New Data on COVID-19 in US Children and Viral Load, Infectivity and Serologic Response to COVID-19
Angela Myers, MD, MPH | Director, Division of Infectious Diseases | Associate Director, Infectious Diseases Fellowship Program; Associate Professor of Pediatrics, UMKC School of Medicine
The Centers for Disease Control and Prevention (CDC) published a Morbidity and Mortality Weekly Report (MMWR) article providing the first description of clinical characteristics of U.S. children diagnosed with COVID-19 (the disease caused by the SARS-CoV-2 virus).1 Data from >149,082 cases of patients aged 0-64 years old was included from Feb. 12 to April 2 using a voluntary reporting system. Among these cases, 2,572 (1.7%) were children <18 years. The majority of cases were in children aged ≥10 years (table 1), and 57% with known gender were male. COVID-19 exposure information was available for 184 cases, and 91% had exposure to a known COVID-19 patient in either their household or community. While data regarding symptoms, underlying conditions, and hospitalization were only available for a small portion of children, it does provide some insight into the clinical characteristics and course of disease that children infected with SARS-CoV-2 are experiencing.
Among 213 children for whom clinical data was obtained, 73% presented with fever, cough, or shortness of breath, compared to 93% of 10,167 adults. Children were more likely to have fever (56%), or cough (54%), compared to shortness of breath (13%). One-quarter of these children had myalgias (23%), sore throat (24%), and/or headache (28%), but only a small portion had rhinorrhea (7.2%), abdominal pain (5.8%), nausea or vomiting (11%), or diarrhea (13%). In contrast, the clinical data in the adult population revealed that fever (71%), cough (80%), and shortness of breath (43%) were common. Additionally, nearly two-thirds of adults developed myalgias, 58% had headache, and 31% had diarrhea. These data are consistent with the data from other countries that speaks to milder disease in children compared to adults. 2-5
To compensate for not having hospitalization and intensive care unit (ICU) status data for two-thirds of both children and adults, the CDC used an interesting approach in their calculation of the estimated ranges for overall hospitalization/ICU care in different age groups. The lower number of the range was calculated using a denominator that was the sum of the number of cases whose hospitalization/ICU status was known, plus the number whose status was unknown. In contrast, the denominator for the higher number in the range used only the number whose hospitalization status was known.
Using this method, children were less likely to be hospitalized and to require ICU care than adults. Between 5.7% (sum of those with unknown and known status as denominator) and 20% (only those with known status as denominator) of children were hospitalized, compared to 10-33% of adults; between 0.5-2% required PICU care compared to 1.4-4.5% of adults needing ICU care. Hospitalization and PICU care were highest in infants <1 year. There were three pediatric deaths reported; review is ongoing to determine whether the cause of death was COVID-19.
Nearly one-quarter (n=80) of 345 children for whom data were available, had at least one underlying condition. The most common underlying condition was chronic lung disease, which included asthma in 40 children, followed by cardiovascular disease in 25 and immune suppression in 10. Need for hospitalization was higher among children with known underlying conditions, 28/37 (77%), and all PICU admissions had an underlying condition.
While these data are not complete, there is an inkling that the clinical characteristics in U.S. children are similar to those in China.2-4
Table 1. Ages of children diagnosed with COVID-19 infection
In another report, a German research team obtained samples from nine hospitalized patients as early as the onset of symptoms from the nasopharynx or oropharynx to establish the diagnosis. Most patients had very mild or prodromal symptoms at the onset. Of note, four out of nine patients noted olfactory and taste alterations. Following hospitalization, samples were repeatedly obtained from oro- and nasopharynx, and sputum. Peak viral concentration varied over time based on specimen site, but all respiratory swabs from all patients were positive for the first five days. The viral concentration on upper respiratory tract samples taken after day 5 dropped precipitously and only had a 40% positivity rate (meaning that virus was no longer being shed).6 There was no difference between viral concentration in oropharyngeal compared to nasopharyngeal swabs.
There were seven patients that had paired upper airway and sputum samples between days 2-4 of illness. Of these, five paired samples had similar viral loads, two had higher viral load in the oro- or nasopharyngeal swab, and two had higher viral load in the sputum. The last positive swab was taken on day 28 post-symptom onset.
Twenty-seven urine and 31 blood samples were obtained for molecular testing and all were negative. Stool samples were positive with generally high concentrations of virus and remained positive, along with sputum, for over three weeks in six of nine patients, despite full resolution of symptoms typically by the end of the first week. Co-infection with common respiratory viruses was assessed for all patients and was not detected.
The authors then looked at infectivity. To do this, they performed live virus isolation from positive clinical samples. Virus was able to be cultured (isolated) from 17% of oro- or nasopharyngeal swabs, and from 83% of sputum samples. However, virus was not able to be cultured in any sample taken after day 8 of illness, even though patients were still shedding viral RNA in high concentrations. Additionally, the ability to isolate virus in culture was dependent upon the viral RNA concentration in the molecular test. For patients who had a molecular test result with <106 RNA copies/mL virus in respiratory samples, viable virus was not detected in culture. Interestingly, viable virus was not detected by culture from any stool specimen. The researchers also evaluated active virus replication (another but indirect way to evaluate activity of virus) by detection of viral messenger RNA (mRNA). Viral mRNA is only present in host cells in which the virus entered and hijacked the essential machinery to replicate. Messenger RNA is not in the whole virion itself. Active viral replication (mRNA) was found in human airway cells of oropharyngeal samples for the first five days, with a steep decline by day 8, and none was found by day 10.
Finally, the authors evaluated seroconversion using an IgM and IgG assay. They found that 50% of patients seroconverted by day 7, mostly with IgG, and 100% by day 14. However, they did note some cross-reactivity or cross-stimulation from the seasonal coronaviruses in several patients.
What does this tell us in terms of ability to detect virus, infectivity, replication and serology? It tells us that the upper airway has a high amount of virus initially and that an upper airway swab provides good sensitivity for detection of infection within the first week. The virus is actively replicating during the first week and likely infectious, as the ability to culture virus is a good surrogate for transmissibility. Although patients may have viral RNA detectable by PCR from airway specimens for a prolonged time, viable viral shedding decreases more quickly over time and patients are unlikely to be infectious for the entire length of time that viral RNA is detectable by PCR. Virus clearly infects upper airway cells, as well as lower airway cells (seemingly through the ACE2 receptor common on epithelial cell surfaces), but this appears to resolve by day 10. While patients are shedding large quantities of viral RNA in the stool, cultures of stool were negative (COVID-19 virus needs its fragile envelope intact to be infectious and envelope integrity is not readily maintained in stool) and thus may not represent a transmission risk. Finally, while there is cross-reactivity with more common coronaviruses among serology specimens, COVID-19 IgG seems a good indicator of infection, if tested at least two weeks post illness onset.
- Accessed on April 6, 2020 at https://www.cdc.gov/mmwr/volumes/69/wr/mm6914e4.htm?s_cid=mm6914e4_w.
- Guan W, Ni Z, Hu Y, et al. Clinical Characteristics of Coronavirus Disease 2019 in China. Accessed on March 23, 2020 https://www.nejm.org/doi/full/10.1056/NEJMoa2002032 .
- Lu X, Zhang L, Du H, et al. SARS -CoV-2 Infection in Children. NEJM. Accessed on March 23, 2020. https://www.nejm.org/doi/pdf/10.1056/NEJMc2005073.
- Dong Y, Mo X, Hu Y, et al. Epidemiological Characteristics of 2143 Pediatric Patients with 2019 Coronavirus Disease in China. Pediatrics. Accessed on March 23, 2020. https://pediatrics.aappublications.org/content/pediatrics/early/2020/03/16/peds.2020-0702.full.pdf.
- Korean Society of Infectious Diseases. Report on the Epidemiological Features of Coronavirus Disease 2019 (COVID-19) Outbreak in the Republic of Korea from January 19 to March 2, 2020. 2020; 35:e112.
- Wolfel R, Corman VM, Guggemos W, et al. Virologic Assessment of Hospitalized Patients with COVID-2019. Nature. Accessed on April 6, 2020 at https://www.nature.com/articles/s41586-020-2196-x_reference.pdf.