Special article

Journal of the Intensive Care Society

2015, Vol. 16(4) 330–338
! The Intensive Care Society 2015
Managing acute central nervous system Reprints and permissions:
sagepub.co.uk/
infections in the UK adult intensive care journalsPermissions.nav
DOI: 10.1177/1751143715587927
unit in the wake of UK encephalitis jics.sagepub.com

guidelines

DJ Stoeter1, BD Michael2,3, T Solomon2,3 and L Poole1

Abstract
The acute central nervous system infections meningitis and encephalitis commonly require management on intensive
care units. The clinical features often overlap and in the acute phase–altered consciousness and seizures may also need to
be managed. In April 2012, the first UK national guideline for the management of suspected viral encephalitis was
published by the British Infection Association and Association of British Neurologists, and other key stakeholders,
and included a simple management algorithm. The new guideline results from evidence demonstrating a number of
common oversights in the standard management of suspected viral encephalitis in many settings. In combination with
British Infection Association meningitis guidelines, evidence-based approaches now exist to facilitate the non-expert
managing patients with suspected central nervous system infections. Here we bring together these guidelines and the
supporting evidence applicable for intensivists into a single resource.

Keywords
Encephalitis, intensive care, critical care, central nervous system infection, guidelines, acute, adult

because delays are associated with a significant

Introduction: defining the problem
Infective meningitis and encephalitis are the most
common forms of suspected central nervous system
(CNS) infections that present to intensivists.1 Less
commonly CNS infections cause intracranial abscess,
cystic disease (such as neurocysticercosis), or lead to
cerebrovascular sequelae such as the diffuse micro-
vascular occlusion of ‘cerebral malaria’ or a vasculitis
complicating meningitis or encephalitis).2,3 Other
causes of CNS inflammation, such as antibody-
associated autoimmune encephalitis, are increasingly
recognised and should be considered in those with
appropriate clinical features.4
In the western world, the incidence in adults is
estimated to be 0.6–4/100,000/year for bacterial men-
ingitis and 5.2–7.6/100,000/year for viral meningitis.
Encephalitis demonstrates a similar incidence of 2.73–
8.66/100,000/year.2,5–10 Whilst cases of proven CNS
infection are relatively uncommon, suspected cases
are common; in some series approximately four
patients are routinely investigated for every one case
diagnosed.9 Nevertheless, a low index of suspicion,
with urgent investigation and treatment, is required
increase in morbidity and mortality.9–17
Infective meningitis is predominantly due to viral
or bacterial infection, with viral forms usually follow-
ing a more benign course.18 In contrast, in the
30–63% of cases for whom a cause is identified,
encephalitis is usually viral and results in serious
sequelae.19 In the UK, of the cases in whom a viral
cause is identified, 90% are due to herpes simplex
(HSV), varicella zoster (VZV) and enteroviruses.5
With the launch of the British meningitis C and
pneumococcal vaccination programmes and the British
Infection Association (BIA) meningitis guidelines and
1
Department of Intensive Care, Royal Liverpool University Hospital,
Liverpool, UK
2
Institute of Infection and Global Health, and NIHR Health Protection
Research Unit in Emerging and Zoonotic Infections, University of
Liverpool, UK
3
Walton Centre NHS Foundation Trust, Liverpool, UK
Corresponding author:
DJ Stoeter, Department of Intensive Care, Royal Liverpool University
Hospital, Prescot Street, Liverpool, L7 8XP, UK.
Email: davidstoeter@nhs.net
Stoeter et al.
algorithm in 2003, clinicians now appear to be aware of
the importance of urgent investigation and treatment for
meningitis.9,17,20 Mortality rises from 7% to 26% if anti-
biotics are not administered within the immediate hours
following admission, findings mirrored in the intensive
care cohort of patients with bacterial meningitis.16,17,21
Perhaps less widely recognised is that delays in treatment
for viral encephalitis dramatically increases morbidity
and mortality, particularly if >48 hours from admis-
sion.12,13 Indeed current guidelines recommend treat-
ment within 6 hours of admission.8 Time to antiviral
therapy represents the single most important modifiable
risk factor for poor outcome.12,13 Without aciclovir treat-
ment, mortality from HSV encephalitis is at least 70%
and only 3% of patients survive without sequelae.7
However, with timely aciclovir, the mortality is reduced
to 10–20% and 40–50% survive without sequelae.7
Nevertheless, comparative studies have demonstrated
that there are significantly longer delays in suspicion of
the diagnosis, delays in investigation and delays in treat-
ment in comparison to suspected meningitis, largely due
to a lack of awareness of standards of management.9
With an incidence that parallels bacterial meningi-
tis, more severe sequelae if treatment is delayed, and
evidence that existing management often falls short of
best practice, a national guideline has been developed
for the non-specialist (summarized in an algorithm:
Figure 1).8,9,16 As many of these patients will require
intensive care input, it is important that intensivists
are well versed in how to manage these patients.
Clinical features
A thorough assessment of clinical features seeks to
answer three questions:
1. Could this be a CNS infection?
2. If a CNS infection, what is the aetiology?
3. Can a diagnostic lumbar puncture (LP) be per-
formed safely to establish the aetiology?
1. The classical clinical features of encephalitis include
headache, altered mental state, focal neurological
deficits and seizures in the context of fever or
recent febrile illness. However, these clinical fea-
tures have been shown to have too low a negative
predictive value alone to reliably exclude CNS
infection.8,9 Much of the ambiguity arises from
patients presenting with encephalopathy: the clin-
ical syndrome of altered mental status manifest as
altered consciousness, cognition, personality or
behaviour, which has a very wide differential diag-
nosis. Psychiatric phenomena are not uncommon
in infective and non-infective encephalitis and can
be misleading early on in the illness.8
Encephalopathy may also be erroneously attribu-
ted to drugs or alcohol.7 This ambiguity leads to a
significantly longer delay to suspicion and investi-
gation as compared with meningitis especially in
331
the elderly for whom altered mental state is often
attributed to systemic sepsis but in whom the inci-
dence of HSV encephalitis is higher, given the
bimodal distribution.7,8 Fever at admission is
often used to help rule-in CNS infection under
these circumstances. However, fever is absent in
10% of patients with HSV encephalitis.12–14 A his-
tory of a prodromal febrile illness or subsequent
documented fever may be useful where fever is
absent at initial assessment.8 Notably, of patients
with proven encephalitis a proportion have diar-
rhoea, vomiting, urinary or respiratory symp-
toms.4 Therefore, investigation should be
undertaken to exclude CNS infection even if
these symptoms are present, unless there is an
established extra-CNS infection with significant
complications (e.g. hypoxia or severe metabolic
acidosis) to explain encephalopathy.7 Infection
outside of the CNS should not cause significant
encephalopathy in an otherwise fit individual
unless significant complications have arisen.
Owing to the clinical (and histopathological) over-
lap that may occur between cases of meningitis and
encephalitis, both may present in a similar manner,
giving rise to the concept of meningoencephalitis.
This is especially true in the intensive care setting
since a patient with meningitis referred for higher
level care will often have been referred for the same
reasons as a patient with encephalitis.
2. Clinical features may point towards the potential
aetiology of encephalitis or meningitis. Foreign
travel history must be established due to the geo-
graphical restrictions of many pathogens and sub-
sequent investigation should be guided by expert
advice.8 This will also include urgent investigation
for common diseases that mimic encephalitis.
Malaria is the most common and geographically
widespread example, and is important to consider
as outcomes are also related to time to commen-
cing antimalarials.22
Identifying patients with immunocompromise
broadens the number of potential pathogens and
HIV itself can cause an acute meningitis or encephal-
itis at seroconversion or a more insidious dementing
illness characterized by psychosis and seizures.8,23
Therefore, the British HIV Association (BHIVA)
guideline, supported by the ABN/BIA encephalitis
guideline, recommend HIV testing in all patients pre-
senting with encephalitis or meningitis.8,23
Finally, subacute presentation and atypical fea-
tures (such as extrapyramidal signs e.g. dystonias
involving the face and arm, or choreoathetosis) sug-
gest an autoimmune cause.8,24,25
3. The role of clinical features in determining the
safety of a LP is outlined below.
332 Journal of the Intensive Care Society 16(4)

Figure 1. National guidelines algorithm for the management of suspected viral encephalitis.
Reproduced from the Association of British Neurologists and British Infection Association National Guidelines with permission from
Elsevier.8

advise that an LP is performed urgently unless clear

Initial investigation and management
CSF analysis is by far the most useful tool in estab-
lishing or excluding meningitis and encephalitis from
the list of potential diagnoses. National guidelines
clinical contraindications are present.8 Initial CSF
findings of opening pressure, white cell count
(WCC) and differential, protein and CSF: blood glu-
cose ratio provide quick evidence for or against a
Stoeter et al.
CNS infection and direct treatment towards a viral or
bacterial pathogen. Moreover, early CSF Gram stain
has a sensitivity and specificity of 60–90% and 97%,
respectively, for bacterial meningitis, and subsequent
culture can also identify antibiotic sensitivities; CSF
PCR has an even greater sensitivity and specificity in
excess of 95% for viral encephalitis, when performed
in the appropriate timeframe.8
CSF findings in viral encephalitis are generally
more modest than those seen in bacterial meningitis.
CSF pressures are usually only modestly elevated in
viral encephalitis in contrast to very high pressures in
up to 40% of those with bacterial meningitis.3,8
In viral infection the CSF WCC is around
10–100  106/L and lymphocyte predominant in con-
trast to 100–10,000  106/L and polymorph predom-
inant in bacterial meningitis.3,8 A normal WCC may
occur within the first few days of illness in viral
encephalitis (and in those with immunocompromise,
in whom cell counts cannot be relied upon).5 Equally
a raised WCC with polymorph predominance more
suggestive of bacterial disease may occur early on in
viral encephalitis.8 Early, atypical findings usually
become typical on repeat LP 24–48 hours later.
Conversely, a lymphocyte predominance with
modest cell counts can occur in bacterial disease: bru-
cellosis, listeriosis, tuberculosis and partially treated
bacterial meningitis.8 In these circumstances, except
for partially treated bacterial meningitis, CSF lactate
<2 mmol/l has been reported to be useful to exclude
bacterial disease with a negative predictive value
(NPV) of 94–96%.7,26,27 Prior treatment with anti-
biotics, particularly for prolonged periods, signifi-
cantly reduces this NPV and in one study produces
a sharp decline in culture positivity to zero at 8 hours,
although bacterial PCR remains reliable and cost-
effective for up to 96 hours.20,28,29
Viral PCR may be negative in early encephalitis.
However, the PCR is likely to be positive if the LP is
repeated 24–48 hours later, if still within 2–10 days of
onset of the illness.8 PCR requests should be for HSV,
VZV and enteroviruses. If there is immunocompro-
mise or foreign travel, early expert advice should be
sought to guide other requests. In patients presenting
more than 10 days after illness onset, CSF: serum
viral antibody ratio (Rieber’s formula) is calculated
using CSF and blood samples paired in time to dem-
onstrate intrathecal synthesis of virus-specific anti-
body to provide post hoc evidence of CNS infection
(‘paired oligoclonical bands’ in the algorithm).8
Beyond 48 hours of treatment with aciclovir, levels
of microbial DNA/RNA fall and PCR results may be
reported as ‘low levels’. Under these circumstances,
CSF: serum antibody ratios may again be useful to
clarify the diagnosis after day 10 of onset of illness.
Some have raised questions around the risk of cere-
bral herniation when performing LP in critically ill
patients with suspected CNS infection.30 The inci-
dence of brain herniation occurring in patients with
333
bacterial meningitis (with or without lumbar puncture
as part of their management) is estimated at around
5%.30 There is little data to support a corresponding
incidence for encephalitis.8 The incidence of hernia-
tion following LP as an investigation for any path-
ology, in which a temporal but not necessarily
causal relationship is demonstrated, is estimated by
a number of authors at around 1%.30–32 However,
this is based on case series and retrospective cohort
studies many of which predate the widespread avail-
ability of CT and consist of a heterogeneous group of
patients with both infectious and non-infectious
aetiologies, and includes many with features of
raised intracranial pressure (ICP).30–32
Clinical features have been demonstrated to be at
least as accurate at predicting this risk as CT imaging
of the brain.32 These clinical contraindications are
listed in Figure 1 alongside those indicating cardio-
respiratory instability making LP impractical, and
coagulopathy increasing the risk of a vertebral canal
haematoma. Given that most intensive care patients
are likely to have such clinical features, almost all will
require CT and/or stabilisation prior to LP according
to the guidelines. A number of small studies of
patients with bacterial meningitis suggest the sensitiv-
ity of CT for warning impending herniation after LP
is surprisingly low and varies between 20% and
80%.30 As such, the guidelines advise performing LP
on a case by case decision in those with clinical signs
suggestive of possible brain shift causing or caused by
raised ICP contraindicating LP but with a normal CT.
Given the high mortality and morbidity associated
with inadequately treated CNS infection, particularly
in those already requiring intensive care support, the
potential benefits of performing the LP may outweigh
the potential risks in many with clinical signs suggest-
ive of possible brain shift but a normal CT scan.
Particularly so that, in an era of increasing antibiotic
resistance, targeted antimicrobial therapy can be
given.
CT scan changes consistent with encephalitis are
unreliable early on, for example a lesion consistent
with HSV encephalitis only occurs in 20–85%.8
A second scan will almost always show an abnormality
but may take up to a week to develop. In HSV enceph-
alitis this typically involves the temporal lobes. The
primary role of CT is in quickly identifying alternative
diagnoses (e.g. intracranial bleeding), particularly
where history is unavailable. In those patients who
demonstrate hydrocephalus, investigation for add-
itional pathogens should be undertaken, particularly
TB and Cryptococcus, when there is the appropriate
travel/exposure history and/or immunocompromise.33
Diffusion-weighted magnetic resonance imaging
(MRI) within 48 hours (ideally within 24 hours) of
admission is recommended in the encephalitis guide-
lines.8 It is markedly more sensitive than CT within
this time frame for detecting focal changes consistent
with encephalitis, for example HSV encephalitis
334
lesions are present in 90%, and other subtle focal
changes may help quickly redirect investigation and
treatment where CT is normal and basic initial CSF
investigations are equivocal.8 A neuroradiologist
opinion may be of value. The guidelines advise that
all patients with suspected encephalitis should be
referred to the neurology team within 24 hours of
admission and this will facilitate such interactions at
a time when decisions about further investigation and
treatment are being made based on initial findings.8
The same neuroprotective targets and strategies
used to prevent secondary brain injury in traumatic
brain injury are recommended for all patients (such
as maintaining a mean arterial pressure of 80 or
more to maintain cerebral perfusion pressure).5,34,35
Acute glucose control avoiding >10 mmol/l, amongst
patients with CNS infections, potentially reduces the
burden of functional disability in patients who sur-
vive.36 High-grade fever appears to be an independent
risk factor for poor outcome, increased length of stay
and mortality in a range of neurocritical care
patients.37 Randomised controlled trials are lacking
in this patient cohort to strongly recommend targeted
temperature management (to normothermia) with sur-
face or intravascular cooling devices but many experts
advocate it.5 ICP monitoring is not routinely recom-
mended even if there are clinical signs of raised pres-
sure.34 Repeat CT imaging and early consultation with
a neurosurgical unit is recommended if there are clin-
ical signs of raised ICP with a view to monitoring and
medical and/or surgical management targeting main-
tenance of cerebral perfusion pressure.34 No strong evi-
dence exists for this approach improving mortality or
morbidity and this recommendation is based on case
reports and series, and extrapolation from limited stu-
dies in patients with meningitis. Some advocate decom-
pressive craniectomy in those with rapid non-dominant
hemispheric swelling as a cohort who may respond well
to such surgical management.5,35 Indications for inten-
sive care admissions are similar as for any neurologic-
ally obtunded patient – a depressed conscious level
(10–25% develop coma), status epilepticus, respiratory
failure from secondary lung insults such as aspiration
and severe electrolyte disturbances than can accom-
pany intracranial disease.5,35
Transfer to tertiary neurological units is considered
appropriate in the following circumstances: where
face-to-face specialist neurological review is not pos-
sible within 24 hours of admission; where access to
MRI and EEG (see below) is not available (EEG is
many hospitals remains unavailable); where signifi-
cant diagnostic delay occurs, is inconclusive or the
patient fails to respond to therapy.8 There is convin-
cing evidence that neurocritical care units both reduce
mortality and improve functional outcomes in
patients with neurological disease, as compared with
the general ICU.38
It is clear that rising to these standards (specifically
expert neurological input, MRI and EEG facilities)
Journal of the Intensive Care Society 16(4)
remains a problem. A recent ABN survey of acute
neurology services in UK based on standards set by
a 2011 Royal College of Physicians publication
demonstrated that the average availability of an
acute expert neurological opinion remains as low as
30% of days, with MRI and EEG access at best
around 22% and 52% of hospitals, respectively, in
most district general hospitals.39 Wide service provi-
sion variation exists across the country and barriers
include fundings systems that fail to recognize multi-
disciplinary involvement in patient caseloads; an
existing unbalanced focus of resources on neurology
outpatient services; limited access MRI and EEG
facilities including equipment and training facilitating
MRI in critically ill patients (MRI safe and condi-
tional equipment), consultant neurologist numbers
across the UK and tertiary neurological unit bed
availability.39 Neuro-intensive care facilities in many
regions have large neurosurgical patient burdens
which may not infrequently present a barrier to trans-
fer to tertiary units. The recent 2014–2015
Department of Health Mandate sets out a directive
to address these wide variations in service provision
from hospital to hospital.40
Further investigation and management
A repeat LP is indicated 24–48 hours after the first if
initial CSF findings are normal or atypical despite
clinical suspicion, particularly early in the disease.
Close liaison with experts in microbiology/virology/
infectious diseases is valuable at this stage to consider
whether any additional tests are worth performing.
Seizures and status epilepticus, including non-con-
vulsive status (CNS infections are one of the most
common causes), can occur in a proportion of
patients with CNS infections.24 However, there is no
evidence to support primary prophylaxis.24 The
approach is the same as for any case of status.5
Relatively strong evidence favours lorazepam as first
line, followed by a non-sedating maintenance drug,
even if the first-line agent terminates the seizure.35,41
These second-line options are phenytoin, valproate or
levetiracetam, with more evidence available for the
former two.42 Status epilepticus that is refractory to
these medications should be managed in the ICU set-
ting with infusions of sedatives. No one sedative is
favoured by evidence. Where available, 24 hours of
EEG-monitored deep sedation, with evidence of burst
suppression or no EEG evidence of seizures, can be
followed by a wake-up trial.42 EEG is of particular
use in the diagnosis and establishment of seizure-free-
dom, in those with non-convulsive (NCSE) or subtle-
motor status epilepticus7.
EEG has two important roles in the ICU setting.
Firstly, determining the presence or absence of seizure
activity in a patient failing to wake from a sedation
break.24 In addition, a normal EEG suggests a non-
organic primary psychiatric cause for failure to wake.8
Stoeter et al. 335

Interestingly, in a small retrospective study of 236
ICU patients with coma due to a number of
common aetiologies ranging from anoxic-ischaemic
encephalopathy to complications of alcohol abuse,
8% demonstrated EEG features of NCSE suggesting
EEG is an underutilized tool in ICU.43 Encephalitis is
one of the most common causes of NCSE.5
With the expansion in the availability of blood
serum testing, it is increasingly important to exclude
non-infectious mimics of CNS infections, particularly
autoimmune encephalitis (Table 1).5 This can occur
de novo or be a paraneoplastic or post-infectious
phenomenon. The most common form is limbic
encephalitis associated with either anti-N-methyl-D-
aspartate (NMDA) receptor or anti-voltage gated
potassium channel (VGKC)-complex antibodies.5,8
CSF protein is often normal with a normal WCC in
anti-VGKC cases and a modest lymphocytosis in anti-
NMDA cases.8 Other non-infectious encephalitides
are described in Table 1. Other mimics to consider
are CNS vasculitis and pseudomigraine with pleocy-
tosis.47 Encephalopathy or delirium (including ‘sepsis
associated encephalopathy’) amongst patients emer-
ging from severe, complicated critical illness of any
cause is well-recognised and can sometimes raise the
question of encephalitis even where the presenting ill-
ness was not neurological in nature. The clinical spec-
trum can range from acute to long-term and mild
cognitive deficits through to severe obtundation with
features of non-specific encephalopathy on EEG and
a variety of changes on MRI.48 It is a diagnosis of
exclusion.
Treatment
Just as initial empirical treatment exists for bacterial
meningitis, empirical treatment with aciclovir within 6
hours of admission is recommended based on the
most common treatable pathogens, HSV and VZV.8
As aforementioned, bacterial meningitis is

Table 1. Autoimmune encephalitides.3,5,8,24,25,44–46

Autoimmune
mechanism Name Distinguishing features Useful investigations

Antibodies to Anti-NMDA receptor -associated malignancy in 20–50%: usu- Serum anti-NMDA receptor
neuronal Limbic Encephalitis ally teratomas (often ovarian, hence antibodies
membrane female predominance)
proteins -Psychiatric phenomena
-orofacial dyskinesia and choreoatheo-
tosis
-autonomic dysfunction
Anti-VGKC-complex -associated malignancy in 10% (thym- Serum anti-VGKC-complex antibo-
Limbic Encephalitis oma or small-cell lung cancer) dies: Antibodies to LGI1 subunit
-psychiatric features typically cause CNS manifest-
-focal unilateral faciobrachial dystonic ations, whilst those targeting
seizures, may be pathognomic in contactin 2 associated protein
those with antibodies to the leucine- (CASPR2) predominantly cause
rich glioma 1 (LGI1) subunit peripheral nerve hyper-
-SiADH excitability
Bickerstaff encephalitis -brainstem features: ataxia, ophthalmo- Anti-GQ1b antibodies (a type of
plegia anti-ganglioside antibody)
-Guillain-Barre variant (post-infectious)
Antibodies to Paraneoplastic Limbic -features of limbic involvement: psychi- Onconeuronal antibodies:
intracellular encephalitides atric and memory deficits Anti-Hu
neuronal -may preceed malignancy Anti-Ma2
proteins Anti-CV2
Anti-amphyphysin
Anti-Ri
Associated with Hashimoto’s -encephalopathy alongside gait -Anti-thyroid peroxidase antibodies
antibodies but Encephalopathy abnormalities, tremor, seizures, psy- (thyroid function may still be
unknown chiatric phenomena normal)
mechanism -steroid responsive
No known Acute Disseminated -post-infectious Typical demyelinating lesions on
antibody Encephalomyelitis -associated myelopathy MRI
association -age <50 years
Rasmussen’s -may be post-infectious Typical MRI unilateral
Encephalitis -strict unilateral hemispheric hemispheric changes
features/seizures
336
concomitantly treated in these critically ill patients
according to BIA guidelines based on age and UK-
prevalent sensitivities amongst the major pathogens
until CSF results are available and conclusive.
Occasionally local sensitivities may apply and indicate
the need for the addition of vancomycin to ceftriax-
one to cover beta-lactam-resistant pneumococcal
strains. Early liaison with local microbiology services
is again advised. Comprehensive, new UK guidelines
for the management of meningitis are due to be pub-
lished later this year.
Of additional note, VZV may require higher doses
of aciclovir, for patients in whom renal function can
tolerate this. Due to bioavailability, the intravenous
route is the only route currently recommended as cap-
able of generating adequate CSF concentration, and
therefore there is currently no recommended role for
the enteral pro-drug valaciclovir. For patients with
suspected encephalitis who are immunocompromised,
the initial treatment approach with aciclovir is the
same, with the exception that early expert advice is
advised to guide additional investigation and treat-
ment for opportunistic infections. For example, if
cytomegalovirus is suspected or identified then gan-
ciclovir is recommended.8 However, no equivalent
antiviral therapy has been established for many
other forms of viral encephalitis, such as entero-
viruses, and therefore treatment is mainly supportive.
To determine when aciclovir can be stopped in
adults with HSV encephalitis a repeat LP is advised
after 14 days of treatment in the immunocompetent
and after 21 days in those with immune compromise.8
If the CSF remains positive for HSV by PCR, then it is
advised that the intravenous aciclovir is continued with
the LP repeated after 7-day intervals to determine that
the CSF has become negative for HSV by PCR.
Aciclovir can precipitate within the distal convoluted
tubule leading to crystal nephropathy and acute renal
failure in between 12% and 49% of patients.49 ICU
patients are at risk and in such cases volume resuscita-
tion to euvolaemia and loop diuretics are advised and
renal replacement therapy may be required, but in gen-
eral the nephropathy is reversible.49
Steroids are not recommended routinely early in
the management of suspected encephalitis or in
proven HSV encephalitis, although they may be con-
sidered for use by experts under specific circum-
stances, whilst the results of a current randomized
controlled trial are awaited. In other encephalitides,
steroids may be recommended, particularly when
there is a vasculitic component, for example in some
cases of VZV encephalitis.
For those with a travel history to a malaria-ende-
mic area, empirical antimalarial therapy should be
considered early. Again, expert advice should be
sought early to advise on any additional investigation
and empirical treatment.
Autoimmune encephalitis is treated with intraven-
ous methylprednisolone and/or immunoglobulins or
Journal of the Intensive Care Society 16(4)
plasmapharesis in the acute phase, with subsequent
adjunctive immunosuppressants, although no long-
term protocols have yet been established. Delays in
treatment of autoimmune encephalitis are important
in determining outcomes although the urgency is less
so than for viral encephalitis. For example, in the
largest series of patients with autoimmune encephal-
itis due to NMDA receptor antibodies, delays of >40
days to starting immune-suppression were associated
with a worse outcome in those with a primarily auto-
immune process, and delays of >3 months have been
associated with a worse outcome in those with para-
neoplastic disease. Nevertheless, as it may take 2–3
weeks from obtaining the serum sample to receiving
the antibody result, in many cases when an auto-
immune encephalitis is suspected, and an infectious
process has been excluded, treatment may need to
commence before the antibody results are known.50
Conclusions
CNS infections, meningitis and encephalitis are med-
ical emergencies requiring early identification and
treatment, often requiring intensive care input.
Significant clinical and histopathological overlap
exist and a broad differential often complicates assess-
ment in those with encephalitis. LP is pivotal to estab-
lishing a diagnosis and directing treatment towards
the responsible pathogen. The last 10–20 years has
seen dramatic improvements in the management of
meningitis with vaccination programmes and national
guidelines. It is hoped that the development of UK
national guidelines for encephalitis will see improve-
ments in the management of these conditions in which
delays in diagnosis and treatment are associated with
significant morbidity and mortality
Learning points
A low index of suspicion is required early amongst
critically ill patients presenting with encephalopathy
to allow prompt, simultaneous investigation for both
meningitis and encephalitis using UK management
algorithms for both.
BIA UK algorithms and guidelines for managing
meningitis and encephalitis are available at: http://
www.britishinfection.org/content/clinical-guidelines#
emmen
It is appropriate to refer patients with suspected
encephalitis to a neurologist within 24 hours of admis-
sion. Transfer to a specialist neurocritical care unit is
indicated where: access to a neurologist in <24 hours
is not possible; there is no access to MRI/EEG; diag-
nostic failure or poor response to therapy occurs.
For cases of encephalitis for whom a cause can be
identified (up to 63%), 90% are due to HSV, VZV
and enteroviruses.
CSF should be sent for standard microscopy/cul-
ture, biochemistry and a sample sent for HSV, VZV
Stoeter et al.
and enterovirus PCR. All patients should be tested
for HIV.
Immunocompromise and/or foreign travel should
prompt early discussion with a neurologist and
microbiologist to determine additional CSF testing.
CSF testing with normal/equivocal initial results
should be repeated at 24–48 hours to avoid false
negatives.
Subacute or atypical features (psychiatric phenom-
ena, orofacial dyskinesias, faciobrachial dystonic seiz-
ures or choreoathetosis) prompt consideration of
autoimmune encephalitis. The most common is
Limbic encephalitis due to anti-NMDA or to anti-
VGKC antibodies tested with serum blood samples.
These take 2–3 weeks to process. Treatment may com-
mence prior to acquisition of results and has good
outcomes.
Empirical treatment of adults with encephalitis is
with intravenous aciclovir 10 mg/kg (dose adjusted in
renal failure) within 6 hours of suspicion. Standard
medical techniques to maintain cerebral perfusion
pressure should be employed.
Patients with clinical or radiological signs of raised
ICP should be referred to a neurosurgeon although
routine ICP monitoring is not currently recom-
mended. Decompressive craniectomy may be con-
sidered in selected patients.
CNS infection is one of the most common causes of
non-convulsive status. EEG should be used in a patient
failing to wake from sedation or with refractory seiz-
ures to optimize anticonvulsant/sedative therapy.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of
this article.
Funding
The authors received no financial support for the research,
authorship, and/or publication of this article.
References
1. Beckham JD, and Tyler KL. Neuro-intensive care of
patients with acute CNS infections. Neurotherapeutics
2012; 9: 124–138.
2. van de Beek D, de Gans J, Tunkle AR, et al.
Community-acquired bacterial meningitis in adults. N
Engl J Med 2006; 354: 44–53.
3. Solomon T. Neurological infection. In M Donaghy (ed.)
Brain’s diseases of the nervous system. Oxford: Oxford
University Press, 2011, pp. 1327–1355.
4. Granerod J, Ambrose HE, Davies NWS, et al. Causes of
encephalitis and differences in their clinical presentations
in England: a multicentre, population-based prospective
study. Lancet Infect Dis 2010; 10: 835–844.
5. Kramer AH. Viral encephalitis in the ICU. Crit Care
Clin 2013; 29: 621–650.
6. Chadwick D. Viral Meningitis. Br Med Bull 2005; 75–76:
1–14.
337
7. Solomon T, Hart I, and Beeching NJ. Viral encephal-
itis: a clinician’s guide. Pract Neurol 2007; 7: 288–305.
8. Solomon T, Michael BD, Smith PE, et al. On behalf of
the National Encephalitis Guidelines Development and
Stakeholder Groups. Management of suspected viral
encephalitis in adults. Association of British
Neurologists and British Infection Association
National Guidelines. J Infect 2012; 64: 347–373.
9. Michael BD, Meneka S, Stoeter D, et al. The
epidemiology and management of adult suspected cen-
tral nervous system infections – a retrospective cohort
study in the NHS Northwest Region. Q J Med 2010;
103: 749–758.
10. Granerod J, Cousens S, Davies NWS, et al. New esti-
mates of the incidence of encephalitis in England.
Emerg Infect Dis 2013; 19: 1455–1462.
11. Proulx N, Frechette D, Toye B, et al. Delays in the
administration of antibiotics are associated with mor-
tality from adult acute bacterial meningitis. Q J Med
2005; 98: 291–298.
12. McGrath N, Anderson NE, Croxson MC, et al. Herpes
simplex encephalitis treated with acyclovir: diagnosis
and long-term outcome. J Neurol Neurosurg
Psychiatry 1997; 63: 321–326.
13. Raschilas F, Wolff M, Delatour F, et al. for the French
Herpes Simplex Encephalitis Study Group. Outcome of
and prognostic factors for herpes simplex encephalitis
in adult patients: results of a multicenter study. Clin
Infect Dis 2002; 35: 254–260.
14. Kennedy PGE. A retrospective analysis of forty-six
cases of herpes simplex encephalitis seen in Glasgow
between 1962 and 1985. Q J Med 1988; 255: 533–540.
15. Bell DJ, Suckling R, Rothburn MM, et al. Management
of suspected herpes simplex virus encephalitis in adults
in a UK teaching hospital. Clin Med 2009; 9: 1–5.
16. Dellinger RP, Levy MM, Rhodes A, et al. Surviving
Sepsis Campaign: International Guidelines for
Management of Severe Sepsis and Septic Shock: 2012.
Crit Care Med 2013; 41: 580–637.
17. Chadwick DR, and Lever AML. The impact of new
diagnostic methodologies in the management of menin-
gitis in adults at a teaching hospital. Q J Med 2002; 95:
663–670.
18. Chadwick DR. Viral meningitis. Br Med Bull 2005;
75–76: 1–14.
19. Whitley RJ. Herpes simplex encephalitis: adolescents
and adults. Antiviral Res 2006; 71: 141–148.
20. National Institute of Clinical Excellence. Bacterial
Meningitis and Meningococcal Septicaemia in children
2010, http://www.nice.org.uk/guidance/cg102/resources/
guidance-bacterial-meningitis-and-meningococcal-septic
aemia-pdf (accessed 25 October 2014).
21. Auburtin M, Wolff M, Charpentier J, et al. Detrimental
role of delayed antibiotic administration and penicillin-
nonsusceptible strains in adult intensive care unit
patients with pneumococcal meningitis: the
PNEUMOREA prospective multicenter study. Crit
Care Med 2006; 34: 2758–2765.
22. Lalloo DG, Shingadia D, Pasvol G, et al. UK malaria
treatment guidelines. J Infect 2007; 54: 111–121.
23. British HIV Association (BHIVA), British Association
for Sexual Health and HIV (BASHH), British Infection
Society (BIS). UK National Guidelines for HIV Testing,
338
http://www.bhiva.org/documents/guidelines/testing/
glineshivtest08.pdf (2008, accessed 25 October 2014).
24. Michael BD, and Solomon T. Seizures and encephalitis:
clinical features, management and potential pathophy-
siologic mechanisms. Epilepsia 2012; 53: 63–71.
25. Irani SR, Vincent A, and Schott JM. Autoimmune
encephalitis. Br Med J 2011; 342: d1918.
26. Sakushima K, Hayashino Y, Kawaguchi T, et al.
Diagnostic accuracy of cerebrospinal fluid lactate for
differentiating bacterial meningitis from aseptic menin-
gitis: a meta-analysis. J Infect 2011; 62: 255–262.
27. Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice
guidelines for the management of bacterial meningitis.
Clin Infect Dis 2004; 39: 1267–1284.
28. Saravolatz LD, Manzor O, VanderVelde N, et al.
Broad-range bacterial polymerase chain reaction for
early detection of bacterial meningitis. Clin Infect Dis
2003; 36: 40–45.
29. Michael BD, Menezes BF, Cunniffe J, et al. The effect
of delayed lumbar punctures on the diagnosis of acute
bacterial meningitis in adults. Emerg Med J 2010; 27:
433–438.
30. Joffe AR. Lumbar puncture and brain herniation in
acute bacterial meningitis: a review. J Intensive Care
Med 2007; 22: 194–207.
31. van Crevel H, Hijdra A, and de Gans J. Lumbar punc-
ture and the risk of herniation: when should we first
perform CT? J Neurol 2002; 249: 129–137.
32. Hasbun R, Abrahams J, Jekel J, et al. Computed tom-
ography of the head before lumbar puncture in adults
with suspected meningitis. N Engl J Med 2001; 345:
1727–1733.
33. Tunkel AR, Glaser CA, Bloch KC, et al. The manage-
ment of encephalitis: clinical practice guidelines by the
Infectious Diseases Society of America. Clin Infect Dis
2008; 47: 303–327.
34. Kumar G, Kalita J, and Misra UK. Raised intracranial
pressure in acute viral encephalitis. Clin Neurol
Neurosurg 2009; 111: 399–406.
35. Howard R, Kullmann D, and Hirsch N. Admission to
neurological intensive care: who, when, and why?
J Neurol Neurosurg Psychiatry 2003; 74: iii2–iii9.
36. Kramer AH, Roberts DJ, and Zygun DA. Optimal
glycemic control in neurocritical care patients: a
systematic review and meta-analysis. Crit Care 2012;
16: R203.
37. Diringer MN, Reaven NL, Funk SE, et al. Elevated
body temperature independently contributes to
Journal of the Intensive Care Society 16(4)
increased length of stay in neurologic intensive care
unit patients. Crit Care Med 2004; 32: 1489–1495.
38. Kramer AH, and Zygun DA. Do neurocritical care
units save lives? Measuring the impact of specialized
ICUs. Neurocrit Care 2011; 14: 329–333.
39. Fuller G and Lawrence J. ABN Acute Neurology ser-
vices survey 2014. Association of British Neurologists,
http://www.theabn.org/media/Acute%20Neurology
%20Survey%20-%20FINAL%20Dec14.pdf (accessed
10 March 2015).
40. Department of Health. The Mandate. A Mandate from
the Government to NHS England April 2014 to March
2015, http://www.gov.uk/government/publications/nhs-
mandate-2014-to-2015 (accessed 10 March 2015).
41. Pandey S, Rathore C and Michael BD. Antiepileptic
drugs for the primary and secondary prevention of seiz-
ures in viral encephalitis. Cochr Database Syst Rev
2014; 10: CD010247.
42. Brophy GM, Bell R, Claasen J, et al. Neurocritical Care
Society Status Epilepticus Guideline Writing Committee.
Guidelines for the Evaluation and Management of
Status Epilepticus. Neurocrit Care 2012; 17: 3–23.
43. Towne AR, Waterhouse EJ, Boggs JG, et al. Prevalence
of nonconvulsive status epilepticus in comatose
patients. Neurology 2000; 54: 340–345.
44. Sonneville R, Klein I, de Brouker T, et al. Post-
infectious encephalitis in adults: diagnosis and manage-
ment. J Infect 2009; 58: 321–328.
45. Bien CG, Widman G, Urbach H, et al. The natural
history of Rasmussen’s encephalitis. Brain 2002; 125:
1751–1759.
46. Castillo P, Woodruff B, Caselli R, et al. Steroid-respon-
sive encephalopathy associated with autoimmune thyr-
oiditis. Arch Neurol 2006; 63: 197–202.
47. Steiner I, Budka H, Chaudhuri A, et al. Viral enceph-
alitis: a review of diagnostic methods and guidelines for
management. Eur J Neurol 2005; 12: 331–343.
48. Widmann CN, and Heneka MT. Long-term cerebral
consequences of sepsis. Lancet Neurol 2014; 13:
630–636.
49. Perazella MA. Crystal-induced acute renal failure. Am J
Med 1999; 106: 459–465.
50. Dalmau J, Lancaster E, and Martinez-Hernandez E.
Clinical experience and laboratory investigations in
patients with anti-NMDAR encephalitis. Lancet
Neurol 2011; 10: 63–74.