SCCM Resource Library-Pediatric COVID-19-Associated Hyperinflammation Score: The VIRUS Registry Predictive Analytic Study

Video Transcription

Hi. I am Meghana Nadigar. On behalf of my colleagues, Dr. Sende, Dr. Bhallala, Dr. Todapalli, and all the investigators associated with the SCCM Discovery Virus Registry Network,  I am presenting our research. This is about the development of the Pediatric COVID-19 Elevated Hyperinflammation Score, a predictive analytic study from the virus registry.  Several adult and pediatric studies have demonstrated a correlation between elevated inflammatory  markers and COVID-19 disease severity. The paper by Webb et al. in adults and the paper  by Todapalli et al. in pediatrics highlights the importance of hyperinflammation and COVID-19.  The objective of our study was to develop an additive pediatric COVID hyperinflammation score using clinical and laboratory markers of inflammation and organ dysfunction relevant  to COVID-19 illness. This SCCM virus registry was used to develop score cutoffs and evaluate the association between the pediatric COVID hyperinflammatory score, or PHS, and severe  COVID-19 illness in children. Methods Patient selection. Children under 18 years of age and who had COVID-19 infection were  filtered from the virus registry. Neonates and children incidentally detected to be positive for COVID-19 were excluded. The data from March 2020 to February 2021 was pulled at  the time. And of a total of 1,123 patients aged under 18 years of age, 722 were included for PEACHES development. Rest had missing data.  The cohort was divided into two groups, those with severe illness and those with non-severe  illness. Severe COVID-19 illness was defined as any patient receiving one of the following supportive interventions. Oxygen via face mask or high flow nasal cannula, self-proning,  invasive or non-invasive mechanical ventilation, presence of a tracheostomy, need for inhaled nitric oxide or prostant therapy, need for vasoactive medications, renal replacement  therapy, need for CPR or ECMO during the hospital stay. 33% of the children had severe COVID-19 illness. Children with severe illness were significantly  older. They were more frequently obese, more frequently had comorbidities, and had longer ICU and hospital length of stay compared to the cohort with non-severe illness.  For development of the PEACHES score, we used seven system variables, which are all markers of inflammation as demonstrated in studies by Webb et al and previously used by Totopalli  Totopalli et al for the development of their score. These included fever, hematologic dysfunction, elevated ferritin, elevated D-dimer, cytokinemia, markers of hepatic injury, and  markers of cardiac injury. Receiver operating characteristic curves were generated for each variable to choose the best cutoff value for identification of severe COVID-19 illness.  Each abnormal value got one point, and the additive PEACHES score was calculated, and  a best discriminatory PEACHES score for identification of severe disease was again determined using  ROC curves. The J-point of UDIN index was used to identify the cutoff with the formula sensitivity plus specificity minus one. This slide shows the median and interquartile range  values of various inflammatory markers in the severe illness versus the non-severe illness  groups. On comparison, the markers of illness were significantly more abnormal in the severe  illness group compared to the non-severe illness group. IL-6 levels, however, did not reach significance, probably because of the very low number of reported IL-6 levels. As such,  IL-6 levels were not included in the development of the PEACHES score.  This diagram shows the ROC curves for the various lab parameters that were used in the development  of the PEACHES score. The maximum or the minimum lab values during the first 72 hours after  admission were used to generate the ROC curves. ROC curves of the PEACHES score for development  of severe COVID-19 illness and for the need for ICU admission are also shown here. This table  shows the discriminant values and the odds ratios by univariate analysis of each of the components.  Odds of severe COVID-19 illness were higher with fever, macrophage activation markers, including elevated ferritin, presence of any of the parameters of hematologic dysfunction,  presence of coagulopathy, hepatic dysfunction, cytokinemia, or elevated cardiac enzyme.  This column shows the cutoff values that were obtained for each of the inflammatory markers by using ROC curves. With these cutoffs, the additive PEACHES score was calculated.  The median PEACHES score for the cohort with severe illness was 3.7 with an interquartile range of 2.1. The median PEACHES score for the non-severe illness cohort was 2.2  with an interquartile range of 1.9, and this difference was statistically significant. The area under ROC for PEACHES was 0.7. When PEACHES score cutoff was set at greater than  or equal to 2, the sensitivity was 80% and the specificity was 46% for development of severe COVID-19 illness. When for a PEACHES of greater than or equal to 3, the sensitivity was  69% and the specificity was 62% for development of severe COVID-19 illness. On binomial logistic  regression analysis based on age, obesity, and presence or absence of MIS-C, a PEACHES score of 2 or more was significant for development of severe COVID-19 illness.  Strengths and limitations. The strength of our study is that it reflects a large multi-center registry involving 43 centers and is a very large pediatric cohort.  Coming to limitations. Inflammatory markers such as IL-1 and IL-6 have been described to be  elevated in COVID-19-associated hyperinflammation, but these are not readily available in all centers  and results may not be immediately available for calculation of the score early in the disease  process. Therefore, IL-1 and IL-6 were not included in the development of the PEACHES score. Similar to all multi-center registry, data analysis may be limited by data integrity  and validation issues at individual participating sites. There may be some inclusion bias because  it's unknown whether all patients with COVID-19 from each site were entered into the database.  Future directions to the study include further statistical analysis for missingness in effect sizes, breaking the cohort up into derivation and validation, and the use of a new  derivation and validation cohorts. Eventually, the PEACHES score needs to be externally validated  with a prospective study before it can be clinically applied. Future studies are needed to evaluate disease patterns caused by recent weight. In conclusion, we have described a  rational additive scale for characterizing the pediatric COVID-19 hyperinflammatory state  using clinical and lab markers of inflammation and organ dysfunction relevant to COVID-19 illness.  The score may be calculated from early laboratory data and may be useful in predicting severe  disease in children with COVID-19. Its role in clinical practice needs to be determined and validated in a prospective study with the potential future use of incorporating it into  risk stratification processes or treatment algorithms. Thank you.

Video Summary

Researchers from the SCCM Discovery Virus Registry Network have developed the Pediatric COVID-19 Elevated Hyperinflammation Score (PEACHES) to predict the severity of COVID-19 in children. The score, based on clinical and laboratory markers of inflammation and organ dysfunction, was developed using data from 722 children with COVID-19. Higher PEACHES scores were associated with severe illness and the need for intensive care unit admission. The score has the potential to be used in risk stratification and treatment algorithms for children with COVID-19, but further validation and prospective studies are needed. The study involved a large multi-center registry, but limitations include the availability of certain inflammatory markers and data integrity issues.

Asset Subtitle

Worldwide Data, Infection, Pediatrics, 2022

Asset Caption

INTRODUCTION: Several adult and pediatric studies demonstrate a correlation between elevated inflammatory markers and COVID-19 disease severity. The Society of Critical Care Medicine (SCCM) Discovery Viral Infection and Respiratory Illness Universal Study (VIRUS) COVID-19 registry was used to develop Pediatric COVID Hyperinflammation Syndrome (PcHIS) score and evaluate the association between PcHIS and severe COVID illness in children. METHODS: Children under 18 years of age hospitalized due to COVID-19 were filtered from VIRUS registry (NCT 04323787). Neonates and children incidentally positive for COVID were excluded. For the development of PcHIS score we used 7 variables: fever, hematologic dysfunction (platelet, leucocyte count), elevated ferritin, elevated D-dimer, cytokinemia (CRP, procalcitonin, IL-6), hepatic injury (ALT, AST, albumin) and elevated cardiac enzymes (BNP, Troponin). ROC curves were generated for each variable to choose the best discriminatory (J point of Youden Index) value for identification of severe disease (anyone requiring respiratory support more than O2 by NC, vasoactive meds, ECMO, or dialysis). Each abnormal value got one point and the additive PcHIS score was calculated for the best discriminatory score for identification of severe disease (using ROC). RESULTS: Out of a total of 1123 patients aged < 18 years with COVID-19 in the registry, 722 were included for PcHIS development; rest had missing data. A 1/3rd in the cohort had severe COVID disease. Odds of severe disease were higher with fever > 39ºC (OR1.5;CI1.05-2.14), presence of any hematologic dysfunction (platelets < 250k/µL or WBC > 6650/µL) (OR 7.12;CI 2.52-20.05), cytokinemia (CRP >6.7 mg/dL or procalcitonin > 3.4) (OR 4.99;CI 3.05-8.17), ferritin level > 270 mg/dL (OR 3.87;CI 2.38-6.28), elevated cardiac enzymes (BNP > 685 or Troponin > 0.03) (OR 3.08;CI1.96-4.85), hepatic injury (AST >50 or ALT >40 or albumin < 3.5 g/dL) (OR 3.25;CI 2.16-5.0), D-dimer > 2000 ng/ml (OR 2.46;CI1.59-3.8). A PcHIS score of 2.5 had a sensitivity of 69.2% and a specificity of 62.1% with ROC area under curve of 0.70 (95% CI: 0.66-0.74; p < 0.001). CONCLUSIONS: PcHIS score may be calculated from early laboratory data and is useful in predicting severe disease in children with COVID-19. Its role in clinical practice needs to be determined in a prospective study.

Meta Tag

Content Type Presentation

Knowledge Area Infection

Knowledge Area Pediatrics

Knowledge Area Worldwide Data

Knowledge Level Advanced

Membership Level Select

Tag Epidemiology Outcomes

Tag Pediatrics

Tag COVID-19

Tag Infectious Diseases

Year 2022

Keywords

Pediatric COVID-19 Elevated Hyperinflammation Score

PEACHES

severity of COVID-19 in children

clinical and laboratory markers

organ dysfunction

	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.