4 DISCUSSION
In the present study involving 182 pediatric COVID-19 patients, we found that the most common symptoms were fever and dry cough, consistent with other reports.25 A slightly male dominance was found in these patients, which was close to that reported in the USA26 and China.27 Most children were infected by family cluster, as observed previously28, which was different from that of adults who were infected during social activities. A small proportion of pediatric patients were normal in CT images. As for those with abnormal CT images, ground glass opacity and patchy shadowing were the most common features, and consolidation was rare in pediatric patients when compared to adult patients.
The proportion of patients with allergies and asthma in the current study was 23.6% (43/182), which was similar to the general prevalence reported in China.16 Age distribution, sex ratio, contact history, clinical symptoms, radiological alterations, severity, treatment and outcomes were not different between allergic and non-allergic patients. The proportion of COVID-19 children with allergic rhinitis was 19.8% (36/182), which was also close to the prevalence of AR recently reported (17.6%) in children of Wuhan.29In consideration of the increasing prevalence of AR in China,30 it could be concluded that AR is not a predisposing factor for infection of SARS-CoV-2. Intranasal corticosteroids should be continued for COVID-19 children with AR, as stated in a position paper of the European Academy of Allergy and Clinical Immunology (EAACI).31 There was no evidence to show that intranasal corticosteroids had any impact on the outcome and the virus shedding of children with AR.
There was only one asthma children in the current study, suggesting that asthma is not a predisposing factor for SARS-CoV-2 infection in children, the same as in adults.22,32This is probably because angiotensin-converting enzyme-2 (ACE2), the cellular receptor of SARS-CoV-2, is less expressed in airway epithelia of individuals with allergic asthma and rhinitis.33 Another possible reason for rare asthma patients in COVID-19 is that most of asthmatic patients infected with SARS-CoV-2 were asymptomatic and thus could not be included in the study. Noticeably, this study involved 10 children (median age, 3.5 years old) with a history of recurrent wheezing, a common symptom in preschool children, which is mainly caused by bronchiolitis and asthma.34 It is not easy to make a definite diagnosis of asthma in these children, although treatments such as bronchodilators or inhaled corticosteroids will relieve the symptom.35
Lymphocytes play an important role in antiviral immunity and contribute to the cytokine storm responsible for pulmonary and systematic inflammation of COVID-19 patients. Lymphopenia was common in adult COVID-19 patients and the degree of lymphopenia was associated with the severity and outcome of the patients.22,36 However, fewer patients with lymphopenia were observed in pediatric patients, as show in the current and previous studies.10,11 This is consistent with and also the cause of the high prevalence of asymptomatic and mild courses in COVID-19 children. Interestingly, eosinopenia was also observed in a significant fraction (29.5%) of infected children, though the percentage was less than that of adults, as reported in a previous study that half of the adult COVID-19 patients had decreased count of eosinophils.22 Given the positive correlation between lymphocyte counts and eosinophil counts, as we demonstrated previously, we speculate that decreased Th2 lymphocytes and cytokines such as IL-5 may be associated with eosinopenia. The ratios of patients with eosinopenia were not different between allergic and non-allergic patients, as well as between younger children and older children. On the other hand, the ratios of eosinophilia were similar between allergic and non-allergic COVID-19 children. Thus, eosinopenia could not be used as a diagnostic indicator of COVID-19 in children, in contrast to that suggested in adults.22
The median levels of total serum IgE in all patients, including allergic and non-allergic patients, were within the normal range. The ratios of patients with increased IgE levels were not different between allergic and non-allergic as well as younger and older children with COVID-19. Most of the allergic disease in these children was well-controlled AR in this study, since only few patients were still using intranasal corticosteroids during the COVID-19. It must be pointed out that no data about specific IgE, skin prick test or other provocation tests were available in these patients, which indicates that a few AR patients might be “misdiagnosed”. Taken together, our results suggest that allergy has no obvious impact on the disease course of COVID-19 in children. Other immunoglobins such as total IgG, IgA, IgM, and complements C3 and C4 were all in normal ranges, without difference between allergic and non-allergic patients, although higher ratio of patients with increased IgA, C3 and C4 were seen in older children, and both ratios of patients with increased and decreased IgG were observed in younger children.
As far as we know, there is no study focusing on the pediatric COVID-19 children with different severities. Considering that most children infected with SARS-CoV-2 were asymptomatic or presented as AURI and mild pneumonia, this is rational. In our cohort, only four children were categorized as severe or critical cases, including one death due to intussusception and secondary multiple-organ failure. The other three severe children were recovered.
Since the rarity of severe and critical pediatric COVID-19 cases, we compared the clinical and laboratory features of mild pneumonia cases with those patients without pneumonia. Clinically, mild pneumonia was associated with more patients with manifestations, especially fever and cough; in addition, allergy prevalence was not different between the two groups. In pediatric COVID-19 patients, most of these parameters were in the normal ranges, except for PCT and CK-MB, both were slightly elevated in nearly half of the cases, without clinical significance.
Most laboratory findings were comparable between the children with mild pneumonia and without pneumonia. Some biochemical and immunological parameters differed between the two groups, but all were in the normal ranges. These differences may reflect the inflammatory responses induced by pneumonia. The incidence of elevated PCT, ALP, IL-10 and decreased complement C3 in pneumonia was higher in mild pneumonia patients than those without pneumonia. This is a little different from adult COVID-19 patients, that inflammatory biomarkers CRP and PCT, coagulation indicator D-dimer, myocadiac injury indicator CK/CK-MB, and liver and renal function indicators ALT, AST, ALP and BUN in the blood were all elevated and were more prominent in severe and critical cases, as studies previously demonstrated.22,37
Peripheral lymphocyte subsets alteration has been reported in studies on adults COVID-19 patients. CD8+ cytotoxic T cells was an independent predictor for COVID-19 severity and treatment efficacy.38 CD4+ and CD8+ T cells counts were also correlated with the disease severity and outcome.39 So far, there were few studies focusing on the lymphocyte subsets in pediatric cases. In the present study, we found that in COVID-19 children, both the numbers and percentages of T cells, CD4+ T cells, CD8+ T cells, B cells and NK cells were mostly in the normal ranges, as well as the levels of cytokines involved in inflammation, immune regulation and antiviral immunity such as IL-2, IL-4, IL-6, IL-10, TNF-α and interferon-γ in the serum. The slight or unchanged immunological function may contribute to the clinical features of pediatric COVID-19 patients such as lower incidence, milder symptoms, shorter course of disease and fewer severe cases.40 The median numbers of total lymphocytes and lymphocyte subsets were different between younger and older children, but still in the corresponding normal ranges of their ages. Older children had more changes in the lymphocyte subsets, including more cases with increased percentages of T cells and CD8+ T cells, and decreased numbers of B cells. Younger children were more frequently associated with increased CD4+ T cells numbers and percentages. Our results provide deep insight into the immunological features of COVID-19 children. Consistent with the clinical symptoms and laboratory findings, the lymphocyte subsets and cytokines were not different in allergic and non-allergic COVID-19 patients. Thus, allergy plays a negligible role in the incidence, disease course and outcomes of COVID-19 in children.
Only a minority of children were with evidences of possible co-infection with other pathogens, most of which was MP, and few were EBV, CMV, adenovirus and influenza B virus. COVID-19 children with evidence of MP co-infection were associated with relatively lower monocyte counts and more azithromycin use, and also lower ratio of ground glass opacity in chest CT when compared to those without evidence of MP co-infection (Table S1 and S2). The severity and percentages of children with allergy were not different between children with or without evidence of MP co-infection. Thus, MP co-infection might play a minor role in the disease course of COVID-19 in children. It is noticeable that there were no patients with evidence of coinfection with rhinovirus (RV) and respiratory syncytial virus (RSV), which are more commonly found in children with AR and wheezing.41 During the quarantine period, many tests performed in commercial institutions were stopped including RV and RSV detections, which resulted in no RV and RSV identified in this study. In addition, the physicians tended not to find other pathogens if the children had been confirmed with SARS-CoV-2 infection by RT-PCR assays. We could not exclude the co-infection in COVID-19 children with other respiratory viruses, which may also play a role in the immunity against SARS-Cov-2.
Treatments of COVID-19 children during hospitalization were diverse and mostly empirical, without evidence of randomized controlled trials (RCTs). Almost all (97.8%) children were treated with inhalation of interferon-α, based on the studies and experiences of its usage in treating other respiratory infection and viral pneumonia.42 Therefore, we could not know the exact clinical effects of interferon-α by comparing the outcomes of patients treated with or without interferon-α. There is no clinical evidence to address the effect of IFN-α use in COVID-19 children. Therefore, the treatment of interferon-α should be evaluated with well-designed RCTs in the future.
In the present cohort, almost all pediatric patients had good outcomes, except one death. This is consistent with previous report of the outcome in pediatric COVID-19 patients.9-11 Both patients with mild and without pneumonia recovered well and were discharged from the hospital. Duration of hospitalization in this study was not different between patients with mild and without pneumonia, but shorter than that of adult patients.37 This is due to the large proportion of patients without pneumonia, which could be discharged after a few days of medical surveillance in the hospital. In addition, the median time from first positive to first negative RT-PCR results of SARS-CoV-2 were 7 days in both patients with mild and without pneumonia. This is also shorter than the time needed for virus clearance in adult patients.37,43 The cause of this difference is unclear, but it was reported that older age and presence of chest tightness were independent factors affecting negative conversion of the virus RNA.43 It may also be associated with different immunological response between pediatric and adult patients. The negative conversion of RT-PCR (two consecutive negative results at least with a 24-hour interval) was one of the discharge criteria, which caused its positive correlation with the duration of hospitalization. Interestingly, in patients with pneumonia, the duration of hospitalization had mildly negative correlation with the level of IFN-γ, which indicated the possible antiviral roles of interferon. More importantly, most of children were treated with IFN-α, and its effect on the negative conversion of virus RNA in respiratory tract sample could not be excluded.
There are a few limitations of this study. Firstly, not all children in the designated hospital could be included in the present study; secondly, the allergic status of most children was according to medical history, but not with medical records or allergen testing; and the allergens responsible for the allergy were reported only in 30% of allergic patients. Thirdly, the dominance of asymptomatic patients and mild pneumonia prevented most patients from repetitive laboratory tests, thus the estimation of dynamic variations in blood cell counts and biochemical parameters was impossible.
In conclusion, pediatric COVID-19 patients tended to have mild clinical course, and severe cases were rare. Patients with mild pneumonia had higher proportion of fever and cough and increased inflammatory biomarkers than those without pneumonia, and the both were with favorable prognosis. Allergic and non-allergic COVID-19 children were not different in aspects of incidence, clinical characteristics, laboratory and immunological findings. Allergy is not a predisposing factor for SARS-CoV-2 infection and plays no role in the disease course of COVID-19 at least in children.