I Welcome to Skeptics guide to emergency medicine. Meet Greet, treat him and Freedom him, Today's date is 07/17/2024 and I'm your skeptical host Can Mill. The title of today's podcast is just 1 year. Pre oxygenation prior to innovation, and our guest skeptic is doctor An Y. She is an emergency physician practicing in Seattle, Washington area for over 2 decades.
She is the former president of the Washington chapter of the American College of emergency physicians and her career focus outside of clinical practice has been largely devoted to health care policy. Welcome back to the s An. Thanks, Ken delighted to be here. Well, I know people just love it when we talk about the weather and all that stuff to start up podcast. So I'm not gonna do that. It's my mom's birthday today. Oh, happy birthday, mom, if you're listening. Happy birthday, mom Mil.
Yep. So I'm gonna be going over toro, we're getting together for a little cake and Ice cream with some of my brothers and sisters. So... Yeah. Happy birthday mom, But let's get to a case. So, Ken, I'm working in my emergency department and a 70 year old man presents with an ex with Copd. He's hypoxia on arrival with an oxygen saturation of 80 percent of room air.
Although his oxygenation improves to 92 percent with supplemental oxygen by nasal cannula, he des saturate with minimal exert, and it shows increasing fatigue, indicating a need for greater respiratory support, but he refuses non invasive mechanical ventilation as he's not tolerated in the past, but he is agreeable intubation, His wife who's been watching the monitors like a hawk says she's worried. Doctor, his oxygen levels keep dropping. How will you in debate him without them going too low?
Yes. Having an audience. Pre oxygenation is a critical step in the management of airway interventions. Especially in the critically ill patients undergoing p. This process involves the administration of supplemental oxygen before the ind of anesthesia to increase the oxygen reserve in the lungs, thereby reducing the risk of hypo. Hypo during intubation can lead to severe complications including d, cardiovascular collapse, Brain injury and death.
Therefore, effective pre oxygenation is essential to enhance patient safety during this high risk procedure. And traditionally, pre oxygenation has been achieved using an oxygen mask, which can deliver a high action of inspired oxygen or F 2 under ideal conditions. However, oxygen masks have limitations, They do not provide positive pressure or
ventilator support. And the actual F 2 delivered can be significantly lower than expected if the mask does not fit well or if the patient is wearing it over their eyes, which we've all seen. This can be particularly problematic in critically ill patients who often present with compromised respiratory function and may not tolerate the procedure well.
So an alternative method of pre oxygenation, is the use of non invasive ventilation, which includes devices like continuous positive airway pressure, cpap, or bi level positive airway pressure. Bi, non invasive ventilation offers several advantages over traditional oxygen mask. It delivers high F 2 and provides positive pressure support, which can help maintain airway weight pate
and improve ventilation. This is particularly beneficial for patients who are critically ill and the risk of saturation is high. It requires more time set up, though, and it may increase the risk of aspiration in certain patient populations. Well Ken, recent studies explored the efficacy of non invasive account ventilation compared to traditional oxygen masks in critically ill
patients. The findings suggest that non invasive ventilation significantly reduces the incidence of hypo during intubation. These results are promising and may influence fe future guidelines and clinical practices, emphasizing the importance of optimizing pre oxygenation strategies to improve outcomes in critically ill patients. Hover, the optimal pre oxygenation strategy to reduce the risk of hypo and potential harm from it has not yet been determined.
Alright, On. What's the clinical question We're gonna try to answer on today's podcast? So among critically ill adults, undergoing trachea intubation, will, pre oxygenation with non invasive positive pressure ventilation reduce the incidence of hypo between induction to 2 minutes after trachea intubation compared to pre oxygenation with face mask oxygen. I need, like, some music there like that goes, some dramatic music or something like that. Alright. What's the reference?
The references gives it all non invasive ventilation for pre oxygenation during emergency intubation or the pre trial. From the new England Journal of Medicine June 20 24. Alright. Let's run through the peacock. What was the population? Critically ill adults, 18 years and older undergoing trachea intubation with a laryngeal and s sudan.
And they had some exclusions, pregnant women, wall prisoners, those already receiving positive pressure ventilation, and those with conditions including non invasive ventilation, which we will talk a little bit more in the Talk nerdy section. How about the intervention? The intervention was pre oxygenation with a non invasive ventilation using a tight fitting mask connected either a conventional or dedicated non invasive ventilator
for 3 to 5 minutes before induction. And what did they compare it to? The comparison was pre oxygenation using the oxygen mask delivered var non mask or using a bag valve mask device without ventilation. Alright. Let's run through their outcomes. What was their primary outcome of interest in this trial?
The primary outcome was hypo during intubation which was defined as an oxygen saturation, less than 85 percent during the interval between induction of anesthesia and 2 minutes after trachea intubation. And their secondary outcomes? Lowest oxygen saturation during the interval between induction and 2 minutes after trachea intubation. There were also several exploratory outcomes, including outcomes related to procedural complications and clinical outcomes including mortality.
There were also outcomes related to patient safety. These are listed in supplemental materials. Alright. And the t of the p cut, what type of study was this? This was a pragmatic multi centered un blinded, randomized parallel group trial. Alright. The author's conclusions were, quote, among critically ill adults undergoing trachea intubation, pre oxygenation with non invasive ventilation resulted in a lower incidence of hypo during intubation, than pre oxygenation with an oxygen mask end quote.
Alright. Let's run through the quality checklist for Rc. First question is always, are these the patients that we see in the emergency department. So yes, and also no. This included emergency department patients, but it didn't exclusively include Ed patients. And in fact, the patient cohort came from 7 emergency departments, which accounted for 27 percent of the patients enrolled, but 17 intensive care units which included 73 percent of the population.
So it was like, 1 quarter Ed patients and 3 quarters Icu patients or there were 3 times as many Icu patients compared to Ed patients. But they had Ed patients in the mix. Correct. The patients were they adequately randomized. Yes. Patients were randomized to either non invasive ventilation or an oxygen mask. Randomization was performed with the use of perm muted blocks of variable size and was strat according to the trial site. Did they conceal the process of randomization?
Yes. The group assignments replaced and sequentially numbered opaque envelopes and remained concealed until after enrollment. The patients were analyzed in the groups to which they were randomized? Yes. They did an intention to treat analysis. And did they recruit patients consecutively into the trial? We're not a hundred percent sure about that. They don't document explicitly consecutive recruitment, but all eligible patients were screened,
and those meeting inclusion criteria were enrolled. The clinician, though could exclude patients at their discretion. If they felt that the patient required intubation so urgently that enrollment and pre oxygenation processes could not be performed. In the patients in both groups, you think that they were similar with respect to pro factors? Yes. They did a very thorough analysis there and both groups were very similar with respect to baseline characteristics and pro factors.
And let's talk about masking. Were all participants that means the patients, clinicians and outcome assessor Were they unaware of what groups they were allocated to? Well, the patients were unaware unaware because they were given anesthesia. But once randomized, it was impossible to conceal the nature of the inter... Intervention.
Most outcome assessment including the primary outcome was performed at the bedside, by the operator and other clinical personnel who are acting as, quote, designated observers, unquote, using standardized data collection tools, and this could possibly have introduced some bias. And do you think all groups were treated equally except for the intervention? Yes. And protocol appearance adherence was extremely good with 97 percent of patients receiving the treatment to which they were randomized.
And how is the follow up? Well, considering the study endpoint with 2 minutes after trachea intubation. You might be surprised to learn that of 1301 patients enrolled, 1301 patients were included in the primary outcome analysis. So they lost a nice round number of patients. That's right. 0 patients were lost. Alright. All patient important outcomes do you think they were considered? No, Ken, and I gotta set off here. The primary outcome here in this study was a monitor oriented outcome or?
No. But it wasn't a patient oriented outcome. Pew. No. Sadly. Important poo such as death, cardiac arrest and severe pre peri intubation hypo were recorded and analyzed, but the study wasn't powered to determine a difference in these outcomes. Which are relatively unfortunately very rare. It is reasonable to consider the primary outcome to be sort of a surrogate a proxy for the important pools but this is not definitively established. Alright. The eleventh question, the treatment effect wasn't
large enough. And precise enough to be clinically significant. Yes. And a final question, talk to me about financial conflicts of interest. Where did the funding come from? The authors had no financial conflicts of interests interest to disclose. The study was primarily financed by a grant from the Us Department of Defense. Alright. Let's get to result, then you already mentioned that they recruited just over 1300 critically ill adults. The mean age was 61 years, and it
was a 60 40 split male female. What was the key result? Hypoxia was significantly lower in patients treated with non invasive ventilation before intubation compared with usual pre oxygenation care. Alright. Let's drill down into some numbers here, briefly. For their primary outcome that hypo. What could they fine? So for the primary outcome, hypo occurred with in 9.1 percent of patients in the non invasive ventilation group versus 18.5
percent in the oxygen mask group. The absolute risk difference of 9.4 percent, with 95 percent confidence interval between a 13 percent and 5.6 percent with a p of less than point 001. So what you're saying is it was statistically significantly different with that absolute risk difference of 9 percent. How about the secondary outcomes?
The secondary outcomes were less citing the median lowest o 2 saturation was statistically different at 99 percent versus 97 percent with a median difference of 2 percent. With a confidence intervals ranging between 1 and 3 percent. But several exploratory outcomes had very interesting findings. Okay. Well, maybe we'll talk a little bit about those in the talk nerdy section, which I love talking nerdy to guest skeptics.
So... But before we talk Nerdy, really wanna recognize that this was a really well conducted trial. I know we're Skeptics. I know we do, quote, critical appraisal of the literature, but we're not gonna be critical Of the study design, they did a really good job. The authors are to be complemented on successfully designing, conducting, completing and publishing the trial. So well done.
Yeah. Ken I've commented before that you're as much a methodology as anything else and the methodology of the study is really very rigorous and very clean. Yeah. And I understand why they made certain choices like going for surrogate outcomes when a primary outcome of a patient oriented outcome of mortality, which would obviously be very clinically significant or survival with good neurologic function. They would be so rare that the study size, the population, like the actual
number. It would have to be this huge study, and that may have made it just very, very unreasonable to do it. So they have to use a surrogate. So I get it, it's certainly not a criticism. It's just... We need to know that when we're evaluating the study. Having said that, I think it is reasonable to focus our first nerdy point on the exclusions and some potential for selection bias. First of all, it's worth noting that they excluded pregnant women, which is a common problem
in medical research. And again, we understand why that choice is made, but certainly does limit the general of a lot of research to this important and vulnerable population. They also excluded 840 out of about 4500 eligible patients, which is about 18 percent because they were already receiving non invasive ventilation. And this arguably could have been the group might have most benefited from the intervention.
But if that speculation is correct, It implies the study might have actually been biased against the intervention, weakening the magnitude of the observed benefit, which perhaps should increase our confidence in the study results. And given the widespread use of non invasive ventilation in patients with respiratory failure, It's actually difficult to conceive of how such a study
without that solution could be ethically constructed. Yeah and we have to be careful because arguably, speculation, may, all, those words in there. Sure I can understand it would be a hypothesis to think that way. But certainly, we can't make that conclusion. The largest group to be excluded was if the clinician, and I'm not rating clinicians, clinicians
should be involved to make clinical judgments. But if the clinicians felt the patients needed urgent intubation, they were excluded, and of the 4500 plus patient screened 20 percent of the population was excluded for this reason so 1 in 5.
Patients could also be excluded if the conditions that preclude non invasive ventilation, and 1 condition was if the clinician felt that the patient had severe agitation, and that represented another 5 percent or 1 in 20 of the possible patients that could have been enrolled that were eligible to be enrolled. They were excluded for this subjective reason.
These exclusions could have introduced some selection bias, making us less confident in the true point estimate of the observed effect size of non invasive ventilation in critically ill patients. A second nerdy point is that 1 limitation of the p trial is of masking or blinding. Masking patients and clinicians would be very difficult this type of study design, and this can include introduce several potential biases that could impact the validity of the study results.
1 potential bias would be performance bias, and this occurs when differences in the care provided to the participants in the different intervention groups are due to the knowledge of the intervention they are receiving. Clinicians who know that a patient is receiving non invasive ventilation may be more attentive and provide additional care potentially improving the outcomes.
Conversely, they might pay less attention to patients in the oxygen mask group which could negatively impact their outcomes, and this could result in overestimate of the effect size. Alright The third nerdy point is about emergency department patients we mentioned this earlier, only a quarter of patients who were included in the cohort came from the Ed. The authors of the p trial did perform a pre planned, excellent, subgroup analysis of Ed patients.
According to the protocol and statistical analysis plan registered in clinical trials dot gov. Oh yes, I go back and I look at that. They wanted to explore whether the effect of the intervention vary across different patient subgroups, including those in the Ed versus those in the Icu, very reasonable subgroup to consider. And the results from the Ed did show a point estimate favoring non invasive ventilation, but it was not statistically significant as the 95 percent confidence interval crosses
the line of 1 or no difference. And we will put the data in the show notes for people to look at. But I do wanna remind people that, well, subgroups could be hypothesis generating and stuff The overall trial effect is more likely to be true than any subgroup 1 way or the other. And then we come to what I think is 1 of the more meaty parts of this study, which is the secondary analysis and the exploratory outcomes. And there were quite a few predefined subgroups analyzed,
and that's important. This wasn't a retrospective ad hoc subgroup analysis, and that improves our confidence in them, but they're still not really powered to be considered as strongly conclusive. But this included patients with acute hypo respiratory failure, which was just under half of all the enrolled patients patients with very high body mass index. Patients with very high apache 2 scores, and looking at patients very by their
FF2 in the previous hour before intubation. But it's striking that all of the point estimates favored non invasive ventilation and all but 1 we're statistically significant. Now, while this is pretty interesting, we shouldn't be too confident in any conclusions from this data, and it really should be thought of as hypothesis generating. The authors collected data on many outcomes other than the primary and secondary outcome, so they
had exploratory outcomes. These 2 are best understood is hypothesis generating, and there were a few which were interesting as 1 might expect since the intervention reduced the incidence of hypoxia in the peri procedural time frame, it also reduced the incidence of severe hypoxia defined as a saturation of less than 80 percent and less than 70 percent by about 7 percent and 3 percent respectively. There was also an observed reduction in the incidence of cardiac arrest by about 1 percent.
For all of these outcomes that confidence interval did not cross 0 and they have some logical face validity in their relationship to the primary outcome. However, there was no observed difference in mortality. So the final thing we touched on this already was the moods and the pews. Which is the primary and secondary outcome where monitor oriented outcomes or moves or surrogate outcomes, not patient oriented outcomes pools.
And again, we've said before we understand why the authors chose hypo as a primary outcome, it could be clinically relevant while also being objective, measurable, immediate and actionable. In contrast, mortal mortality are relatively infrequent event compared to hypo especially within the short timeframe frame of an intubation procedure. To detect statistically significant difference in mortality rates between groups a much larger sample size will we required.
As my friend, Seth T likes to say, it's difficult to measure the efficacy of lifeboats on ships that don't sync. Oh, I like that. Seth, He's he's a smart guy. He is. Oh, shit. I said it online. U. This would increase the complexity duration and cost of the trial. Well, those were the 5 nerdy points we wanted to go through, can you comment now on the authors conclusions and compare them to the Conclusions?
Well, can, we agree with the conclusions, though, whether this translates to a patient oriented benefit is not established, and caution should be exercised to avoid over interpreting these results. How about giving an Sn sms gem bottom line? Non invasive ventilation can be used as an adjunct in preparing appropriate patients for anti intra intubation and can reduce transient hypo. And can you resolve the case you presented?
So I discussed a situation with the patient and his wife, and I recommended non invasive ventilation before intubation. He was in integrated successfully without any hypo? A good clinical outcome. Okay. And how about, the clinical application? How are you gonna take this pre trial and apply clinically in your practice. Well, the pre trial supports the use of non invasive ventilation for pre oxygenation during emergency
intubation and critically ill patients. So by adopting this strategy as our preferred oxygenation method, physicians can reduce the incidence of hypo during this high risk procedure. And how we you translate this at the bedside? For talking to the patient or talking to the patient's wife. Well, I tell the Wife I share her concerns about his oxygen level, and I tell the patient because I know you hated that tight mask.
But I think if you could just stand it for just 5 minutes, we can get your oxygen levels high enough that we can in intimate you more safely. And he's agreeable to this and I'm successful in debating without any complications. Okay. It's time to talk about the Keen contest. Dennis Ren did a pediatric emergency Medicine episode last week and so we don't have a keen contest attached to that. So we have a brand new keen contest question for this week. What is it on you?
Okay, Ken. So the k contest this week is Wow, mouth to mouth ventilation has been described for many centuries, when and by whom was the first mechanical non invasive positive pressure ventilation device invented. Oh, we could send some people down some rabbit holes for this 1. So if you know who and when the first mechanical non invasive positive pressure ventilation device was invented and send an email to me. The s
j at gmail dot com. With Keen in the subject line, the first correct answer will get a big shout out on the next s episode. There is some other fo on the pre trial from our friends over at first 10AM. That's justin Mo. And, of course, E Cri with Scott Wei. Take a look at what they think about the Trio trial. Well, thanks, An. You you suggested this trial, so thanks for suggesting it, and I'm glad we got you back on the est. Thanks, Ken. It was a lot of fun.
And when I saw this kind of roll across my Twitter feed, I still call it Twitter. It was kind of exciting to see a study that actually showed positive benefits that were actionable in the Ed, so I thought this would be a fun 1 to do. Appreciate you having me. I'm not surprised that you still call it Twitter because I do too. I'm just surprised you're still on it. I lurk. Yeah. It's a... There's there's been a big decrease in the number of people on X or Twitter.
Anyways, it's time for you to read the shi tagline line. Remember to be skeptical of anything you learn even if you've heard it on the skeptics guide to a emergency medicine. Back to everyone next week.