Neurologic manifestations of influenza in children

It is estimated that each year, 3% to 11% of adults and children develop symptomatic influenza, while 5% to 20% are infected (1). Almost all symptomatic cases manifest as upper or lower respiratory tract infections. Children are more likely than adults to also exhibit non-respiratory signs and symptoms, including neurologic manifestations. As evidence for this disparity, about three-quarters of all cases of influenza encephalopathy or encephalitis occur in children (2).

Seizures, encephalopathy, or encephalitis (including acute necrotizing encephalopathy [ANE]) typically present within 48 hours of influenza symptom onset (2). In a recent study from Colorado, 33 of 182 (18%) children hospitalized with influenza had neurologic manifestations (3), with 18 having seizures alone and 15 having encephalopathy or encephalitis with or without seizures. In the same centre, from 2009 through 2017, the percentage of hospitalized children who presented with neurologic manifestations ranged from 9% to 24% (3). The relative risk of neurologic presentations with influenza A (H1N1), influenza A (H3N2), and influenza B has not been established but may account for some of the year-to-year variability. As in a previous study (2), underlying neurologic disease appeared to increase the risk of neurologic presentations with influenza infection.

The incidence of febrile seizures from influenza is unknown, as viral detection is not routinely pursued in children with febrile seizures. A study from Japan reported that in 16 million proven influenza infections, 2,603 patients (94% of whom were children) were hospitalized for febrile seizures (4). However, the majority of patients with febrile seizures would presumably have been managed as outpatients, and so were not captured in this study.

Influenza is a well-recognized etiology of encephalopathy and encephalitis in children (5). The term encephalopathy implies that there is brain dysfunction in the absence of inflammation (delirium). The term encephalitis implies that there is “inflammation of the brain.” The 2013 International Encephalitis Consortium defines encephalitis as unexplained decreased level of consciousness for more than 24 hours, combined with two (for possible encephalitis) or three or more (for probable or confirmed encephalitis) of the following features: fever, seizures, focal neurologic findings, cerebrospinal fluid (CSF) pleocytosis, compatible imaging, and compatible EEG findings (6). Unfortunately, influenza encephalitis and encephalopathy cannot be clearly distinguished in most case series in the literature, as definitions are not consistently applied and investigations may be incomplete. For example, a study of children admitted to 12 Canadian tertiary care centres with proven influenza from 2004 through 2013 reported that 78 of 4,155 children (1.9%) had encephalitis; however, the study did not provide a definition for encephalitis (7).

The clinical manifestations, CSF parameters, and head imaging findings of influenza encephalitis overlap with other etiologies of viral encephalitis. When suspected encephalitis occurs with influenza, CSF pleocytosis is often absent and the virus is typically not detected in the CSF, even with molecular assays (with the notable exception of a 1998 study from Japan, where virus was detected in 5 of 7 cases (8)). Possible explanations for failure to detect virus in the CSF in a given case are that 1) the neurologic manifestations are due to encephalopathy rather than encephalitis, 2) the virus has already cleared in the CSF by the time encephalitis becomes apparent, or 3) the encephalitis is an autoimmune, rather than an infectious, process.

Applying the correct label of encephalopathy, viral encephalitis, or autoimmune encephalitis may aid in selecting therapy. Treatment is not indicated for encephalopathy, as prompt and full recovery is expected (5). All studies have had an inadequate sample size to demonstrate whether oseltamivir or baloxavir prevent the development of, or shorten, the course of viral encephalitis or lead to improved long-term outcomes. However, oseltamivir appears to have limited penetration into CSF (9), which would limit utility for influenza infection of the central nervous system; CSF penetration appears not to have been studied yet for baloxavir. Immunosuppressive agents generally improve the short-term neurologic outcomes of autoimmune encephalitis, but the optimal regimen is yet to be established and it is not clear if the choice of therapy should be influenced by the precipitating infection.

ANE is a particularly severe form of encephalitis that manifests as acute, severe neurologic deterioration. First described in Japan in 1995 (10), there are now reports from North America of cases in non-Asian children (11). ANE is often accompanied by seizures and is estimated to account for up to 10% of influenza-associated encephalitis in Japan (12). It is thought to be a metabolic or mitochondrial disorder precipitated by an infection (most commonly influenza but can occur with other viruses or with Mycoplasma infection) (12). CSF protein is high in the absence of CSF pleocytosis (13). Pre-mortem diagnosis requires magnetic resonance imaging (MRI). Bilateral symmetrical changes occur on MRI, typically involving the thalami but often with extensive involvement of other parts of the brain and even on occasion the spinal cord (13). Management remains controversial (12), but immune-modulatory agents are typically given. It is noteworthy that some patients with ANE have a fairly complete neurologic recovery despite extensive changes on MRI (14). Recurrent and familial ANE occur and are most commonly due to mutations in the gene RANBP2 (13) transmitted via autosomal dominant inheritance with an estimated penetrance of 40% (12).

Acute disseminated encephalomyelitis (ADEM), transverse myelitis (TM), and Guillain-Barré syndrome (GBS) have also been linked to influenza but, unlike the previously described complications, can present weeks after influenza symptoms have resolved. This makes it difficult to establish a link, as the influenza symptoms may have been mild and non-specific and the influenza virus is typically no longer detectable by the time ADEM, TM, or GBS is diagnosed. It is therefore not surprising that ADEM (15), TM (16), and GBS (17) have rarely been linked to influenza infection in children. A causal link between GBS and influenza vaccine was first described in adults in Canada and the United States in 1976, when a vaccine for a virulent strain of swine influenza that could readily spread person-to-person was widely administered. The risk of GBS is thought to be clearly lower with modern influenza vaccines than with natural infection (18), but it is not clear whether the severity (19) or the timing of GBS varies when precipitated by vaccine versus by influenza infection. Large epidemiologic studies would be required to show the upper time limit between influenza infection and the onset of ADEM, TM, or GBS.

Cerebrovascular accidents are very rarely associated with influenza in children, with authors claiming that their 2019 case was only the third ever reported (20).

Despite the fact that molecular testing for influenza has improved the sensitivity and rapidity of diagnosis, it seems likely that the link is not always established when children are admitted with neurologic manifestations of influenza. When unexplained neurologic manifestations occur in children during influenza season, testing for influenza virus should be considered even if respiratory tract symptoms are mild or resolved, recognizing that virus may no longer be detectable by the time that complications manifest.

Competing Interests:

None declared.

Ethics Approval:

N/A

Informed Consent:

N/A

Registry and the Registration No. of the Study/Trial:

N/A

Funding:

No funding was received for this work.

Disclosures:

The authors have nothing to disclose.

Peer Review:

This article has been peer reviewed.

Animal Studies:

N/A