The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial (EMPULSE)
Source: View publication →
In patients hospitalized for acute heart failure, early in-hospital initiation of empagliflozin resulted in significant clinical benefit at 90 days across the spectrum of left ventricular ejection fraction and regardless of diabetes status.
Key Findings
Study Design
Study Limitations
Clinical Significance
EMPULSE provided crucial evidence that it is safe and highly beneficial to initiate SGLT2 inhibitors early during hospitalization for acute heart failure, once the patient is clinically stable. This challenged the previous practice of delaying disease-modifying therapies and established early in-hospital initiation as a standard of care to prevent early post-discharge readmissions and rapidly improve symptom burden.
Historical Context
Prior to EMPULSE, trials like DAPA-HF and EMPEROR-Reduced established SGLT2 inhibitors as foundational therapy for chronic heart failure, but patients hospitalized for acute decompensated heart failure were largely excluded. While the SOLOIST-WHF trial showed early benefits in worsening heart failure, it only included patients with diabetes. EMPULSE was the first major trial to demonstrate the safety and efficacy of early in-hospital initiation of an SGLT2 inhibitor for acute heart failure across the full spectrum of ejection fractions and regardless of diabetes status.
Guided Discussion
High-yield insights from every perspective
How does the mechanism of action of SGLT2 inhibitors like empagliflozin differ from traditional loop diuretics, and why might this explain the early clinical benefits and safety profile seen in acute heart failure patients in the EMPULSE trial?
Key Response
SGLT2 inhibitors cause osmotic diuresis and natriuresis at the proximal tubule without activating the sympathetic nervous system or the renin-angiotensin-aldosterone system (RAAS) as aggressively as loop diuretics. They also improve myocardial energetics, reduce interstitial edema preferentially over intravascular volume, and decrease preload and afterload. This multifaceted mechanism helps explain the early, multi-domain benefits seen in EMPULSE (including mortality, HF events, and quality of life) without increasing the risk of acute kidney injury or severe hypotension.
Before the EMPULSE trial, clinicians often delayed starting SGLT2 inhibitors until an acute heart failure patient was discharged and euvolemic out of concern for AKI or hypotension. Based on EMPULSE, what specific clinical stability criteria were required to initiate empagliflozin in the hospital, and what were the safety findings?
Key Response
In EMPULSE, patients were deemed stable enough for initiation if they had a systolic blood pressure over 100 mmHg, no increase in IV diuretics in the last 6 hours, no IV vasodilators, and no IV inotropes in the last 24 hours. The safety findings demonstrated no excess risk of acute kidney injury, symptomatic hypotension, or volume depletion compared to placebo. This shows residents that once these basic stability criteria are met, initiating SGLT2 inhibitors in-hospital is safe and helps overcome clinical inertia.
The EMPULSE trial utilized a stratified win ratio for its primary endpoint, incorporating mortality, heart failure events, and Kansas City Cardiomyopathy Questionnaire (KCCQ) scores. How does the hierarchical win ratio approach differ from traditional time-to-first-event analyses, and why is this particularly advantageous in acute heart failure populations with heterogeneous ejection fractions?
Key Response
Traditional time-to-first-event composites treat all events equally (e.g., a rehospitalization counts the same as death) and ignore subsequent events. The win ratio compares pairs of patients in a hierarchical manner (death > HF events > KCCQ change), prioritizing mortality as the most important outcome. In a diverse acute heart failure population, this provides a more comprehensive assessment of the net clinical benefit, capturing improvements in quality of life via KCCQ for the many patients who do not experience a hard clinical event during the short 90-day follow-up.
EMPULSE demonstrated significant clinical benefit at just 90 days post-discharge. How does this finding shift our paradigm regarding the 'vulnerable phase' of post-discharge heart failure, and how should this influence our rounding discussions with junior trainees about timing the initiation of guideline-directed medical therapy (GDMT)?
Key Response
The first 90 days after an acute heart failure hospitalization represent a highly 'vulnerable phase' with peak rates of rehospitalization and death. EMPULSE proved that early SGLT2i initiation rapidly alters this trajectory. Attendings must teach trainees that the hospital admission is a critical, captive window to optimize GDMT rather than deferring it to the outpatient setting, thereby dramatically reducing early post-discharge morbidity and overcoming the well-documented problem of outpatient clinical inertia.
Scholarly Review
Critical appraisal through the lens of expert reviewers and guideline development
The EMPULSE trial enrolled a relatively modest sample size (n=530) to assess outcomes across diverse strata (de novo versus decompensated chronic HF). From a statistical and design perspective, how did the use of a stratified win ratio provide adequate power for this sample size, and what are the limitations in making subgroup-specific inferences from this trial design?
Key Response
The win ratio increases statistical power by utilizing the full spectrum of patient outcomes, including continuous variables like the KCCQ symptom score, rather than relying solely on relatively rare hard events (like deaths) in a short 90-day window. However, because it is an aggregate measure of pairs, a modest sample size means the study is underpowered to detect significant treatment interactions in smaller subgroups. It demonstrates an overall direction of benefit across strata but lacks granularity for specific hard-outcome reductions within individual subgroups.
In reviewing the EMPULSE trial, a critical peer reviewer might flag the inclusion of KCCQ scores in the primary win ratio endpoint. Given the physiological effects of SGLT2 inhibitors, what are the potential threats to blinding, and how might unblinding bias the patient-reported KCCQ component of the hierarchical endpoint?
Key Response
SGLT2 inhibitors have distinct physiological effects, such as increased urine output, potential genital mycotic infections, and detectable glycosuria on routine lab panels, which could inadvertently unblind patients or clinicians. Since the KCCQ is a subjective, patient-reported outcome of symptoms and quality of life, unblinding could lead to a placebo effect or reporting bias. A critical editor must evaluate whether the hierarchical win ratio's success was heavily driven by the subjective KCCQ ties broken in favor of the treatment arm, rather than strictly objective clinical endpoints.
Prior to 2022, heart failure guidelines primarily recommended SGLT2 inhibitors for chronic HFrEF in the outpatient setting. How do the results of EMPULSE compel a guideline committee to update the Class of Recommendation (COR) and Level of Evidence (LOE) specifically for the in-hospital initiation of SGLT2 inhibitors in acute heart failure?
Key Response
EMPULSE provides strong evidence supporting early in-hospital initiation of SGLT2 inhibitors in acute heart failure once the patient is hemodynamically stable, regardless of ejection fraction or diabetes status. The guidelines must be updated to a Class 1 recommendation (Level of Evidence B-R, or A when combined with similar trials like SOLOIST-WHF) to explicitly instruct clinicians to initiate SGLT2 inhibitors prior to hospital discharge to reduce the high risk of early post-discharge rehospitalization and mortality.
Clinical Landscape
Noteworthy Related Trials
EMPEROR-Reduced Trial
Tested
Empagliflozin 10 mg daily
Population
Patients with heart failure and reduced ejection fraction (HFrEF)
Comparator
Placebo
Endpoint
Composite of cardiovascular death or hospitalization for heart failure
SOLOIST-WHF Trial
Tested
Sotagliflozin 200-400 mg daily
Population
Patients with type 2 diabetes and recent worsening heart failure
Comparator
Placebo
Endpoint
Total cardiovascular deaths, hospitalizations for heart failure, and urgent heart failure visits
EMPEROR-Preserved Trial
Tested
Empagliflozin 10 mg daily
Population
Patients with heart failure and preserved ejection fraction (HFpEF)
Comparator
Placebo
Endpoint
Composite of cardiovascular death or hospitalization for heart failure
Tailored to your role
Want this tailored to you?
Add your specialty or training stage to get role-specific takeaways and more questions.
Personalize this analysis