SCCM Resource Library-Late-Breaking Studies That Will Change Your Practice

Video Transcription

Ladies and gentlemen, please welcome Dr. Timothy Buchman, Editor-in-Chief of Critical Care Medicine.  Welcome to the 2021 Plenary Late Breaker Session at our first ever virtual Critical Care Congress.  The Society of Critical Care Medicine has been presenting late-breaking studies from well-regarded peer-reviewed medical journals at the Critical Care Congress for many years.  I'm Tim Buchman, Editor-in-Chief of Critical Care Medicine, and joining me today to introduce  late-breaking studies which will change your practice are Dr. Howard Bauchner, who's Editor-in-Chief  of JAMA and the JAMA Network, and Dr. Darren Tashman, Deputy Editor and Online Editor, New England Journal of Medicine and Executive Strategy Editor of the New England Journal  of Medicine Group. Ladies and gentlemen, please welcome Dr. Darren Tashman. Thank you very much for being part of the Critical Care Congress Plenary Late Breaker Session.  Dr. Christopher Hughes is a Professor of Anesthesiology at Vanderbilt University Medical Center.  He's a practicing anesthesiologist and intensivist who currently serves as Chief of the Anesthesia  Critical Care Medicine Division, as Medical Director of the NeuroICU, and as a member of the Vanderbilt IRB.  Dr. Hughes is a Principal Investigator in the Critical Illness, Brain Dysfunction, and Survivorship Center at Vanderbilt.  His research focuses on evaluating mechanisms, prevention strategies, and potential therapies  of delirium and long-term cognitive impairment after critical illness, including the role of sedative regimens.  He will be sharing highlights of the NIH-funded MENDS-II study, a multicenter trial of dexmedetomidine  versus propofol for sedation in mechanically ventilated adults with sepsis, which we are simultaneously publishing in the New England Journal of Medicine.  My name is Christopher Hughes and I'm an anesthesiologist and intensivist at Vanderbilt University Medical Center.  I'm going to be discussing our trial of dexmedetomidine versus propofol for sedation in mechanically ventilated patients with sepsis. So first, I want to say thank you.  Thank you, thank you, thank you to all the members of the critical care community out there really rising to the challenges of 2020 and 2021 and the COVID pandemic.  I also want to say thank you to our patients that we enrolled in the trial. This is a picture of our last patient that we enrolled.  As you can see, a sick guy on a ventilator, multiple infusions, and one of them is covered with black tubing over that IV tubing, and we'll talk a little bit more about why that is.  The project was funded by the NIH and studied drug was provided by Pfizer, which was dexmedetomidine for this study. There have been several key trials of dexmedetomidine sedation.  They started with benzodiazepine trials. The men's trial was dexmedetomidine versus lorazepam, and the CETCOM trial was dexmedetomidine versus midazolam.  And those first trials showed improvement with dexmedetomidine with regard to delirium, time receiving mechanical ventilation, and cost.  In some trials, the MIDEX and PRODEX studies were non-inferiority studies. They found no significant impact with dexmedetomidine.  Patients were more interactive, but risk of delirium was unclear in these studies. They did not evaluate for cognitive impairment or long-term mortality.  There's been a couple of smaller studies in cardiac surgery specifically, but these may not be applicable to the general population of medical and surgical ICU patients.  More recently, there's been some open-label trials of dexmedetomidine. The DESIRE trial was sedation with dexmedetomidine versus sedation without dexmedetomidine in  patients with sepsis. They found no difference in outcomes here. They did limit the dexmedetomidine dose to 0.7 micrograms per hour, and 20% to 30% of  the dexmedetomidine group received propofol, while also 10% to 20% received midazolam. The SPICE-3 trial was the most recent trial.  It is a huge landmark trial, open-label as well, where patients were randomized to early sedation with dexmedetomidine or sedation that did not include dexmedetomidine.  As you can see here, the dexmedetomidine and control groups had very similar propofol and midazolam use, and there was not a big separation between groups with regard to sedative  exposure. They also had a high percentage of deep sedation, and they found no difference in mortality, delirium coma-free days, or ventilator-free days between groups.  Why are we interested in dexmedetomidine in sepsis patients specifically? That's because dexmedetomidine has been found to have superior anti-inflammatory and bacterial  clearance properties compared to benzodiazepines and propofol. There's been lower rates of subsequent infection versus midazolam in the CEDCOM trial, and  reduced 28-day mortality in sepsis patients in the MEN-2 trial versus lorazepam, and that's what's shown here in this figure.  The Society of Critical Care Medicine currently recommends slight sedation techniques when critically ill mechanically ventilated adults need sedation, and it's just using either  propofol or dexmedetomidine over benzodiazepines for the sedation. Now, there's not a recommendation of propofol versus dexmedetomidine, and part of that is  due to the limitations in the trials that we discussed. And that's where the MEN-2 trial comes in.  So MEN-2 stands for Maximizing the Efficacy of Sedation and Reducing Neurological Dysfunction and Mortality in Sepsis.  Our article is just being published in the New England Journal of Medicine, so please check it out. The MEN-2 study was a double-blind, multi-sensor, randomized trial.  We included patients who are medical or surgical ICU patients with suspected or known infection. They had to require continuous sedation with mechanical ventilation.  We excluded folks with severe cognitive impairment, second- or third-degree heart block, persistent  bradycardia requiring therapy, allergy to either dexmedetomidine or propofol, necessity for benzodiazepines, meaning they're either on chronic benzodiazepines or having acute  alcohol withdrawal, and mechanical ventilation for greater than 96 hours to avoid having a lot of sedative exposure prior to randomized to one of our sedative study drugs.  We allowed dexmedetomidine doses up to 1.5 mics per kilo per hour and propofol doses up to 50 mics per kilo per minute.  These medications were prepared in IV bags by an investigational pharmacist, so there are different concentrations in what we use clinically.  This allowed us to administer them in a milliliter-per-hour basis for equipotent dosing to be able to blind the study drugs. These study drugs were given up to 14 days.  Now, the bedside nurse was the one part of the care team who was unblinded, so the research team and the care team were blinded besides the bedside nurse, and he or she administered  the study drug to meet the clinical RAS target. Here's some examples of how we blinded the MENS 2 study drug.  As you can see, we had black IV bags to cover the IV bags themselves, and then we had black tubing to cover the IV tubing. Here is our study drug weight-based titration table.  As you can see, we ranged from propofol from 5 mics per kilo per minute up to 50, and dexmedetomidine  from 0.15 mics per kilo per hour up to 1.5, and the columns here represent what we feel like were equipotent doses of the sedative study drugs.  And we had started these medications at doses that were similar to where the patient was receiving before they were randomized.  So, for example, an 80-kilo patient who was on 15 mics per kilo per minute of propofol  would be started on 7.2 milliliters per hour of study drug, and then this would be titrated up or down to left or right on this table based on whether or not the patient needed  more or less sedation. Our research personnel performed RAS and CAM-ICU CPOD assessments up to 14 days, hospital discharge or death, whichever came first.  Again, sedation targets were set by our clinical teams, primarily light sedation with RAS scores of 0 to negative 2.  And fortunately, all centers performed the ABCDE bundle and up-to-date sedation care. And then six months post-randomization, we performed a phone assessment.  This included the telephone interview for cognitive status, additional cognitive battery,  the CATS ADLs and a functional activity questionnaire to evaluate for functional impairment, and quality of life with the EQ5D survey.  Our primary outcome for our study was days alive without delirium or coma in 14 days. This allowed us to account for duration of acute brain dysfunction while also avoiding  bias due to death. Our secondary outcomes included ventilator-free days in 28 days, death at 90 days, and TIC-T score at six months.  We had a modified attention-to-treat approach, which we defined as all patients were randomized and received study drug.  We were originally aiming for 530 patients to provide over 90% power to detect the difference  in acute brain dysfunction and also mortality, but due to slower-than-anticipated enrollment, our trial was subsequently resized and repowered to 420 patients.  This resizing was approved by the DSMB and the NIH, and importantly, we still had greater  than 85% power to detect a difference in 1.5 days alive without delirium or coma, and greater than 80% power to detect the absolute difference in mortality of 12% at 90 days.  We had 13 study sites involved in this multi-center trial, as shown here, across the United States. Shout-out to all of our study sites.  Thank you for participating and all that you did to enroll patients. Overall, we randomized 432 patients, 216 to dexmetatomine and 216 to propofol.  We had 10 patients that had rapid mechanical ventilation weaning and no longer needed sedation and therefore no longer needed study drug, so 214 received dexmetatomidine and  208 received propofol. We had 108 people follow up in the six-month evaluation in dexmetatomidine arm, and 101  underwent six-month evaluation in the propofol arm, and more than 90% of our eligible patients had a six-month follow-up.  This table shows our baseline characteristics of the trial population, and some key points from this table.  We had a high severity of illness with APACHE-2 scores of 27 and SOFA score of 10. Approximately two-thirds of these patients were medical patients and one-third were surgical  patients, and we enrolled them approximately one day from their ICU admission. This table shows some of our study drug information.  Some key points from here is we started a study drug a median of 22 hours from the start of their mechanical ventilation. The study drug was administered for a median of three days.  The RAF score while on study drug median was negative two. We had median low doses of our sedative study drugs, so the median dose of dexmetatomidine  was 0.27 micrograms per kilo per hour, and our propofol was 10.2 micrograms per kilo per minute.  We also had low rates of unblinding, especially given the physical differences in these study drugs.  So we had 14% overall of either a clinical team member or a research team member being unblinded. Some additional sedation regimen information provided here.  We had high compliance with the ABCDE bundle. We found similar rescue medication use between groups.  This included low doses of midazolam and also antipsychotics, and importantly, we found low rates and low doses of the open-label propofol and open-label dexmetatomidine.  So we had very good study adherence to our protocol and also low cross-contamination of our sedatives.  This slide shows the results of our study from Panel A, Days Alive Without Delirium or Coma in the 14-Day Study Period.  You can see the adjusted median of days of dexmetatomidine was not significantly different than propofol, 10.7 versus 10.8.  With regard to ventilator-free days and 28 days, we also found no significant difference between groups with dexmetatomidine at 23.7 and propofol at 24.0.  For the TICS T-score at six months, higher scores on the T-score here equal better cognition,  and we found no difference between the dexmetatomidine and propofol groups, 40.9 and 41.4, respectively.  Here we show the outcomes of our mortality analysis. So this is the Kaplan-Meier curve for mortality up to 90 days, and also the hazard ratio crosses one.  So no significant difference in our mortality outcomes at 90 days between the dexmetatomidine group shown in red and the propofol group shown in blue.  Some additional efficacy endpoints that we examined that were performed at our six-month assessments included the Functional Activities Questionnaire, the CAHPS ADL, Median EQ 5e,  and we found really no differences between groups. We also found no significant differences between groups in the other cognitive batteries that we performed.  We wanted to look at some safety outcomes given the high severity of illness in our study population.  This table shows our organ dysfunction based on study group, and overall we found similar rates of organ dysfunction between the groups and across the different systems.  Along the same lines, we looked at several different safety endpoints, including hypotension,  amount of days requiring pressors, bradycardia, lactic acidosis, ARDS, self-extubation, triglyceride levels.  And we see subtle differences here between the groups, but overall the safety profiles seem to be pretty similar between the groups.  Of note, the symptomatic bradycardia requiring study drug discontinuation was only 2% in dexmetatomidine and 1% in the propofol group, so not much difference there either.  Strength of our study included enrolling a representative population of septic patients in centers across the United States.  We made every effort to blind the delivery of propofol and dexmetatomidine, despite the different physical properties.  All centers performed up-to-date sedation care with the ABCDE bundle. Our clinicians set sedation targets, and so patients were generally managed at lighter  levels of sedation with low doses of sedative medications.  We achieved good separation between groups regarding sedative exposure, and we had robust follow-up in our cohort.  Limitations include an episode of unblinding in 14% of patients, but this blinding compliance was still higher than reported in similar clinical trials.  We also had some cross-contamination of sedative use, which was substantially lower than similar sedation studies.  Our study drug was started at a median of 22 hours for meeting all inclusion criteria, so this may have limited our ability to impact study outcomes, as opposed to if we were able  to start earlier in the critical illness course. We did have to adjust our sample size, but we still had adequate power to address the study questions at hand.  We had some exclusions due to clinicians not having equipoise regarding sedation for a given patient, so this may have limited our ability to generalize this out to other patients.  In conclusion, among mechanically ventilated adults with sepsis who are being treated with recommended light sedation approaches, use of dexmedetomidine did not improve days alive  without acute brain dysfunction, ventilator-free days, death at 90 days, or global cognition at six months, when compared to our control agent of propofol.  I'd like to give a shout-out again to our team. Here's some pictures from our start-up meeting. We've got everyone out dancing out in Nashville at the time.  A few pictures from the different sites and teams from our study start-ups at their sites. Unfortunately, I don't have everyone pictured, but thank you again.  A few more pictures here. Several members of the Society of Critical Care Medicine were at their respective sites.  And finally, I'd like to draw attention to our website, www.icudelirium.org, for a lot of information on this topic.  It also pairs well with the ICU Liberation and Society of Critical Care Medicine's information out there on their websites as well. So thank you very much for listening.  Also, thank you, thank you, thank you for all the work that you're doing taking care of our patients. Really appreciate you.  The Society of Critical Care Medicine's Critical Care Research Network, Discovery, is leading  the efforts for the global COVID-19 registry, the Viral Infection and Respiratory Illness Universal Study, or VIRUS.  VIRUS is a prospective cross-sectional observational study registry of all eligible patients who were admitted to civilian and military hospitals.  The registry is designed to create a real-time COVID-19 registry of current ICU and hospital care patterns to allow evaluation of safety and observational effectiveness of COVID-19  practices. Dr. Amos Lau is a Critical Care Fellow at the Mayo Clinic, Rochester, Minnesota. His research interests include sepsis and outcomes in the critically ill, COVID-19 and  the VIRUS registry while working with SCCM Discovery. Dr. Lau will present outcomes of patients with COVID-19 receiving organ support therapies,  the International VIRUS Registry, which will be simultaneously released in Critical Care Medicine, available now at the journal website. Good morning, friends.  Thank you for joining us for this interesting talk on a late-breaking paper about outcomes  of patients with COVID-19 receiving organ support therapies, which is part of International  VIRUS Infection and Respiratory Illness University Study, which is part of VIRUS registry. My name is Amos.  I'm one of the Chief Fellows in Critical Care Medicine at Mayo Clinic, Rochester, and I'm honored to present this talk on behalf of VIRUS Investigators, SCCM Discovery Network.  We are also thankful to our collaborators from Boston University, Mayo Clinic, and other places worldwide.  Moving on, I have no relevant disclosures to mention at this point relevant to this talk. The overview of our talk will include a few things which will have objectives of our study.  This is our first outcome paper that has come out from VIRUS Registry. I'm going to be briefly talking about the design of the study, the setting, and the  patients that we included in the study, what were the interventions, outcomes of our patients based on organ support therapies, as well as distribution within the age categories.  We will talk about the measurements and finally the results and the conclusion from our first outcome paper.  Our main objective for our first outcome paper was to describe the outcome of patients who  are hospitalized, not just in the United States, but our worldwide collaborators in a multicenter  international COVID-19 registry, inception of which took place early in the pandemic, early in 2019, under the leadership of Dr. Kashyap, Dr. Kumar, and Dr. Walkie, who have  worked tirelessly to bring it to existence. We are also very thankful to all the patients and our research collaborators who have worked with us on this.  The design of our first outcome paper was a cross-sectional observation study. It's a live registry where we included patients who were confirmed SARS-CoV-2 positive, and  we included patients from February 15th all the way to November of 2020. Primary outcome for the study was to assess hospital mortality based on the type of organ  support that these patients received. Secondary outcomes that were measured were discharge home with or without any assistance  and the length of stay, whether the hospital length of stay or the ICU length of stay. The stratification of our study group or the patients that were included in the study  was stratified based on the age as well as the type of organ support. Patients who did not require any sort of organ support versus patients who required invasive  mechanical ventilation versus patients who required renal replacement therapy or patients who required invasive mechanical ventilation, renal replacement therapy, vasoactive drugs,  and ECMO. Setting, until now we have included almost 179 hospitals in 16 countries with the collaboration  of SCCM's Discovery Viral Infection Registry, and this number continues to grow. On the left side of the screen is the world map and the points that are highlighted show  the centers which have partnered with us. On the right side we highlight the U.S. locations. So we have the whole breadth of hospitals including from small teaching hospitals to  community hospitals to large university hospitals, not just in the United States but also in South America, parts of Asia, parts of Africa, Europe.  The only continent that is left at the moment is just Australia. So this is not just representative of the practice in the United States but also across the continents.  This is the stroke diagram for the eligibility criteria for the patients that were included in this study.  There were a total of 49,000 patients who were included in this iteration of our outcome paper. 29,000 of those were assessed for eligibility.  Almost 9,000 were excluded because they did not qualify. The patients who were finally included in the analysis were close to 21,000, 20,608 patients.  Out of these patients, 15,000 patients did not require any sort of organ support. 5,000 required at least mechanical ventilation and these patients were further stratified  into patients who required only mechanical ventilation which was about 1,700, patients required invasive mechanical ventilation and vasoactive drugs, and then patients who required  invasive mechanical ventilation, vasoactive drugs, and renal replacement. So as you see, what we tried to do is stratify the patients from least support to maximum support.  And then the patients who required all sort of organ support therapy in addition to ECMO. As I mentioned, the outcomes for this study that we wanted to focus on were primarily  hospital mortality. The secondary outcomes were length of stay, discharge home with or without assistance,  and for length of stay we were interested to assess the length of stay in the hospital as well as length of stay in the ICU.  And these were the baseline characteristics of these 20,000 plus patients in the first outcome paper. We had about 4,000 patients who were below the age of 44.  About 5,000 patients were in the age bracket of 45 to 60 years. We had about 6,000 or 6,500 patients between 60 to 74 years of age, and beyond 75 years  of age we had about 5,000 patients as well. And as you see, the comorbid conditions continue to increase as the age progresses.  So the youngest age group, which was 18 to 44, had minimum number of comorbidities such as hypertension, diabetes, obesity, chronic kidney disease, and as you progress on this  table towards the right side, as the age grows the number of comorbidities increases as well.  When we did the analysis for these patients, these groups were, although these groups differed in terms of their comorbid conditions, but in terms of the SOFA score they were fairly  well matched. Also the use of steroids were noted to be high in the group between 45 to 60 as well as 60 to 74.  However, the youngest group and the oldest group did not have the same amount of steroid use as the age groups in between.  The overall length of stay, as expected, the youngest group had the lowest length of stay, and then the group between 45 to 75 ranged from about 7 to 8.6 days, and the similar  range for group of people with 75 and above. Moving on to the results section, as we briefly discussed about the steroids, you can see  most of the age group received about close to 15% of their, all the patients received some sort of steroids during their hospitalization and treatment plan.  Length of stay progressed or increased in number as your age increases from the youngest age group between 18 to 44 to almost 75, your length of stay increased from five days to  almost 8.6 or nine days. The important finding is here, discharge disposition. As you can see, the younger age group had about almost 70% chances of getting discharged  home from the hospital after being acutely infected, whereas this percentage falls significantly to about 28% in the 75 and above age group.  This significantly translates into the mortality. When you come here and see the number of patients who were deceased, in the younger age group  the mortality was only 4.6%, that is 18 to 44, and that rose to almost 35% in the population which was 75 and above.  The number of patients who were invasively ventilated or patients who required invasive mechanical ventilation was close to 45 to about 55% in all the four groups in terms  of the age stratification.  Number of days on invasive mechanical ventilation also did not vary significantly between these four age groups.  And what we just iterated in the previous slide, as you can see, as your support increases as well, the chances of mortality rise significantly.  When you have no support, no organ support, and then you transition over to only mechanical ventilation support, then you add invasive mechanical ventilation and vasopressors, and  finally invasive mechanical ventilation, vasopressors, NRR, and renal replacement, the mortality goes up in all the age groups.  So this slide highlights the groups by colors, as you can see, but there is a significant trend in mortality as your organ support increases.  And that slide is really the key to the next slide. As your mortality is increasing, the chances of you getting discharged home are less and less.  So as your requirement for organ support therapy increases, the chances of discharge home without  support fall down significantly in all age groups, but more so for people in the older age group.  As I mentioned, the hospital length of stay for patients who require organ support therapy did not vary significantly in terms of the age stratification.  And finally, this is the important finding that we were able to see in our analysis.  There was a significant inter-hospital mortality that we found, and this begs the question that why did a lot of hospitals have inter-hospital variability  in terms of mortality and outcome? And to address that, the group is actually exploring other options and there are ancillary  studies which are underway which will be evaluating the inter-hospital mortality variation as well. So stay tuned for that.  Finally then, implications and conclusions based on our first outcome paper. This is a paper which gives very novel and important information for the clinicians to  discuss with the patients and the family, for patients who come in with acute infection with COVID-19, based on the organ support therapy that they're requiring, what is the  expected outcome? Also helpful at the hospital and administrative level in terms of resource planning and resource  allocation, especially in the places where the resources are limited. Also there are countless opportunities to improve the care that we provide to our patients  in terms of QI endeavors and further clinical research that can be carried out to improve the care of these patients. Similarly implications to research.  As I mentioned, this is a lab registry and we are continuing to add more and more patients.  Last time I logged on to the system there are almost 47,000 patients which we have already enrolled in virus registry and the number continues to grow as the days go by and we  are adding more and more collaborative sites, not just in the United States but worldwide. The advantage of that is it should not only reflect the practices that we have in the  United States but also gives us the idea of what people in the hospitals outside the United States are doing and to share knowledge across the borders to improve the quality of care  that is provided to these patients. It also gives a lot of opportunities for the young investigators and the senior investigators  for international collaboration across the board, not just to share our knowledge but also to learn from them.  As I mentioned, near real-time dashboard which gives us the amount of patients that we have enrolled as we go along day by day it changes.  It gives us the breakdown of how many patients are requiring mechanical ventilation, non-invasive  mechanical ventilation, other types of oxygen support, patients who are requiring dialysis, ECMO, et cetera.  I want to finish my presentation by thanking most importantly the patients and the families who in the difficult times of this illness have allowed us to conduct the research so  that we can improve the delivery of care not just to them but also patients who are going to come and time to come.  We are thankful to the team at Mayo Clinic as well as Boston University, team at SCCM, Weiss Registry and all our collaborators.  We are also very thankful to our sponsors who have helped us so that the research like this on a global, multinational, international platform can be conducted. Thank you very much.  Ladies and gentlemen, please welcome Dr. Howard Bochner, Editor-in-Chief of JAMA. Hello and welcome to the next presentation.  This is Howard Bochner, Editor-in-Chief of JAMA and I'm delighted to introduce John Laffey.  John is a Professor of Intensive Care Medicine and Anesthesia at the School of Medicine of the National University of Ireland, Galway and a consultant in intensive care medicine  and anesthesia at Galway University Hospitals. He is the Director of Clinical Research at NUI Galway and the Salata University Hospital  Group and Vice Dean for Research at the College of Medicine, Nursing and Health Sciences at NUI Galway.  His basic and translational research is focused on acute respiratory distress syndrome and sepsis with a focus on regenerative medicines for the critically ill.  John is presenting the findings of the INTUBE, I-N-T-U-B-E study today. INTUBE is a prospective observational study of adverse events relating to emergent tracheal  intubation in critically ill patients. The title of the article is Intubation Practices and Peri-Intubation Adverse Events in Critically Ill Patients from 29 Countries.  It has been accepted for publication in a forthcoming issue of JAMA.  And for all of you who are listening today, I want to remind you that Chris Seymour has joined the JAMA family as the Associate Editor in Critical Care Medicine.  Derek Angus remains in the JAMA family as a Senior Editor. The three of us and others, the peer reviewers, decided that this article was certainly important  and worthy of publication in JAMA. It details critical, it details important events for critically ill patients in a common  procedure intubation that occurs thousands of times around the world. Thanks for tuning in today, John. It's all yours. Hello. Good morning. My name is John Laffey.  I'm based at the National University of Ireland in Galway, and I'm delighted to present this  study on Intubation Practices and Peri-Intubation Adverse Events in Critically Ill Patients from 29 Countries.  I'm doing this on behalf of the Intube investigators, a global network of investigators. I have no conflicts of interest in relation to the subject matter of this talk.  So tracheal intubation is among the most commonly performed urgent procedures in critically ill patients.  There are a limited number of prospective or national-level studies available, and of those that do exist, they indicate that about 28% of procedures can be complicated by major  adverse events. However, there's relatively limited information available on practices and peri-intubation events in routine clinical practice from large international cohorts.  The NAP study, or the National Audit Project of the Royal College of Anaesthetists in the United Kingdom, performed in 2008 and 2009, is perhaps the largest existing study in this  field. It looked at major events, that's death, brain injury, emergent surgical area, ICU admission,  arising from airway management in the intensive care unit, the operating rooms, or the emergency departments of hospitals across the United Kingdom.  They found that 20% of all major adverse airway-related events occurred in the intensive care unit.  And perhaps of most concern was the fact that high morbidity resulted from these adverse airway events in the critically ill.  In fact, 61% of all airway complications in the ICU led to death or to brain damage. And the comparison rate for these events in the operating room was 14%.  So our study was an international, multicenter, prospective cohort study. We enrolled critically ill patients undergoing intubation in the intensive care unit, the  emergency department, or the ward during an eight-week period from the 1st of October 2018 to the 31st of July 2019.  The inclusion criteria were broad and they included all adults who were critically ill undergoing in-hospital tracheal intubation.  And the exclusion criteria were out-of-hospital tracheal intubation procedures, tracheal intubation  for elective measures such as general anesthesia for surgery, and patients in cardiac arrest prior to intubation.  The primary outcome was the presence of at least one of the following events within 30 minutes of the start of the tracheal intubation procedure.  Severe hypoxemia, that is an SpO2 of less than 80%, cardiac arrest, or cardiovascular instability, which was one of the following.  A systolic arterial pressure of less than 65 millimeters of mercury recorded at least once. A systolic arterial pressure of less than 90 millimeters of mercury for over 30 minutes.  And the new requirement for, or the increased requirement for pressors are fluid bowls over 15 mils per kilo to maintain a target blood pressure.  These are the secondary outcomes of the in-tube study and they vary from ICU mortality, arrhythmia,  intubation characteristics, pulmonary aspiration, and injury to the area or dental. We had a global representation of participating sites.  There were 197 centers from all five continents across 29 countries in the in-tube study. So this is our inclusion algorithm. We screened 3,659 patients for eligibility.  2,964 patients were enrolled, undergoing 3,058 intubations, meaning that 94 of these patients were re-intubated.  2,964 patients were followed for the full 30 minutes post-intubation, and 2,943 patients were followed up until ICU discharge.  So these are the demographic characteristics of the patient cohort. In age of 63, female patients were just over one-third of the cohort, and in terms of reasons  for intubation, I just want to highlight that over 50% were intubated for respiratory failure and 30% were intubated for neurologic impairment.  Cardiovascular instability, airway obstruction, and emergent or urgent procedure were the other indications.  In terms of the peri-intubation severe adverse events, these are represented on this graph. What you can see is the dominant event was severe cardiovascular instability occurring  in over 40% of patients. Severe hypoxia occurred in 9% of patients and cardiac arrest in 3%. Other secondary events, such as esophageal intubation, new-onset cardiac arrhythmia,  difficult intubation, and aspiration of gastric contents occurred in about 3% to 5% of patients.  And this is a table showing the primary outcome events that I just showed you on the graph previously. So as we said, severe cardiovascular instability, 42% of patients.  This was mainly seen as a new or an increased need for vasopressors. Cardiac arrest occurred in 3% and severe hypoxia in 9.3%.  So this is a series of Kaplan-Meyers showing the effect of a major adverse event on survival rates.  So in the first panel here, what we see is that patients who suffered any of these major intubation-related events had a significantly lower survival probability.  This was true for cardiovascular instability alone. Again, this relationship was highly statistically significant. This was also true for intubation-related cardiac arrests.  Interestingly, in regards to severe hypoxia, the survival rates were not significantly different.  Repeated intubation attempts in these critically ill patients was associated with an increased risk of adverse events.  This was particularly true for cardiac arrest and for severe hypoxia.  In regard to the details of the intubation procedure, the pre-oxygenation method used in most cases, over 60%, was bag-mask ventilation.  Apneic oxygenation was used in approximately 10% of patients, and I want to draw your attention to two specific points here.  The first is that video laryngoscopy was used only in a small minority of patients, 17% overall.  Waveform capnography, which is the best evidence-based method to confirm tracheal intubation, was used in one quarter of all patients.  What we found in regard to the use of waveform capnography, or end-tidal CO2 measurement, was that its use varied depending on the location in which tracheal intubation occurred.  The ICU saw the highest levels of capnography use, but it was still only 40% of all tracheal intubations in the intensive care unit.  In the emergency department, it was less than one quarter of all intubations, at 23.3%, and on the ward, it was 6.9% of all tracheal intubations.  I return now to the National Airway Audit carried out in the United Kingdom in 2008-2009, and they found that the use of capnography was important in avoiding severe adverse events.  In fact, they reported in this paper that the lack of capnography contributed to 74% of airway-related deaths in the intensive care unit.  So this audit, while confined to the United Kingdom, was carried out just over 10 years  ago and reported in 2011, and so the continuing underuse of capnography is a particular concern.  So in conclusions then, our prospective observational study of intubation practices was performed  in 2,964 critically ill patients undergoing emergent intubation at 197 sites across 29 countries.  So we feel well positioned to—we feel that this is a very good global picture of the current approaches to emergent tracheal intubation.  We found a high frequency of intubation-related major adverse events. Cardiovascular instability was seen in 42.6%, severe hypoxemia in 9.3%, and cardiac arrest in 3.1% of patients.  Of importance, these major perioperative events were significantly more frequent in patients that died later during their intensive care unit stay.  And future study directions should focus on interventions targeting perioperative or periintubation hemodynamics.  I'd like to finish by thanking our participants, the almost 3,000 patients who participated in this study and our intubed collaboration of researchers.  This truly was a global research effort. I would also like to thank the societies that endorsed the intubed study, and I would  particularly like to draw attention to the endorsement of the Society of Critical Care Medicine and the Critical Care Research Discovery Network within the SCCCM.  We are particularly grateful for the help given to us in regard to this study. I would like to conclude this presentation now, and I would be very happy to take questions  at the end. Dr. Erik von Ulrich is an intensivist at the Karolinska University Hospital and researcher at the Karolinska Institute in Stockholm, Sweden.  His primary research interests are in long-term outcomes after severe trauma and intensive care, including the use of opioids in trauma patients and ICU survivors.  The current study emerged from an interest in the increasing use of opioids in Europe, including Scandinavia, during the past decade.  Whether or not intensive care treatment per se is contributing to this current increase in opioid use is not clear. Neither is the prevalence of chronic opioid use after ICU care.  Dr. von Ulrich will present Opioid Use After Intensive Care, a nationwide cohort study, which will be simultaneously published in Critical Care Medicine and available now at  the journal website. So good morning, everyone. I want to start by thanking the Society of Critical Care Medicine for allowing us to present our data today.  My name is Erik von Ulrich, and I'm an intensivist at the Karolinska University Hospital in Stockholm, Sweden.  And I'm here today to talk about opioids before and after critical care. And I want to start by showing our disclosures, mine and my co-authors, and also our funding.  So as we all know, the misuse of opioids has turned into a major public health problem in many countries.  And starting in the 1990s in the U.S., marketing and promotion of prescription opioids led  to generous prescribing and misuse of prescription opioids. And really, no matter what statistic you look at, the numbers are pretty alarming, and we  know that we have a large number of deaths every year due to misuse of prescription opioids. In contrast to the discussion about opioids in the U.S. and Canada, we don't really know  that much what the situation is like in Scandinavia or in Sweden. We don't have any evidence that we have an opioid epidemic like in the U.S., but we know  that the use of opioid is increasing. And just to give you an idea, these are numbers from the OECD showing opioid-related deaths per million inhabitants from 2011 to 2016.  And we can see that the U.S. are pretty far ahead, but the rate is increasing even faster in Sweden.  So when it comes to critical care, we know that opioids are commonly used, and we use morphine, fentanyl, and remifentanyl to treat pain or for sedation purposes.  We also know that a majority of ICU patients report pain and discomfort for many years after discharge.  And even though opioids are the primary choice to treat moderate to severe pain, there's really no evidence for any long-term treatment with opioids, and to treat chronic pain with  opioids, that's controversial today. So with all this in mind, our goals with this study were to look at and to describe opioid use before and after critical care.  And we also wanted to look at factors associated with the risk of becoming a chronic opioid  user after discharge from critical care, and to determine if such chronic use is associated with an increased mortality.  So this was a cohort study, and the primary outcome was chronic opioid use after critical care. And the secondary outcome, all-cause mortality six to 18 months after ICU admission.  And to identify our study cohort, we turned to the Swedish Intensive Care Registry. This is a national quality register that collects individual data from all 83 ICUs in Sweden.  And we identified all patients registered from 2010 to 2018. And then we linked this information with other Swedish health registers to get information  on comorbidities and socioeconomic factors and prescribed drugs. And what these Swedish health registers have in common is that they are all validated,  they have low rates of missing data, and almost a complete follow-up. A couple of definitions important to this paper.  We define pre-ICU opioid use as at least one prescription in the year preceding ICU admission.  Chronic opioid use after discharge was defined as repeated prescriptions in the first and the second quarter after ICU admission.  And to be able to compare different opioids with varying potency, we recalculated all opioids into oral morphine equivalents.  And also we excluded patients with prescriptions of methadone or buprenorphine we excluded because those medicines are also used in opioid agonist therapy.  And we also excluded all individuals dying within the first six months after admission. And that's because we wanted to look at long-term outcomes and they were not able to fulfill  our outcomes. Statistics, we used different kinds of regression models. When we looked at opioid use over time, we used GE models.  And looking at factors associated with chronic use, we used logistic regression models. And for the hazard ratio, we used Cox regression models.  Moving on to the results, we identified a little more than 265,000 patients. We excluded 61,000-something due to use of methadone, buprenorphine, or because they  died within the first six months.  So, we ended up with 204,402 patients in our final study cohort that we ultimately analyzed. These are some demographics of the study cohort. The median age was 63.  We had around 60% males, 22% used opioids the year before admission, and we had a fairly short median length of stay of one and a half days.  So, this graph shows opioid consumption over time in relation to ICU admission. On the x-axis are cal and quarters, and on the y-axis, oral morphine equivalents.  And more specifically, this shows the mean opioid consumption per person per quarter. So, we can see that on average, the opioid consumption increased already before ICU admission.  Then we had an initial peak and a continuous decline. But we also saw that even after 24 months, the opioid use was not back to the baseline use.  So, we had a significantly increased consumption even at two years after admission. We also looked at a subset of patients not using opioids in the year preceding ICU admission,  and they showed a similar pattern with an initial peak and a continuous decline. But here it's even more obvious that they had a substantial use of opioids two years  after admission. Here are some just general characteristics comparing non-chronic users with the chronic users.  We can see that we had around 10% of chronic users, so 22,000 individuals. They were older, more likely to be female. They had a lower level of education.  They also had more comorbidities in terms of somatic, psychiatric, and substance abuse.  But most importantly, around 70% of the chronic users did use opioids already before admission. So here we can see associations with chronic opioid use presented as odds ratios.  We can see that the most important factor was pre-ICU opioid use. Besides that, also female sex, somatic and psychiatric comorbidity, and ICU length of  stay for more than two years also had increased odds. And when looking at the individuals not using opioids the year before admission, there were  only minor changes, and the odds ratios were about the same as for the full cohort. So finally, the hazard ratio for death for the chronic users, unadjusted, it was 2.2.  When adjusting for some baseline ICU admission characteristics, it was still 1.7. And when looking at the individuals not using opioids the year before admission, the adjusted  hazard ratio was 1.9. So we concluded that the mean opioid consumption was increased 24 months after ICU admission,  even though we don't really have any evidence for any benefits with long-term use of opioids.  We also saw that several factors were associated with the chronic opioid use, including female sex and ICU length of stay, but most importantly, the pre-ICU opioid use.  And finally, we also saw that chronic opioid use after critical care was associated with an increased mortality. A few limitations leading us into future directions.  We cannot be sure to what extent these individuals actually took their medication. We know that the opioids were prescribed and dispensed, but if they consumed their opioids,  we cannot be sure. More importantly, I think, unfortunately, we don't have any information on in-hospital use of opioids. We can only see the prescribed opioids after discharge.  This is something, however, we are looking into, and hopefully we will be able to present that data with it soon.  And finally, we don't have any data on non-prescribed or sweet opioids like heroin. So again, I want to thank my co-authors, thank our funders, and again, say thank you  to the Society of Critical Care Medicine for allowing us 2% today. Thank you very much.  Please join the Continue the Conversation session with the Late Breaker faculty later this morning at 11 a.m.

Video Summary

Dr. Timothy Buchman, Editor-in-Chief of Critical Care Medicine, introduces the Plenary Late Breaker Session at the 2021 virtual Critical Care Congress. The session features presentations on various topics related to critical care medicine. Dr. Buchman is joined by Dr. Howard Bauchner, Editor-in-Chief of JAMA, and Dr. Darren Tashman, Deputy Editor and Online Editor of the New England Journal of Medicine, who introduce late-breaking studies that will be presented. Dr. Christopher Hughes presents the highlights of the MENDS-II study, a multicenter trial comparing dexmedetomidine and propofol for sedation in mechanically ventilated adults with sepsis. The study found no significant difference between the two medications in terms of outcomes such as days alive without delirium or coma, ventilator-free days, and mortality. Dr. Amos Lau presents the outcomes of patients with COVID-19 receiving organ support therapies, based on data from the VIRUS registry. The study found a high frequency of major adverse events such as cardiovascular instability, severe hypoxia, and cardiac arrest during tracheal intubation in these patients. Dr. Erik von Ulrich presents a nationwide cohort study on opioid use after intensive care, finding that opioid consumption increased even after two years post-ICU admission, and that chronic opioid use was associated with increased mortality.

Asset Subtitle

Pulmonary, Pharmacology, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Dexmedetomidine versus propofol for sedation in mechanically ventilated adults with sepsis
Christopher Hughes

Outcomes of Patients with COVID-19 Receiving Organ Support Therapies: The International VIRUS Registry
Amos Lal

Intubation Practices and Peri-Intubation Adverse Events in Critically Ill Patients from 29 Countries
John Laffey

Opioid use after Intensive Care: A nationwide cohort study
Erik von Oelreich

Meta Tag

Content Type Presentation

Knowledge Area Pulmonary

Knowledge Area Pharmacology

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Mechanical Ventilation

Tag Analgesia and Sedation

Year 2021

Keywords

Critical Care Medicine

Plenary Late Breaker Session

MENDS-II study

sedation

mechanically ventilated adults

COVID-19

organ support therapies

major adverse events

opioid use

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.