The EMPA-REG OUTCOME trial is one of the most important cardiovascular trials in diabetes history. Published in the New England Journal of Medicine in 2015 by Zinman and colleagues, it was the first cardiovascular outcomes trial of an SGLT2 inhibitor — and produced unprecedented mortality reductions in type 2 diabetes patients with established cardiovascular disease. The trial launched the SGLT2 inhibitor era and transformed diabetes care from a glucose-centric model to a cardio-renal-protective model. Today the ADA Standards of Care recommend SGLT2 inhibitors and GLP-1 receptor agonists for patients with established CV disease, heart failure, or chronic kidney disease — a direct consequence of EMPA-REG and the follow-on trials it inspired.
Background: Why EMPA-REG Was Needed
After ACCORD (2008) and ADVANCE (2008) failed to show clear cardiovascular benefit from intensive glucose control alone, the FDA in 2008 required cardiovascular outcomes trials for all new type 2 diabetes drugs. The goal was to rule out cardiovascular harm — not necessarily to show benefit. Early trials of DPP-4 inhibitors (SAVOR-TIMI 53 with saxagliptin, EXAMINE with alogliptin, TECOS with sitagliptin) showed neutrality. EMPA-REG OUTCOME was the cardiovascular outcomes trial for empagliflozin, the SGLT2 inhibitor developed by Boehringer Ingelheim and Eli Lilly. Expectations were modest — neutrality would have been acceptable. The actual results stunned the field.
Trial Design
| Design Feature | Detail |
|---|---|
| Trial type | Multicenter, randomized, double-blind, placebo-controlled |
| Years | 2010 to 2015 |
| Participants | 7,020 adults with type 2 diabetes and established CV disease |
| Mean age | 63 years |
| Mean diabetes duration | ~10 years |
| Mean A1C at baseline | 8.1 percent |
| Established CV disease | 99 percent (prior MI, CAD, stroke, or PAD) |
| Arms | Empagliflozin 10 mg, empagliflozin 25 mg, placebo (1:1:1) |
| Background therapy | Standard care including metformin, insulin, statins, ACE/ARB, antiplatelets |
| Median follow-up | 3.1 years |
| Primary outcome | 3-point MACE: CV death, nonfatal MI, nonfatal stroke |
| Key secondary outcome | 4-point MACE plus hospitalization for unstable angina |
Primary Results
| Outcome (Empagliflozin vs Placebo) | Hazard Ratio | P value | Relative Risk Reduction |
|---|---|---|---|
| Primary 3-point MACE | 0.86 | 0.04 (superiority) | 14 percent |
| Cardiovascular death | 0.62 | less than 0.001 | 38 percent |
| All-cause mortality | 0.68 | less than 0.001 | 32 percent |
| Hospitalization for heart failure | 0.65 | 0.002 | 35 percent |
| Nonfatal myocardial infarction | 0.87 | NS | 13 percent |
| Nonfatal stroke | 1.24 | NS | 24 percent increase (not significant) |
| Composite renal outcome | 0.61 | less than 0.001 | 39 percent |
| Doubling of serum creatinine | 0.56 | less than 0.001 | 44 percent |
| Renal replacement therapy initiation | 0.45 | 0.04 | 55 percent |
| A1C reduction | 0.5 to 0.6 percent | — | — |
| Weight loss vs placebo | ~2 kg | — | — |
| Systolic BP reduction | ~3 mmHg | — | — |
Why the Mortality Benefit Was Surprising
Several aspects of the EMPA-REG findings shocked the field:
- Mortality reduction at 32 percent — no glucose-lowering drug had ever shown this magnitude of all-cause mortality benefit in type 2 diabetes
- Speed of effect: The mortality and heart failure curves separated within 3 to 6 months, far faster than any plausible glucose-mediated effect (which takes years to accumulate, per UKPDS)
- Heart failure benefit: Heart failure was not even a pre-specified focus of the trial; the 35 percent reduction in HF hospitalizations was discovered in the analyses
- Modest A1C reduction: Only ~0.5 percentage point — too little to explain the CV benefit
- Stroke trend wrong direction: Numerically more strokes in the empagliflozin arm (not statistically significant) — an unresolved finding that has not been fully replicated in other SGLT2i trials
Proposed Mechanisms of Benefit
The disconnect between modest glucose lowering and large CV mortality benefit prompted intense mechanistic study. Current thinking attributes the benefit to a combination:
- Osmotic diuresis and natriuresis: SGLT2 inhibitors cause modest glucose-and-sodium loss in urine, reducing preload and afterload — beneficial for heart failure
- Blood pressure reduction: ~3 mmHg systolic without reflex tachycardia
- Weight loss: ~2 kg sustained
- Ketone metabolism: Mild increase in beta-hydroxybutyrate may improve cardiac energetics (the “thrifty substrate” hypothesis)
- Renal hemodynamics: Restoration of tubuloglomerular feedback reduces hyperfiltration and intraglomerular pressure
- Reduced cardiac fibrosis and remodeling in animal models
- Reduced sympathetic nervous system activity
- Anti-inflammatory effects on cardiac and renal tissues
The Follow-on Trials
EMPA-REG OUTCOME launched a wave of confirmatory and expanding trials:
- CANVAS (2017): Canagliflozin reduced 3-point MACE by 14 percent in type 2 diabetes with CV disease or high CV risk
- DECLARE-TIMI 58 (2018): Dapagliflozin reduced HF hospitalization in broader type 2 diabetes population
- DAPA-HF (2019): Dapagliflozin reduced HF events in heart failure with reduced ejection fraction (HFrEF) — with or without diabetes
- EMPEROR-Reduced (2020): Empagliflozin confirmed HFrEF benefit
- DAPA-CKD (2020): Dapagliflozin reduced renal events in chronic kidney disease — with or without diabetes
- EMPEROR-Preserved (2021): Empagliflozin reduced HF events in heart failure with preserved ejection fraction (HFpEF)
- EMPA-KIDNEY (2022): Empagliflozin extended renal benefit to broader CKD population
The class effect is now firmly established. SGLT2 inhibitors are cardio-renal-protective drugs that happen to also lower glucose modestly, not glucose-lowering drugs that happen to also help the heart and kidneys.
Safety Profile
| Adverse Event (Empagliflozin vs Placebo) | Empagliflozin | Placebo |
|---|---|---|
| Genital infection | 6.4 percent | 1.8 percent |
| Urinary tract infection | 18.0 percent | 18.1 percent (NS) |
| Diabetic ketoacidosis | ~0.07 percent | ~0.06 percent (rare) |
| Acute kidney injury | 1.0 percent | 1.6 percent (lower with empa) |
| Volume depletion / dehydration | 5.1 percent | 4.9 percent |
| Fracture | 3.8 percent | 3.9 percent (NS) |
| Severe hypoglycemia | 1.3 percent | 1.5 percent (NS) |
Genital infections (mostly mild yeast in both men and women) are the most common adverse event and the most common reason patients discontinue. DKA is rare but documented — clinicians should counsel patients about euglycemic DKA risk especially with sick days, surgery, or low-carb diets.
How EMPA-REG Changed Care
- ADA Standards of Care: SGLT2 inhibitors are now recommended in type 2 diabetes patients with established CV disease, heart failure, or CKD — independent of A1C and metformin status
- Cardiology and nephrology adoption: SGLT2 inhibitors are now used by cardiologists for HF and by nephrologists for CKD — even in patients without diabetes
- Mechanism-of-action thinking: Diabetes care now centers on cardio-renal-protective drug classes rather than just glucose lowering
- FDA label expansions: Empagliflozin, dapagliflozin, and canagliflozin now carry indications for CV risk reduction, HF, and/or CKD beyond diabetes
- Quadruple therapy for HFrEF: SGLT2 inhibitor is one of the four pillars (ACE/ARB/ARNI, beta-blocker, MRA, SGLT2i)
What EMPA-REG Did Not Show
- It did not test patients without established CV disease — DECLARE-TIMI 58 and CANVAS later extended findings to higher-risk primary prevention
- It did not demonstrate stroke benefit (trend was actually unfavorable, unresolved)
- It did not test combination with GLP-1 receptor agonists (later trials suggest additive benefit)
- It did not directly compare SGLT2 inhibitor versus GLP-1 RA head-to-head
Implications for Patients and Clinicians
The EMPA-REG message for patients with type 2 diabetes and CV disease, heart failure, or CKD is direct: an SGLT2 inhibitor is among the most cardio-protective and reno-protective drugs available — likely on par with statins and ACE inhibitors. For clinicians, EMPA-REG marked a paradigm shift. The question is no longer just “what is your A1C?” — it is “what are your cardiovascular and renal risks, and which drug class addresses them?” Many patients now do well on a metformin plus SGLT2 inhibitor backbone, sometimes adding a GLP-1 RA for additional CV benefit and weight loss.
Related Reading
See our broader guides on diabetes treatment, the UKPDS trial, the ACCORD trial, and the Look AHEAD trial.
The Bottom Line
The EMPA-REG OUTCOME trial showed that empagliflozin reduced cardiovascular mortality by 38 percent, all-cause mortality by 32 percent, heart failure hospitalization by 35 percent, and major renal events by 39 percent in patients with type 2 diabetes and established cardiovascular disease — at modest 0.5 percent A1C reduction. The mortality benefit emerged within months and could not be explained by glucose lowering alone, implicating hemodynamic, diuretic, ketone, and renal mechanisms. EMPA-REG launched the SGLT2 inhibitor era, and follow-on trials (CANVAS, DECLARE, DAPA-HF, EMPEROR, DAPA-CKD, EMPEROR-Preserved, EMPA-KIDNEY) established cardio-renal protection as a class effect. SGLT2 inhibitors are now recommended in ADA Standards of Care for type 2 diabetes patients with established CV disease, heart failure (any ejection fraction), or chronic kidney disease — independent of A1C — and have transformed the diabetes treatment paradigm from glucose-focused to organ-focused.
