Starting SGLT2 Inhibitors in the Hospital for Acute HF Gains Support

MADRID, Spain—The DAPA ACT HF – TIMI 68 trial has failed to show a significant improvement in outcomes when starting dapagliflozin (Farxiga; AstraZeneca) during an acute heart failure (HF) hospitalization.

But trends toward benefit, in addition to positive findings of a meta-analysis incorporating the results, had experts here at the European Society of Cardiology Congress 2025 backing early initiation of sodium-glucose cotransporter 2 (SGLT2) inhibition.

“The totality of randomized trial data really suggests that starting SGLT2 inhibitors during heart failure hospitalization reduces the early risk of cardiovascular death or worsening heart failure and all-cause mortality in the early postdischarge period,” David Berg, MD (Brigham and Women’s Hospital, Boston, MA), a principal investigator of DAPA ACT HF – TIMI 68, said during a press conference.

Filippo Crea, MD, PhD (Catholic University, Rome, Italy), who moderated the press conference, agreed. There is already convincing evidence that when patients are hospitalized with HF, all four pillars of guideline-directed medical therapy (GDMT)—including SGLT2 inhibitors—should be started, he said.

“We know from the history of medicine that if something works, it’s better to give it sooner than later,” Crea told TCTMD.

Pulling the Evidence Together

Patients hospitalized with HF have a high risk of death and other adverse outcomes during the index admission and shortly after discharge, Berg noted. Even though SGLT2 inhibitors are indicated for HF regardless of LVEF, there are limited data on in-hospital initiation of the drugs, he added.

DAPA ACT HF – TIMI 68, published simultaneously in Circulation, was conducted at 210 sites in the United States, Canada, Czech Republic, Hungary, and Poland. Investigators explored the safety and efficacy of in-hospital initiation of dapagliflozin 10 mg daily compared with placebo in 2,401 patients (median age 69 years; 34% women) hospitalized for acute heart failure.

Most (72%) had an LVEF of 40% or less, and for 45%, hospitalization was the first presentation of HF. Patients were well treated at both randomization and discharge with beta-blockers, renin-angiotensin-aldosterone-system inhibitors, mineralocorticoid receptor antagonists, and loop diuretics.

The primary endpoint was a composite of cardiovascular death or worsening HF, defined as decompensation during index admission, rehospitalization for HF, or urgent HF visit, through 60 days. At that time, 10.9% of patients treated with dapagliflozin and 12.7% of those who received placebo had a clinical event (HR 0.86; 95% CI 0.68-1.08). There were nonsignificant trends toward reductions in both individual components, as well as in all-cause death and a composite of CV death or rehospitalization for HF.

Outcomes at 60 Days

Results for the primary endpoint were consistent across subgroups defined by LVEF, renal function, NT-proBNP/BNP levels, and whether HF was newly diagnosed or chronic.

Compared with placebo, dapagliflozin was associated with higher rates of symptomatic hypotension (3.6% vs 2.2%) and worsening kidney function (5.9% vs 4.7%) but no increase in major hypoglycemia (0.2% vs 0.3%). There were no cases of diabetic ketoacidosis in the trial. The rate of adverse events leading to study drug discontinuation was 4.8% with dapagliflozin and 4.7% with placebo.

These results are consistent with the known safety profile of SGLT2 inhibitors as a class, Berg said.

To provide a broader look at the impact of in-hospital initiation of SGLT2 inhibition, Berg and his colleagues combined the results of DAPA ACT HF – TIMI 68 with those from EMPULSE, a trial of empagliflozin (Jardiance; Boehringer Ingelheim/Eli Lilly), and from the in-hospital cohort of SOLOIST-WHF, a trial of sotagliflozin (Zynquista; Sanofi Aventis/Lexicon), a dual inhibitor of SGLT1 and SGLT2. The prespecified meta-analysis showed that early treatment was associated with lower risks of CV death or worsening HF (HR 0.71; 95% CI 0.54-0.93) and all-cause mortality (HR 0.57; 95% CI 0.41-0.80).

‘We Have an Answer’

Regarding the lack of statistical significance for the difference in the primary outcome of DAPA ACT HF – TIMI 68, Berg said the trial was underpowered due to the combined effects of a short follow-up (chosen to limit the amount of time patients in the control group would be on placebo) and a lower-than-expected number of events.

That is why the meta-analysis was particularly helpful, he said. “Bringing the totality of data together in a pooled estimate of the treatment effect probably provides the best insight of the true treatment effect.”

We have good motivation to start these therapies in the hospital and no reason not to. David Berg

Biykem Bozkurt, MD, PhD (Baylor College of Medicine, Houston, TX), the discussant for the study, agreed that the short duration of follow-up and the low event rates likely explain the neutral results of the trial, noting that the point estimates for the endpoints are directionally consistent with benefit.

She indicated that these findings reinforce that in-hospital initiation of SGLT2 inhibition is safe. In addition, “we do know that SGLT2 inhibitor therapy is effective across the continuum of heart failure and across the spectrum of presentations” based on prior studies, which was validated by the meta-analysis, she said.

“The totality of evidence indicates that the therapy is safe and effective across the spectrum of heart failure, including when initiated during hospitalization, and the clinicians should not be dissuaded [from starting] SGLT2 therapy in hospitalized patients when the patient is stabilized,” Bozkurt said.

Berg said: “I think we have an answer. . . . We have good motivation to start these therapies in the hospital and no reason not to.”

Continuing reluctance to do so is an educational problem, Crea said. “This is the challenge for science and for medicine in particular—new discoveries do not reach the target immediately. It takes time to implement them,” he said. “And this is the reason why we have guidelines and then surveys to verify that the new knowledge at the end of the day is implemented in clinical practice.”