0268_0272_Kennedy_JON_1770 16.03.

2005 12:13 Uhr Seite 268

J Neurol (2005) 252 : 268–272

DOI 10.1007/s00415-005-0770-7 ENS TEACHING REVIEW

Peter G. E. Kennedy Viral encephalitis

Received: 19 November 2004
Accepted: 24 November 2004
Published online: 11 March 2005
Prof. Peter G. E. Kennedy, MD, PhD, DSc,
FRSE, FMedSci
Department of Neurology
Division of Clinical Neurosciences
Southern General Hospital
Institute of Neurological Sciences
Glasgow G51 4TF, Scotland, UK
E-Mail: P.G.Kennedy@clinmed.gla.ac.uk
■ Abstract Acute viral encephali-
tis may be caused by a wide range
of viruses but the most important
is herpes simplex encephalitis
(HSE) because of its severity, espe-
cially if untreated, and its good re-
sponse to specific treatment with
acyclovir. The outcome of any CNS
viral infection is dependent on
both the immune status of the host
and the virulence of the infecting
virus. In evaluating a patient with
suspected viral encephalitis there
are 3 essential steps, namely the
identification of a true parenchy-
mal virus infection of the brain
rather than a non-infective en-
cephalopathy, the distinction of an
infectious viral encephalitis from
an acute disseminated en-
cephalomyelitis (ADEM), and then
the determination, where possible,
of the specific virus involved. In
practice, the precise viral cause of
the encephalitis may never be es-
tablished. Analysis of the CSF for
herpes simplex virus (HSV) DNA
using the Polymerase Chain Reac-
tion (PCR) has been a significant
advance in the diagnosis of HSE as
this test has a very high sensitivity
and specificity especially with ap-
propriate sample timing. It is es-
sential to commence early treat-
ment with intravenous acyclovir in
patients suspected of having HSE
because of the remarkable safety
and efficacy of this drug and the
dangers of delaying potentially ef-
fective treatment of life threatening
disease. This review outlines the
general management approach in
patients suspected of having viral
encephalitis.
■ Key words viral encephalitis ·
acyclovir · herpes simplex virus ·
CSF PCR

predisposes the individual to infection with particular

General aspects of viral encephalitis
Viral encephalitis occurs as a result of a viral infection
of the brain parenchyma, usually accompanied by a
meningitis, thereby producing a meningoencephalitis.
The usual scenario is an acute illness with a diffuse in-
flammatory process which can seriously affect brain
function, sometimes irreversibly. The result of a viral in-
sult to the central nervous system (CNS) is determined
by a combination of the host immune response and the
JON 1770
herpes viruses such as varicella-zoster virus (VZV) and
cytomegalovirus (CMV), and organ transplant patients
are more likely to develop listeria infection than im-
munocompetent individuals. Viruses also show differ-
ences in their propensity to infect the CNS and to infect
particular brain cells. The latter property, known as neu-
rotropism, provides the basis for certain clinical pheno-
types shown by different viruses. For example, po-
liovirus is neurotropic for spinal cord anterior horn
cells, resulting in a lower motor neuron presentation of

neurovirulence of the infecting virus [8]. Thus a state of poliomyelitis. The JC virus, a papovavirus, has an affin-
host immunosuppression, due, for example, to AIDS, ity for oligodendrocytes, the myelin-producing glial cell
0268_0272_Kennedy_JON_1770 16.03.2005 12:13 Uhr Seite 269
within the CNS, resulting in multifocal demyelination
[8]. Determining the molecular mechanisms of viral
neurotropism is important in understanding the neu-
ropathogenesis of viral encephalitis. The host and viral
factors determining the outcome of viral infection are
likely to be complex, and a recent experimental study of
lethal HSV-1 infection in mice has elegantly demon-
strated that Toll-like receptors such as TLR2 play an im-
portant role in the induction of inflammatory cytokine
response to HSV-1 which is important in determining
the inflammatory reaction in the virally infected brain
[13]. Perhaps similar mechanisms might also operate in
human patients with HSE. It is important to recognise
that another form of encephalitis is not due to direct vi-
ral infection of the brain parenchyma but to an alter-
ation of normal immune function following a recent
vaccination or viral infection; this type is known as
acute disseminated encephalomyelitis (ADEM) [5, 6, 8].
There are numerous potential causes of viral en-
cephalitis [3, 10], and some of these are shown in Table 1
which is not exhaustive. It should be appreciated that the
cause of a particular episode if viral encephalitis is not
always determined unequivocally, with probably
30–50 % of cases being of unknown cause based on evi-
dence to date [2]. In the author’s experience, however, a
50 % success rate of virus identification during virus en-
cephalitis is probably optimistic. Insight into this issue
was recently provided by a large study from Finland
where PCR was carried out on the CSF of over 3000 pa-
tients who had CNS infections including encephalitis,
meningitis and myelitis [12]. It was found, that VZV in-
fection, the cause of chicken pox and herpes zoster, oc-
curred in 29 %, herpes simplex virus (HSV) and en-
teroviruses each accounted for 11 % of cases, and
influenza A virus was detected in 9 % of the samples.
One assumes a cause-and-effect relationship between
the virus and the disease although this may not neces-
sarily be the case in every situation. Epidemiological es-
timates of certain viruses have been made, and a fre-
Table 1 Some Important Causes of Viral Encephalitis*
Herpes simplex virus (HSV-1, HSV-2)
Other Herpes viruses – Varicella-Zoster virus (VZV), Cytomegalovirus (CMV),
Epstein-Barr virus (EBV), Human Herpes virus 6 (HHV6)
Adenoviruses
Influenza A
Enteroviruses. Poliovirus
Measles, Mumps and Rubella viruses
Rabies
Arboviruses eg Japanese B Encephalitis, West Nile encephalitis virus, tick-borne
encephalitis viruses
Bunyaviruses eg La Crosse strain of California virus
Reoviruses eg Colorado tick fever virus
* Table modified from refs 3 and 10
269
quently quoted incidence of herpes simplex encephalitis
(HSE), the most frequent cause of fatal sporadic en-
cephalitis in humans, is about 1 case per million per year
[11]. However, in the author’s opinion this is probably an
underestimate, and there is some recent evidence, al-
though based on a small number of cases, to suggest that
milder forms of HSE may occur, possibly in up to 20 %
of cases and caused by HSV-2, and are more common in
the immunosuppressed patient [7]. About 2000 cases of
HSE are said to occur in the USA annually [8]. The inci-
dence of ‘cold sores’ in patients developing HSE is no dif-
ferent from the rest of the population [8]. Pathway speci-
ficity in HSE is probably more important than viral
cell-type susceptibility in the brain. Whether the path-
way used by HSV-1 to the fronto-temporal regions of the
brain is primarily via the olfactory pathway, or some
other mechanism, is as yet unclear [8]. It is important to
keep in mind the wide spectrum of viral encephalitis
which may occur globally. For example, Japanese en-
cephalitis causes 10–15000 deaths annually [14]. There
has also been great interest over the last few years in
‘emerging infections’ of the CNS such as West Nile en-
cephalitis due to transport of a neurovirulent viral
strain from Israel to the USA, and Nipah virus en-
cephalitis which resulted from a paramyxovirus cross-
ing species barriers from bat to pig to humans [9].
Diagnostic approach to a patient with suspected
viral encephalitis
There are essentially three components to the approach
to a patient with suspected viral encephalitis, sum-
marised in Table 2.
The first requirement is to distinguish between a true
acute encephalitis and an encephalopathy due to any
one of a multitude of non-infective conditions which
can mimic a viral encephalitis. The latter include such
metabolic conditions as liver failure, renal failure,
diabetic coma and mitochondrial cytopathies, anoxia/
Table 2 Summary of Diagnostic Approach to Viral Encephalitis
Distinguish infective encephalitis from encephalopathy
• History eg onset, known metabolic problem
• Signs eg fever, focal neurology
• Investigations eg bloods, CSF, EEG, MRI/CT scan.
Distinguish infectious encephalitis from ADEM
• History eg age, previous vaccination, illness
• Signs eg visual loss, spinal cord signs
• Investigations eg blood, CSF, MRI
Determine if possible the specific viral cause of acute encephalitis
• History eg prodrome, pattern of onset, recent travel, immunosuppression,
occupation
• Signs eg focal neurology, peripheral signs
• Investigations eg bloods, CSF, EEG, MRI/CT
0268_0272_Kennedy_JON_1770 16.03.2005 12:13 Uhr Seite 270
270
brain ischaemia, systemic infection, toxic states, parane-
oplastic disorders, critical illness, nutritional deficiency,
malignant hypertension, traumatic brain injury and
non-convulsive status in epilepsy [3]. There are no hard
and fast rules for making this distinction which can
sometimes be difficult, but the assessment is based on a
number of different clues which the physician considers
in their entirety to make a judgement. In general terms
a viral encephalitis is typified by headache, fever and
clouding of consciousness together with possible
seizures and focal neurology. Factors which would tend
to make one think of an acute encephalitis rather than
an encephalopathy would include a history of headache,
fever, focal neurological signs, focal seizures, a CSF
pleiocytosis, focal abnormalities seen on EEG and focal
changes seen on either CT or MRI [6]. In an en-
cephalopathy one would not expect to detect focal signs
or focal changes on EEG or neuroimaging and the CSF is
usually acellular.
Having established that the patient does indeed have
an encephalitis, it is then necessary to distinguish an in-
fective viral encephalitis from ADEM. Again, it is the
combination of typical features in both scenarios that
can be used as clues to reach a diagnosis. Factors that
would tend to point towards a diagnosis of ADEM in-
clude a history of recent vaccination in children, a pro-
dromal illness, visual impairment in one or both eyes,
spinal cord signs, and multifocal white matter changes
seen in both hemispheres seen on MRI [6]. A CSF lym-
phocytic pleiocytosis with elevated protein and normal
glucose would be common in both ADEM and infective
viral encephalitis.
After ruling out as far as possible a diagnosis of
ADEM, it is then necessary to attempt to identify the
most likely cause of the virus encephalitis, but as has
been indicated above this is often not possible in prac-
tice.Again the approach is straightforward, consisting of
taking a full history, carrying out a systemic and CNS ex-
amination and then performing appropriate general
and neurological investigations. One of the aims of this
approach is also to exclude the possibility of any one of
numerous possible non-viral causes of acute encephali-
tis which are summarised in Table 3.
There are a variety of clues from the history and pre-
sentation that may be helpful. The mode of onset of the
illness should be noted. The most important cause of vi-
ral encephalitis, HSE, has a characteristic abrupt onset
and then rapidly progresses, whereas enterovirus en-
cephalitis tends to be biphasic [10]. A history of recent
travel may be highly relevant, especially if the patient
has recently returned from an area where a particular
infection is known to be endemic eg Africa in the case of
malaria, particular regions in the US and Europe in the
case of Lyme disease, and Asia in the case of Japanese
Encephalitis [3, 10]. Recent animal bites may be relevant
in the case of rabies or tick borne encephalitis, as may
Table 3 Some Non-Viral Causes of Infectious Encephalitis
Bacterial
Mycobacterium tuberculosis
Mycoplasma pneumoniae
Listeria monocytogenes
Borrelia burgdorferi
Brucellosis
Leptospirosis
Legionella
Tropheryma whippeli (Whipple’s disease)
Nocardia actinomyces
Treponema pallidum
Salmonella typhi
All causes of pyogenic meningitis
Rickettsial
Rocky Mountain spotted fever
Endemic and epidemic typhus
Q fever
Ehrlichiosis
Fungal
Cryptococcus
Aspergillosis
Candidiasis
Parasitic
Human African Trypanosomiasis
Cerebral Malaria
Toxoplasma gondii
Schistosomiasis
Table modified from refs 3 and 10
the patient’s occupation, geographical location and the
season during which the illness presents. The last named
is particularly important in the case of arbovirus infec-
tions.
The examination may also reveal important clues al-
though in the author’s experience this is not always the
case. Systemic examination should include a search for
skin lesions or other general signs of possible relevance.
If the neurological examination reveals the presence of
focal features, especially fronto-temporal signs, then this
is suggestive of HSE, especially if there is also a history
of generalised and/or focal seizures. If there is evidence
of peripheral nervous system involvement then this
points towards the possibility of ADEM or infection with
certain herpes viruses such as CMV, EBV and VZV, all of
which have been reported to produce a combination of
central and peripheral nervous system signs [10].
The investigations in suspected acute viral en-
cephalitis may be helpful in trying to establish a diag-
nosis. General investigations such as routine haematol-
ogy and biochemistry must always be performed in
order to exclude an obvious but unsuspected systemic
disturbance which may be masquerading as an acute
neurological illness. A chest radiograph is also useful to
exclude such conditions as tuberculosis, legionella, my-
coplasma and malignancy [3, 10]. Serological tests may
0268_0272_Kennedy_JON_1770 16.03.2005 12:13 Uhr Seite 271
be helpful in providing evidence of a recent viral infec-
tion of relevance, but the time taken to perform these
may be a disadvantage in the context of making imme-
diate management decisions.
Cranial imaging with CT or, preferably MRI if it is
available, should be carried out as soon as possible after
presentation. MRI gives more detailed relevant informa-
tion than CT. It is vital to exclude a space-occupying le-
sion such as a brain tumour or abscess which would re-
quire immediate neurosurgical attention. In HSE the
typical MRI features include focal oedematous areas in
the temporal lobes and the orbital surface of the frontal
lobes, the abnormal areas sometimes enhancing with
gadolinium [6, 11]. Evidence of raised intracranial pres-
sure with possible midline shift may also be evident in
HSE and may require treatment in its own right (see be-
low). An EEG is usually abnormal in viral encephalitis
and may show diffuse non-specific abnormalities or
typical focal ‘PLEDS’ (periodic lateralising epileptiform
discharges) in HSE, although they are not diagnostic for
this condition. Provided there is not a space-occupying
lesion or significant cerebral oedema with midline shift
then a lumbar puncture is indicated although it may oc-
casionally be normal in viral encephalitis. A CSF lym-
phocytic pleiocytosis of 10–200/mm3 with a protein in-
crease around 0.6–6 g/l can be expected [11]. CSF PCR
has been a significant advance in the diagnosis of viral
encephalitis and is particularly valuable in suspected
HSE. In an experienced laboratory, PCR to detect HSV-1
has a specificity of about 95 %. If the CSF sample is taken
between 48h and 10 days then the sensitivity of PCR is
also about 95 % [1, 11]. False negative results are most
Fig. 1 Suggested management flow chart for pa-
tients with suspected herpes simplex encephalitis
(Reprinted from reference 10)
271
likely within the first 24–48h after the illness, and then
10–14 days after the onset.
Management of viral encephalitis
General management of the acutely ill patient with viral
encephalitis will likely occur in a high dependency med-
ical unit, and will include the treatment of raised in-
tracranial pressure with IV mannitol and/or steroids,
control of seizures with appropriate anticonvulsants
and correction of fluid and electrolyte imbalances.
Where a non-viral diagnosis has been established spe-
cific anti-microbial therapy should be given where pos-
sible. Acyclovir treatment should be commenced as
soon as a diagnosis of HSE is a credible possibility. In-
travenous acyclovir at a dose of 10 mg/kg three times
daily should be continued for at least 14 days where the
diagnosis is certain, usually confirmed by CSF PCR. In
an immunosuppressed patient the duration of acyclovir
treatment should be extended to 21 days to help prevent
possible relapses [10, 11]. Once started, acyclovir should
only be discontinued if a definite alternative diagnosis
has been established. In the case of possible, but not def-
inite, HSE the author usually chooses the option of giv-
ing acyclovir for 10 days, with a further possible option
of giving oral therapy for another 4 days in such cases.
The author’s suggested management scheme for HSE is
shown in a flow chart in Fig. 1. Acyclovir is remarkably
safe, but renal function should nevertheless be carefully
monitored as renal toxicity may occur. Untreated, the
mortality rate of HSE is over 70 % [15], but acyclovir
0268_0272_Kennedy_JON_1770 16.03.2005 12:13 Uhr Seite 272

272

treatment reduces the mortality to 20–30 %, which is still
unacceptably high. There is current discussion as to
whether oral valacyclovir should be given at the end of
IV acyclovir treatment, and whether the CSF should be
re-examined following 14 days treatment to determine
if it is still positive for HSV-1 by PCR [4]. These issues
should be resolved unambiguously in a few years after
completion of the Collaborative Antiviral Study Group
trial. In this study, which is ongoing and will take several
years to complete, patients with PCR-proven HSE are
given standard intravenous acyclovir therapy for 14–21
days (14 days minimum) and at the end of this period a
further CSF sample is taken and analysed by PCR for
HSV DNA following which patients are randomised to
receive a further 90 days treatment with either oral vala-
cyclovir or a placebo. Neurological and neuropsycho-
logical follow-up testing then occurs both during the 90
days treatment and then annually for 5 years. This ex-
tremely well designed and important trial will establish
unequivocally whether giving a follow-up course of oral
valacyclovir is associated with a better outcome, and
also whether any patients on valacyclovir with a better
outcome than placebos are more or less likely to have
had a positive CSF PCR for HSV DNA after the initial
course of intravenous acyclovir. Effective as acyclovir is
for HSE, the overall morbidity and mortality seen in
treated patients is still far too high. There is a pressing
need to develop more effective antiviral regimes for pa-
tients with all forms of viral encephalitis.
■ Acknowledgement This review article is based on a lecture given
by the author at an ENS Teaching Course on June 26th 2004.

References
1. Baringer JR (2002) Herpes Simplex
virus encephalitis. In: Davis LE,
Kennedy PGE (eds) Infectious Dis-
eases of the Nervous System, 1st edn.
Butterworth-Heinemann, pp 139–164
2. Booss J, Esiri MM (2003) Viral En-
cephalitis in Humans. ASM Press,
Washington
3. Chaudhuri A, Kennedy PGE (2002) Di-
agnosis and treatment of viral en-
cephalitis. Postgrad Med J 78:575–583
4. Cinque P, Cleator GM, Weber T, et al.
(1996) The role of laboratory investi-
gation in the diagnosis and manage-
ment of patients with suspected her-
pes simplex encephalitis: a consensus
report. J Neurol Neurosurg Psychiatry
61:339–345
5. Coyle PK (2000) Postinfectious en-
cephalomyelitis. In: Davis LE, Kennedy
PGE (eds) Infectious diseases of the
Nervous System, 1st edn. Butterworth-
Heinemann, pp 83–108
6. Davis LE (2000) Diagnosis and treat-
ment of acute encephalitis. The Neu-
rologist 6:145–159
7. Fodor PA, Levin MJ, Weinberg A, et al.
(1998) Atypical herpes simplex virus
encephalitis diagnosed by PCR ampli-
fication of viral DNA from CSF. Neu-
rology 51:554–559
8. Johnston RT (1998) Viral infections of
the nervous system, 2nd edn. Lippin-
cott-Raven, Philadelphia
9. Johnson RT (2003) Emerging viral in-
fections of the nervous system. J Neu-
rovirol 9:140–147
10. Kennedy PGE (2004) Viral encephali-
tis: causes, differential diagnosis, and
management. J Neurol Neurosurg Psy-
chiatry 75(Suppl 10):i10–i15
11. Kennedy PGE, Chaudhuri A (2002)
Herpes simplex encephalitis. J Neurol
Neurosurg Psychiatry 73:237–238
12. Koskiniemi M, Rantalaiho T, Piipari-
nen H, et al. (2001) Infections of the
central nervous system of suspected
viral origin: A collaborative study from
Finland. J Neurovirol 7:400–408
13. Kurt-Jones EA, Chan M, Zhou S, Wang
J, Reed G, Bronson R, Arnold MM,
Knipe DM, Finberg RW (2004) Herpes
simplex virus 1 interaction with Toll-
like receptor 2 contributes to lethal en-
cephalitis. Proc Natl Acad Sci, USA
101:1315–1320
14. Solomon T (2003) Recent advances in
Japanese encephalitis. J Neurovirol
9:274–283
15. Whitley RJ, Gnann JW (2002) Viral en-
cephalitis: familiar infections and
emerging pathogens. Lancet
359:507–514