platt R, Shrier I. Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses. J Clin Epidemiol 2016;79:70-75.
- 17. Lee HH, Lee HA, Kim EJ, Kim HY, Kim HC, Ahn SH, et al. Metabolic dysfunction-associated steatotic liver disease and risk of cardiovascular disease. Gut 2024;73:533-540.
- 18. Tajima A, Tobe K, Eiki JI, Origasa H, Watada H, Shimomura I, et al. Treatment patterns and satisfaction in patients with type 2 diabetes newly initiating oral monotherapy with antidiabetic drugs in Japan: results from the prospective Real-world Observational Study on Patient Outcomes in Diabetes (RESPOND). BMJ Open Diabetes Res Care 2022;10:e003032.
- 19. Yoshida K, Solomon DH, Kim SC. Active-comparator design and new-user design in observational studies. Nat Rev Rheumatol 2015;11:437-441.
- 20. Baumgartner PC, Haynes RB, Hersberger KE, Arnet I. A systematic review of medication adherence thresholds dependent of clinical outcomes. Front Pharmacol 2018;9:1290.
- 21. Luijken K, Spekreijse JJ, van Smeden M, Gardarsdottir H, Groenwold RHH. New-user and prevalent-user designs and the definition of study time origin in pharmacoepidemiology: a review of reporting practices. Pharmacoepidemiol Drug Saf 2021;30:960-974.
- 22. Arnott C, Li Q, Kang A, Neuen BL, Bompoint S, Lam CSP, et al. Sodium-glucose cotransporter 2 inhibition for the prevention of cardiovascular events in patients with type 2 diabetes mellitus: a systematic review and meta-analysis. J Am Heart Assoc 2020;9:e014908.
- 23. Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994;81:515-526.
- 24. Aalen OO, Stensrud MJ, Didelez V, Daniel R, Røysland K, Strohmaier S, et al. Time-dependent mediators in survival analysis: modeling direct and indirect effects with the additive hazards model. Biom J 2020;62:532-549.
- 25. Neal B, Perkovic V, Mahaffey KW, de Zeeuw D, Fulcher G, Erondu N, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017;377:644-657.
- 26. Wiviott SD, Raz I, Bonaca MP, Mosenzon O, Kato ET, Cahn A, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med 2019;380:347-357.
- 27. Zelniker TA, Wiviott SD, Raz I, Im K, Goodrich EL, Bonaca MP, et al. SGLT2 inhibitors for primary and secondary prevention of cardiovascular and renal outcomes in type 2 diabetes: a systematic review and meta-analysis of cardiovascular outcome trials. Lancet 2019;393:31-39.
- 28. Hur KY, Moon MK, Park JS, Kim SK, Lee SH, Yun JS, et al. 2021 clinical practice guidelines for diabetes mellitus of the Korean Diabetes Association. Diabetes Metab J 2021;45:461-481.
- 29. Lipscombe LL, Gomes T, Lévesque LE, Hux JE, Juurlink DN, Alter DA, et al. Thiazolidinediones and cardiovascular outcomes in older patients with diabetes. JAMA 2007;298:2634-2643.
- 30. Papagianni M, Tziomalos K. Cardiovascular effects of dipeptidyl peptidase-4 inhibitors. Hippokratia 2015;19:195-199.
- 31. Ehlers LH, Lamotte M, Ramos MC, Sandgaard S, Holmgaard P, Kristensen MM, et al. The cost-effectiveness of subcutaneous semaglutide versus empagliflozin in type 2 diabetes uncontrolled on metformin alone in Denmark. Diabetes Ther 2022;13:489-503.
- 32. Johnson CE, Sussman WB, Weeda ER. Medication adherence to sodium-glucose cotransporter-2 inhibitors versus glucagon-like peptide-1 receptor agonists: a meta-analysis. Diabetes Obes Metab 2024;26:4544-4550.
- 33. Lopaschuk GD, Verma S. Mechanisms of cardiovascular benefits of sodium glucose co-transporter 2 (SGLT2) inhibitors: a state-of-the-art review. JACC Basic Transl Sci 2020;5:632-644.
- 34. Verma S, Garg A, Yan AT, Gupta AK, Al-Omran M, Sabongui A, et al. Effect of empagliflozin on left ventricular mass and diastolic function in individuals with diabetes: an important clue to the EMPA-REG OUTCOME trial? Diabetes Care 2016;39:e212-e213.
- 35. Preda A, Montecucco F, Carbone F, Camici GG, Lüscher TF, Kraler S, et al. SGLT2 inhibitors: from glucose-lowering to cardiovascular benefits. Cardiovasc Res 2024;120:443-460.
- 36. Wei R, Wang W, Pan Q, Guo L. Effects of SGLT-2 inhibitors on vascular endothelial function and arterial stiffness in subjects with type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials. Front Endocrinol (Lausanne) 2022;13:826604.
- 37. Ikonomidis I, Thymis J, Pavlidis G, Birba D, Kalogeris A, Kousathana F, et al. The effects of empagliflozin on arterial stiffness, endothelial function and ventriculoarterial coupling in type 2 diabetes mellitus: 1 year follow up. Eur Heart J 2020;41:ehaa946.3354.
- 38. Khawaja T, Nied M, Wilgor A, Neeland IJ. Impact of visceral and hepatic fat on cardiometabolic health. Curr Cardiol Rep 2024;26:1297-1307.
- 39. Masson W, Lavalle-Cobo A, Nogueira JP. Effect of SGLT2-inhibitors on epicardial adipose tissue: a meta-analysis. Cells 2021;10:2150.
- 40. Scisciola L, Cataldo V, Taktaz F, Fontanella RA, Pesapane A, Ghosh P, et al. Anti-inflammatory role of SGLT2 inhibitors as part of their anti-atherosclerotic activity: data from basic science and clinical trials. Front Cardiovasc Med 2022;9:1008922.
- 41. Takahara S, Soni S, Maayah ZH, Ferdaoussi M, Dyck JRB. Ketone therapy for heart failure: current evidence for clinical use. Cardiovasc Res 2022;118:977-987.
- 42. Su S, Ji X, Li T, Teng Y, Wang B, Han X, et al. The changes of cardiac energy metabolism with sodium-glucose transporter 2 inhibitor therapy. Front Cardiovasc Med 2023;10:1291450.
- 43. Brata R, Pascalau AV, Fratila O, Paul I, Muresan MM, Camarasan A, et al. Hemodynamic effects of SGLT2 inhibitors in patients with and without diabetes mellitus-a narrative review. Healthcare (Basel) 2024;12:2464.
- 44. Chen X, Yang Q, Bai W, Yao W, Liu L, Xing Y, et al. Dapagliflozin attenuates myocardial fibrosis by inhibiting the TGF-β1/Smad signaling pathway in a normoglycemic rabbit model of chronic heart failure. Front Pharmacol 2022;13:873108.
- 45. Herat LY, Magno AL, Rudnicka C, Hricova J, Carnagarin R, Ward NC, et al. SGLT2 inhibitor-induced sympathoinhibition: a novel mechanism for cardiorenal protection. JACC Basic Transl Sci 2020;5:169-179.
- 46. Kawaguchi T, Nakano D, Okamura S, Shimose S, Hayakawa M, Niizeki T, et al. Spontaneous regression of hepatocellular carcinoma with reduction in angiogenesis-related cytokines after treatment with sodium-glucose cotransporter 2 inhibitor in a cirrhotic patient with diabetes mellitus. Hepatol Res 2019;49:479-486.
- 47. Nakano D, Kawaguchi T, Tsutsumi T, Hayakawa M, Yoshio S, Koga H, et al. Effects of SGLT2 inhibitor on tumor-releasing chemokines/cytokines in Hep3B and Huh7 cells. JGH Open 2022;6:270-273.
- 48. Nakano D, Kawaguchi T, Iwamoto H, Hayakawa M, Koga H, Torimura T, et al. Effects of canagliflozin on growth and metabolic reprograming in hepatocellular carcinoma cells: multiomics analysis of metabolomics and absolute quantification proteomics (iMPAQT). PLoS One 2020;15:e0232283.
- 49. Nakano D, Akiba J, Tsutsumi T, Kawaguchi M, Yoshida T, Koga H, et al. Hepatic expression of sodium-glucose cotransporter 2 (SGLT2) in patients with chronic liver disease. Med Mol Morphol 2022;55:304-315.
, Stefano Romeo4
