SCCM Resource Library-Lessons Learned From Clinical Studies in Pediatric ARDS

Video Transcription

So thank you all, it's a real honor to be here. Today I'm going to talk about lessons learned from clinical studies in pediatric ARDS. I have no real relevant disclosures.  Some of my research has been funded by the following entities, but they're not relevant for this talk.  So the Pediatric Acute Lung Injury Consensus Conference was convened from 2012 to 2015 and created a pediatric-specific definition for ARDS, as well as clinical recommendations  in eight different areas related to ARDS management. Now unfortunately, there was not a lot of evidence to form many of these recommendations  and there was a real call for future research. The goal of this talk are to review the top 10 lessons that we've learned from recent  clinical trials in pediatric ARDS, really since the publication of the initial pallet recommendations, as well as introduce some new concepts that are emerging as we're creating  the update to the pallet recommendations, i.e. pallet two, which are due to come out this summer.  So the first thing that became clear, really, even as the first pallet was created, was that we needed a pediatric-specific definition for ARDS.  While there's a lot of overlap between adult and pediatrics ARDS definitions with respect to pathobiology and physiology, there are many specific components for both adult and  pediatric practice, as well as comorbidities that really do necessitate a pediatric-specific definition.  This prompted creation of the Pallet Definition, which was published in 2015. And there are many similarities between the Pallet Definition and the Berlin Definition,  with some notable exceptions related to simplification of the chest imaging-based criteria, the use  of oxygenation index instead of the PF ratio for those on invasive mechanical ventilation, as well as the use of pulse oximetry-based criteria with oxygen saturation index, or  the SpO2 to FiO2 ratio. The use of pulse oximetry-based criteria were one of the major changes with the Pallet Definition as compared to the Berlin Definition.  And it's important to consider why this happened. Our practice patterns in pediatric ventilation management have changed, and arterial lines  are no longer standard of care for a lot of patients that are on mechanical ventilators. And this is a study, in fact, that we published around the time that the Pallet Definition  came out that highlights that if we were to limit the definition of ARDS to those that had an arterial line, we would be selecting for a different patient population.  Those that have arterial lines are much more likely to be on vasoactive agents, nearly  50% of them, compared to only 10% of those that do not have an arterial line. And so the use of the arterial blood gas-based criteria will select for patients that have  this combination of hypoxemia and cardiovascular dysfunction. This will underestimate the prevalence of ARDS and potentially overestimate ARDS mortality  by selecting for a more ill patient population. While the Pallet Definition was based largely on consensus opinion, there were some older  studies that were used to derive the cut points for mild, moderate, and severe ARDS. But really, it was unclear how well this definition would perform in actual practice.  And as a consequence, the PARTY study came about over the last several years in an attempt  to try to understand the real implications of the Pallet Definition with a point prevalence type design conducted in 145 international pediatric ICUs.  As part of this study, we screened over 23,000 ICU patients, about half of which were on mechanical ventilation, and identified 744 new cases of pediatric ARDS.  This represented 3% of all ICU patients and 6% of those on mechanical ventilation with an overall mortality rate of 17%.  When looking at the overlap between the Berlin Definition and the Pallet Definition, what we see is that at the time of PARD's diagnosis, only about a third of the patients that met  Pallet criteria for ARDS would also meet Berlin criteria for ARDS, i.e. with a PF ratio less than 300 in bilateral infiltrates, although that group had the highest mortality  of over 25%. There's a large subset of patients that also have bilateral infiltrates but do not meet  the PAO2-based criteria, but would meet criteria if an SpO2-FiO2 ratio, for example, was permitted.  By the time we get out to three days, there's a lot more overlap between the Berlin Definition and the Pallet Definition, with now up to 60% of patients having an arterial line with  a PF ratio less than 300 in bilateral infiltrates. But there's still a substantial number that either never have an arterial line placed,  but there is a relatively small number of patients, only about 10% of the cohort, that persist with bilateral infiltrates.  So this highlights that the Pallet Definition is likely identifying patients with ARDS sooner than would be identified with the Berlin Definition.  The second important lesson is that not all outcomes in ARDS are about the PARD severity. While that's part of the diagnostic criteria and the risk stratification of mild, moderate,  and severe based upon the oxygenation index groups with PALLIC or their PF ratio groups  with the Berlin, what we see in our PARD cohort is, in fact, that there's very little discrimination  of mortality risk between mild and moderate patients, at least at PARD's onset, with a large step-up in mortality that occurs with patients that have more severe disease.  And Nader Yahia led the V1 sub-study of PARDI, which tried to identify risk factors, independent  risk factors for mortality in pediatric ARDS patients, and what we can see here is that although the PF ratio is a marker, certainly, of mortality risk, being immune-compromised  or having other organ dysfunction on the day of PARD's diagnosis were much more relevant with the highest BIC for the risk of mortality in multivariable modeling.  Lesson number three really surrounds non-invasive ventilation, and it has been increasingly used in ARDS populations, both in adult and pediatrics, with the advent of new interfaces  and modes. But the question still remains about, should we be using non-invasive ventilation in ARDS patients, and when is the right time to intubate?  And if we look to our adult colleagues, based upon the Lung Safe study, there's a suggestion that the use of NIV is associated with harm.  There was higher ICU mortality in patients that were exposed to NIV, and this was particularly  evident in those with the more severe hypoxemia, with a PF ratio less than or equal to 150. Now what pediatric data do we have?  There was an interesting analysis that came out of the RESTORE clinical trial that was focused only on patients that were invasively ventilated, but looking at pre-intubation  NIV exposure. And here what we see is that patients that had pre-intubation NIV exposure had a longer  duration of ventilation, longer PICU length of stay, and in fact, even higher in-hospital mortality.  However, if we only limit our analysis to those that end up getting invasively ventilated, we don't understand the population that never went on to get intubated.  So that's what was addressed here in another secondary analysis of the PARDI study. We see that certainly the rates of intubation are very high amongst those with non-invasive  ventilation PARDS, where 50% of them go on to get intubated. But regardless of the initial PF ratio, especially when the initial PF ratio is greater than  100, children in the NIV group had an overall shorter duration of ventilation, counting both NIV and IMV together.  For those with a PF ratio less than 100, there was no clear difference between the two.  But regardless, there was no independent association between the use of NIV and ICU survival in any of the multivariable models or in propensity-matched analysis.  So this suggests that potentially using NIV, if done in an appropriate manner, may not increase the risk of harm for pediatric ARDS.  Lesson number four concentrates on the idea of lung strain or choice of tidal volume. And as we think about strain as a concept, it's the change in size of a structure in  relation to its original size. So as we think about this with respect to the lung, this would be the tidal volume over functional residual capacity.  So if we think of the illustrative example of a normal patient that's being ventilated with 10 mLs per kilo, who has a functional residual capacity of 20 mLs per kilo, the  strain is 0.5. If that patient develops ARDS and the functional residual capacity is, let's say, cut in half and the tidal volume is kept the same, now the strain has doubled.  So that's part of the reason to reduce the tidal volume if that functional residual capacity is also reduced. If it's dropped to 6 mLs per kilo, now the strain is lower once again.  And this forms the basis for the PALICC recommendations with respect to tidal volume, that tidal volume should be patient-specific.  Those with the most severe impairments and respiratory system compliance should receive the lowest tidal volumes of 3 to 6 mLs per kilo.  And those with more preserved compliance, more in the 5 to 8 mL per kilo range. And Anoop Bala has found that, really, in our actual practice in pediatrics, tidal volume  doesn't really vary much based upon lung disease severity, and certainly not based upon delta pressure, which, again, would be a surrogate here for lung compliance, that we tend to  ventilate patients between 7 to 8 mLs per kilo, sort of regardless of lung disease severity, and we're very non-adherent, particularly in those that have higher delta pressures  of greater than 18, where, in fact, even the tidal volume is suggested to be slightly higher.  In addition, we're often using PEEP levels well below what would be recommended by PALICC, which would say greater than 10, up to 15, or even higher for those with more severe  disease, and lower than what is typically recommended by, for example, the ARDSnet even low PEEP, high FiO2 table, which is what's shown here with the bars.  What we do in actual practice, especially as we look over here to the right, is that we rarely escalate PEEP as FiO2 is increased, really capping the PEEP levels at about 10  centimeters of water. And Anoop found in this secondary analysis from the party data that failure to adhere  to that PEEP FiO2 table, for example, i.e. the use of high FiO2 with relatively low PEEP, was associated with worse outcome, higher mortality, as was failure to reduce the tidal  volume in those patients with more severe disease, i.e. those that had more non-adherence to the tidal volume recommendations were much more likely to die.  Lesson number seven looks at ancillary therapies, another party study that Courtney Rowan took the lead at.  And what we see is that, particularly the use of neuromuscular blockade, nitric oxide, prone positioning, et cetera, have a lot of variation from region as well as income  status of the country. With prone positioning, for example, used very infrequently in North America or in high  income countries, but used very frequently in middle income countries and in areas like South America, Australia, Asia, and the Middle East.  And this highlights the need to think about global implications for even things like clinical practice recommendations based upon the availability of these therapies.  Now unfortunately, we still lack a lot of information about what patients are going to benefit from some of these rescue or ancillary therapies.  And here using nitric oxide as an example, there's been a new RCT that was published related to nitric oxide since the initial publication of PALIC, which does suggest that  for some subset of patients, perhaps those that have a high risk of going on to ECMO,  the use of nitric oxide might be beneficial, but there also may be some significant heterogeneity in treatment effect.  And this concept of heterogeneity and treatment effect has been quite well described in adult  ARDS studies, especially as it relates to, for example, PEEP management strategies, where patients with more hypoxemia, i.e. lower PF ratio, or patients that have phenotypes that  are more consistent with hyperinflammation are the ones that are perhaps benefited from the high PEEP strategy, whereas other types of patients may in fact be harmed.  And while we have fewer randomized control trials in pediatric ARDS, certainly this concept of heterogeneity and treatment effect is likely important for us too, and we've certainly  seen it as we look back to the calfactin study from 2005. We see there was a clear difference in terms of the potential benefit of calfactin being  more beneficial potentially in those with direct lung injury as compared to those with indirect lung injury, although, of course, that did not pan out in the subsequent follow-up  study for calfactin done only in those with direct lung injury. But Mary Dahmer and the Bali group from the RESTORE clinical trial have really added to  our understanding of these phenotypes with this latent classifier approach to really characterize that the hyperinflammatory phenotype that was described in adult ARDS patients  may also exist in pediatric ARDS patients, and importantly, that just the severity of  ARDS alone does not capture this hyperinflammatory versus non-hyperinflammatory phenotype, although  those with more severe disease appear to be more likely to fall into that hyperinflammatory group.  And our last lesson here is that there are a lot more outcomes than mortality and length of mechanical ventilation, although these are the outcomes that we often focus on in  our clinical trials. We see that there are many post-ICU sequelae for our ARDS survivors, up to 20% of them  have a significant decline in health-related quality of life, up to 60% of them have new  equipment or healthcare needs, and unfortunately, many of the risk factors for these poor outcomes  are not necessarily modifiable by our therapies in the ICU, i.e. these may be sociodemographic factors that affect this as well as diagnostic factors.  In addition, length of mechanical ventilation or even ARDS severity may not necessarily  have a strong association with post-ICU health-related quality of life, which brings into question the use of these outcomes as our primary metrics in our clinical studies.  So where do we go from here? Certainly, there have been many lessons that we've learned over the last five to seven years since the publication of the initial PALIC.  And as a preview, as we're updating PALIC, there are many new things that we need to consider that these groups are focused on.  For the definition subgroup, for example, there's this idea of do we need to have persistent  PARDS, i.e. there are many patients that after some period of time, 6, 12, 24 hours, they no longer meet PARDS criteria.  Should these patients still be classified as having PARDS or not? Moreover, what should we do about our evolution in practice with respect to non-invasive ventilation  and high-flow nasal cannula? As has become clear with the COVID pandemic in adults, many patients with ARDS are managed on high-flow nasal cannula.  And so where provider preference for choice of interface can have a major impact on whether or not somebody is being diagnosed with a particular disease.  And Steve Schein did some nice work that highlights that the SF ratio, which can be calculated either on nasal or non-nasal modes of non-invasive ventilation, has a near equal  performance in terms of the risk of identifying patients that develop subsequent ARDS, i.e. with full-face mask or go on to be intubated.  We also need to have an increased focus on trying to better understand heterogeneity of treatment effect in pediatric ARDS.  And perhaps this can be explained with phenotypes, whether those be based upon pathobiological  profiles as Nader Yehia has highlighted here in some of his work this last year, or the work that Mary Dahmer has done from the Bali cohort, looking at a combination of biomarkers  and clinical variables to really understand phenotypes that might be more or less likely to benefit from therapies.  There are some important new concepts that we probably need to be thinking about with respect to ventilator management, specifically in relation to mechanical power and driving  pressure. These were concepts that were not really discussed in the initial PALIC recommendation, and certainly  there have been some studies in pediatrics that have looked at these, at least observational studies over the last several years.  And there have been some updates to thoughts about recruitment maneuvers and tidal volume,  as well as PEEP management, based on observational studies that have emerged in pediatrics. Pulmonary ancillary therapies are particularly relevant for this concept of heterogeneity  in treatment effect. There certainly do appear to be subsets of patients that are likely to benefit from some of these therapies.  So while they cannot be recommended universally for all patients with ARDS because of lack of evidence, there certainly are some subpopulations that might benefit.  And these are important concepts that we have to consider in thinking about applying these therapies to our patients.  In addition, there have been many new studies focused on the importance of understanding non-pulmonary treatments as they may relate to patients with ARDS, i.e., concepts related  to delirium, early mobilization, fluid management, and transfusion strategies, really building  upon work that's recently been done, for example, with the TAXI collaborative to rethink  how we should approach transfusing our pediatric ARDS patients and our general critically ill patients.  Key new concepts in the area of respiratory monitoring have emerged, particularly with respect to concepts of patient self-inflicted lung injury, i.e., that the patient might  be an important factor as we think about the risk of progression of lung injury. So coming up with methods to estimate patient respiratory effort and calculate the overall  mechanical power or driving pressure within the system might be important for us to think about moving forward to understand patients' risk for lung injury.  There are many new topics and concepts that are important in the area of non-invasive ventilation with the real increase in the use of high-flow nasal cannula, for example,  and how this compares to non-invasive ventilation and or CPAP, as well as the potential influence  of other factors like synchrony and availability of these resources in resource-limited settings.  In the area of extracorporeal support, there are many new technologies that are now available and more and more observational data that's emerged about ventilation strategies while  on ECMO, as well as how to achieve and how fast to achieve oxygenation and CO2 targets for pediatric patients with ARDS.  For post-ICU outcomes, there are many new domains that we should likely be considering to really describe the morbidity of pediatric ARDS patients after they've left the hospital  with new measures for post-ICU status and domains of focus, with an understanding that many of the factors that may influence this post-ICU morbidity may not be immediately  modifiable by our actions in the intensive care unit. There is a new section that's focused on informatics, which is trying to understand how we can improve  the diagnosis and recognition of pediatric ARDS, as well as improve compliance with, for example, lung protective recommendations for ventilator support using electronic screening  tools, EMR integration, as well as computerized decision support. And finally, there's a section that's really focused on how we need to modify both our  definition and therapy considerations based upon resource availability, i.e., how our definition performs in low-resource settings and additional therapeutic options in some  of these settings where the capabilities may not be the same. So I'll conclude by just saying that there have been a wealth of new studies in pediatric  ARDS in the last five to seven years since publication of the original PALIC, and many important lessons learned from both observational studies and randomized control trials.  And that has shaped updated clinical practice guidelines and areas for research that will be forthcoming as part of PALIC II, expected to be published this summer.  Thank you very much.

Video Summary

The speaker discusses the lessons learned from clinical studies in pediatric ARDS (acute respiratory distress syndrome). The Pediatric Acute Lung Injury Consensus Conference (PALICC) created a pediatric-specific definition for ARDS and made clinical recommendations related to its management. The speaker highlights the need for a pediatric-specific definition due to the differences between adult and pediatric ARDS. The PALICC Definition was based on consensus opinion and some older studies, but its performance in practice was unclear. The PARTY study attempted to understand the implications of the PALICC Definition and found that it identified patients with ARDS earlier than the Berlin Definition. Other lessons learned include the limited discrimination of mortality risk between mild and moderate ARDS, the use of non-invasive ventilation without increasing the risk of harm, the importance of tidal volume and PEEP management, the variation in ancillary therapies usage, the need to consider post-ICU outcomes, and the importance of considering resource availability in defining and treating pediatric ARDS. The speaker finishes by highlighting upcoming updates to the PALICC recommendations.

Asset Subtitle

Pulmonary, Pediatrics, 2022

Asset Caption

Pediatric acute respiratory distress syndrome (PARDS) is a manifestation of severe lung injury with a mortality rate of up to 35%. Despite this significance of PARDS in critically ill mechanically ventilated children, respiratory management remains largely supportive, with no data to support one approach over another. This session will highlight advances in the understanding of PARDS by relating experimental models to clinical experience and discussing recent advances in respiratory monitoring and management of PARDS and the identification of biomarkers that can be used for disease stratification and prognostication. Speakers will also discuss how to design clinical trials of PARDS, delivering much-needed scientific knowledge

Meta Tag

Content Type Presentation

Knowledge Area Pulmonary

Knowledge Area Pediatrics

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Acute Respiratory Distress Syndrome ARDS

Tag Pediatrics

Year 2022

Keywords

pediatric ARDS

PALICC Definition

PARTY study

mortality risk

non-invasive ventilation

resource availability

	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.