All right, good afternoon, everybody. Thanks so much for being here. I'm Deepshika Ishana. I'm an assistant professor in the Division of Pulmonary and Critical Care at Duke University. Today I'll be talking to you about three trials related to mechanical ventilation. I don't have any relevant disclosures, just that I receive research funding from the National Institutes of Health and the Doris Duke Charitable Foundation. And despite having picked three French trials, I would like to disclose that I do not have any connection to France. It just happened to be that they produced very good evidence this year. So the first of the three trials is the NICO trial. This trial investigated non-invasive airway management of comatose patients. This was published in JAMA just a few months ago in November of 2023. The central study question here was as follows. Is a conservative airway strategy of withholding routine intubation beneficial in patients with suspected poisoning and a Glasgow Coma Scale score of less than 9? So just for those of you who may not use GCS in your practice routinely, this would be a patient who, for example, only withdraws to pain, only opens their eyes to pain, and speaks but uses words that are inappropriate. So a patient that has quite a depressed mental status. Let's talk a little bit about the prior evidence base upon which this trial rests. So we know that greater than 20,000 patients with acute poisoning are intubated each year in the US alone. So this is a clinical problem of significant magnitude. There's not a great scientific rationale for intubating these patients. Really the reason that we do it is because we believe that they cannot protect their airway and there's a hypothesized reduction in aspiration pneumonia that may come from intubating these patients. But there's equipoise about this question because there are also many risks of intubation and mechanical ventilation, which is the trade-off that the authors sought to study. Prior evidence was observational in nature and, as a result, had a high risk of confounding by indication and had very mixed evidence. So some studies suggested benefit to intubation and others did not. Going through the details of the study design, the authors randomized 225 adult patients with suspected acute poisoning. So this is an important component of the study design that patients need not have confirmed intoxication but just suspected acute poisoning, which mimics our clinical practice. A GCS of less than 9 and no other indication for tracheal intubation except for the depressed mental status related to the suspected poisoning. They enrolled patients from 20 emergency departments and one ICU in France. Interestingly, pre-hospital recruitment was also possible. So emergency medical services enrolled a large majority of these patients. So the randomization was not done in a computerized manner. It was done by sealed envelopes that were also given to EMTs. Patients were assigned an equal proportions to the intervention and control arms. Randomization was stratified by the hospital and the two arms were to have intubation withheld at presentation versus usual care, meaning that the attending physician could decide whether or not to intubate the patient. The outcome, the primary outcome, was a composite of in-hospital death, length of ICU stay, and length of hospital stay truncated at 21 days. This was a hierarchical composite outcome, meaning that the outcomes were valued differently from each other. So hospital death was more important, in other words, than length of ICU stay and length of hospital stay, respectively. A couple of additional points about the study design. Pregnant and incarcerated patients were excluded, as they often are, because informed consent was waived for this trial. And then another important point is that patients who were suspected of having ingested a cardiotropic drug, such as a beta blocker or calcium channel blocker, were also excluded, as were patients who had an intoxication with a readily reversible substance, such as an opioid or a benzodiazepine. All patients, regardless of the trial arm, were able to be intubated if they met these criteria. So even a patient in the control arm for whom routine intubation was withheld, if they met any criteria for emergency intubation, they could still be intubated. So those included things like seizure, respiratory distress, the definition of which you can see there, vomiting, and shock. So these are the components of the primary outcome. You can see here that there were no deaths in either arm. Jumping down just to the last row for a moment here, 18% of patients in the intervention arm were intubated, and 60% of patients in the control group. I highlight that just to say that it wasn't 80% or 90% or 100% of patients in the control group who were intubated, which shows that there really is equipoise about this question. So even for patients for whom they could be intubated per the discretion of the attending physician, only 60% of those patients were intubated. As a result of that, only 40% of patients in the restricted group were admitted to the ICU, compared to 66% in the control group. And the median length of stay, of course, as a result, was zero for the patients in the restricted group, and 24 hours for the patients in the control group. So not substantially higher, but this may be important to patients and health systems. And the median length of stay was different by about 16 hours as well. This was an interesting component of the trial. So because this was a hierarchical composite outcome, the investigators calculated a win ratio, or win odds, to determine the efficacy of the restricted intubation arm. And not to go through this in too much detail, but I think that this is probably an analytic approach that we'll be seeing in trials more, so I just wanted to explain it briefly. So what happens in this sort of analysis is that each patient in the intervention group is compared to each patient in the control group. So basically, if there's 100 patients in each arm, you'd have 100 times 100 pairs of patients that could be compared. And for each of the outcomes, so for death, for example, for each of those pairs, the investigators determine who wins in this case. Is it the intervention, meaning that the patient survives, the patient in the intervention group survives, the patient in the control group dies? That would be an example of a win for the intervention. Or does the intervention lose for that pair? So the intervention patient dies, and the control group patient survives. So that's essentially what these proportions represent for each component of the outcome. They're tabulating where the intervention wins, where it loses, and where it's tied for each pair that they're investigating. So you can see that for deaths, because there were no deaths, all of these are ties. For length of ICU stay, the intervention is favored. For length of hospital stay, there's perhaps a slight benefit to receiving intubation. But overall, the win ratio and the win odds favor the intervention. And remember that I said, you know, one of the reasons that we intubate these patients is because there's a hypothesized benefit that we may be reducing aspiration pneumonia. And in fact, we see the opposite in this trial. So patients in the control arm, 14.7% of them developed pneumonia versus only 6.9% in the restricted intubation group. So I think this is just a really nice example of why it's necessary to challenge dogma and do trials. So limitations of this trial, it was an unblinded study, of course. There were no deaths in the study population. I think this is probably because the sample size was too small to capture such a rare event. There was a very pragmatic approach to including patients in the trial, because it was solely based on suspicion of acute poisoning. But that also mirrors our clinical practice. And so I don't think that this is necessarily a limitation. And some patients were intubated in the ED or the ICU, not the pre-hospital setting. So that may have influenced first pass failure rates and some of the other secondary outcomes. So in conclusion, restricted intubation of patients with toxic congestions appears to be beneficial primarily due to less use of ICU resources. And just remember that cardiotoxic ingestions were excluded from this trial, as were readily reversible ingestions. The second trial compared pressure support ventilation versus TPs for spontaneous breathing trials. And the central study question here was whether pressure support ventilation trials result in a shorter time to extubation without a higher risk of reintubation among patients who have a high risk of extubation failure. There was a prior RCT. I'm sorry. This was published in JAMA in the spring of last year. And there was a prior RCT also published in JAMA a few years ago that showed a higher successful extubation rate for pressure support ventilation compared to TPs. But there were two main limitations of that study that this current trial sought to address. So that prior trial included all comers who were mechanically ventilated, not patients who were high risk only. And I think in clinical practice, many people only use TPs trials for patients who are high risk. So there was a little bit of a discrepancy there. And then the duration of the interventions also differed, which I think really was the main limitation. Patients were exposed to 30 minutes of pressure support ventilation versus two hours of TPs. So that was a major limitation. And many people thought that the benefit of pressure support really may have just been from the difference in the duration of the intervention. So in this trial, they randomized 969 patients who had been mechanically ventilated for at least one day and were deemed ready for a spontaneous breathing trial using standard criteria and considered high risk for extubation, defined as either having an age greater than 65 years or chronic cardiac or respiratory disease from ICUs in France almost entirely during the pandemic, which certainly affected the patient population that was enrolled in this trial. Again, this was a one-to-one random assignment to SBT using one of those modes. And the pressure support settings were 8 centimeters of water pressure support with no PEEP. The TPs patients received approximately 40% FiO2. And both patients in both arms were exposed to these for one hour. So that duration question is answered or sort of addressed in this trial. This was repeated daily until the SBT was deemed successful by the investigators. So that's, I think, another really interesting piece of the study design is that the investigators, not the treating physicians, were the ones who deemed whether the patient had successfully passed their SBT or not. The decision to extubate, of course, was by the clinical team. The primary outcome here was ventilator-free days from the initial SBT for 28 days following the initial SBT. I'll just say here that there was very high fidelity to the intervention, only 1% to 2% crossover between groups. And one important thing to note is that prophylactic use of non-invasive ventilation or high-flow oxygen for 48 hours was highly encouraged in this group, or in this trial, rather. So here you see in the first row the primary outcome, median total time alive and without exposure to invasive ventilation, in other words, ventilator-free days was identical between the two groups. And importantly, was there a higher extubation failure among these groups? So that's the second red box down here. This is reintubation within seven days after extubation, again, very similar between the two groups. Many other secondary outcomes, which you can look through on your own time. But some limitations to this study, again, it was an open-label study. Patients were not blinded to the assignment of the patients, and it may not be generalizable to practice settings that do not use prophylactic non-invasive ventilation or high-flow oxygen following extubation. So the conclusion from this trial really is that there's no significant difference in outcomes between pressure support ventilation and TPS trials for SBTs and patients at high risk of extubation failure. So both could be effective methods for conducting SBTs in this patient population. And then the last trial investigated the continuation of enteral nutrition until extubation versus fasting before extubation. This was published in Lancet Respiratory Medicine in the winter of 2023. So just to reiterate the question there, how does continued enteral nutrition until extubation compare to fasting before extubation for patients in the intensive care unit? The prior evidence here, we know that extubation is a common cause of nutrition interruption in critically ill patients. And current approaches to nutrition preceding extubation are highly heterogeneous. So one study showed that on average, or I guess I should say the median duration of fasting prior to extubation is two hours. But you can see the interquartile range there ranges from 0 to 6. So 25% or more of practitioners are not interrupting at all. So this trial randomized 1,130 adult patients receiving mechanical ventilation for at least 48 hours and were receiving prepyloric enteral nutrition for at least 24 hours and judged by their attending physician to have met criteria for extubation in 22 ICUs in France. This was, again, an open label trial. This was cluster randomized, meaning that entire ICUs were randomized rather than individual patients. The reason that this was done is because there was a high risk of contamination between the two arms. So to continue enteral nutrition versus fasting for six hours with gastric suctioning if the lumen of the tube allowed. This was, again, a composite outcome of reintubation or death within seven days after extubation. And this was a non-inferiority trial, so the non-inferiority margin was set at 10%. This is a patient population. So they're quite high risk for extubation failure. Generally adult patients in medical ICUs with a median duration of mechanical ventilation of more than a week, and one third had one or more risk factors for extubation failure. So as you can see here, this is the primary outcome. And the dashed line represents the non-inferiority margin. Their 95% confidence intervals are substantially above that, demonstrating that these strategies are likely to be equivalent to each other. These are some of the secondary outcomes. You might wonder, for patients in the continued enteral nutrition group, was there a higher rate of pneumonia? And hopefully you can see this first row here, which shows that the rate of nosocomial pneumonia following extubation was quite similar between the groups. The only other one that is worth pointing out is the duration from the first successful SBT to extubation was only two hours in the continued enteral nutrition group, and was 18 hours in the fasting group. Some of this may have been imposed by the investigators, because after the SBT was done, that's when fasting began, and fasting needed to be at least six hours. So there was sort of a built-in delay into the trial. So I'm not sure that it's entirely fair to say that a fasting strategy necessarily increases duration of time to extubation. This was very important to note that this was part of the study design. And you can see that patients in the continued enteral nutrition group received more nutrition on the day of extubation. They received 482 calories versus 236, and had lower rates of hypoglycemia, as may be expected. So again, the limitations of this. This was another open-label trial. And there really was no justification for their non-inferiority margin, which I think is a very important component of non-inferiority trials. So that was the only thing that I wondered about a little bit. But regardless, their results suggest that the two strategies are equivalent to each other. So the main conclusion for this is that continued enteral nutrition until extubation seems to be a safe and potentially beneficial practice, because it results in patients receiving more nutrition, and critically ill patients receiving invasive mechanical ventilation. Thanks very much for your attention.

mechanical ventilation

intubation strategies

spontaneous breathing trials

enteral nutrition

ICU protocols