Does SARS-CoV-2 Cause Seizures and Epilepsy in COVID-19 via Inflammation or by Direct Infection?

Abstract

Incidence of Epilepsy and Seizures Over the First 6 Months After a COVID-19 Diagnosis: A Retrospective Cohort Study

Taquet M, Devinsky O, Cross JH, Harrison PJ, Sen A. Neurology. 2023;100:e790-e799. doi:10.1212/WNL.0000000000201595.

Background:

The relationship between COVID-19 and epilepsy is uncertain. We studied the potential association between COVID-19 and seizures or epilepsy in the 6 months after infection.

Methods:

We applied validated methods to an electronic health records network (TriNetX Analytics) of 81 million people. We closely matched people with COVID-19 infections to those with influenza. In each cohort, we measured the incidence and hazard ratios (HRs) of seizures and of epilepsy. We stratified data by age and by whether the person was hospitalized during the acute infection. We then explored time-varying HRs to assess temporal patterns of seizure or epilepsy diagnoses.

Results:

We analyzed 860,934 electronic health records. After matching, this yielded 2 cohorts each of 152,754 patients. COVID-19 was associated with an increased risk of seizures and epilepsy compared with influenza. The incidence of seizures within 6 months of COVID-19 was 0.81% (95% CI, 0.75-0.88; HR compared to influenza 1.55 (1.39-1.74)). The incidence of epilepsy was 0.30% (0.26-0.34; HR compared to influenza 1.87 (1.54-2.28)). The HR of epilepsy after COVID-19 compared with influenza was greater in people who had not been hospitalized and in individuals younger than 16 years. The time of peak HR after infection differed by age and hospitalization status.

Discussion:

The incidence of new seizures or epilepsy diagnoses in the 6 months after COVID-19 was low overall, but higher than in matched patients with influenza. This difference was more marked in people who were not hospitalized, highlighting the risk of epilepsy and seizures even in those with less severe infection. Children appear at particular risk of seizures and epilepsy after COVID-19 providing another motivation to prevent COVID-19 infection in pediatric populations. That the varying time of peak risk related to hospitalization and age may provide clues as to the underlying mechanisms of COVID-associated seizures and epilepsy.

Commentary

After the emergence of COVID-19 in December 2019, the resulting pandemic caused the deaths of millions and long-term health consequences among even greater numbers of people. There was evidence early on that severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) can infect brain tissue. ¹ Early studies on whether SARS-CoV-2 infection could cause seizures and resultant epilepsy and whether COVID-19 and its sequelae were more prevalent or severe in persons with preexisting epilepsy were difficult to interpret because SARS-CoV-2 antigen tests had not yet been developed or widely deployed.^1-3 In 2020 and 2021, it was observed that COVID-19 patients hospitalized with severe pulmonary signs sometimes had encephalopathy, but acute seizures and status epilepticus appeared to be relatively uncommon.^4-6

Taquet et al questioned whether patients with COVID-19 were more likely than controls to develop seizures or epilepsy. ⁷ They used TriNetX Analytics to examine 81 million (mostly US) electronic health records and identified 152 754 patients diagnosed with COVID-19 and an identical number with influenza between January 2020 and May 2021. Patients with preexisting epilepsy or seizures were excluded. The 2 groups were matched for demographics (mean age = 30.6 years) and medical comorbidities. Primary outcome was the 6-month cumulative incidence of seizures and of epilepsy. In the COVID-19 cohort compared to the control group, the incidences of seizures (0.81% vs 0.51%, respectively; hazard ratio [HR] 1.55, p <0.0001) and of epilepsy (0.30% vs 0.17%, respectively; HR 1.87, p <0.0001) were increased. The time of HR peak elevation for the whole cohort was 23 days after COVID-19 diagnosis.

The authors studied whether the increased risks of seizures and of epilepsy were age dependent. Children aged ≤16 years (n = 43 231) were compared to patients >16 years of age (n = 108 116). The risks of seizures 6 months after diagnosis of COVID-19 for children ≤16 years and for patients > 16 years of age were both elevated compared to infuenza (HR 1.83, p <0.0001 vs 1.64, p < 0.0001, respectively). Similarly, the risks of epilepsy 6 months after COVID-19 versus influenza were elevated for children and for older patients, but was numerically greater among the children (HR 2.85, p < 0.0001 vs 1.48, p = 0.00056, respectively). Further, the time after COVID-19 onset at which the HR was greatest was 50 days for the children and 21 days for those >16 years of age.

Finally, they examined whether the increased risks of seizures and of epilepsy were related to COVID-19 severity. Surprisingly, the risks of seizures 6 months after COVID-19 versus influenza was greater for outpatients (HR 1.37, p < 0.0001) but was not significantly greater for patients requiring hospitalization (HR 1.18). Similarly, the risk of epilepsy after COVID-19 versus influenza for outpatients was raised (HR 1.98, p < 0.0001) but the risk for those who were ill enough to be hospitalized was not significantly elevated (HR 1.06). Interestingly, the peak incidence of seizures or epilepsy in the 6 months after COVID-19 infection occurred at 41 days in the nonhospitalized group yet at 9 days in the hospitalized group.

In summary, at 6 months after infection, although COVID-19 was associated with increased relative risks both of seizures and of epilepsy the absolute incidence was low (less than 1% of all COVID-19 patients). There was a near-doubling of the risk of seizures, and near-tripling of the risk of epilepsy, among children with COVID-19 compared to influenza. There was also a greater risk of both seizures and of epilepsy among COVID-19 patients treated in the outpatient setting, but not among hospitalized patients.

Whereas pulmonary disease is the main manifestation of COVID-19 in adults, immune-mediated inflammation with or without multisystem inflammatory syndrome in children (MIS-C) often occurs in children. ⁸ The authors speculate that immunologic or inflammatory mechanisms due to SARS-CoV-2 infection could contribute to epileptogenesis or unmask a predisposition to seizures in the young brain. They argue that the delayed peak HRs of seizures and epilepsy in children versus adults supports this hypothesis. They further speculate that the reason seizures or epilepsy had a peak incidence 9 days after infection in hospitalized patients may have been due to various factors including fever, metabolic disturbances, and sleep deprivation, and that a delayed peak incidence in outpatients may be due to immunological factors.

The study had several limitations. Owing to the use of TriNetX Analytics, it was not possible to assess completeness of the records, to validate diagnoses, to ascertain vaccination status, to know the SARS-CoV-2 variants involved, or to determine outcomes in patients infected with the virus but who were not tested or diagnosed with COVID-19. Also, the numbers of patients acquiring influenza were small in the early years of the pandemic. As alternative control groups, the general population (which may have had lower risks of seizures or epilepsy than the influenza group) was not chosen, and patients diagnosed with herpes simplex virus infections (an epileptogenic virus) were too few to study.

In another report by some members of the current group, the TriNetX network was also used to determine longer-term outcomes and found that there was not only elevated risk of epilepsy or seizures but also elevated risks of cognitive disorder, dementia, and psychotic disorder 2 years after COVID-19. ⁹

In a very recent report, complete autopsies on 44 patients who died from COVID-19 included systematic tissue testing from multiple organs using droplet digital polymerase chain reaction (ddPCR) for detection and quantification of SARS-CoV-2 nucleocapsid gene targets and in situ hybridization (ISH) to determine the cellular tropism of the virus. ¹⁰ Immunofluorescence and chromogenic immunohistochemistry were used to further validate the presence of SARS-CoV-2 in the brain. Eleven brains were comprehensively examined. They detected and quantified SARS-CoV-2 RNA levels by ddPCR and ISH, as well as isolated virus in cell culture from multiple nonrespiratory tissues including the brain, as late as 230 days after symptom onset in 1 case. ¹⁰ High SARS-CoV-2 viral burden was found in the brain, yet few histopathological changes were observed. SARS-CoV-2 protein was expressed in neurons in samples from hypothalamus, basal ganglia, cerebellum, and cervical spinal cord. Very strong evidence was found that SARS-CoV-2 is capable of infecting and replicating in the brain (e.g., thalamus) and multiple other tissues and organs of the “entire body,” and that viral replication may occur in these tissues for several months. ¹⁰ One juvenile patient with an underlying neurological condition had no evidence of MIS-C, suggesting that children may develop systemic infection with SARS-CoV-2 without a generalized inflammatory response. ¹⁰

From the highlighted study, it appears that, although the relative risks of seizures and of epilepsy are under 1% six months after COVID-19, the risks are mildly elevated overall and are increased to an even greater extent in children ≤16 years of age and among patients who are not ill enough to require hospitalization. Whether the increased risk in children is due to MIS-C or inflammation or whether it is caused by persistent SARS-CoV-2 infection and viral replication in various brain regions remains to be determined. At a minimum, the findings of the current study, and those of a very recent study of 2168 children hospitalized with COVID-19 or MIS-C, ⁸ provide further evidence supporting the vaccination of children against SARS-CoV-2.

David G. Vossler, MD, FAES, FAAN, FACNS
Neurology University of Washington