Head-to-head Comparison of Dapagliflozin vs Empagliflozin: Cardiovascular and Safety Events

Background: Sodium-glucose co-transporter 2 inhibitors (SGLT-2i) are recommended for adults with type 2 diabetes (T2D) at high cardiovascular (CV) risk. However, the comparative effectiveness and safety of dapagliflozin and empagliflozin are unknown.

Objectives: We investigated CV and safety events comparing dapagliflozin versus empagliflozin among adults with T2D.

Methods: Using claims data from two commercial (Clinformatics and MarketScan) and one public (Medicare) US health insurance, we identified individuals aged 18 years and older (older than 65 years in Medicare) with T2D who initiated dapagliflozin or empagliflozin between August 2014 and June 2020. Endpoints evaluated were 2-point MACE (a composite of myocardial infarction [MI] or stroke), hospitalization for heart failure (HHF), MI, stroke, all-cause death, and safety outcomes (diabetic ketoacidosis, amputations, fractures, urinary tract infections, genital infections, acute kidney injury, and hypoglycemia). Follow-up began the day after initiating the exposure of interest and continued as long as individuals continued the initial drug. Database-specific models were weighted using propensity score matching-weights to adjust for confounding, and hazard ratios (HRs) were pooled across databases.

Results: We identified 107,602 dapagliflozin and 272,014 empagliflozin initiators. Compared with empagliflozin, dapagliflozin initiators were younger (mean age 56 vs. 61 years) and had less CV disease at baseline (19% vs 31%). After adjusting for 129 confounders, compared with empagliflozin at both 10 mg and 25 mg doses, dapagliflozin showed similar risk for most CV outcomes, including MI/stroke (HR=0.96, 95% CI=0.88–1.03). However, dapagliflozin initiators had a higher risk of HHF (HR=1.17, 1.02–1.33), driven by the lower dose of dapagliflozin (HR5mg=1.23, 1.04–1.46 for dapagliflozin 5mg; HR10mg=1.08, 0.88–1.32 for dapagliflozin 10mg). Compared to those on empagliflozin, dapagliflozin initiators had a slightly lower risk of genital infections (HR=0.93, 0.90–0.96) across all doses and a similar risk of the other safety outcomes.

Conclusions: The findings of this large multi-database cohort study suggest that dapagliflozin at the dose proven effective in cardiovascular outcome trials, i.e., the 10mg dose, offers similar CV benefits compared with empagliflozin at both 10 mg and 25 mg doses. However, the lower dose of dapagliflozin, 5 mg, which is indicated for the management of hyperglycemia, may not offer similar CV effectiveness. Further studies are needed to confirm the CV effectiveness of the lower dose of dapagliflozin.