Episode 11: Hyperoxia

In this episode I discuss hyperoxia and review the evidence in regards to the damage it can cause in patients with a wide variety of illnesses.

 

Slides to accompany the episode can be found here: hyperoxia slides

6 Replies to “Episode 11: Hyperoxia”

  1. This comment is a little late; I just heard the podcast. As a practicing anesthesiologist, I’m mostly interested in perioperative medicine, rather than ICU. However, a couple comments. You discussed a number of observational studies that indicate patients with hyperoxia (PaO2 > 200) and worse outcomes with STMI, COPD, and TBI. These are intriguing, but without RCTs to back up this data, I am quite skeptical of these results because it is likely that sicker patients were given more oxygen in these situations. So patients with hyperoxia appear to have worse outcomes because of the oxygen, not because they were sicker to begin with. You don’t mention any control for comorbidity or severity of illness in any of these studies, and it would really take large RCTs to really tease this factor out.

    Also, as pertains to perioperative use of oxygen, the data are in conflict with those from the ICU. A recent meta-analysis showed hyperoxia is associated with REDUCTIONS in surgical site infections (SSIs), mild improvements in PONV, with no significant worsening of atelectasis. (Anesthesiology. 2013 Aug;119(2):303-16). it would be important to provide this information as a counter balance to the data from the ICU, indicating that relatively short periods of hyperoxia in the OR may be beneficial.

    Thank you for your excellent podcasts!

    1. Hi Andrew,

      Thanks for the excellent comment. You make some really good points. I agree that there are not many RCTs looking at hyperoxia. I hope they get done. We do, however, have a good RCT in COPD: Effect of high flow oxygen on mortality in chronic obstructive pulmonary disease patients in prehospital setting: randomised controlled trial BMJ 2010;341:c5462. We also have the recent AVOID trial looking at hyperoxia in STEMI published in Circulation 2017: Air Versus Oxygen in ST-Segment Elevation Myocardial Infarction 135(10) which found hyperoxia was harmful in STEMI patients.

      Also, many of the observational studies looking at hyperoxia did try to control for comorbidities and severity of illness, though I agree with you that the only way to really eliminate these as confounders would be to do more RCTs.

      The article from Anesthesiology in 2013 is interesting but definitely has some drawbacks. For example, the PONV effect was mostly in those patients who didn’t get antiemetics. And while the SSI portion is interesting, there is definitely conflicting data and the effect that the meta-analysis found was borderline significant.

      But regardless, I think you bring up really good points that we should be talking and thinking about. Thanks again.

      Best,
      Jed

  2. Hi Dr Wolpaw,

    I’m a consultant anesthesiologist and critical care physician from Belgium (EU).
    I’ve discovered your podcast two weeks ago and I have to tell you that it’s a great podcast.
    The topics you’ve been presenting are very interesting.
    In fact, this one is one of my favourite because in all of the ICU and operating room, it is always the same problem. All of our patients are in a relative hyperoxia status…
    I definitely, have to explain it every single day to the nurses, residents and sometimes colleagues that hyperoxia is harmful to our patients…
    We obviously need to discuss about that issue more during the meeting, probably by producing RCTs in this field …
    Thanks again for all the time, energy,… you put in the podcast.

  3. This was awesome, I have considered hyperoxia my soap box as well in ICU. I mentioned a week ago, I’m an ICU nurse and CRNA student. I’ve had doctors shocked and excited when I weaned patients from 4 to 2 L/min or room air. I find anecdotally that people don’t find it important to make that jump from 2L to room air, maybe less exciting than your discuss but still a necessary and often overlooked step. 40% FiO2 is “minimal” in my unit, I am looking forward to starting to suggest 30% or even 21%, and seeing people look at me like I have 3 heads, because “thats just not how we do it”.
    How do you wean patients? I know you mentioned PaO2 >100 as hyperoxia, but how low do you tolerate PaO2 before you consider titrating FiO2 back up? I remember specifically a case when a patient was supposed to be extubated, the PaO2 was 55, and the attending (pulmonologists in my area) said that’s ok, extubate anyway because I can’t even prescribe home O2 unless the PaO2 is less than 50 on room air. Also do you require frequent ABG’s or is SPO2 enough?

    I look forward to trying 80% FiO2 in the OR instead of 100%, and I’m also excited to explain the risks of hyperoxia to everyone who calls me crazy because “that’s not how we do it”

    Another similar topic: how do you deal with metabolic acidosis in regards to ventilator patients? I’ve had a lot of discussions about this with mixed opinions. Do you turn the respiratory rate up to correct the pH or what is the conventional wisdom on this? Just like you, I’ve had respiratory therapists say they are increasing the rate on the ventilator, I say “Why, his PCO2 is normal, it’s metabolic?” and they say Dr. X is here this week and he wants it that way.

    As always, great podcast, very interesting, informative and well researched!

    1. Hi Mike,

      I’m glad to hear you are pushing for avoidance of hyperoxia. A lot of progress can be made just by bringing it up and bringing it to people’s attention. I talked to my residents and fellows about it the last time I was attending in the ICU and over the subsequent few days I found 2 or 3 patients on 21% fio2 on the vent on rounds. I was thrilled!

      I wean patients based on sats unless we’re getting ABGs for another reason in which case I’ll use those as well. We don’t know the answer to whether patients do better on room air even if their pao2 is in the 50s, but we do know that hypoxia is worse than normoxia, so with the exception of COPD patients I try to keep sats above about 92-94%. If sats fall below that, we’ll titrate the fio2 back up.

      Acidosis is another complicated issue and depends a lot on the patient and situation. In general, there’s no need to correct a metabolic acidosis all the way back to 7.4 by hyperventilating. In fact, oxygen delivery to tissues is improved in an acidotic environment. Better to find and treat the underlying cause. However, if the acidosis is severe, or the patient has a reason that they may not tolerate it (pulmonary hypertension) then some amount of hyperventilation may be necessary while pursuing the underlying cause.

      I hope that’s helpful. Thanks for all you do!

      Best,
      Jed

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.