Video

HFA 2026: Restrictive vs Liberal Oxygenation in Acute HF — REDOX-AHF

Published: 15 May 2026

  • Views:

    Views Icon 40
  • Likes:

    Heart Icon 0
Average (ratings)
No ratings
Your rating
View Transcript

HFA Congress 2026 — Dr Ida Arentz Taraldsen (Amager and Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, DK) discusses REDOX-AHF, a prospective, randomised, blinded, multicentre pilot trial evaluating restrictive versus liberal oxygenation targets in patients hospitalised with acute heart failure and pulmonary congestion.

Although oxygen therapy is widely used in the acute management of heart failure, there has been a lack of randomised controlled trials comparing different oxygenation strategies using clinically relevant endpoints in this population. REDOX-AHF enrolled 122 patients in the emergency department and randomised them 1:1 to a liberal oxygenation strategy (SpO2 target 96%) or a restrictive strategy (SpO2 target 90%), delivered through automated oxygen titration using the O2MATIC system. The primary endpoint was change in pulmonary parenchymal fluid content at 24 hours, measured non-invasively by remote dielectric sensing.

Findings showed a greater reduction in lung fluid content in the restrictive group compared with the liberal group, without differences in serious adverse events, and with a signal toward more days alive and out of hospital at 30 days in patients treated with a restrictive oxygenation strategy, in the context of a small pilot sample.


Interview Questions:
1. Why has the question of optimal oxygenation targets in acute heart failure remained unanswered for so long, and what was the hypothesis driving REDOX-AHF?
2. Provide a brief outline of the trial design and address how the O2MATIC automated delivery system ensured blinding and precision.
3. What were your primary endpoint findings — how did the two oxygenation strategies compare in terms of pulmonary fluid content at 24 hours?
4. What did the secondary outcomes reveal, including mortality, days alive out of hospital, and biomarker and imaging data?
5. Were there any safety signals or cases where the intervention needed to be stopped early, and what does this tell us about the feasibility of a restrictive approach?
5. How should these findings inform the way clinicians approach oxygen administration in acute heart failure in the emergency department?
6. What would a larger, definitive trial need to look like, and what endpoints would you prioritise?

Recorded on-site at Heart Failure Association Congress 2026, Barcelona.

Editors: Jordan Rance, Mirjam Boros
Videographer: David ben-Harosh, Tom Green


Support: This is an independent interview produced by Radcliffe Cardiology.

Transcript

My name is Ida Arentz Taraldsen, and I’m a doctor from Amager and Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, Denmark.

Well, I think the first part is remembering to ask the simple questions about the things that we are doing, uh, that are not that complicated but still a part of the treatment. For example, oxygen treatment.

Uh, and then, well, we were just looking at the guidelines of acute heart failure, and the evidence level is so low in so many points, and here we don’t really have any randomized controlled trials that can direct us. So that was the reason.

So, it’s a randomized controlled trial, and we randomized them during 24 hours to a restrictive or a liberal oxygenation target. The restrictive target was 90%, and the liberal oxygenation target was 96%.

And we blinded it with automatic, which is an automated oxygen delivery system that has a finger clamp measuring SpO2, and then it automatically adjusts the oxygen levels from this point.

Yeah. So our primary outcome was pulmonary fluid content measured by ReDS, which is remote dielectric sensing, which is this wearable device that you have on for like 30 seconds, and it measures lung fluid content.

Uh, and we found a greater reduction in the restrictive group compared with the liberal group, and the difference was 2.5%, and the p-value was 0.01.

So our secondary outcomes included arterial blood gas oxygenation levels after 24 hours, which was preceded by 10 minutes without oxygen, uh, and then showed a tendency towards a better oxygenation in the restrictive group as compared to the liberal, but was not statistically significant. The p-value was 0.06.

The other secondary outcomes, which were a change in respiratory rate, um, the length of, um, uh, until when you were off oxygen, um, and, um, that was not statistically significant. There was no change.

And the last one was the change in dyspnoea and the Likert score, and that was also not, uh, not statistical.

Uh, well, we looked at the number of serious adverse events. There was no change between the groups, and, uh, there were tendencies to, uh, more deaths in the liberal group as well as longer in-hospital stay, but without statistically significant differences.

Uh, but then in days alive and out of hospital, there were two days more in the restrictive group as compared to the liberal group to 30 days, with a p-value of 0.04.

Well, I think that we definitely don’t have to insist on giving a liberal oxygenation strategy. The guidelines already say that we should just give it if the saturation level is below 90%. So going more for that than the liberal strategy that is often what’s happening outside in the hospitals is probably better.

Uh, well, I think it should be clinical outcomes that would be very nice. So, powered studies on days alive and out of hospital, which did show a difference but in a small sample size, and then of course length of in-hospital stay, mortality, cardiovascular death, and so on.

Thank you very much.”

Dr Ida Arentz Taraldsen
Amager and Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, DK

Notes
The affiliation and name match publicly available records for Dr Ida Arentz Taraldsen, including her identification as an MD/PhD student working on acute heart failure and oxygen treatment, and conference-related references to her REDOX-AHF presentation. The trial topic and oxygen-target framing are also consistent with published material associated with her work.

I can also turn this into a verbatim transcript, a clean publication transcript, or a caption-ready version.

remove spoken uh and hesitation

Certainly — here is the cleaned full transcript with spoken fillers and hesitation removed, while keeping the original meaning and flow intact. Dr Ida Arentz Taraldsen is affiliated with Copenhagen University Hospital–Amager Hvidovre Hospital, and her work concerns oxygen treatment in acute heart failure.

Dr Ida Arentz Taraldsen, Amager and Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, DK:

“My name is Ida Arentz Taraldsen, and I’m a doctor from Amager and Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, Denmark.

I think the first part is remembering to ask the simple questions about the things that we are doing that are not that complicated, but are still part of the treatment. For example, oxygen treatment.

We were looking at the guidelines for acute heart failure, and the level of evidence is very low at many points. We do not really have any randomized controlled trials that can guide us, and that was the reason for doing this study.

It is a randomized controlled trial, and we randomized patients during 24 hours to a restrictive or a liberal oxygenation target. The restrictive target was 90%, and the liberal oxygenation target was 96%.

We blinded it with an automated oxygen delivery system that has a finger clamp measuring SpO2, and then automatically adjusts the oxygen levels from that point.

Our primary outcome was pulmonary fluid content measured by ReDS, which stands for remote dielectric sensing. This is a wearable device that you wear for about 30 seconds, and it measures lung fluid content.

We found a greater reduction in the restrictive group compared with the liberal group. The difference was 2.5%, and the p-value was 0.01.

Our secondary outcomes included arterial blood gas oxygenation levels after 24 hours, preceded by 10 minutes without oxygen. This showed a tendency toward better oxygenation in the restrictive group compared with the liberal group, but it was not statistically significant. The p-value was 0.06.

The other secondary outcomes were change in respiratory rate and time until patients were off oxygen. These were not statistically significant, and there was no difference between the groups.

The final secondary outcome was change in dyspnoea on the Likert scale, and that was also not statistically significant.

We looked at the number of serious adverse events. There was no difference between the groups, and there were trends toward more deaths in the liberal group, as well as longer in-hospital stay, but without statistically significant differences.

However, for days alive and out of hospital at 30 days, there were two more days in the restrictive group compared with the liberal group, with a p-value of 0.04.

I think these findings suggest that we definitely do not need to insist on giving a liberal oxygenation strategy. The guidelines already say that we should give oxygen only if the saturation level is below 90%. So moving more toward that strategy than the liberal approach that is often used in hospitals is probably better.

I think it would be very valuable to study clinical outcomes in a larger trial. Days alive and out of hospital did show a difference, but in a small sample size. Then, of course, outcomes such as length of in-hospital stay, mortality, cardiovascular death, and so on would also be important.

Thank you very much

Comments

You must be to comment. If you are not registered, you can register here.