O3016: The Impact of BMI on Clinical Trajectory Among Critically Ill Children Hospitalized with COVID-19

Background: Identifying modifiable risk factors of poor clinical course among critically ill children hospitalized with COVID-19 is vital to inform public health initiatives and healthcare providers. Obesity has been associated with worse clinical prognoses for a variety of critical illnesses affecting children and among adults hospitalized for COVID-19. Yet, it remains unclear how obesity relates to clinical trajectory in the pediatric intensive care unit (PICU) setting. To address this critical knowledge gap, we aim to describe body mass index (BMI) among children admitted to the PICU, determine if BMI can discriminate poor clinical trajectory, and explore for differences and risk-factors among those with and without a poor clinical trajectory.

Methods: We performed a single-center, retrospective cohort study among children < 20 years of age hospitalized in a PICU from June 2021 through October 2021 including those with documented acute COVID-19 infection, multisystem inflammatory syndrome in children (MIS-C), and asymptomatic disease identified during screening. Cohorts were stratified by obesity (BMI >30) for comparative analyses. The primary outcome was a composite outcome for poor clinical trajectory including (1) index mortality, (2) invasive mechanical ventilation, (3) non-invasive ventilation, (4) acute venous thromboembolism, (5) and extracorporeal life support. Descriptive data included demographics, anthropometrics, COVID-19 specific treatments, and hospital length of stay (LOS). Descriptive, comparative (fisher’s exact, students t, and Wilcoxon rank sum tests), discriminatory (receiver operator characteristic [ROC] curve analyses), and exploratory (adjusted logistic regression) statistics were employed.

Results: Of the 80 subjects eligible for study, 44 (55%) comprised the composite outcome for poor clinical trajectory. As compared to those with a good clinical outcomes (Table 1), those with a poor clinical course had greater BMI (31.8 ± 17.3 versus 21.1 ± 6.2, P< 0.001), were more commonly Hispanic/Latino (38.6% versus 16.7%, P=0.046), had greater frequency of acute COVID-19 infections as compared to MIS-C (88.6/11.4% versus 66.7/13.9%, P=0.019), and longer median hospital LOS (7.4 [interquartile range: 4.7,13.9] versus 4.5 [interquartile range:2.3,7.4], P< 0.001). An ROC curve analysis of BMI on poor clinical trajectory (Figure 1) yielded an AUC of 0.693 (95% confidence interval [CI]: 0.576-0.810). Sensitivities and specificities were obtained for threshold BMI values of 30 (50% and 86.1%) and 35 (40.9% and 97.2%). A multiple logistic regression assessing BMI and the composite study outcome accounting for peak D-dimer levels and COVID-19 illness phenotype yielded an odds ratio of 1.09 (95%CI: 1.03-1.15, P=0.002)

Conclusion: The clinical trajectory for critically ill children hospitalized with COVID-19 was associated with admission BMI with threshold values for obesity and severe obesity reflecting high discriminatory capacity. These data should be used to inform next-step public health initiatives aimed to address the burden of the COVID-19 pandemic on the healthcare system as well as psychosocial and socioeconomic features that may mitigate pediatric obesity.