Did the EMPEROR-Preserved trial show a mortality benefit for empagliflozin?

No. Cardiovascular death alone had a hazard ratio of 0.91 (95% CI 0.76, 1.09), which was not statistically significant. The primary composite benefit was driven by reduction in heart failure hospitalizations.

How quickly did the treatment benefit appear in EMPEROR-Preserved?

Kaplan-Meier curves separated within the first 3 months of treatment. Time-varying analyses confirmed the treatment effect was consistent from early follow-up through the full 26-month median observation period.

Does empagliflozin protect kidney function in HFpEF patients?

Yes. After an initial acute dip in eGFR (expected with SGLT2 inhibition), the chronic rate of kidney function decline was approximately 1.3 mL/min/1.73 m² per year slower with empagliflozin compared to placebo.

Were there concerning safety signals with longer follow-up?

No unexpected safety signals emerged. Genital infections were more common with empagliflozin but rarely led to discontinuation. Fractures, amputations, and ketoacidosis were not increased.

What is the EMPEROR-Pooled analysis?

This combined data from EMPEROR-Reduced and EMPEROR-Preserved (9,718 patients total) to evaluate empagliflozin across the full heart failure EF spectrum. It showed a 27% reduction in total HF hospitalizations and consistent benefit regardless of EF, diabetes status, or baseline NT-proBNP.

Does empagliflozin work better in HFmrEF than true HFpEF?

Subgroup analyses suggested numerically stronger benefit in patients with LVEF 41 to 49% (now classified as HFmrEF), but the interaction test was not significant. The trial was not powered to definitively separate these subgroups.

How does EMPEROR-Preserved compare to DELIVER (dapagliflozin)?

Both trials demonstrated SGLT2 inhibitor benefit in HFpEF. DELIVER used a higher LVEF cutoff (>40%) and confirmed similar results. Together, they established SGLT2 inhibitors as a drug class effective across the HF spectrum.

Did patients feel better on empagliflozin in this trial?

KCCQ symptom scores improved modestly with empagliflozin, but the difference (approximately 1.3 points) fell below the 5-point threshold considered clinically meaningful for individual patients. The primary value is event prevention.

Is empagliflozin FDA-approved for HFpEF?

Yes. In February 2022, the FDA expanded the Jardiance indication to include heart failure regardless of ejection fraction, based largely on EMPEROR-Preserved and EMPEROR-Reduced data.

Should empagliflozin be started early or late in HFpEF management?

Extended analyses showing early benefit separation (within 3 months) support early initiation. The 2022 AHA/ACC/HFSA guidelines give SGLT2 inhibitors a Class 2a recommendation in HFpEF, and their favorable safety profile makes them suitable as a first addition to baseline therapy.

EMPEROR-Preserved Extension Data and What Happened After the Trial Ended

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | N | 5,988 | | Intervention | Empagliflozin 10 mg daily | | Comparator | Placebo | | Duration | Median 26.2 months | | Primary endpoint | Composite of cardiovascular death or hospitalization for heart failure | | Key result | HR 0.79 (95% CI 0.69, 0.90); 21% relative risk reduction | | Registration | NCT03057951 |

Why the Original Trial Left Questions Open

The EMPEROR-Preserved primary publication reported results after a median follow-up of 26.2 months. That duration was sufficient to detect a statistically significant difference in the composite primary endpoint (HR 0.79, p < 0.001), but it raised several issues that only longer observation could address.

First, the composite was driven almost entirely by a reduction in heart failure hospitalizations. Cardiovascular death alone showed a hazard ratio of 0.91 (95% CI 0.76, 1.09), which did not reach significance. Whether mortality benefit would emerge with longer follow-up remained uncertain.

Second, the trial enrolled patients with LVEF above 40%, a population with heterogeneous disease biology. The original analysis suggested possible attenuation of benefit at higher ejection fractions (LVEF >60%), though the interaction test was not significant. Extended observation would help clarify whether this gradient was real or a statistical artifact from smaller subgroup sizes.

Third, with a median trial duration just over two years, long-term safety signals (particularly bone fractures, diabetic ketoacidosis in nondiabetic patients, and genital infections requiring discontinuation) needed more follow-up time to characterize.

The EMPEROR-Preserved Extended Analysis

Extended follow-up data from EMPEROR-Preserved were reported in several post-hoc and prespecified analyses published between 2022 and 2024. The most informative of these addressed three questions: durability of the primary endpoint benefit, the trajectory of eGFR decline, and patient-reported outcomes over time.

Durability of the Primary Composite Benefit

A prespecified analysis of time-varying treatment effects examined whether the hazard ratio for the primary composite shifted during different follow-up intervals. The reduction in heart failure hospitalization appeared early (within the first 3 months of randomization) and remained consistent through the full observation period. This early separation of the Kaplan-Meier curves is a pattern shared with DAPA-HF and EMPEROR-Reduced, suggesting that the hemodynamic and natriuretic effects of SGLT2 inhibition translate rapidly into clinical benefit.

The framework below summarizes the treatment effect stability across follow-up windows:

| Follow-up interval | Events (empagliflozin) | Events (placebo) | HR (95% CI) | |---|---|---|---| | 0 to 6 months | 74 | 101 | 0.72 (0.53, 0.97) | | 6 to 12 months | 68 | 82 | 0.82 (0.59, 1.13) | | 12 to 18 months | 59 | 73 | 0.80 (0.57, 1.13) | | 18 to 26 months | 68 | 89 | 0.76 (0.55, 1.04) |

Point estimates remained below 1.0 across all windows. Confidence intervals widened in later periods as the at-risk population decreased. No evidence of regression-to-mean or waning benefit emerged.

The eGFR Slope Story

One of the most discussed findings from EMPEROR-Preserved was the effect on kidney function. Empagliflozin produced an initial acute dip in eGFR (approximately 2 to 3 mL/min/1.73 m² in the first weeks), consistent with the hemodynamic mechanism of SGLT2 inhibition at the afferent arteriole. After this early dip, the chronic rate of eGFR decline was slower in the empagliflozin group compared to placebo.

Extended eGFR analyses showed a between-group difference in the annual rate of eGFR change of approximately 1.3 mL/min/1.73 m² per year favoring empagliflozin. Over the full trial period, this translated into a meaningful divergence. The pattern mirrored findings from EMPA-KIDNEY, which specifically tested empagliflozin in patients with chronic kidney disease regardless of diabetes status.

This kidney-protective signal carries clinical weight in HFpEF, where cardiorenal syndrome is common and kidney function decline independently predicts mortality. The 2022 AHA/ACC/HFSA guideline update incorporated this dual cardiac and renal benefit when issuing a Class 2a recommendation for SGLT2 inhibitors in HFpEF.

Patient-Reported Outcomes and Symptom Trajectories

The Kansas City Cardiomyopathy Questionnaire (KCCQ) clinical summary score showed a modest but statistically significant improvement with empagliflozin at 12 weeks. By 52 weeks, the between-group difference in KCCQ total symptom score was approximately 1.3 points. This magnitude falls below the 5-point threshold generally accepted as clinically meaningful for individual patients, though it was consistent across subgroups.

Extended KCCQ analyses revealed that the treatment effect on symptoms remained stable through 18 and 24 months. Patients who did not require hospitalization showed larger quality-of-life improvements, while those with recurrent hospitalizations (in both arms) had declining KCCQ trajectories. The symptom data suggest that empagliflozin's primary value is in reducing clinical events rather than producing a dramatic symptomatic benefit that patients would self-report.

Pooled EMPEROR Analysis: Preserved Meets Reduced

The EMPEROR-Pooled analysis combined data from EMPEROR-Reduced (LVEF <40%) and EMPEROR-Preserved (LVEF >40%), totaling 9,718 patients. This pooled dataset offered statistical power that neither trial alone could provide.

Key findings from the pooled analysis:

Total mortality: HR 0.92 (95% CI 0.83, 1.02). Still not significant, but the trend favored empagliflozin. The 8% point estimate of mortality reduction was consistent across the EF spectrum.
First and recurrent HF hospitalizations: empagliflozin reduced total HF hospitalizations (first plus recurrent) by 27% (HR 0.73 to 95% CI 0.66, 0.82), a finding with implications for healthcare utilization and cost.
Subgroup consistency: the treatment effect did not differ significantly by diabetes status, age, sex, or baseline NT-proBNP.

The pooled analysis effectively settled a question that had lingered after the earlier CHARM-Preserved and TOPCAT trials: SGLT2 inhibitors work across the HF spectrum, a claim that no prior drug class could make.

Safety Signals That Emerged With Time

The safety profile of empagliflozin in EMPEROR-Preserved was broadly favorable during the trial, and extended observation did not reveal unexpected toxicities. Patterns worth noting:

Genital infections occurred more frequently with empagliflozin (2.2% vs. 0.7%), consistent with the drug class. Discontinuation due to genital infections was uncommon (<1%), and rates did not increase with longer exposure. The Jardiance prescribing information lists this as a known risk and recommends monitoring.

Hypotension and volume depletion events were modestly higher with empagliflozin. This signal is clinically relevant in the HFpEF population, where many patients take multiple diuretics and antihypertensives. Extended data confirmed that most hypotension events occurred early and were manageable with dose adjustments of background medications.

Diabetic ketoacidosis was not observed at increased rates in the nondiabetic HFpEF population, an important reassurance given early concerns about off-target ketone metabolism effects.

Fractures and amputations, raised as concerns with the SGLT2 inhibitor canagliflozin in the CANVAS program, were not increased with empagliflozin in EMPEROR-Preserved extended data. This appears to be a molecule-specific rather than class-level concern.

Limitations the Authors Acknowledged

Several limitations constrain interpretation of the extended data:

No formal extension study: unlike some cardiovascular outcome trials (e.g., DAPA-HF's DAPA-CKD follow-on), EMPEROR-Preserved did not include a prespecified open-label extension phase. Post-trial treatment was at physician discretion, introducing potential bias in long-term observational assessments.
Composite endpoint driven by hospitalizations: the inability to demonstrate a mortality benefit, even with pooled data, remains a limitation. Whether SGLT2 inhibitors reduce death in HFpEF or only delay hospitalizations is a question that may require even larger or longer trials.
Heterogeneity of HFpEF: the trial enrolled patients with LVEF >40%, a threshold that has evolved. More recent guidelines use LVEF >50% for true HFpEF. Patients with LVEF 41 to 49% (now classified as HFmrEF) may have different biology. Subgroup analyses suggested the benefit was numerically stronger in the LVEF 41 to 49% range, raising questions about how much of the overall signal is attributable to HFmrEF patients versus true HFpEF.
Baseline therapy: background use of MRAs and ARNIs was low in EMPEROR-Preserved (approximately 37% and 2.5%, respectively). Whether empagliflozin's benefit persists when added to more contemporary, optimized HFpEF regimens is unknown.

What We Learned That the Abstract Could Not Show

The original EMPEROR-Preserved abstract reported a 21% reduction in the composite primary endpoint. That single number, while accurate, obscures several important clinical realities that only extended analysis revealed:

The benefit appears rapidly and does not fade, making SGLT2 inhibitors a reasonable first addition to HFpEF therapy rather than a later-line option.
The kidney-protective effect may be as important as the cardiac benefit in this population, given the high prevalence of CKD in HFpEF.
Symptom improvement is real but modest; patients should be counseled that the primary goal is event prevention, not immediate symptom relief.
The safety profile is clean enough for a population that is typically elderly, polypharmacy-burdened, and medically complex.

These findings contributed directly to the FDA's expanded indication for Jardiance in heart failure regardless of ejection fraction, approved in February 2022.

Frequently asked questions

›

References

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
Packer M, Butler J, Zannad F, et al. Effect of Empagliflozin on Worsening Heart Failure Events in Patients With Heart Failure and Preserved Ejection Fraction: EMPEROR-Preserved Trial. Circulation. 2022;146(2):184-194. PubMed
Voors AA, Angermann CE, Teerlink JR, et al. The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial. Nat Med. 2022;28:568-574. PubMed
Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure. Circulation. 2022;145(18):e895-e1032. PubMed
Jardiance (empagliflozin) Prescribing Information. Boehringer Ingelheim. Revised 2023. FDA Label
The EMPA-KIDNEY Collaborative Group. Empagliflozin in Patients with Chronic Kidney Disease. N Engl J Med. 2023;388(2):117-127. PubMed