Your browser is unsupported

We recommend using the latest version of IE11, Edge, Chrome, Firefox or Safari.

What evidence supports the initiation of sodium-glucose cotransporter 2 inhibitors during acute decompensated heart failure?

In 2013, the first sodium-glucose cotransporter-2 (SGLT2) inhibitor, canagliflozin, was approved, representing a novel mechanism of action and a new target for medical therapy of type 2 diabetes mellitus (T2DM).1 Following the approval of other agents in the class, the SGLT2 inhibitors quickly became an option for oral pharmacotherapy of many patients with T2DM who require additional glycemic control.1-5 Now, nearly a decade after their entry to the US market, the SGLT2 inhibitors have accumulated a wealth of high-quality data supporting their ability to improve major adverse cardiovascular (CV) and renal outcomes, including in patients without T2DM.6

Patients with heart failure (HF) have been a major subpopulation included in clinical trials of SGLT2 inhibitors. In fact, the prescribing information of dapagliflozin and empagliflozin indicates they may be used in patients with HF, regardless of the presence of T2DM, based on several well-designed trials.2,3 Guidelines for treatment of HF now provide recommendations on use of SGLT2 inhibitors in patients with HF at various stages of disease, including HF with reduced ejection fraction (HFrEF) and HF with preserved ejection fraction (HFpEF).6

Given their impressive benefit in chronic treatment of HF, hypotheses have arisen regarding the role of SGLT2 inhibitors during acute decompensated heart failure (ADHF). However, during the early phases of ADHF, fluctuations in fluid and electrolyte balance may occur, which may be affected by the transient effects of SGLT2 inhibitors on diuresis and estimated glomerular filtration rate (GFR).1-4,7 Because questions may arise regarding the safety and efficacy of these drugs in this setting, this review discusses current evidence regarding initiation of SGLT2 inhibitors during ADHF, with a focus on studies reporting clinical endpoints.

Clinical trials
To date, evidence describing the effect of SGLT2 inhibitors on clinical endpoints is available for various agents. Studies that report cardiorenal endpoints and rates of hospital readmission are described below.

The SOLOIST-WHF trial, the largest trial of SGLT2 inhibitor use in ADHF to date, randomized 1222 patients with T2DM and recent hospitalization for heart failure (HHF), regardless of left ventricular ejection fraction (LVEF), to sotagliflozin or placebo.8 The primary endpoint, a composite of death from CV causes and hospitalizations and urgent visits for heart failure, was significantly improved with sotagliflozin after a median follow-up of 9 months (hazard ratio [HR], 0.67; 95% confidence interval [CI], 0.52 to 0.85). The component endpoint likely most responsible for this finding was hospitalizations and urgent visits for HF (HR, 0.64; 95% CI, 0.49 to 0.83). Death from CV causes did not significantly differ between groups. In subgroup analyses, sotagliflozin demonstrated consistent benefit on the primary endpoint in patients with baseline LVEF <50% or ≥50%, and whether administration of the first dose occurred before versus after discharge.

The incidences of treatment-emergent adverse events were similar between sotagliflozin (69.4%) and placebo (67.4%).8 Although hypoglycemia was more common with sotagliflozin (4.3% vs 2.8%), incidence of hypotension (6.0% vs 4.6%, respectively), hyperkalemia (4.3% vs 5.1%), and acute kidney injury (4.1% vs 4.4%) was similar between groups.

The EMPULSE trial randomized 530 patients with T2DM and ADHF to double-blind dapagliflozin 10 mg once daily or placebo.9 EMPULSE calculated a “win ratio” to compare effects on the primary endpoint of clinical benefit, defined as a hierarchical composite of all-cause death, number of HF events and time to first HF event, or a ≥5-point difference in change from baseline in the Kansas City Cardiomyopathy Questionnaire Total Symptom Score at 90 days. The win ratio is an alternative method of analyzing composite endpoints, which places greater weight on endpoints of greater clinical importance. This addresses a limitation of traditional composite endpoints, in which events of lesser importance (eg, HHF) typically occur earlier than those of greater importance (eg, CV death). Empagliflozin significantly improved the win ratio compared with placebo (1.36; 95% CI, 1.09 to 1.68), “winning” in 53.9% of matched comparisons. This benefit was consistent in subgroups of patients with or without diabetes, and regardless of age, sex, LVEF, and baseline measures of N-terminal-pro hormone B-natriuretic peptide (NT-proBNP) and GFR.

EMPULSE revealed no major imbalances in adverse events between empagliflozin and placebo.9 Incidence was comparable for any adverse event (70.0% vs 77.3%, respectively), acute renal failure (7.7% vs 12.1%), volume depletion (12.7% vs 10.2%), serious symptomatic hypotension (1.2% vs 1.5%), and hypoglycemia (1.9% vs 1.5%).

Prior to the larger EMPULSE trial, EMPA-RESPONSE-AHF also evaluated empagliflozin in in a pilot study.10 Like EMPULSE, this trial randomized 80 patients with ADHF, regardless of comorbid T2DM, to double-blind empagliflozin 10 mg once daily or placebo. The primary endpoints of this trial were change in visual analogue scale dyspnea score, diuretic response, change in NT-proBNP, and length of stay. No differences were detected between groups in these primary endpoints. However, patients treated with empagliflozin experienced improvements in a composite of in-hospital worsening HF, death, or HF readmission at 60 days (10% vs 33%; p=0.014). Frequencies of adverse events were similar between empagliflozin and placebo overall (55 vs 63 events, respectively) and for renal/urinary events (38% vs 33%).

One substudy of an ongoing trial reports outcomes with dapagliflozin in patients with ADHF, regardless of LVEF, who were planned to received intravenous loop diuretics.11 This trial, by Charya et al, randomized 151 patients to open-label dapagliflozin 10 mg once daily or placebo. The primary endpoint, deterioration of renal function (defined as increased serum creatinine of 0.3 mg/dL or more within 48 hours), was numerically more frequent with dapagliflozin (24.4% vs 15.2%, respectively; p=0.07). Secondary endpoints revealed that dapagliflozin was associated with lower mean daily doses of loop diuretics (79 vs 103 mg furosemide equivalents per day, respectively; p=0.001) and greater weight loss (4100 vs 3000 g; p=0.02), but did not influence the need for another class of diuretic therapy (10% vs 15%; p=0.66). Mortality, readmission at 30 days, and length of hospitalization were unaffected.

In this trial, no major differences between dapagliflozin and placebo were noted in symptomatic hypotension (6.0% vs 9.6%, respectively) or oligoanuria (2.0% vs 1.9%), and no patients in either group experienced severe hypoglycemia.11 However, the small size of this trial and the rare frequency of adverse events complicate interpretation of these safety analyses.

Limitations of current trials

Important limitations are present in current trials evaluating use of SGLT2 inhibitors in ADHF. First, SOLOIST-WHF was stopped early because of loss of funding during the COVID-19 pandemic.8,12 To maintain power, the primary endpoint was changed from first occurrence of CV death or HHF to instead measure CV death and hospitalization and urgent visits for HF. Additionally, instead of using adjudicated events as intended, investigator-reported events were considered. Moreover, the first treatment dose was administered after discharge to approximately half of patients. Because of the dual mechanism of action of sotagliflozin (inhibiting SGLT2 and SGLT1, which delays intestinal glucose absorption), it is unclear whether inhibitors of only SGLT2 would offer benefit similar to that demonstrated in SOLOIST-WHF. Lastly, sotagliflozin is not approved for use in the US, and findings of SOLOIST-WHF cannot be directly implemented in practice. Although approved in Europe, the regulatory status of sotagliflozin in the US is in flux, and may become clearer in 2023.13

The EMPULSE trial provided additional information through its reporting on outcomes both for patients with and without diabetes, indicating consistent benefit in a broader HF population.9 Unlike SOLOIST-WHF, the results of EMPULSE may better reflect the effects of early treatment initiation because patients were randomized at a median of 3 days after admission.8,9 Nonetheless, EMPULSE required enrolled patients who were clinically stable and thus may not represent more frail patients with severe disease. EMPA-RESPONSE-AHF, like EMPULSE, suggest empagliflozin initiation during ADHF may be safe, but is limited by its small size and lack of major clinical endpoints.

The trial by Charya and colleagues is limited by its unblinded nature, and because information on death and readmission were gathered via telephone survey and not adjudicated.11 However, patients were enrolled earlier than in other trials, at a median of 17.5 hours after admission. Additionally, most patients had NYHA class III and IV symptoms, providing more information on effects in patients with more severe symptoms. The mean GFR was approximately 53 to 56 mL/min, representing effects in patients with greater baseline renal function.

Other studies
Several observational studies have evaluated the effect of SGLT2 inhibitors as a class effect in patients with ADHF.14,15 One retrospective study by Kambara et al compared outcomes in patients who initiated (n=12) versus did not initiate SGLT2 inhibitor therapy (n=19) during ADHF at a single institution in Japan.14 Patients were treated with empagliflozin (n=9) or canagliflozin (n=3). Major findings indicated that SGLT2 inhibitor use was associated with less frequent diuretic use at discharge (67% vs 100%; p=0.016), lower diuretic dosages (13 vs 34 mg furosemide equivalents per day; p=0.008), and lower incidence of acute kidney injury (16% vs 58%; p=0.031). Mean length of hospital stay did not significantly differ between SGLT2 inhibitor and conventional therapy (18 vs 20 days; p=0.512). No hypoglycemic events occurred in either group. While these findings comport with those of well-designed interventional trials, several major limitations remain. First, the retrospective nature precludes conclusions regarding causality. Additionally, SGLT2 inhibitors were initiated based on uncontrolled diabetes (defined as hemoglobin A1c ≥6.5%) rather than HF, which may not represent practical decision criteria when initiating SGLT2 inhibitors during ADHF.

Another retrospective study, reported in abstract form, evaluated the effect of SGLT initiation during ADHF in patients with HFrEF on the rate of SGLT2 prescription at 30 days.15 Using data from a single institution, 177 patients (6.6%) began SGLT2 inhibitor therapy during hospitalization. Compared with matched patients not initiating SGLT2 inhibitors, 30-day HF readmission rates were significantly lower in those initiating SGLT2 inhibitors (4.9% vs 22.7%; p<0.05). While also limited by its retrospective nature, this study offers additional information on HF readmission rates in a real-world sample.

Place in therapy
Currently, the 2022 guideline for management of HF by the American Heart Association, American College of Cardiology, and Heart Failure Society of America (AHA/ACC/HFSA) recommends chronic use of SGLT2 inhibitors in a variety of patient groups to reduce HHF and CV mortality.6 The SGLT2 inhibitors receive a class 1 recommendation (strong recommendation; high-quality evidence) in patients with HF and T2DM, as well as in those with symptomatic HFrEF without regard to comorbid T2DM. Similarly, in patients with HF with mildly reduced EF (HFmrEF), SGLT2 inhibitors receive a class 2a recommendation (moderate strength of recommendation; moderate-quality evidence) to decrease HHF and CV mortality, higher than the 2b recommendation (weak strength of recommendation; moderate-quality evidence) for angiotensin receptor-neprilysin inhibitor (ARNi), angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, mineralocorticoid receptor antagonists (MRAs), and beta blockers in this population. They also receive a class 2a recommendation in patients with HFpEF, again at a recommendation class higher than that for MRAs and ARNi.

The 2022 AHA/ACC/HFSA guideline provides no recommendations on use of SGLT2 inhibitors in patients with ADHF.6 Currently, traditional guideline-directed medical therapy is recommended, including loop diuretics for decongestion, adjuvant vasodilators for dyspnea relief, and inotropic support for cardiogenic shock. It is conceivable that the RCTs studying SGLT2 inhibitors in ADHF may support their inclusion in future guideline updates.

Future directions
Additional RCTs are underway that will expand the evidence base for use of SGLT2 inhibitors in ADHF. The double-blind, placebo-controlled DAPA-ACT HF-TIMI 68 trial will evaluate dapagliflozin 10 mg once daily or placebo.16 The primary endpoint will be CV death or worsening HF, which will add substantially to the evidence on SGLT2 inhibitors’ effect on major clinical endpoints. DAPA-ACT HF-TIMI 68 has an estimated enrollment of 2400 patients is estimated to conclude in May 2023.

The double-blind EMPA-AHF trial will evaluate patients with HHF at high risk of adverse events, comparing empagliflozin 10 mg and placebo.17 The primary endpoint of EMPA-AHF will also evaluate a win ratio for a hierarchical composite of death within 90 days, HF rehospitalization within 90 days, worsening HF during hospitalization, and urine output up to 48 hours after treatment initiation. This trial is expected to enroll 500 patients and to be completed in April 2023.

The multicenter, open-label DICTATE-AHF trial will compare dapagliflozin 10 mg once daily with usual care in patients with T2DM.18 Treatment in this trial is planned to begin within the first 24 hours of hospitalization and continue until day 5 or hospital discharge. Diuretic response, defined as the cumulative change in weight per cumulative loop diuretic dose, will be the primary endpoint. DICTATE-AHF includes a protocolized initiation and titration scheme for background loop diuretic therapy, which will be an improvement over protocols utilized in other trials. DICTATE-AHF is anticipated to enroll 240 patients and to be completed in January 2023.

Lastly, a trial of canagliflozin use in ADHF will randomize 180 patients, regardless of comorbid T2DM or LVEF, to open-label canagliflozin 100 mg or usual care within 24 hours of admission.19 The primary endpoint will measure cumulative mean daily diuresis up to day 5 or discharge, and additional outcomes will include laboratory measures, mortality, readmission, and safety events. Although the small sample size may provide limited information for clinical endpoints, this trial will provide information on canagliflozin, which is not represented in current evidence. This trial is anticipated to complete in November 2023.

The available studies on initiation of SGLT2 inhibitors during ADHF add to their broader body of evidence. Like that in T2DM, chronic HF, and chronic kidney disease, the evidence in ADHF reflects a clinically meaningful benefit that persists in important subgroups. The general consistency in findings of benefit for several agents may indicate a class effect, although the SGLT2 inhibitors differ in the amount of supporting evidence and outcomes assessed. Some studies are small and may not be powered to detect rare events such as ketoacidosis. However, the generally similar adverse event rate with SGLT2 inhibitors compared with placebo or standard care suggests the potential for their safe initiation during ADHF. Until definitive guidance is offered by professional society guidelines, use of SGLT2 inhibitors may be a consideration in patients hospitalized for ADHF who are properly selected and monitored. Additional evidence from ongoing RCTs beginning in early 2023 will provide greater clarity.


Prepared by:
Ryan Rodriguez, PharmD, MS, BCPS
Clinical Associate Professor, Drug Information Specialist
University of Illinois at Chicago College of Pharmacy

July 2022

The information presented is current as June 15, 2022. This information is intended as an educational piece and should not be used as the sole source for clinical decision-making.


  1. Invokana. Package insert. Janssen Pharmaceuticals Inc; 2021.
  2. Farxiga. Package insert. AstraZeneca Pharmaceuticals LP; 2021.
  3. Jardiance. Package insert. Boehringer Ingelheim Pharmaceuticals Inc; 2022.
  4. Steglatro. Package insert. Merck Sharp & Dohme Corp; 2022.
  5. American Diabetes Association. 9. Pharmacologic Approaches to Glycemic Treatment: Standards of Medical Care in Diabetes-2020. Diabetes Care. 2020;43:S98-S110. doi:10.2337/dc20-S009
  6. Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA Guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2022;145(18):e895-e1032. doi:10.1161/CIR.0000000000001063
  7. Reed BN, Gale SE, Cox ZL. Acute decomopensated heart failure. In: DiPiro JT, Yee GC, Michael Posey L, Haines ST, Nolin TD, Ellingrod VL, eds. DiPiro: Pharmacotherapy A Pathophysiologic Approach. 12th ed. McGraw Hill; 2021.
  8. Bhatt DL, Szarek M, Steg PG, et al. Sotagliflozin in patients with diabetes and recent worsening heart failure. N Engl J Med. 2021;384(2):117-128. doi:10.1056/NEJMoa2030183
  9. 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. doi:10.1038/s41591-021-01659-1
  10. Damman K, Beusekamp JC, Boorsma EM, et al. Randomized, double-blind, placebo-controlled, multicentre pilot study on the effects of empagliflozin on clinical outcomes in patients with acute decompensated heart failure (EMPA-RESPONSE-AHF). Eur J Heart Fail. 2020;22(4):713-722. doi:10.1002/ejhf.1713
  11. Charaya K, Shchekochikhin D, Andreev D, et al. Impact of dapagliflozin treatment on renal function and diuretics use in acute heart failure: a pilot study. Open Heart. 2022;9(1). doi:10.1136/openhrt-2021-001936
  12. American College of Cardiology. SCORED/SOLOIST-WHF: sotagliflozin proves beneficial in reducing CV events in hard-to-treat patients with diabetes, HF. American College of Cardiology. Published May 17, 2021. Accessed June 21, 2022.
  13. Liu A. Lexicon pulls FDA submission for sotagliflozin to fix ‘technical issue,’ plans prompt refiling. Fierce Pharma. Published June 15, 2022. Accessed June 21, 2022.,full%20rights%20to%20the%20drug.
  14. Kambara T, Shibata R, Osanai H, et al. Importance of sodium-glucose cotransporter 2 inhibitor use in diabetic patients with acute heart failure. Ther Adv Cardiovasc Dis. 2019;13:1753944719894509. doi:10.1177/1753944719894509
  15. Okoroike HC, Patel SV, Simone P, Lavelle RI, Szwak JA. Impact of inpatient initiation of sodium-glucose cotransporter 2 inhibitors on guideline-directed medical therapy for heart failure with reduced ejection fraction patients. J Am Coll Cardiol. 2022;79(9):282. doi:10.1016/S0735-1097(22)01273-6
  16. A multicenter, randomized, double-blind, parallel group, placebo-controlled trial to evaluate the effect of in-hospital initiation of dapagliflozin on clinical outcomes in patients who have been stabilized during hospitalization for acute heart failure dapagliflozin and effect on cardiovascular events in acute heart failure -thrombolysis in myocardial infarction 68 (DAPA ACT HF-TIMI 68). identifier: NCT04363697. Updated December 20, 2021. Accessed May 20, 2022.
  17. A multicenter, randomized, double-blind, placebo-controlled study to evaluate the efficacy of empagliflozin in patients with acute heart failure. identifier: NCT05392764. Updated May 26, 2022. Accessed May 20, 2022.
  18. Cox ZL, Collins SP, Aaron M, et al. Efficacy and safety of dapagliflozin in acute heart failure: Rationale and design of the DICTATE-AHF trial. Am Heart J. 2021;232:116-124. doi:10.1016/j.ahj.2020.10.071
  19. Efficacy and safety of early initiation of canagliflozin in patients with acute decompensated heart failure. identifier: NCT05364190. Updated June 8, 2022. Accessed May 20, 2022.