SCCM Resource Library-PARDS in Children With Lower Respiratory Tract Infection: A BACON (Bronchiolitis and Co-DetectioN) Sub-Study

Video Transcription

Hello, everyone, and thank you for taking the time to listen to our presentation on PARDS in children with lower respiratory tract infections.  My name is Ben White, and I'm an assistant professor of pediatric critical care medicine at the Penn State Health Children's Hospital. I have no conflicts of interest to disclose.  If you have ever worked in the pediatric wards or PICU, then you are very familiar with the yearly surge of children admitted with respiratory illnesses.  Bronchiolitis represents about 17% of all pediatric hospitalizations, and approximately 5% of these patients require mechanical ventilation.  Data over the last two decades has shown that the incidence of lower respiratory tract infections is increasing. Pediatric ARDS, or PARDS, may also be common in this population.  PARDS is present in 6% of all mechanically ventilated children, as found by the PARDY study, and a recent single center study of patients with bronchiolitis showed that this  rate was 27%. Even though this population has been well-described in an epidemiological sense, the clinical  management of bronchiolitis and lower respiratory tract infections, especially in the PICU, has been understudied.  To address that, a group of investigators led by Todd Carseys, Steve Schein, and Steve Pahn, as part of the Polisi Research Collaborative, created the Bronchiolitis and Codetection,  or BACON, study. What started off as a pilot study quickly grew to a large, international-perspective  observational study of mechanically ventilated children with lower respiratory tract infections. The data collection for the study spanned from December 2019 through November 2020,  and 44 sites had completed data reporting at the time of this presentation. Specifically, the objectives of the BACON study were to determine the prevalence of  codetection of bacterial pathogens in patients mechanically ventilated for lower respiratory tract infections, and to also determine the impact of the presence of those pathogens.  Here we see a map of countries where data was gathered to demonstrate the wide reach of this project and also highlight all the generous intensivists and investigators who  donated their time and energy to the effort. Transitioning from the main BACON study, our study is a secondary analysis of the BACON  cohort where we look to describe those subjects who were found to have PARDS on study day one using the PALIC-OI or OSI guidelines and comparing their characteristics and outcomes  to those patients without PARDS. Here you can see the data collected during the study, and of note, all participating  centers attested that their standard of care was to obtain a respiratory culture at the time of intubation for children with suspected lower respiratory tract infection.  Importantly for this presentation, also collected were the first three days of ventilator settings, oxygenation indices, as well as supplemental therapies.  The BACON study included patients less than two years of age, diagnosed with a lower respiratory tract infection, and who were intubated for acute respiratory failure.  Patients were excluded for post-conceptual age less than 38 weeks, presence of only URI symptoms, isolated apnea, intubation only for surgery, intubation greater than 24 hours  from time of admission to the study center, cardiac arrest, presence of DNR order, or prior BACON enrollment. Some clarifying points on our methodology.  Lower respiratory tract infection type was determined by the blood site provider. This design was chosen as there is a wide variation in our field and naming conventions,  and terms such as pneumonia and bronchiolitis are often used differently or interchangeably depending on location.  Bacterial co-detection was defined as any respiratory culture that had both moderate or greater PMNs on gram stain and moderate or greater bacterial growth.  The statistical analyses that follow used either Chi-squared, Kruskal-Wallis, or Wilcoxon Rank Sum tests for comparisons.  556 subjects were enrolled in the study, and we found that 41% met PARDS criteria on the first full calendar day of mechanical ventilation.  This diagnosis was made using OI in 56% of the cohort, and OSI in the remaining 44%. Stratified by severity, 51% had mild disease, 36% moderate, and 12% had severe PARDS.  Here we see the basic demographics of our cohort, with the finding that those with PARDS tended to be older, with a median age of 5 months versus 3 months for the non-PARDS group.  There was no difference between the groups on the day of illness at hospital admission. Next, we can see the comparison of selected comorbidities between the two groups, though  most notably, the lack of comorbidities was equally common between them, representing about half of each cohort.  Presence of chronic lung disease was not associated with a statistically increased frequency of PARDS, nor was the rate increased in children born prematurely.  In contrast, major developmental delay was significantly more common in the PARDS group. There was no difference in the use of home medications between groups, though there was  a trend towards increased inhaled steroid use in the PARDS group. Family history of asthma or ATP also was not found to be more prevalent in those with PARDS,  but there was an association between living in a home with a smoker and PARDS. The differences in admission diagnosis are seen here.  There is no difference in the rates of bronchiolitis or lower respiratory tract infection, but  there was a significantly greater percentage of those carrying a diagnosis of pneumonia in the PARDS group, 38% versus 20%.  Of note, these diagnoses were subjective and were not exclusive of each other. For example, 44 patients carried the diagnosis of both bronchiolitis and pneumonia.  Bacterial positivity was more common in the PARDS cohort, but no individual virus was associated with a higher rate of PARDS.  There was also no difference in the rate of positive blood, urine, or CSF cultures between the two groups.  Here we see the respiratory culture results and the rates of the composite outcome of bacterial co-detection, which to remind you, was defined as both moderate or greater PMNNs  on gram stain and moderate or greater bacterial growth. PARDS was associated with a higher rate of moderate or greater PMNs, and notably with bacterial co-detection.  30% of those in the PARDS group met that criteria, compared to 21% in the non-PARDS cohort.  When stratified by PARDS severity, there was no association between bacterial co-detection or its components with increased severity, though there was a significant decrease in  culture positivity with increasing severity. Here you can see the use of adjunctive therapies on day one between the two groups.  Use of these therapies was generally more common in the PARDS cohort, including one-third of that group utilizing a vasopressor, and 25% with initiation of steroids.  The ventilator parameters for the cohort showed expected differences between the groups. Median PEEP in the PARDS cohort was 8 compared to 7 in the non-PARDS group, and MAP was 14  compared to 12. Similarly, the PARDS group also experienced higher FiO2 and delta pressure. There were no differences in the median tidal volume between the groups.  When stratified by severity, these relationships mostly continued to hold, with PEEP, MAP, and FiO2 all increasing with increased PARDS severity.  The lack of a difference in tidal volume strategies between the PARDS and non-PARDS groups is  notable, and further analysis shows that the proportion of subjects in different groupings of tidal volumes were equivalent between the PARDS and non-PARDS group, and between  the different levels of PARDS severity, suggesting that the PALIC low tidal volume recommendations were not widely implemented.  Our main outcomes also follow the trend of a PARDS diagnosis being associated with a more severe disease process, here showing that mortality is over two times more likely  in the PARDS cohort. PARDS was also associated with a significantly longer duration in mechanical ventilation,  ICU length of stay, and hospital length of stay, accounting for a median of one and a half more days intubated, three more days in the ICU, and five more days in the hospital.  When analyzed by those with and without comorbidities, we see that those with comorbidities had longer  ICU and hospital lengths of stay, and that there was a trend towards increased mortality, though it did not reach significance.  Again, stratifying by PARDS severity, we see that PARDS mortality is highly centered in those with moderate and severe disease, with mortality less than one percent in subjects  with mild PARDS. Increasing severity of illness is also associated with longer mechanical ventilation and ICU  length of stay, with these measures increasing between groups, as would be expected. To conclude, our secondary analysis of the Bacon Study cohort shows that PARDS is common  in children intubated for lower respiratory tract infections, being present in 41 percent on day one in mechanical ventilation following intubation.  Additionally, the diagnosis of PARDS is associated with more severe disease, including increased mortality, duration of mechanical ventilation, and length of stay.  These results are consistent with previous studies of PARDS in the general pediatric population, though the 8 percent mortality rate here is much lower than the 17 percent  rate for the general pediatric cohort reported in the PARDS study. We've also found that there is a signal to suggest that PARDS in this population may  be related to secondary bacterial infections, with both the diagnosis of pneumonia and the finding of bacterial co-detection more common in the PARDS cohort.  Future prospective investigations will be needed to determine the significance of these relationships, and whether they represent modifiable effects that interventions such  as lung protective ventilation or early antibiotic administration can affect. Thank you everyone for taking the time to listen to our presentation, and a huge thank  you to all the Bacon investigators who made this research possible.

Video Summary

The speaker discusses the prevalence and impact of Pediatric Acute Respiratory Distress Syndrome (PARDS) in children with lower respiratory tract infections. They present data from the Bronchiolitis and Codetection (BACON) study, which collected data from over 40 sites internationally. The study found that 41% of children with lower respiratory tract infections met PARDS criteria on the first day of mechanical ventilation. PARDS was associated with more severe disease, including increased mortality, longer durations of mechanical ventilation, ICU stay, and hospital stay. There was also a suggestion that PARDS may be related to secondary bacterial infections. Further research is needed to explore these relationships and potential interventions.

Asset Subtitle

Pulmonary, Pediatrics, 2022

Asset Caption

INTRODUCTION: Bronchiolitis and other lower respiratory tract infections (LRTI) are the most common causes of pediatric respiratory failure. Because there are insufficient data characterizing Pediatric ARDS (PARDS) in young children with LRTI to inform clinical management, we aimed to describe their clinical and demographic characteristics and identify factors associated with PARDS development. METHODS: We performed a sub-analysis of data from Bronchiolitis And CO-detectioN (BACON), an IRB-approved international prospective observational study of critical bronchiolitis at 43 sites from 12/2019 to 11/2020. Inclusion criteria were age < 2 years, endotracheal intubation, and acute LRTI. We used Chi squared and Wilcoxon rank sum to compare PARDS patients (meeting PALICC OI/OSI criteria during the first full calendar day following intubation) to non-PARDS patients. Bacterial co-detection was defined as a respiratory culture with both moderate/many PMNs and growth of moderate/many bacteria. Data are shown as n (%) and median (IQR). RESULTS: Complete data were available for 438 children from 41 sites. PARDS was diagnosed in 290 (66%) subjects and was associated with increased mortality (7.3% vs 1.4% [p=0.009]), greater duration of mechanical ventilation (166 [110.8, 261] vs 116 [72.6, 181.6] hours [p < 0.0001]), and PICU length of stay (11.1 [7.6, 16] vs 7.5 [5, 11.6] days, [p < 0.0001]). PARDS was associated with older age (5 [1.6, 11.1] vs 2.4 [1.3, 7.9] months [p=0.005]) and history of previous admission with wheezing (13% vs 6.1% [p=0.025]). PARDS was not associated with comorbidities, home medications (bronchodilators, steroids, or diuretics), or family history of atopy. Children with PARDS were less likely to have a positive respiratory culture (43 vs 54% [p=0.030]), but had no difference in meeting bacterial co-detection criteria (31% vs 26% [p=0.286]). PARDS subjects were treated with higher PEEP (8 [6, 9] vs 7 [6, 8] [p < 0.0001]) and mean airway pressures (14 [12, 16] vs 12 [10, 13] [p < 0.0001]). CONCLUSIONS: PARDS is common in patients intubated for LRTI and is associated with increased mortality, duration of mechanical ventilation, and PICU length of stay. Many factors were associated with the development of PARDS and prospective studies are needed to elucidate optimal management of critical bronchiolitis.

Meta Tag

Content Type Presentation

Knowledge Area Pulmonary

Knowledge Area Pediatrics

Knowledge Level Advanced

Membership Level Select

Tag Acute Respiratory Distress Syndrome ARDS

Tag Pediatrics

Tag Ventilation

Tag Respiratory Syncytial Virus RSV

Year 2022

Keywords

Pediatric Acute Respiratory Distress Syndrome

PARDS

lower respiratory tract infections

BACON study

mechanical ventilation

	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.