Is the use of SGLT2 (sodium-glucose cotransporter 2) inhibitors warranted in both heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction (HFrEF), particularly in patients with comorbid diabetes mellitus?

SGLT2 Inhibitors in Heart Failure: Use Across the Ejection Fraction Spectrum

Yes, SGLT2 inhibitors are strongly warranted in both heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction (HFrEF), regardless of diabetes status, with Class I evidence for HFrEF and Class 2a evidence for HFpEF. 1, 2

Evidence-Based Recommendations by Heart Failure Type

HFrEF (EF ≤40%): Foundational Therapy

• SGLT2 inhibitors receive a Class I, Level A recommendation for all patients with HFrEF (NYHA class II-IV) to reduce cardiovascular death and heart failure hospitalizations, independent of diabetes status. 2, 3
• Dapagliflozin reduces the composite outcome of worsening heart failure or cardiovascular death by 26% (HR 0.74,95% CI 0.65-0.85) in HFrEF patients. 2
• Empagliflozin reduces cardiovascular death or hospitalization for worsening heart failure by 25% (HR 0.75,95% CI 0.65-0.86) in HFrEF. 2
• These agents should be initiated as cornerstone therapy alongside ACE inhibitors/ARNIs, beta-blockers, and mineralocorticoid receptor antagonists. 2

HFpEF (EF >40%): Proven Benefit

• SGLT2 inhibitors receive a Class 2a, Level B-R recommendation for HFpEF to decrease heart failure hospitalizations and cardiovascular mortality. 2, 3
• The EMPEROR-Preserved trial demonstrated empagliflozin reduced the composite of cardiovascular death or hospitalization for heart failure by 21% (HR 0.79,95% CI 0.69-0.90) in patients with LVEF >40%. 1
• The effects were consistent in patients with or without diabetes—approximately 50% of EMPEROR-Preserved subjects had type 2 diabetes at baseline. 1
• Dapagliflozin reduced the composite of worsening heart failure or cardiovascular death by 18% in the DELIVER trial for HFpEF patients. 2

Clinical Implementation Algorithm

Step 1: Identify Eligible Patients

• All patients with symptomatic heart failure (NYHA class II-IV) regardless of ejection fraction should be considered for SGLT2 inhibitor therapy. 2, 4
• Check eGFR: Must be ≥25 mL/min/1.73 m² for dapagliflozin or ≥30 mL/min/1.73 m² for empagliflozin. 2, 5
• Diabetes status is irrelevant—benefits occur with and without diabetes. 1, 2

Step 2: Choose the Agent

• Use dapagliflozin 10 mg daily or empagliflozin 10 mg daily—these are the only SGLT2 inhibitors with dedicated heart failure outcome trials showing benefit across the EF spectrum. 2
• Ertugliflozin lacks evidence from dedicated heart failure trials and should not be used. 2
• No dose titration is required—the dose is fixed at 10 mg daily for all indications. 2, 5

Step 3: Timing of Initiation

• Initiate early, ideally during hospitalization in stabilized patients (no IV diuretics for 6 hours, no IV vasodilators/inotropes for 24 hours). 2
• Benefits occur within weeks of initiation, with empagliflozin showing 58% relative risk reduction at just 12 days. 2
• Do not wait for "optimal" glycemic control or background therapy optimization—start immediately. 2

Step 4: Monitoring and Continuation

• Expect a mild, transient eGFR decline of 3-5 mL/min/1.73 m² within the first 1-4 weeks—this is expected and provides long-term kidney protection. 2, 5
• Do not discontinue for mild eGFR decline unless there are signs of hypovolemia. 2
• If eGFR falls below 25 mL/min/1.73 m² during treatment, continue 10 mg daily until dialysis is required. 5

Key Advantages Unique to SGLT2 Inhibitors

• No dose titration required, unlike ACE inhibitors, ARBs, or beta-blockers. 2
• No significant effect on blood pressure, heart rate, or potassium levels, making them safe to combine with other guideline-directed medical therapy. 2
• Benefits are maintained regardless of age, sex, background medical therapy, and diabetes status. 2
• Can be used with eGFR as low as 20-30 mL/min/1.73 m² for dapagliflozin. 2

Comparison with Other Diabetes Medications in Heart Failure

Avoid in Heart Failure

• Thiazolidinediones (TZDs) are contraindicated in all heart failure patients—they increase heart failure risk through volume expansion. 1
• Certain DPP-4 inhibitors (except sitagliptin) may increase heart failure hospitalization risk and should be avoided. 1

Neutral Effect

• GLP-1 receptor agonists show no effect on heart failure hospitalization but have cardiovascular benefits in other domains. 1
• Avoid GLP-1 receptor agonists if recent heart failure decompensation. 1

Superior Choice

• SGLT2 inhibitors are the only glucose-lowering agents that consistently reduce heart failure hospitalizations by 27-39% across multiple trials. 1, 2

Common Pitfalls to Avoid

• Do not discontinue SGLT2 inhibitors solely because eGFR falls below 45 mL/min/1.73 m²—cardiovascular and renal protective benefits persist even when glycemic efficacy is lost. 2, 5
• Do not withhold SGLT2 inhibitors in patients without diabetes—the benefits are independent of diabetes status. 1, 2
• Do not delay initiation waiting for optimization of other heart failure medications—start immediately. 2
• Withhold SGLT2 inhibitors during acute illness with reduced oral intake, fever, vomiting, or diarrhea to prevent euglycemic diabetic ketoacidosis. 5
• Withhold at least 3 days before major surgery or procedures requiring prolonged fasting. 5

Safety Considerations

• Hypotension may occur in approximately 5.7% of patients, especially those with volume depletion—assess volume status before initiation. 2, 5
• Genital mycotic infections occur in 1.5-6% of patients but are generally manageable. 1, 2
• Euglycemic diabetic ketoacidosis risk is significantly lower in non-diabetic populations but counsel all patients on sick day rules. 2, 5

Strength of Evidence Hierarchy

• Highest quality evidence (Class I, Level A) exists for HFrEF based on DAPA-HF and EMPEROR-Reduced trials. 2
• Strong evidence (Class 2a, Level B-R) exists for HFpEF based on EMPEROR-Preserved and DELIVER trials. 2, 3
• The evidence quality is consistently high across both diabetes and non-diabetes populations. 1, 2