Presentations

Follow

• 

  Lessons From the Pandemic: How a Children’s Hospital Responded to the Challenges of COVID-19

  John Lantos, Paul Kempinski, Laurie Ellison, Jennifer Watts, and Angela Myers

  The COVID-19 pandemic challenged doctors and hospital administrators as did no other event in our lifetimes. Leaders needed to develop a pandemic command structure with the agility to respond to rapidly evolving situations. They had to deal with drastic financial implications, develop new methods of delivering health care, and collaborate regionally. They learned the importance of communication with staff, policy makers, the local medical community, and the public. They had to allocate of scarce resources internally and externally, and balancing rational policy making against irrational fears. For children’s hospitals, some specific challenges included determining our role in a pandemic that predominately affected adults, doing research on the unique pediatric manifestations of disease, and dealing with questions about schools and daycare. In this workshop, leaders from hospital administration, infectious disease, and disaster preparedness will review and analyze some of our experiences and responses at a large quaternary care children’s hospital. We will use specific events and decisions to illustrate the unique challenges and our retrospective analysis of whether we could have done better. The goal of the interactive workshop is to learn together from our collective experience in order to be better prepared for future events.

  Learning Objectives:

  1. Describe the mechanisms that were put in place to respond to the emergent demands of the pandemic

  2. Analyze options that were available to decision makers regarding specific choices, the reasons for the choices that they made, and downstream implications of those choices

  3. Speculate about preparedness for the next phase of this pandemic, or the next major crisis.

  Presented at the 2021 PAS Virtual Conference

• 

  Medicaid Expenditures Among Children with Documented Obesity

  Kathyrn Kyler, Matt Hall, Jessica L. Bettenhausen, Sarah Hampl, and Ann M. Davis

  Background: Obesity rates continue to rise among children, but knowledge regarding spending patterns of Medicaid enrollees with documented obesity are lacking.
  
  Objective: We aimed to describe Medicaid expenditure patterns and determine the degree to which specific clinical characteristics and conditions contribute to high expenditures among children with obesity.
  
  Design/Methods: We performed a retrospective cross-sectional analysis of children aged 2-17 years with a diagnosis code (ICD-10) for obesity continuously enrolled in the nationally-representative 2017 Medicaid Marketscan database. Expenditures were measured as median per member per year (PMPY) spending and categorized based on prior literature from low to high PMPY expenditure groups: <80th%, 80-<95th%, 95-<99th%, and ≥99th%. Inpatient, outpatient, and pharmacy expenditures were analyzed. Covariates included demographic factors, common obesity co-morbid conditions (e.g., hypertension), number of complex chronic conditions (CCCs), and number of mental health conditions. Chi square tests were used to compare PMPY spending across expenditure groups and logistic regression analyses were used to measure demographic and clinical characteristics associations for patients in the high spending groups (≥95th%).
  
  Results: We identified 300,286 children with a diagnosis of obesity. Children aged 12-17 years, of non-Hispanic white race/ethnicity, with obesity comorbid conditions, at least 1 CCC, or mental health condition were most likely to be in the highest spending group. (Table 1). The highest overall median PMPY spending was from inpatient and outpatient therapy and treatment ($6,018 and $800, respectively). Mental health therapy and treatment drove the PMPY spending in the higher spending groups (≥99th% group $16,471) (Table 2). Characteristics found to be associated with being in the higher spending groups included: age 12-17 years, having an obesity comorbid condition, having ≥ 1 CCC and mental health condition, with these associations increasing considerably as the number of CCCs or mental health conditions increased (Table 3).
  
  Conclusion(s): Inpatient and outpatient mental health expenditures made up a large proportion of spending among Medicaid-enrolled children with obesity. Important drivers of cost in this population included having obesity comorbid conditions and mental health conditions. Future research is needed to determine if some of these costs are avoidable in children with obesity.
  

  Presented at the 2021 PAS Virtual Conference

• 

  Neonatal DNA methylation as a predictor of cognitive, language, and motor performance at 24 months adjusted age, among children born very preterm

  Stefan Graw, Marie Camerota, Brian S. Carter, Jennifer Helderman, Julie A. Hofheimer, Elizabeth C. McGowan, Charles R. Neal, Steven Pastyrnak, Lynne Smith, Michael O'Shea, Barry Lester, Carment Marsit, and Todd M. Everson

  Background: Infants born prematurely are at increased risk for neurodevelopmental impairments during childhood and early neurodevelopment can be influenced by multiple factors (perinatal and environmental exposures, genetics). DNA methylation (DNAm) can also be influenced by these factors and be reflective of health and development, and thus may provide an integrated measure of early life risk for neurodevelopmental outcomes.
  
  Objective: To test whether neonatal DNAm is predictive of cognitive, language, and/or motor performance at 24 months of age in children that were born very preterm.
  
  Design/Methods: We studied 433 neonates born < 30 weeks postmenstrual age in the Neonatal Neurobehavioral Outcomes in Very Preterm Infants (NOVI) study. Neonatal DNAm was measured from buccal swabs at NICU discharge via the Illumina MethylationEPIC BeadArray. Cognitive, motor, and language performance were assessed by the Bayley Scales of Infant and Toddler Development-III (BSID-III) at 24 months adjusted age, with mild and moderate impairment defined as composite scores < 85 and < 70. We used an elastic net regression and leave-one-out cross-validation to limit overfitting and selection bias, to identify sets of DNAm sites that were predictive of composite Bayley scores. For the leave-one-out cross-validation, a model was trained on all but one sample to make a prediction on that held-out sample. This step is repeated until each sample was held out and predicted once. In addition, for a given hold-out sample, all siblings were removed from its training set. The performance of predicted composite scores were evaluated by their correlation with the measured Bayley scores and the area under the receiver operating characteristic (ROC) curve (AUC) based on dichotomized Bayley scores (< 85 and < 70).
  
  Results: We identified sets of DNAm sites (including 33-306 sites) that predicted composite scores for the BSID-III at 24 months of age, which were positively (r>0.3) and significantly (p-value <1e-11) correlated with reported scores (Figure 1A-C). Additionally, neonatal DNAm at these sites was predictive of mild cognitive (AUC=0.68), language (AUC=0.71), and motor impairments (AUC=0.66) (Figure 1D-F), and yielded similar predictive statistics for moderate impairments.
  
  Conclusion(s): DNAm patterns measured at NICU discharge were indicative of cognitive, language, and motor performance at two years of age, among children born very preterm.

  Presented at the 2021 PAS Virtual Conference

• 

  Neonatal Gut Microbiota Alterations and Local Inflammation Induced by Escherichia coli Infection are Modified by Lactobacillus rhamnosus Prophylaxis

  Susana Chavez-Bueno, Hao Xuan, Shahid Umar, Concong Zhong, Wei Yu, and Venkatesh Sampath

  Background: E. coli is a leading cause of neonatal sepsis. Newborns ingest E. coli, which transcytoses the gut producing bacteremia. Lactobacillus spp. decrease E. coli gut transcytosis but the mechanisms involved in this protective effect are not well understood.
  
  Objective: To determine the effects of Lactobacillus pretreatment on the intestinal microbiota and inflammation in neonatal rats orally infected with E. coli.
  
  Design/Methods: Newborn rats were orally pretreated on day of life (DOL) 1 and 2 with four doses of 107 colony forming units (CFU) of Lactobacillus rhamnosus GG (LGG) or PBS. On DOL 2, pups received orally 106 CFU of neonatal E. coli bacteremia strain SCB34 or PBS. On DOL 7, distal colon with stool were collected for 16S sequencing. AbundantOTU+ was used to generate de novo operational taxonomic units (OTUs). Observed, ACE, Shannon, and Simpson indices were used for alpha diversity. Bray-Curtis and Jaccard indices were used for beta diversity. Linear discriminant analysis effect size (LEfSe) was used to determine differences at the genus level. Expression of ICAM-1, GRO1, Toll-like receptor 4 (TLR4), and Single-Immunoglobulin Interleukin-1 Related Receptor (SIGIRR) was measured in ileal homogenates by real-time PCR.
  
  Results: Phyla distribution clearly distinguished the SCB34-infected experimental groups (Fig. 1). We also observed greater alpha diversity in the SCB34-infected groups (Fig. 2; A-D). Bray-Curtis analyses showed differences in microbial abundance between the groups receiving LGG pretreatment vs. PBS prior to SCB34 infection. However, the microbial composition of these two groups was similar per Jaccard index (Fig. 2; E-F). LEfSe results showed greater abundance of Lactobacillus in both SCB34-infected groups, and surprisingly, lower abundance of Escherichia-Shigella genera in these groups compared to controls (Fig. 3; A, B). LGG pretreatment produced a significant decrease in anaerobes including Clostridium, Romboutsia and Veillonella (Fig. 3; C-E). LGG prophylaxis significantly suppressed expression of ICAM-1, GRO-1 and TLR4, while inducing the TLR4-inhibitor, SIGIRR (Fig. 4).
  
  Conclusion(s): LGG pretreatment significantly modified clinically relevant microbiota features of neonatal pups orally infected with E. coli, and attenuated E. coli-induced intestinal inflammation. Harnessing these relevant mechanisms afforded by probiotics such as LGG will provide novel preventive and therapeutic interventions against gut-derived neonatal sepsis.

  Presented at the 2021 PAS Virtual Conference

• 

  Neonatal Neurobehavior, Medical Risk, and 2-year Developmental Outcomes in Infants Born <30 Weeks>Gestation

  Elizabeth McGowan, Marie Camerota, Julie A. Hofheimer, Michael O'Shea, Brian S. Carter, Howard Kilbride, Steven Pastyrnak, Charles R. Neal, Lynne Smith, Jennifer Helderman, Jennifer Check, Lynne Dansereau, Sheri A. DellaGrotta, and Barry Lester

  Background: Among preterm infants, risk of poor outcomes has been linked to neonatal medical illness and neurobehavior. The combined contribution of neonatal medical morbidities and neonatal neurobehavior to 2 year outcomes has yet to be studied.
  
  Objective: To determine associations between NICU Network Neurobehavioral Scale (NNNS) profiles, medical risk, and 2 year developmental outcomes.
  
  Design/Methods: Multi-center study of 704 infants born <30 weeks gestation enrolled in the Neonatal Neurobehavior and Outcomes in Very Preterm Infants>(NOVI) Study. NNNS were completed prior to NICU discharge and six profiles of neonatal neurobehavior were calculated by latent profile analysis. Two profiles (5 and 6) were considered atypical. Medical risk included brain injury, CLD, ROP, and NEC/sepsis. Outcomes at 2 years were Bayley Scales of Infant and Toddler Development-III (BSID-III) composite scores and Child Behavior Checklist (CBCL) T-scores. Generalized estimating equation (GEE) models were used to test the associations among NNNS profiles, neonatal medical risk, and 2 year developmental outcomes. GEE models accounted for multiple births. Covariates were research site, maternal socioeconomic status (SES), race or ethnicity, maternal primary language, partner status, maternal psychopathology, and infant sex.
  
  Results: Follow-up data were available for 556/704 (78.9%) infants, of which 157 (28.2%) were Profile 5-6. Mother of Profile 5-6 infants were more likely to be non-English speaking and lower SES compared to Profiles 1-4 mothers (Table 1). In a model that included both atypical NNNS and number of medical risks, each independently predicted cognitive and motor composite scores <1 >(mild delay) and <2 SDs>(moderate delay) below the mean (Table 2). Medical risk alone predicted language composite scores <1 SD. Atypical NNNS patterns alone predicted CBCL internalizing and total problems in the clinical range>(Table 2).
  
  Conclusion(s): Neonatal medical risk remains a consistent concern for poor cognitive, language, and motor performance. Atypical neonatal neurobehavioral patterns predicted both adverse developmental outcomes and significant behavioral problems at 2 years, thereby offering an early predictive clinical tool to identify atypical neurobehavior and target NICU and post-discharge interventions.

  Presented at the 2021 PAS Virtual Conference

• 

  Neurodevelopmental profiles of infants born < 30 weeks' gestation at 2 years of age

  Marie Camerota, Elisabeth C. McGowan, Julie A. Hofheimer, T. Michael O'Shea, Brian S. Carter, Jennifer Helderman, Jennifer Check, Charles R. Neal, Steven L. Pastyrnak, Lynne Smith, Cynthia Loncar, Stephen Sheinkopf, Lynne Dansereau, Sheri A. DellaGrotta, and Barry Lester

  Background: Infants born (PMA) are at increased risk for neurodevelopmental impairment by age 2. Prior studies have tended to examine individual outcomes separately, rather than investigating whether there are subgroups of children with distinct neurobehavioral profiles.
  
  Objective: To determine distinct neurodevelopmental profiles in 2 year old children born cognitive, language, motor, and behavioral characteristics.
  
  Design/Methods: NOVI (Neonatal Neurobehavior and Outcomes in Very Preterm Infants) is a multi-center study of infants born (BSID-III) subscale scores, Child Behavior Checklist (CBCL) syndrome scores, diagnosis of cerebral palsy (CP; Gross Motor Function Classification System and abnormal neurologic exam), and positive screen for autism spectrum disorder (ASD risk; Modified Checklist for Autism in Toddlers, Revised, with Follow-Up [MCHAT-R/F]). Latent profile analysis (LPA) was applied to these measures to group children into mutually exclusive profiles.
  
  Results: Follow-up data were available for N=587/704 (83%) infants. LPA model selection statistics (e.g., Bayesian information criterion [BIC]) were used to identify a four-profile solution (Table 1; Figure 1). Children in profiles 1 (black) and 2 (blue) exhibited the best outcomes at age 2 (e.g., highest Bayley; lowest CBCL scores). Children in profile 3 (red) had the lowest Bayley scores, whereas children in profile 4 (purple) had the highest CBCL scores. Rates of CP diagnosis and ASD risk were highest in profiles 3 and 4 and lowest in profiles 1 and 2. About one-third (27%) of children were classified in one of the atypical profiles (3 or 4).
  
  Conclusion(s): We identified four qualitatively different neurodevelopmental profiles ofcognitive, language, motor, and/or behavioral impairment. The study of profiles offers a nuanced “whole child” approach to understanding outcomes for children born very preterm.

  Presented at the 2021 PAS Virtual Conference

• 

  Obesity does not increase risk for kidney replacement therapy (KRT), but weight reduction improves kidney function in children with chronic kidney disease (CKD)

  Amy Kogon, J Roem, Mark Mitsnefes, Babette Zeme, Bradley A. Warady, Susan Furth, and Nancy Rodig

  Background: Obesity is prevalent in pediatric CKD, but its impact on CKD progression is unclear.
  
  Objective: To determine the relationship between obesity/weight changes with CKD progression among children and adolescents enrolled in the Chronic Kidney Disease in Children (CKiD) study.
  
  Design/Methods: Participants were categorized based on body mass index (BMI) as normal, overweight (OW) or obese and those who were underweight were excluded. Kaplan-Meier survival curves and parametric failure time models determined the association of baseline BMI category on time to KRT stratified by non-glomerular (NG) and glomerular (G) etiology of CKD. The distribution of changes in BMI categories within sequential visit pairs (e.g., obese to OW or OW to normal BMI) and the effect of a one-unit change in BMI category on the annualized change in estimated glomerular filtration rate (eGFR) were determined. Three separate regression models were used for participants with NG and G CKD; each model included all pairs of visits with the same BMI category at the initial visit. Generalized estimating equations, adjusted for age, sex, race, proteinuria and hypertension, were used to account for repeated visit pairs within the same participant.
  
  Results: 160 (27%) of 600 children with NG and 77 (31%) of 246 children with G CKD progressed to KRT. At baseline, 15% of children with NG CKD and 26% of children with G CKD were obese. Times to KRT did not associate with baseline BMI category (Figs 1 and 2). For most, BMI category did not change over time (Figure 3). For those with NG CKD who did not change weight category, there was a similar annualized eGFR change for those who were normal weight, OW and obese. A hypothetical participant with NG CKD who is a 10-year old non-Black female without nephrotic range proteinuria and hypertension and remains obese within a pair of visits has an average annualized eGFR change of -1.0% (95% CI: -3.7%, 1.7%); similar to -1.3% (95% CI: -3.7%, 1.2%) and -0.7% (95% CI: -1.8%, 0.5%) for those who remain OW or normal weight, respectively. Among those with NG CKD who were obese, each decrease in BMI category over time was associated with a concurrent 3.7% increase (95% CI: 0.8%, 6.5%) in annualized eGFR. In those with G CKD, there was no significant difference in annualized change in eGFR by weight category or category change (Fig 4).
  
  Conclusion(s): Baseline obesity does not affect time to KRT, but weight loss in those who are obese may improve kidney survival.

  Presented at the 2021 PAS Virtual Conference

• 

  Reducing alarm burden by promoting judicious ordering of continuous pulse oximetry

  Kathleen Berg, David Johnson, Ginny Nyberg, Andrew Ausmus, Christine Claeys, Emily Wilkinson, and Nicholas Clark

  Background: Overutilization of continuous pulse oximetry (CPO) contributes to overdiagnosis and prolonged supplemental oxygen utilization. It may also negatively impact patient sleep, mobility, length of stay (LOS), and cost. Response times to actionable alarms increase with the number of non-actionable alarms. Our baseline data revealed an average of 29.6 pulse oximetry monitor alarms sounded for each admitted patient each day of their hospitalization. Most notably, 38.7% of pulse oximetry alarms were for ≥88% which is generally considered non-actionable.
  
  Objective: We aimed to decrease both total pulse oximetry alarms per patient day and alarms for ≥88% per patient day by 20%, each by September 2020.
  
  Design/Methods: This single-center quality improvement study included patients admitted to inpatient pediatric units from January 2019 to September 2020. Those in intensive care or cardiology units were excluded. Process measures were 1) percentage of patients with CPO order and 2) percentage of LOS with CPO orders in place. Outcome measures were 1) total pulse oximetry alarms per patient day and 2) alarms for ≥88% per patient day. Frequencies of high acuity transfers to intensive care and code blue events without CPO ordered served as balancing measures. Plan-Do-Study-Act cycles included: 1) changing default alarm limits from <90% to <88%, 2) changing pulse oximetry order default to intermittent rather than continuous monitoring, and 3) requiring selection of an indication for CPO from a new list within the order. Statistical process control charts monitored improvement.
  
  Results: Our project included 18,080 patients. Process measures of percentages of patients with CPO order and of LOS with CPO order in place decreased by 29.9% (44.2% to 31.0%; Fig 1) and 29.2% (45.2% to 32.0%; Fig 2), respectively. Outcome measures of total pulse oximetry alarms and alarms ≥88% per patient day decreased by 37.2% (29.6 to 18.6; Fig 3) and by 54.0% (12.4 to 5.7; Fig 4), respectively. Balancing measures were unchanged.
  
  Conclusion(s): Change in default pulse oximetry alarm limits was associated with decreased pulse oximetry alarms, particularly of non-actionable alarms. Changes to pulse oximetry order defaults and requirements impacted provider ordering behavior with subsequent decrease in frequency and duration of CPO orders, but had a lesser impact on alarms. Such system-level changes may be applied to reduce non-actionable cardiorespiratory monitor alarms and further reduce overall alarm burden and fatigue.

  Presented at the 2021 PAS Virtual Conference

• 

  The Association Between Age and Unrecognized and Untreated Hypertension in Children with Chronic Kidney Disease

  Chloe Douglas, J Roem, Joseph Flynn, Susan Furth, Bradley A. Warady, and Susan Halbach

  Background: Younger age has been associated with unrecognized hypertension (HTN) in both the general pediatric population and in children on dialysis. An increased lifetime risk of cardiovascular disease and the association of HTN with chronic kidney disease (CKD) progression make optimal treatment of elevated blood pressure (BP) a key component of CKD management. However, the role of age in the recognition and treatment of HTN in non-dialysis pediatric CKD remains unknown.
  
  Objective: Using data from the Chronic Kidney Disease in Children (CKiD) Cohort Study, we examined the relationship between age and recognized vs. unrecognized HTN, and the frequency with which stages I and II HTN are pharmacologically treated. Among children with unrecognized HTN, we also examined the relationship between age and rates of uncontrolled BP.
  
  Design/Methods: Subjects included children <18 years of age with CKD stages>2-4, for whom BP measurements were available. Subjects were stratified by age (0 to <7 >years, ≥7 to <13 >years, ≥13 to ≤18 years) and BP readings were classified by percentile per AAP clinical guidelines. Unrecognized HTN was defined as HTN stage 1 or 2 based on clinic BP measurement without a self-reported diagnosis of HTN. Uncontrolled BP was determined by self-reported lack of antihypertensive treatment among those with HTN. Generalized estimating equations to account for repeated measures were applied to logistic regression analyses to evaluate the associations of age with unrecognized HTN and medication use.
  
  Results: 890 CKiD Study participants with 3,442 annual study visits met inclusion criteria. Children <7 years of age had higher rates of stage 1 or 2 HTN and lower rates of antihypertensive use compared to older children>(Table 1). 46% of children age <7 years with hypertensive BP readings had>unrecognized, untreated HTN compared to 21% of hypertensive children >13 years of age. The youngest age group was associated with higher odds of unrecognized HTN (adjusted OR 2.03) and lower odds of taking antihypertensive medication (adjusted OR 0.48). With the exception of glomerular disease etiology of CKD, other covariates were not associated with unrecognized HTN.
  
  Conclusion(s): Children with CKD younger than age 7 years are more likely to have both underdiagnosed and undertreated HTN compared to older children. Given the impact of hypertension on CKD progression and cardiovascular disease, efforts to improve BP control in these young children are needed.

  Presented at the 2021 PAS Virtual Conference

• 

  The Effect of COVID-19 on the Detection and Prevalence of Suicidality in Adolescents

  John Lantos, Hung-Wen Yeh, Fajar Raza, Mark Connelly, and Shayla Sullivant

  Background: The COVID-19 pandemic saw changes to the delivery of health services and concerns about psychological morbidity not directly related to COVID infection. We analyzed the effects of the pandemic on our hospital’s program to screen adolescents for suicidality.
  
  Objective: To determine the effect of the COVID-19 pandemic on 1) a hospital-wide screening program for suicidality in teens; and 2) the prevalence of positive screens for suicidality.
  
  Design/Methods: Our study population was all patients > 12yo seen as inpatients or in the ED or clinics at a children’s hospital in Apr-June in 2019 (T1) and 2020 (T2). Eligible pts were screened for suicide risk using the 4-item ASW (Ask Suicide-Screening Questions). A positive response to any ASQ question was defined as positive. Results and de-identified visit characteristics were maintained in a research repository. For T1 and T2, we compared overall visits, number of telemedicine (TM) visits, proportion of eligible patients screened, proportion of positive screens, and the effect of TM. We used generalized linear mixed-effects models to test for statistical significance.
  
  Results: During T1, 24,860 patients had 38,455 visits, of which 15 (0.04%) were by TM. During T2, 16,359 patients had 24,073 visits, of which 3,372 (26.7%) were by TM. A higher percentage of eligible patients were screened at T1 relative to T2 (66.4% vs. 55.8%, z = -26.1, p < 0.001). The rate of screening in TM visits (T2 only, due to paucity of TM at T1) was lower than in in-person visits (39.3% vs 65.1%, z = -39.2, p
  Conclusion(s): In the early months of the COVID-19 pandemic, patient encounters decreased by >50%. Even with this dimished volume, a lower percentage of eligible patients were screened for suicidality. The rate of positive screens was higher than a year earlier. Screening was lowest in TM visits. The higher rate of positive screens during the pandemic could reflect doctor selection in encouraging higher risk patients to keep appointments, patient self-selection in seeking care, and/or an actual higher rate of suicidality among teens. Any of these explanations is concerning. Fewer screenings and a higher rate of positivity suggest that, during the pandemict teens at risk for suicide may not have been identified, and the prevalence of suicidality may have increased.

  Presented at the 2021 PAS Virtual Conference

• 

  Use of Clinical Criteria for Prediction of Invasive Bacterial Infection in Febrile Infants: Evaluation of the PROS Criteria

  Lyubina Yankova, Mark Neuman, Marie Wang, Adrienne G. DePorre, Sanyukta Desai, Laura Sartori, Richard Marble, Lise Nigrovic, Rianna Leazer, Christopher Pruitt, Christopher Woll, Sahar Rooholamini, Frances Balamuth, and Paul Aronson

  Background: In a 2004 study, the American Academy of Pediatrics Pediatric Research in Office Settings (PROS) network developed clinical criteria that had a sensitivity of 93.6% for identification of febrile infants at low-risk for invasive bacterial infection (IBI). Although application of the PROS criteria would reduce routine testing and emergency department (ED) evaluation for febrile infants in outpatient settings without readily available access to laboratory testing, these criteria have not been externally validated in a contemporary cohort.
  
  Objective: To evaluate the sensitivity of the PROS criteria for IBI in febrile infants ≤60 days old.
  
  Design/Methods: We performed a secondary analysis of a retrospective cohort study of febrile infants ≤60 days old with IBI who presented to 11 EDs between 7/1/2011-6/30/2016. Infants were included if 1) they had a temperature of ≥38°C at home, in the office, or at triage in the ED; 2) had a bacterial pathogen isolated in blood and/or cerebrospinal fluid (CSF); and 3) were treated for IBI. We calculated the sensitivity of the PROS criteria (age <25 >days, ill-appearance, maximum temperature ≥38.6°C) for identification of IBI.
  
  Results: Among 313 febrile infants ≤60 days old with IBI, 249 infants (79.6%) had bacteremia without meningitis and 64 (20.5%) had bacterial meningitis. The sensitivity of the PROS criteria for IBI overall was 87.9% (95% confidence interval [CI] 83.7-91.3), and was 90.6% (95% CI 80.7-96.5) for bacterial meningitis. Overall, 38 infants with IBI (12.1% of the cohort) were misclassified as low-risk, including 6 infants with bacterial meningitis (Table 1).
  
  Conclusion(s): The sensitivity of the PROS criteria for identification of IBI was lower in our multicenter ED cohort than in the original PROS study. Although the PROS criteria may help inform management of febrile infants in settings with limited access to laboratory testing, newer low-risk prediction rules for IBI are preferable when laboratory testing is available.

  Presented at the 2021 PAS Virtual Conference

• 

  Using machine learning to identify metabolomic signatures based on pediatric chronic kidney disease etiology

  Arthur Lee, Alison Abraham, Yunwen Xu, Jian Hu, Xia Or, Josef Coresh, Casey Rebholz, J Chen, Eugene Ree, Harold Feldman, Paul Kimmel, Bradley A. Warady, Susan Furth, and Michelle Denburg

  Background: Untargeted plasma metabolomic profiling combined with novel machine learning (ML) tools may lead to the discovery of metabolic profiles that inform our understanding of pediatric chronic kidney disease (CKD) etiologies and identify potential therapeutic targets.
  
  Objective: We sought to identify metabolomic signatures in pediatric CKD based on etiology: focal segmental glomerular sclerosis (FSGS), obstructive uropathy (OU), aplasia/dysplasia/hypoplasia (A/D/H), & reflux nephropathy (RN).
  
  Design/Methods: Untargeted GC/MS2 and LC/MS2-based metabolomics quantification (Metabolon) was performed on baseline plasma samples from 702 Chronic Kidney Disease in Children (CKiD) participants. Participants per etiology were: FSGS (n=63), OU (n=122), A/D/H (n=109), & RN (n=86). Lasso penalized logistic regression was used for feature selection, adjusting for age, sex, race, BMI z-score, proteinuria, estimated glomerular filtration rate, hypertension, medication usage, & CKD duration. Four methods were then applied to the selected metabolites to stratify significance; logistic regression, support vector machine, random forest, and extreme gradient boosting. Important features were selected based on being significant in at least 2 out of the 4 modeling approaches. ML significance was designated as being top 10-percentile weighted input features.
  
  Results: Participant characteristics differed based on CKD etiology. Metabolomic profiles were identified based on CKD etiology (Table 1). ML models were evaluated on hold-out validation subsets with 4 metrics; receiver-operator & precision-recall area-under-the-curve, F-1 score, & Matthews correlation coefficient. All metrics demonstrated the ML models outperformed no-skill prediction (Table 2). FSGS had strong lipid signals that remained significant when comparing FSGS within the glomerular CKD cohort. Histidine metabolites were associated with OU.
  
  Conclusion(s): We successfully trained ML models on the CKiD metabolomics data to identify metabolomic signatures based on CKD etiology. By using newer techniques such as Lasso, SVM, RF, & XGB in conjunction with traditional statistical approaches, we increased our confidence in these findings. Sphingomyelin dysmetabolism has been previously described in smaller FSGS studies. This is the largest cohort of pediatric FSGS showing associations with lipid dysmetabolism. We showed unique histidine signals in OU, which were previously undescribed.

  Presented at the 2021 PAS Virtual Conference

• 

  Utility of Diagnostic Testing in Patients who Present with Brief Resolved Unexplained Event

  Allayne Stephans, Kathryn Westphal, Erin Sullivan, Matt Hall, Risa Bochner, Adam Cohen, Jennifer Y. Colgan, Atima C. Delaney, Amy Delaroche, Thomas Graf, Beth Harper, Ron L. Kaplan, Hannah C. Neubauer, Mark I. Neuman, Nirav Shastri, Victoria Wilkins, Joel S. Tieder, and Manoj Mittal

  Background: Well appearing infants commonly present with a history of a brief event characterized by changes in color, breathing, muscle tone, or level of alertness. In 1986 the National Institutes of Health coined the term ALTE (Apparent Life-Threatening Event) to describe these events. In 2016 the American Academcy of Pediatrics (AAP) published a guideline that updated the term ALTE to BRUE (Brief Resolved Unexplained Event). This guideline includes a more precise definition and limited recommendations for the evaluation of patients who are designated as lower risk for serious underlying diagnoses. A few small studies have examined the diagnostic yield of specific testing in the BRUE population. There are no large, multi-institutional studies that focus on this research question.
  
  Objective: The primary aim was to determine if specific diagnostic tests done in BRUE patients are linked to a discharge diagnosis or if that diagnosis could have been made by history, physical, and non- invasive monitoring alone. Secondary aim is to report how frequently diagnostic testing takes place.
  
  Design/Methods: Multi institutional, retrospective cohort study of 2036 infants < 1 year of age who presented to 14 children’s hospitals between Oct 1, 2015 and Sep 30, 2018 with an event consistent with the American Academy of Pediatrics (AAP) definition of BRUE. The frequency and contributory rates of diagnostic testing were determined, distinguishing false positive vs true positive findings.
  
  Results: In this cohort 63.2% (1286) patients were hospitalized. The mean age was lower in hospitalized patients 36 days [16,77] vs 72 days [28, 165], 12.8% (262) patients met lower risk AAP criteria, and 45% (923) received an explanation for their event sometime during their ED or inpatient stay. Hospitalized patients and AAP higher risk patients were more likely to undergo more diagnostic testing. A single diagnostic test or consultation contributed to the final diagnoses in 1.89% (132/6951) cases with laboratory testing contributing in 0.9% (36/3605) tests, imaging in 1.8% (45/2408) images, and consults contributing in 5.4% (51/938) consults.
  
  Conclusion(s): Diagnostic testing and consultations contribute to final diagnoses with low overall frequency in patients that present with BRUE. History, physical exam, and non-invasive monitoring remain the most important aspects of initial management. These findings may inform AAP higher risk BRUE evaluation and management guidelines and offer providers a contextual framework for approaching initial evaluation.

  Presented at the 2021 PAS Virtual Conference

• 

  Congenital Abnormalities of the Lower Airways and Lungs

  Christopher M. Oermann

  Goals and Objectives. Participants will be able to: Describe the 5 stages of fetal lung development. Summarize the prenatal identification and management of congenital abnormalities of the lower airways and lung (CALAL). Explain the classification of CALAL. Compare the clinical characteristics, pathology, and management of common CALAL. Discuss the long term prognosis of CALAL

• 

  Missouri Department of Social Services Rare Disease Advisory Committee Update On Ivacaftor/Tezacafto/Elexacaftor

  Christopher M. Oermann

  Rare Disease Advisory Committee. Established 2019 to advise Drug Utilization Review Board. Multidisciplinary committee comprised of allied healthcare professionals (physicians, PharmD, RN, PhDs) and MO HealthNet staff. Quarterly meetings. Appointment based on experience researching, diagnosing, and treating rare diseases. Provide expert recommendations or determinations regarding access to drugs and/or biological products for rare disease treatment.

• 

  Human Trafficking and Today's Youth

  Rachel Whitfield

  This presentation is designed to help health care providers

  • Dispel myths around US domestic trafficking
  • Define the different types of human trafficking
  • Understand the contributing factors for trafficking
  • Understand the effect of pornography on the developing brain
  • Recognize risk factors and signs and symptoms of trafficking
  • Know how to report human trafficking and where to find local resources

• 

  Impact of a Mobile Device-Based Clinical Decision Support Tool on Guideline Adherence and Mental Workload Among Trainee and Attending Physicians

  Katherine M. Richardson

  This study evaluated PedsGuide, an app used to evaluate serious bacterial infections in febrile infants.

• 

  Comparison of Post-Operative Pain Control Modalities for Pectus Excavatum Repair

  Charlene Dekonenko

  This study compares patient outcomes following pectus excavatum repair using cryoablation, epidural, or patient-controlled analgesia for post-operative pain control. Primary outcome: length of stay
  Secondary outcomes: operative time, pain scores, time to only oral analgesics.
  Conclusion: Intercostal cryoablation during minimally invasive pectus excavatum repair significantly reduces length of stay and time to oral analgesics alone.

• 

  Improved Coordination of Care for PICU Patients with Newly Diagnosed Anterior Mediastinal Masses

  Laura McCarthy

  This presentation describes a quality improvement project to develop a multi-disciplinary standardized approach for prompt surgical intervention to establish a diagnosis of anterior mediastinal mass. The aim was to decrease the time from presentation to surgical diagnostic procedure totreatment.

• 

  21st Century Clinical Trials: Rethinking Best Practices Challenges and Opportunities

  William E Truog

  Panel discussion at the Pediatric Academic Societies 2019 annual meeting.

• 

  Congenital Abnormalities of the Lower Airways and Lungs

  Christopher M. Oermann

  Participants will be able to: Describe the 5 stages of fetal lung development. Summarize the prenatal identification and management of congenital abnormalities of the lower airways and lung (CALA). Explain the classification of CALAL. Compare the clinical characteristics, pathology, and management of common CALAL. Discuss the long term prognosis of CALAL.

• 

  Decreasing Unplanned Extubations by Taping Technique & Creating a Culture of Safety

  Karishma Rao, Beckie Palmer, Christopher R. Nitkin, Christian Anthony Schneider, Brandy Huitt, Molly Terhune, Ashley Orwick, Dianne Wilderson, Sarah Carboneau, Jenny McKee, Kerrie A. Meinert, and Eugenia K. Pallotto

  Describes strategies used to decrease the unplanned extubation rate in the Level IV neonatal intensive care unit to less than 1.0 event per 100 ventilator days by December 2018.

• 

  Mutant p53 Depletion by Natural Compounds

  Mohamed A.A. Alalem, Sana Farooki, and Tomoo Iwakuma

  The goal of this research project is to identify and characterize potentially novel antineoplastic agents that could specifically degrade conformational mutP53 protein in cancer cells. We concluded that treatment of cancer cells with curcumin or PLINH compounds inhibit proliferation of cancer cells in a DNAJA1-misfolded mutp5-dependent manner. Delineation of the exact mechanism through which curcumin and PLINH exert their anti-cancer effects could pave the way for promising novel preventive and therapeutic modalities for cancer.

  Erratum/Corrigendum:

  Corrigendum submitted by Mohamed Alalem

  Error on slide 6: Figure 4 updated in GREEN. Cell line is HT29 not Mia-paca-2.

  Access Restricted by Mohamed Alalem and Tomoo Iwakuma.

• 

  The Effect of Antihypertensive Dose on Blood Pressure Control in Children with Chronic Kidney Disease

  Benjamin A. Matta

  This presentation describes a study to determine the effect of antihypertensive dosing on blood pressure control in children with chronic kidney disease. The original version of this presentation is a prezi show at https://prezi.com/view/4qluDL1gMKFELOf1nLVA/

• 

  The Physiology and Hormonal Control of Calcium, Phosphate and Vitamin D

  Uri Alon

  Describes calcium, phophate, and vitamin D in the context of pediatric rickets.