In a recent systematic review posted to the medRxiv* preprint server, researchers discussed the validity of reported health outcomes post-acute infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in children.
Coronavirus disease 2019 (COVID-19) usually causes mild or no symptoms in children, and hospitalization and fatalities are rarely observed. However, persistent symptoms post-acute infection phase have been documented in adults and children; this condition is termed post-COVID syndrome or long COVID. Most studies assessing long COVID in children reported heterogeneity in (their) study designs with several limitations, including the lack of control groups. It is crucial to provide valid and precise risk estimates for assessing the benefits and harms of COVID-19 preventive measures among children.
In the current study, researchers systematically investigated the validity of reported health outcomes of infection with SARS-CoV-2 on post-COVID syndrome in children. The authors identified research studies that reported long-term health consequences post-acute SARS-CoV-2 infection from the PubMed and Web of Science databases.
Peer-reviewed or preprint studies were considered if they included children aged below 18 years with a follow-up period of two months or more post-COVID-19 diagnosis, hospitalization, or one month or more post-recovery from acute disease (or hospital discharge). Meta-analyses, abstract-only publications, evidence syntheses, and conference proceedings were excluded from the study.
Studies that evaluated the quality-of-life outcomes, symptoms and other relevant outcomes regarding how children felt were eligible. Those with laboratory and imaging analyses were not included. The most frequent symptoms reported among all studies and the proportion of children demonstrating the primary outcome were determined.
The authors determined whether in these studies the SARS-CoV-2 infection and the primary outcome were causally interpreted or a conceptual causal model described the causal mechanisms or made any explicit causal disclaimers. Confounding bias consideration (in studies) was assessed based on what the findings meant, and the limitations were. The risk of bias was also assessed using the risk of bias in non-randomized studies of interventions (ROBINS-I).
The researchers obtained 21 eligible research studies. Six of them were controlled studies, and the remaining lacked controls. Nineteen studies reported ethical approval, and only two studies were registered. Most (13) of the studies were conducted in Europe, and 18 were published in 2021. These selected studies encompassed 81,896 children, with 1335 children diagnosed with COVID-19 in uncontrolled studies. In those with control groups, 15651 children were infected with SARS-CoV-2, and 64910 were not.
Seventeen studies exclusively included hospitalized children, and five included children with multisystem inflammatory syndrome (MIS-C). The SARS-CoV-2 diagnosis was confirmed exclusively with reverse transcription polymerase chain reaction (RT-PCR) tests across 11 studies. Among the controlled studies, SARS-CoV-2-negative status was ascertained by serological investigations in two studies, based on the absence of symptoms in three studies, and RT-PCR results in one study.
Twelve studies collected outcome data two to seven months after initial diagnosis and nine studies with two to 11.5 months of follow-up period post-recovery. Only two studies clearly defined the post-acute symptoms. The principal outcome of 16 studies was a composite of any symptom; in others, the primary outcome was fatigue, gastrointestinal, respiratory, olfactory, and cardiac symptoms. The three most frequent symptoms reported were headache, fatigue, and cough. The reported percentage of children experiencing post-COVID syndrome ranged from 0 to 66.5% in COVID-19-infected children.
Among the controlled studies, two to 53.3% of children without COVID-19 showed symptoms associated with the post-COVID syndrome. Only one controlled study provided a statistical comparison between controls and the infected children. Sixteen studies did not address confounding bias, while three studies mentioned confounding bias, and two others statistically considered the confounding bias. Other biases were discussed in 10 studies, and none of the 21 studies stated cautious interpretation or mentioned limitations in the abstract.
Only two controlled studies clearly interpreted the causality associated between COVID-19 and the primary outcome and recommended preventive measures, with one of them suggesting their findings be confounded. None of the studies used a conceptual causal mechanism of interpretation. The risk of bias was critical due to confounding in all studies. Nineteen studies reported severe to critical risk of bias in participant selection. Twelve studies that employed self-reporting methods observed serious risk of bias in measuring outcomes, while four studies that utilized structural clinical assessment data observed moderate risk.
In conclusion, the present study that systematically reviewed 21 studies noted that the validity of reported health outcomes post-COVID-19 in children was critically limited. No study offered concrete evidence suggesting that SARS-CoV-2 infection impacted post-COVID-19 health outcomes. Understanding the severity and frequency of post-COVID syndrome is crucial. These observations warrant more robust and reliable methodological data so that children and their families can address and improve their health concerns.
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.