Objective To study seizures in patients hospitalised due to SARS-CoV-2 infection, and compare their severity with seizures in patients hospitalised due to other viral respiratory tract infections (RTIs).
Design Observational population-based cohort study.
Setting Northern Stockholm.
Patients Patients aged 1 month–18 years hospitalised due to SARS-CoV-2 with and without seizures, and patients of the same age hospitalised due to other viral RTIs with seizures, between 1 March 2020 and 30 June 2022.
Main outcome measures The prevalence of seizures in hospitalised patients due to SARS-CoV-2, the evaluation of assumed predictors of seizures and the comparison of severity markers in patients with SARS-CoV-2 versus other RTIs.
Results 32 of 239 included patients (13.4%) admitted due to SARS-CoV-2 infection had seizures. Central nervous system (CNS) disease and the omicron period had significantly increased OR for seizures (OR: 5.12; CI: 2.06 to 12.72 and OR: 3.01; CI: 1.15 to 7.88, respectively). Seizures in patients with SARS-CoV-2 were more common in children older than 5 years (p=0.001), even in the absence of fever (p=0.007), as compared with other viral RTIs. The duration of hospitalisation was longer in patients with seizures due to other viral RTIs (p=0.023). There was no significant difference regarding severity markers of seizures between the two groups.
Conclusions CNS disease and the omicron period were risk factors for seizures in patients with SARS-CoV-2, who were older than patients with other RTIs. The severity of seizures was comparable between the two groups; hospitalisation was however longer in patients with other RTIs.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplemental information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
WHAT IS ALREADY KNOWN ON THIS TOPIC
Seizures are a relatively common complication of SARS-CoV-2 infection in children. Their prevalence increased since the dominance of the omicron variant, especially in patients with epilepsy.
The prophylactic SARS-CoV-2 immunisation through vaccination of risk patients has been suggested, but data are scarce.
WHAT THIS STUDY ADDS
This study supports previous limited data.
It additionally suggests that seizures in patients with SARS-CoV-2 may have a more complex pathogenesis compared with seizures in patients with other respiratory viruses, which are more frequent in children younger than 5 years with fever.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
These findings may be considered prior to decisions regarding prophylactic vaccination against SARS-CoV-2 of at-risk patients, and in epidemiological studies on properties of respiratory viruses.
Seizures are one of the most frequently reported neurological symptoms of SARS-CoV-2 infection in hospitalised paediatric patients, along with more uncommon neurological symptoms such as stroke, encephalopathy, olfactory dysfunction and headache.1–6 Current data refer to both SARS-CoV-2-related febrile seizures and other provoked seizures.7 8 Reports on seizures in children have increased during the latest period of the COVID-19 pandemic, dominated by the omicron variant, and some authors have suggested that omicron may be more neurotropic than previous SARS-CoV-2 variants as well as that prophylactic SARS-CoV-2 immunisation through vaccination in children should be considered, but data are still scarce.5 6 8–11 According to our best knowledge, while reports on seizures in children with SARS-CoV-2 accumulate, comparisons between the severity of seizures in patients with SARS-CoV-2 and other viral respiratory tract infections (RTIs) are lacking,
In this study, we aimed to map the prevalence of seizures in paediatric patients admitted to a tertiary paediatric hospital due to SARS-CoV-2 and explore potential differences between the omicron and pre-omicron period, risk factors for seizures, and to compare the severity of seizures in patients with SARS-CoV-2 with the severity of seizures in hospitalised patients due to other viral RTIs.
Patients aged 1 month–18 years from northern Stockholm who were admitted to Astrid Lindgren Children’s Hospital (ALB) with verified SARS-CoV-2 or other respiratory viruses between 1 March 2020 and 30 June 2022 were included in this observational retrospective population-based cohort study. Patients younger than 1 month were excluded since seizures in this patient group are mostly reactive to underlying conditions.12
From the start of the pandemic in March 2020, all children at triage in the emergency department who were considered eligible for admission or general anaesthesia were screened for SARS-CoV-2. From February 2021, this screening panel also included respiratory syncytial virus (RSV) and influenza. The screening was performed with real-time PCR (RT-PCR), and the results were available before admission. To test for other viruses, a new nasopharyngeal sample had to be obtained for examination by an extended RT-PCR panel including 14 viruses, with results available within 1–2 days. Screening-positive children were only occasionally tested this way, while children with negative screening results but clinical symptoms of RTI were routinely tested with the extended virus panel. Apart from screening of patients with RTI symptoms at admission and increased hygiene routines at the paediatric hospital to limit the spread of SARS-CoV-2 among staff and other patients, there were no specific admission criteria for patients with SARS-CoV-2. Patients were admitted according to their current clinical status, need of pharmacological, nutritional or other support of vital parameters and need of monitoring as usual. In this study, we included all admitted children with any positive viral test and presumed virus-induced disease, identified by laboratory data and diagnoses codes of electronic journals.
The pandemic periods regarding the dominance of significant variants of SARS-CoV-2 (alpha, delta, omicron) were defined according to data from the Stockholm Region Medical Microbiology Laboratory, as well as the COVID-19 Data Portal, Sweden as pre-omicron: 1 March 2020–23 December 2021, and omicron: 24 December 2021–30 June 2022.13
Patients’ records were reviewed by two child neurologists (SA, ES). Some patients were admitted more than once due to an RTI, each admission was included in the total count. Patients who tested positive for SARS-CoV-2 but were admitted for other reasons than RTI (mainly trauma), or had a simultaneous bacterial infection, or had reported seizures with unclear or doubtful semiology were excluded. Data on patients living in other geographical regions in Stockholm or outside Stockholm who were admitted to ALB are not shown.
Age was evaluated as age groups 1 month–4.9 years, 5–11.9 years and >12 years to be in line with previous reports.5 We defined age group, sex, underlying central nervous system (CNS) disease and pandemic period as potential risk factors for seizures according to current literature on SARS-CoV-2.5 8
In order to make an indicative comparison of the severity of seizures in patients admitted to our paediatric department due to SARS-CoV-2 or other viral RTIs, we compared epidemiological and clinical features including underlying CNS disease, occurrence of status epilepticus (SE), admission to paediatric intensive care unit (PICU), diagnostic work-up, presence of fever and total days of admission to the paediatric department in patients with either SARS-CoV-2 or viral RTI. Evaluation of seizure semiology and SE was made according to the International League Against Epilepsy current guidelines.14–16
The association of period of the pandemic, sex, age and underlying CNS disease was evaluated by logistic regression. Comparisons of epidemiological and clinical features of seizures between patients with SARS-CoV-2 and other viral RTIs were made by Χ2, Fisher’s exact test or Mann-Whitney U test as appropriate. Statistical analyses were performed by SPSS V.26 for Windows.
Patients with seizures due to SARS-CoV-2
Of 437 hospitalised patients from northern Stockholm with SARS-CoV-2, 239 met the study inclusion criteria (figure 1). Thirty-two of 239 included patients had seizures (13.4%). The prevalence of seizures during the pre-omicron period was 8.1% (7 of 86 hospitalised patients), while the prevalence of seizures during the omicron period was 16.3% (25 of 153 hospitalised patients). Median age of patients was 1.6 (IQR: 0.3–6.9) years. Forty-four (18.4%) patients had an underlying neurological disease with CNS involvement, including 29 patients with epilepsy. In six cases, the epilepsy diagnosis was made during or shortly after admission for SARS-CoV-2 infection. Underlying CNS disease was significantly related to seizures in both univariable analyses and after simultaneous multiple adjustments for all assumed risk factors. The omicron period of the pandemic period was not related to increased risk of seizures in univariable analysis, but was significantly related to seizures in multivariable analyses including all assumed predictors, which may indicate that the increased risk of seizures during the omicron period was multifactorial. Sex and age were not risk factors for seizures in univariable or multivariable analyses (table 1).
Thirteen of 32 patients with SARS-CoV-2 and seizures had a single seizure and 19 had recurrent seizures (n=2–>10). Twenty-one of 32 patients had generalised seizures, six had focal seizures and five had focal to bilateral seizures. Data on seizures’ duration were available for 28 of 32 patients, of whom 14 met the criteria of SE and 12 had seizures with median duration 1 (IQR: 1–1.75) min. Electroencephalogram (EEG) was performed in 25 of 32 cases and was abnormal in 17 of 25 registrations. Pathological EEG findings included epileptic activity for patients with an epilepsy diagnosis and unspecific slow activity for most of the others. Fifteen of 18 patients (83.3%) with seizures without an underlying CNS disease had fever, while 7 of 14 patients (50.0%) with seizures with an underlying CNS disease had fever.
Three patients with SARS-CoV-2 and seizures had additional CNS complications related to their disease: posterior reversible encephalopathy syndrome, infarcts and encephalitis. All these cases occurred during the omicron period and none of these patients had an underlying CNS disease. None of the 32 patients with SARS-CoV-2 and seizures died due to the disease, but the patient with infarcts was left with a hemiparesis.
Patients with seizures due to other viral RTIs
There were totally 727 extended virus panel-positive admissions of 670 children. In 213 of 727 admissions (29.3%), >1 virus was detected. Seizures were present in 55 of 727 (7.6%) admissions of 51 children, in 23 of which (41.8%), >1 virus was detected (online supplemental table 1).
Of screening-positive children, there were totally 450 admissions of children with influenza or RSV. Seizures were present in 7 of 70 (10.0%) of patients who tested positive for influenza (one of them had co-detection of RSV) and in 7 of 380 (1.8%) patients who tested positive for RSV (two had single RSV infection, one had co-detection with influenza and four had co-detection of another respiratory virus (these four patients were also tested with extended panel)) (online supplemental table 1).
Of the total 64 patients with seizures due to other RTIs, 29 had one single seizure and 35 had recurrent seizures (n=2–>10). Data on seizures’ semiology were available for 62 of 64 patients, of whom 37 had generalised seizures, 20 had focal to bilateral seizures and 5 had focal seizures. Data on seizures’ duration were available for 63 of 64 patients, of whom 31 met the criteria of SE and 32 had seizures with median duration 2 (IQR: 1–3) min. EEG was performed in 41 of 64 cases (64.1%) and was abnormal in 23 of 41 (56.1%) registrations with pathological findings having the same pattern as in patients with SARS-CoV-2. Thirty-four of 38 patients (89.5%) with seizures without an underlying CNS disease had fever, and 24 of 26 patients (92.3%) with seizures and an underlying CNS disease had fever. One patient with RTI due to rhinovirus combined with metapneumonovirus had encephalitis. None of the patients with other RTIs and seizures died or had late effects of the disease after discharge.
Comparison of seizures between patients with SARS-CoV-2 or other viral infections
Comparison of the epidemiological and clinical features of patients admitted to our hospital due to SARS-CoV-2 or other viral RTIs showed that patients with SARS-CoV-2 and seizures were significantly more often older than 5 years and more often had seizures also in the absence of fever. In contrast, the duration of hospitalisation was longer in patients admitted due to other viral RTIs. Notably, five of seven patients (71.4%) with other RTIs and hospitalisation longer than 1 week had underlying epileptic encephalopathy syndromes. This may be compared with four patients with SARS-CoV-2 who were admitted longer than 1 week, of whom two (50%) had another underlying CNS disease. There was no statistically significant difference in the severity of seizures regarding occurrence of SE, diagnostic work-up or duration of admission to PICU (table 2).
Underlying CNS disease and the omicron period of the COVID-19 pandemic were risk factors for seizures in hospitalised patients due to SARS-CoV-2 in this study. Patients with SARS-CoV-2 were significantly more susceptible to seizures during infection even in the absence of fever and they were older compared with patients with other viral RTIs. On the other hand, duration of hospitalisation of patients with seizures due to other viral infection was longer.
The overall prevalence of seizures in admitted patients with SARS-CoV-2 in this cohort was higher than currently described which may reflect different local epidemiological characteristics and different criteria for hospital admission in different countries.5 6 8 Our finding of an increased prevalence of seizures in patients with SARS-CoV-2 during the omicron period is in line with other currently available studies.8–10 We additionally showed that SARS-CoV-2 was related to significantly increased rate of seizures in the absence of fever and age >5 years compared with other RTIs, which suggests a more complex pathogenesis of seizures due to SARS-CoV-2 as compared with febrile seizures in children younger than 5 years old.17
Most children with seizures due to other viral infections were younger than 5 years and had fever, which is compatible with the established definition of febrile seizures in children between 1 month and 6 years, commonly between 6 months and 5 years of age.17 18 Rhinovirus was the most common pathogen in children with other viral RTIs here, which is in line with available data on viruses related to febrile seizures that include rhinovirus among viruses related to febrile seizures.19 20 However, other viruses, especially adenoviruses, epidemic coronaviruses and influenza virus, had higher relative rates of seizures than rhinovirus (online supplemental table 1).
The longer duration of hospitalisation of patients with seizures due to other viruses probably reflects that other viruses as a group are more frequent and thereby the risk to affect children with underlying epileptic syndromes, who could be assumed to need longer hospitalisation, is higher. There was not a significant difference in severity of seizures between the two patient groups otherwise, but one patient with SARS-CoV-2 has severe sequelae due to CNS complications. Seizures were not significantly increased in older children within the SARS-CoV-2 patient group as previously shown, but this may reflect different cohort sizes.5 8 We made no comparisons of incidence of seizures or additional severe neurological complications between the two patient groups or between SARS-CoV-2 and specific other viruses because the study period is limited and longer observational periods should be merited for such analyses.
It is currently unclear if SARS-CoV-2 and especially the omicron variant is more neurotropic compared with other viruses, or if the increased incidence of seizures in patients with SARS-CoV-2 during the omicron period is multifactorial with involvement of more epidemiological factors.8 Our findings indicate that the omicron variant of SARS-CoV-2 may be more neurotropic than other viruses affecting the respiratory system. However, more data are warranted in order to understand if the omicron variant implicates seizures through enhanced activation of inflammatory response or through affecting more individuals because of its higher contagiousness resulting to more total admissions due to seizures, even if the overall prevalence of seizures in patients with SARS-CoV-2 per se would not be affected, at least during the study period.8 10 21 22 Awareness of an increasing risk of seizures and hospitalisation in patients with SARS-CoV-2 may also be taken under consideration prior to decision of vaccination of at-risk paediatric patients. Follow-up studies on SARS-CoV-2-related complications during longer periods and comparison with other RTIs are merited to be able to sort out contributing epidemiological factors to seizures, frequency of hospitalisation and accompanying severe neurological complications.23–25
The prevalence of seizures in admitted paediatric patients with SARS-CoV-2 in this study cannot be used as an indicator of the prevalence of seizures in children with SARS-CoV-2 in the community. This, because the threshold of families to seek help at the emergencies and physicians to admit children for monitoring is lower for children with seizures, while most children with SARS-CoV-2 and mild RTI symptoms will never go to the emergencies. Not all children with seizures in this study were tested with the extended viral panel and we thereby may have missed some coinfections. The comparison of seizures in patients with SARS-CoV-2 with seizures in patients with diverse viruses as a group may be overestimating or underestimating the severity of seizures due to specific viruses in this diverse viruses group. Finally, frequent co-detections complicated the analysis of the role of different viruses; enterovirus is particularly problematic since virological differentiation against rhinovirus is difficult. Nevertheless, our aim here was to make a rough indicative comparison between seizures in patients with SARS-CoV-2 with viruses already better mapped in paediatric respiratory infections, and we acknowledge that separate comparisons between SARS-CoV-2 and specific viruses are merited to make safe conclusions on convulsive properties of each virus.
Seizures are common in hospitalised paediatric patients with SARS-CoV-2, especially children with underlying CNS disease, and even more so with the omicron variant. Children with SARS-CoV-2 have higher risk of displaying seizures even in the absence of fever, while febrile seizures are more common in children with other viral RTIs. The severity of seizures is comparable between patients with SARS-CoV-2 and other viral RTIs, but patients with SARS-CoV-2 may present with more unfavourable CNS complications, which is why further studies are warranted.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplemental information.
Patient consent for publication
Ethics approval was obtained from the regional ethics committee in Stockholm (2017/1918–31).
Correction notice The licence has been updated to Open Access since this article was first published.
Contributors SA contributed to the conceptualisation of the study, analysis and interpretation of the data and drafting the initial manuscript, approved the manuscript and is accountable for this work as the guarantor. ES contributed to the conceptualisation of the study, analysis and interpretation of the data, revised the initial manuscript, approved the manuscript and is accountable for this work. RW contributed to the conceptualisation of the study and interpretation of data, critically revised the initial manuscript, approved the manuscript and is accountable for this work. OH contributed to the conceptualisation of the study and interpretation of data, critically revised the initial manuscript, approved the manuscript and is accountable for this work. MRR contributed to the conceptualisation of the study, acquisition and interpretation of data, critically revised the initial manuscript, approved the manuscript and is accountable for this work. RB contributed to the conceptualisation of the study, acquisition, analysis and interpretation of data, critically revised the initial manuscript, approved the manuscript and is accountable for this work. ME contributed to the conceptualisation of the study, acquisition, analysis and interpretation of data, critically revised the initial manuscript, approved the manuscript and is accountable for this work.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.