SGLT2 Inhibitors in Acute Heart Failure, The Earlier The Better? DICTATE AHF

Ileana L. Piña, MD, MPH; Zachary L. Cox, PharmD


September 20, 2023

This transcript has been edited for clarity.

Ileana L. Piña, MD, MPH: Hello, and welcome. I'm Ileana Piña, professor of medicine at Thomas Jefferson University, and the chair of Quality. I am here at the European Society meetings with a large amount of excitement and many new trials that are being presented. That's exactly why we have my friend here, Dr Cox, who comes from Vanderbilt, to talk to us bit about DICTATE-AHF.

If it starts with a D, then we know it's dapagliflozin. Tell me, why did we think a sodium-glucose cotransporter 2 (SGLT2) [inhibitor] would do anything in acute heart failure? We're terrible at acute decompensation. We're awful. Why did you put it together that way?

DICTATE-AHF: Dapagliflozin in Acute Heart Failure

Zachary L. Cox, PharmD: Thank you so much for the opportunity to talk with you. When we looked at the space of acute heart failure, as you just laid out, we know that starting an SGLT2 inhibitor in the ambulatory setting or at hospital discharge and continuing it chronically has widespread benefit across patients, regardless of ejection fraction.

However, what we didn't know was that when starting it acutely on the first day of heart failure hospitalization, (1) does it provide any acute efficacy, and (2) is it safe? There have been concerns that it may not be safe and concerns that there may be increased risk for hypoglycemia or ketoacidosis in an acute hospital setting.

Piña: What mechanism are you targeting? We don't even know why all the benefits of the SGLT2s happen. We know about glucosuric effects, and we know that the curves split apart very quickly. But what's the drug really doing in an acute setting?

Cox: That's a great question because although we don't know all the mechanisms that probably lead to the chronic benefits, for acute efficacy, we looked at the benefits across diuresis, natriuresis, and diuretic response, looking for any additive benefit acutely for a diuretic decongestive outcome and then safety across all the other outcomes.

Piña: Interesting. We're going for the diuretic. What happened? Tell me a little bit more about what patients you recruited.

Cox: We took 240 patients across six hospital sites in the US, randomized them within 24 hours of presenting to the hospital, to either starting dapagliflozin 10 mg daily or structured usual care. Both groups had structural usual care, meaning that their intravenous (IV) diuretic regimen was protocolized and titrated to 3-5 L of urine output a day. We continued that until hospital day 5 or discharge if sooner, with 30-day follow-up for safety events.

Piña: The diuretic that you protocolized could have been bumetanide or furosemide; it didn't matter?

Cox: That's correct. The IV loop diuretic was protocolized as was when the addition of thiazide could happen.

Piña: You followed the DOSE trial protocol — that if it was insufficient, then you could add the hydrochlorothiazide.

Cox: That's exactly right. 

Diuretic Efficiency and Safety Findings

Piña: Your goal was not a diuretic, rather your goal was the diuresis however you got there. Tell us the key findings.

Cox: The primary efficacy outcome that we looked at was diuretic efficiency. That's been calculated in many different ways, but we looked at change in weight over the study period, divided by the cumulative loop diuretic dose that was required to get this change in weight, which has been shown in many different cohorts to be prognostically significant. For safety, we had a blinded endpoint committee look at prespecified outcomes across diabetes, worsening renal function, across cardiovascular safety.

Piña: Did you look at things like hyperkalemia or hypokalemia?

Cox: We looked at severe electrolyte deficiencies. We also looked at genitourinary tract infections and really tried to provide the entire spectrum of potential concerns.

Piña: That was a busy safety committee.

Cox: It was. They had many events to look at. We also asked them to look at not only events but severity: Was this a severe event that prolonged hospitalization?

Piña: Tell me what happened.

Cox: We're pleased to say that, first, starting dapagliflozin on the first day was very safe. It did not worsen any cardiovascular endpoints nor worsen hypotension or severity of hypotension. It's also safe across all diabetes endpoints.

Piña: Did it change blood pressure at all? Did it have a vasodilatory affect at all?

Cox: There was no significant difference in mean blood pressure? There was no increased incidence of hypoglycemia or severity of this compared to usual care. There were no new urinary tract infections and no incidence of diabetic ketoacidosis.

Piña: How many diabetics did you have, or was everybody a diabetic?

Cox: In all, 71% of patients had type 2 diabetes.

Piña: Was it planned that way?

Cox: We initially planned for only type 2 diabetes patients to be in the study, and we amended the protocol a little over halfway through.

Piña: To allow others?

Cox: To allow people with or without type 2 diabetes; type 1 diabetes was excluded. Lastly, on renal outcomes, we found that this was safe and that it didn't worsen estimated glomerular filtration rate (eGFR) relative to usual care, which was a concern with SGLT2 inhibitors during aggressive diuresis. That was a very positive finding that it's safe to optimize guideline-directed medical therapy on the first day.

Relative to the efficacy, we found for the primary outcome of diuretic efficiency, this was in a P0 logistic regression model adjusted for baseline weight. The odds ratio favored dapagliflozin, at 0.65, but the 95% confidence interval upper limit exceeded 1. It was 1.01 with a P value of 0.06, so we just missed statistical significance for the primary outcome of diuretic efficiency.

Piña: There were not many patients when you put it into the larger context. Do you know what other medications the doctors were giving during this time?

Cox: At baseline, we looked at the three classes of guideline-directed medical therapy. On the first day of hospitalization, approximately 50%-60% of patients were on each of the other guideline-directed medical therapies, beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, or mineralocorticoid receptor agonists. Some of those, of course, may have been temporarily held or not yet initiated, but that's what it was on the first day of hospitalization.

Piña: You have the data so you can really get into the weeds and find out. You barely missed your statistically significant endpoint. What about symptoms? How did the patients feel?

Cox: We did not directly measure patient-reported symptoms. We did measure other measures of diuretic outcomes, as you mentioned. I think some that are important to highlight include that dapagliflozin increased 24-hour natriuresis. There was more salt being eliminated per 40 mg of IV furosemide. Importantly, we also found a shorter time to IV diuresis completion and a shorter time to hospital discharge. 

Mechanism of Action?

Piña: Is it acting to maybe, block the kidney somewhere that you get more sodium in the loop of Henle? There's something going on there in the receptors.

Cox: That's exactly what we think. The short-term diuretic acute benefits come from its effects on the proximal convoluted tubule blocking sodium reabsorption.

Piña: Somewhat like acetazolamide?

Cox: Yes. As you know, acetazolamide and SGLT2 inhibitors share a common pathway of the sodium–hydrogen exchanger 3.

Piña: The pH didn't change?

Cox: We did not measure urine pH in everyone reliably.

Piña: You didn't get any hyperchloremic effects?

Cox: No, we looked at all serum electrolytes. There was no severe worsening.

Piña: What was the potassium? You've got more natriuresis, so you're bound to have more hypokalemia, right?

Cox: That's right. We see that particularly with thiazides, but with adding an SGLT2 inhibitor, we only had one patient who ever had a serum potassium < 3 mEq/L. This was virtually no difference from usual care, where there was also one person.

Piña: The investigators could give potassium any time.

Cox: Investigators were free to do so. We measured B-type natriuretic peptides twice daily and potassium was repleted per the treating team.

Piña: What was the ejection fraction?

Cox: The ejection fraction was across the spectrum: 52% had heart failure with reduced ejection fraction, and the rest had mildly reduced to preserved.

Piña: You do have some preserves?

Cox: We do.

Piña: Male, female?

Cox: It was 60% male, so 40% female.

Piña: That's not bad. We should be very happy because 40% female is very good. We usually get 20% female.

Cox: We're very pleased with the representation of other groups of patients in this study.

Piña: What are you going to do next?

Cox: I think what's probably next is we hope that these findings will embolden clinicians that it's safe to start an SGLT2 inhibitor on the very first day, that there are some diuretic benefits that accrue immediately.

Piña: Natriuresis being better, I think, is a very significant point.

Cox: It's an important thing, and it was measured 24-hour natriuresis.

Piña: You missed that endpoint probably because your confidence intervals crossed one.

Cox: That's exactly right. We feel like the totality of the evidence supports that this is having a meaningful diuretic effect and that our sample size was just not large enough for the diuretic efficiency.

Piña: Is the next trial now planned a larger trial?

Cox: I don't think so. I think that demonstrating this was probably enough, and, resting on the expansive literature to date of SGLT2 inhibitors, we know that starting them on the first day is safe and that IV loop diuretics plus an SGLT2 inhibitor should be the first thing patients are started on. This really opens up the remainder of the hospital stay for optimizing other guideline-directed medical care.

Piña: We don't have an acute heart failure standard of care. We really don't. Everybody does what they've been trained to do. I'm one that gives IV nitroglycerin to lower central venous pressure, and I give constant infusion diuresis. That's just clinical experience of many years. It would be nice to think that, if we add this, you can actually get to better volume quicker that will allow you to get the patient out of the hospital faster.

Cox: That's right. We saw less IV loop diuretic dose was needed, fewer titrations, and shorter time to discontinuation of diuresis. Hopefully, like you said, we're expediting this.

Piña: I hope you write that up. We probably have some journals that would like to have that.

We've heard some very interesting data about a difficult time, where nobody knows what to do. We're telling our residents and fellows to get the patients out quickly. Everybody's looking now at length of stay, saying that we're doing so well with readmissions. Are we really? We need to get the patients out quickly. Maybe if we improve diuresis early in the course of the stay of the patient, we will get them out sooner. I think there is more to come.

Thank you for joining me today. This is Ileana Piña, signing off.

Ileana L. Piña, MD, MPH, is a heart failure and cardiac transplantation expert. She serves as an advisor/consultant to the FDA's Center for Devices and Radiological Health and has been a volunteer for the American Heart Association since 1982. Originally from Havana, Cuba, she is passionate about enrolling more women and minorities in clinical trials. She also enjoys cooking and taking spin classes.

Follow | Medscape Cardiology on X (formerly Twitter)

Follow Medscape on Facebook, X (formerly known as Twitter), Instagram, and YouTube


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.