7c. Acute decompensated heart failure

7c.1 Background

7c.1.1 2023 Update to guideline structure

In the UKKA 2021 guideline, this section summarised evidence in populations with heart failure with preserved ejection fraction (HFpEF) and in acute decompensated heart failure (ADHF). Since then, an additional large randomized controlled trial has been reported which demonstrated the efficacy of dapagliflozin at reducing risk among people with heart failure with mildly reduced or preserved ejection fraction (the DELIVER trial (23)). Based on the findings from both DELIVER and the EMPEROR-PRESERVED trials, there is now sufficient evidence to provide grade 1A recommendations for use in HFpEF alongside heart failure with reduced ejection fraction (HFrEF) in sections 2 and 3. Consequently, this section now focuses primarily on the more limited evidence available for recommendation of use in ADHF.

7c.1.2 Summary of trial evidence in acute decompensated heart failure

Since publishing the UKKA 2021 guideline, there is new evidence supporting a role for SGLT 2 inhibitors within this context. The trials can be divided into those reporting clinical benefit versus those reporting on clinical surrogates for fluid status, biochemical markers of heart failure status or safety information, such as worsening kidney function. Relevant trials are summarised in table 7c.1. Clinical benefits of SGLT 2 inhibition in ADHF The largest reported trial is EMPULSE (24), which randomized 530 participants to empagliflozin 10mg once daily versus placebo. EMPULSE eligibility aimed to recruit participants with an acute heart failure admission irrespective of diabetes status, and included eGFRs down to 20 mL/min/1.73m². ~10% of participants had an eGFR <30 (mean eGFR 52) mL/min/1.73m². 32% of participants had HFpEF (i.e. left ventricular ejection fraction [LVEF] >40%). The primary outcome was a hierarchical composite based on death from any cause, number of heart failure events, time to first heart failure event, and a change in Kansas City Cardiomyopathy Questionnaire Total Symptom Score (KCCQ-TSS). The main result was a favourable win ratio for empagliflozin of 1.36 [95%CI 1.09-1.68], p=0.005. This result can be thought of as, for any random pair of participants, the odds of a better outcome was increased by 36% among those allocated to empagliflozin compared to placebo. Primary outcome events included 11 deaths (4.2%) in the empagliflozin group and 22 (8.3%) in the placebo group, and 28 participants (10.6%) in the empagliflozin group had a heart failure event versus 39 participants (14.7%) in the placebo group. For the outcome of a ≥5 point difference in the KCCQ-TSS (from baseline to day 90), there were 35.9% wins for the empagliflozin group versus 27.5% for the placebo group. The information on the benefits of empagliflozin were therefore based mostly on participant-reported KCCQ score, but there were effects on mortality and heart failure outcomes higher in the outcome hierarchy that were also consistent with benefit on these clinical outcomes. These benefits were consistent regardless of eGFR and diabetes status. The trial found no difference in serious adverse events in the empagliflozin versus the placebo group (32.3% versus 43.6%), and “acute renal failure” occurred in 7.7% of participants in the empagliflozin group versus 12.1% of participants in the placebo group, consistent with other evidence than SGLT 2 inhibition reduces risk of acute kidney injury by about a quarter (25). Effects of SGLT 2 inhibition on markers of cardiac congestion in ADHF The EMPULSE trial also demonstrated that several markers of cardiac congestion were improved by allocation to empagliflozin. These included weight loss, improved clinical congestion score, and reduced NT-proBNP at day 15 after randomization (26). Of note, improvements in body weight change from baseline (including when adjusted per total daily loop diuretic dose) were maintained up to day 90 post-randomization. Several small prospective trials (27-31) have reported on similar surrogate markers of cardiac congestion, including NT-proBNP, weight loss, diuresis and diuretic dose. Benefits are typically identified within the first week of admission or at discharge. They indicate that the addition of SGLT 2 inhibition to standard treatment leads to reduced diuretic doses and improved diuresis, but there are variable effects on NT-proBNP levels (28, 29, 32). The effects of SGLT 2 inhibition in ADHF on kidney function and CKD Several trials (24, 27-32) have addressed the kidney safety of SGLT 2 inhibition in ADHF, indicating that the addition of the agent does not adversely affect kidney function when used in participants with decreased eGFR. Where declines in eGFR were observed, these do not exceed expected levels (section 1.3.2) and follow-up data from multiple trials have proven the longer term benefits on kidney disease progression outcomes and acute kidney injury in people with CKD (17, 18, 33, 34). Participants with kidney failure (i.e. eGFR <15 ml/min/1.73m² or on dialysis) have not been included in the ADHF trials reported to date. Table 7c.1 provides a summary of the relevant data pertaining to kidney safety by trial. Taken together, these data support the use of SGLT 2 inhibitors in ADHF to assist diuresis during admission and indicate that these agents are safe and efficacious for those presenting with an eGFR >15 mL/min/1.73m². Given the positive effects of these agents in chronic stable heart failure (23, 35-38), initiating their use in ADHF will help maximise implementation for grade 1 recommendations. Upcoming trials of SGLT 2 inhibition in ADHF Several randomized trials are currently underway to further evaluate the role of SGLT 2 inhibition in people presenting to hospital with ADHF. The largest of these, DAPA ACT HF-TIMI 68 (39) aims to randomize ~2400 participants to two months of treatment with dapagliflozin or placebo. Participants will be enrolled within 24 hours to 14 days of hospital admission if they have ADHF with intensification of therapy during admission. Individuals with eGFR ≥25 mL/min/1.73m² will be enrolled, irrespective of diabetes status or ejection fraction. The estimated study completion date at the time of writing is Q2/2023.

7c.1.3 Quality of the evidence

There are relatively few large randomised trials investigating SGLT 2 inhibition for the treatment of ADHF. With the exception of EMPULSE, reported trials are generally small, often open-label and with limited power to assess clinical outcomes.

Table 7C.1

7c.2 Recommendations for implementation

We suggest initiating SGLT 2 inhibition in people with CKD (eGFR ≥20 mL/min/1.73m²) with acute decompensated heart failure. (2B) Rationale: Multiple small-scale randomized controlled trials provide evidence that SGLT 2 inhibition is safe in a population presenting to hospital with ADHF. Furthermore, the moderate-sized EMPULSE trial demonstrates that individuals treated with SGLT 2 inhibition for ADHF are more likely to yield clinical benefit than those treated with placebo. Trials in ADHF enrolled people with eGFR >15 mL/min/1.73m², with the largest trial (EMPULSE) enrolling those with eGFR ≥20 mL/min/1.73m². More evidence from large, well-conducted randomized controlled trials (such as the DAPA ACT HF-TIMI 68 trial) will provide more comprehensive evidence to support this recommendation. See section 2 & 3 for recommendations for use in other forms of heart failure or to modify cardiovascular risk.

7c.3 Clinical Research Recommendations

Large randomised placebo-controlled clinical trials powered to assess hard clinical outcomes in people with ADHF.