DELIVER

Published

Solomon SD, et al. "Dapagliflozin in heart failure with mildly reduced or preserved ejection fraction". The New England Journal of Medicine. 2022. 387(12):1089-1098.
PubMed • Full text • PDF • ClinicalTrials.gov

Clinical Question

In patients with heart failure with mid-range or preserved LVEF (i.e., LVEF >40%), does an addition of the SGLT2 inhibitor medication dapagliflozin reduce HF hospitalizations, urgent HF visits, or CVD mortality?

Bottom Line

Regardless of diabetes status, among patients with HFmrEF or HFpEF (i.e., LVEF >40%), addition of dapagliflozin reduces HF hospitalizations, urgent HF visits, or CVD mortality.

Major Points

Originally intended to treat diabetes, SGLT2 inhibitors (e.g., dapagliflozin, empagliflozin) were shown to reduce HF events among adults with LVEF ≤40% (i.e., HFrEF) in the 2019 DAPA-HF and 2020 EMPEROR-Reduced trials. Adults with HFmrEF (i.e., LVEF >40 to <50%) or HFpEF (i.e., LVEF ≥50%) comprise upwards of 25% and 50% of HF, respectively. There are limited evidence-based treatments for HFmrEF and HFpEF. The 2021 EMPEROR-Preserved trial showed benefit for the SGLT2 inhibitor empagliflozin among adults with HFmrEF or HFpEF.

Published in 2022, the Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure (DELIVER) trial randomized 6,263 adults with HFmrEF or HFpEF to the SGLT2 inhibitor dapagliflozin or placebo. At 2.3 years, there was a reduction in HF hospitalization, urgent HF visits, or CVD mortality in the dapagliflozin group relative to placebo (16.4% vs. 19.5%; HR 0.82; 95% CI 0.73-0.92). This outcome was similar across prespecified subgroups, including by diabetes status. This finding confirms the benefit of SGLT2 inhibitor therapy in HFmrEF and HFpEF.

Guidelines

ACC/AHA/HFSA heart failure (2022, adapted)

Note: These guidelines were released a few months before the publication of DELIVER.

• If symptomatic chronic HFrEF, SGLT2i recommended to reduce HF hospitalizations and CVD mortality, regardless of diabetes status (COR 1, LOE A)
• If HFpEF or HFmrEF, SGLT2i can be useful for lowering HF mortality and HF hospitalizations (COR 2a, LOE B-R)

Design

• Multicenter, double-blind, randomized controlled trial
• N=6,263
  • Dapagliflozin (n=3,131)
  • Placebo (n=3,132)
• Setting: 353 centers in 20 countries
• Enrollment: 2018-2020
• Median follow-up: 2.3 years
• Analysis: Intention-to-treat
• Primary outcome: HF hospitalization, urgent HF visit, or CVD mortality

Population

Inclusion Criteria

• Aged ≥40
• Stable/chronic HF with LVEF >40%; with NYHA class II-IV symptoms
• ≥1 bout of HF ≥6 weeks before enrollment, characterized by HF signs and symptoms
• Intermittent or chronic need for diuretics
• NT-pro BNP ≥300 pg/mL (or ≥600 pg/mL if AF)
• No intravenous HF treatment in the prior 12-24 hours.

Exclusion Criteria

• On an SGLT2 inhibitor
• T1DM
• eGFR <25
• SBP <95 mm Hg
• CAD event, AF ablation, valve procedure in prior 12 weeks, or one of these interventions planned
• Stroke/TIA in prior 12 weeks
• BMI >50 kg/m²
• Certain conditions: Primary pulmonary hypertension, chronic PE, severe COPD or other severe lung disease, prior cardiac transplant, complex congenital heart disease
• Planned CRT
• Atypical HF cause (e.g., amyloid)
• Life expectancy <2 years for reason other than cardiac
• Liver disease
• Usual pregnancy/childbearing potential exclusion

Baseline Characteristics

From the dapagliflozin group.

• Demographics: Age 72 years, 45% female, 71% White adults, 20% Asian adults, 3% Black adults
• Region: 14% N America, 19% L America, 48% Europe or Saudi Arabia, 19% Asia
• HF details:
  • NYHA class: II 74%, III 26%, IV <1%
  • Mean LVEF 54%; 40-49% range 34%; 50-59% range 36%, ≥60% range 30%
• Medical problems: T2DM 45%, HTN 88%, prior LVEF ≤40% 18%
• eGFR: 61

Interventions

• Randomly assigned in a 1:1 ratio to receive 10 mg dapagliflozin or matching placebo
  • Control group was provided with a placebo dosage as there are no approved pharmacological treatments for HFpEF that could be utilized as a comparator.
• In person visits were necessitated 30 days after randomization, 4 months after randomization, and then every 4 months after until study completion.

Outcomes

Comparisons are intensive therapy vs. standard therapy.

Primary Outcomes

Components of the primary outcome are shown here for simplicity, but individually were not primary outcomes.

HF hospitalization, urgent HF visit, or CVD mortality

16.4% vs. 19.5% (HR 0.82; 95% CI 0.73-0.92; P<0.001)

HF events: 11.8% vs. 14.5% (HR 0.79; 95% CI 0.69-0.91)

HF hospitalization: 10.5% vs. 13.3% (HR 0.77; 95% CI 0.67-0.89)

Urgent HF visit: 1.9% vs. 2.5% (HR 0.76; 95% CI 0.55-1.07)

CVD mortality: 7.4% vs. 8.3% (HR 0.88; 95% CI 0.74-1.05)

Secondary Outcomes

# of initial and recurrent HF events or CVD mortality

11.8 vs. 15.3 events/100 P-Y (RR 0.77; 95% CI 0.67-0.89; P<0.001)

Change in KCCQ total symptom score at 8 mo, higher being better

No comparison given (win ratio 1.11; 95% CI 1.03-1.21; P=0.009)

All-cause mortality

15.9% vs. 16.8% (HR 0.94; 95% CI 0.83-1.07)

Subgroup Analysis

The primary analysis was similar when stratified by age, sex, race, region, NYHA symptoms, LVEF range, NT-proBNP, recent HF hospitalization, T2DM status, AF/flutter status, BMI, eGFR, SBP, and prior LVEF ≤40%.

Adverse Events

Adverse events were similar between groups. These are detailed in Table 2 on page 1093.

Criticisms

• Low enrollment of Black adults

Funding

AstraZeneca, the manufacturers of Farxiga, the brand name of dapagliflozin.