Current Concepts in Pediatric Critical Care-7: Pediatric Trauma

7: Pediatric Trauma - Current Best Evidence (Hot Topic)

Video Transcription

Hello and welcome. My name is Chris Watson, and I'm a pediatric intensivist on faculty at the Medical College of Georgia. I'm excited to have the opportunity to share with you today  a snapshot of some of the current best evidence on the topic of pediatric trauma. My presentation is not meant to be a comprehensive review of all new evidence in this field,  but instead is a smaller selection of topics that I hope are of interest and inspire your further exploration. To help you along the way,  you'll find QR codes on key slides throughout my talk that you can scan to link you to the articles being discussed. First, as far as disclosures,  I have no actual or potential conflicts of interest in relationship to this presentation. This talk has five objectives. First, to outline the impacts of the COVID-19 pandemic on  pediatric trauma epidemiologic trends. Second, to discuss firearm injury trends and resource utilization in the pediatric ICU. Third, to illustrate the patterns of association  between trauma and pediatric acute respiratory distress syndrome or PARDS. Fourth, to describe the role of extracorporeal life support in trauma.  Finally, to discuss functional outcomes and modifiers after PICU admission for trauma. Let's begin here with the case for us to consider.  We have a five-year-old male presenting following a motor vehicle collision brought in by EMS to the emergency department. He was an unrestrained passenger  and has no other known history. On exam, his Glasgow Coma Scale or GCS is currently nine. Imaging of the head reveals multi-compartmental intracranial hemorrhages,  and no other serious injuries are noted. At the macroscopic level, as we're still meeting virtually even now in 2022, how has the sustained COVID-19 pandemic  affected the epidemiology of pediatric trauma?  An important place for us to begin answering this question is here by understanding what the major causes of death are for children in the United States.  The children represent roughly one-quarter of the US population by number. They account for only about two percent of the annual deaths.  This 2018 retrospective observational cohort study of the CDC wonder database describes the leading causes of death in children 1-19 years of age  categorized by mechanism and intent. In this cohort, representing more than 20,000 children and adolescents, 60 percent resulted from injury-related causes,  including six of the top 10 causes of death. Among injury-related deaths, unintentional injuries were most common, representing 57 percent.  Motor vehicle crashes were the leading cause of death and firearm-related injuries were the second leading cause of death. Finally, among intentional injuries,  suicide was slightly more common than homicide. For additional context, before discussing the effects of the pandemic, this infographic from the CDC depicts  key summary findings from the analysis of the National Vital Statistics System from 2010-2019. Now, overall, unintentional injury deaths decreased 11 percent during this period.  They remain the leading cause of death for children 1-19 years of age in the United States and the third leading cause of death for infants less than one.  Again, motor vehicle crashes cause more deaths than other causes of unintentional injury. Drowning was the leading cause of injury death for children ages 1-4.  Additionally, drowning death rates were higher among black children 5-14 years when compared with white children of the same age.  Geographically, overall unintentional injury death rates in rural areas were higher than metro and urban areas. Unfortunately, despite overall decreases  in child unintentional injury death rates, racial and ethnic disparities persist and rates actually increased among some groups specifically.  Suffocation death rates increased 20 percent among infants overall and 21 percent among black children. Motor vehicle death rates among black children increased by  nine percent and poisoning death rates increased by 50 percent among Hispanic and 37 percent among black children. This added clarity and who unintentional injury death affects  offers opportunities for us for more focused strategies to protect those who are at disproportionate risk. Illness, school closures, social distancing,  sick family members, and financial stressors may have had a negative impact on the emotional and physical well-being of children and their parents,  but the overall impact of the pandemic on pediatric trauma remains unclear. This retrospective cross-sectional study of children less than  18 years evaluated in 15 trauma centers from 2019 to 2020 in LA County describes the pandemic impact on regional trauma epidemiology.  As compared to 2019 as the pre-pandemic timeframe, several trends were noted in 2020 reflective of the initial lockdown.  More trauma activations occurred in non-Hispanic black children. The rates of injury from MVCs and higher speed MVCs increased.  The number of children who sustained injury from firearms and burns both increased and sports related injuries and assault both decreased.  Notably though, overall trauma activations in 2020 did not significantly differ during the COVID-19 pandemic from the pre-pandemic timeframe in this study.  Shown in this graph is the rate of pediatric trauma activations with the solid line and the rate of COVID cases with the dashed line depicting this overall trend.  With regards to outcomes, several key findings were noted though. Median injury severity score, ISS and GCS were not different between the two time periods,  nor was the proportion admitted to the PICU. Intubations actually declined from 12-6 percent and mortality from 3-2 percent. But ED, hospital, and ICU lengths of  stay remained similar between pre-pandemic and pandemic timeframes.  The second study is a retrospective observational cohort study using the virtual pediatric systems or VPS database of more than 160,000 children at  77 sites comparing the first two quarters of the two years prior to the pandemic to those of 2020. Overall, the investigators found that admissions decreased by  32 percent in quarter 2 of 2020 as compared to the pre-COVID-19 baseline with the largest decrease being in respiratory conditions.  The frequency of brain trauma and general trauma both increased and attempted suicides decreased. However, the raw patient volumes actually decreased in all three diagnoses.  Finally, severity of illness and raw mortality rates for both traumatic and non-traumatic illness were approximately 15 percent higher in quarter 2 of 2020.  The authors concluded that this may have represented hesitation in seeking healthcare as well as healthcare system strain, but causality remains unclear.  I believe it's also important for us to talk about the effect of the pandemic on child maltreatment as well. There has been open question and concern that during the pandemic,  whether or not a silent or secondary pandemic of child maltreatment was also occurring. First, as background, child maltreatment affects  nine in every 1,000 children in the United States. In 2019, there were 656,000 reports of child maltreatment, of which 61 percent were for neglect, 10 percent for physical abuse,  7 percent for sexual abuse, and 16 percent represented multiple types. Tragically, 1,840 deaths were attributed to maltreatment.  Children in the first year of life are more often victims, as are girls, more often than boys. If we look at racial and ethnic trends,  American Indian or Alaskan Native children have the highest rate of victimization, and African-American children have the second highest rate.  Risk factors for maltreatment can be described as intrinsic to the child, such as age or special needs, the abuser, such as substance abuse, mental health issues, or low education,  or society at large reflecting factors such as parental perception of lack of community support, job loss, life stressors, and poverty. Digging in deeper on the question of  child maltreatment during the pandemic, this recent study is a nine-center retrospective review of more than 300 suspected child physical abuse admissions in  2020 as compared to 2019 that included 69 cases during the pandemic. There was no difference in terms of age, gender, ethnicity, race,  insurance, morality, GCS, ISS, or pre-hospital arrest. Investigators found that there was actually no difference in the number of cases,  injury type, severity, hospital, or ICU length of stay, procedures, disposition, or mortality. These two additional studies help provide some further insight, I believe.  The first is a retrospective cross-sectional study of more than 20,000 child physical abuse cases at 47 hospitals, comparing 2020 to the prior three years.  The investigators found that child physical abuse cases actually decreased significantly in the pandemic period. However, those that presented were more likely to be severe,  requiring ICU admission or having traumatic brain injury in those less than five years. The second study here is a systematic review of child maltreatment during the  pandemic that identified 11 articles. Five articles found child maltreatment increased, six found that it decreased, and one  study found no difference in rates. Based on these mixed findings, the authors of this review conclude that the possibility exists that a silent  pandemic of abuse still occurred. This potential may have been exacerbated by limited exposure to mandatory reporters, health care avoidance, and increased caregiver stressors.  So what do we know about the impact of the pandemic on trauma epidemiology? First, local pediatric trauma surges are likely multifactorial  and attributable to the interaction of both traditional risk factors, such as age, gender, and family environment, as well as pandemic-related factors such  as social distancing, school closures, stay-at-home orders, other public health measures, and even COVID fatigue. Second, interpretation of these trends is  certainly confounded by inequities in COVID-19 impact by race, competing factors of race in socioeconomic systems that disadvantage minorities, as well as recognized racial bias in  child maltreatment reporting. Though ongoing regional differences in lockdown restrictions and risk of new novel coronavirus variants may further  drive pediatric trends, these patterns can potentially help to inform pre-hospital and hospital preparedness and planning.  So at this juncture, I'd actually like to return to the first study we looked at describing the major causes of death in children to highlight some extremely  concerning trends regarding firearm-related injuries in children. By cause, firearm-related injuries were the second leading cause of death  responsible for 15 percent of deaths in children less than 19 years in 2016. Of these, 59 percent were homicides, 35 percent were suicides, and four percent  were unintentional. This figure shows the 10 leading causes of mortality from 1999 through 2016. The 38 percent decline in MVC-related  deaths is likely multifactorial and attributable to improved safety standards and widespread adoption of seat belts and child safety seats,  though the later increase from 2013 on may reflect an increasing trend in distracted driving. Looking at firearm-related deaths alone,  the overall mortality rate is 3.5 per 100,000 children and there's been a 28 percent relative increase between 2013 and 2016,  which reflects increases in rates of firearm homicide and firearm suicide during this period. Now for global context, the rate of  firearm-related deaths among children and adolescents in the United States was nearly 37 times higher than the overall rate observed in 12 other high-income countries.  So now let's specifically look at firearm-related injuries in the pediatric ICU. This recent retrospective review of  children 1 month up to 18 years of age used the VPS database to look at firearm-related injuries from 2009 through 2017.  Of more than 1,400 cases identified, 12 percent died in the PICU, which is five times higher than the overall all-cause pediatric mortality rate.  Males represented more than three quarters of the injuries and unintentional injury and assault comprised 90 percent of the cases.  Among children attempting suicide with a firearm, mortality was higher than other causes of firearm injury.  Additionally, though Black children represent about 14 percent of the U.S. child population, they suffered 45 percent of critical firearm injuries and 31 percent of firearm deaths.  White children had the highest rate of suicide and, as a subgroup, had a higher mortality, though this is disproportionately low  based on population representation. This study also helps draw attention to the issue of firearm injuries in children.  It's a retrospective review of 6 million admissions to 28 U.S. children's hospitals in the Pediatric Health Information Systems database from 2004  to 2017. The investigators found that roughly 3,700 admissions were firearm injuries. Most patients were 10 to 19 years old,  male, and Black. Regionally, there was variation with the South overtaking the Midwest in 2016. As is shown in the figure here,  unintentional injuries predominated up until 14 years of age and then assault leads for 15 to 19. Of the 3,700 injuries, 30 percent  required PICU admission. Continuing to look at the same study, the authors found that in this cohort, the median PICU length of stay was two  days, which remained unchanged during the study, though median cost of admission increased. Across the study period, there was also an increased use of mechanical ventilation,  surgical procedures, vasopressors, blood transfusions, ICP monitoring, and CT imaging. ICU mortality was 13 percent, again higher than the anticipated all-cause PICU mortality.  These recent studies are essential in helping us to better describe an ongoing epidemic of firearm-related injury among children and adolescents.  These quotes from an accompanying editorial poignantly highlight this, and the full editorial is worth the time to read.  The authors state, when we searched the archives of pediatric critical care medicine for past publications on firearm violence,  the best matches for firearm and gun were forearm and gut, but this is changing. These data again highlight violence and  firearm injury as a health disparity, reflecting systemic racism and structural inequality experienced by Black youth. And pediatric intensivists can intervene on multiple levels,  individual, hospital, community, state, and national, to advance pediatric critical care firearm injury prevention research and advocacy.  So now let's transition our focus a bit and return back to our initial case. We have the same five-year-old male with moderate TBI and a GCS of 9,  presenting following an MBC without other serious injuries. Imaging in the ED revealed no focal abnormality on chest X-ray, but CT of  the head revealed multi-compartmental intracranial hemorrhage. Six hours after presentation, now in the PICU, he is intubated for a declining neurological exam.  Repeat CT of the head at that time shows progressive cerebral edema with  stable hemorrhages. An EVD is placed at the bedside, however his intracranial pressure, or ICP, progressively remains elevated despite  first-tier TBI therapies. Additionally, progressive hypoxia develops and a repeat chest X-ray shows diffuse bilateral alveolar infiltrates.  So first, are there risk factors that predict neurological deterioration in moderate TBI? And second, what key pathophysiologic arrangements help to best describe this new hypoxia?  Certainly, early risk stratification and identification of children who warrant further evaluation and more aggressive management,  as in our case, seem likely to help us to decrease morbidity and potentially mortality. But who is at risk for decline in moderate TBI? In this study, the authors conducted a  single institution retrospective cohort analysis to describe the characteristics of children with moderate TBI, defined as a GCS of 9 to 13, and identify factors associated with  deterioration to severe TBI with the GCS less than or equal to 8.  From 2010 to 2017, 177 patients were identified, with 37 patients, or 21 percent, experiencing deterioration in the first 48 hours.  Children who deteriorated had a 47 percent increase in intubations and were more likely to have generalized edema, subdural hematoma, or a contusion on  CT skin. They were also more likely to receive hypertonic saline, undergo intracranial pressure monitoring, be transferred to inpatient rehab  following discharge, and incur greater costs of care. Factors associated with deterioration on multivariable regression analysis  included higher injury severity score, higher INR, and a lower GCS of 9 to 10 on presentation.  In our case, not only has our patient experienced neurological decline following TBI, but we suspect he is also developing  trauma-related pediatric acute respiratory distress syndrome, or PARDS. So what do we know about PARDS and trauma? This retrospective cohort of  children less than 18 years admitted to the PICU with trauma from 2009 to 2017 identified nearly 2,500 pediatric trauma admissions, with approximately four percent meeting  complete PALICC PARDS criteria within seven days post-injury. Notably, among those with PARDS, mortality was 34 percent versus two  percent without. Also, as is shown in the bar graph here, mortality was 50 percent for severe PARDS at onset, 33 percent for moderate, and 31 percent for mild.  Among PARDS patients in this study, though the cause of death was primarily neurologic, death from multi-organ failure was more  common than among those without PARDS, perhaps suggesting a role of refractory hypoxia. Overall, as is shown in this graph,  PARDS patients had a longer duration of mechanical ventilation, longer ICU length of stay, and longer hospital length of stay than patients  without PARDS. Finally, among survivors, 77 percent of PARDS patients had functional disability at discharge versus 31 percent of patients without PARDS.  Similarly, this 10-year retrospective review of approximately 7,400 trauma patients highlights the morbidity and mortality of trauma-associated PARDS.  Of 646 patients meeting inclusion criteria, nine percent had PARDS. Patients with TBI, non-accidental trauma, and higher injury severity were more likely to have  trauma-related PARDS. Finally, new or progressive multiple organ dysfunction syndrome was eight times higher, and mortality  was five times higher in trauma-related PARDS patients than those without trauma-related PARDS. So now, back to our case again. What  would you do? Given higher first 24-hour resuscitation volumes and altered inflammatory profiles in TBI patients, in TBI patients with trauma-related PARDS, how should we weigh  neuroprotection versus lung protection physiologic targets and management strategies? Should PARDS management be tailored to  allow for a period of neuroprotection immediately following TBI? At this point, we really just don't know. Further research is necessary to help  guide us in this common clinical scenario. So let's shift our discussion a bit once more by considering this case.  A three-year-old male presents after house fire entrapment with approximately 30 percent total body surface area, or TBSA, burns of the face,  chest, and upper extremities with evidence of inhalational injury. Following intubation in the ED, he subsequently develops PARDS, refractory  to conventional medical management and maximal ventilatory support. So as we pause to consider this case, what do we know about ECMO outcomes for  trauma, and specifically for burn patients? A prudent place for us to begin to answer this question, I believe, is here,  looking at the general characteristics of burn-injured children in the U.S. The authors of this study conducted a retrospective review of burn-injured  children from 2009 through 2017, and 117 U.S. PICUs using the VPS database. Of more than 2,000 patients, most were  male and less than six years of age. Almost all had cutaneous burns. Nearly six percent had oral and pharyngeal injury,  and five percent had inhalational injury. Flame burns were most common in children older than two years, and scald injury was most common in  those less than two years. Mortality occurred in five percent and was associated with five-fold increased risk in patients with greater than 30  percent TBSA burns. Additionally, most patients were cared for at facilities that were not ABA pediatric verified.  Finally, of those with data reported, 22 percent had a new designation of moderate or severe disability or persistent vegetative state. So from  this, we can clearly see that burn-injured children have a substantial burden of organ failure, morbidity, and mortality. Now if we think back to our case, what's  the role of extracorporeal life support in the pediatric trauma patient? This recent retrospective review of the ELSO registry from 1989 to 2018  identified 573 trauma patients with a median age of almost five years. The majority of patients were male and placed on veno-arterial support.  Drowning was the most common mechanism, followed by burns and thoracic trauma. Complications, including cardiovascular, mechanical, and hemorrhagic were high,  but actually similar to historical cohorts. Overall survival was 55 percent and was even higher for patients on veno-venous ECMO compared to those on veno-arterial ECMO,  even when controlling for mechanism. So from this study, we can see that the indications, characteristics, and outcomes of children placed on ECMO with trauma  differ significantly from that previously reported for adult populations, and that at least in children, trauma may  not be a contraindication for ECMO. What if we just look at ECMO in burn injured children alone? This recent retrospective review, also  of the ELSO registry database from 1990 to 2016, identified 113 pediatric burn patients who were predominantly male with a median age of approximately two years.  In fact, only 11 children were older than 11 years. Approximately two-thirds of the patients were supported on VA ECMO and one-third  on VV ECMO. Overall cohort survival was 52 percent, and mortality for respiratory failure was actually similar to that previously reported for all pediatric respiratory indications.  However, survival was lower for cardiac support and eCPR than that reported for all pediatric patients requiring ECMO for these indications. Cardiac arrest prior to cannulation was  associated with increased mortality, though inhalational injury was not. Of this small cohort, overall survival was 70 percent when  less than 60 percent TBSA was involved with improved survival rates associated with lower percent TBSA involvement. So certainly  further research is necessary, but it seems likely that there is a subset of pediatric trauma and specifically burn injured patients  who may be optimal candidates for more aggressive use of extracorporeal technologies. It's important for us to also consider functional outcomes following trauma in children.  With declining mortality, metrics that can meaningfully be used to assess morbidity which are rapid, reliable, and easy to use are necessary.  One promising metric is the Functional Status Score, or FSS, which was assessed in this study of more than 550 children  with at least one injury surviving to discharge from the Collaborative Pediatric Critical Care Research Network. Summary statistics of this cohort are shown in the table here.  Most were male with a median age of almost six years. Most were at least moderately injured with an ISS greater than nine. The median length of stay in the ICU was 1.5 days  and three days in the hospital overall. The functional status based on the FSS, which assesses mental status, sensory function, communication,  motor function, beating, and respiratory status for most patients was normal before being injured. In this cohort based on the FSS,  new domain and overall morbidity were observed in 17 and 11 percent of patients respectively.  Building on this evidence base, this prospective observational study at seven level one pediatric trauma centers looked at long-term outcomes of children  treated for at least one serious injury. Among 323 injured children with complete discharge and follow-up assessments,  one quarter had functional impairment at discharge, and at six months the FSS was abnormal in 10 percent. Factors associated with six-month FSS  impairment included increased impairment at discharge, older age, penetrating injury type, severe head injury,  and spine injuries. Factors associated with worse health-related quality of life included older age and higher FSS impairment at discharge.  So, the most seriously injured children return to normal. A minority suffer ongoing disability and reduced health-related quality of life beyond discharge.  Based on what we know about outcomes from pediatric trauma, perhaps future efforts to minimize morbidity with increased survivorship may focus on  decreasing PICS or the post-intensive care syndrome. PICS is a group of cognitive, physical, and mental health impairments that commonly occur in patients after ICU discharge.  PICS in children also integrates baseline status and social health and has variable recovery patterns that may demonstrate improvement, decline,  vacillation, or a static state following ICU stay. Based on our discussion today and using the PICS framework,  one key future challenge is how we can better address the long-term impacts of critical injury on children and families. One new set of guidelines will certainly  help us in this goal.  Building on the SCCM ICU liberation campaign, the recent 2022 SCCM clinical practice guideline on prevention and management of pain, agitation, neuromuscular blockade, and  delirium in critically ill pediatric patients with considerations of the ICU environment and early mobility, otherwise known as PEDS PANDEM, should be  studied further in critically injured children and integrated into clinical care pathways for this population specifically.  So what are the key take-home points today? First, as was the focus of a wonderful recent special article on pediatric critical care medicine,  we should consciously take an anti-racist approach to pediatric research, inclusive of trauma research, to identify and reduce health care  disparities in marginalized populations. Second, pediatric trauma trends during the COVID-19 pandemic have varied by region and likely are attributable to  multiple pandemic and non-pandemic related factors. Additionally, the impact on child maltreatment is unclear, but concerning  for possible under-recognition. Third, pediatric firearm injury represents a major public health crisis with significant associated health care  cost, morbidity, and mortality. Further research and advocacy is critically needed here. Fourth, trauma-related PARDS has a  significant impact on mortality. ECMO remains a viable option for trauma patients with recoverable injury, inclusive of burn-injured children.  Finally, improving identification of trauma-associated functional morbidity is an important aspect of recognizing, treating, and hopefully reducing  pediatric PICS. Clinical care pathways should integrate the recent PEDS PANDEM guidelines, though additional research in the critically injured child remains necessary.  So as we've looked at some recent highlights in the pediatric trauma literature, I hope that my presentation may have  challenged you to ask your own questions and explore a bit further. Again, my name is Chris Watson and thank you.

Video Summary

Summary: The video presents a snapshot of the current best evidence on the topic of pediatric trauma. It covers various aspects, including the impact of the COVID-19 pandemic on pediatric trauma epidemiology, firearm injuries and resource utilization in the pediatric ICU, patterns of associations between trauma and pediatric acute respiratory distress syndrome (PARDS), the role of extracorporeal life support in trauma, and functional outcomes after PICU admission for trauma. The impacts of the COVID-19 pandemic on pediatric trauma epidemiology are varied and likely influenced by multiple factors. Pediatric firearm injuries represent a significant public health crisis, and further research and advocacy are needed in this area. Trauma-related PARDS is associated with increased mortality and longer hospital stays. ECMO can be a viable option for trauma patients, including burn-injured children. Identifying and reducing trauma-associated functional morbidity is important for recognizing, treating, and reducing pediatric post-intensive care syndrome (PICS).

Asset Caption

Christopher M. Watson, MD, MPH

	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.