296 - Defining Clinical Phenotypes of COVID Viral Infections and Post-Viral Syndromes

Poster Number: 296
Publication Number: 296.215

Christopher L. Carroll, Connecticut Children's Medical Center, Avon, CT, United States; Alexander H. Hogan, Connecticut Children's Medical Center, Hartford, CT, United States; Katherine W. Herbst, Connecticut Children's Medical Center, Hartford, CT, United States; Celina M. Porcaro, Connecticut Children's Medical Center, West Hartford, CT, United States; Schulman R. Noah, Connecticut Children's Medical Center, West Hartford, CT, United States; Hassan El Chebib, Connecticut Children's Medical Center, Hartford, CT, United States; Juan C. Salazar, University of Connecticut School of Medicine, Hartford, CT, United States

Background: SARS-CoV-2 infection exhibits a wide range of clinical outcomes, from asymptomatic and mild disease to severe viral infection and severe post-viral syndromes. Epidemiologic, clinical, and laboratory predictors are urgently needed in the fight against the pandemic in children, where treatment for these different conditions requires prompt and accurate diagnoses.

Objective: To define the phenotypic characteristics of children hospitalized with acute, and post-infectious sequela of SARS-CoV-2 (COVID) compared to children hospitalized with non-COVID infections and inflammatory conditions.

Design/Methods: Prospective cohort study of children enrolled from 4/1/2020 through 12/20/2021. Three physicians categorized subject’s primary diagnosis as COVID viral infections, non-COVID viral infections, multisystem inflammatory syndrome in children (MIS-C), and Kawasaki cohorts and agreement were assessed by Fleiss’ Kappa. Acute viral infection cohorts (COVID vs. non-COVID viral) were compared, as were post-viral syndromes (MIS-C vs. Kawasaki), by Wilcoxon rank-sum or chi square.

Results: There were 178 children enrolled; 56 with COVID viral infections, 73 with non-COVID viral infection, 39 with MIS-C, and 10 with Kawasaki (Figure 1 and Table 1). Interrater reliability was substantial (kappa = 0.75). Presenting symptoms varied by cohort (Figure 1). Demographics, treatments, and outcomes also varied by cohort (Table 1). Comparing the acute viral infection hospitalizations, the COVID cohort was significantly older (Table 1). Antimicrobials and remdesivir were given more often in acute COVID patients, although non-COVID viral patients received steroids more frequently (Table 1). The length of stay was also significantly longer in the acute COVID cohort (3 vs. 2 days, P=0.002) compared to the non-COVID viral cohort. When comparing the post-viral syndromes, the children with MIS-C were significantly older (Table 1). All children in the post-viral syndrome cohorts received intravenous immunoglobulin (IVIG), but MIS-C patients routinely received steroids (95%) and antimicrobials (87%) while Kawasaki patients did not (30% and 10% respectively, both p < 0.001). MIS-C patients were significantly more likely to require an ICU admission (50% vs 0%), however the groups had similar length of stays (p=0.14).Conclusion(s): Children in these cohorts differed by demographics, outcomes, work-ups and treatments received. These cohort phenotypes may be useful to help identify potential biomarkers in this population.
Figure 1. Median Age & Length of Stays by Cohort with three most common presenting symptoms of each cohort.Circle centered on median value and size of circle proportionate to number of children in each cohort.
Table 1. Clinical Data by Cohort.COVID vs non-COVID viral cohorts, and MIS-C vs Kawasaki cohorts compared using non-parametric statistics. Data presented as frequency or median (25-75% IQR).