EMPEROR-Preserved Results in Detail: Numbers, Subgroups, and Time Course

At a glance

| Parameter | Detail | |-----------|--------| | N | 5,988 (empagliflozin 2,997; placebo 2,991) | | Intervention | Empagliflozin 10 mg once daily | | Comparator | Matching placebo | | Duration | Median 26.2 months | | Primary endpoint | Time to first CV death or HF hospitalization | | Key result | HR 0.79 (95% CI 0.69, 0.90), p <0.001 |

Trial Design and Population

EMPEROR-Preserved enrolled adults with chronic heart failure, NYHA class II, IV symptoms, and a left ventricular ejection fraction (LVEF) >40%. Patients required elevated NT-proBNP levels (≥300 pg/mL, or ≥900 pg/mL if atrial fibrillation was present) and evidence of structural heart disease. The trial was event-driven, requiring 841 primary endpoint events for adequate power (Anker et al., NEJM 2021).

Mean age was 71.9 years. Median LVEF sat at 54% (IQR 49 to 60%). About 45% of participants were female. Approximately 33% had diabetes at baseline, and roughly 36% carried atrial fibrillation. Mean eGFR was 60.6 mL/min/1.73 m².

Background therapy reflected contemporary practice: 80% on renin-angiotensin system inhibitors, 86% on beta-blockers, 37% on mineralocorticoid receptor antagonists.

Primary Endpoint: Composite of CV Death or HF Hospitalization

The primary composite occurred in 415 patients (13.8%) in the empagliflozin arm versus 511 (17.1%) on placebo. The hazard ratio was 0.79 (95% CI 0.69, 0.90, p <0.001), representing an absolute risk reduction of 3.3 percentage points over the trial period (Anker et al., NEJM 2021).

The number needed to treat (NNT) to prevent one primary endpoint event over the median follow-up: 31 patients.

Component Breakdown

| Endpoint | Empagliflozin (n, rate/100 pt-yr) | Placebo (n, rate/100 pt-yr) | HR (95% CI) | |----------|-----------------------------------|-----------------------------|-------------| | CV death or HF hospitalization | 415, 7.3 | 511, 8.9 | 0.79 (0.69, 0.90) | | First HF hospitalization | 259, 4.5 | 352, 6.1 | 0.71 (0.60, 0.83) | | CV death | 219, 3.4 | 244, 3.8 | 0.91 (0.76, 1.09) | | Total HF hospitalizations (recurrent) | 407 | 541 | RR 0.73 (0.61, 0.88) | | All-cause mortality | 422 | 427 | HR 1.00 (0.87, 1.15) |

The signal was clear: empagliflozin's benefit came from suppressing HF hospitalizations. The 29% relative reduction in first HF hospitalization (HR 0.71) was highly significant. Cardiovascular death alone did not reach statistical significance (HR 0.91, p = 0.34), and all-cause mortality was neutral.

Time-Course of Benefit

Kaplan-Meier curves for the primary endpoint began separating within the first 3 months and continued diverging throughout follow-up. By 12 months, the absolute difference was approximately 1.5 percentage points. By 24 months it exceeded 3 percentage points (Anker et al., NEJM 2021).

This early and sustained separation pattern mirrors what was seen in EMPEROR-Reduced (the companion trial in HFrEF) and in DAPA-HF with dapagliflozin. The SGLT2 inhibitor class appears to exert rapid hemodynamic and volume-related effects that translate into early event reduction, distinct from the slower remodeling pathways of neurohormonal blockade.

Secondary Endpoints and Supportive Analyses

eGFR Slope

The annual rate of decline in eGFR was significantly slower with empagliflozin: −1.25 mL/min/1.73 m²/year versus −2.62 mL/min/1.73 m²/year on placebo (between-group difference +1.36 mL/min/1.73 m²/year, p <0.001). This renal preservation signal has been reproduced across SGLT2i trials in heart failure and diabetic kidney disease, consistent with data from EMPA-KIDNEY (2023).

Kansas City Cardiomyopathy Questionnaire (KCCQ)

At week 52, empagliflozin produced a statistically significant 1.32-point greater improvement in KCCQ Clinical Summary Score compared with placebo (95% CI 0.45, 2.19). While modest in absolute terms, a higher proportion of empagliflozin patients achieved clinically meaningful improvements (≥5-point gain): 31.5% versus 28.6%.

Composite of Renal Events

The prespecified renal composite (sustained ≥40% decline in eGFR, renal replacement therapy, or renal death) showed a numerical trend favoring empagliflozin but did not reach significance in this population (HR 0.95, 95% CI 0.73, 1.24). The trial was not powered for this endpoint.

Subgroup Analyses

A defining feature of EMPEROR-Preserved was the consistency of effect across subgroups. The following table summarizes key interactions:

| Subgroup | HR (95% CI) | p-interaction | |----------|-------------|---------------| | LVEF 41 to 49% (HFmrEF) | 0.71 (0.57, 0.88) | 0.21 | | LVEF ≥50% (true HFpEF) | 0.83 (0.71, 0.98) |, | | LVEF ≥60% | 0.87 (0.69, 1.10) |, | | Diabetes | 0.79 (0.67, 0.94) | 0.95 | | No diabetes | 0.78 (0.64, 0.95) |, | | eGFR ≥60 | 0.78 (0.64, 0.94) | 0.63 | | eGFR <60 | 0.80 (0.67, 0.96) |, | | Female | 0.73 (0.59, 0.90) | 0.26 | | Male | 0.83 (0.70, 0.98) |, | | Age ≥75 years | 0.74 (0.62, 0.89) | 0.17 | | Age <75 years | 0.84 (0.70, 1.01) |, |

The p-values for interaction were uniformly non-significant, indicating no clear effect modification. One notable finding: the point estimate appeared slightly more favorable at lower EF ranges within the trial (41 to 49%), consistent with a pooled analysis published later in Packer et al., Circulation 2021 that combined EMPEROR-Preserved and EMPEROR-Reduced data.

Safety Profile

Empagliflozin was well tolerated. Discontinuation rates were similar between groups (17.6% vs 17.4%). Key safety observations:

| Event | Empagliflozin | Placebo | |-------|---------------|---------| | Urinary tract infections | 9.9% | 8.1% | | Genital infections | 2.2% | 0.7% | | Hypotension | 10.4% | 8.6% | | Diabetic ketoacidosis | 0 events | 0 events | | Lower limb amputations | No signal |, |

Genital mycotic infections were the only adverse event with a clear excess, consistent with the known SGLT2i class effect. The FDA prescribing information for empagliflozin lists genital infections as an identified risk and recommends monitoring.

Limitations and Interpretive Caution

The investigators acknowledged several limitations:

1. No mortality benefit. The trial was powered for the composite, not for death alone. All-cause mortality was flat (HR 1.00), which tempers the clinical enthusiasm for patients whose primary risk is death rather than hospitalization.

2. EF >60% attenuation. The point estimate above 60% (HR 0.87, CI crossing 1.0) raised questions about whether the drug's hemodynamic mechanism has a ceiling in truly stiff, non-dilated hearts.

3. NT-proBNP enrichment. The biomarker threshold excluded patients with mild symptoms and low natriuretic peptides. Generalizability to community-diagnosed HFpEF with lower biomarker levels remains uncertain.

4. Short follow-up for renal outcomes. At 26 months, the renal composite was underpowered. Longer-term data from EMPA-KIDNEY addressed this gap for CKD populations.

5. Diuretic co-management. Protocol-mandated diuretic adjustment was allowed; the trial cannot isolate the drug's effect from optimization of loop diuretics.

Clinical Translation

The 2022 AHA/ACC/HFSA Heart Failure Guidelines incorporated EMPEROR-Preserved data to issue a Class 2a recommendation for SGLT2 inhibitors in HFpEF. This was a major shift. Before this trial, no drug class had demonstrated reliable reduction of HF events in patients with EF >40%.

In practice, empagliflozin 10 mg can be initiated regardless of diabetes status, does not require dose titration, and can be added on top of existing neurohormonal therapy. Monitoring for volume depletion and genital infections is the primary clinical concern.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Anker SD, Butler J, Filippatos G, et al. Empagliflozin in Heart Failure with a Preserved Ejection Fraction. N Engl J Med. 2021;385(16):1451-1461. PubMed
2. Packer M, Butler J, Zannad F, et al. Empagliflozin and Major Renal Outcomes in Heart Failure. Circulation. 2021;144(16):1284-1294. PubMed
3. The EMPA-KIDNEY Collaborative Group. Empagliflozin in Patients with Chronic Kidney Disease. N Engl J Med. 2023;388(2):117-127. PubMed
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. PubMed
5. FDA. Jardiance (empagliflozin) Prescribing Information. Revised 2022. FDA Label