Cardiorenal Effectiveness of Empagliflozin vs. GLP-1 Receptor Agonists in Patients with Advanced Chronic Kidney Disease: Results from the EMPRISE study

Introduction: Patients with chronic kidney disease (CKD) are at an increased risk of cardiovascular disease, heart failure, and mortality. We reported the final year results of the EMPRISE study program in patients with type 2 diabetes (T2D) and advanced CKD. Here, we aimed to assess the cardiovascular effectiveness of empagliflozin compared to glucagon-like peptide-1 receptor agonists (GLP-1RA) in patients with T2D and CKD stages 3-4.

Methods: Using claims data from 2 commercial insurance datasets (Optum Clinformatics (CDM) and IBM®​ MarketScan®​) and Medicare (2014-2019), we evaluated the cardiovascular effectiveness of empagliflozin compared to glucagon-like peptide-1 receptor agonists (GLP-1RA) in patients with T2D and CKD stages 3-4. The Study design is a new-user active-comparator cohort study and included adults ≥ 18 years (≥ 65 years in Medicare) with T2D and diagnoses for CKD stages 3-4. The exposure and comparator were Initiators of empagliflozin relative to GLP-1RA. The measured outcomes were hospitalizations for myocardial infarction, ischemic or hemorrhagic stroke, and heart failure, which were identified using validated claims-based algorithms with high specificity. End-staged kidney disease (ESKD) included CKD stage 5, dialysis, kidney transplant, and replacement therapy.

Confounding was addressed via 1:1 propensity score (PS) matching adjusting for 143 baseline patient characteristics. PS was estimated as the predicted probability of initiating empagliflozin vs. GLP-1RA using multivariable logistic regression, and PS estimation and matching were conducted separately within each data source. Hazard ratios (HR) and rate differences (RD), accounting for mortality as a competing risk, were estimated in the final database pooled across 3 data sources.

Results: Using Medicare, Optum, and Marketscan data (2014-19), we identified 10,930 pairs of 1:1 propensity score-matched patients aged > 18 years with T2D and CKD stage 3-4 who initiated empagliflozin (EMPA) or a glucagon-like peptide-1 receptor agonist (GLP-1RA). Primary outcomes were hospitalization for heart failure (HHF), a composite of myocardial infarction (MI) or stroke, and end-stage renal disease (ESRD). Secondary outcomes were MI, stroke, and all-cause mortality. We estimated hazard ratios (HR) and rate differences (RD), adjusting for 143 baseline covariates. Compared to GLP-1RA, EMPA was associated with a reduced risk of HHF [HR 0.68 (0.55-0.85); RD -9.44 (-14.78, -4.10)], and ESRD [HR 0.70 (0.56-0.87); RD -9.01 (-14.30, -3.72)], and a trend towards reduced risk for the composite outcome [HR 0.88 (0.72-1.07); RD -3.84 (-9.60, 1.93)]. Estimates for the secondary outcomes were: MI [HR 0.78 (0.61-1.00)], stroke [HR 1.16 (0.83-1.62)], and all-cause mortality [HR 1.08 (0.86-1.36)].

Conclusion: In patients with T2D and advanced CKD in routine care, we observed risk reductions for cardiorenal outcomes with EMPA vs. GLP-1RA.

Funding: Boehringer Ingelheim (116283)

Publication History

Article published online:
18 April 2024