BACKGROUND: In type 2 diabetes, cardioprotective glucose-lowering drugs, including sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists, and statins reduce the risk of secondary major adverse cardiovascular events. No trials examined the combination of these drugs as withholding statins in high-risk individuals would be unethical. We tested the hypothesis that cardioprotective glucose-lowering drug and statin combined is associated with lower risk of secondary major adverse cardiovascular events than either drug alone.
METHODS: Individuals with type 2 diabetes and established cardiovascular disease from December 2012 through 2021 were identified via national Danish health registers. They were analyzed in an active comparator cohort including 15,404 individuals followed from treatment intensification with cardioprotective glucose-lowering drug or dipeptidyl peptidase-4 inhibitor and additionally in a time-varying cohort including 76,853 individuals with yearly updated treatment and covariate status. The treatment groups were: (i) no cardioprotective drug (no use of cardioprotective glucose-lowering drug or statin), (ii) cardioprotective glucose-lowering drug, (iii) statin, and (iv) cardioprotective glucose-lowering drug and statin. The primary outcome was a new major adverse cardiovascular event (myocardial infarction, stroke, or cardiovascular death).
RESULTS: During mean 2.7 and 4.7 years of follow-up, 1,843 and 23,051 had major adverse cardiovascular events in the active comparator and time-varying cohorts. In the active comparator cohort, when compared to nonusers of cardioprotective glucose-lowering drug or statin, multivariable adjusted hazard ratios of major adverse cardiovascular events were 0.82 (95% confidence interval: 0.67 to 1.02) for cardioprotective glucose-lowering drug, 0.85 (0.74 to 0.97) for statin, and 0.71 (0.60 to 0.84) for cardioprotective glucose-lowering drug and statin combined. Corresponding hazard ratios in the time-varying cohort were 0.77 (0.69 to 0.86), 0.73 (0.70 to 0.75), and 0.57 (0.54 to 0.60), respectively. When restricting the active comparator cohort to individuals entering observation between 2019 and 2021 with reduced statistical power, the corresponding hazard ratios were 0.86 (0.57 to 1.31), 0.89 (0.60 to 1.30), and 0.79 (0.54 to 1.14), respectively. In both cohorts p for interaction between cardioprotective glucose-lowering drug and statin was > 0.05.
CONCLUSIONS AND RELEVANCE: In individuals with type 2 diabetes and established cardiovascular disease, treatment with a cardioprotective glucose-lowering drug and statin in combination was associated with lower risk of secondary major adverse cardiovascular events than using either drug alone. This is important given the persistently suboptimal uptake of both drug classes in real-world practice. Some biases can never be completely excluded in real-world pharmacotherapy use studies such as this one, including confounding by indication, time-related or immortal time biases, and shifting standards of care, rendering causal interpretation unattainable; however, our results seemed consistent across numerous sensitivity analyses and designs.
