Do SGLT2 Inhibitors DELIVER in All Patients With Heart Failure?

John M. Mandrola, MD


August 27, 2022

It's easy to forget that a medicine works by altering a biochemical signal. Whether this translates to a clinical benefit may depend on who takes the medication. The same medicine can produce different outcomes in different patients.

Take adrenergic inhibition with β-blockers. When given to patients with heart failure due to reduced ejection fraction, β-blockers provide amazing results, substantially reducing the probability of dying. But when given to patients with essential hypertension, β-blockers underperform against other drugs.

This is the background against which I will interpret the results of the DELIVER trial of the sodium-glucose cotransporter 2 (SGLT2) inhibitor dapagliflozin in more than 6200 patients with heart failure and a mildly reduced or preserved ejection fraction.

At the 2022 European Society of Cardiology (ESC) meeting, we learned (almost all) the specifics of the DELIVER trial.

The patients were 72 years old on average, and the median left ventricular ejection fraction was 54%. After 2.3 years of follow-up, patients receiving dapagliflozin had a lower rate of the primary outcome of cardiovascular death or worsening heart failure (16.4%) than did those receiving placebo (19.5%). The 3.1–percentage point absolute risk reduction equated to an 18% relative risk reduction (hazard ratio [HR], 0.82; 95% CI, 0.73 - 0.92; P < .001).

The statistically robust result was driven largely by a reduction in heart failure hospitalization or urgent heart failure visits (HR, 0.79; 95% CI, 0.69 - 0.91). The DELIVER investigators did not present the number of total hospitalizations, nor were they listed in the accompanying report in the New England Journal of Medicine.

The reduction in cardiovascular death with dapagliflozin was 0.9 percentage points in absolute terms and 12% in relative terms (HR, 0.88; 95% CI, 0.74 - 1.05).

Despite a high mortality rate in this older population, dapagliflozin did not result in a statistically significant reduction in overall death.


The DELIVER trial produced results that were similar to those of EMPEROR-Preserved. SGLT2 inhibition reduced heart failure events, mostly hospitalizations. There was no reduction in overall mortality, and the reduction in cardiovascular death was modest (the upper bound of the 95% CI crossed 1.0, so it's possible there is no reduction in cardiovascular death).

I am troubled by the lack of data on total hospitalizations. Total hospitalizations are critical to understanding the effect of these drugs. In EMPEROR-Preserved, heart failure hospitalizations represented less than 20% of total hospitalizations, which explains why the trial did not find a significant reduction in hospitalization for any cause.

If you are 72 years old, you don't distinguish that much between types of hospitalizations. You care only about not being in the hospital—for any reason. Similarly, you don't care about preventing one sort of death; you care about living longer.

Heart failure hospitalization can be an important surrogate endpoint only if its reduction is large enough to change total hospitalizations. This is generally the case for patients with reduced ejection fraction because heart failure is their dominant issue. But patients with heart failure and better ejection fractions are far more complex.

DELIVER and EMPEROR-Preserved reveal that SGLT2 inhibition in patients with better ejection fractions produces substantially less benefit than it does in patients with lower ejection fraction.

A word on the meta-analyses of trials that will come from this meeting. It makes no sense to me to combine trials of heart failure with reduced ejection fraction with those of heart failure and preserved ejection fraction. These patients are just too different.


SGLT2 inhibitors, like any drug, can have differential benefits depending on the patients. If I have a patient with heart failure and a normal ejection fraction who has another indication for SGLT2 inhibition, such as diabetes or chronic kidney disease, I will try to use the drug. I write "try" because, in the United States, it depends on insurance coverage.

If, however, the only indication is heart failure with preserved ejection fraction, I remain unconvinced that these drugs warrant the current expense—the price is the same no matter the reason for use.

If you care about disparities in health outcomes, you should oppose the reflexive or algorithmic use of low-value interventions. Because the pool of money to provide healthcare is finite, the argument to use SGLT2 inhibitors in all patients with heart failure belies the concern to improve equity of healthcare outcomes for all.

For more from | Medscape Cardiology, join us on  Twitter  and  Facebook

Meeting hashtag: #ESCCongress

John Mandrola, MD, practices cardiac electrophysiology in Louisville, Kentucky, and is a writer and podcaster for Medscape. He espouses a conservative approach to medical practice. He participates in clinical research and writes often about the state of medical evidence. Follow John on Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.