Thank you very much for the introduction. So, I'm Stephan Ehrman, I'm a professor of critical care medicine at Tours University Hospital in France, and I thank the organizers, the Congress Committee, CCM, and the Lancet Respiratory Medicine who gave me the opportunity to present this work. It's a study which was performed within the CRICS-TRIGACEP research network with the support of the REWA network. It was funded by a grant from the French Ministry of Health in a call in west of France, inter-region. Here are my disclosures, I don't consider them related to this work. So, fasting prior to extubation in the intensive care is quite a common practice. If we look at this web survey performed in the UK, most ICUs impose two to six hours fasting prior to extubating their patients. Actually nearly all ICUs impose some sort of fasting. In this U.S. survey among training centers of anesthesia, anesthesia critical care, surgical critical care, and medical critical care, we can see that the duration of fasting imposed prior to extubation is actually decreasing from purely anesthesia departments to medical critical care. And so, this fasting practice is based on guidelines for shuttle surgery, but it has not really been evaluated in the ICU. Why do practitioners impose this fasting? It's because they have a perceived risk of aspiration at the time of extubation, and they're probably right to perceive such a risk. And this physiological study performed in 20 patients who had a systematic fiber optic examination within 24 hours of extubation, where some colored dye liquid is injected during the fiber optic examination, and those authors noted that half of the patient would aspirate, and in half of the cases, this aspiration was silent, meaning that there's not even a clinical sign. So the dye goes in the trachea, and nothing more happens. So patients are at high risk of aspiration after extubation in the ICU. In this larger study, about 1,000 patients, they also looked clinically to see if patients had dysphagia and risk of aspiration in the post-extubation period, and yes, it's very common. 12% of the patients three hours after extubation had clinically significant dysphagia, 12% at six hours. So you need to add the silent aspiration on top of that. So dysphagia is very frequent after extubation in the ICU. The question is, does it really matter to patients? And the response is probably yes. In this retrospective study on patients who underwent swallowing evaluation, you notice that those who have moderate or severe dysphagia after extubation will have significantly more pneumonia than those who had non- or mild dysphagia, and the same was observed for reintubation and even mortality. So dysphagia is associated with extubation failure, pneumonia, and mortality, so that's not something to neglect. In those epidemiological data on more than 100,000 patients in France over five years, you can see that there's a strong link between reintubation and nosocomial pneumonia, and we know that aspiration plays a role in nosocomial pneumonia. So there is a risk of aspiration. The question is, is fasting the proper response? In that study, people performed gastric echography, and they measure the volume of liquid or solid you may have in the stomach, and they did that just before extubation in the ICU. And they compared patients who were fasted and those who were not fasted, and you see that the gastric antral area, which is a surrogate of gastric volume, was exactly the same in the two groups, and the stomach was considered full in a quarter of the patients, whatever the fasting practice before. So fasting may not reduce the gastric volume. And it's not necessarily devoid of side effects. In that study, which looked at the gap between prescribed calorie intake and what is actually delivered, well, the gap in one third of the cases was due to airway management procedures, including extubation. So basically, fasting is frequently performed prior to extubation in the ICU. It's a, aspiration's a serious risk among those patients. Fasting may not be efficient to reduce gastric content, and may not be devoid of potential side effects. So our objective was to compare continued antral nutrition until extubation with fasting prior to extubation of patients in the intensive care unit. We designed a pragmatic open-label cluster-randomized parallel group non-inferiority trial. The methodology has been published before. We enrolled adult patients ventilated more than 48 hours, and with antral nutrition for more than 24 hours. We did not include patients with do not re-intubate orders, with a tracheostomy, or postpuloric antral nutrition. The intervention prior to entry in the trial, weaning of mechanical ventilation, antral nutrition strategies, parenteral fluids, glucose, nutrition, aspiration, prevention measures were left at the discretion of the attending physician. Patients were allocated to the continued antral nutrition group, with no change in antral nutrition flow rate, and no suctioning of the gastric tube, or allocated to the fasting group, where we implemented a maximum gastric vacuity strategy during at least six hours prior to extubation. No antral nutrition, and continuous suctioning of the gastric tube, if of sufficient diameter. Once extubated, everything was again left at the discretion of the attending physician. Gastric tube removal or not, nutrition via the tube or not, oral nutrition, parenteral fluids, et cetera, everything was left at the attending physician. We performed cluster randomization, which was stratified on university affiliation and case mix. 11 ICUs were randomized to the continued antral nutrition strategy, and that was applied to all patients recruited in those units during the trial, and 11 ICUs to the fasting strategy. So there was no crossover between the strategies. The primary outcome was extubation failure, reintubation, or death within seven days. We had a sample size calculation based on an expected extubation failure rate of 16% in the fasting group, and an inferiority margin of 10%, and with one-sided type 1 error 2.5, a power of 80%, we needed to include 1,100 patients to be included in 25 clusters. So a little bit more than 3,000 patients were screened in both groups, and finally 1,130 patients were included in the intention to treat population. We also had a pair protocol population. In the fasting group, this were patients who did not receive any antral nutrition six hours prior to extubation, and when the gastric tube suctioning was not interrupted more than 30 minutes. In the continued antral nutrition group, the pair protocol population was patients where nutrition was not interrupted for more than 30 minutes, and the flow rate never reduced more than 50%. At admission, the patient's characteristics at ICU admission were well-balanced between the two groups. Patients were at high risk of extubation failure with a mean age above 60 years, a substitute at around 50, 14% had COPD, 14% ischemic heart disease, arterial fibrillation in 12%, severe valvular disease was slightly unbalanced with 7.3% higher percentage in the fasting group. The recruitment was mostly medical, but 15% of the patients were surgical. At extubation, which occurred in a median of 8.4 and 8.3 days after that duration of mechanical ventilation, patients in the antral nutrition group had received 375 milliliters of nutrition over the six hours, and in the other group, 20 milliliters had been suctioned from the gastric tube. There were some imbalance between groups at the time of extubation. For example, the difficult weaning, the failure of the first spontaneous breathing trial was much more frequent, 35% of the patients of the continued antral nutrition group compared to 25 in the other group, and that increases the risk of extubation failure in the continued antral nutrition group. Steroids were given much less frequently, again, in the continued antral nutrition group, especially for laryngeal edema prevention. Again, this increases the risk of extubation failure. Conversely, in the fasting group, ineffective or weak cough was much more frequent, and abundant or greatly abundant secretions was also more frequent in the fasting group, and this increases the risk of extubation failure in the fasting group. So there were some imbalances between groups. They may compensate each other, but that's what we observed. So the results on the primary outcome, extubation failure was 17.2% and 70.5% in the two groups, and in the pair protocol analysis, the numbers were about the same, and this met the non-inferiority criteria we set. Secondary outcomes, the time between the first successful spontaneous breathing trial and extubation was shorter in the continued antral nutrition group, and so was the time to ICU discharge. So time to extubation, two hours versus 18 hours. Time to ICU discharge, four days towards 6.6 days. Nosocomial pneumonia incidence was low in both groups. The ICU death rate was significantly lower in the continued antral nutrition group, and hypo- and hyperglycemia happened more frequently at some time points in the fasting group. So because of the imbalance between groups, we looked if that could be explained by the clusters, because that is something to take into account, and you have here the extubation failure rate for each of the units which participated, and the size of the circles represents the number of patients included, and you can see that there is not one unit which drives the result, so the things are quite well distributed among the different rates of extubation failure, and we did some sensitivity analysis, adjusting in univariate methods. We calculated the absolute risk difference adjusted on steroids, respiratory physiotherapy, age, substitute, failure of the first SBT, cough strength, secretion abundance, risk factor for extubation failure, and cardiopulmonary comorbidity. In all cases, the non-inferiority was confirmed. We also did a multivariate analysis which also confirmed the non-inferiority. So our conclusions are that continued enteral nutrition until extubation was non-inferior to a six hours fasting maximum gastric vacuity strategy in terms of extubation failure at seven days, a patient-centered outcome evaluating the overall safety of the procedure. These results support changes in clinical practice towards this safe and simple alternative. Continued enteral nutrition until extubation may be associated with faster extubation and ICU discharge. Thank you very much for your attention.

fasting prior to extubation

ICU patients

dysphagia

extubation failure

continued enteral nutrition