DAPA-HF Trial: A Plain-English Overview of What It Established

At a glance

Field	Detail
Trial name	DAPA-HF
N	4,744
Intervention	Dapagliflozin 10 mg once daily
Comparator	Matched placebo
Duration	Median 18.2 months
Primary endpoint	Composite of worsening HF (hospitalization or urgent visit) or CV death
Key result	HR 0.74 (95% CI 0.65, 0.85), p <0.001, absolute risk reduction ~5%
Published	NEJM, 2019

---

What Question Was DAPA-HF Actually Trying to Answer?

Before 2019, SGLT2 inhibitors had earned attention largely because cardiovascular outcomes trials like EMPA-REG OUTCOME showed mortality benefits in people with type 2 diabetes at high cardiovascular risk. The assumption was that glucose lowering, or at least the metabolic effects tied to diabetes, drove those gains.

DAPA-HF was designed to challenge that assumption head-on. The central question was whether dapagliflozin, a sodium-glucose cotransporter-2 inhibitor that blocks glucose reabsorption in the proximal tubule, would reduce serious heart failure events and cardiovascular death in patients who already had symptomatic heart failure with reduced ejection fraction, regardless of whether they had type 2 diabetes. About 45% of participants did not have diabetes at baseline. That deliberate design choice is what makes the primary publication stand apart from earlier SGLT2i outcomes data.

---

Who Was Enrolled, and Why That Matters

Understanding the enrollment criteria is essential to knowing how far the results travel.

Eligible patients were adults with established, symptomatic heart failure. Key eligibility thresholds included:

• Left ventricular ejection fraction (LVEF): ≤40%
• NYHA functional class: II, III, or IV
• NT-proBNP: ≥600 pg/mL; or ≥400 pg/mL if hospitalized for HF within 12 months; or ≥900 pg/mL with atrial fibrillation or flutter
• Background therapy: Optimized guideline-directed medical therapy (GDMT), meaning patients were expected to be on ACE inhibitors, ARBs, or sacubitril/valsartan, plus beta-blockers and mineralocorticoid receptor antagonists unless intolerant

The NT-proBNP threshold is worth noting. It functioned as a biological severity filter, excluding lower-risk patients who might dilute any treatment signal. The requirement for optimized GDMT also means the trial was asking whether dapagliflozin added value on top of a best-available background, not in a GDMT vacuum.

At baseline, mean LVEF was approximately 31%, and about 68% of participants were NYHA class II. Mean age was 66 years, and roughly 23% were women. Patients with an eGFR <30 mL/min/1.73 m² were excluded, which becomes relevant when interpreting the renal subgroup data.

---

What the Intervention Actually Looked Like

Participants were randomized 1:1 to dapagliflozin 10 mg orally once daily or a matched placebo. Both groups continued their background heart failure therapy. There was no dose titration, no run-in period to select tolerant patients, and no restriction on concurrent diuretic use. The simplicity of the regimen was intentional: if dapagliflozin was going to earn a place in clinical practice, it needed to show it could layer onto complex multi-drug regimens without requiring management acrobatics.

The FDA label for dapagliflozin (Farxiga) was subsequently updated to include the HFrEF indication based substantially on this data, reflecting what the trial's randomization architecture was built to support.

---

The Primary and Secondary Endpoints in Detail

The primary composite endpoint was the first occurrence of either:

1. A worsening heart failure event (unplanned hospitalization or an urgent outpatient visit requiring IV therapy), or
2. Cardiovascular death

Secondary endpoints were hierarchically tested and included total (recurrent) hospitalizations for HF plus CV death, change in Kansas City Cardiomyopathy Questionnaire total symptom score (KCCQ-TSS), the rate of eGFR decline, and all-cause mortality.

Primary Endpoint Results

The number needed to treat (NNT) for the primary composite over a median of 18 months was approximately 21. For context, that sits favorably alongside many accepted cardiovascular therapies.

The Diabetes-Status Subgroup

This is where McMurray et al. made the most clinically consequential observation. The hazard ratio for the primary composite was 0.75 (95% CI 0.63, 0.90) in patients with diabetes and 0.73 (95% CI 0.60, 0.88) in those without. The interaction p-value was 0.80, meaning no statistically meaningful difference between the subgroups. The drug worked similarly regardless of metabolic status.

Symptom and Quality-of-Life Results

The KCCQ-TSS improved by a clinically meaningful margin. Dapagliflozin increased the odds of a ≥5-point improvement in KCCQ-TSS (OR 1.15, 95% CI 1.08, 1.23) and reduced the odds of a ≥5-point deterioration. Symptom-level data often gets less attention than survival curves, but in a condition defined by breathlessness and functional limitation, patient-reported improvements carry real clinical weight.

---

Methodological Strengths Worth Understanding

Several design choices elevated the credibility of DAPA-HF's findings.

Pre-specified hierarchical testing guarded against inflated type I error across secondary endpoints. The trial team tested outcomes in a fixed sequence, so the symptom data and mortality data were not independently fished for significance.

Event-driven design. The trial was powered to a target event count rather than a fixed calendar duration. This meant that if the event rate in the placebo group was lower than anticipated (which could happen if background GDMT was particularly effective), the trial would simply run longer.

Site diversity. Centers across 20 countries enrolled patients, which reduces the risk that results are artifacts of one healthcare system's prescribing habits or patient population.

Consistent background therapy documentation. The proportion of patients on sacubitril/valsartan (approximately 11%) was tracked. The benefit of dapagliflozin held across sacubitril/valsartan users and non-users, an early signal that SGLT2 inhibition and neprilysin inhibition address different physiological pathways.

---

Honest Limitations the Authors Acknowledged

No trial should be read without its limitations table, and the DAPA-HF investigators were reasonably candid about several constraints.

Low female enrollment. Women made up only about 23% of the cohort. Heart failure epidemiology and phenotype differ by sex, and the relative underrepresentation limits confidence that the point estimates translate cleanly to women at the same magnitude.

Limited representation of NYHA class IV. Only about 1% of participants had class IV symptoms. This is partly a practical issue (very sick patients drop out or die before endpoints accumulate), but it also means the most severely ill patients are extrapolating from a trial that mostly studied class II and III disease.

Relatively short median follow-up. Eighteen months is enough to capture recurrent hospitalizations but is not long enough to fully characterize long-term renal protection signals or late-curve separation in mortality. The DAPA-HF trial was not designed to be a renal outcomes trial.

Exclusion of eGFR <30. This cutoff was conservative relative to what later trials explored. The DAPA-CKD trial subsequently demonstrated dapagliflozin's renal benefits at lower GFR ranges, but DAPA-HF cannot speak to that population directly.

Open-label escape. If a patient's condition deteriorated significantly, investigators could introduce new therapies, potentially diluting between-group differences in hospitalization rates.

---

What Changed in Clinical Practice After DAPA-HF

The trial was published in September 2019, and the 2022 AHA/ACC/HFSA Heart Failure Guidelines gave SGLT2 inhibitors a Class I recommendation (Level of Evidence A) for patients with HFrEF, regardless of the presence of type 2 diabetes. That is the highest recommendation class the guideline system offers, reflecting the quality and consistency of evidence that DAPA-HF, alongside the EMPEROR-Reduced trial, provided.

Practically speaking, the addition of dapagliflozin to the four-pillar GDMT framework (ACE inhibitor/ARB/ARNI, beta-blocker, MRA, and SGLT2i) has become the standard conversation in heart failure clinic. Clinicians managing patients with HFrEF who do not have diabetes now have explicit trial-level support for prescribing a drug that was previously categorized as a diabetes medication.

The EMPEROR-Reduced trial with empagliflozin, published in 2020, independently replicated the core finding of DAPA-HF, reducing the same composite endpoint by 25% with a different SGLT2 inhibitor in a similar population. The concordance across two large, well-run trials in the same drug class has made the mechanism-level conclusion far more secure than either trial alone could have established.

---

Frequently asked questions

›What type of heart failure did DAPA-HF study?

DAPA-HF studied heart failure with reduced ejection fraction (HFrEF), defined as an LVEF of 40% or below. It did not enroll patients with preserved ejection fraction (HFpEF). A separate program, DELIVER, later addressed that population.

›Did patients need to have diabetes to be in the trial?

No. Approximately 45% of DAPA-HF participants did not have type 2 diabetes at baseline. This was a deliberate design choice, and the results showed the benefit was similar regardless of diabetes status.

›What was the absolute risk reduction, and why does it matter alongside the relative risk?

The absolute risk reduction for the primary composite was approximately 5 percentage points (21.2% vs 16.3%). This translates to an NNT of about 21 over 18 months. Relative risk reductions look impressive in isolation; the absolute number contextualizes what adding this drug actually means for a given patient.

›Was the mortality reduction statistically significant?

Cardiovascular death was reduced (HR 0.82, 95% CI 0.69, 0.98, p=0.029), and all-cause mortality trended in the same direction but was a secondary endpoint. The trial was not independently powered for mortality as a standalone outcome.

›How did dapagliflozin perform in patients on sacubitril/valsartan?

About 11% of participants were taking sacubitril/valsartan at baseline. The benefit of dapagliflozin was consistent in this subgroup, suggesting additive effect rather than redundancy between the two drug classes.

›What kidney function level was required to enroll?

Patients needed an eGFR of at least 30 mL/min/1.73 m² at enrollment. People below that threshold were excluded, so the trial results are not directly applicable to patients with advanced chronic kidney disease.

›Did DAPA-HF show any meaningful quality-of-life benefit?

Yes. The dapagliflozin group had significantly greater odds of a clinically meaningful improvement in KCCQ total symptom score and lower odds of symptom deterioration. This is important because managing breathlessness and functional capacity is central to heart failure care.

›What did DAPA-HF mean for the drug's regulatory status?

The FDA approved an HFrEF indication for dapagliflozin (Farxiga) in 2020, largely on the strength of this trial. The label update reflected the non-diabetes population benefit specifically.

›How does DAPA-HF compare to EMPEROR-Reduced?

Both trials showed approximately 25% relative risk reduction in the same composite endpoint using different SGLT2 inhibitors. EMPEROR-Reduced used empagliflozin and showed a numerically larger reduction in total HF hospitalizations. The two trials together form the evidential backbone for the Class I guideline recommendation covering the entire drug class in HFrEF.

›Are the DAPA-HF results generalizable to women?

With caution. Only about 23% of participants were women. The point estimates appeared consistent by sex in subgroup analyses, but the trial was not powered to independently confirm this. Wider representation of women in future HF trials remains an ongoing methodological gap.

---

References

1. McMurray JJV, Solomon SD, Inzucchi SE, et al. Dapagliflozin in patients with heart failure and reduced ejection fraction. N Engl J Med. 2019;381(21):1995 to 2008. https://pubmed.ncbi.nlm.nih.gov/31535829/

2. FDA prescribing information for dapagliflozin (Farxiga), including HFrEF indication. U.S. Food and Drug Administration. https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/202293s018lbl.pdf

3. Packer M, Anker SD, Butler J, et al. Cardiovascular and renal outcomes with empagliflozin in heart failure (EMPEROR-Reduced). N Engl J Med. 2020;383(15):1413, 1424. https://pubmed.ncbi.nlm.nih.gov/32865377/

4. Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure. J Am Coll Cardiol. 2022;79(17):e263, e421. https://pubmed.ncbi.nlm.nih.gov/35379504/

5. Heerspink HJL, Stefánsson BV, Correa-Rotter R, et al. Dapagliflozin in patients with chronic kidney disease (DAPA-CKD). N Engl J Med. 2020;383(15):1436, 1446. https://pubmed.ncbi.nlm.nih.gov/32970396/