Diabetes is a big issue in our nation. There are 37.3 million people in the United States with diabetes, and diabetes can cause micro- or macrovascular complications such as myocardial infarction and stroke. Unfortunately, patients with type 2 diabetes are 2 times more likely than those without diabetes to suffer from heart attack and stroke. It is really important for us to try to lower patients' A1c and prevent the risk of these catastrophic events.

For our study, we compared GLP-1 receptor agonists to sulfonylureas in an integrated health care system. Many clinical studies show GLP-1 receptor agonists have cardiovascular (CV) benefits, but no trials have compared GLP-1 receptor agonists head-to-head with specific glucose-lowering agents such as sulfonylureas. 

Our retrospective longitudinal study was conducted in the Kaiser Permanente California and Colorado regions. Our primary outcome was a composite of CV events including myocardial infarction and stroke. 

We also did a subgroup analysis of the primary outcome, so we looked at the incidence of CV events in patients at high vs low risk for CV events. High-risk patients had a history of CV events, and low-risk patients had no history. Secondary outcomes included change in A1c and a 5% decrease in weight within 1 year.

We found no statistically significant difference in the incidence of CV events between groups or in the subgroup analysis.

Regarding the secondary outcome, the sulfonylurea group had a significantly greater reduction of A1c compared to those receiving GLP-1 receptor agonists. Across the entire cohort, the GLP-1 receptor agonist group had a significant reduction in weight compared to the sulfonylurea group, and this was also found in patients with a body mass index of less than 30. In patients with a body mass index of 30 or more, there was no difference in CV events between the two groups.

GLP-1 receptor agonists are a hot topic right now, so these findings pose the question: do we really need to put our patients on a GLP-1 receptor agonist if this agent does not provide any CV benefit? 

This poster has some strengths and limitations. In terms of strengths, this was multi-center study, and patients were well-matched using Poisson regression. In terms of limitations, this was a retrospective study. Ideally, we would do a randomized, controlled trial and prospective study. 

Additionally, the GLP-1 receptor agonists were analyzed together as a class rather than as individual medications, and clinical trials have shown that specific GLP-1 receptor agonists reduce CV events while others do not. I think future studies can analyze the GLP-1 receptor agonists individually, and that can build on the findings from this study.

In the future, my coinvestigators hope to compare GLP-1 receptor agonists to other diabetic therapies.