You are on page 1of 9

Epilepsia, 53(Suppl.

4):63–71, 2012
doi: 10.1111/j.1528-1167.2012.03615.x


Seizures and encephalitis: Clinical features,

management, and potential pathophysiologic
*yBenedict D. Michael and *yTom Solomon

*Institute of Infection and Global Health, University of Liverpool, Liverpool, United Kingdom; and
yThe Walton Centre Neurology NHS Foundation Trust, Liverpool, United Kingdom

evidence for seizures occurring in the context of

antibody-associated encephalitis, an increasingly
Encephalitis is an inflammation and swelling of the important condition. Finally, we describe the
brain, which is often caused by a viral infection; it is acute and longer-term management of encephali-
an important cause of acute symptomatic seizures tis-related seizures and their potential pathophysi-
as well as subsequent epilepsy. Herein we describe ologic mechanisms, concluding with the emerging
the definition, epidemiology, and etiology of etiologic role of human herpesvirus 6.
encephalitis as a cause of seizures. We then focus KEY WORDS: Encephalitis, Seizures, Epilepsy,
on encephalitis due to herpes simplex virus (the Herpes simplex virus, Japanese encephalitis virus,
most common sporadic viral cause of encephalitis) Human herpesvirus 6, Antibody-associated enceph-
and Japanese encephalitis virus (the most com- alitis, Voltage-gated potassium channel complex,
mon epidemic viral cause). We also discuss the N-Methyl-D-aspartate receptor.

Encephalitis is inflammation and swelling of the brain, Seizures and Epilepsy

which is often caused by a viral infection; it is an impor-
Acute symptomatic seizures, defined by the Interna-
tant cause of acute symptomatic seizures as well as subse-
tional League Against Epilepsy (ILAE) as seizures occur-
quent epilepsy (Misra et al., 2008).
ring within 7 days of an acute central nervous system
Herein we describe the definition, epidemiology, and
(CNS) infection, occur in 2–67% of patients with encepha-
etiology of encephalitis as a cause of seizures. We then
litis (Misra et al., 2008). However, this is probably an
focus on encephalitis due to herpes simplex virus (the
underestimate because seizures are often subtle and
most common sporadic viral cause of encephalitis) and
because nonconvulsive status epilepticus is often difficult
Japanese encephalitis virus (the most common epidemic
to diagnose in confused or comatose patients (Solomon
viral cause) (Solomon et al., 2008; Bauer & Bien, 2009;
et al., 2007; Misra et al., 2008). Underdiagnosis is a partic-
Granerod et al., 2010). We also discuss the evidence for
ular problem where there is difficulty with electroenceph-
seizures occurring in the context of antibody-associated
alography (EEG) access. The incidence both of acute
encephalitis, because of its increasing importance (Gran-
symptomatic seizures and of subsequent epilepsy varies
erod et al., 2010). Finally, we describe the acute and
with the cause of encephalitis, the patient’s age, delays in
longer-term management of these seizures and their
starting treatment, and possibly the degree of cortical
potential pathophysiologic mechanisms, concluding with
inflammation (Misra et al., 2008; Bauer & Bien, 2009).
the emerging etiologic role of human herpesvirus 6.
There are only limited detailed data from epidemiologic
studies of the seizure occurrence in viral encephalitis, the
likely underlying mechanisms, and the prognosis: this is
particularly so in resource-poor settings, where data are at
Address correspondence to Tom Solomon, Institute of Infection and
Global Health, University of Liverpool, The Apex Building, West Derby
best available only from passive surveillance.
Street, Liverpool L69 7BE, U.K. E-mail: Patients with encephalitis risk developing seizures
Wiley Periodicals, Inc. acutely, but also are more likely to develop later unpro-
ª 2012 International League Against Epilepsy voked seizures; the ILAE defines these as occurring

B. D. Michael and T. Solomon

>7 days after acute CNS infection (Berg et al., 2010).

Studies from Western industrialized countries show that
patients with encephalitis overall are about 16 times more
likely than the general population to develop later unpro-
voked seizures (Misra et al., 2008). Those with seizures
during the acute illness are up to 22 times more likely to
develop later unprovoked seizures than the general popu-
lation; even those with no acute seizures have a 10 times
greater risk. Encephalitis accounts for similar numbers of
new causes of epilepsy per year in the United States, as
does head injury, even though there are 1.4 million CNS
trauma cases per year, compared to fewer than 50,000
cases of CNS infection (Misra et al., 2008). If seizures
develop, it is usually within the first 5 years following the
encephalitic illness, but can occur up to 20 years later.
The pathogen causing the encephalitis appears important
in predicting the likelihood of later developing epilepsy.
For example, in those with La Crosse virus encephalitis,
the cumulative incidence may be 10–12%, whereas in
encephalitis caused by Nipah virus, this may be as low as
2.2% (Misra et al., 2008). Nevertheless, in countries where Figure 1.
there is limited access to antiepileptic drugs (AEDs) and Classical asymmetrical temporal lobe T2 hyperinten-
where seizures may have significant socioeconomic con- sity in a patient with herpes simplex virus type 1
sequences, postencephalitis epilepsy may present a con- encephalitis. (Courtesy of Dr Ian Turnball).
siderable lifelong burden. Epilepsia ILAE

‘‘Encephalitis’’ encompasses a broad range of patho- et al., 2010; Solomon et al., 2012). This equates to
physiologic processes that result in inflammation of the approximately 700 cases per year in the United Kingdom;
brain parenchyma (Solomon et al., 2001; Michael et al., in the United States, it corresponds to approximately
2010; Kneen et al., 2012; Solomon et al., 2012). Strictly, 19,000 hospitalizations and 1,400 deaths per year (Misra
therefore, the diagnosis is established only by histopatho- et al., 2008). Encephalitis is relatively rare but is important
logic examination of brain tissue. However, tissue is avail- for two reasons; first, encephalitis is often suspected even
able only for those patients who have a postmortem or in though many cases are not confirmed, and second, delays
the minority where biopsy is clinically justified antemor- in diagnosis and in starting treatment can cause significant
tem. Therefore, proxy markers of brain inflammation are morbidity and mortality. In encephalitis due to herpes sim-
routinely used (Kneen et al., 2012; Solomon et al., 2012). plex virus (HSV) type 1, those untreated or treated late
Most important of these is an elevated cerebrospinal fluid have a mortality as high as 90%; this falls to 20–30% if
(CSF) leukocyte count, although neuroimaging may also treatment is started early. If treatment starts before there is
be supportive. significant clouding of consciousness, the mortality may
Encephalitis typically presents with an encephalopa- be zero (Solomon et al., 2007, 2012). There are no detailed
thy syndrome, that is, altered consciousness. However, epidemiologic data on encephalitis in many lower-income
encephalopathy has a broad differential diagnosis, countries, where paradoxically, the global burden of dis-
including systemic metabolic disturbance and infection ease is most heavy (Misra et al., 2008).
outside the CNS (Michael et al., 2010). The absence of
evidence of inflammation on imaging or in the CSF
helps to distinguish these from encephalitis (Solomon
Etiology of Encephalitis
et al., 2012). Infection is by far the most common cause of encephali-
tis, either by direct CNS infection or through parainfec-
tious or postinfectious immune-mediated processes
Epidemiology (Kneen et al., 2012; Solomon et al., 2012). An immune-
Incidence mediated encephalitis can also occur as part of a paraneo-
The incidence of encephalitis is reported variably as plastic disease, with specific antibodies directed against
between 0.7 and 13.8 per 100,000 per year (Michael CNS antigens. Some forms of antibody-associated

Epilepsia, 53(Suppl. 4):63–71, 2012

doi: 10.1111/j.1528-1167.2012.03615.x
Seizures and Encephalitis

The most commonly identified sporadic cause of viral
encephalitis is herpes simplex virus; this is usually type 1,
although 10% of cases are type 2; it is particularly impor-
tant in neonates where encephalitis may be as part of a dis-
seminated infection (Michael et al., 2010; Solomon et al.,
2012). Other important causes of sporadic viral encephali-
tis include the varicella zoster virus and enteroviruses.
Immunocompromised people risk infection from a wider
range of pathogens, including viruses such as cytomegalo-
virus, and small intracellular bacteria and parasites such as
A Toxoplasma gondii (Solomon et al., 2012; Kneen et al.,
2012). Moreover, immunocompromised patients who
develop CNS infections from common CNS pathogens
may present subacutely or chronically (Solomon et al.,
2007; Solomon et al., 2012).
Epidemic encephalitis is usually due to arthropod-borne
viruses (arboviruses), which vary greatly by geographic
distribution. Globally, the most important epidemic cause
is Japanese encephalitis virus (Solomon et al., 2001); many
others are of local importance, such as West Nile virus,
tick-borne encephalitis virus and Nipah virus (Solomon
et al., 2007). Table 1 summarizes the global distribution of
B causes of epidemic viral encephalitis (Table 1).

HSV encephalitis
HSV encephalitis has an annual incidence of 1 in
250,000–500,000 (Solomon et al., 2007; Michael et al.,
2010). HSV is an alpha herpes DNA virus; most people
are exposed to HSV type 1 during childhood, and almost
everybody has been infected by adulthood (Whitley,
2006). The virus is transmitted by droplet spread and it
then crosses the mucous membrane of the nasal and oral
cavity. From there, it travels by retrograde axonal trans-
port, using the cellular cytoskeleton, along the trigeminal
C nerve to the trigeminal ganglion, where it establishes
Figure 2. latency. Periodically, the virus reactivates and travels by
Immunohistochemical staining of brain sections in a anterograde axonal transport to be shed; this is usually
patient who died of herpes simplex encephalitis asymptomatic, but can manifest as herpes labialis (a cold
showing (A) evidence of herpes simplex infection, (B) sore) in a small proportion (Whitley, 2006). The patho-
diffuse microglial activation, and (C) perivascular physiologic processes underlying latency and reactivation
inflammatory infiltrates. (Courtesy of Dr. Daniel are not fully understood, but toll-like receptors and the
Crooks, Walton Centre NHS Foundation Trust, resultant interferon response are probably important.
United Kingdom. Rarely, the virus replicates in the brain, causing encephalitis;
Epilepsia ILAE it is not clear if this follows directly on from further retro-
grade axonal transport after reactivation in the trigeminal
ganglion, or whether this is due to reactivation of virus
already latent within the brain. Indeed, a proportion of
encephalitis can also develop as part of a primary auto- people who die without any previous neurologic disease
immune process, without neoplasia (Vincent et al., have HSV nucleic acid within the brain parenchyma. HSV
2004). However, this is relatively uncommon compared to type 1 encephalitis has a bimodal age distribution, with
infective encephalitis, and the exact underlying patho- peaks in incidence in young adults and the elderly (Whitley,
genic processes are unknown (Granerod et al., 2010). 2006; Solomon et al., 2007).
Acute infections causing encephalitis group broadly into Seizures occur in approximately 40% of patients with
those occurring sporadically or epidemically. encephalitis due to HSV type 1 and may be the presenting
Epilepsia, 53(Suppl. 4):63–71, 2012
doi: 10.1111/j.1528-1167.2012.03615.x
B. D. Michael and T. Solomon

Table 1. Causes of acute viral encephalitis, with geographic clues

Groups Viruses Comments
Sporadic causes
(not geographically
restricted) listed by group
Herpes viruses Herpes simplex virus type 1 Most commonly diagnosed sporadic encephalitis
(family Herpesviridae) Herpes simplex virus type 2 Causes meningitis in adults (esp. recurrent); meningoencephalitis
occurs typically in the immunocompromised. Also causes a radiculitis
Varicella zoster virus Postinfective cerebellitis, or acute infective encephalitis or
Epstein–Barr virus Encephalitis in the immunocompromised
Cytomegalovirus Encephalitis in the immunocompromised; also retinitis
or radiculitis; often neutrophilic CSF with low glucose
Human herpesviruses 6 and 7 Febrile convulsions in children (after roseola); encephalitis in
Enteroviruses (family Picornaviridae) Enterovirus 70 Epidemic hemorrhagic conjunctivitis, with CNS involvement
Enterovirus 71 Epidemic hand foot and mouth disease, with aseptic meningitis,
brainstem encephalitis, myelitis
Poliovirus Myelitis
Coxsackie viruses, Mostly aseptic meningitis
Echoviruses, parechovirus
Paramyxoviruses (family Measles virus Causes acute postinfectious encephalitis, subacute encephalitis
Paramyxoviridae) and subacute sclerosing panencephalitis
Mumps virus Parotitis, orchitis or pancreatitis may occur before, during or after
Others (rarer causes) Influenza viruses, adenovirus,
parvovirus B19, lymphocytic
choriomeningitis virus, rubella virus
Arthropod-borne and zoonotic
Flaviviruses (family Flaviviridae) West Nile virus North America, Southern Europe, Africa, Middle East, West and
Central Asia associated with flaccid paralysis and parkinsonian
movement disorders
Japanese encephalitis virus Asia, associated with flaccid paralysis and parkinsonian
movement disorders
Tick-borne encephalitis virus Travel in Eastern Europe, Former USSR; tick bite;
upper limb flaccid paralysis
Dengue viruses (types 1–4) Causes fever, arthralgia, rash and hemorrhagic disease,
occasional CNS disease
Alphaviruses (family Togaviridae) Western, Eastern and Venezuelan Found in the Americas; encephalitis of horses and humans
equine encephalitis viruses
Chikungunya virus Asia Pacific, Africa
Bunyaviruses La Crosse virus Encephalitis in America
Coltiviruses Colorado tick fever virus North America
Rhabdoviruses Rabies, virus other lyssaviruses Non–arthropod-borne zoonotic viruses transmitted by dogs,
cats, bats, depending on location
Chandipura virus Transmitted by sandflies, causing outbreaks in India
Henipaviruses Nipah virus Transmitted in faeces of fruit bats in Malaysia, Bangladesh

Most are zoonotic, that is, animals rather than humans are the main natural hosts, the exceptions being dengue and
chikungunya viruses. Note viral causes of chronic encephalitis such as JC viruses are not included.

symptom (Whitley, 2006; Solomon et al., 2007). The tra- 2010; Solomon et al., 2012). These include alterations in
ditional view was that herpes simplex virus encephalitis cognition, consciousness, personality, or behavior, in the
(HSE) might present with simple partial olfactory sei- context of concurrent or recent febrile illness. The fever
zures, reflecting orbitofrontal cortex infection, or simple may be low-grade, and approximately 11% of HSE cases
and complex partial seizures, reflecting temporal lobe are afebrile on admission (Solomon et al., 2001; Granerod
infection; we now recognize that these occur only in the et al., 2010; Michael et al., 2010). Associated clinical fea-
minority and often later in the disease (Solomon et al., tures include headache, which may be severe, nausea, and
2007). Instead the focus has shifted to the early clinical vomiting, or features suggesting either meningism (neck
features, typically preceding any seizures (Michael et al., stiffness,) or raised intracranial pressure (e.g., papilledema)

Epilepsia, 53(Suppl. 4):63–71, 2012

doi: 10.1111/j.1528-1167.2012.03615.x
Seizures and Encephalitis

(Solomon et al., 2001; Granerod et al., 2010; Michael

Japanese encephalitis
et al., 2010).
Japanese encephalitis virus (JEV) is thought to infect
The relatively higher rate of acute symptomatic seizures
7 million people per year, with nearly 70,000 cases of
in patients with encephalitis due to HSV type 1 may be
encephalitis and at least 14,000 deaths, although these
result from the predilection of HSV to areas of high
may be underestimates (Solomon et al., 2008; Campbell
epileptogenic potential: the mesial temporal lobes, partic-
et al., 2011). Most cases are in South East Asia, but some
ularly the hippocampus, and, to a lesser extent, the orbito-
occur in the Western Pacific and Eastern Mediterranean
frontal cortices (Misra et al., 2008). This concurs with the
regions. JEV is an enzoonotic flavivirus transmitted by
findings that ex vivo rat brain tissue infected with HSV
Culex mosquitoes; the main cycle of transmission is
type 1 shows acute electrographic evidence of seizure
between wading birds and pigs, with the pigs acting as
activity and longer term hippocampal changes with neuro-
amplifying hosts. Humans are dead-end hosts, becoming
nal loss, predominantly in the cornu ammonis area 3
infected when bitten by a virus-carrying mosquito while
(Chen et al., 2004). Both these findings suggest that the
living in or traveling to endemic areas (Solomon et al.,
hippocampus is uniquely susceptible. In addition the nec-
2008). A mosquito bite results in an acute viremic illness,
rotizing nature of HSV encephalitis, the extent of leuko-
and in some people leads to the virus spreading across the
cyte infiltration, and the degree of cortical involvement
blood–brain barrier. Once in the brain parenchyma, there
may all be important in promoting seizures (Misra et al.,
is rapid viral replication in neurons. Leukocytes cross the
2008). Acute symptomatic seizures predict a worse out-
blood–brain barrier into the brain, leading histologically
come, as do a lower Glasgow coma scale score, delays in
to perivascular cuffing. It is unclear how much of the neu-
commencing acyclovir treatment, and older age (Misra
ronal injury results from direct viral cytopathy and how
et al., 2008; Whitley, 2006).
much is secondary to the inflammatory response. There is
Following HSV encephalitis, 40–65% of patients
currently no specific treatment for JEV encephalitis and
develop later unprovoked seizures (Misra et al., 2008;
so the management of its complications, such as seizures
Whitley, 2006). Treatment of seizures is similar to that in
and raised intracranial pressure, is paramount. Vaccina-
patients with other focal epilepsies. However, epilepsy
tion against JEV has advanced considerably in recent
following HSV encephalitis is typically refractory to
years, but delivering wider vaccination programs is diffi-
AEDs, often requiring combination treatment (Misra
cult in resource-poor settings; therefore, JEV encephalitis
et al., 2008). Early refractory epilepsy, particularly with
will likely remain an important disease for the foreseeable
progressive neurocognitive or neuropsychiatric impair-
future (Solomon et al., 2008).
ment, may indicate encephalitis relapse or chronic infec-
Seizures occur in 7–67% of JEV encephalitis cases (Mis-
tion, especially in those with immune compromise or
ra et al., 2008; Solomon et al., 2008); however, many stud-
suboptimal antiviral treatment. Although most relapses
ies have been limited by poor access to diagnostic
occur immediately following the acute encephalitic ill-
investigations. Furthermore, the seizure semiology in JEV
ness, some patients show progressive deterioration due to
encephalitis is often subtle and therefore easily missed if
chronic infection (De Tiege et al., 2003). Relapse of
there is limited access to EEG. Acute symptomatic seizures
encephalitis develops in 5–26% of patients, particularly
are more likely in younger patients with JEV encephalitis,
when the duration of acyclovir treatment is insufficient
those with reduced level of consciousness on admission,
(Whitley, 2006; Solomon et al., 2007). The recently pub-
and those with cortical inflammatory changes on neuroi-
lished United Kingdom guidelines recommend that all
maging (Solomon et al., 2001; Misra et al., 2008).
immunocompetent adults should receive 14 days of intra-
The development of seizures in JEV encephalitis corre-
venous acyclovir; they should then undergo repeat lumbar
lates with there being clinical signs of cerebral herniation,
puncture. If this CSF remains positive for HSV by poly-
high CSF opening pressure, and mortality (Solomon et al.,
merase chain reaction (PCR), acyclovir should continue
2001; Misra et al., 2008). In a prospective 3-year study of
for further 7 days cycles and the lumbar puncture repeated
144 adults and children with JEV encephalitis in Southern
at the end of each to demonstrate viral clearance before
Vietnam, 59 (41%) had acute symptomatic seizures; those
stopping the treatment (Solomon et al., 2012). Patients
with a witnessed seizure while an inpatient were at least
with immune compromise, or those aged between
four times more likely to have a poor outcome (odds ratio
3 months and 12 years, should receive 21 days of intrave-
[OR] 4.50, 95% confidence interval [95% CI] 1.94–10.52,
nous acyclovir before the lumbar puncture is repeated, but
p < 0.0001). In addition, those patients who developed
otherwise the approach for stopping acyclovir in children
status epilepticus were more likely to die than those with
is the same (Kneen et al., 2012; Solomon et al., 2012). Fur-
self-limiting seizures (11/25 [14%] vs. 0/15 [0%]
ther work is needed to delineate the complex mechanisms
p = 0.003). In this cohort, seizures were more common in
by which HSV establishes neurotropism, evades the host
children. Of interest, an isolated seizure before admission
immune system, and results in acute symptomatic seizures
was common and did not predict increased mortality. The
and later unprovoked epileptic seizures.
Epilepsia, 53(Suppl. 4):63–71, 2012
doi: 10.1111/j.1528-1167.2012.03615.x
B. D. Michael and T. Solomon

combination of symptoms for at least 7 days, seizures, a About 60% of patients with VGKC-complex disease
reduced Glasgow Coma Scale score, and signs of brain- have low serum sodium levels, and about 60% have hippo-
stem herniation had a composite positive predictive value campal high signal on MR scans; however, this may be
of 77% for poor outcome and a negative predictive value well localized and subtle (Irani et al., 2010a,b). CSF
of 91% (OR 35.88, 95% CI 12.46–108.15, p < 0.001) abnormalities are common with NMDA receptor antibody
(Solomon et al., 2001). Others have reported that seizures encephalitis, including a lymphocytosis and detectable
during encephalitis correlate with morbidity, but there is no antibodies; in those with VGKC-complex antibodies,
association with mortality overall; however, these are unpub- there are usually no significant CSF abnormalities and no
lished observations and the lack of statistical significance detectable CSF antibodies. Indeed, it is a moot point as to
may reflect smaller patient numbers (Misra et al., 2008). whether, by the strictest definition, these are truly exam-
The literature on the neurologic sequelae of JEV ples of encephalitis, or rather an encephalopathy. All
encephalitis has focused on neurocognitive impairments patients need appropriate imaging to identify an associ-
and movement disorders, particularly those reflecting sub- ated tumor. In women with NMDA receptor disease,
cortical involvement, for example, of basal ganglia and 20–50% have a tumor, almost always an ovarian teratoma;
thalamus. Another focus has been the ‘‘polio-like’’ flaccid in men the rates of neoplasia are lower (Buckley &
paralysis (Misra & Kalita, 1997; Solomon et al., 1998). Vincent, 2005; Dalmau et al., 2011). In those with
We clearly need more detailed cohort studies of the VGKC-complex disease, up to 10% have a tumor, usually
longer-term sequelae of this important cause of viral a thymoma or small cell lung cancer (Buckley & Vincent,
encephalitis, with a particular focus on late unprovoked 2005; Irani et al., 2010a,b).
seizures. The optimum treatment regimens are still uncertain for
these recently identified conditions. In the acute phase,
Antibody-associated encephalitis immunomodulatory therapy with corticosteroids and
In patients with an acute—or, more commonly, sub- either intravenous immunoglobulin or plasma exchange is
acute—onset of encephalitis, immune-mediated encepha- advised (Kneen et al., 2012; Solomon et al., 2012). Sei-
litis should be considered, particularly as the treatment is zures (which can lead to status epilepticus) should be
very different and early intervention may significantly managed as described below.
improve outcome (Buckley & Vincent, 2005; Irani et al., The important point is that a recurrence or worsening in
2010a,b. seizures may reflect relapse or progression of the disease;
The two most frequently identified causes of antibody- in such cases, antibody titers should be reassessed and, in
associated encephalitis are those due to antibodies to the consultation with an appropriate specialist, more aggres-
N-methyl-D-aspartate (NMDA) receptor and (voltage- sive immunomodulatory therapy considered. Although
gated potassium channel) VGKC-complex. There are relapse is uncommon in those with VGKC antibodies, up
important clinical differences between these conditions. to one third of patients with NMDA receptor antibodies
For VGKC-complex antibody-associated disease, the may relapse, even those without a tumor. Such patients
median age at presentation is 65 years and the male-to- may need long-term immunosuppression and annual
female ratio is 2:1 (Buckley & Vincent, 2005), although tumor screening (Kneen et al., 2012; Solomon et al.,
children are now beginning to be identified. In NMDA 2012).
receptor-associated disease, the median age at presenta-
tion is 25 years and the male-to-female ratio is 1:2.
Patients with VGKC-complex disease do not usually
Management of Seizures
have a fever or headache, whereas both of these are com- Acute symptomatic seizures
mon in those with NMDA receptor antibodies; indeed With such a wide range in incidence of acute sympto-
many have a febrile flu-like prodrome. Both can present matic seizures in encephalitis, there is currently no evi-
with seizures, often refractory to treatment, as well as dence to support routine prophylactic AEDs in all patients
confusion, amnesia, and psychosis. However, NMDA with encephalitis (Misra et al., 2008; Kneen et al., 2012;
receptor disease often has a second phase of presentation, Solomon et al., 2012). This is particularly the case in set-
days to weeks later, in which there are involuntary move- tings with limited resources to manage the sedative conse-
ments, including orofacial dyskinesias and choreoatheto- quences of some AEDs. However, if a high-risk group for
sis. Faciobrachial dystonic seizures were recently acute symptomatic seizures were identifiable, then pro-
described in patients with VGKC-complex disease and phylactic AEDs might be justified. Furthermore, if there is
are associated with leucine-rich glioma inactivated 1 an excitotoxic component responsible for the increased
(Lgi1) antibodies (Irani et al., 2010a,b). Indeed, faciobra- incidence of morbidity and mortality in this group, then
chial dystonic seizures appear pathognomonic for this this gives a stronger argument for using AEDs prophylac-
condition, sometimes preceding the other clinical fea- tically. However, there have been only a few and poor
tures by several weeks. quality studies to provide data by which clinicians can
Epilepsia, 53(Suppl. 4):63–71, 2012
doi: 10.1111/j.1528-1167.2012.03615.x
Seizures and Encephalitis

reliably identify such a high-risk group. Until we have quently refractory to AEDs and therefore many will
data from large multicenter stratified randomized con- require combination therapy or neurosurgery to attempt to
trolled trials, we cannot recommend routine prophylactic control seizures. Patients and their relatives should be
AEDs (Misra et al., 2008; Kneen et al., 2012; Solomon made aware of support groups, such as the Encephalitis
et al., 2012). Nevertheless, all patients should be managed Society (, which can provide
in a clinical setting where neurologic observations can be great support to cope with the sequelae of encephalitis
performed so that, should seizures develop, they can be (Kneen et al., 2012; Solomon et al., 2012).
treated promptly (Kneen et al., 2012; Solomon et al.,
2012). Convulsive status epilepticus is treated in the same
way as when due to other causes, with stabilization of the
Potential Pathogenic
patient’s airway, oxygenation, and administration of intra- Mechanisms of Seizures
venous lorazepam, or rectal diazepam if intravenous in Encephalitis
access is not available. There is little evidence to guide the There are several potential mechanisms for the develop-
AED choice as second-line therapy. Patients who do not ment of seizures in viral encephalitis. In particular, there
respond to these treatments should be sedated and intubated is growing support for the role of the extensive parainfec-
as for other causes of status epilepticus. tious inflammatory response (Winter et al., 2004; Kamei
The paradox of managing seizures associated with et al., 2009). The pathophysiologic mechanisms driving
encephalitis is that the developing world, where there is the the inflammatory response have yet to be fully elucidated,
greatest global burden of disease, has the least sufficient but there is growing evidence that the innate cytokine/
services to manage these patients (Misra et al., 2008). chemokine responses are important. Moreover, these
inflammatory peptides, which are produced by neurons,
Differential diagnoses not to miss astrocytes, and microglia, are increasingly recognized to
When managing a patient with encephalitis who promote excitatory neurotransmitter release and conse-
develops seizures, it is important not to miss parainfec- quent depolarization. In addition, depolarization itself can
tious processes, which could potentially cause the sei- result in the release of certain proinflammatory cytokines,
zure. These include hyponatremia and cerebral venous in the context of costimulation (Tsakiri et al., 2008).
sinus thrombosis (Solomon et al., 2007). In patients with Therefore, a dynamic interplay between cytokines,
acute encephalitis with reduced level of consciousness, chemokines, neurotransmitters, and other factors is likely
particularly if there are fluctuations or subtle repetitive to be important.
motor features, clinicians need a low threshold for
arranging EEG to exclude complex partial status epilep-
ticus or subtle motor status epilepticus, as these require Potential novel roles for encephalitis in epilepsy
specific treatment (Kneen et al., 2012; Solomon et al., pathogenesis: Human herpesvirus 6 and
2012), febrile seizures
Human herpesvirus 6 (HHV-6) is a relatively recently
Management of epilepsy following encephalitis isolated ubiquitous, lymphotrophic beta herpes DNA
The treatment of postencephalitic epilepsy depends on virus. It is the most common cause of roseola infantum, a
the natural history of the disease. For example the seizure febrile rash syndrome, in children younger than 2 years of
frequency and rates of relapse may vary greatly, and there age (Campadelli-Fiume et al., 1999). HHV6 achieves
are no clear predictors of postencephalitic epilepsy (Solo- 90–100% seroprevalence following transmission from the
mon et al., 2007; Misra et al., 2008). Two randomized con- mother and possibly siblings, predominantly during the
trolled trials showed that acyclovir reduces the short-term first 2 years of life (Yamanishi, 1992; Hall et al., 1994;
mortality from encephalitis. However, there are few data Campadelli-Fiume et al., 1999; Zerr et al., 2005). It is
to clarify the role and timing of administering acyclovir, now known to account for 20% of emergency depart-
or other antivirals, in preventing long-term sequelae such ment admissions for febrile illnesses in those aged
as postencephalitic epilepsy (Skoldenberg et al., 1984; 6–12 months, of whom approximately 13% develop sei-
Whitley et al., 1986). It is clear that we urgently require zures (Hall et al., 1994). HHV6 infection accounts for one
further investigation into the association between CNS third of febrile seizures in patients younger than 2 years;
infections and epilepsy to assess who is at risk, what are there is serologic persistence in approximately 66% and
the structural correlates, what is the clinical course/prog- reactivation in 6–16% as soon as 1–2 years later (Studahl
nosis, and what is the effect of antimicrobial and antiepi- et al., 2000). HHV6 is present in the CSF in approximately
leptic treatment. 10% of cases of suspected viral CNS disease, regardless of
As for other causes of epilepsy, AED treatment should age (Studahl et al., 2000). Moreover, following primary
be directed toward the seizure semiology. However, infection, approximately 60% of individuals have viral
patients with epilepsy following encephalitis are fre- DNA detectable in CSF as the sole source of latency,

Epilepsia, 53(Suppl. 4):63–71, 2012

doi: 10.1111/j.1528-1167.2012.03615.x
B. D. Michael and T. Solomon

suggesting that this is a highly neurotrophic virus (Chen & Campadelli-Fiume G, Mirandola P, Menotti L. (1999) Human herpes
virus 6: an emerging pathogen. Emerg Infect Dis 5:353–366.
Hudnall, 2006). Campbell GL, Hills SL, Fischer M, Jacobson JA, Hoke CH, Hombach
PCR detects HHV6 viral DNA in 50–69% of surgical JM, Marfin AA, Solomon T, Tsai TF, Tsu VD, Ginsburg AS. (2011)
resection specimens for mesial temporal lobe epilepsy, Estimated global incidence of Japanese encephalitis: a systematic
review. Bull World Health Organ 89:766–774.
predominantly the variant HHV6B subtype (Donati et al., Chen T, Hudnall SD. (2006) Anatomical mapping of human herpesvirus
2003; Fotheringham et al., 2007). In these studies, there reservoirs of infection. Mod Pathol 1:726–737.
was no HHV6 DNA in control specimens of neocortical Chen SF, Huang CC, Wu HM, Chen SH, Liang YC, Hsu KS. (2004) Sei-
zure, neuron loss, and mossy fiber sprouting in herpes simplex virus
temporal lobe epilepsy and autopsy. However, these type 1–infected organotypic hippocampal cultures. Epilepsia
studies recruited only 8–16 patients. In the largest and 45:322–332.
most recent study, including 33 cases of mesial temporal Dalmau J, Lancaster E, Martinez-Hernandez E, Rosenfeld MR, Balice-
Gordon R. (2011) Clinical experience and laboratory investigations
lobe epilepsy and 7 nonepileptic controls, HHV6 viral in patients with anti-NMDAR encephalitis. Lancet Neurol 10:63–74.
DNA was identified by PCR in 3 of the 33 mesial temporal De Tiege X, Rozenberg F, Des Portes V, Lobut JB, Lebon P, Ponsot G,
lobe cases (Karatas et al., 2008). Two of the National Insti- Heron B. (2003) Herpes simplex encephalitis relapses in children:
differentiation of two neurologic entities. Neurology 61:241–243.
tute for Neurology and Stoke Key Benchmarks for Donati D, Akhyani N, Fogdell-Hahn A, Cermelli C, Cassiani-Ingoni R,
research into epilepsy are to identify infectious etiologic Vortmeyer A, Heiss JD, Cogen P, Gaillard WD, Sato S, Theodore
risk factors for epilepsy and the pathologic development WH, Jacobson S. (2003) Detection of human herpes virus-6 in mesial
temporal lobe epilepsy surgical brain resections. Neurology 61:1405–
of epileptogenic foci. In this effort, HHV6 is clearly a 1411.
pathogen that justifies further research. Fotheringham J, Donati D, Akhyani N, Fogdell-Hahn A, Vortmeyer A,
Heiss JD, Williams E, Weinstein S, Bruce DA, Gaillard WD, Sato S,
Theodore WH, Jacobson S. (2007) Association of human herpes
Conclusions virus-6B with mesial temporal lobe epilepsy. PLoS Med 4:e180.
Granerod J, Ambrose HE, Davies NW, Clewley JP, Walsh AL, Mor-
Seizures are important in encephalitis, both in the acute gan D, Cunningham R, Zuckerman M, Mutton KJ, Solomon T,
Ward KN, Lunn MP, Irani SR, Vincent A, Brown DW, Crow-
presentation and the long-term neurologic sequelae. The croft NS; UK Health Protection Agency (HPA) Aetiology of
exact risks of developing seizures are poorly understood, Encephalitis Study Group. (2010) Causes of encephalitis and differ-
but appear to relate to the pathogen, the degree of cortical ences in their clinical presentations in England: a multicentre, popu-
lation-based prospective study. Lancet Infect Dis 10:835–844.
involvement, and the cytokine-mediated inflammatory Hall CB, Long CE, Schnabel KC, Caserta MT, McIntyre KM, Costanzo
response. This is clearly an area of great clinical impor- MA, Knott A, Dewhurst S, Insel RA, Epstein LG. (1994) Human Her-
tance requiring further investigation. However, there are pesvirus-6 infection in children. A prospective study of complications
and reactivation. N Engl J Med 331:8.
barriers to undertaking this work, particularly in resource- Irani SR, Bera K, Waters P, Zuliani L, Maxwell S, Zandi MS, Friese MA,
poor settings, which must be addressed if we are to Galea I, Kullmann DM, Beeson D, Lang B, Bien CG, Vincent A.
improve our understanding of the pathophysiologic pro- (2010a) N-methyl-D-aspartate antibody encephalitis: temporal pro-
gression of clinical and paraclinical observations in a predominantly
cesses underlying this and ultimately to translate this into non-paraneoplastic disorder of both sexes. Brain 133:1655–1667.
meaningful benefits for our patients. Irani SR, Michell AW, Lang B, Pettingill P, Waters P, Johnson MR,
Schott JM, Armstrong RJE, Zagami AS, Bleasel A, Somerville ER,
Smith SMJ, Vincent A. (2010b) Faciobrachial dystonic seizures pre-
Acknowledgments cede Lgi1-antibody limbic encephalitis. Ann Neurol 69:892–900.
Kamei S, Taira N, Ishihara M, Sekizawa T, Morita A, Miki K, Shiota H,
The authors acknowledge the on-going work of the members of Brain Kannno A, Suzuki Y, Mizutani T, Itoyama Y, Morishima T, Hirawy-
Infections United Kingdom ( whose anagi K. (2009) Prognostic value of cerebrospinal fluid cytokine
work led up to this article. changes in herpes simplex encephalitis. Cytokine 46:187–193.
Karatas H, Gurer G, Pinar A, Soylemezoglu F, Tezel GG, Hascelik G,
Akalan N, Tuncer S, Ciger A, Saygi S. (2008) Investigation of HSV1,
Disclosures HSV2, CMV, HHV6 and HHV 8 by real time PCR in surgical resec-
tion materials of epilepsy patients with mesial temporal lob sclerosis.
The authors have no conflicts of interest to declare. We confirm that J Neurol Sci 264:151–156.
we have read the Journal’s position on issues involved in ethical publica- Kneen R, Michael BD, Menson E, Mehta B, Easton A, Hemmingway C,
tion and affirm that this report is consistent with those guidelines. Klapper PE, Vincent A, Lim M, Carrol E, Solomon T; National Ence-
phalitis Guidelines Development and Stakeholder Groups. (2012)
Management of suspected viral encephalitis in children – Association
of British Neurologists and British Paediatric Allergy, Immunology
References and Infection Group national guidelines. J Infect 64:449–477.
Michael BD, Sidhu M, Stoeter D, Roberts M, Beeching NJ, Wilkins E,
Bauer J, Bien CG. (2009) Encephalitis and epilepsy. Semin Immunopa- Hart I, Kneen R, Miller A, Solomon T. (2010) The epidemiology and
thol 31:537–544. management of adults with suspected central nervous system infec-
Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross JH, van Emde tions – a retrospective cohort study in the NHS northwest region.
Boas W, Engel J, French J, Glauser TA, Mathern GW, Moshe SL, Q J Med 103:749–758.
Nordli D, Plouin P, Scheffer ZE. (2010) International League Against Misra UK, Kalita J. (1997) Anterior horn cells are also involved in
Epilepsy Revised terminology and concepts for organization of sei- Japanese encephalitis. Acta Neurol Scand 71:441–459.
zures and epilepsies: report of the ILAE Commission on Classifica- Misra UK, Tin CT, Kalita J. (2008) Viral encephalitis and epilepsy.
tion and Terminology, 2005–2009. Epilepsia 51:676–685. Epilepsia 49(Suppl. 6):13–18.
Buckley C, Vincent A. (2005) Autoimmune channelopathies. Nat Clin Skoldenberg B, Forsgren M, Alestig K, Burman L, Dahlqvist E, Forkman
Pract Neurol 1:22–33. A, Frydn A, Lçvgren K, Norlin K. (1984) Acyclovir versus vidarabine

Epilepsia, 53(Suppl. 4):63–71, 2012

doi: 10.1111/j.1528-1167.2012.03615.x
Seizures and Encephalitis

in herpes simplex encephalitis. Randomised multicentre study in con- tation between alpha, beta and gamma herpesviruses. Scand J Infect
secutive Swedish patients. Lancet 2:707–711. Dis 32:237–248.
Solomon T, Whitley RJ. (2004) Arthropod-borne viral encephalitides. Tsakiri N, Kimber I, Rothwell NJ, Pinteaux E. (2008) Mechanisms of
In Scheld M, Whitley RJ, Marra C (Eds) Infections of the central ner- interleukin-6 synthesis and release induced by interleukin-1 and cell
vous system. Lippincott Williams and Wilkins, Philidelphia, PA, depolarisation in neurones. Mol Cell Neurosci 37:110–118.
pp. 205758230. Vincent A, Buckley C, Schott JM, Baker I, Dewar BK, Detert N, Clover
Solomon T, Kneen R, Dung NM, Khanh VC, Thuy TTN, Ha DQ, Day L, Parkinson A, Bien CG, Omer S, Lang B, Rossor MN, Palace J.
NPJ, Nisalak A, Vaughn DW, White NJ. (1998) Poliomyelitis-like (2004) Potassium channel antibody-associated encephalopathy: a
illness due to Japanese encephalitis virus. Lancet 351:1094–1097. potentially immunotherapy-responsive form of limbic encephalitis.
Solomon T, Dung MN, Kneen R, Thao LTT, Gainsborough M, Nisalak Brain 127:701–712.
A, Day NPJ, Kirkham FJ, Vaughn DW, Smith S, White NJ. (2001) Whitley RJ. (2006) Herpes simplex encephalitis: adolescents and adults.
Seizures and raised intracranial pressure in Vietnamese patients with Antiviral Res 71:141–148.
Japanese encephalitis. Brain 125:1084–1093. Whitley RJ, Alford CA, Hirsch MS, Schooley RT, Luby JP, Aoki FY,
Solomon T, Hart IJ, Beeching NJ. (2007) Viral Encephalitis: a clinician’s Hanley D, Nahmias AJ, Soong SJ. (1986) Vidarabine versus acyclo-
guide. Pract Neurol 7:288–305. vir therapy in herpes simplex encephalitis. N Engl J Med 314:
Solomon T, Thao TT, Lewthwaite P, Ooi MH, Kneen R, Dung NM, White N. 144–149.
(2008) A cohort study to assess the new WHO Japanese encephalitis Winter PDN, Loan HT, Kneen R, Wills B, Thu LT, House D, White NJ,
surveillance standards. Bull World Health Organ 86:178–186. Farrar JJ, Hart A, Solomon T. (2004) Proinflammatory cytokines and
Solomon T, Michael BD, Smith PE, Sanderson F, Davies NWS, Hart IJ, chemokines an humans with Japanese encephalitis. J Infect Dis
Buckley C, Holland M, Easton A, Kneen R, Beeching NJ, On behalf 190:1618–1626.
of the National Encephalitis Guidelines Development Group. (2012) Yamanishi K. (1992) Pathogenesis of human herpes virus 6. Infect Agents
Management of suspected viral encephalitis in adults: Association of Dis 1:149–155.
British Neurologists and British Infection Association National Zerr DM, Meier AS, Selke SS, Frenkel LM, Huang ML, Wald A, Rhoads
Guideline. J Infect 64:347–373. MP, Nguy L, Bornemann R, Morrow RA, Corey L. (2005) A popula-
Studahl M, Hagberg L, Rekabdar E, Bergstrçm T. (2000) Herpesvirus tion based study of primary human herpes virus 6 infection. N Engl J
DNA detection in cerebrospinal fluid: differences in clinical presen- Med 352:768–776.

Epilepsia, 53(Suppl. 4):63–71, 2012

doi: 10.1111/j.1528-1167.2012.03615.x