Long COVID in children and adolescents

Purpose of review

Although acute COVID-19 has been milder in children and young people compared with adults, there is a concern that they may suffer persistent symptoms. There is a need to define the clinical phenotype, determine those most at risk, the natural course of the condition and evaluate preventive and therapeutic strategies for both mental health and physical symptoms.

Recent findings

More recent studies with control groups reported a lower prevalence of persistent symptoms in children and young people exposed to SARS-CoV-2. A systematic review and meta-analysis found that the frequency of the majority of reported persistent symptoms is similar in SARS-CoV-2 positive cases and controls. Children and young people infected with SARS-COV-2 had small but significant increases in persisting cognitive difficulties, headache and loss of smell. Factors associated with persisting, impairing symptoms include increased number of symptoms at the time of testing, female sex, older age, worse self-rated physical and mental health, and feelings of loneliness preinfection.

Summary

This review highlights the importance of a control group in studies following SARS-CoV-2 infection, the need for case definitions and research to understand the outcomes of long COVID in children and young people.

INTRODUCTION

SARS-COV-2 is the RNA virus, which causes Coronavirus Disease 2019 (COVID-19). Ongoing symptomatic COVID-19 and post COVID-19 syndrome are referred to as long COVID, but the term post COVID-19 condition is also used as is PASC (postacute sequelae of COVID-19) and ‘long haulers’ in the USA. This review was commissioned using the term long COVID and this is the term used by the public, healthcare professionals and in literature searches and systematic reviews.

There is no diagnostic test or agreed international definition of long COVID in children. The WHO has defined post-COVID-19 condition for adults: ‘with a history of probable or confirmed SARS CoV-2 infection, usually 3 months from the onset of COVID-19 with symptoms that last for at least 2 months and cannot be explained by an alternative diagnosis. Common symptoms include fatigue, shortness of breath, cognitive dysfunction and others, which generally have an impact on everyday functioning. Symptoms may be new onset following initial recovery from an acute COVID-19 episode or persist from the initial illness. Symptoms may also fluctuate or relapse over time’ [1].

England's National Institute for Health and Clinical Excellence has defined post-COVID-19 syndrome for adults as: ‘signs and symptoms that develop during or after an infection consistent with COVID-19, continue for more than 12 weeks and are not explained by an alternative diagnosis. It usually presents with clusters of symptoms, often overlapping, which can fluctuate and change over time and can affect any system in the body. Post-COVID-19 syndrome may be considered before 12 weeks while the possibility of an alternative underlying disease is also being assessed’ [2]. 

Box 1.

no caption available

WHAT WE KNEW IN 2020?

Almost a year into the SARS-COV-2 pandemic, in December 2020, we submitted a grant application to NIHR/UKRI to undertake a study of long COVID in children and young people less than 18 years old (CYP). At that time, very little was known about the characteristics, risk factors, progression or outcomes of long COVID in children. Risk factors of severe COVID-19 at all ages included obesity, comorbid long-term conditions, learning and neurological disabilities, mental health problems and ethnic minority status. However, whether these would also be the CYP most at risk from long COVID was unknown. We postulated that the population of CYP likely to be most at risk of long COVID would be teenagers rather than younger children based on our previous clinical experience with other persistent postviral syndromes. We noted that the five case reports by Ludvigsson [3] with symptoms longer than 6 months had a median age 12 years and that depression, anxiety and fatigue may occur in teenagers after other viral illnesses including influenza. In adults, there was emerging evidence suggesting that long COVID was commoner in girls and with two predominant symptom groupings: one dominated by respiratory symptoms such as cough and shortness of breath, as well as fatigue and headaches, and the second being multisystemic, including the brain, gut and heart [4]. Our experience and the wider literature told us that, in both adults and CYP, a wide range of long-term conditions can increase the risk of mental ill-health after acute illness and so our study aimed to research both persisting physical and mental health problems after SARS-CoV-2 infection [5].

WHAT WE KNEW A YEAR LATER

In November 2021, we published a systematic review and meta-analysis of published studies of long COVID in children [6^▪▪]. We included studies of CYP with confirmed or probable COVID-19 and who had persistent symptoms as defined by authors of the individual studies. We excluded studies wherein all participants were admitted to intensive care to increase generalizability of our findings.

The literature search was completed in July 2021 and 22 studies were eligible, including 23 141 children and young people [3,7–27]. Median duration of follow-up was 125 days (IQR 99–231). Fifteen were cohort studies, six cross-sectional studies and one included case reports. Eight studies included population-based control groups. Sample size ranged from five (the original Ludvigsson [3] study) to 6804 CYP, with a total of 23 141 participants (median, 109). Eleven studies included less than 100 participants. All studies assessed outcomes at more than 4 weeks after infection (range 28–324 days), with 15 assessing outcomes at more than 12 weeks. Across all studies, 101 symptoms were reported, with 46 symptoms reported in at least two studies and 32 symptoms reported in at least three studies. Overall, the frequency of the majority of reported persistent symptoms was similar in SARS-CoV-2 positive cases and controls. This systematic review highlighted the critical importance of a control group in studies on CYP after SARS-CoV-2 infection.

CYP in the control group of the largest of the controlled studies [25] had a high rate of at least one persisting symptoms (53.3% compared with 66.5% in CYP with confirmed SARS-CoV-2 infection) 3 months after the time of SARS-COV-2 testing. For multiple persisting symptoms, 30.3% of PCR test-positive CYP and 16.2% of PCR test-negative controls had three or more symptoms, with 13.4 and 6.4%, respectively, reporting at least five symptoms. The most common symptoms among test-positive CYP were tiredness (39.0%), headache (23.2%) and shortness of breath (23.4%), and among test-negative controls were tiredness (24.4%), headache (14.2%) and other unspecified symptoms (15.8%). Although the prevalence of multiple physical symptoms was greater in CYP with confirmed SARS-COV-2 virus, their mental health, wellbeing and fatigue scores were similar to both the control group and to prepandemic norms. Notably, around 40% of both cases and controls reported feeling worried, sad or unhappy, consistent with parent-reported surveys of mental health of CYP during the pandemic [28], highlighting the overall toll of the pandemic, school closures and social isolations on the mental health and well being of CYP, irrespective of their SARS-CoV-2 infection status.

From the meta-analysis, there were some small but significant differences in the pooled risk of some persistent symptoms between cases and controls: cognitive difficulties (3%), headache (5%), loss of smell (8%), sore throat (2%) and sore eyes (2%) but not abdominal pain, cough, fatigue, myalgia, insomnia, diarrhoea, fever, dizziness or dyspnoea. Pooled prevalence of symptoms in post-COVID participants ranged from 15% (diarrhoea) to 47% (fatigue). The older the child, the higher the prevalence of all symptoms except cough.

The lack of an agreed case definition meant that it was difficult to comment on the prevalence of long COVID in children and young people. We have used a Delphi consensus methodology to propose a research definition [29] for CYP aligned to the WHO clinical definition for adults: ‘Post-COVID-19 condition occurs in CYP with a history of confirmed SARS-CoV-2 infection, with one or more persisting physical symptoms for at least 12 weeks after testing that cannot be explained by an alternative diagnosis. The symptoms have an impact on everyday functioning, may continue or develop after SARS-CoV-2 infection, and may fluctuate or relapse over time’. We believe this definition is an advance that is critical to comparing future research studies, which must also include an appropriate control group.

A sensitivity analysis was used to estimate the range of frequencies for persisting symptoms attributable to SARS-COV-2 infection [25]. For UK, 1.8–14% of the 234 803 teenagers who tested positive for SARS-COV-2 between September 2020 and March 2021 would have three or more physical symptoms and 0.9–7% five or more physical symptoms attributable to the SARS-COV-2 virus.

Using a methodology involving randomly surveying households, the United Kingdom Office for National Statistics (ONS) estimated the percentage of young people in England with self-reported ‘long COVID’ (i.e. without a specified case definition) of any duration as 0.51% for those aged 12–16 years and 1.21% for those aged 17–24 years, equating to 31 080 teenagers aged 11–17 years across UK [30].

As the case definition of long COVID becomes more refined, along with the inclusion of appropriate control groups, prevalence estimates have been gradually revised downwards [31]. Compared with April 2021, for example, when the ONS-reported long COVID prevalence of 7.4% of primary-school aged and 8.2% of secondary school aged children, this was revised down to 3.3% [95% confidence interval (95% CI) 2.5–4.5] and 4.6% (95% CI 3.5–6.0), respectively, in September 2021. Importantly, too, in the more recent ONS analyses, a matched-control group of children had similar prevalence of ‘long COVID’ despite being likely to have never had COVID-19 disease [3.6% (95% CI 2.7–4.8) and 2.9% (95% CI 2.1–4.0), for primary and secondary-school aged children, respectively].

WHAT WE KNOW NOW?

At least 14 further studies reporting prolonged symptoms in CYP have been published since the 22 reviewed in the systematic review [6^▪▪]. Four included a control group [32–34,35^▪] and 10 were uncontrolled [36–45]. These newer studies have not changed the main messages from the original systematic review. The dominant persisting symptoms were again tiredness, headache and shortness of breath. CYP in the control group of the largest of these controlled studies [35^▪] again had a high rate of persisting symptoms (33.6% compared with 43.7% in CYP with previous proven infection).

Three large Scandinavian registry studies on children provide population data. One from Norway [46^▪] showed an increase in primary care use in those with COVID-19, which persisted for up to 6 months. One study from Denmark [47^▪] showed that in 25–46% of children (depending on age) ‘long COVID’ symptoms had not resolved within 1–5 months and a second Danish study [48^▪] showed that participants with SARS-CoV-2-positive tests had more long-lasting symptoms and sick leave.

The apparently high rates of reported symptoms in control groups should be interpreted against published prepandemic norms. For example, before the pandemic, 30% of 842 adolescents aged 11–15 years reported fatigue over a 4 to 6-month period [49] and a cross-sectional survey reported 19.9% of 7977 adolescents to have headache, fatigue or asthma [50].

Recently, we developed and internally validated a model to predict which teenagers are most likely to experience at least one impairing physical symptom 3 months after a positive SARS-CoV-2 PCR-test [51]. We found that a persisting, impairing symptom is more likely with increased number of symptoms at the time of testing, female sex, older age and worse self-rated physical and mental health and feelings of loneliness before their SARS-CoV-2 PCR test. Nevertheless, it is important to emphasise that the majority of CYP reporting persisting symptoms post-COVID did not have overt mental health problems pre-COVID.

CRITICISMS OF THESE STUDIES

In our systematic review [6^▪▪], we wrote that the majority of the included studies were of poor quality, mostly uncontrolled and retrospective, and open to selection bias. Ten studies were assessed to have high risk of bias, six moderate and six low risk of bias. These sources of bias are also present in the more recent studies, particularly problems of not correcting for confounders, recall bias, referral bias and missing data.

However, these criticisms need to be considered in the context of doing real-world research at pace and scale in the midst of a global pandemic. Epidemiological research starts with a recognition of a new condition with initial case reports, then single-centre studies biased towards the most unwell, and subsequently larger population studies with more rigorous data collection from a more diverse and representative population, ideally with appropriate control groups for comparison.

Many of the CYP long COVID studies derived data from community-based surveys asking parents about their child's experiences or asking children themselves. This maximizes the size of the study population, can access hard to reach minorities and avoids selection bias from recruiting only CYP who seek medical attention. Such studies, however, have three aspects, which are in tension:

• (1)The researchers want to ask about as many variables as possible to allow their analysis to address as many potential confounders as possible.
• (2)The CYP or their proxy respondents (parents) have views on how much time they will be willing to spend providing this information. In our study, the researchers’ initial draft questionnaire asked about a great deal more than the final version but took over an hour to complete. In our pilot study, the teenagers said they would only be willing to spend 20 min maximum completing the survey, despite a financial reward at nationally recommended rates.
• (3)Young people can be deterred from completing surveys by being asked for information they consider confidential or intrusive. This can include questions about their parents’ or siblings’ health, mental health, income and so on. If the same data are asked directly from parents, that may further reduce response rates for similar reasons of concern about confidentiality around sensitive data.

Hence, it would be easy to design the perfect epidemiological study, but whether it can be done in a real world with real young people is another matter. Our response rate of 13.4% [25] was comparable to other COVID-related studies. Office for National Statistics studies had an enrolment rate 12% (July 2020--November 2021) when sampling randomly [52]. Such biases can be explored by comparing the characteristics of the respondents versus nonrespondents. Finding control groups of CYP who have never been infected will be increasingly difficult in future; vaccinated children who had no previous proven infection would be the next best alternative. Children with other proven viral infections may be an alternative control group. Historical, prepandemic data could be used for comparison where needed. These alternatives are better than no controls at all.

Moreover, all research studies which describe symptoms (whether in person, telephone or online) are, by definition, open to bias. Symptoms are the subjective description of what a person experiences. In any study of a condition without a diagnostic biomarker or confirmation by imaging or biopsy, this is an inherent feature of asking patients about their symptoms.

POTENTIAL MECHANISMS TO EXPLAIN LONG COVID IN CHILDREN

Four potential mechanisms have been postulated to explain long COVID. See Table 1[53–58].

Table 1.

Four potential mechanisms that have been postulated to explain long COVID

1.	Damage occurs to organs during acute infection and some of the damage persists leading to long-term symptoms. For example, the ‘CoverScan’ study in adults showed persisting multiorgan impairment in adults [53]. Relatively small numbers of children have been admitted to intensive care with severe organ damage as a result of acute SARS-COV-2 infection. The majority of those with paediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2 appear to be making a full recovery. Therefore, this mechanism seems unlikely to explain persisting post-COVID syndrome in children and young people.
2.	The virus persists in the body, and this leads to the persisting symptoms. This would be analogous to the mechanism of persisting or recurrent symptoms from double stranded DNA herpes viruses causing recurrent genital and labial herpes and shingles [54]. There has been a case report of biopsy proven persistence of the nucleopcapsid protein from the SARS-COV-2 RNA virus in the gut lining of a child with persisting gut symptoms [55]. In autopsies on 44 adults with COVID-19, SARS-CoV-2 was widely distributed, even among patients who died with asymptomatic to mild COVID-19 [56]. Persistent SARS-CoV-2 RNA was detected in multiple anatomic sites, including the brain, for up to 230 days following symptom onset.
3.	There is an underlying autoimmune mechanism whereby antibodies raised against the virus cross react against host tissues. This could be analogous to the mechanism of ataxia some weeks after a varicella infection due to an autoimmune inflammation of the cerebellum [57]. We are not aware of any studies supporting this mechanism in post COVID syndrome in children and young people although autoimmunity has been suggested as a factor in adults [58].
4.	Finally, perhaps none of the above biological mechanisms explain long COVID. The persistence of symptoms is unexplained as in many postviral syndromes in children.

THE FUTURE

There are a number of priorities going forward. There is still a need for ‘deep phenotyping’ CYP who describe the persisting symptoms of long COVID. This has been undertaken in adults and detailed imaging, functional physiological testing, cognitive testing and measurement of biomarkers have shown persistent abnormalities in some adults suffering from long COVID [53].

Apps allowing those with persisting symptoms to self-report symptoms and severity on a regular basis (patient-reported outcome measures) have been developed for adults and could be adapted for CYP but need appropriate control groups for comparison.

The broad ‘definition’ of long COVID in CYP, or indeed the lack of one, brings a whole range of morbidities and severities under one heading. For example, loss of smell and taste constitutes a significant proportion of CYP with long COVID. This is not as great a concern as the more serious but rare outcomes which can leave CYP unable to function. Chronic fatigue syndrome can occur after any severe illness, often with a poor prognosis. We need better long COVID definitions of the different syndromes within this condition so that we can develop appropriate referral, diagnostic and management pathways as well as better outcome studies to inform clinicians, parents and policy makers.

Finally, there is a pressing need for intervention studies. Given the multiple and varied symptoms described in over 35 studies reporting persisting post-COVID symptoms in CYP, a multicomponent intervention will be required, building on existing management of common symptoms. Studies of whether vaccination reduces the risk of long COVID in CYP who develop COVID after vaccination would be valuable.

As we continue to improve our understanding of long COVID, there is a need to support the mental health and wellbeing of CYP irrespective of their SARS-CoV-2 infection status. Although the precise number of young people who need help is difficult to ascertain, the demand is likely to exceed resources in many health systems [59,60]. For example, NHS data show that in 2021, there was the highest number of referrals to child and adolescent mental health services between April and October in 2 years. Referrals were up by 52% from 2020 to 2021, with almost 400 000 being referred [61]. YoungMinds, a charity which offers a chat line, online advice and a helpline for parents, recorded a 48% increase in demand in its services between 2019 and 2021. Papyrus, a charity dedicated to the prevention of young suicide, had a 25% increase in calls, texts and emails to its confidential HopelineUK service during the pandemic. Rapid, manualized online assistance may be necessary as a first ‘port of call’ for early intervention and then triage for face-to-face consultations with hard pressed services for the most severely affected.

CONCLUSION

There remains much uncertainty about the prevalence, natural history, risk factors, mechanisms and outcomes of long COVID in CYP. Nevertheless, we are much better informed now than in late 2020 when estimates of the risk of long COVID CYP ranged from 1 to 51% after SARS-CoV-2 infection. Although some CYP are severely affected and incapacitated long after the initial illness, the majority of the studies reviewed show a milder phenotype in CYP. Long COVID appears to be commoner in female teenagers and those with preexisting physical and mental health problems at the time of infection.

Acknowledgements

The authors acknowledge Natalia Rojas who formatted the manuscript and references.

Financial support and sponsorship

The CLoCk Study funded by the Department of Health and Social Care, in their capacity as the National Institute for Health Research (NIHR), and by UK Research & Innovation (UKRI) who have awarded funding grant number COVLT0022. All research at Great Ormond Street Hospital NHS Foundation Trust and UCL Great Ormond Street Institute of Child Health is made possible by the NIHR Great Ormond Street Hospital Biomedical Research Centre. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, UKRI or the Department of Health.

Conflicts of interest

Terence Stephenson is also the Chair of the Health Research Authority for England and therefore recused himself from the Research Ethics Application for the CLoCk Study. All other authors report no conflicts of interest.

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

• ▪ of special interest

• ▪▪ of outstanding interest