Association of SARS-CoV-2 Seropositivity and Symptomatic Reinfection in Children in Nicaragua

Key Points Question What is the burden of COVID-19 among young children, and how does risk of reinfection vary with age? Findings In this cohort study of 1964 children in Nicaragua aged 0 to 14 years, children younger than 2 years had the highest rates of symptomatic and severe COVID-19, particularly compared with children aged 5 to 14 years. Meaning These findings suggest that the burden of COVID-19 and associated severe illness may not be evenly distributed across age groups in children.


Introduction
The SARS-CoV-2 pandemic has caused severe disease and death globally. 1,2Although most pediatric SARS-CoV-2 infections are mild or asymptomatic, severe illness can occur in children, including typical severe respiratory infections and multisystem inflammatory syndrome. 3,4Previous studies [5][6][7] have suggested that younger children, older adolescents, and children with underlying health conditions more frequently present with symptomatic or severe COVID-19, although less commonly than adults.These groups may also be more likely to have long-lasting postacute sequelae, often referred to as long COVID, although still less likely than adults. 8,9However, most of the literature regarding the burden of SARS-CoV-2 in children comes from hospital-based studies, 4,6,[10][11][12][13][14] which underestimate the disease burden in children, particularly those with mild infections.6][17] Despite understanding that differences in infection severity exist between adults and children, the burden and characteristics of SARS-CoV-2 infection among children remain poorly characterized. 3,18derstanding the implications of SARS-CoV-2 for children is particularly important because children will be among the last to be vaccinated.At the time of writing, only 1 vaccine has been approved in the US and Europe for children younger than 11 years, 19 and vaccine supplies for many countries, particularly low-and middle-income countries, remain limited. 20In fact, as of November 12, 2021, only 6.5% of people in low-income countries have received at least 1 vaccine dose. 20,21wever, even in high-income countries where older children are being vaccinated, it is critical to understand the natural history of SARS-CoV-2 in children.
As a consensus has grown around the likelihood of SARS-CoV-2 becoming endemic, researchers have begun exploring what the transition to endemicity might look like so that interventions might be tailored to be more effective.The strength and durability of immune protection against SARS-CoV-2 are perhaps the most important factors in this consideration.A February 2021 study by Lavine et al 22 suggests that even though sterilizing immunity may wane quickly, if protection against severe infections remains relatively stable, SARS-CoV-2 may become no more severe than the known seasonal human coronaviruses.However, uncertainty regarding the natural history of SARS-CoV-2 infections limits our ability to determine whether this will occur.Of particular importance is the strength and durability of immune protection against illness after infection across the severity spectrum and whether the increased circulation of variants of concern has affected COVID-19 severity in children.These questions are also of great relevance to future emerging pathogens with pandemic potential.Using data from a community-based prospective pediatric cohort in Managua, Nicaragua, we aimed to assess the burden of infection and disease in this cohort and the risk of symptomatic reinfection.

Study Population and Sample Collection
Participants were from the Nicaraguan Pediatric Influenza Cohort, the methods of which have been described in detail previously. 23Briefly, from March 1, 2020, to October 15, 2021, children aged 0 to 14 years were enrolled when visiting the Health Center Sócrates Flores Vivas or through home visits and followed up until their 15th birthday or loss to follow-up.The study was approved by the institutional review boards of the Nicaraguan Ministry of Health and the University of Michigan.
Written informed consent was obtained from a parent or guardian of all participants, and verbal assent was obtained from children 6 years or older.This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. 24rents agreed to bring their children to the study clinic (Health Center Sócrates Flores Vivas) for any illness, and all participants were provided with free medical care for the duration of their participation.Initially, respiratory samples were collected from participants aged 2 to 14 years presenting with fever or parent-reported fever and at least 1 of the following symptoms: cough, sore throat, or rhinorrhea.Respiratory samples were collected from participants younger than 2 years who presented with fever or parent-reported fever.In June 2020, these criteria were expanded so that samples were also collected from children who presented with loss of taste or smell, rash, conjunctivitis, or fever without a defined focus.Samples were collected using nasal and oropharyngeal flocked plastic swabs.Blood samples were obtained from children on enrollment and yearly thereafter (in February or March 2020 and February or March 2021).A random selection of children in the cohort in 2017 (n = 509) were enrolled into a household cohort study and had an additional blood sample collected in September or October 2020.

Laboratory Methods
RNA was extracted from respiratory samples (QIAamp Viral RNA Mini Kit, Qiagen) and tested for SARS-CoV-2 via real-time reverse transcriptase-polymerase chain reaction (PCR). 25Blood samples were tested for antibodies against the SARS-CoV-2 receptor-binding domain and spike proteins via enzyme-linked immunosorbent assay (ELISA). 26Samples were tested for an end point using 4-fold dilution from 100 until 6400, and the titer for each sample was obtained using the Reed and Muench method.

Data Collection and Case Definitions
Yearly surveys were administered at the individual and household levels every March or April, collecting data on a wide variety of social and environmental factors.A SARS-CoV-2-specific survey was also completed in September or October 2020 and February or March 2021.Finally, comprehensive medical consult forms were collected on each visit to the study health clinic along with data from follow-up visits conducted until the resolution of a child's illness.
We classified the severity of COVID-19 as subclinical, mild, moderate, or severe based on criteria used in a household transmission study within the same community. 27These definitions differed from those of the National Institutes of Health in being less stringent in the classification of moderate or severe disease.Participants who reported no symptoms associated with their infection were classified as having subclinical illness.Those with any symptoms, excluding difficulty breathing, rapid breathing, or shortness of breath, were classified as having mild illness.Moderate COVID-19 was considered illness with associated difficulty breathing, rapid breathing, or shortness of breath.
Finally, those requiring hospitalization were classified as having severe COVID-19.A more detailed description of the case definitions used in this analysis can be found in the eMethods in the Supplement.
Reinfection with SARS-CoV-2 was defined as a having a positive PCR result after an ELISApositive result and/or a positive PCR result occurring at least 60 days after an earlier PCR-positive result.We considered PCR-positive results occurring within 59 days of each other to be from the same infection, because a previous study 28  Symptoms reported between August 1, 2020, and February 15, 2021, were considered not related to SARS-CoV-2 unless there was a clear epidemiologic link (eg, confirmed infection in the household at the same time).

Statistical Analysis
We calculated incidence rates for COVID-19 and COVID-19-associated hospitalization using a Poisson distribution to estimate 95% CIs. 29,30

ELISA-Confirmed Symptomatic SARS-CoV-2 Infection
ELISA results for anti-SARS-CoV-2 antibodies were obtained for 1824 participants (92.9%), with 908 (49.8%; 95% CI, 47.5%-52.1%)having a seropositive status throughout the study.Antibody titers were obtained from 877 participants (96.6%) with a seropositive status.Seroreversion was rare, with 11 participants (1.2%) with a seropositive status subsequently testing negative by ELISA.Of these, 10 (90.9%) were older than 5 years, whereas 1 was younger than 6 months, suggesting a possible loss of maternal antibodies.In addition, among children who seroreverted, only 2 had experienced illness episodes that were symptomatic (both were mild); the rest were subclinical.Titers were highest among young children, decreasing between 4 and 6 years of age before reaching a plateau, where they remained relatively stable (Figure 2A).We saw evidence of waning maternal antibodies among young infants, with titers steadily decreasing in the first 4 months of life before beginning to increase again (Figure 2B).We observed no meaningful difference in titers by sex (eFigure 1 in the Supplement).because the youngest and oldest age groups in our study displayed the highest frequency of symptoms lasting 28 days or longer.The most common symptoms that persisted beyond 28 days of onset in our study also differed from previous reports, 9,15 with upper respiratory tract symptoms being the most common rather than the more nonspecific symptoms of fatigue and headache reported by others.
Given the frequently mild presentation of pediatric SARS-CoV-2 infections, the relative dearth of community-based cohort studies of SARS-CoV-2 infection in children has resulted in a poor understanding of the risk of reinfection.A previous study 27 in this same community found that risk after infection was similar to that after vaccination; however, this study was performed before variants of concern were widely circulating in Nicaragua and was not powered to assess risk of reinfection among young children.In the current study, we found that symptomatic second infections were common, making up approximately 20% of all PCR-confirmed episodes of symptomatic SARS-CoV-2 infection, with children 6 to 59 months and 5 to 9 years displaying lower risk of symptomatic second infection (Table 3, Figure 1C), although these differences were not statistically significant.Notably, some of these second infections resulted in illnesses classified as moderate or severe.We were underpowered to sufficiently explore severity of second infections compared with first infections.Given its implications for pediatric vaccination strategies, the relative severity of second SARS-CoV-2 infections in children should be monitored closely moving forward. 22

Strengths and Limitations
This study has several strengths.First, as a prospective, community-based pediatric cohort study, it provides valuable insight into the burden of symptomatic SARS-CoV-2 infection within a population that is likely to be among the last to be vaccinated-children in low-and middle-income countries.
Second, by using well-established systems for sample collection and testing combined with robust data collection, we were able to characterize key aspects of SARS-CoV-2 presentation and severity that remain poorly described among children.Third, by incorporating serologic testing and genetic sequencing, we were able to estimate the risk of reinfection in the context of variants such as Gamma and Delta.Fourth, although durability of immunity against SARS-CoV-2 infection after vaccination is being widely studied, fewer data are available regarding immunity after SARS-CoV-2 infection, particularly among children.
This analysis also has several limitations.The initial serologic samples were collected in February or March 2020, largely before the first wave of SARS-CoV-2 infections in Nicaragua.In the absence of a laboratory test, we also required an epidemiologic link for those reporting symptoms between August 1, 2020, and February 15, 2021, for an illness to be considered SARS-CoV-2 related.As such, it is possible that some infections, particularly milder presentations or those with less common symptom presentations (eg, gastrointestinal symptoms), were missed; however, given that a large respiratory syncytial virus epidemic occurred during that time, our definition leads to the least misclassification.In addition, most of these infections were captured via PCR or ELISA testing of samples collected in late 2020 and early 2021.To assess illness severity of infections detected only by ELISA, we relied on retrospective surveys that may have been subject to recall bias and limited our ability to pinpoint the exact time of infection for many mild cases.Fortunately, this likely affected only the mildest of cases given the expansive PCR testing criteria used.The initial testing criteria (before June 2020) were slightly broader for children younger than 2 years, which could have biased our estimates of the incidence of symptomatic COVID-19 by age.However, very few children younger than 2 years were identified using these broader criteria, so we would expect any bias to be small.
As a community-based, prospective cohort study, this study was underpowered to assess the burden of multisystem inflammatory syndrome and death associated with SARS-CoV-2 infection because of their rarity.Finally, this study was conducted before the introduction of the Omicron variant to the population.

JAMA Network Open | Infectious Diseases
SARS-CoV-2 Seropositivity and Symptomatic Reinfection in Children in Nicaragua

Downloaded From: https://jamanetwork.com/ on 09/15/2023
SARS-CoV-2 Seropositivity and Symptomatic Reinfection in Children in Nicaragua observed shedding detectable by PCR for longer than 30 days.For participants with positive ELISA results who did not have positive PCR results, we used surveys conducted in October or November 2020 and February or March 2021 to retrospectively assess COVID-19 illness severity.We observed almost no transmission of influenza or respiratory syncytial virus during periods of elevated SARS-CoV-2 transmission; thus, we assumed that symptoms reported by participants within these time periods were related to SARS-CoV-2.
seropositive starting the day their first ELISA-positive sample was collected.They were assumed to remain seropositive unless they had a subsequent sample that tested negative by ELISA.To compare against vaccine effectiveness, we estimated the protection from symptomatic reinfection as 1 − incidence rate ratio, hereafter referred to as protection.The incidence rate ratio was obtained by fitting age-stratified generalized linear models with a Poisson distribution.All analyses were conducted using R software, version 4.1.1(R Foundation for Statistical Computing).

Table 1 .
Characteristics of Study Participants