Jardiance Evidence Base Graded by GRADE: What the Trials Actually Show

What GRADE Means and Why It Matters for Empagliflozin

GRADE (Grading of Recommendations Assessment, Development, and Evaluation) rates both the quality of evidence and the strength of the recommendation built on that evidence. Four certainty levels apply: High, Moderate, Low, and Very Low. Randomized controlled trials start at High and are downgraded for risk of bias, inconsistency, indirectness, imprecision, or publication bias.

Empagliflozin has been studied in four major Phase III cardiovascular-renal outcome trials enrolling more than 23,000 patients. That volume, combined with consistent directional effects across independent trial programs, means most of its headline endpoints survive GRADE scrutiny at the High level. The sections below apply the GRADE framework domain by domain.

Why GRADE starts high for RCT evidence

Regulatory-grade outcome trials are double-blind, placebo-controlled, and powered for hard clinical endpoints. The EMPA-REG OUTCOME, EMPEROR-Reduced, EMPEROR-Preserved, and EMPA-KIDNEY trials all meet these criteria. Starting certainty is therefore High before any downgrading factors are considered.

Where downgrading occurs

Downgrading occurs when a result is imprecise (wide confidence interval crossing no effect), when the trial population does not match clinical practice (indirectness), or when results conflict across trials (inconsistency). For empagliflozin, the glycemic endpoint is the area most affected by indirectness, because surrogate outcomes (HbA1c) predict but do not guarantee clinical benefit.

EMPA-REG OUTCOME: Cardiovascular Death in Type 2 Diabetes

EMPA-REG OUTCOME enrolled 7,020 adults with type 2 diabetes and established cardiovascular disease across 42 countries. Participants received empagliflozin 10 mg, empagliflozin 25 mg, or placebo on top of standard care. The primary endpoint was major adverse cardiovascular events (MACE): CV death, nonfatal myocardial infarction, or nonfatal stroke. The trial results were published in the New England Journal of Medicine in 2015.

Primary MACE endpoint

The trial met its primary endpoint. Empagliflozin reduced the rate of MACE by 14% relative to placebo (hazard ratio 0.86, 95% CI 0.74 to 0.99; P<0.001 for noninferiority, P=0.04 for superiority) [1]. The number needed to treat for MACE over 3.1 years was 63.

The cardiovascular death signal

The CV death reduction drove the result. Empagliflozin cut CV mortality by 38% relative to placebo (HR 0.62, 95% CI 0.49 to 0.77; P<0.001) [1]. All-cause mortality fell by 32% (HR 0.68, 95% CI 0.57 to 0.82; P<0.001). These effect sizes were larger than most investigators anticipated, and they appeared within the first few months of the trial, suggesting a hemodynamic mechanism rather than an atherosclerosis-modifying one.

Heart failure hospitalization as a secondary endpoint

Hospitalization for heart failure was reduced by 35% (HR 0.65, 95% CI 0.50 to 0.85; P<0.001) despite heart failure not being an inclusion criterion [1]. This finding seeded the dedicated heart failure trial program that followed.

GRADE rating for CV death: High

No downgrading factors apply to the CV death endpoint. The trial was adequately powered, blinded, and consistently replicated in pre-specified subgroups. The confidence interval is precise and does not cross the null. GRADE certainty: High.

EMPEROR-Reduced: Heart Failure with Reduced Ejection Fraction

EMPEROR-Reduced enrolled 3,730 patients with heart failure and left ventricular ejection fraction (LVEF) <40%, with or without type 2 diabetes. The trial was published in the New England Journal of Medicine in 2020. Full trial record available via PubMed.

Primary endpoint result

Empagliflozin 10 mg daily reduced the composite of CV death or hospitalization for heart failure by 25% relative to placebo (HR 0.75, 95% CI 0.65 to 0.86; P<0.001) [2]. The benefit was consistent across participants with and without diabetes.

Rate of total heart failure hospitalizations

Total (first plus recurrent) heart failure hospitalizations fell by 30% (HR 0.70, 95% CI 0.58 to 0.85; P<0.001) [2]. The kidney function decline rate was also slower in the empagliflozin group, a secondary finding later confirmed in the dedicated CKD trial.

GRADE rating for HFrEF: High

The population closely matches the labeled indication. The confidence interval is precise. Consistent effects were seen across diabetes status, age, LVEF category within the inclusion range, and baseline kidney function. GRADE certainty: High.

EMPEROR-Preserved: Heart Failure with Preserved Ejection Fraction

EMPEROR-Preserved enrolled 5,988 patients with heart failure and LVEF >40%. It was the first large outcomes trial to demonstrate a pharmacologic benefit specifically in heart failure with preserved ejection fraction (HFpEF). Results were published in the New England Journal of Medicine in 2021. See the primary publication.

Primary endpoint and the HFpEF breakthrough

Empagliflozin reduced the composite of CV death or hospitalization for heart failure by 21% relative to placebo (HR 0.79, 95% CI 0.69 to 0.90; P<0.001) [3]. This was the first pharmacologic agent to show statistically significant benefit on this endpoint in patients with LVEF above 40%, an area where spironolactone and other agents had previously failed.

Subgroup consistency

Benefit was consistent across sex, diabetes status, and LVEF subgroups spanning 41% to 80% and above. The 2022 FDA approval for HFpEF and heart failure with mildly reduced ejection fraction (HFmrEF) was based on these data. FDA prescribing information for empagliflozin.

GRADE rating for HFpEF: High

The trial is the only large Phase III RCT with a positive result in this population. A single trial in principle warrants scrutiny, but the sample size (N=5,988), the precise confidence interval, and biologically plausible mechanisms consistent with HFrEF results support High certainty. GRADE certainty: High.

EMPA-KIDNEY: Chronic Kidney Disease Progression

EMPA-KIDNEY enrolled 6,609 adults with CKD across a wider eGFR range than previous trials, from eGFR as low as 20 mL/min/1.73 m² up to 45, plus patients with eGFR 45 to 90 with albuminuria. The trial was stopped early for efficacy and published in the New England Journal of Medicine in 2023. Primary publication.

Primary composite endpoint

Empagliflozin reduced the primary composite of kidney disease progression or CV death by 28% (HR 0.72, 95% CI 0.64 to 0.82; P<0.001) [4]. Kidney disease progression was defined as sustained >40% decline in eGFR, initiation of dialysis, kidney transplant, or death from kidney disease.

Benefit at very low eGFR

Benefit was consistent even in patients with eGFR 20 to 45 mL/min/1.73 m², a group previously excluded from most SGLT2 inhibitor trials. This substantially broadens the clinically applicable population. About 45% of enrolled participants did not have type 2 diabetes, confirming that the kidney benefit is independent of glucose-lowering.

GRADE rating for CKD progression: High

Early stopping for benefit can inflate effect estimates (a reason to downgrade for imprecision), but the confidence interval remains precise and the stopping was guided by pre-specified rules. Consistency across subgroups without diabetes, with low eGFR, and with varying albuminuria levels prevents downgrading for inconsistency. GRADE certainty: High.

Glycemic Efficacy: HbA1c Reduction in Type 2 Diabetes

Empagliflozin lowers HbA1c approximately 0.5% to 0.8% from baseline when added to existing therapy at the 10 mg dose, and approximately 0.6% to 0.9% at 25 mg, based on the Phase III registration program reviewed by the FDA. FDA drug approval for empagliflozin. The GRADE analysis for this endpoint differs materially from the cardiovascular and renal endpoints.

Why glycemic evidence is rated Moderate, not High

HbA1c is a validated surrogate for microvascular complications, but it is not itself a clinical outcome. GRADE requires downgrading when evidence relies primarily on surrogates unless the surrogate has been validated against hard outcomes in the same drug class. For SGLT2 inhibitors, the cardiovascular data came from a population with established CVD, creating indirectness when the HbA1c endpoint is applied to a lower-risk newly diagnosed population. One level of downgrading for indirectness is appropriate. GRADE certainty: Moderate.

Comparison with other antidiabetic agents

The 2024 ADA Standards of Care assign empagliflozin a Class I recommendation for patients with type 2 diabetes and established CVD or high CVD risk, placing it ahead of agents with glycemic-only evidence. ADA Standards of Medical Care in Diabetes 2024. The glycemic benefit is a secondary consideration in patients who meet criteria for the cardiovascular or renal indications.

Guideline Positions and Strength of Recommendation

The following framework summarizes how major guidelines translate the GRADE evidence into actionable recommendations as of 2024 to 2025.

ADA 2024 Standards of Care

The American Diabetes Association 2024 Standards of Care recommend empagliflozin (or another SGLT2 inhibitor with proven CV benefit) as a preferred add-on agent in type 2 diabetes with established atherosclerotic cardiovascular disease, heart failure, or CKD regardless of baseline HbA1c or metformin use [5]. The recommendation is graded Level A (highest evidence level in ADA nomenclature), based on multiple large RCTs. The guideline states: "For patients with T2D and established CVD or indicators of high CVD risk, an SGLT2 inhibitor with demonstrated CV benefit is recommended to reduce the risk of major adverse cardiovascular events and/or hospitalization for heart failure." ADA 2024 full text.

ESC 2023 Guidelines on Diabetes and Cardiovascular Disease

The European Society of Cardiology 2023 guidelines on diabetes, pre-diabetes, and cardiovascular diseases recommend SGLT2 inhibitors (including empagliflozin) as first-line agents in patients with type 2 diabetes and established CVD or multiple risk factors, with a Class I, Level A recommendation. ESC 2023 guidelines. For heart failure with reduced ejection fraction, SGLT2 inhibitors receive a Class I, Level A recommendation regardless of diabetes status.

KDIGO 2022 CKD Guidelines

The Kidney Disease: Improving Global Outcomes 2022 guideline recommends SGLT2 inhibitors for adults with type 2 diabetes and CKD with eGFR >20 mL/min/1.73 m² and urinary albumin-to-creatinine ratio >200 mg/g. KDIGO CKD 2022 summary. The EMPA-KIDNEY data, published after the 2022 guideline, are expected to expand this recommendation in the next update to include patients without diabetes and those with lower albuminuria thresholds.

Mechanism, Safety, and Practical Prescribing Considerations

Mechanism relevant to outcomes

Empagliflozin blocks the sodium-glucose cotransporter-2 in the proximal tubule, reducing renal glucose reabsorption by approximately 60 to 80 grams per day and lowering plasma glucose independent of insulin. The cardiovascular and renal benefits appear driven by reductions in cardiac preload and afterload (via osmotic diuresis and natriuresis), a reduction in intraglomerular pressure, and possible direct cardiac and renal energetic effects. These mechanisms are well-described in a 2018 NEJM review of SGLT2 inhibitor pharmacology. See mechanistic review.

Key safety signals from the outcome trials

Genital mycotic infections occurred in approximately 6% to 10% of empagliflozin-treated patients versus 1% to 3% with placebo across the major trials [1][2][3][4]. Diabetic ketoacidosis (DKA) was rare but real in type 2 diabetes populations, with an absolute rate below 0.5% per year. Volume depletion events were more common with empagliflozin, particularly in patients over 75 or on loop diuretics. The Fournier gangrene signal (necrotizing fasciitis of the perineum) carries an FDA black-box warning, though absolute incidence across the trial program was below 1 per 10,000 patient-years. FDA safety communication on SGLT2 inhibitors and Fournier gangrene.

Dose selection in clinical practice

The standard starting dose for type 2 diabetes is 10 mg orally once daily in the morning, with or without food. Uptitration to 25 mg is available for additional glycemic control but did not produce a meaningfully different cardiovascular signal compared with 10 mg in EMPA-REG OUTCOME. For heart failure and CKD, 10 mg once daily is the approved dose; 25 mg is not indicated in those settings. Empagliflozin should not be initiated if eGFR is below 20 mL/min/1.73 m², though it may be continued below 45 in patients already on the drug.

Evidence Gaps and GRADE Low / Very Low Areas

Not every question about empagliflozin has High-quality evidence. Three areas currently carry GRADE Low or Very Low certainty.

Primary prevention of cardiovascular events

EMPA-REG OUTCOME enrolled only patients with established CVD. No large RCT has enrolled type 2 diabetes patients at lower cardiovascular risk as the primary population. Applying the EMPA-REG OUTCOME results to a 45-year-old with newly diagnosed type 2 diabetes and no vascular disease is indirect. GRADE certainty for primary prevention: Low.

Type 1 diabetes

Empagliflozin is not FDA-approved for type 1 diabetes. Trials in type 1 populations showed glycemic benefit but also a disproportionately elevated DKA rate (approximately 2.5-fold higher than placebo), which caused the FDA to decline approval. FDA complete response letter context. GRADE certainty for type 1 diabetes use: Very Low for net clinical benefit.

Combination with GLP-1 receptor agonists

Large RCTs comparing dual SGLT2-inhibitor-plus-GLP-1-RA therapy against either agent alone for hard cardiovascular or renal endpoints do not yet exist. The DURATION-8 trial (N=695) showed additive glycemic effects for exenatide plus dapagliflozin, but it was not powered for clinical outcomes. DURATION-8 citation. Combination use is common in clinical practice but supported only by GRADE Low evidence for outcomes beyond glycemia.

Summary GRADE Table for Empagliflozin

| Endpoint | Key Trial | N | Effect Size (HR or RR) | GRADE Certainty | |---|---|---|---|---| | CV death (T2D + established CVD) | EMPA-REG OUTCOME | 7,020 | HR 0.62 (95% CI 0.49 to 0.77) | High | | MACE (T2D + established CVD) | EMPA-REG OUTCOME | 7,020 | HR 0.86 (95% CI 0.74 to 0.99) | High | | CV death or HF hosp. (HFrEF) | EMPEROR-Reduced | 3,730 | HR 0.75 (95% CI 0.65 to 0.86) | High | | CV death or HF hosp. (HFpEF) | EMPEROR-Preserved | 5,988 | HR 0.79 (95% CI 0.69 to 0.90) | High | | CKD progression or CV death | EMPA-KIDNEY | 6,609 | HR 0.72 (95% CI 0.64 to 0.82) | High | | HbA1c reduction (T2D, surrogate) | Registration program | Various | 0.5 to 0.9% reduction | Moderate | | Primary CV prevention (no CVD hx) | No dedicated RCT | N/A | Extrapolated | Low | | Net benefit in type 1 diabetes | Small phase II RCTs | <1,000 | Elevated DKA risk | Very Low |