Sodium–Glucose Cotransporter 2 Inhibitors in Cardiovascular and Renal Outcomes in Patients With Diabetes but Without Established Cardiovascular Disease: A Nationwide Population-Based Cohort Study

Type 2 diabetes is associated with a substantially increased risk for cardiovascular diseases. Reports on the cardioprotective effects of sodium–glucose cotransporter 2 (SGLT2) inhibitors in cardiovascular outcome trials (1,2) have brought about a paradigm shift in the understanding of the role of antidiabetic agents beyond glycemic control. However, as the majority of available evidence pertains to patients with established cardiovascular diseases or at high risk for them, whether similar benefits can be extended to individuals without established cardiovascular disease as a primary prevention strategy is unclear.

We conducted a nationwide population-based cohort study using retrospective data from the National Health Insurance Research Database in Taiwan. A total of 85,579 patients with diabetes without prior established cardiovascular diseases were identified (aged 61.1 ± 11.9 years, 50.3% men), including 2,497 patients on SGLT2 inhibitors and 83,082 patients on other glucose-lowering drugs. Patients with diabetes were identified from the registry in 2016 using the International Classification of Diseases, 10th revision, clinical modification code E08. Patients aged <18 years with prior history of heart failure (HF), chronic kidney disease (CKD), stroke, coronary artery disease, or peripheral arterial occlusive disease were excluded. Each patient receiving an SGLT2 inhibitor (prescribed for ≥180 days) was matched with patients taking other glucose-lowering drugs (at a 1:2 ratio) using propensity score matching (PSM). All study participants were followed up until death, the occurrence of cardiovascular or renal outcomes, or censoring at the administrative date of 31 December 2020. Cardiovascular outcomes include cardiovascular death (I00–I99), hospitalization for HF (HHF) (I50), and stroke (I60–I68), whereas renal outcome was the progression to end-stage renal disease (ESRD) requiring either hemodialysis or peritoneal dialysis. The conditional Cox proportional hazards models were used to estimate the hazard ratios (HR) of SGLT2 inhibitors versus other glucose-lowering medications after adjusting for age, sex, comorbidities, and prescribed medications in the fully adjusted model.

Following PSM, 2,497 patients receiving SGLT2 inhibitors were compared with 4,994 patients receiving other glucose-lowering agents. Since the PSM encompassed age, sex, comorbidities, and prescribed medications, most imbalanced variables between the two groups became statistically insignificant. Compared with patients on other glucose-lowering agents, patients on SLGT2 inhibitors had lower all-cause mortality (incidence rate 2.39 vs. 5.17 per 100 person-years), HHF (1.97 vs. 2.66 per 100 person-years), and progression to ESRD (0.39 vs. 2.16 per 100 person-years). However, there were no significant differences between the two groups in terms of cardiovascular death (0.64 vs. 0.92 per 100 person-years) and stroke (2.04 vs. 2.89 per 100 person-years). In the fully adjusted Cox regression model, the use of SGLT2 inhibitors was associated with a reduced risk for all-cause mortality (HR 0.51, 95% CI 0.38–0.67; P < 0.001), HHF (0.71, 0.52–0.99; P = 0.04), and the progression to ESRD (0.24, 0.13–0.46; P < 0.001) (Table 1).

In this study, we found that among patients with diabetes without prior established cardiovascular diseases, patients taking SGLT2 inhibitors had a lower risk of all-cause mortality, HHF, and progression to ESRD than those taking other glucose-lowering medicines. As patients with diabetes in clinical practice are frequently heterogeneous and differ from subjects in the highly selected randomized controlled trials, we should be cautious when inferring the cardiovascular benefits of SGLT2 inhibitors to patients with diverse baseline risk profiles. The Comparative Effectiveness of Cardiovascular Outcomes in New Users of SGLT-2 Inhibitors (CVD-REAL) study has extended the findings to broader groups of cardiovascular risk patients (3); however, the heterogeneous study population could limit its generalizability. Our study therefore mainly focused on the clearly defined low-cardiovascular-risk patients with diabetes, testing the role of SGLT2 inhibitors on adverse cardiovascular events from a primary prevention perspective. In addition to cardiovascular events, patients with diabetes remain at considerable risk for progression to ESRD despite the effectiveness of renin-angiotensin system blockade. As previous trials have only enrolled high-risk patients with either established cardiovascular diseases or CKD (4,5), the preventive role of SGLT2 inhibitors in patients yet to develop diabetic kidney disease remains unclear. By enrolling low-cardiovascular-risk patients without known CKD, our study demonstrates that early use of SGLT2 inhibitors may reduce the risk of progression to ESRD.

The current study has some limitations. Despite this study demonstrating that administration of SGLT2 inhibitors may lead to favorable outcomes, further prospective randomized controlled trials are mandatory to validate the findings. Furthermore, given the nature of an observational cohort study, and despite robust PSM and full adjustment in the multivariable model, unmeasured confounders may still exist.

Our study provided clinical evidence that the cardiovascular and renoprotective benefits of SGLT2 inhibitors could be consistently extended to primary prevention strategies among patients with diabetes with low cardiovascular risk. To validate the superiority of SGLT2 inhibitors over other diabetes medications, additional prospective, randomized controlled trials will be required.