SOURCE: Zinman B, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 2015;373:2117-2128.

Current FDA regulations require that new pharmacologic agents to treat diabetes must demonstrate cardiovascular safety in addition to glucose lowering. The rationale for this stipulation is that some very early clinical trials with sulfonylureas in type 2 diabetes (T2DM) suggested a negative effect on cardiovascular outcomes, and successive clinical trials have failed to demonstrate statistically significant improvements in cardiovascular outcomes, despite excellent glucose control. Indeed, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial suggested a potential worsening of cardiovascular outcomes and mortality in tightly controlled T2DM patients (A1c < 6.5).

The cardiovascular safety trial for empagliflozin, a sodium-glucose co-transporter 2 (SGLT2) inhibitor, is known as the EMPA-REG trial. Investigators randomized T2DM patients (n = 7020) to empagliflozin or placebo for 3 years. At the conclusion of the trial, the primary outcome (a composite of death from cardiovascular cases, nonfatal myocardial infarction, and nonfatal stroke) was 14% better in the treatment group than placebo (absolute risk reduction 1.7%).

While these results were the source of much celebration when presented at the European Association for the Study of Diabetes meeting in Stockholm, Sweden, in September 2015, there are reasons for pause. First, clinical trials with two other SGLT2 inhibitors (dapagliflozin, canagliflozin), which work by essentially comparable mechanisms, have not yet completed their cardiovascular safety trials. On the other hand, the cardiovascular outcomes from trial data thus far accrued with both of these other agents does not suggest cardiovascular risk reduction. Until FDA-mandated cardiovascular risk trials are completed with each of the agents in this class, whether individual agents might provide particular cardiovascular benefits or not will remain speculative. Second, it is difficult to reconcile how a trial that did not show a reduction in fatal or nonfatal myocardial infarction and did not show a reduction in fatal or nonfatal stroke achieved a reduction in cardiovascular mortality. Isn’t the combination of stroke + myocardial infarction the primary constituent of cardiovascular mortality?

For the time being, physicians will have to defer to the statistical wisdom of clinical trial data experts to decipher this intuitively self-contradictory result. While some physicians may want to celebrate the possible discovery of a pharmacologic treatment for T2DM associated with improved cardiovascular outcomes, I am not there yet.