By Michael H. Crawford, MD, Editor

SYNOPSIS: By adding dapagliflozin to maximally tolerated standard therapy for heart failure with reduced left ventricular ejection fraction, researchers noted the reduction in mortality and recurrent heart failure began within one month of starting this therapy.

SOURCE: Berg DD, Jhund PS, Docherty KF, et al. Time to clinical benefit of dapagliflozin and significance of prior heart failure hospitalization in patients with heart failure with reduced ejection fraction. JAMA Cardiol 2021;6:499-507.

Hospitalization of patients with heart failure caused by reduced left ventricular ejection fraction (HFrEF) is a known adverse prognostic event. However, little is known about the relative benefits of adding treatment with dapagliflozin (dapa) to standard therapy in such patients in relation to a prior hospitalization. Investigators from the DAPA-HF trial examined the data to analyze the timing of the demonstrated clinical benefits from the onset of therapy and whether the risk of clinical events varied as function of proximity to a heart failure hospitalization.

DAPA-HF was a multinational, double-blind, placebo-controlled, randomized trial of HF patients with symptoms, EF < 40%, and elevated NT-proBNP on optimal standard medical therapy. Those enrolled were free from renal dysfunction, type 1 diabetes, and short-expected survival. Patients were excluded if they were in the hospital for HF — or, within four weeks of said hospitalization. The main trial exhibited a reduced risk of the combined primary endpoint of worsening HF and cardiovascular death (HR, 0.74; P < 0.001). For the Berg et al analysis, the patients were divided into three groups: no prior hospitalization (16%), a prior hospitalization more than one year ago (27%), or a prior hospitalization less than one year ago (36%). The total population included 4,744 patients, 23% of whom were women (average age was 66 years). There was an early separation of the Kaplan-Meier curves for the primary endpoint after 28 days of dapa therapy (HR, 0.51; 95% CI, 0.28-0.94; P = 0.03), which was driven mainly by a reduction in worsening HF. A prior HF hospitalization had occurred in 27% of patients in < 12 months from trial onset. These patients presented with higher baseline New York Heart Association class and used more diuretics.

In the placebo arm, there was a stepwise gradient of risk of the primary outcome by timing of the most recent hospitalization: 21% with no prior hospitalization, 25% with a prior hospitalization more than one year ago, and 34% with a prior hospitalization less than one year ago (P = 0.04). A multivariate analysis of the primary endpoint in the placebo group showed an increase in the hospitalization more than one year ago group (HR, 1.08) and a larger increase among the prior hospitalization less than one year ago group (HR, 1.30; P = 0.04) vs. the no prior hospitalization group, again driven by worsening HF rather than mortality. Dapa lowered the relative risk of the primary outcome by 16% in the no prior HF hospitalization group (HR, 0.84), by 27% in the prior hospitalization more than one year ago group (HR, 0.73), and 36% in the prior hospitalization less than one year ago group (HR, 0.64; P = 0.07).

The authors concluded that in the DAPA-HF study population, dapa therapy produced a rapid decrease in the risk of worsening HF or death, with a sustained benefit emerging within one month of treatment onset. Also, the patients with a more recent hospitalization for HF before the trial onset were at a particularly high risk and benefitted most from the addition of dapa therapy.


A significant problem in the management of HFrEF, especially in the outpatient arena, is clinical inertia. The patient is doing well, so why rock the boat and risk adverse effects? We often do not have the resources to start a new drug that requires more intensive follow up. Then, there is the cost of new pharmaceuticals that are not generic yet. Finally, patients hate taking more drugs. Yet here is a new treatment that clearly not only benefits HFrEF patients, but the clinical benefit is rapidly apparent. If the patients have been hospitalized recently, this course helps prevent readmission for worsening heart failure. Thus, we may need to up our game and figure out how to efficiently put patients on dapa or one of the other sodium glucose cotransporter 2 (SGLT2) inhibitors. As the Berg et al study shows, the sooner we start, the better.

Another stumbling block for clinicians is how to sequence the many drugs now recommended for HFrEF patients. Most patients are treated with a renin-angiotensin inhibitor and a beta-blocker, but the addition of other agents has lagged for some of the reasons mentioned already. Also, physicians are loath to just slam a patient with a handful of drugs all at once and for good reasons. Yet there is unclear guidance on what to start first and what to add next in the HF patient. Recently, McMurray and Packer suggested a somewhat radical approach: start with a beta-blocker and an SGLT2 inhibitor, then add an angiotensin receptor neprilysin inhibitor, then add a mineralocorticoid receptor antagonist.1 This is sage advice from two respected authorities in the field, but it has not been tested in a randomized trial.

The Berg et al analysis of the DAPA-HF trial was post hoc and must be considered hypothesis-generating. Another limitation is the time after hospitalization was dichotomized at one year rather than considered as a continuous variable. Also, patients in the hospital for HF or within four weeks of said hospitalization were excluded. By this trial design, the authors are not recommending starting an SGLT2 inhibitor until after the early post-discharge medication adjustments have been accomplished and renal function has stabilized. The schema proposed by McMurray and Packer would not be applicable since we apparently must wait until one month after a HF hospitalization to start an SGLT2 inhibitor. Clearly, we need more information to assimilate these new data into our treatment algorithms. 


  1. McMurray JJV, Packer M. How should we sequence the treatments for heart failure and a reduced ejection fraction?: A redefinition of evidence-based medicine. Circulation 2021;143:875-877.