Sodium–Glucose Cotransporter 2 Inhibitors in Hospitalized Heart Failure Patients: No Time Like Now

This article refers to ‘Mineralocorticoid receptor antagonist use and the effects of empagliflozin on clinical outcomes in patients admitted for acute heart failure: Findings from EMPULSE’ by J.P. Ferreira et al., published in this issue on pages xxx. Sodium–glucose cotransporter 2 inhibitors (SGLT2i) have emerged as a cornerstone of heart failure management in the last decade. While initially heralded as a breakthrough for heart failure with reduced ejection fraction (HFrEF), subsequent randomized trials have shown this class to perform remarkably well in reducing non-fatal and fatal outcomes in patients with heart failure with mildly reduced ejection fraction (HFmrEF) and heart failure with preserved ejection fraction (HFpEF). Therefore, it is no surprise that subsequent investigations have focused on how to generalize the use of this drug class for heart failure patients. In this regard, EMPULSE was an important trial that showed clinicians that SGLT2i commencement in the inpatient setting is both safe and efficacious.1 In this double-blind randomized controlled trial, hospitalized patients with acutely decompensated heart failure were randomized to empagliflozin or placebo. Importantly, patients presenting with their first presentation of heart failure were included in this trial. The empagliflozin group had a superior win ratio compared to the placebo group for the primary outcome of time to all-cause death, number of heart failure hospitalizations, time to first heart failure hospitalization, and ≥5 point change in the Kansas City Cardiomyopathy Questionnaire after 90 days of treatment. Within this trial, the empagliflozin group had a 7.7% risk of acute renal failure, but there were no significant differences in adverse effects between the empagliflozin and placebo arms. Despite the promising results of the trials like EMPULSE, there are still key perceived limitations to starting neurohormonal medications among patients hospitalized for decompensated heart failure. Patients with ambulatory heart failure have been consistently shown to be on suboptimal levels of guideline-directed neurohormonal blockade.2, 3 Thus, it is likely that clinicians would also be hesitant to co-administer SGLT2i and mineralocorticoid receptor antagonists (MRA) in hospitalized heart failure patients due to their dynamic electrolyte balance, fluid status, and renal function. But do these concerns translate to real risks? Perhaps not. In this issue of the Journal, Ferreira et al.4 aimed to address this very concern with their post-hoc analysis of the EMPULSE trial. They sought to evaluate the safety and efficacy of empagliflozin according to usage of MRA. They used the win-ratio method to allow for hierarchical evaluation of the composite endpoint from the EMPULSE trial. Empagliflozin was superior to placebo in reducing the primary composite endpoint, regardless of MRA use. There were no differences in the incidence of hyperkalaemia, acute kidney injury, or adverse events leading to discontinuation between the MRA groups. This analysis does have several important limitations. Post-hoc analyses such as this one, which involve studies of subgroups, are generally not subject to power calculations in the original trial. Secondly, more empagliflozin users were receiving MRA at baseline. Additionally, in those on MRA at baseline, the duration of MRA therapy was also not known. It is plausible that the duration of MRA use may have a time-varying interaction with the effect of SGLT2i. Moreover, the inclusion of large numbers of HFmrEF and HFpEF patients produce important differences in the adjunctive therapies that the patients had at the time of the trial. This is borne out in the significant differences in the angiotensin blockade and beta-blocker use among the patients in the trial. This may also limit the ability to find differences in the endpoint between the MRA groups. So how does this analysis stack up against the available data? In short, clinicians should still be optimistic about the overall message of this investigation. Firstly, the conclusions from this analysis are supported by the EMPULSE trial itself. Post-hoc analyses of the DAPA-HF (HFrEF)5 and DELIVER6 (HFpEF/HFmrEF) trials also evaluated a similar question. There, the authors found that dapagliflozin use reduced the hazard of the primary composite endpoint of worsening heart failure or cardiovascular death without any interaction by MRA use. The incidence of renal adverse events and hyperkalaemia was not significantly different by MRA use in either analysis. Ferreira and colleagues also found similar results in post-hoc analyses of the EMPEROR-Preserved7 (HFpEF) and EMPEROR-Reduced8 (HFrEF) trials. In both post-hoc analyses, empagliflozin reduced the primary outcome of cardiovascular death and heart failure hospitalization similarly irrespective of baseline MRA use. Again, there was no significant difference in the incidence of adverse events. Thus, the directionality and magnitude of the effect seen in the other trial data are generally consistent with the current investigation. Finally, the current analysis also yields several avenues for future investigation. The order and rapidity with which neurohormonal blockade is instituted for heart failure patients is probably important but remains unclear. This practical consideration will affect the uptake of the SGLT2i class across the heart failure subtypes. Moreover, whether patient subgroups that have a greater tendency towards metabolic dysregulation9-11 might have a greater benefit from co-administration of SGLT2i with other cardio- and nephroprotective therapies requires further investigation. Finally, the mean estimated glomerular filtration rate of the abovementioned post-hoc analyses was approximately 60 ml/min/1.73 m2. Knowing how co-administration of SGLT2i with MRA in larger cohorts of patients with greater renal dysfunction is another important clinical consideration. In summary, the use of SGLT2i with MRA and their inpatient commencement appears to be efficacious and associated with few excess adverse events. The analysis by Ferreira et al.4 adds to the body of data that shows that clinicians should consider the commencement of SGLT2i across the spectrum of heart failure in both inpatient and outpatient settings. Ultimately, when it comes to SGLT2i initiation for the heart failure patient, there is no time like now. This work was supported by the American Heart Association (23TPA1140962). Conflict of interest: none declared.