The EMPA-REG OUTCOME study investigated the impact of empagliflozin on CV morbidity and mortality in patients with type 2 diabetes who have CV disease.7
A total of 7,020 patients with established CV disease (prior myocardial infarction, coronary artery disease, stroke, unstable angina or occlusive peripheral arterial disease) were randomised to receive either empagliflozin (10 or 25 mg) or placebo once daily, in addition to standard of care treatments for type 2 diabetes.7 The median observation time was 3.1 years.
The mean age of the cohort was approximately 63 years, 70%–72% were men, mean HbA1c was approximately 8% and mean BMI was approximately 30.6 kg/m². Within the cohort, 80%–81% were using ACE inhibitors, and 76%–78% were using statins.7
Reduced incidence of death from CV causes, non-fatal MI or non-fatal stroke
The primary composite outcome of death from CV causes, non-fatal myocardial infarction, or non-fatal stroke, occurred in a significantly lower proportion of patients taking empagliflozin (pooled) compared to placebo (10.5% vs 12.1%, respectively, hazard ratio (HR) 0.86, 95% CI 0.74 to 0.99, p = 0.04 for superiority).7
When expressed as a number needed to treat (NNT), 62 patients would need to be treated with empagliflozin (pooled across 10–25 mg doses) over a 3-year period to prevent death from CV causes, non-fatal myocardial infarction, or non-fatal stroke in 1 patient.
When the individual components of the composite primary outcome were assessed separately, a significantly lower rate of death from CV causes was observed in those treated with empagliflozin (pooled) versus placebo (3.7% vs 5.9%, HR 0.62, 95% CI 0.49 to 0.77). However, no significant difference was found for incidence of non-fatal myocardial infarction (4.5% vs 5.2%, respectively) or non-fatal stroke (3.2% vs 2.6%, respectively) between the empagliflozin and placebo groups, respectively.7
There was also no significant difference in the key secondary composite outcome (the primary outcome plus hospitalisation for unstable angina) between the pooled empagliflozin and placebo groups (14.3% vs 12.8%, HR 0.89, 95% CI 0.78 to 1.01, p = 0.08).
Other secondary outcomes measured separately indicated a reduction in deaths from any cause (HR 0.68, 95% CI 0.57 to 0.82, p < 0.001) and hospitalisation for heart failure (HR 0.65, 95% CI 0.50 to 0.85, p < 0.002) in the pooled empagliflozin group compared to placebo.7
The mechanism by which empagliflozin may influence CV outcomes has not yet been established.8 It has been speculated that it may involve osmotic diuresis, effects on hyperglycaemia, weight and blood pressure.7,9,10
Limitations of findings
When considering the expanded indications for empagliflozin based on the EMPA-REG study results in 2016, the Federal Drug Administration Advisory Committee noted a number of issues which require consideration when interpreting the robustness of the primary endpoint results.8
Prospective design of the trial was to assess safety, not CV benefit
The FDA noted that the primary objective of the EMPA-REG OUTCOME study was to exclude the possibility that use of empagliflozin to control glycaemia increased CV risk (predominantly absolute CVD risk) by 30% or more compared to the use of alternate, standard of care, glycaemic lowering agents.8
It was designed to assess non-inferiority of empagliflozin versus placebo for the 3-point major adverse CV endpoint (CV death, non-fatal MI or non-fatal stroke).11
The primary intent of the EMPA-REG OUTCOME study was not to establish a benefit on a specific outcome, and it was not prospectively designed (eg, sized) with the expressed intent of demonstrating a CV benefit of a glucose-lowering therapy.8
Influence of ‘non-assessable’ deaths on primary endpoint
A subset of deaths (n = 124) were categorised as ‘non-assessable’ and adjudicated as presumed CV deaths; this accounted for 71 deaths in the empagliflozin group, and 53 in the placebo group. Those ‘non-assessable’ deaths that were presumed to be CV deaths comprised 40% of CV deaths, and 27% of overall deaths in the trial.8
A sensitivity analysis that removed these deaths from the primary composite endpoint (3-point major adverse cardiac events, or MACE) found that empagliflozin was not superior to placebo for reducing death from CV causes, non-fatal MI and non-fatal stroke (HR 0.90, 95% CI 0.77 to 1.06).8
Sensitivity analyses on the single component of CV death, excluding the 124 non-assessable deaths, suggested there was a still a reduction in CV deaths (HR 0.59, 95% CI 0.44 to 0.79) and all-cause mortality (HR 0.68, 95% CI 0.57 to 0.82) in those treated with empagliflozin compared to placebo.8