SCCM Resource Library-Dawning of a New Era in Sepsis

Video Transcription

Welcome to today's session, The Dawning of a New Era in Sepsis. I'm Dr. Margaret Parker, Professor Emerita of Pediatrics from the Renaissance  School of Medicine in Stony Brook, New York. I will be co-moderating the session along with Jerry Zimmerman from Seattle Children's Hospital in Seattle, Washington.  Dr. Zimmerman will start off the session with a talk on organ dysfunction in sepsis. What's new with the OD in MODS? The next talk will be Heidi Flory  from Mott Children's Hospital in Michigan. She will talk about how have guidelines changed practice? Scott Weiss from Children's Hospital of Philadelphia  in Pennsylvania will do his talk on gaps in the literature and research currently being done. Following these three presentations, there will be time for questions and answers.  I hope you enjoy this session, which I'm sure will be very interesting and informative. Good afternoon, my name is Jerry Zimmerman.  I am a faculty member with the Division of Pediatric Critical Care Medicine at Seattle Children's Hospital and Harborview Medical Center in Seattle.  And it is my privilege to provide you with an update of organ dysfunction as it relates to pediatric sepsis. For this presentation, I have no conflicts of interest  or other disclosures that impact on my presentation.  Today, I'm going to discuss contemporary definitions of sepsis as well as pediatric organ dysfunction. I'll spend a little time talking about sepsis pathophysiology and end with how  organ dysfunction is related to clinically meaningful patient-centered outcomes in the short term, as well as long term.  So beginning with definitions, you probably are aware that in 2016, the Sepsis-3 Working Group published three articles  in JAMA, and the first of these was important  in a novel definition of sepsis, at least for adults, as a life-threatening organ dysfunction caused by a dysregulated host response to infection.  You can take this apart mathematically just to look at the elements where Y is sepsis  as a function of really four different variables, the first being suspected or documented infection, the second being a dysregulated host response  to this infection, resulting importantly for our discussion in organ dysfunction, and all of this being a real threat to life.  Closely aligned with sepsis, at least in the history of pediatric critical care is pediatric organ dysfunction. And you may also know that the first definition  for organ dysfunction derived from a forum that was conducted in Boston, Massachusetts in 2004,  as a collaboration of the PoliSea Research Network and the International Sepsis Forum in preparation for an interventional trial of activated protein C for pediatric sepsis.  This conference considered not only some initial definitions for pediatric SERS, sepsis, severe sepsis, and septic shock, but importantly also provided definitions  for pediatric organ dysfunction.  This activity at this conference reflected a number of well-known experts in the field,  as well as literature review and consensus building.  And the output of this working group was published in Pediatric Critical Care Medicine, and to this day stands as one of the most cited articles for that journal.  Around this same time, 2004, 2005, investigators in both France as well as Canada  were deriving and then validating pediatric logistic organ dysfunction score, better known as a PILOD. And it's of interest that the score initially  was validated on almost 600 children with sepsis.  These investigators honed down on organ dysfunction, derived a score to quantitate organ dysfunction, and ultimately assigned a hazard ratio of 2.37  for each organ dysfunction that might have been present. In addition, they also defined an additional risk factor for the different severity stages of sepsis,  SERS versus sepsis with organ dysfunction versus septic shock. And ultimately, they congealed both of these ideas into a cumulative risk factor for sepsis  that involved not only the category of sepsis, but the degree of organ dysfunction. So for example, for a child with septic shock and three organ dysfunctions,  the hazard ratio of death in the hospital from the septic event would be 2.37 to the third power  for the three organ dysfunctions multiplied by this other risk factor in the setting of septic shock. So for this hypothetical child, a huge risk, almost 600 fold  as compared to a critically ill child without sepsis or organ dysfunction.  What's going on right now has been going on since 2015 as a grassroots group called the Podium Collaborative. This is the Pediatric Organ Dysfunction  Information Update Mandate Collaborative  that came together at a meeting at NIH in 2015 around a conference focused on organ dysfunction in children. This group was organized to address the question  with contemporary information, what are the performance characteristics of currently used scoring systems and other assessments screen for both single  as well as multiple organ dysfunction in critically ill children?  The design of this work was initially reviewed by the Quality Effective Healthcare Program at Johns Hopkins Medical Center.  Working group included 88, I would say expert panelists from seven countries, 47 different institutions. They derived the PICO questions around 11 individual organ failures  as well as cumulative or composite organ dysfunction  and undertook systematic literature reviews including deriving evidence as well as the quality of that evidence ultimately to derive new definitions with new information since 2005  for single and composite organ dysfunction.  This work product hopefully will be published in Pediatric Critical Care Medicine as an electronic supplement during the coming year.  Now just a word about Sepsis Pathophysiology because that's what we intensivists deal with. This is a schematic showing that on the one hand,  inflammation which is well known in sepsis drives other process ischemia, reperfusion, dysoxia  which in turn drives inflammation. And these two processes in concert sort of a catch 22 situation are the underpinning for sepsis multiple organ dysfunction syndrome.  And what we do as intensivists is to focus ourselves  on resolving this organ function to move these patients ultimately out of the PICU and back into the community.  Well, it turns out that organ dysfunction is not all one color and the distribution of organ dysfunction is critically important in determining outcomes.  But some of the seminal work in this area has been done by Joe Carcillo at Pittsburgh Children's Hospital.  And for example, in this cohort of 100 children describes very different risk of mortality based on the type of organ dysfunction that might be present.  So really with one organ failure, the risk of mortality is very small, almost zero as compared to children's who demonstrate a macrophage activation syndrome  where the mortality is highest 70%. And then there are other organ dysfunction phenotypes like immune paralysis or thrombocytopenia associated multiple organ failure  that are associated with intermediate risk of mortality.  Similarly, more recently,  Chris Seymour and his colleagues  utilized very granular information from the electronic health records of patients who were enrolled in three different adult sepsis interventional trials.  And again, the focus here was how different organ dysfunction phenotypes affect outcomes in this case in adult sepsis patients.  So this is a schematic of what this looks like. On the bottom, you can see the alpha, beta, gamma, delta sepsis phenotypes and their relationship  to a variety of different types of organ dysfunction,  both qualitatively and quantitatively.  This is important because different sepsis phenotypes based on differential organ dysfunction is associated with different outcomes. Shown here is need for mechanical ventilation,  vasoactive inotropic support, duration of intensive care, and importantly, in-hospital mortality. So for these four different phenotypes, risk of mortality differed substantially  between the alpha, beta, gamma, and delta groups ranging from two up to 32%.  So what about outcomes and the relationship to organ dysfunction and sepsis? Well, the first thing that I wanna point out is this concept of organ dysfunction  is not only of interest to researchers,  but families of critically ill children and the healthcare professionals in the ICU who take care of these children, this concept is also important to them.  This is a figure which summarizes a survey study  in four pediatric intensive care units and participants were asked a simple question.  In a hypothetical trial, what would you choose as an outcome measure in addition to survival as something that's really important to you and your family?  And interestingly, both groups chose  what happens to the patient in terms of quality of life and functional status after recovering from a critical illness but very importantly, both groups also chose  duration of organ dysfunction. Both groups also identified a number of other important outcome measures but for the purposes of this discussion,  organ dysfunction is an important variable for both families, patients, and healthcare professionals.  John Lin utilized the Sprout Database which was an international point prevalence study  of sepsis again, demonstrating that risk of mortality  is clearly associated in a dose response fashion with a number of organ dysfunctions. This study ascertained previously published data  with a similar research design. In a more granular fashion, Lauren Schlappach and his colleagues in Australia and New Zealand actually drilled down  to look at different scoring systems, service criteria, the SOFA score adjusted  for pediatric variables, here's PILOD again, and the shortened version of the sequential organ failure assessment. This is a figure demonstrating  the receiver operator characteristic curves for these four different ways of assessing by scoring  in terms of looking at the risk for mortality. And as you can see here, the pediatric SOFA and PILOD scores demonstrated the best performance  and ultimately these investigators chose the updated version of PILOD  with a threshold of eight points.  Here, the score had good sensitivity, not so good specificity, but it did provide a negative predictive value of 97%,  which is pretty good, meaning that a score of less than eight points predicted a very low risk of mortality from sepsis.  Lastly, I would be remiss if I did not provide you with some information from the Life after Pediatric Sepsis Evaluation trial, which was published earlier this year  in Critical Care Medicine. This study really focused on long-term outcomes and was particularly interested in poor outcome at three months,  either death or severe quality of life impairment. And in multiple variable regression analysis, here's what was found. Summation of daily PILOD scores during ICU admission,  surrogate for intensity and duration of organ dysfunction,  highest vasoactive inotropic score, perhaps a surrogate for severity of septic shock and any new neurologic event. Each of these measures of organ dysfunction  independently predicted poor outcome from at three months following admission for pediatric septic shock.  So that's the update on pediatric organ dysfunction. Organ dysfunction, what we take care of and try to resolve in the intensive care unit clearly has an effect on patient-centered,  clinically meaningful short and long-term outcomes. Please look for the new evidence-based organ dysfunction definitions as a supplement to pediatric critical care medicine  coming to you soon during the next year. Thank you very much for your attention and please enjoy the rest of the Congress.  Hello, my name is Heidi Flory from the University of Michigan, and I'm really pleased to be able to talk to you today about how the Pediatric Surviving Sepsis Campaign Guidelines  have changed practice. My disclosures here are really related to my NIH funding right now and related to this work, working with SCCM and ESICM on the guidelines themselves.  So for today, we're going to run through briefly differences related to screening, the one and three hour bundle and initial management, as well as some revised management  as particularly relates to respiratory support and hemodynamics.  So why should we change our practice? Well, I think we've had data for a long time now, and this is from Dr. Hahn and colleagues, that when providing pediatric  advanced life support guidelines for sepsis early on in community hospitals, mortality is dramatically reduced. This data from Fran Ballamuth more recently  shows that when adding electronic health record related alerts appropriately, you can decrease the number of missed sepsis cases dramatically, and in this study,  sepsis identification increased from 83 to 96% with this change. Perhaps most importantly, however, is how and when it may in fact be more important than what.  And although we have recommendations related to vital signs and advanced hemodynamic monitoring, we're probably not going to go back to the age of the swan GANS catheter,  at least with the current set of literature we have.  So I would say that the Pediatric Surviving Sepsis Campaign Guideline in relation to screening does suggest implementing systematic screening for timely recognition of septic shock.  And this needs to be tailored to your own resources and procedures within each institution. However, I'm not necessarily going to call this one a change  because actually in 2017, our ACCM recommendation also suggested for a recognition bundle.  Moving on to timelines, fluids, antibiotics and diagnostics. So I think a major distinction today to point out is that we now have moved the one hour algorithm  for septic shock patients, and we have a three hour algorithm for suspected sepsis patients. But you'll notice this little arrow here  suggesting that if the patient is deteriorating rapidly and moving into shock, then you need to move quicker as we all suspected.  So this is indeed a change.  What about fluids? Well, how much fluid? Well, in PED surviving sepsis, we've softened the initial impact of 60 mLs per kilo in the first hour to now being 40 to 60 mLs per kilo  in the first hour with a clear reassessment between each bolus. And this is in particular in environments where resources are available.  We also qualified now that if there are no available intensive care resources, and the patient is not hypotensive, we are recommending against this fluid bolus,  and this is largely related to the FEAST trial results. We may have more information soon, as the SQUEEZE trial sponsored by Canadian Clinical Trials Group  is now underway with a trial for early goal-directed therapy versus usual care. So this is indeed a change.  What type of fluid? Well, in PED surviving sepsis, we have recommended crystalloids rather than albumin, and within crystalloids, we recommended balanced or buffered crystalloids  like plasmolyte or lactated ringers rather than normal saline. And while we don't have large RCTs about this, one is underway, and I'm sure my next colleague, Dr. Weiss,  will talk about this more. The PROMPT bolus trial is trying to enroll 8,800 children  with a pragmatic effectiveness trial. And this recommendation is, again, a change.  What about the one and three-hour distinctions? Well, some of this is driven by the antibiotic data that we have to date. And this data from Scott Weiss and colleagues in CHOP,  now over five years old, does show that the length of time of delay with antibiotic administration is definitely associated with increased odds of death.  This is observational data as well. But further data suggest that once this is risk-adjusted, this distinction isn't necessarily as severe as in prior study.  So this is indeed a change in terms of time management. And our recommendation here, you can see these forest plots that are all found in the Supplemented  PEDE Surviving Sepsis Campaign, says that for shock, please hold tight with that one-hour recommendation. And as well with antibiotics, a shout out to our IDSA colleagues  who have been involved with this work from the start. Without immune compromise or risk for multidrug-resistant organisms, we are suggesting against the routine use  of empiric multidrug therapy. Again, and also we are recommending a daily reassessment for de-escalation, and this is a best practice statement. So this is indeed a change.  What about diagnostics? So for diagnostics, we are suggesting not to use just bedside clinical signs in isolation, and to categorize warmer cold shock,  and we'll talk about that in a second, as well as some use of advanced hemodynamic variables as able, including trends in lactate, and we'll talk about that next.  So this data published recently shows the results of inter-rater reliability with warm and cold shock assessment using clinical variables. You can see high inter-rater reliability  for temperature, capillary refill, and pulse strength,  but not good inter-rater reliability at all, and disagreement between the clinical signs and shock type of warm or cold related to pulse pressure and diastolic blood pressure.  And indeed, in our review of the data for PEDS surviving sepsis, we have indicated that cardiogenic component is definitely in part related both to cold and warm shock categories.  And in this study, the subgroup of 86 patients requiring vasoactives, when vasoactives are begun between 30 minutes before an end of the documentation,  there was mismatch between five clinical signs and vasoactive match very often, but it was not associated with the worst clinical course.  So we don't have a recommendation for first-line vasoactive in children with septic shock, a distinct change.  Lactate measurements have been more recently shown to be quite effective in identifying patients and classifying severity. This very large observational study  showing highest lactate levels associated with intubation, admission, length of stay, even mortality, in contradistinction to lower levels.  As well in trending lactate beyond the initial assessment,  multiple observational studies, although not that large in number, are also favoring lactated guided measurement, a distinct change. And this study from Colorado in 2016,  showing that normalization of lactate over the first four hours was associated with decreased mods and decreases in hospital and piculance of stay.  Going on to mixed venous saturations, this study published very recently shows around 150 pediatric patients with septic shock that even with intermittent mixed venous saturation  assessment within 19% of intermittent versus 36% of continuous mixed venous saturation monitoring achieved shock resolution within six hours.  In contradistinction within 24 hours, even with intermittent mixed venous saturation monitoring, over two thirds of patients achieved shock resolution.  This data from De Oliveira and colleagues also shows that the target of 70% for mixed venous saturation seems to be beneficial in discriminating high-risk patients,  as well as potentially improving mortality as depicted in this Kaplan-Meier graph.  So there are gonna be many opportunities for monitoring and diagnostics in the future. The gauntlet is down for us, and this is a distinct gap, which I'm sure that Dr. Weiss  in the next talk will talk about.  What about therapies? Extracorporeal treatments. There are some slight changes here as well that are quite worth noting. We definitely suggest against using plasma exchange  and septic shock unless perhaps in the thrombocytopenia associated group. And we do suggest using renal replacement therapy, CRRT,  if unresponsive to fluid restriction and diuretic therapy to treat fluid overload. But we are suggesting against high volume hemofiltration strategies at this point,  barring further evidence.  Within extracorporeal membrane oxygenation, we definitely are suggesting VV for refractory hypoxemia in sepsis, and VA for refractory for both respiratory and hemodynamic treatment.  This data, which is a propensity score analysis of restore population patients by my colleague, Dr. Barbaro at University of Michigan, shows actually that there was not  a distinct mortality difference in those patients who received ECMO versus non-ECMO, although only around a quarter of the group, of each of these groups had sepsis.  And there was some suggestion that there was a lower mortality in the ECMO group with a propensity score matched, and perhaps even more with a sensitivity propensity score match,  but it did not reach statistical significance. So data supporting ECMO use in those refractory for treatment, but more data needed to come.  We cannot recommend against certain therapies at this juncture, and hopefully we'll have more information, but the Half-Pint trial and the various antioxidant trials  have indicated so far that tight glycemic control, antioxidant therapy, et cetera, are not necessarily beneficial in this group. The TAXI studies and the ABC-PICU trials  have definitely also supported favoring a lower hemoglobin threshold at this point, but certainly certain subgroups are definitely warranting these things,  for example, IVIG in the MIS-C shock patients. So caution is encouraged, but in general, routine use is not encouraged at this time. And the SHIPS trial is underway right now  to test the hydrocortisone and septic shock aspect enrolling over 1,000 patients.  Moving on to respiratory management, I think this is very important to recognize the timing and reassessment is very important in all the respiratory aspects of shock patients.  We do have recommendations now that noninvasive ventilation does seem indicated at least as a trial, and this is definite change. And this very recent data  published by Guillaume Emeraud and colleagues with the large NEAR for Kids group does actually show that despite a trial of noninvasive ventilation that failed,  severe adverse events are not increased in those patients subsequently requiring intubation, but great caution is advised in this group.  Again, we are also recommending higher PEEP strategies, and this is a large observational study, and therefore this will not reach strong recommendation criteria,  but we did show that in large multicenter studies that when you're not adherent to the ARDSnet-FIO2 grid in those patients with severe PARDS,  there is a distinct mortality difference, and so this is a clear recommendation and a change.  With regard to nitric oxide, we wanted to mention that clearly in those patients with very severe oxygenation index early on, INO could be trialed, but in all comers,  particularly with persistent hypoxemia, there does not seem to be an effect, so we suggest using INO as a rescue therapy in sepsis-induced PARDS.  With regard to oscillation, and we're nearing the end of our talk now, we have had some lackluster support of high frequency in post-hoc analyses of larger studies,  and RCTs in high frequency are very small in number. So right now we cannot issue a recommendation, and this is a distinct change,  but some recent data from the Netherlands does show in severe PARDS patients with a very tightly protocolized high frequency algorithm,  oxygenation indices can be dramatically improved, including in the severe group, and ventilation and CO2 clearance does seem to be quite effective as well. With prone positioning,  we are now suggesting a trial of prone positioning, and I caution you because the initial studies in prone positioning enrolled mild, moderate, and severe PARDS patients,  and this was a no-difference pediatric study, but many data have now supported in the adults, particularly for severe ARDS mortality benefit,  and based on that, we are suggesting a trial in severe PARDS patients. However, both of these aspects, oscillation and prone positioning,  are currently being tested in a large multicenter trial.  Finally, we will speak to a neuromuscular blockade, and we are suggesting using neuromuscular blockade with children in sepsis and severe PARDS,  although we will look towards this randomized control trial undergoing now in the Netherlands to help give us additional information in the future. So again, a change.  With all this, I thank you for your time and support in allowing me to present this data to you. I look forward to questions and comments,  and here is my email and my cell phone, and bye.  Hi, my name is Scott Weiss, and I appreciate the opportunity to present on gaps in the literature and current research related to pediatric sepsis.  I have no disclosures that are relevant to the content that I will be presenting on today.  So the Surviving Sepsis Campaign Children's Guidelines were published earlier this year in February of 2020 in Pediatric Critical Care Medicine and Intensive Care Medicine.  And there were 77 statements issued  as a result of the panel's work, but you'll notice that only 15 were recommended as strong recommendations or best practice statements, while the majority, back to over 80%,  were either weak recommendations or suggestions, or there was no sufficient evidence to issue any kind of recommendation at all.  So as a result, a separate subgroup of the panel that was chaired by Dr. Robert Tasker set out to identify key pathophysiological gaps and topics for clinical trials.  And ultimately, 52 research questions were identified as priorities to try to address some of the lack of sufficient data  to issue stronger guidelines, and 23 of these were related to pathophysiology, and 23 were related to potential clinical trials.  So what I'm gonna do is go through some of the key statements that were  either weak recommendations, indicated by we suggest, or no recommendation was able to be offered, and I'm gonna highlight the pathophysiology  and clinical trial research gaps that were identified, and in some cases, show where new research is being done that will help to answer some of these research questions,  hopefully for subsequent iterations of these guidelines. So the very first statement was a suggestion to implement systematic screening for timely recognition  of children with septic shock and other sepsis-associated organ dysfunction. And this was based on very low quality of evidence. The research questions that were identified  were the need to develop better algorithms to improve recognition of clinical deterioration, as well as to test whether systematic screening can improve early recognition  and ultimately patient-centered outcomes compared to usual care.  And some of this work is definitely ongoing. So this is a study recently published from the group at Nationwide Children's Hospital where they compared manual documentation  of sepsis recognition with a embedded best practice electronic alert that was not visible to the clinicians. And what they were able to find  is that all 29 cases that were manually documented to having sepsis were identified by the alert,  but the alert identified these children much sooner than the manual documentation, suggesting that clinical decision support embedded within the electronic health record  may actually help us to improve early recognition of children with sepsis.  Screens like this need to be tested further in clinical trials.  There were two statements about lactate. There was no recommendation issued about whether or not blood lactate could be useful to stratify children into low versus high risk  of having septic shock or sepsis if they presented with a concern for infection. And there was a suggestion to trend blood lactate levels in addition to other clinical assessments  to help guide resuscitation, again, based on very low quality of evidence. Research gaps highlighted were the need to identify what the optimal level of hyperlactatemia is,  both in the ED and ICU setting, as well as to determine and test whether resuscitation guided by lactate measurements and improvements in lactate levels over time  can actually improve outcomes relative to using alternative hemodynamic indicators of improved shock.  With regards to antibiotics, two key recommendations were a strong recommendation to institute antimicrobial therapy as soon as possible for those children with septic shock,  ideally within one hour of recognition. You'll note, however, despite being a strong recommendation, this was also based on very low quality of evidence.  And in those children with suspected sepsis, but without clinical evidence of shock, it was a suggestion to start antimicrobial therapy as soon as possible,  allowing for an appropriate expedited diagnostic evaluation such that antimicrobials will be started no later than three hours after initial suspicion of sepsis  for those children ultimately determined to likely have infection and or sepsis.  However, a research gap that was identified was the need to really determine the optimal timing of antimicrobial therapy, which should be embedded as part of an overall care bundle  that include other elements of resuscitation. And these two recommendations were based largely on two observational studies, and there's clearly a need for additional studies  to provide more high quality evidence to determine how best to guide antimicrobial therapy in children with sepsis.  For fluid therapy, two key recommendations were that in those healthcare systems where intensive care is accessible, either locally or via transport to a nearby facility,  there was a weak recommendation or a suggestion to provide up to 40 to 60 mLs per kilo in bolus fluid over the first hour. Whereas in healthcare systems  where intensive care was not able to be provided either locally or via transfer, the recommendation was to withhold bolus fluid and instead utilize maintenance fluids.  That latter recommendation was based on high quality of evidence that stemmed from the FEAST trial, which was a randomized clinical trial. Whereas the first recommendation there  was really based largely on observational data, and so it was qualified as low quality of evidence.  So clearly there is a need to help figure out how best to recognize early symptoms of fluid overload, as well as the best clinical markers for cardiac output and perfusion  to guide optimal fluid resuscitation in all settings.  Similarly with vasoactive medications, there was a weak suggestion to utilize either epinephrine and norepinephrine over dopamine  as first line therapy for children with septic shock. And to begin vasoactive after 40 to 60 mLs per kilo of fluid was given if abnormal perfusion persists  or sooner if fluid overload or myocardial dysfunction was present.  There's clearly a need and a gap to identify the best choice, either epinephrine or norepinephrine, and the optimal timing to start vasoactive infusions  in combination with fluid resuscitation. And also if and when to use other vasoactive medications such as vasopressin or inodilators such as milrinone.  Here we have two trials, one completed and one ongoing, that helped to inform some of these research gaps. So the FISH study conducted by Dr. David Inwald  and colleagues in England was a multi-center trial that showed that if you use 10 mLs per kilo instead of 20 mLs per kilo boluses, that patients received less fluid  and there was no difference in patient outcomes, suggesting perhaps we could get away with smaller fluid boluses. And the SQUIEZE trial being performed in Canada  is looking at early initiation of vasoactives compared to usual care for children with septic shock and will help to provide important information about the timing of vasoactives  versus ongoing fluid resuscitation.  An additional suggestion for fluid therapy was to use balanced or buffered crystalloids such as lactated ringers or plasma light for initial resuscitation,  again based on very low quality of evidence.  And here the research gap identified was a comparative effectiveness trial between balanced crystalloids and saline.  There has been a recently completed multi-center trial in India at four centers that did this very trial with a primary outcome of acute kidney injury and was completed in 2020  and we're actually awaiting the results of that study. And there's a larger ongoing multi-center international trial that had recently started enrolling  at sites in the US and will soon include sites in Canada, Australia, and New Zealand that will seek to enroll close to 9,000 children that will hopefully provide definitive evidence  for balanced fluid versus saline in the treatment of children with suspected shock.  Similarly, hemodynamic monitoring was an issue where the panel was unable to issue a recommendation about the optimal blood pressure targets  and therefore our research gap identified was the identification of blood pressure targets such as either the fifth or 50th percentile  mean arterial pressures that should be targeted for children with septic shock.  There was a recent study that demonstrated  that showed if you compared current documents that provide guidance about optimal blood pressure targets such as Pew Score, PALS, or the Goldstein definitions for pediatric sepsis,  if you compare those to the population-based ranges of fifth percentile systolic blood pressure shown in the gray shaded boxes across different ages on the x-axis,  you can see that the targets shown in different colors vary substantially. And so there's no general agreement about what the best blood pressure target should be  for initial resuscitation of children with shock. And this is especially relevant given the increasing amount of indirect data from adult studies that suggest  that slightly lower blood pressure targets might actually be more effective and may spare patients unnecessary increases in exposure to vasoactive medications.  For hydrocortisone, there was a suggestion to avoid hydrocortisone if fluid resuscitation and vasoactives can restore hemodynamics. But it's another suggestion to either use  or not use hydrocortisone if fluid and vasoactives are unable to restore hemodynamic stability. And this latter suggestion, this weak recommendation based on low quality of evidence  reflected the ongoing equipoise about whether or not hydrocortisone as an adjunctive therapy for children with refractory shock is beneficial.  Of course, the research gap was a clinical trial to answer this question, and such a trial is ongoing. The SHIPS trial, a multi-center randomized controlled trial  in the US and Canada, comparing adjunctive hydrocortisone to placebo for fluid and vasoactive refractory septic shock.  For metabolic support, there were many weak or no recommendations. I point out two here with weak suggestions against the use of vitamin C  or thiamine to treat children with sepsis or septic shock.  And in here, clinical trials to test these therapies has been recommended.  There was a recently published retrospective observational study from Lurie Children's Hospital in Chicago that suggested that resuscitation that included hydrocortisone,  ascorbic acid, and thiamine or HAT therapy was superior to resuscitation that did not include that triad of therapies. However, this was an observational study  with issues of confounding by indication. There's now an ongoing randomized controlled trial by the ANZICS group in Australia that is comparing early vasoactives  with and without HAT therapy versus usual care. We actually await results from that pilot study and that will go on to do a larger study for more definitive results  following completion of the pilot study.  For red blood cell transfusion, there was no recommendation made regarding the optimal level of hemoglobin transfusion thresholds for clinically ill children  who remained unstable with septic shock. And so a research gap was identified as the optimal hemoglobin level or are there other targets one should use in order to optimize  when to utilize red blood cell transfusion. And there's some work that's being started to try to identify this. Some studies through the PALISI group  that hopefully will be coming down the pike in the next few years.  Finally, there was no recommendation that could be made for or against the use of plasma exchange in children with septic shock who experienced thrombocytopenia  associated multiple organ failure or TAMOF  despite some observational studies suggesting that these may be helpful.  But we really need better trials to test these therapies in order to more definitively determine whether or not they're truly beneficial.  So the Adult Surviving Sepsis Campaign following their latest update guidelines  went through a process where they were able to take their initial research gaps and questions and hone those down to 10 high priority clinical questions.  And this was reported in a recently published separate manuscript. There's an ongoing process now to do a similar set of steps to identify the key top research priorities  for children with sepsis and septic shock. That should be hopefully available sometime later in 2021.  So in conclusion, I like the quote by the French physiologist Claude Bernard who said, it is what we know already that often prevents us from learning.  Many of us who take care of patients with sepsis and septic shock feel very strongly about the way in which we resuscitate these children  but we must remember that many of the things  that we do are based on either no or low quality data at the moment. And so we must continue to push ourselves to come up with, to perform better studies  to come up with more high quality data so that we can ensure that we're providing the best possible care to our patients. To start this process,  the Pediatric Surviving Sepsis Campaign made explicit over 50 knowledge gaps and will hopefully distill these down to a smaller number of high priority  clinical and basic research questions. And I've tried to show a small sampling of ongoing studies, including some clinical trials that will provide us with new and exciting data  that will continue to inform and improve the way that we provide care to children with sepsis and septic shock. And so I'm excited about the future and hearing about these new data.  Thank you very much. I appreciate the opportunity to present today.  All right, let's move on to our question and answer session. We have a great question for Dr. Zimmerman.  Could you comment on what you feel to be the most important organ dysfunction scores to know at 3 a.m. when dealing with a critically ill child and why, meaning how that or those  scores are helpful in adjudicating short-term and immediate management plans? Jerry? Thanks, Margaret. Well, there are a number of scores.  I think right now, overall, the best validated one is the second version of PILOD. I think what this score does for you is it may not direct your management in that moment,  but all of these organ dysfunction scores, if they are high, if they are prolonged, should alert you to the fact that this is a child that is going to be in trouble, either in  terms of risk for mortality or they need to be watched because they are at high risk for long-term morbidity.  So I can pass the baton on here for a question that I have, and I believe that assessing adequacy of fluid replacement or resuscitation is still a big deal.  So Scott or Heidi, in your opinion, do we have evidence of best practice for assessing this or what's your own practice?  What would you recommend before we go to that vasoactive inotropic agent? I'll cede to Scott in a quick second, but thanks for the question.  I think one of the biggest takes I took from working on Pete's surviving sepsis again and reviewing the literature is this do something, assess, do something, assess. It's so basic.  It's something we do all the time. It's part of all of our CPR challenges, and I think the same holds true for fluid management.  Each person, you're at the bedside, you're with the patient, you give a bolus, you see the response, you do it again.  You have to be very active and dynamic in that process, and I think there is some nice literature talking about clinical signs and symptoms, and thankfully being at the bedside  helps. And I think you can't get good blood pressure or hemodynamics if the intravascular space is empty. Yeah.  It's an important question, and one that I think that we as a field I think still struggle with how best to do those assessments and reassessments at the bedside.  Everything starts with the clinical exam, and so you want to be attentive to those things  that suggest the patient's in shock, although be cautious about trying to infer a state of high or low SVR or presence or absence of myocardial dysfunction using clinical signs  alone, the so-called cold versus warm shock, because they could be sometimes misleading. But once you've assessed that a patient has some abnormalities and perfusion, giving a  fluid bolus and reassessing to see the impact of that fluid bolus on those abnormalities is important. But then moving towards more advanced monitoring is also important.  So currently there's increasing evidence to suggest that bedside ultrasound can be very  helpful, although the best measures to assess fluid responsiveness are still somewhat debated.  And then moving forward, more other options to continuously measure hemodynamics such as pulse wave Doppler and other technology that can assist clinical decision-making at  the bedside is coming to pediatrics and should be things that we seek to implement through a rigorous data-driven manner where we test the benefits of using those new technologies.  Both of you, Scott and Heidi, talked about using alerts from the EHR for earlier diagnosis of sepsis, because clearly earlier diagnosis is one of the keys to improving outcomes and  survival. You showed some studies that demonstrated that, in fact, diagnosis is earlier. Looking at the other side of that question, are there problems with false alerts, you  know, alarm fatigue, if you will, and are some of the alerts, are they too sensitive? You know, what's the false positive rate in the studies, in the literature, and is there  a way to sort of redesign the alert to make them more effective, more specific without losing too much sensitivity?  Yeah, I think that we definitely need to have data from PDCA-ing these alerts and making sure that the false positivity isn't too high.  And I know the Ipso collaborative within the United States hosted by Children's Hospital Association is trying to enable and PDCA the alerts, and each institution is doing  something slightly differently, but learning from each other in the process. I think that's absolutely key.  And we also talked about antibiotic stewardship during the PEAT surviving sepsis campaign. Earlier is presumed to be better, but we have to be smart about it.  Scott, do you have other additions to that? Yeah, I agree with those points, Heidi. Absolutely. I think I'll add two things.  One is that I think we need to continue to work with our bioinformatics colleagues to work towards more specific alerts that utilize trends that are otherwise invisible to the  human eye, but that are available through algorithmic re-evaluation of data that's available from either bedside monitors or in the EHR.  And second is that screens to alert the clinician that a patient may have sepsis are best used  in that manner, rather than to trigger automatic treatment that I think they're most effective when they're used to suggest that a patient has a clinical change that may or may not  be due to sepsis, and helps to reset somebody's decision-making and thought process in a multi-professional  bedside huddle, so that the screen augments decision-making as opposed to drives it. Margaret, could I add to that? Yes, of course.  I also want to just point out that the electronic medical record is also a source of information for in real time, eventually, identifying sepsis phenotypes.  And I neglected to mention a very important article in my presentation by Nelson Sanchez-Pinto and the hospitals at Lurie and Palmer in Chicago.  And for example, this is a really important paper to read, he identified this cohort of patients with persistent hypoxemia and shock that are at significant risk for resource  utilization and poor outcomes. So another way that we are moving with the electronic health record. And the EHR can keep an eye on that fluid balance we were talking about, too.  Heidi, there's a question for you. You talked about recommendations for ECMO in refractory septic shock patients.  Pushback from surgeons is often based on bleeding risk with DIC or thrombocytopenia and so forth, and source control after placing new plastic.  Can you comment on these concerns based on currently available evidence?  Yeah, that's a good question. But I think the evidence is pointing more and more toward being able to attain hemostasis  and being able to get these patients on when needed to, and to being able to adjust anticoagulant  therapies and having protocols to be able to be off anticoagulants to allow for procedures to be done, and then judiciously started again.  So that is important to assess, but I think the data is suggesting we're doing better at it over time through the ELSO cohort.  Scott, if you had unlimited resources and collaborators, how would you design the ideal pediatric sepsis trial, and what would you test?  Unlimited resources and collaborators, sounds nice. I could use that. So, it's a great question.  You know, there's been a lot of discussion over the recent years, including at this virtual  Congress, about taking advantage of prognostic and predictive enrichment strategies so that  we can selectively target patients for therapies that are most likely to benefit their pathobiology.  And so I think in the ideal sepsis trial, we would have rapid diagnostic tests that allow us to differentiate patients into these different phenotypes, some of which Jerry  described, although there are others emerging as well. And then we would test a series of therapeutics that are most likely to benefit those patients.  And we not only need those tests to be available at a single point in time, but we also know that some patients change phenotypes, sometimes rather quickly.  And so if there's a delay between initial eligibility assessment and initiation of that interventional therapeutic, we need to confirm that the patient still has that phenotype  and avoid treating patients with therapies that are unlikely to be beneficial. So we can reduce the noise in trials and help to hone our therapeutics better to align with  the patient's pathobiology. Thank you.  That concludes our session today. And thank you all for your excellent, excellent talks. And we'll follow up with the additional questions as they come in on the website. Thank you.  Thanks, everyone.

Video Summary

In this video transcript, Dr. Margaret Parker, Dr. Jerry Zimmerman, and Dr. Heidi Flory discuss various aspects of sepsis, including organ dysfunction, guidelines, gaps in the literature, and ongoing research. Dr. Parker introduces the session and highlights the topics that will be covered by each speaker. Dr. Zimmerman discusses the importance of organ dysfunction in pediatric sepsis and the need for accurate definitions and scoring systems. Dr. Flory focuses on how the Pediatric Surviving Sepsis Campaign guidelines have changed practice, including screening, fluid therapy, antibiotics, and hemodynamic monitoring. Dr. Weiss discusses the gaps in the literature and current research related to pediatric sepsis, emphasizing the need for more evidence-based practices to guide clinical decision making. The speakers highlight the importance of early recognition and diagnosis of sepsis, the role of fluid resuscitation and antibiotic therapy, the need for better methods to assess organ dysfunction and fluid balance, and ongoing research to address these gaps. They also discuss the challenges of using electronic health record alerts for sepsis diagnosis, concerns regarding ECMO in refractory septic shock patients, and the design of an ideal pediatric sepsis trial. Overall, the session provides valuable insights into recent advances and ongoing efforts in the field of pediatric sepsis.

Asset Subtitle

Pediatrics, Sepsis, 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.

Learning Objectives

Describe the newer paradigms for definition of new or progressive multiple organ dysfunction syndrome (MODS) and recurrent MODS in pediatric sepsis
Review the most recent Surviving Sepsis Campaign pediatric sepsis guidelines
Highlight ongoing research and existing gaps in the literature

Organ Dysfunction in Sepsis: What's New With the OD in MODS?
Jerry J. Zimmerman

How Have Guidelines Changed Practice?
Heidi R. Flori

Gaps in the Literature and Current Research
Scott L. Weiss

Meta Tag

Content Type Presentation

Knowledge Area Pediatrics

Knowledge Area Sepsis

Knowledge Level Foundational

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Guidelines

Tag Scoring Systems

Tag Shock

Year 2021

Keywords

sepsis

organ dysfunction

guidelines

ongoing research

pediatric sepsis

fluid therapy

antibiotics

hemodynamic monitoring

clinical decision making

advances

	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.