Clinical spectrum of disease and outcomes in children with Omicron SARS-COV-2 infection in Cape Town, South Africa

SUMMARY INTRODUCTION Children with underlying comorbidities and infants are most severely affected by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection, including in low- and middle-income countries with a high prevalence of HIV and TB. We describe the clinical presentation of SARS-CoV-2 infection in children during the Omicron wave, in Cape Town, South Africa. METHODS We analysed routine care data from a prospective cohort of children aged 0–13 years, with a positive SARS-CoV-2 real-time reverse-transcription polymerase chain reaction (rRT-PCR) or SARS-CoV-2 antigen test, admitted to Tygerberg Hospital between 1 November 2021 until 1 March 2022. Risk factors for severity of disease were assessed. RESULTS Ninety-five children tested positive for SARS-CoV-2, of whom 87 (91.6%) were symptomatic. Clinical data were available for 86 children. The median age was 11 months (IQR 3.0–60.0), 37 (43.0%) were females, 21 (24.7%) were HIV-exposed and 7 (8.1%) were living with HIV (CLHIV). In total, 44 (51.2%) children had at least one underlying comorbidity. TB co-infection was seen in 11 children, 6 children were newly diagnosed and 5 children were already on TB treatment at the time of admission. CONCLUSION There was no evidence of more severe disease in children living with HIV or TB.

Omicron (B1.1.529),the fifth variant of concern (VoC) of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), was first reported to the WHO by South African researchers on 24 November 2021. 1,2Omicron caused a sharp rise in infections, leading to the fourth peak during the COVID-19 pandemic in South Africa with, for the first time, a decoupling of the number of infections from the numbers of hospitalisations and deaths, especially in adults, likely due to high seroprevalence in the population, either due to natural infection or COVID-19 vaccination resulting in some immunity against the virus. 3egardless of VoC, the SARS-CoV-2 disease spectrum in children has been described as mild, with mostly out-of-hospital management. 3,46][7] Underlying comorbidities and young age are considered risk factors for more severe disease.Thus, far limited data are available on the role of TB and HIV in severity of COVID-19 disease in children and how this impacted the clinical presentation during the Omicron wave.
The aim of the present study was to describe the clinical presentation and risk factors associated with disease severity in children admitted in a tertiary hospital with SARS-CoV-2 infection during the Omicron wave in Cape Town, South Africa.

METHODS
This study describes routine care data from children aged 0-13 years with a laboratory-confirmed diagnosis of SARS-CoV-2 presenting to Tygerberg Hospital (TBH) in the period from 1 November 2021 until 1 March 2022.According to the National Institute of Communicable Diseases (NICD) Surveillance report, the Omicron wave in South Africa lasted from 28 November 2021 to 30 January 2022. 8We defined laboratory confirmation as a positive real-time reverse-transcription polymerase chain reaction (rRT-PCR) or antigen test for SARS-CoV-2, performed on respiratory samples.

Setting
The study was conducted in TBH, Cape Town, South Africa, which is an academic teaching hospital that provides secondary-level care to the surrounding geographical areas, as well as tertiary care to the children of the Western Cape Province. 9In this resource-limited setting, people experience a high burden of infectious diseases, including TB-HIV and live in poor socioeconomic circumstances. 9None of the children in this study received a vaccine.

SARS-CoV-2 molecular testing
Respiratory samples were eluted in phosphate-buffered saline.Total nucleic acid content was isolated using the Nuclisens EasyMag system (bioMerieux, Marcy l'Etoile, France), the NIMBUS automated extraction system (Seegene, Seoul, Republic of Korea), and the MICROLAB ® STARlet instrument (Hamilton, Reno, NV, USA) and NucleoMag Pathogen (Macherey-Nagel, Duren, Germany) extraction kit.Positive cases were identified using molecular assays in routine use, including

Data collection and definitions
Routine care clinicians prospectively completed a case report form for each child with SARS-COV-2 on admission.The hospital laboratory system was used to ensure no children with a positive SARS-COV-2 PCR or antigen test were missed, and we retrospectively completed data of these children from hospital records.Radiological data were collected through the digital site, Picture Archiving and Communication System (PACS).
Anterior-posterior (AP) and/or lateral view chest X-ray (CXR) was routinely done in children requiring oxygen or at the discretion of the treating clinician.Single CXR reading was done retrospectively according to the WHO guidelines for pneumonia 10 by a paediatric pulmonologist (PG).
SARS-CoV-2 testing strategies changed during the pandemic.During the study period, children were only tested if they required admission.Incidental infections were defined as surveillance cases, usually done routinely before planned medical procedures or surgery.

Severity of disease
We adapted the WHO severity grading for SARS-COV-2 infection in our setting as follows: 1) critical: requiring mechanical ventilation, shock or multi-organ dysfunction/multisystem inflammatory syndrome in children (MIS-C); 2) severe: all children requiring continuous positive airway pressure (CPAP)/high-flow nasal canula (HFNC); 3) moderate: children needing lowflow nasal prong oxygen (NPO 2 ) or without oxygen need but with any CXR changes; 4) mild: if symptomatic but without a need for oxygen.

Data analysis
De-identified data were entered into a RedCap database (Vanderbilt University, Nashville, TN, USA) and analysed using SPSS v27 (IBM SPSS, Armonk, NY, USA). 11,12The clinical characteristics of the children with laboratoryconfirmed COVID-19 were summarised with descriptive statistics.Pearson χ 2 test was used to compare the differences between the different age groups.The Yates continuity correction was used if the expected cell size was <5.Non-parametric tests were used to compare differences between the age groups and continuous/dichotomous variables (Kruskal-Wallis test/Mann-Whitney U-test).For risk factor analysis, severity of disease was divided into severe disease (critical and severe) and non-severe disease (mild, moderate).Logistic regression was used to analyse risk factors for disease severity using univariate analysis.Variables with P < 0.20 were included in the multivariate model.
The Health Research Ethics Committee of Stellenbosch University, South Africa, approved this study (HREC N20/04/013_COVID).The data were entered without patient identifiers using only routinely collected data, a waiver of consent for this process was obtained.

RESULTS
During the Omicron wave, a total of 95 children tested positive for SARS-CoV-2.In total, there were 87 (91.6%) children with symptoms compatible with COVID-19 and 8 (8.4%) incidental (asymptomatic) detections (Figure 1).Clinical data were lacking in 1 child who was excluded from the analysis, and clinical description was done for the 86 symptomatic cases only.The median age was 11.0 months (interquartile range [IQR] 3.0-60.0);37/86 (43.0%) were females and 45/86 (52.3%) had one or more underlying comorbidities (Table 1).Prematurity was the most common comorbidity in children under the age of 1 year.Older children more often had a recent diagnosis or history of TB or malignancy.In total, 21/86 (24.7%) children were HIV-exposed, with 7/86 (8.1%) living with HIV (CLHIV).Baseline CXRs were done in 49/86 (57.0%) children, with abnormal findings in 28/49 (57.1%).CXRs were consistent with pneumonia in 22/49 (44.9%) cases and pneumonia with effusion in 2/49 (4.1%) cases.Most children with radiological evidence of pneumonia were less than 1 year old, with 8 (50.0%) who were less than 3 months old.
In total, 13 (15.1%)children were admitted to PICU, nine of whom were younger than 1 year of age (P = 0.40).Two children were diagnosed with MIS-C, but both were managed outside of the PICU setting.Respiratory support was required in 38 (44.2%) children, with the highest proportion (78.9%) in children younger than 1 year (P < 0.01).The overall median oxygen duration need was 5 days (IQR 1.8-9.0),without no observed differences between groups.Eight children, of whom six had at least one comorbidity, required oxygen for ≥10 days.
Using the adapted WHO severity grading system for SARS-COV-2 infection, overall, 14/86 (16.3%) children were assessed as having critical, 12/86 (14.0%) severe and 60/86 (69.8%) non-severe disease (Table 1).Most children (19/26, 73.1%) with critical or severe disease were younger than 1 year.Univariate analysis showed that age, prematurity and features of pneumonia on CXR were associated with more severe disease.In multivariate analysis, this association disappeared (Table 2).HIV exposure, infection or TB disease was not associated with more severe disease.

Table 2. Univariate and multivariate logistic regression analysis for severity of COVID-19 disease* * Only variables that were significant in univariate analysis were included in the multivariate regression. OR = odds ratio; CI = confidence interval; CLHIV = children living with HIV.
Table 3 shows the clinical presentation and underlying comorbidities seen in children with TB.TB co-infection was seen in 11 children: 6 children were diagnosed with TB during admission and 5 children were recently started on TB treatment.In total, 5 (45.5%) were CLHIV and 4 (36.4%) were confirmed TB, while 7 (63.6%)were clinically diagnosed.Typical features for TB were seen on the CXR of 3/5 (60.0%) children, with hilar lymphadenopathy being the most common.Of the 5 children who were recently diagnosed with TB, 2 had possible TB-immune reconstitution syndrome (IRIS) with clinical and or radiological worsening of disease.The median hospital admission duration was 6 days (IQR 1.0-11.0),without no observed differences between the different age groups.Three (3.5%) children died: one death (1.2%) was considered related to COVID-19.This child was less than 3 months old, and, as previously described, lung tissue was SARS-COV-2-positive at autopsy. 13The other two children had comorbidities, one 3-month-old with a cardiac defect and the other almost 9 years old with a pelvic rhabdomyosarcoma complicated by anasarca.

DISCUSSION
This study describes the clinical presentation and outcomes of children in Cape Town during the Omicron wave.In general, outcomes were good, but more severe disease continued to be seen in infants and young children with underlying comorbidities.
The first Omicron study from Tshwane, South Africa, found that 20.0% of children in their setting required oxygen therapy; however, this was not analysed according to age groups. 14Our study found that a third of children aged 0-3 months required some form of respiratory support, which was lower than we previously found during the first SARS-CoV-2 wave in our hospital. 7This confirms a trend of less severe SARS-CoV-2 disease seen in our population, with the youngest children remaining at highest risk for severe disease.
6][17][18] These include very young or older children, children with persisting symptoms and those requiring repeated doses of dexamethasone and other steroids. 15,18,19In our study, one 3-month-old child required intubation and ventilation for 4 days, and subsequently needed oxygen for 8 more days.The child received a prolonged course of steroids.
Furthermore, a third of children in our study presented with atypical febrile seizures. 20A possible explanation for this finding could be that cells within the central nervous system express ACE2, which is a binding site for SARS-CoV-2. 21,22A quarter of the children in this cohort were HIV-exposed, with 8% of CLHIV.Previous studies have shown that HIV-exposed, non-infected infants generally have a higher risk of hospitalisation due to a lower respiratory tract infection when compared to non-HIVexposed infants. 23Our data do not show more severe disease in children living with HIV or with TB disease but there was a relative high proportion of children who were identified with a new diagnosis of TB and two children with possible TB-IRIS.Both SARS-CoV-2 and Mycobacterium tuberculosis (M.tb) can elicit a hyperinflammatory state in the lung and the hyperinflammatory environment, induced by either M. tb or SARS-CoV-2 infection, could potentially accelerate TB disease progression or cause more severe COVID-19 disease. 24The exact role of viral co-infections in the inception, progression and severity of TB disease is unknown, and whether SARS-CoV-2 plays a role requires further investigations.
The COVID-19 mitigation strategies during the early phase of the pandemic initially led to temporary reduction in the occurrence of some other respiratory viruses. 25,26ith the easing of non-pharmaceutical interventions such as lockdown measures and mask-wearing, respiratory virus infections increased. 27A relatively low proportion of children had viral co-infections compared to previous studies. 28This is partially because the Omicron wave occurred during summer in South Africa, with a general lower circulation of respiratory viruses.There was no evidence that viral co-infections were associated with more severe disease.
The interpretation of our data needs to be cognizant of some limitations.This study was conducted in a tertiary referral hospital and could therefore represent more severe disease.We did not include all cases of MIS-C seen during the Omicron wave; however, the role of Omicron in MIS-C cases in Cape Town have been previously described and clinical presentation and outcomes of MIS-C during the Omicron wave were similar to the previous three waves. 29urthermore, we did not record COVID-19 vaccination coverage in the mothers of the children, which could potentially mitigate severe outcomes in infants.However, widespread SARS-CoV-2 seroprevalence was seen before the Omicron wave due to a combination of natural infection and vaccination immunity. 30Finally, due to changes in testing strategy, it is difficult to know how many people were infected during the Omicron wave and the estimated proportion of children and adults requiring hospitalisation.
In conclusion, the clinical presentation and outcomes of children during the Omicron wave in Cape Town were generally favourable, but severe respiratory disease was seen especially in infants and young children with underlying comorbidities.There was no evidence for more severe disease in children living with HIV or TB.Vaccination strategies for vulnerable young children need to be reviewed, especially in some African countries where there is not yet a policy for this age group.Children born premature might benefit from maternal vaccination during pregnancy.

Figure 1 .
Figure 1.Flowchart of children with SARS-CoV-2 infection during the Omicron wave.In total, 95 SARS-CoV-2 positive RT-PCR and antigen tests were done; 8 were done for surveillance purposes and 87 were persons under investigation.In total, 81 samples were PCRpositive.RT-PCR = reverse transcription polymerase chain reaction.

Table 1 .
Demographics and clinical presentation of SARS-COV-2 positive children by age group.*Add up to more than 100% as 13 children had more than 1 underlying comorbidity.† Includes all children with new TB diagnosis on admission (n = 6), as well as children who recently started TB treatment but are still in the intensive phase.IQR = interquartile range; NA = not available; BCG = 3acilli Calmette-Guerin; PICU= paediatric intensive care unit; CPAP = continuous positive airway pressure.HFNC = high-flow nasal cannula; NPO2 = nasal prong O2; O2 = oxygen.The clinical findings varied among the different age groups.Most children presented with fever (31/82, 37.8%), dyspnoea (29/85, 34.1%) or dry cough (23/80, 28.8%).Respiratory symptoms were more frequently seen in younger children.The children aged >5 years presented more often with vomiting (3/20, 15.0%), abnormal neurology (9/20, 45.0%) and seizures (3/20, 15.0%).One 16-month-old child presented with typical febrile seizures.The other eight children with seizures were either very young or older children with a fever and seizure (atypical febrile seizure) or in the typical age group, but without associated fever.None of these children had a previous history of febrile seizures.Two children, both 3 months old, presented with a croup-like illness.One child was ventilated with oxygen requirements for 14 days and the other required CPAP for 3 days.Laboratory findings of the different age groups were similar.The median C-reactive protein (CRP) of all children was 10 (IQR 3.0-41.8),with the lowest values noted in in the group of children between 1 and 5 years old, and the highest values noted in the oldest age group (≥5 years).