Severity of SARS-CoV-2 Infection in Hospitalized Children With and Without Obesity

AFFILIATIONS:
¹Division of Infectious Diseases, Department of Pediatrics, Baylor College of Medicine, Houston, TX
²Department of Pediatrics, Center for Research, Innovation and Scholarship in Medical Education, Baylor College of Medicine; Texas Children’s Hospital, Houston, TX
³Department of Infection Control and Prevention, Texas Children’s Hospital, Houston, TX

Abstract. Obesity is a known risk factor for severe SARS-CoV-2 infection in adults, whereas comorbidities affecting COVID-19 outcomes among children are less well defined. Our primary purpose of the study was to evaluate whether children with obesity were more likely to be at risk of severe outcomes when hospitalized with SARS-CoV2 infection when compared to those without obesity. We performed a retrospective cohort study of children with overweight and obesity (OW) compared with children with underweight and normal weight (NW) who were hospitalized with symptomatic SARS-CoV-2 infection from April to December 2020. A total of 145 children met inclusion criteria (55 in NW group; 90 in OW group). Baseline characteristics did not differ significantly between groups, except for age; the children in the OW group were older (P = .005). For older children (>10 years of age) admitted with symptomatic COVID-19, overweight or obesity was significantly associated with having pneumonia (relative risk [RR], 2.2; 95% CI, 1.2-4.1; P = .01) and an oxygen requirement (RR, 1.8; 95% CI, 1.1-2.9; P = .03). Differences in intensive care unit admission, length of hospitalization, and mortality were not observed. Larger, multisite studies are needed to further evaluate the role of obesity on COVID-19 outcomes among children. Given the prevalence of children with OW in the United States, obesity prevention efforts may lessen COVID-19 related morbidity in this population.

Introduction. Obesity is a known independent risk factor for severe COVID-19 infection in adults. Studies in adults have shown that overweight and obesity are risk factors for hospitalization, intensive care unit (ICU) admission, mechanical ventilation, and death.^1,2 Adult patients with severe COVID-19 infection show higher visceral adipose tissue accumulation than those with non-severe COVID-19.³ Obesity in adults younger than 60 years is also a risk factor for COVID-19 hospital admission.⁴ However, less is known about the effect of obesity on the severity of COVID-19 disease among children and adolescents.

The potential mechanism by which obesity increases the severity of COVID-19 is largely unknown, although it is speculated that the impact of obesity on COVID-19 infection is multifactorial. Obesity increases the expression of angiotensin-converting enzyme 2 receptors, which bind SARS-CoV-2 in adipose tissue, dysregulates the renin-angiotensin-aldosterone pathway, and promotes activation of proinflammatory cytokines due to dysregulated cellular and adaptive immunity.^2,5 Both obesity and aging lead to increased white adipose tissue deposition and white adipose tissue-derived proinflammatory cytokines such as tumor necrosis factor-α, interleukin-6, and interferon-γ.⁶

In addition, obesity increases the risk for baseline inflammation and other comorbid conditions, such as diabetes, dyslipidemia, and cardiovascular diseases in aduIts, which are also known risk factors for severe COVID-19. Among children, however, associated underlying risk factors are less common.

Therefore, we sought to further investigate the impact of obesity on severity of COVID-19 in hospitalized children at a single tertiary care pediatric hospital. Our primary aim was to determine whether hospitalized pediatric patients with overweight or obesity have more severe outcomes with symptomatic COVID-19 disease compared with those who are not overweight or obese. Our secondary aim was to determine risk factors that predict severity of COVID-19 outcomes in pediatric patients with overweight or obesity admitted with symptomatic COVID-19 disease.

Patients and methods. We performed a retrospective cohort study of children with overweight and obesity (OW) compared with children with underweight and normal weight (NW) with SARS-CoV-2 infection. Individuals aged between 2 and 18 years who were admitted to our hospital from April 1, 2020 through December of 31, 2020 with a positive SARS-CoV-2 polymerase chain reaction test and symptomatic COVID-19 were included. Asymptomatic patients undergoing surveillance testing for SARS-CoV-2 were excluded. Information abstracted from the electronic medical records included age, sex, race, ethnicity, height, weight, BMI, BMI percentile, admission and discharge dates, length of hospitalization, and discharge disposition. A manual chart review was performed by five investigators for information about underlying medical conditions, symptomatology, radiologic findings, laboratory results, admission unit, supportive and therapeutic treatments, prophylactic agents, and outcomes.

BMI categorization was determined by the CDC BMI-for-age growth charts. Patients were categorized as follows, based on age and sex: obese (BMI ≥ 95%), overweight (BMI 85% to < 95%), normal weight (BMI 5% to < 85%), and underweight (BMI < 5%).7 We defined a fever as a temperature of 38°C or greater. Presence of pneumonia was determined by radiographic report using the terms infiltrate, opacity, or consolidation and could represent infection with either a viral or bacterial pathogen. Hospitalized status required a length of stay of at least 1 day. A standardized assay for qualitative detection of SARS-CoV-2 RNA was performed at TCH Molecular Microbiology Laboratory for all patients admitted to the hospital.⁸

We identified 145 total children who met inclusion criteria, 55 of whom (38%) were NW and 90 (62%) were OW. Demographics, clinical characteristics, and outcomes of children with NW compared with children with OW and COVID-19 illness are shown in Table 1 and Figure 1.

Table 1. Demographic and clinical characteristics and outcomes of hospitalized pediatric patients with COVID-19 infection, stratified by weight categories and age.

Abbreviations: ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; MIS-C, multisystem inflammatory syndrome among children; MV, mechanical ventilation; NC, nasal cannula; NW, children with underweight and normal weight; OW, children with overweight and obesity.

^aThe percentages may not round to 100 due to rounding errors.

^bP < 0.05 are bolded.

^cIncludes patients with home noninvasive MV (2) or tracheostomy and home MV (4).

Figure 1. Distribution of BMI categories by age.

Children in the OW group were significantly older. The median age (Q1-Q3) was 10 years (range, 4-14) vs 13 years (range, 9-16) for the NW and OW groups, respectively (P = .005). Demographic variables did not vary significantly between groups. Twenty-two (40%) of the NW group were male vs 46 (51%) of the OW group (P = .2). Most children were Hispanic with 30 individuals (56%) in the NW group vs 56 (62%) in the OW group (P = .5). Ten patients were born prematurely, defined as less than 37 weeks' gestational age. Underlying asthma or chronic lung disease was present in 13 (24%) vs 31 (34%) of the NW and OW groups, respectively (P = .17). There were no differences in mean white cell count, C-reactive protein, absolute lymphocyte counts, or absolute neutrophil counts between the two groups (Table 2).

Table 2. Comparison of mean laboratory values among children with underweight or normal weight and children with overweight or obesity.

^aCalculated using Wilcoxon rank sum test.

Differences in demographics and clinical characteristics between groups were analyzed using the Wilcoxon rank-sum test for continuous variables and a chi-squared test or Fisher exact test for categorical variables. Age was dichotomized to separate younger children from both preadolescents (aged ≤ 10 years) and adolescents (age > 10 years). To estimate relative risk and confidence interval, a modified Poisson approach using robust error variance was used.9 Statistical analyses were performed using SAS 9.4 (Cary, North Carolina). A P value of less than 0.05 was considered statistically significant.

Results. For children 10 years of age or older (n = 88), being in the OW group was significantly associated with pneumonia (RR, 2.2 95% CI, 1.2-4.1; P = .01) and having an oxygen requirement (RR, 1.8, 95% CI, 1.1-2.9; P = .03). This association was not seen in younger children. Patients with pneumonia or an oxygen requirement had a statistically higher mean BMI percentile than patients without pneumonia or an oxygen requirement (P = .16 and P = .04, respectively; Table 3). The median length of hospitalization was 4 days for the NW group vs 5 days for the OW group (P = .6). Admission to the ICU was similar between the groups (P = .6). Twenty-four patients (8 NW and 16 OW) had an acute kidney injury, including two patients who required dialysis (not statistically significant, not shown in Table 1). Four patients in the OW group (three with obesity and one with overweight) developed pulmonary hemorrhage. Other adverse events among the OW group included multiorgan failure (n = 2), pericarditis (n = 1), and development of intra-cardiac thrombi (n = 1). One individual from the NW group developed rhabdomyolysis. Three patients (one NW and two OW) required extracorporeal membrane oxygenation, including one patient in the OW group who required placement of a left ventricular assist device.

Table 3. Presence of pneumonia and oxygen requirement by BMI percentile.

There was no significant difference in treatments administered to patients in the two groups, although there was a trend toward increased steroid use (29 [53%] vs 59 [67%]; P = .13) and remdesivir use (12 [22%] vs 30 [33%]; P = .14) in the OW group vs NW group. There was no difference in the receipt of either prophylactic or treatment doses of anticoagulants in either group. Four children in each group died.

Discussion. We performed a retrospective cohort study to assess the impact of OW on the severity of COVID-19 in hospitalized children. Our study demonstrates that in older children (>10 years) with SARS-CoV-2 infection, being OW was significantly associated with having pneumonia and hypoxia. This was not observed in younger children (<10 years). However, there was no significant association between obesity and ICU admission or death.   

Globally, there has been a rise in the prevalence of overweight and obesity among children and adolescents from 4% in 1975 to over 18% in 2016.¹⁰ From 2017 to 2018, the prevalence of obesity among children aged 2 to 9 years in the United States was 19.3%.¹¹ Childhood obesity was more common among Hispanic children (25.6%) and non-Hispanic Black children (24.2%).11 In a recent pediatric study, younger age group, obesity, higher white cell count, and bilateral infiltrates on chest radiograph at admission were found to independently predict disease severity.⁶ In another pediatric study, obesity and asthma were highly prevalent among hospitalized children but did not increase pediatric ICU admission.¹²

In a New York–based study of 494 pediatric patients with SARS-CoV-2 infection, Guzman and colleagues¹³ demonstrated that obesity was an independent risk factor for critical illness in adolescents aged 13 to 21 years but not among children aged 0 to 12 years. Our study similarly showed that older children with OW were more likely to be hypoxic and have pneumonia than younger children, but this did not increase their risk for ICU admission or mechanical ventilation compared with adult studies. Perhaps the protective effect in younger age children relates to age-related differences in concentrations of brown and white adipose tissue.⁶ There are mixed findings among pediatric studies analyzing the association of SARS-COV-2 infection and obesity. Obesity was not associated with increased ICU admission in one study whereas another found it to be an independent risk factor for severe COVID-19.^12,14 Most of these studies were retrospective in nature, had a smaller sample size, and did not include a control group.

With regard to rates of ICU admission, adult rates are known to be much higher with an estimate of up to 32.0%, compared with pediatric rates which range between 3.9% and 19.0%.^15-18 Mortality rates among children (including children with OW) continue to be lower compared with adults. Four children in each group died in our study, all of whom had underlying risk factors.

Importantly, the COVID-19 pandemic can also contribute to or even accelerate the epidemic of childhood obesity. Several studies have shown how children with and without SARS-CoV-2 infection were at risk for increased weight gain due to COVID-19 isolation measures in the form of school closures and lockdowns (eg, they were less physically active, engaged in increased screen time, and had changes in healthy dietary habits).^19,20 In a study of Austrian primary school-aged children, Jarnig and colleagues²¹ found that COVID-19 restrictions were associated with increases in mean BMI and the proportion of children with overweight and obesity. We suggest that promotion of healthy eating habits and physical activity strategies with patient-provider interactions as well as broader-reaching public health initiatives may help lessen COVID-19 related morbidity, especially among older children.

We acknowledge that our study has several limitations, including those inherent in a retrospective review. We included only hospitalized patients with COVID-19, which could have contributed to a selection bias and limited our sample size. This study was performed at a single pediatric referral center; thus, geographic generalizability might be low. Lastly, our cohort preceded the surge of cases due to delta and omicron variants, which may have impacted the association of obesity and COVID-19 disease severity.

To summarize, we found that among hospitalized children with SARS-CoV-2 infection, older children with overweight or obesity were at significantly increased risk for pneumonia and an oxygen requirement. A deeper understanding of the role of obesity on COVID-19 outcomes among children is needed to help risk-stratify patients and to formulate targeted prevention strategies.