Symposium: neurology
Evaluation of headaches
in children
Sourabh Mukhopadhyay
Catharine P White
Abstract
This review provides a practical guide to the common causes of headache
and their assessment in children. Contrary to popular belief, headaches
are very common in children. The primary headache disorders, which
include migraine and tension-type headache, account for the majority of
headaches, while secondary headache, i.e. those with underlying pathol-
ogy, are much less common. A thorough history and examination is the
key to determining the cause and should be the most important means
of reassuring the child and family that there is no serious cause for
the headaches. To manage childhood headache you need to be able to
distinguish the painful from the harmful, and therefore must recognize
the common headache patterns and the signs and symptoms that may
indicate serious intracranial disease. Most non-acute headaches do not
need further investigation. Neuroimaging is rarely necessary. Recurrent
headaches, of whatever cause, are a cause of considerable morbidity,
especially in terms of school absence.
Keywords acute headache; brain imaging; differential diagnosis;
­migraine; tension-type headache
Introduction
Headaches are common in children and the prevalence increases
with increasing age. In our practice, almost half of referrals from
primary care are because of headache. Unfortunately, most par-
ents think that headache is an uncommon symptom in children,
hence their understandable concern. As well as hoping to relieve
the pain, parents are often seeking reassurance that their child’s
headache is not a sign of serious intracranial disease, such as
a brain tumour. If this is understood, then we do not need to
explain every headache, but we must be able to reassure the
child and family that it is not a sign of serious illness.
Migraine and tension-type headache are by far the common-
est causes of headache.1 Other rarer causes include hemicrania
continua, cluster headaches, idiopathic intracranial hypertension
and, of course, the headache associated with raised intracranial
pressure secondary to a tumour.
Studies of Swedish school children have indicated that 39%
of children experience headache by 7 years of age and 70%
Sourabh Mukhopadhyay MB BS MRCPCH is a Specialist Registrar in
Paediatric Neurology at Morriston Hospital, Swansea, Wales, UK.
Catharine P White MB BS FRCP FRCPCH is a Consultant Paediatric
Neurologist at Morriston Hospital, Swansea, Wales, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
by age 15.2 Other studies have reported similar incidences and
trends.3 The prevalence of headache increases with age and
reaches an adult population prevalence in the early teens.
Migraine affects 1–3% of children by age 7 years and 4–11%
by age 15 years.4 Abu-Afereh and Russell estimated the preva-
lence of tension-type headaches in school children to be 0.9%.5
Secondary headaches are rare, and brain tumour as a cause of
headache even rarer. For every child with a brain tumour there
are around 5000 children with recurrent headaches, including
2000 children with migraine.1
Headaches have a significant impact on the lives of children
and adolescents, resulting in school absence,5 decreased extra-
curricular activities and poor academic achievement.
A good history will provide a diagnosis in the vast majority
of children. This together with a careful examination will ensure
that serious causes are unlikely to be missed. This process should
also be therapeutic and reassuring to the parents and child.
Classifying paediatric headaches
In 1988 The International Headache Society6 published a clas-
sification scheme for headaches, including complex diagnostic
­criteria. In essence, it divided headache into two categories –
­primary and secondary. Primary headache disorders, i.e. those
that have no other underlying cause, include migraine, tension-
type headache and cluster headache. Secondary headaches are
associated with underlying central nervous system (CNS) or
other pathology. More recently, it was recognized that this clas-
sification needed fine tuning, especially in relation to the classifi-
cation of headaches in children and adolescents, and the revised
classification was published in 2004.7
Although standard teaching is to consider migraine and ten-
sion headache as completely different entities, it is much more
likely that they lie on a continuum.8 This theory fits better with
clinical practice.
Clinically it can be more helpful initially to divide headache
into one of four broad types depending on the temporal pattern9:
• isolated acute;
• acute recurrent (episodic);
• chronic progressive;
• chronic non-progressive.
 Acute headache is defined as a recent onset headache with no
prior history of similar episodes. In children, this pattern is most
commonly due to febrile illness related to upper respiratory tract
infection,10 but severe acute headache may also be the present-
ing symptom of a variety of serious intracranial pathologies
such as meningitis, raised intracranial pressure or haemorrhage
(Table 1).
Attacks of head pain separated by symptom-free intervals
are classified as acute recurrent headache. Primary headache
syndromes, such as migraine or tension-type headache, usually
cause this pattern. Infrequently, recurrent headaches are attribut-
able to epilepsy or cluster headache.
In chronic progressive headache the frequency and severity
of the headaches gradually increases with time. This is the most
ominous of the temporal patterns and is commonly correlated
with increasing intracranial pressure. Causes include idiopathic
intracranial hypertension, tumour, hydrocephalus and subdural
collections.
 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology

the headache is important but unfortunately children under 10

Important causes of acute headache years are not good at describing pain, its frequency, severity or
distribution. This does not mean that they should not be asked
• Migraine or listened to. Rothner has created a checklist to aid clinicians
• Tension headache in eliciting the important features of any headache11 (Table 2).
• Infection Whilst more applicable to the outpatient setting, it is still a useful
○ Local aide memoir in the acute situation.
• Eyes In addition, questions need to be specifically asked about other
• Ears symptoms that suggest raised intracranial pressure or progressive
• Teeth neurological disease, such as unsteadiness, seizures or visual dis-
• Sinuses turbances. Subtle behavioural disturbances or school difficulties
• Skin can be important early warning symptoms of a structural aetiol-
• Lymph nodes ogy, but may also occur when the pain becomes chronic. A past
○  Systemic history of head injury or other neurological problems may be
• Viraemia relevant. School absence can be a useful proxy measure for the
• Bacteraemia severity of the problem. A family member with headaches may
• Meningitis give a clue to the cause but may also be acting as a role model
• Encephalitis for the headache behaviour.
• Septicaemia Symptoms that suggest a secondary cause for the headaches
• Arterial hypertension are given in Table 3.
• Inflammatory disease
○ Local Examination
• Cervical The focus of the examination will be determined by the history
• Musculoskeletal and clinical context. It is helpful to divide the child presenting
○  Systemic with an acute severe headache as an emergency from the child
• Kawasaki disease with other temporal patterns of headache as the causes are some-
• Lupus what different (Tables 1 and 4). Lewis and Qureshi10 found that
• Other collagen vascular disease children presenting to the emergency room with an acute head-
• Intracranial ache most commonly had a febrile illness related to an upper
○  Hydrocephalus respiratory tract infection. A serious underlying neurological diag-
○ Intracranial haemorrhage nosis was uncommon, and all these patients had clear objective
○  Brain tumour neurological signs. Signs of an infective aetiology, both intra- and
○  Vascular anomaly
○ Idiopathic intracranial hypertension

○ Post traumatic
Table 1
Chronic non-progressive headaches differ from acute recur-
rent headaches by their greater frequency and persistence. They
may last for years with no associated neurological symptoms or
change in headache severity. A common headache in this cat-
egory is chronic tension-type headache. An important newly
recognized entity that also occurs with this temporal pattern is
chronic daily headache.
Clinical assessment of headache in children
The cornerstone of headache management remains good history
taking and careful physical and neurological examination. This
invariably allows a diagnosis to be made, identifies those chil-
dren who have a secondary cause for their headache and recog-
nizes the few who require further investigation.
History
The history is the key to diagnosing the cause of the headache
and to identifying those children with symptomatic (second-
ary) headache. Asking both the child and their parents about
Things to ask about headache11
Do you have more than one type of headache?
How did the headache begin?
Was there associated trauma or infection?
How long has the headache been present?
Are the symptoms getting better, worse or staying the same?
How often do the symptoms occur?
How long do they last?
Do the headaches occur at any particular time or circumstance?
Is the headache preceded by a warning?
Where does it hurt?
What sort of pain is it? Is it pounding or sharp?
Do you have any associated symptoms during the headache?
Is there any nausea or vomiting?
Do you stop what you are doing during the headache?
Are there activities that make the headache worse?
Does anything make the headache better?
Do you have any other medical problems?
Are you taking any medication?
Does any one in your family have headaches?
What do you think is causing your headache?
Table 2

PAEDIATRICS AND CHILD HEALTH 18:1  © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology

­ ressure. Specific features of the neurological examination that

p
Symptoms suggestive of a secondary cause may indicate a secondary cause are given in Table 5.

Headache history
Investigation
• Short history
○  ‘First or worst’ headache Careful history taking and examination should allow identifica-
○ Recurrent severe headache(s) for a few weeks tion of the few children who require further investigation. It is
• Accelerated course again helpful to divide the investigation of the child with an acute
○ Increasing frequency severe headache from the child with other temporal patterns of
○  Worsening usual headache headache as the investigations that need to be considered are
• Headache timing and posture different.
○ Mainly from sleep

○ In the morning before getting up Acute severe headache

○ Mainly or worse when lying down, relieved when
upright (suggests raised pressure)
○  Worse with bending, coughing, etc
○ Mainly upright, relieved when lying down (suggests
low pressure headache)
Associated symptoms
○ Vomiting from sleep or before getting up
○ Confusion, impaired consciousness
○ Altered personality
○ Focal weakness
○ Diplopia
○ Fever, rigors
○ Seizures
Table 3
extra-cranial, must therefore be specifically sought; as well as
signs of acutely raised pressure and intracranial haemorrhage.
In the outpatient setting most children will have acute recur-
rent or chronic non- progressive headaches, but it is the rare
child with chronic progressive headaches that we need to iden-
tify. Table 4 lists the major causes of headache in the clinic
­population.12
Given this list it is not surprising that the clinical examina-
tion in this situation is invariably normal. Important aspects of
the general physical examination are height, weight and blood
In acute severe headache, routine laboratory investigations may
be helpful given that intercurrent infection is the commonest
cause. Lumbar puncture with measurement of the opening pres-
sure may be needed if subarachnoid haemorrhage, meningitis or
idiopathic intracranial hypertension are diagnostic possibilities.
Skull X-ray and EEG are of extremely limited value. An urgent
initial computerized tomography (CT) scan may be required if
acute hydrocephalus, haemorrhage or a structural lesion are sus-
pected. A CT brain scan with contrast will demonstrate nearly all
structural lesions; one done without contrast is somewhat more
limited in its sensitivity, although it can define hydrocephalus
and haemorrhage easily. Magnetic resonance imaging (MRI) is
less readily available; less good at demonstrating acute blood
and makes monitoring the ill patient more difficult so it is not the
initial method of choice in this situation.13
Table 6 lists the features in the history and examination that
indicate imaging should be strongly considered.
Other temporal patterns of headache
In the outpatient setting the cause for the child’s headache is
usually clear from the history and examination. Neuroimag-
ing is rarely necessary and of little value, unless the history
or examination suggests a structural aetiology. If the history is
typical for migraine and the neurological examination is nor-
mal, no imaging is required. Raised intracranial pressure due
to a tumour is the major concern for the family and the refer-
ring clinician; however, the symptoms and signs should rarely

Major causes of headache in the clinic population12

Signs suggestive of secondary headache
Diagnosis Percentage
Signs of raised intracranial pressure
Migraine without aura 24.3 • Large or accelerating head circumference
Migraine with aura    6.0 • Cracked pot sign
Complicated migraine    2.4 • Papilloedema
Episodic tension-type headache 15.0 • IV nerve palsy
Chronic tension–type headache 23.9
Other signs of CNS disease
Mixed common migraine and episodic tension    4.2
• Other cranial nerve palsies
headache
• Brainstem signs
Mixed common migraine and chronic tension    6.2
• Other focal neurological signs, e.g. hemiplegia
headache
• Cerebellar signs, e.g. ataxia, nystagmus
Non-specific headache 12.3
Other specific diagnoses or combinations    6.0 Signs of other systemic disease

Table 4 Table 5

PAEDIATRICS AND CHILD HEALTH 18:1  © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Indications for neuroimaging in patients with acute
headache
Signs and symptoms of elevated intracranial pressure
Meningeal signs + focal neurological findings or altered
mental status
Progressive or new neurological signs
Significant head trauma
Severe (‘worst headache of life’) headaches of increasing
frequency and duration
Presence of VP shunt
Table 6
be mistaken for other benign causes of headache and brain
tumours account for less than 0.1% of the lifetime prevalence
of ­headaches.14
All patients who present with symptoms and/or signs sug-
gestive of a secondary intracranial cause (Tables 3 and 5) for
their headaches should undergo high-quality imaging. MRI is
generally more costly, takes longer and may require sedation,
but its superior imaging capabilities offer detailed structural
definition not available from CT scanning even with contrast.
Neuroimaging should be considered in those with a significant
change in headache symptomatology or who are younger than
3 years of age.
The Quality Standards Subcommittee of the American Acad-
emy of Neurology and the Practice Committee of the Child
Neurology Society reviewed the literature on the evaluation
of the child with recurrent headaches and concluded that rou-
tine laboratory investigations, lumbar puncture and EEG are
not necessary.15 They also concluded that neuroimaging is not
indicated in children with a normal neurological examination.
In the studies they reviewed, all the children with CNS lesions
who required surgical treatment had definite abnormalities on
examination. They recommended neuroimaging be considered
in children with:
• recent onset of headache (less than 1 month’s duration);
• features in the history suggestive of neurological dysfunction;
• abnormal neurological findings on examination;
• occurrence of seizures.
 Variables that predicted a space-occupying lesion included the
above plus gait abnormalities and the absence of family history
of migraine.
Repeated studies looking at the yield of neuroimaging in chil-
dren with uncomplicated migraine or tension-type headaches
and a normal neurological examination have found small but sig-
nificant numbers of abnormalities, none of which has influenced
the diagnosis, management or outcome for the patient.16,17
It is sometimes suggested that scanning is reassuring to the
patient and their family but a randomized control trial of imaging
in adults with chronic daily headache showed that MRI scan-
ning only temporarily reduced their anxiety about the cause
of their headaches. At 1 year there was no difference between
the scanned and the non-imaged groups and having a scan did
not improve most other measures of health anxiety, illness per-
ceptions or quality of life.18 There is no reason to suspect that
PAEDIATRICS AND CHILD HEALTH 18:1
­children or their parents are any different. Given the potential
risks of neuroimaging, which include incidental findings that
increase the concern of the patient or parents, false reassurance
from an inadequate study, the risks of an allergic reaction to
iodine contrast media with CT scanning, and the risk of sedation
in young children or claustrophobic patients having MRI scans;
we should continue to try to avoid imaging for reassurance.
Specific headache syndromes
As the history is so important in making the diagnosis, this sec-
tion details the features and diagnostic criteria for the two com-
mon causes of headache – migraine and tension-type headache.
It is important to recognize that a mixed pattern of migraine and
tension-type headache is common.
Migraine headaches
Paediatric migraine is now distinctly recognized among the pri-
mary headache disorders. The diagnostic criteria for migraine
are broader in children than they are in adults and allow for
a wider range of duration and localization of the pain. In prac-
tice, paediatric migraines are often bilateral and clear local-
ization of the pain can be difficult to obtain from children. In
general, attacks in children last for a shorter time than in adults.19
They are ­ frequently preceded by a behavioural prodrome with
mood changes or withdrawal from activity. The classical history
obtained in adults of unilateral throbbing/pounding pain lasting
greater than 4 h becomes commoner with age. Migraine with
aura is seen in 14–30% of children with migraine. Typical auras
are spots, colours, image distortions or visual scotoma.
Migraine commonly ‘runs in families’ with approximately
70% of children having an affected first-degree relative.20 Earlier
attempts to define migraine used this strong familial occurrence
as one of the diagnostic features of migraine. The new diagnos-
tic criteria set by the International Headache Society (IHS) are
described in Table 7.
Complicated migraines are headaches that are accompanied or
manifested by transient neurological symptoms. These ­symptoms
may occur immediately before, during or after the headache.
In some situations, the headache may be mild or non-existent.
Hemiplegic migraine, ophthalmoplegic migraine and basilar
artery migraine are typical examples of complicated migraine.
Hemiplegic migraine, while unusual, is seen more commonly in
children than in adults. It is characterized by abrupt onset of
hemiparesis, which usually is followed by a headache. Hemi-
anaesthesia may also precede the headache. Ophthalmoplegic
migraine may occur at any age, and usually is associated with
orbital or periorbital pain, as well as third, fourth or sixth cra-
nial nerve involvement. The headache resolves in hours, but the
ophthalmoplegia may last for days. Basilar artery migraines are
characterized by dizziness, weakness, ataxia and severe occipital
headache (with vomiting).
Tension-type headache
Tension-type headaches are more common in adults than
children. The revised IHS classification distinguishes three
subtypes depending on frequency – infrequent episodic, frequent
episodic and chronic. Tension-type headache is usually bilat-
eral and described as pressing or band-like. These headaches
 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
International Headache Society diagnostic features of
migraine
Migraine without aura
A. At least five attacks fulfilling B–D
B. Duration between 1 and 48 h
C. At least two of the following:
• Bilateral or unilateral
• Pulsating
• Moderate to severe in intensity
• Aggravation by routine physical activity
D. During the headache at least one of the following:
• Nausea or vomiting
• Photophobia or phonophobia
E. Not attributable to another disorder
Migraine with aura
A. In addition to above criteria, at least two attacks fulfilling B
B. At least three of the following:
• One or more fully reversible aura symptoms indicating
focal cortical or brainstem dysfunction
• Aura developing gradually over minutes, or two or more
symptoms occurring in succession
• Aura lasts no more than 1 h
• Pain follows aura after <1 h
Table 7
can last for days and are not aggravated by physical activity.
Nausea, vomiting, photophobia and phonophobia are not typi-
cal accompaniments.
The IHS criteria for the diagnosis of infrequent tension-type
headache are described in Table 8.
Frequent episodic tension-type headache is defined as occur-
ring on greater than 1 but less than 15 days a month for at least 3
months, while chronic tension-type headache occurs for greater
than 15 days per month for at least 3 months. Episodic and chronic
tension-type ­headaches have same physical ­characteristics, but
International Headache Society criteria for the
diagnosis of infrequent tension-type headache
A. At least 10 episodes occurring on <1 day per month on
average (<12 days per year) and fulfilling criteria B–D
B. Headache lasting from 30 min to 7 days
C. Headache has at least two of the following characteristics
1. bilateral location
2. pressing/tightening (non-pulsating) quality
3. mild or moderate intensity
4. not aggravated by routine physical activity such as
walking or climbing stairs
D. Both of the following:
1. no nausea or vomiting (anorexia may occur)
2. no more than one of photophobia or phonophobia
E. Not attributed to another disorder
Table 8
PAEDIATRICS AND CHILD HEALTH 18:1
for the latter the pain may be continuous and mild nausea may
occur. The picture may be complicated and perpetuated by inap-
propriate and ­excessive analgesia use.
Chronic daily headache (CDH) was first described in adults
who reported daily or nearly daily headaches. It was soon rec-
ognized that, although patients were similar in the number of
headaches experienced, the characteristics of their headaches fell
on a continuum between migraine and tension-type headache.
Recent work has demonstrated a similar spectrum in children.
The most common CDH pattern is superimposition of migraines
on a background pattern of frequent tension-type headaches, but
other categories are recognized based on the specific clinical fea-
tures and the pattern of evolution.21 ◆
References
1 Abu-Arafeh I, Macleod S. Serious neurological disorders in children
with chronic headache. Arch Dis Child 2005; 90: 937–940.
2 Bille BS. Migraine in school children. A study of the incidence
and short term prognosis, and a clinical, psychological and
electroencephalographic comparison between children with migraine
and matched controls. Acta Paediatr 1962; 51(suppl 136): 1–151.
3 Sillanpaa M, Abu-Arafeh I. Epidemiology of recurrent headache
in children. In: Abu-Arafeh I, ed. Childhood headache. London:
MacKeith Press, 2002, p. 19–34.
4 Lipton RB. Diagnosis and epidemiology of pediatric migraine. Curr
Opin Neurol 1997; 10: 231–236.
5 Abu-Arefeh I, Russel G. Prevalence of headache and migraine in
school children. BMJ 1994; 309: 765–769.
6 Headache Classification Committee of the International Headache
Society. Classification and diagnostic criteria for headache disorders,
cranial neuralgias, and facial pain. Cephalalgia 1998; 8(suppl 7):
1–96.
7 Headache Classification Committee of the International Headache
Society. The international classification of headache disorders.
Cephalalgia 2004; 24(suppl 1): 1–160.
8 Viswanathan V, Bridges SJ, Whitehouse W, et al. Childhood
headaches; discrete entities or continuum? Dev Med Child Neurol
1998; 40: 544–550.
9 Rothner AD. Headaches in children. A review. Headache 1978; 18:
169–175.
10 Lewis DW, Qureshi F. Acute headache in children and adolescents
presenting to the emergency department. Headache 2000; 40:
200–203.
11 Rothner AD. Headaches. In: Swaiman KF, Ashwal S, eds. Paediatric
neurology: Principles and practice. St Louis: Mosby, 1999, p. 747–758.
12 Abu-Arafeh I, Callaghan M. Headache clinics for children.
In: Abu-Arafeh I, ed. Childhood headache. London: MacKeith Press,
2002, p. 175–184.
13 Stoodley N. Neuroimaging in children. Curr Paediatr 2003; 13:
479–485.
14 Rasmussen BK. Epidemiology of headache. Cephalalgia 1995; 15:
45–68.
15 Lewis DW, Ashwal S, Dahl G, et al. Practice parameter: Evaluation
of children and adolescents with recurrent headaches: report of
the Quality Standards Subcommittee of the American Academy
of Neurology and the Practice Committee of the Child Neurology
Society. Neurology 2002; 59: 490–498.
 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
16 Worber-Bingol C, Wober C, Prayer D, et al. Magnetic resonance
imaging for recurrent headache in childhood and adolescence.
Headache 1996; 36: 83–90.
17 Lewis DW, Dorbad D. The utility of neuroimaging in the evaluation
of children with migraine or chronic daily headache who have
normal neurologic examinations. Headache 2000; 40: 629–632.
18 Howard L, Wessely S, Leese M, et al. Are investigations anxiolytic
or anxiogenic? A randomised controlled trial of neuroimaging to
provide reassurance in chronic daily headache. J Neurol Neurosurg
Psychiatry 2005; 76: 1558–1564.
19 Maytal J, Young M, Shecter A, et al. Pediatric migraine and the
International Headache Society criteria. Neurology 1997; 48: 602–607.
20 Condgen PJ, Forsythe WI. Migraine in childhood: a study of 300
children. Dev Med Child Neurol 1979; 21: 209–216.
21 Gladstein J, Holden EW. Chronic daily headache in children and
adolescents: a 2 year prospective study. Headache 1996; 36: 349–351.
PAEDIATRICS AND CHILD HEALTH 18:1
Practice points
• Headache is common in children
• The history is the key to diagnosis
• Many parents are simply seeking reassurance that their child
does not have a brain tumour
• Neuroimaging is rarely necessary unless the history or
examination suggests a structural aetiology
• Headache due to a space-occupying lesion is very rarely an
isolated symptom and there are invariably accompanying
neurological signs
• Adult definitions of migraine do not always apply to children
and there is likely to be a continuum between migraine and
tension-type headache.
 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: Neurology
Management and outcome
of viral encephalitis in
children
Rachel Kneen
Tom Solomon
Abstract
Although viral encephalitis (VE) is relatively uncommon in children in
Q2
the UK, early recognition and appropriate investigation and manage-
ment are essential because of the devastating nature of the condition.
An estimated 1200 paediatric cases from all causes are admitted each
year. The cause of VE is diagnosed in approximately 30% of cases. The
commonest cause in the UK is herpes simplex virus type 1 but ­globally
the most important cause is Japanese encephalitis virus. Diagnostic
methods have improved recently and the yield can be increased with
a rational approach to investigations and by taking specialist advice.
Morbidity and mortality are high but can be significantly reduced in
herpes simplex ­virus encephalitis by treating for at least 14 days with
intravenous aciclovir. Sequelae are common and include motor, sensory
and cognitive impairments, epilepsy, behaviour problems and psychiatric
disorders. ­ Children should have access to a neurorehabilitation team,
which includes a neuropsychologist, to achieve the best recovery possi-
ble. Long term follow-up and an organized transition into adult services
is required.
Keywords aciclovir; central nervous system infection; herpes simplex
virus; meningoencephalitis; viral encephalitis
Introduction
Viral encephalitis (VE) is uncommon in the UK. A typical district
general hospital could expect to see approximately five paediatric
cases per year.1 However, it is very common for a child to pres-
ent with symptoms for which the differential diagnosis includes
VE. These children are started on intravenous aciclovir and broad-
spectrum antibiotics and treated for a variable length of time before
either an alternative diagnosis is made or the diagnosis of VE is
confirmed. In these children, the approach to making a diagnosis
and the length and route of treatment with aciclovir is often vari-
able. Guidelines for the management of adult patients with sus-
pected encephalitis have been drawn up2,3 and are in the process
of being modified/ratified for national usage. ­Similar guidelines are
in development for children with suspected VE. The management
Rachel Kneen MRCPCH is a Consultant Paediatric Neurologist at the
Department of Neurology, Littlewoods Neuroscience Foundation, Royal
Liverpool Children’s NHS Trust, Alder Hey, Liverpool, L12 2AP, UK.
Tom Solomon FRCP is Chair of Neurology and heads the Brain Infections
Group at the University of Liverpool, Walton Centre for Neurology and
Neurosurgery, Liverpool, L9 7LJ, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
of patients with suspected VE has changed dramatically in recent
years, with improved viral diagnostics and imaging studies, better
antiviral and immunomodulatory therapies, and enhanced neuro-
intensive care and rehabilitation. This review aims to provide a
rational approach to the investigation and treatment of children
and infants older than 28 days with suspected and proven VE.
What do we mean by encephalitis?
Encephalitis means inflammation of the brain parenchyma.
It can be caused directly by a range of viruses (Table 1) and
other microorganisms (Table 2). Encephalitis can also occur as
an immune-mediated phenomenon, e.g. in acute disseminated
encephalomyelitis (ADEM), which often follows infections or
vaccinations (Table 2).
Encephalitis is strictly a pathological diagnosis that should only
be made if there is tissue confirmation (autopsy or brain biopsy).
This is clearly not very practical. Therefore, most patients are
diagnosed with encephalitis if they have the appropriate clinical
presentation and surrogate markers of brain inflammation, such
as inflammatory cells in the cerebrospinal fluid (CSF) or inflam-
mation shown on imaging, especially if an appropriate organism
is detected. Encephalitis is different from encephalopathy, which
is the clinical syndrome of reduced consciousness associated
with other infectious diseases, metabolic disorders and drugs.
Metabolic and toxic causes of encephalopathy can usually be dis-
tinguished from VE by the lack of an acute febrile illness, more
gradual onset, lack of a CSF pleocytosis and absence of focal
changes on magnetic resonance imaging (MRI).4 The organisms
that cause encephalitis often also cause an associated meningeal
reaction (meningitis). In this clinical situation, the term menin-
goencephalitis is sometimes used. However, others use this term
to cover the full spectrum of encephalitis and meningitis; this
should be discouraged because the clinical features, investiga-
tions and treatment differ. Spinal cord inflammation (myelitis)
or nerve root involvement (radiculitis) may also occur in viral
central nervous system (CNS) infections.
What causes the clinical picture in viral encephalitis?
Depending on the virus, the pathogenesis can consist of a mix-
ture of direct viral cytopathology and a para- or post-infectious
inflammatory or immune-mediated response. For most viruses,
the brain parenchyma and neuronal cells are primarily infected,
but for some, the blood vessels can be attacked, giving a strong
vasculitic component. Demyelination following infection can
also contribute. Herpes simplex virus (HSV), for example, pri-
marily targets the brain parenchyma on the temporal lobes,
sometimes with frontal or parietal involvement. Mumps virus
can cause an acute VE or an immune-mediated delayed enceph-
alitis. Measles virus causes a post-infectious encephalitis, which
can sometimes have a severe haemorrhagic component (acute
haemorrhagic leukoencephalitis). For influenza A virus a dif-
fuse cerebral oedema is a major component in the pathogen-
esis, while for varicella zoster virus (VZV), vasculitis is a major
pathogenic process.
Viruses must cross the blood−brain barrier to cause enceph-
alitis. Primary infection with HSV type 1 (HSV-1) occurs in
the oral mucosa. Following primary infection the virus travels
 © 2007 Published by Elsevier Ltd.
 Symposium: Neurology

Causes of acute viral encephalitis in children Diseases mimicking viral encephalitis in children Q3

(modified from ref.28) (modified from ref.28)

Sporadic causes of viral encephalitis (not geographically CNS infections Para/post-infectious causes
restricted) listed by group Bacteria Guillain-Barré syndrome*
Herpes viruses Bacterial meninigitis Acute disseminated
• Herpes simplex virus type 1 and 2 Tuberculous encephalomyelitis*
• Varicella zoster virus meningitis Systemic viral illnesses with
• Epstein-Barr virus Brain abscess complex febrile convulsions
• Cytomegalovirus Typhoid fever Acute haemorrhagic leukoencephalitis
• Human herpes virus types 6 and 7 Parameningeal Acute necrotizing encephalopathy
Enteroviruses infection
 Non-infectious diseases
• Coxsackieviruses Lyme disease
Vasculitic
• Echoviruses Leptospirosis
Systemic lupus erythematosis
• Enteroviruses 70 and 71 Mycoplasma
Acute haemorrhagic
• Parechovirus pneumonia
leukoencephalitis
• Poliovirus Listeria
Acute necrotizing
Paramyxoviruses monocytogenes
encephalopathy
• Measles virus Brucellosis
Polyarteritis nodosa
• Mumps virus Subacute bacterial
ANCA associated vasculitis
Others (rarer causes) endocarditis
Neoplastic
• Influenza viruses (emboli)
Primary brain tumour
• Adenovirus Fungi
Brain metastases
• Parvovirus Cryptococcus
Paraneoplastic limbic encephalitis+
• Lymphocytic choreomeningitis virus Coccidiomycosis
Metabolic
• Rubella virus Histoplasmosis
Diabetic ketoacidosis and coma
Candidiasis
 Geographically restricted causes of encephalitis – mostly Hepatic encephalopathy
Parasites
arthropod-borne* Hypertensive encephalopathy
Cerebral malaria
The Americas Haemolytic uraemic syndrome
Toxoplasmosis
• West Nile Hypoglycaemia
Cysticercosis
• La Crosse Reye syndrome
Rickettsiae
• St Louis Mitochondrial encephalopathy
Typhus
• Dengue Toxic encephalopathy (drugs,
Q fever
• Rabies alcohol)
Erlichiosis
Europe/Middle East Other
Cat-scratch fever
• Tick-borne encephalitis Drug reactions
• West Nile Subarachnoid and subdural
• Rabies haemorrhage
Africa Venous sinus thrombosis and
• West Nile infarction
• Rift Valley fever virus Ischaemic cerebrovascular
• Crimean-Congo haemorrhagic fever accidents
• Dengue Epileptic encephalopathy or non-
• Rabies convulsive status epilepticus
Asia Functional illness
• Japanese encephalitis Voltage-gated K channel limbic
• West Nile encephalitis+
• Dengue
• Murray Valley encephalitis *Guillain–Barré syndrome and acute disseminated encephalomyelitis may
follow viral or bacterial infections or vaccinations.
• Rabies +Very rare in children.
• Nipah
Australasia
Table 2
• Murray valley encephalitis
• Japanese encephalitis
c­ entripetally along the trigeminal nerve to give latent infection in
*All viruses are arthropod-borne, except for Rabies and Nipah virus. the trigeminal ganglion. About 70% of all cases of HSV encepha-
litis already have antibody present, indicating re-activation of
Table 1 virus is the most common mechanism;5 though it is not clear

PAEDIATRICS AND CHILD HEALTH 18:1  © 2007 Published by Elsevier Ltd.

 Symposium: Neurology
whether this is re-­activation of virus in the trigeminal ganglion
or virus that has already established latency in the brain itself.6
Why HSV ­sometimes re-activates is not known. In young adults
and children, HSV encephalitis occurs during primary infec-
tion. Neonates can be infected with HSV-2 during delivery to
cause neonatal herpes, a disseminated infection often with CNS
involvement. This review does not cover neonatal herpes infec-
tion in detail (see Useful websites and further reading). HSV-2
also causes viral meningitis in adults, which may be recurrent.
HSV-2 is predominantly sexually acquired and the diagnosis of
HSV-2 meningitis in a child should alert the paediatrician to the
possibility of sexual abuse.7
The other major route of viral entry into the nervous system
is with a viraemia and subsequent spread across the blood–brain
barrier; this occurs for enteroviruses such as polio, and arbo­
viruses (viruses transmitted via an insect or tick vector) such as
Japanese encephalitis virus (JEV) and West Nile virus (WNV).
Epidemiology
The epidemiology of VE is changing for several reasons:
• Numbers of immunocompromised patients have increased
due to human immunodeficiency virus (HIV) infection, trans-
plant surgery and cancer treatments. These patients are at risk
of encephalitis caused by cytomegalovirus (CMV), Epstein-Barr
virus (EBV) and human herpes virus 6 (HHV-6).
• Arboviruses are spreading to new areas, e.g. WNV across
North America and southern Europe, and JEV across Asia.
• Large and unexpected outbreaks of VE have occurred in Asia
from enterovirus (EV) 71 and Nipah virus.8,9
• Vaccination has reduced encephalitis due to measles and
mumps virus.
Incidence
The incidence of VE depends on the particular viruses that are
prevalent. Currently, accurate figures for the UK are unknown.
This is being addressed in a prospective study (funded by the
Health Protection Agency), which is currently enrolling patients
from many centres in England (www.hpa.org.uk/infections/
topics%5Faz/encephalitis/study.htm). For children, the best
extrapolation we can use at present are data from Finland. In a
2-year prospective study (1993/4), Koskiniemi et al found that
the overall incidence was 10.5 in 100 000 child-years, with the
highest figure in children younger than 1 year of age (18.4 in 100
000 child-years). The most commonly identified agents, based on
virological and serological studies, were VZV, respiratory viruses,
enteroviruses, adenoviruses, EBV, HSV, rotaviruses and HHV-6.1
For adults and children, HSV encephalitis is the most com-
monly diagnosed VE in developed countries, with an incidence
of 1 in 250 000–500 000.5 Most HSV encephalitis is due to HSV-1,
but about 10% are due to HSV-2; the latter causes encephalitis in
immunocompromised adults and neonates, in whom it causes a
disseminated infection.
Presentation
The classical presentation is a brief flu-like prodrome imme-
diately followed by severe headaches, nausea, vomiting and
PAEDIATRICS AND CHILD HEALTH 18:1
altered consciousness. There may be meningism, seizures and
focal neurological signs.
The subtle presentation is with low grade fever, behav-
ioural changes or speech and language disturbance.10 The sub-
acute presentation is more common in immunocompromised
patients.
Important features in the history
As well as being crucial in determining who needs investigation
for VE, the history can provide useful clues as to the ­aetiology:
• recent rash (patient or contacts) – measles, chicken pox
(VZV), slapped cheek syndrome (parvovirus), hand, foot and
mouth disease [enterovirus (EV)] or roseola (HHV-6). See
­examination findings below;
• parotitis, testicular pain or abdominal pain (pancreatitis) –
mumps;
• dog bite, scratch from bat – rabies;
• travel history – South East Asia (JEV); forests in central ­Europe
[tick borne encephalitis virus (TBEV)];
• social history – mother from HIV endemic area or intravenous
drug abuser.
Important examination findings
The priority is to stabilize the patient and treat immediate com-
plications such as seizures. Make an assessment of level of coma
using the modified Glasgow coma score (GCS). Do not ignore
the parents who tell you their child is confused or has altered
behaviour.
Other important examination findings are:
• rashes – exanthems (many viruses), e.g. hand, foot and mouth
disease (EV71), purpuric rash (meningococcal meningitis);
• meningism or photophobia – check for neck stiffness and
Kernig’s sign;
• signs of raised intracranial pressure (RICP) – altered pupillary
responses, raised blood pressure, bradycardia, altered respira-
tory pattern, abnormal posture (decorticate or decerebrate),
papilloedema;
• focal neurological signs – hemiplegia, cranial neuropathies,
abnormal posturing (see above);
• tonic eye deviation, nystagmus, subtle clonic movements of
face or fingers – subtle motor status epilepticus;
• lower cranial neuropathies – rhomboencephalitis/basal
meningoencephalitis (enteroviruses or listeria, tuberculosis).
Initial investigations
A cause is found in approximately 30% of children with VE. It is
likely this figure could be improved if more detailed virological
tests were undertaken (see below). This is another of the aims
of the Health Protection Agency study (see Useful websites and
further reading). Some features and clues may be present in the
baseline investigations:
• full blood count – atypical lymphocytes in EBV;
• hyponatraemia – syndrome of inappropriate antidiuretic hor-
mone (SIADH) common in many encephalitides;
• elevated amylase – mumps;
• elevated liver enzymes – EBV, cytomegalovirus (CMV).
 © 2007 Published by Elsevier Ltd.
 Symposium: Neurology
CSF findings
The use of lumbar puncture (LP) in suspected CNS infection has
been controversial and has declined recently.11 However, the LP
is an essential investigation in suspected CNS infection because:
• Initial findings are available within hours and reveal whether
or not there is an infection. On rare occasions, the initial cell
count may be normal early in the disease process or if the child
is immunocompromised. In this circumstance, if CNS infection is
still strongly suspected, the LP may need to be repeated.
• Initial findings are often able to distinguish between bacterial
and viral infection (see Table 4).
• Culture, antibody studies and polymerase chain reaction
(PCR) of CSF confirm what the organism is. This allows therapies
to be altered and appropriate treatment regimens to be given.
• Positive results allow appropriate disease notification.
• Results may alter follow-up arrangements for the child, e.g.
hearing tests, long term developmental follow-up, etc.
• Parents find it easier to understand their child’s illness if a
diagnosis has been confirmed, especially if the child dies or has
long term sequelae.
Therefore, an LP should be undertaken unless specific contra­
indications are present (Table 3). An LP is not contraindicated
in children who are only mildly confused or lethargic. If focal
neuro­logical signs or coma are present, an LP should still be pos-
sible after performing a CT scan, if the latter shows no evidence
of brain shift. A CT should also be performed in children who
may be immunocompromised as they may not mount an inflam-
matory response to an abscess. Very often, a child is admitted
and treatment for bacterial meningitis and encephalitis is started
before an LP is performed. In these children, it is still essential
to perform an LP when safe to do so because this informs the
diagnosis and guides subsequent management.12 Giving blind
antibacte­rial and antiviral treatment to all patients with ­suspected
CNS infection, and then not investigating with LP because it
‘makes no ­difference to the management’ is unacceptable prac-
tice and should be discouraged. This approach risks missing
other dia­gnoses that may require alternative treatments.13 In
Contraindications to lumbar puncture
• Signs of RICP – altered papillary responses, absent Doll’s
eye reflex, decerebrate or decorticate posturing, abnormal
respiratory pattern, papilloedema, hypertension and
bradycardia
• Recent (within 30 min) or prolonged (>30 min) convulsive
seizures
• Focal or tonic seizures
• Other focal neurological signs – hemi/monoparesis, extensor
plantar responses, ocular palsies
• GCS < 13 or deteriorating level of consciousness
• Strong suspicion of meningococcal infection (typical purpuric
rash in an ill child)
• State of shock
• Local superficial infection
• Coagulation disorder
Table 3
PAEDIATRICS AND CHILD HEALTH 18:1
HSV ­encephalitis, PCR remains positive in about 80% of patients
even 1 week after starting antiviral therapy.5
CSF findings in VE are shown in Table 4. The CSF opening
pressure is often mildly raised and there is usually a mild to mod-
erate CSF pleocytosis of 5 to 1000 cells/mm3, with predominant
lymphocytes. CSF white cell count may be normal in the first few
days of the infection or neutrophils may predominate. The CSF
red cell count is usually normal or mildly elevated, but it may be
­markedly raised in HSV encephalitis, which can be haemorrhagic,
or in acute necrotizing haemorrhagic leukoencephalitis. The
­glucose ratio is usually normal in viral CNS infections, though it
may be mildly reduced, especially with mumps infection. The CSF
protein is often mildly elevated (between 0.5 and 1.0 g/litre).
Specific investigations to consider
The list of possible investigations for VE is considerable. Our
practice is to perform the initial CSF PCR for α-herpes viruses
(HSV-1 and -2, and VZV) (see below) immediately the samples are
received, because of their importance for early management. Fur-
ther investigations are guided by the clinical picture, especially the
patient’s immune status, and initial CSF findings. The opinion of a
clinical virologist should be sought before undertaking these inves-
tigations. A staged approach should be undertaken (Table 5).
PCR of CSF
Many important viruses and other microorganisms can now be
detected using PCR.14 For HSV, PCR has high sensitivity and
specificity (both greater than 95%); however, it may be negative
in the first few days of the illness or after about 10 days.15 Ini-
tial investigations for patients with VE should therefore include
PCR for HSV, and probably VZV, because these are potentially
­treatable with aciclovir. If it is the summer or autumn, PCR for
EV (­coxsackievirus types A and B, echoviruses and the new
enterovirus types such as EV71) should also be performed; mea-
sles and mumps should be looked for if there is a clinical indi-
cation, but they can occasionally cause encephalitis in patients
without the typical features. Other viruses to consider include
HHV-6 and -7 (especially in young children), adenovirus, influ-
enza virus A and B, and rotaviruses. In immunocompromised
patients with EBV and CMV, PCR should also be performed
and HHV-6 considered. PCR can also be used to detect Myco-
plasma pneumoniae in the CSF. CSF viral culture is now rarely
performed, though it has the advantage over PCR of potentially
being able to detect any viruses, whereas PCR is specific to the
virus being looked for.
Further investigations
The following may be helpful in VE and should be considered in
conjunction with a clinical virologist:
• Serum IgM for specific viruses, particularly HSV, EV, EBV and
CMV. Compare with IgM in CSF (see below).
• CSF IgM for specific viruses, particularly HSV, EV, EBV and
CMV. The detection of specific IgM in the CSF, higher than levels in
the serum, is an indication that antibody is being produced ­locally
in the CNS in response to infection. (In general, though, IgM is only
elevated for primary infections rather than re-activation.)
• PCR and/or culture of throat and rectal swabs for EV and
rotavirus.
10 © 2007 Published by Elsevier Ltd.
 Symposium: Neurology

Typical cerebrospinal fluid findings in central nervous system infections

Viral Acute bacterial Tuberculous meningitis Fungal Normal

meningoencephalitis meningitis

Opening pressure Normal/high High High High-very high Age < 8 years
< 13.5 cm3
Age > 8 years
< 20 cm3
Colour ‘Gin’ clear Cloudy Cloudy/yellow Clear/cloudy Clear
Cells/mm3 Normal-high High-very high Mild elevation Normal-high <5
0–1000 1000–50,000 25–500 0–1000
Differential Lymphocytes Neutrophils Lymphocytes Lymphocytes Lymphocytes
CSF:plasma glucose ratio Normal Low Low-very low (e.g. <30%) Normal-low 66%*
Q4

Protein (g/L) Normal-high (0.5–1.0) High (>1) High-very high (1.0–5.0) Normal-high (0.5–5.0) <0.5

A bloody tap will falsely elevate the CSF white cell count and protein. To correct for a bloody tap, subtract 1 white cell for every 700 red blood cells/mm3 in the
CSF, and 0.1 g/dL of protein for every 1000 red blood cells.
*Although a normal CSF:plasma glucose ratio is quoted as 66%, only values < 50% are likely to be significant.

Some important exceptions

In viral CNS infections, an early LP may give predominantly neutrophils, or there may be no cells in early or late LPs.
In patients with acute bacterial meningitis that has been partially pre-treated with antibiotics (or patients < 1 year old), the CSF cell count may not be very high
and may be mostly lymphocytes.

Tuberculous meningitis may have predominant CSF polymorphs early in the infection.
Listeria can give a similar CSF picture to TBM, but the history is shorter.
CSF findings in bacterial abscesses range from near normal to purulent, depending on location of the abscess, and whether there is associated meningitis or rupture.
A cryptococcal antigen test (CRAG) and Indian ink stain should be performed on the CSF of all patients in whom cryptococcus is possible.

Table 4

• PCR, culture, and antigen detection of nasopharyngeal aspirates
for respiratory viruses, e.g. adenoviruses or influenza ­viruses.
• PCR, electron microscopy from vesicular fluid – chicken pox
(VZV), hand foot and mouth disease (EV71), genital herpes
(HSV-2).
• Cold agglutinins and IgM for Mycoplasma pneumoniae if
­respiratory infection is present.
• Brain biopsy for PCR, culture, antigen detection, histology –
consider if this will change management when diagnosis un-
clear. Decision should be made with a paediatric neurologist and
­neurosurgeon.
Imaging and EEG studies
As described above, in patients with encephalopathy and fever, a
computerised tomography (CT) scan (with contrast) is performed
before LP if there are clinical signs suggestive of herniation or a
space occupying lesion. In HSV encephalitis, CT may be normal
initially, or there may be subtle swelling of the fronto-­temporal
region with loss of the normal gyral pattern; subsequently there
is hypodensity, and there may be high signal change if haemor-
rhagic transformation occurs (Figure 1). MRI is generally more
sensitive, showing high signal intensities in the brain areas
affected (Figures 2 and 3), though even MRI can be normal if per-
formed very early.16 Diffusion-weighted MRI may be especially
useful at demonstrating early changes.17 Gadolinium should be
given during the MRI scan.
An EEG usually shows non-specific, diffuse, high amplitude
slow waves of encephalopathy, but can be useful in looking for
subtle motor seizures. Periodic lateralized epileptiform discharges
(PLEDS) were once thought to be diagnostic of HSV encephalitis,
but have since been seen in other conditions ­(Figure 4).
Management
The management of VE covers the acute presentation, recov-
ery and rehabilitation stages. Often children are transferred to a
­tertiary centre which has facilities for acute brain injury rehabili-
tation. Broadly speaking, the management comprises of:
1. Stabilization of the patient – this may require ventilation and
transfer to a paediatric intensive care unit.
2. Control of immediate complications, e.g. seizures, RICP.
3. Antiviral or immunomodulatory treatment.
4. General supportive measures, e.g. hydration, safe feeding, treat-
ment of infections such as pneumonia, urinary tract infections.
5. Prevent later complications, e.g. contractures, venous
­thrombosis.
6. Neurorehabilitation, including all therapy modalities (physi-
otherapy, occupational therapy, speech and language therapy
and psychology).
7. Specialist educational assessment and provision.
8. Treatment of later complications, e.g. epilepsy, spasticity.
9. Continued neuropsychology for specific sequelae (see below).
Seizures, raised intracranial pressure and other complications
Seizures are common in VE and status epilepticus may occur,
which can be refractory.18 Seizures may be obvious (focal or
­generalized tonic–clonic) or subtle, which may manifest as
twitching of a digit or around the mouth or eyes. An EEG should

PAEDIATRICS AND CHILD HEALTH 18:1 11 © 2007 Published by Elsevier Ltd.

 Symposium: Neurology

Staged approach to microbiological investigation of

suspected viral encephalitis in children

CSF PCR
1. All patients
• Herpes simplex virus (HSV) type 1 (type 2 usually also
performed by lab), varicella zoster virus (VZV)
2. If indicated:
• Enterovirus and parechovirus (especially if
meningoencephalitis picture, summer/autumn)
• Adenovirus (ADV), influenza (Inf ) A and B, rotavirus
(winter/spring)
• Epstein-Barr virus/cytomegalovirus (CMV) (especially if
immunocompromised)
• Human herpes virus (HHV) 6 and 7 (young children or
immunocompromised)
• Measles, mumps (if clinically indicated, contact Public
Health Lab for advice)
• Mycoplasma pneumoniae (if respiratory infection present)
3. Special circumstance
• Rabies, Japanese encephalitis virus, West Nile virus, dengue,
tick-borne encephalitis virus (if appropriate exposure
history)

Antibody testing
1. Viruses: IgM IgG, oligoclonal bands of CSF and serum (acute
and convalescent) for antibodies against:
• HSV (1 and 2), VZV, CMV, HHV6, HHV7, enteroviruses,
respiratory synctial virus, ADV, Inf A and B;
2. If associated with atypical pneumonia, test serum for:
• Mycoplasma pneumoniae serology and cold agglutinins
• Chlamydia serology
Ancillary investigations
1. Throat swab (TS), nasopharyngeal aspirate (NPA), rectal swab
(RS):
• PCR/culture of TS, RS for enteroviruses
• PCR of TS for Mycoplasma pneumoniae, chlamydia
• PCR/antigen detection of NPA for respiratory viruses, ADV, Inf
2. Vesicle electron microscopy and culture:
• Patients with herpetic lesions (for HSV, VZV)
• Children with hand, foot and mouth disease (for
enteroviruses)
3. Brain biopsy
Table 5
be performed if there is any uncertainty. The EEG may help with
the diagnosis in HSV encephalitis (see above). Seizures should
be managed with intravenous phenytoin or phenobarbitone.
Usually a child is maintained on one of these anticonvulsants for
a minimum of 6 weeks following the acute illness.
Intracranial pressure should be reduced by nursing the patient
with their head elevated to 30°, keeping the head straight and keep-
ing the patient as still as possible. The patient should be ventilated
to keep a low pCO2 and bolus doses of mannitol should be given.
The risk of deep venous thrombosis/pulmonary embolus
should be reduced with TED stockings and prophylactic heparin;
the risk of pressure sores reduced with appropriate mattresses
Figure 1 CT scan showing oedema in left temporal lobe with effacement
of left lateral ventricle and overlying sulci in HSV type 1 encephalitis.
and positioning; and the risk of joint contractures reduced with
passive and active limb movements, and orthoses.
To reduce the risk of pneumonia, the risk of aspiration should
be assessed. Enteral feeds should be given via a nasogastric tube
until swallow safety is assessed.
Aciclovir
In most immunocompetent patients, aciclovir should be given as
soon as there is a strong suspicion of VE. This suspicion will be
based on the clinical presentation and initial CSF and/or imaging
findings. Aciclovir is a nucleoside analogue that is highly effec-
tive against HSV and other herpes viruses. Given intravenously at
10 mg/kg three times daily it reduces the risk of a fatal outcome
from approximately 70% to less than 20%.19 Renal function
should be monitored closely, and adequate hydration maintained,
because of the rare risk of renal failure. Other rare side effects
include local inflammation, hepatitis and bone marrow failure.
Although the original aciclovir trials were for 10 days’ treatment,
most neurologists would continue treatment for 14 or 21 days, espe-
cially in children with proven HSV encephalitis, because of the risk of
relapse after 10 days.20 Some advocate repeating the CSF examination
at the end of treatment and continuing with aciclovir if HSV is still
detected. This approach is being evaluated by the American Collabor-
ative Antiviral Study Group, which is assessing the prognostic value
of quantitative PCR detection of viral DNA at the end of 3 weeks’
treatment, and prolonged high dose oral valaciclovir for 3 months.
If the initial CSF PCR is negative for HSV, but other features are
consistent with HSV encephalitis, then the aciclovir should not
be stopped, because false negatives can occur, particularly early
on.21 In such patients the LP should be repeated, and treatment

PAEDIATRICS AND CHILD HEALTH 18:1 12 © 2007 Published by Elsevier Ltd.

 Symposium: Neurology
Figure 2 MRI (T2 axial view) showing high signal if left temporal lobe in
HSV type 1 encephalitis.
continued for at least 10 days. However, if a definitive alternative
diagnosis has become apparent, or it seems very unlikely that the
patient has VE, then it is reasonable to stop aciclovir.
Oral valaciclovir
In a proven case of HSV encephalitis or if HSV encephalitis is
strongly suspected, it may be reasonable to convert to oral valaci-
clovir if ongoing intravenous treatment is proving difficult (e.g.
in a child who is now fully conscious).22 However, we would
only consider this after the first 10 days of intravenous treatment.
Occasionally, therefore, a child may require either a long line or
central line to be inserted under general anaesthetic to complete
a full course of treatment. Valaciclovir is an ester of aciclovir,
which is converted to aciclovir after absorption, and has good oral
bioavailability. Oral aciclovir should not be used in HSV encepha-
litis, because the levels achieved in the CSF are not high enough.
Other treatments to consider
Antibiotic treatment with a cephalosporin is usually given in chil-
dren until a proven viral cause is identified. It is wise to treat
until the blood and CSF cultures are known to be negative.
In patients with brain swelling, corticosteroids and mannitol
are often used to control RICP. A recent trial suggests steroids
may be beneficial in adults with HSV encephalitis whether or not
there is marked swelling.23 In patients with severe brain swell-
ing, decompressive hemicraniectomy is sometimes ­performed.
Other antiviral and immunomodulatory treatment
Except for HSV encephalitis, there are few large randomized con-
trolled trials (RCTs) assessing the efficacy of antiviral treatments
PAEDIATRICS AND CHILD HEALTH 18:1
Figure 3 MRI (coronal FLAIR) showing high signal in left temporal lobe
in HSV type 1 encephalitis.
in other viral CNS infections. In VZV encephalitis steroids are used
alongside aciclovir because of the strong vasculitic component of
the disease. Severe HHV-6 disease is treated with ganciclovir or
foscarnet, and plecornaril is used for severe EV infections. Inter-
feron-alpha has been used in WNV and other flavivirus infec-
tions, but an RCT in JEV showed that it was not effective.24
Management in the recovery period
Children may require a prolonged period of rehabilitation both
as an inpatient and as an outpatient. Ideally, this should be with
a tertiary neurorehabilitation team. In our centre, we have a
multidisciplinary acute brain injury team, consisting of paedi-
atric neurologists, community paediatricians, neurophysiothera-
pists, occupational therapists, speech and language therapists,
neuropsychologist and specialist teaching staff, that makes a full
assessment and delivers a package of treatments and therapies
for rehabilitation. It continues to be involved as children are
re-integrated back to their local teams and will liaise with local
services and schools to ensure the child has the best ongoing
management (see Useful websites and further reading).
Neurocognitive impairments are common after VE, including
change in memory, perception and executive function skills, such
as organizing and planning. These changes are most noticeable
in older children and can make re-integration back into school
very difficult. Before discharge a full neuropsychiatric assess-
ment should be performed, including cognitive function, intel-
ligence, memory and speech assessment, because this will help
13 © 2007 Published by Elsevier Ltd.
 Symposium: Neurology
Figure 4 EEG showing left sided periodic discharges every 2 seconds in HSV type 1 encephalitis.
determine the extent of any damage and identify where help is
needed. Regular outpatient assessment following VE is impor-
tant. Behavioural and psychiatric disturbances are common and
may include depression, attention deficit hyperactivity disorder,
aggressive behaviour or disinhibition. A referral to the local child
and adolescent psychiatry team may be necessary.
Memory difficulties can be an especially important prob-
lem in HSV encephalitis. A range of practical approaches can
help to overcome these difficulties; these include simple mea-
sures such as the child keeping a notebook and diary, having
a ­ communication book for school, labelling items around the
house and leaving messages as reminders. For older children and
adolescents, more sophisticated aids may be helpful, including a
neuropage system (www.neuropage.nhs.uk), which sends pager
reminder messages throughout the day. Excellent support and
advice can be obtained from patient support groups, such as the
encephalitis society (www.encephalitis.info).
Other disabilities in the recovery period or afterwards include
post-encephalitic seizures. Following VE, epilepsy is more likely
in children who had recurrent seizures, status epilepticus, severe
disturbance of consciousness, focal neurological signs or who
deteriorated during hospital admission.25 If seizures return, then
a more appropriate long term anticonvulsant can be chosen, such
as carbemazepine or lamotrigine.
Prognostic factors and outcome
Although with aciclovir treatment the mortality of HSV
encephalitis has decreased from about 70% to about 20%,
PAEDIATRICS AND CHILD HEALTH 18:1
the morbidity remains high.5 Poor prognostic factors include
reduced GCS on admission and delays between hospitaliza-
tion and starting aciclovir treatment (especially delays of
greater than 2 days).26 Two-thirds of survivors have signifi-
cant neuropsychiatric sequelae, including memory impairment
(69%), personality and behavioural change (45%), dysphasia
(41%) and epilepsy (up to 25%).27
Summary
Although VE is relatively uncommon in children in the UK,
early recognition and appropriate investigation and manage-
ment is essential, because of the devastating nature of the
­condition.
In the UK, HSV-1 is the commonest cause and treatment with
intravenous aciclovir for at least 14 days is now the standard
management. Diagnostic techniques and areas within the man-
agement have improved (neurointensive care and rehabilitation).
Despite these improvements, many long term sequelae are com-
mon. Therefore, although it is uncommon, VE is a devastating
condition for children and their families and the personal and
financial costs are high. It is therefore very important that VE is
recognized and managed appropriately, and that paediatricians
are up to date with recent developments and follow a rational
approach to investigation and treatment.
Funding
TS is a UK Medical Research Council Senior Clinical Fellow. ◆
14 © 2007 Published by Elsevier Ltd.
 Symposium: Neurology
References
1 Koskiniemi M, Korppi M, Mustonen K, et al. Epidemiology of
encephalitis in children. A prospective multicentre study. Eur J
Pediatr 1997; 156: 541–545.
2 Solomon T, Hart I, Beeching N. Viral encephalitis: a clinician’s guide.
Pract Neurol 2007; 7: 288–305.
3 Univeristy of Liverpool. www.liv.ac.uk/braininfections
4 Kennedy PG. Viral encephalitis: causes, differential diagnosis, and
management. J Neurol Neurosurg Psychiatry 2004; 75(suppl 1):
i10–15.
5 Whitley RJ. Herpes simplex virus infection. Semin Pediatr Infect Dis
2002; 13: 6–11.
6 Esiri MM. Herpes simplex encephalitis. An immunohistological study
of the distribution of viral antigen within the brain. J Neurol Sci
1982; 54: 209–226.
7 Kumar S, Kumar S, Kohlhoff SA. Recurrent HSV-2 meningitis in a
9-year-old girl. Scand J Infect Dis 2006; 38: 570–572.
8 Chua KB, Goh KJ, Wong KT, et al. Fatal encephalitis due to Nipah
virus among pig-farmers in Malaysia. Lancet 1999; 354: 1257–1259.
9 McMinn PC. An overview of the evolution of enterovirus 71 and its
clinical and public health significance. FEMS Microbiol Rev 2002;
26: 91–107.
10 Fodor PA, Levin MJ, Weinberg A, et al. Atypical herpes simplex virus
encephalitis diagnosed by PCR amplification of viral DNA from CSF.
Neurology 1998; 51: 554–559.
11 Kneen R, Solomon T, Appleton R. The role of lumbar puncture in
suspected CNS infection - a disappearing skill? Arch Dis Child 2002;
87: 181–183.
12 Kneen R, Solomon T, Appleton R. The role of lumbar puncture in
children with suspected central nervous system infection. BMC
Pediatr 2002; 2: 8.
13 Chataway J, Davies NW, Farmer S, et al. Herpes simplex encephalitis:
an audit of the use of laboratory diagnostic tests. QJM 2004; 97:
325–330.
14 Jeffery KJ, Read SJ, Peto TE, et al. Diagnosis of viral infections of the
central nervous system: clinical interpretation of PCR results. Lancet
1997; 349: 313–317.
15 Cinque P, Cleator GM, Weber T, et al. The role of laboratory
investigation in the diagnosis and management of patients with
suspected herpes simplex encephalitis: a consensus report. The
EU Concerted Action on Virus Meningitis and Encephalitis. J Neurol
Neurosurg Psychiatry 1996; 61: 339–345.
16 Tien RD, Felsberg GJ, Osumi AK. Herpesvirus infections of the CNS:
MR findings. AJR Am J Roentgenol 1993; 161: 167–176.
17 McCabe K, Tyler K, Tanabe J. Diffusion-weighted MRI abnormalities
as a clue to the diagnosis of herpes simplex encephalitis. Neurology
2003; 61: 1015–1016.
18 Lin JJ, Lin KL, Wang HS, et al. Analysis of status epilepticus related
presumed encephalitis in children. Eur J Paediatr Neurol 2007
Jun 19 (epub ahead of print).
19 Whitley RJ, Alford CA, Hirsch MS, et al. Vidarabine versus acyclovir
therapy in herpes simplex encephalitis. N Engl J Med 1986; 314:
144–149.
20 Yamada S, Kameyama T, Nagaya S, et al. Relapsing herpes simplex
encephalitis: pathological confirmation of viral reactivation. J Neurol
Neurosurg Psychiatry 2003; 74: 262–264.
21 Coren ME, Buchdahl RM, Cowan FM, et al. Imaging and laboratory
investigation in herpes simplex encephalitis. J Neurol Neurosurg
Psychiatry 1999; 67: 243–245.
PAEDIATRICS AND CHILD HEALTH 18:1
22 Chan PK, Chow PC, Peiris JS, et al. Use of oral valaciclovir in a
12-year-old boy with herpes simplex encephalitis. Hong Kong Med J
2000; 6: 119–21.
23 Kamei S, Sekizawa T, Shiota H, et al. Evaluation of combination therapy
using aciclovir and corticosteroid in adult patients with herpes simplex
virus encephalitis. J Neurol Neurosurg Psychiatry 2005; 76: 1544–1549.
24 Solomon T, Dung NM, Wills B, et al. Interferon alfa-2a in Japanese
encephalitis: a randomised double-blind placebo-controlled trial.
Lancet 2003; 361: 821–826.
25 Lee WT, Yu TW, Chang WC, et al. Risk factors for postencephalitic
epilepsy in children: A hospital-based study in Taiwan. Eur J
Paediatr Neurol 2007; 11: 302–309.
26 Skoldenberg B, Forsgren M, Alestig K, et al. Acyclovir versus
vidarabine in herpes simplex encephalitis. Randomised multicentre
study in consecutive Swedish patients. Lancet 1984; 2: 707–711.
27 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.
28 Solomon T, Whitley RJ. Arthropod-borne viral encephalitides. In:
Scheld M, Whitley RJ, Marra C, eds. Infections of the Central Nervous
System, 2nd Edn. Philadelphia: Lippincott, Williams and Wilkins,
2004.
Useful websites and further reading
Appleton R, Baldwin T, eds. Management of brain injured children.
2nd Edn. Oxford: OUP, 2006, Book with useful information about
rehabilitation of children with acute brain injury from any cause.
Encephalitis Society. Can provide support for patients and their
families. Also provides teaching, training and educational materials
for health care professionals. www.encephalitis.info
Health protection Agency funded Prospective Study of Viral Encephalitis.
www.hpa.org.uk/infections/topics%5Faz/encephalitis/study.htm
Kimberlin D. Herpes simplex virus, meningitis and encephalitis in
neonates. Herpes 2004; 11(suppl 2): 65A–76A. Review of neonatal
herpes simplex encephalitis.
Practice points
• Viral encephalitis (VE) is uncommon but suspected CNS
infections are not. Paediatricians should have a rational
approach to investigation and management of children with
suspected CNS infection
• A definitive diagnosis is made in approximately 30% of cases
of VE. This figure can be improved by taking advice from a
virologist or paediatric neurologist
• A lumbar puncture is essential in patients with suspected VE. It
may be delayed in patients with suspected brain shift. Results
from CSF analysis are useful even after treatment has been
started. For HSV encephalitis, PCR remains positive in about
80% of patients even 1 week after starting antiviral therapy
• The commonest cause of VE in the UK is HSV-1. Globally the
commonest cause is JEV
• The mortality and morbidity from VE are high. Both have
decreased for HSV encephalitis since aciclovir became the
15 © 2007 Published by Elsevier Ltd.
 Symposium: Neurology

standard treatment. However, specific treatments are not Acknowledgments

available for most other types of VE
• The incidence of VE may be increasing. Global travel and We thank Dr Nick Beeching and Dr Ian Hart for helpful
environmental changes may mean we see an increase in new ­discussions, Mrs Margaret Beirne and Dr Richard Appleton for
outbreaks of VE in the future the EEG and Dr Lawrence Abernethy for CT and MRI scan images.

PAEDIATRICS AND CHILD HEALTH 18:1 16 © 2007 Published by Elsevier Ltd.

 Symposium: neurology
Evaluation of the floppy
infant
Vasantha Gowda
Jeremy Parr
Sandeep Jayawant
Abstract
This review outlines a clinical approach to the evaluation of the floppy
infant. Attention is drawn to the varied manner in which the condition
can present, and emphasis is placed upon a detailed assessment of
characteristic clinical findings. A distinction is drawn between central
and peripheral causes for hypotonia. Guidance is given regarding the
importance of evaluating the child for signs of weakness, which is an
important marker of neuromuscular pathology. Reference is made to
situations where peripheral pathology may mimic central disorders. A
diagnostic algorithm is outlined for the investigation of neuromuscular
disorders, and reference is made to the discrepancy in findings that often
exists between electromyography and muscle biopsy findings. Attention
is drawn to available therapeutic options, as well as the importance of
addressing ethical issues, which become of particular importance once
a diagnosis is reached.
Keywords hypotonia; neuromuscular; neuropathy
Introduction and definition
The word ‘floppy’ can be used to mean:
• decrease in muscle tone (hypotonia);
• decrease in muscle power (weakness);
• ligamentous laxity and increased range of joint mobility.
Strictly speaking, the term ‘floppy’ should be used to describe
hypotonia. The interconnection between tone, muscle strength
and joint mobility can be appreciated through a consideration
of the definition of tone – the resistance to passive movement
around a joint. Phasic tone is assessed by the response of the
muscle to a rapid stretch, illustrated classically by a tendon
Vasantha Gowda MBBS MRCP is a Specialist Registrar, Department
Q1
of Paediatric Neurology, John Radcliffe Hospital, Headley Way,
Headington, Oxford OX3 9DU, UK.
Jeremy Parr MBChB MD MRCPCH is a Clinical Lecturer, Department
of Paediatric Neurology, John Radcliffe Hospital, Headley Way,
Headington, Oxford OX3 9DU, UK.
Sandeep Jayawant MD FRCP FRCPCH is a Consultant Paediatric Neurologist,
Department of Paediatric Neurology, John Radcliffe Hospital, Headley
Way, Headington, Oxford OX3 9DU, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
reflex, whilst postural tone is measured by the response of
the muscle to a sustained low-intensity stretch, as illustrated
by the body’s ability to maintain posture against the force of
gravity.
Clinical appearance
Some features are common to all floppy infants regardless of the
aetiology and location of the abnormality. A child is generally
said to be floppy if he/she assumes a frog-like posture, is unable
to maintain normal posture against gravity, exhibits diminished
resistance to passive movements and has an excessive range
of joint mobility. Table 1 lists some of the clinical signs with
which a floppy infant may present; these features may or may
not co­exist in the same infant.
Common modes of presentation
The clinical consequences of hypotonia and/or weakness may
be evident even in antenatal life.1 Specific questions in the his-
tory should address whether fetal movements were normal, as
well as whether there was evidence of polyhydramnios. In the
neonatal period, the manner of presentation depends on the
severity of the condition. This ranges from the consequences
of fetal immobilization, such as hip dislocation, arthrogryposis,
talipes and flexion deformity of all limbs, to respiratory and
feeding difficulties (slow feeding, recurrent choking or aspira-
tion episodes). Later in infancy, hypotonia may be more obvi-
ous once delayed achievement of motor milestones becomes
evident, with or without accompanying delay in other areas of
development.
Clinical confirmation of hypotonia
Once the suspicion of hypotonia has been raised, the evaluation
of the floppy infant should proceed by searching for those clinical
signs that corroborate the diagnosis (Figures 1 and 2).
Diagnostic approach
The initial approach to a floppy infant is to determine whether
the problem is of central or peripheral origin. This is of crucial
importance when forming a plan for diagnostic investigations.
As a general rule, an attempt is made to gauge whether there
Clinical signs in a floppy infant
• Observation of a ‘frog-leg’ posture. This generally implies
reduced spontaneous movement, with the legs fully abducted
and arms lying beside the body either extended or flexed
• Significant head lag on traction or pull-to-sit manoeuvre and
excessively rounded back when sitting (>33 weeks)
• Rag-doll posture on ventral suspension
• Vertical suspension test – feeling of ‘slipping through the
hands’ when the infant is held under the arms
• Various associated examination findings such as flat occiput
or congenital dislocation of the hips, arthrogryposis
Table 1
17 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology

Clinical features suggestive of hypotonia of central

origin

• Social and cognitive impairment in addition to motor delay

• Dysmorphic features implying a syndrome or other organ
malformations sometimes implying a syndrome
• Fisting of hands
• Normal or brisk tendon reflexes
• Features of pseudobulbar palsy, brisk jaw jerk, crossed
adductor response or scissoring on vertical suspension
• Features that may suggest an underlying spinal dysraphism
• History suggestive of hypoxic-ischaemic encephalopathy, birth
trauma or symptomatic hypoglycaemia
• Seizures

Table 2

Table 3 shows features in the history and clinical examination

that indicate the hypotonia and/or weakness are more likely due
to peripheral (neuromuscular) causes. During infancy, formal
testing for both tone and weakness is more difficult and relies
more heavily on clinical evaluation as well as the search for
­corroborating risk factors.
In some conditions, both central and peripheral hypotonia
may coexist (Table 4).
Once the decision about central or peripheral (neuromuscular)
aetiology has been made, investigations are directed accordingly
to identify the specific condition causing the hypotonia. Tables
Figure 1 Head lag. 5 and 6 list some of the causes of hypotonia and/or weakness
resulting from central and peripheral (neuromuscular) causes,
­respectively.
is a significant element of weakness. Paralytic hypotonia with
­significant weakness suggests a peripheral neuromuscular prob-
lem, whereas non-paralytic hypotonia without significant weak-
ness points to a central cause which may be neurological, genetic,
syndromic or metabolic.
A central cause is suggested by clinical features as described Indicators of peripheral hypotonia
in Table 2. The presence of these findings does not, however,
exclude a peripheral aetiology of weakness. In some cases fea- • Delay in motor milestones with relative normality of social
tures suggesting both central and peripheral aetiologies may and cognitive development
be seen. • Family history of neuromuscular disorders/maternal myotonia
• Reduced or absent spontaneous antigravity movements,
reduced or absent deep tendon jerks and increased range of
joint mobility
• Frog-leg posture or ‘jug-handle’ posture of arms in
association with marked paucity of spontaneous movement
• Myopathic facies (open mouth with tented upper lip, poor
lip seal when sucking, lack of facial expression, ptosis and
restricted ocular movements)
• Muscle fasciculation (rarely seen but of diagnostic importance
when recognized)
• Other corroborative evidence including muscle atrophy,
muscle hypertrophy and absent or depressed deep tendon
reflexes

Figure 2 Classic posture in hypotonia. Table 3

PAEDIATRICS AND CHILD HEALTH 18:1 18 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Conditions where central and peripheral hypotonia
may coexist
• Familial dysautonomia
• Hypoxic–ischaemic encephalopathy
• Infantile neuroaxonal degeneration
• Lipid storage diseases
• Lysosomal disorders
• Mitochondrial disorders
• Perinatal asphyxia secondary to motor unit disease
Table 4
Investigations where central hypotonia is suspected
The range of investigations that may yield diagnostic clues in
the clinical setting of non-paralytic/central hypotonia are listed
in Table 7. The diagnosis of central hypotonia attributable to
hypoxic-ischaemic encephalopathy is based upon an appropri-
ate clinical history along with concordant imaging studies of the
brain. Metabolic disorders are rare and despite extensive investi-
gations the diagnostic yield may be quite low.
Investigations where peripheral hypotonia is suspected
Investigations likely to yield diagnostic clues in cases where para-
lytic/neuromuscular hypotonia is suspected are listed in Table 8.
Myotonic dystrophy is usually suspected on the basis of
known family history or recognition of the disorder in an affected
mother, and can be confirmed by testing for the expanded CTG
Conditions associated with central (non-paralytic)
hypotonia
Acute encephalopathies
• Birth trauma
• Hypoxic-ischaemic encephalopathy
• Hypoglycaemia
Chronic encephalopathies
• Cerebral malformations
• Inborn errors of metabolism (mucopolysaccharidoses,
aminoacidurias, organic acidurias, lipidoses, glycogen
storage diseases, Menkes syndrome)
• Chromosomal disorders (Prader-Willi syndrome, trisomy 21)
• Genetic disorders (familial dysautonomia, Lowe syndrome)
• Peroxisomal disorders (neonatal adrenoleukodystrophy,
Zellweger syndrome)
• Endocrine (hypothyroidism)
• Metabolic (rickets, renal tubular acidosis)
Connective tissue disorders
• Ehlers-Danlos syndrome
• Osteogenesis imperfecta
• Congenital ligamentous laxity
• Benign congenital hypotonia
Table 5
PAEDIATRICS AND CHILD HEALTH 18:1
Causes of paralytic/neuromuscular hypotonia
Spinal muscular atrophy
Paralytic poliomyelitis
Neuropathies
• Hereditary motor-sensory neuropathy
• Congenital hypomyelinating neuropathy
• Acute demyelinating polyneuropathy
Neuromuscular junction problems
• Botulism
• Transient neonatal myasthenia
• Autoimmune myasthenia
• Congenital myasthenic syndromes
Muscular disorders
• Congenital myopathies (nemaline rod myopathy,
myotubular myopathies, central core disease, minicore
disease, etc)
• Congenital muscular dystrophies (CMD) (Walker-Warburg,
Fukuyama, muscle-eye-brain disease, merosin-positive
CMD, etc)
• Congenital myasthenic syndromes
• Congenital myotonic dystrophy
• Metabolic myopathies (acid maltase deficiency,
phosphorylase deficiency, mitochondrial myopathy
• Endocrine myopathies (hypothyroidism)
Table 6
trinucleotide repeat sequence on chromosome 19q13.2–q13.3.
In the absence of a diagnosis of myotonic dystrophy, we sup-
port the early use of electromyography (EMG)/nerve conduc-
tion studies (NCS) when a peripheral cause is likely.2 Areflexia,
decreased limb movements and denervation on EMG should
prompt investigation for anterior horn cell disorders (­spinal
Investigations in cases where a central cause for
hypotonia is suspected
• Serum electrolytes, including calcium and phosphate, serum
alkaline phosphatase, venous blood gas, thyroid function
tests
• Plasma copper/caeruloplasmin assay (as screening test for
Menkes syndrome)
• Chromosomal analysis (trisomy), testing for Prader-Willi
syndrome(15q11–13)
• Plasma amino acids and urine organic acids
• Urine mucopolysaccharide screen (GAG)
• Molecular/biochemical diagnosis of pro-collagen disorders
• Very long chain fatty acids
• Medical genetics opinion
• Ophthalmology opinion
• Brain imaging (CT/MRI)
Table 7
19 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Investigations of peripheral hypotonia
• Creatinine kinase
• Lactate
• EMG/NCS/repetitive nerve stimulation test
• Muscle biopsy (histology, immunohistochemistry, electron
microscopy, respiratory chain enzyme analysis)
• Genetic testing (SMN gene deletion present in 95% of cases
of spinal muscular atrophy type 1, myotonic dystrophy,
congenital myasthenic syndromes)
• Nerve biopsy (rarely)
• Tensilon test
Table 8
­ uscular atrophy), and can be confirmed by testing for the
m findings in myopathic and neurogenic disorders are outlined in
homozygous deletion of exon 7 in the telomeric survival motor
neurone gene.3 Failure to identify electrophysiological abnor-
malities should prompt testing for Prader-Willi syndrome.
Arthrogryposis, feeding difficulties, recurrent apnoeic/choking
episodes, ­ophthalmoplegia, ­ptosis and fatigability should prompt
­investigations for congenital myasthenic syndromes. Finally,
muscle biopsy is recommended in neonates with weakness,
even if needle EMG is normal (Figure 3). Care should, however,
be taken in patient selection for muscle biopsy because of an
increased risk of anaesthetic complications (postoperative respi-
ratory failure and reactions to anaesthetic agents, particularly
malignant hyperthermia and rhabdomyolysis).
EMG studies are used as a supportive diagnostic tool in
deciding whether there is true weakness due to neuromus-
cular disease, or hypotonia from causes in other systems or
other parts of the nervous system, and whether the process
is due to a myopathic, neuropathic or a denervating process.4
In general, EMG can be performed at any age, although some
caution is required in the interpretation of results within the
first 6–8 weeks of life, particularly if the baby was premature.
Whilst severe myopathy or neuropathy is not usually difficult
to diagnose on EMG, studies on cases of mild weakness are
Left, muscle biopsy (histochemistry) and right, electron microscopy in nemaline rod myopathy (congenital myopathy).
Figure 3
PAEDIATRICS AND CHILD HEALTH 18:1
Useful EMG features in peripheral hypotonia
• EMG/NCS studies may distinguish between neurogenic,
myopathic and myasthenic aetiologies for hypotonia
• Neurogenic – large amplitude action potentials, reduced
interference pattern, increased internal instability
• Myopathic – small amplitude action potentials with increased
interference pattern
• Myotonic – increased insertional activity
• Myasthenic – abnormal repetitive and single fibre studies
Table 9
usually inconclusive.5 In general, EMG and biopsy studies more
often concur in denervation than myopathy. Classic EMG/NCS
Table 9. Recent advances in the diagnosis of congenital myas-
thenic syndromes allow for DNA testing in some of the common
syndromes where there is strong clinical suspicion, particularly
in young children where accurate neurophysiology evidence is
sometimes difficult to obtain.6
Therapeutic approach
Various aspects of function may be affected in a floppy infant.
In most cases, supportive therapies are indicated. Few condi-
tions have specific treatments. These include hypothyroidism
(thyroxine), some types of congenital myasthenic syndromes
(pyridostigmine or neostigmine) and rickets (vitamin D). Some
metabolic disorders may respond to specific dietary modifications
or enzyme replacement therapies. The mainstays of ­ treatment
are outlined in Table 10.
Prognosis and prevention
Rapid as well as accurate diagnosis of individual cases is vital
in order to provide precise prognostic information. Ethical con-
siderations, such as the appropriateness of cardiopulmonary
20 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Principles of management
• Physiotherapy - stretches aimed at prevention of contractures
• Occupational therapy - appliances, improvement of posture
and function, facilitating activities of daily living
• Prevention and correction of scoliosis
• Evaluation and treatment of associated cardiac dysfunction
• Respiratory support - assessment of requirement for invasive
or non-invasive ventilation and/or tracheostomy
• Feeding - nasogastric feeding, caloric supplementation,
gastrostomy
• Management of gastro-oesophageal reflux - medical or
fundoplication
• Orthopaedic intervention in setting of established or evolving
joint contractures
• Encouragement of overall development and stimulation of
learning
• Prevention (influenza and pneumococcal vaccination) and
prompt treatment of respiratory infections
Table 10
resuscitation in the event of cardiac arrest or acute respiratory
failure, need to be addressed sensitively. Informed discussion
around these issues requires detailed knowledge, where avail-
able, of specific conditions and, in particular, their presentation,
clinical features, course and outcomes. Prenatal diagnosis using
amniocentesis or chorionic villus sampling is often feasible if a
definitive diagnosis has been reached in the index case. ◆
References
1 Fenichel GM. Pediatric neurology. Philadelphia: WB Saunders
Company, 2005.
2 Richer LP, Shevell MI, Miller SP. Diagnostic profile of neonatal
hypotonia: an 11-year study. Pediatr Neurol 2001; 25: 32–37.
3 Nicole S, Diaz CC, Frugier T, et al. Spinal muscular atrophy: recent
advances and future prospects. Muscle Nerve 2002; 26: 4–13.
PAEDIATRICS AND CHILD HEALTH 18:1
4 Brett EM. Pediatric neurology. Edinburgh: Churchill Livingstone,
1997.
5 Russell JW, Afifi AK, Ross MA. Predictive value of electromyography
in diagnosis and prognosis of the hypotonic infant. J Child Neurol
1992; 7: 387–389.
6 Parr JR, Jayawant S. Childhood myasthenia:clinical subtypes and
practical management. Dev Med Child Neurol 2007; 49: 629–635.
Recommended reading
Dubowitz V. Muscle disorders in childhood. London: WB Saunders
Company, 1995.
Practice points
• Many neurological and non-neurological disorders can
present with hypotonia
• Central causes of hypotonia must be distinguished from
those due to neuromuscular disease
• Neuromuscular causes are usually characterized by the
presence of significant weakness
• Neurophysiological studies are a helpful initial investigation
• Muscle biopsy is usually required to diagnose muscle disease
• Some conditions can be diagnosed on genetic testing alone
Recent developments
• Recent advances in genetics have uncovered new conditions
causing hypotonia and weakness such as congenital
myasthenic syndromes and spinal muscular atrophy variants
• Advances in immunohistochemistry, electron microscopy and
genetics have led to a more specific diagnosis of myopathies
• Some of these advances have allowed for specific therapeutic
interventions, e.g. use of acetylcholinesterase inhibitors in
some congenital myasthenic syndromes
21 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Recent developments in the
management of Duchenne
muscular dystrophy
Maria Kinali
Adnan Y Manzur
Francesco Muntoni
Abstract
Duchenne muscular dystrophy (DMD) is the most common and severe
childhood muscular dystrophy, resulting in progressive muscle weakness
and wasting, disability and decreased survival. Although the molecular
defect in DMD is known, no curative treatment is available. The treat-
ment options of glucocorticoid corticosteroids and supportive measures,
such as ventilation and management of cardiac insufficiency, have be-
come accepted clinical practice in the last decade, and these are of
major interest to the paediatrician as they have significantly changed the
course of the disease in treated individuals. This has implications not
only for the affected individual and his family, but also for the medical
and social sectors to provide transition to adult medical services and for
provision of suitable employment, and social care.
Several experimental therapeutic strategies, including cell and gene
therapy, are promising, with encouraging results in relevant animal mod-
els and in cultured human cells. As a number of approaches are in early
clinical trials, expectations are raised of their impact as a cure for DMD;
nevertheless, it is not realistic to expect that these approaches will have
a substantial impact on disease course in the short term. While waiting
for a curative therapy to become available, symptomatic and palliative
treatment is essential to enhance the patient’s quality of life. This review
addresses the advances in these therapies aimed at improving func-
tion and quality of life in patients with DMD, and the current status of
­research into the DMD experimental therapies.
Keywords Duchenne muscular dystrophy; management; therapy
Maria Kinali MD MRCPCH is a Senior Clinical Research Fellow at the
Dubowitz Neuromuscular Centre, Department of Paediatrics, Division of
Medicine, Hammersmith Campus, Imperial College, London.
Adnan Y Manzur DCH FRCPCH is a Consultant Paediatrician at the
Dubowitz Neuromuscular Centre, Department of Paediatrics, Division of
Medicine, Hammersmith Campus, Imperial College, London.
Francesco Muntoni MD FRCPCH FMed Scie is a Professor of Paediatric
Neurology at the Dubowitz Neuromuscular Centre, Department of
Paediatrics, Division of Medicine, Hammersmith Campus, Imperial
College, London.
PAEDIATRICS AND CHILD HEALTH 18:1
Introduction
Duchenne muscular dystrophy (DMD) affects 1 in every 3500
live male births. DMD is the first genetic condition for which the
defect was identified by positional cloning. Approximately 65%
of patients with DMD have intragenic out-of-frame (gross rear-
rangements) deletions and approximately 10% have duplications
of one or more exons of the dystrophin gene.1 The remaining
patients have point mutations or other smaller gene rearrange-
ments (pure intronic deletions, insertions of repetitive sequences,
splice site mutations). Depending on the type of mutation, there
may be a severe reduction or absence of dystrophin in the muscle,
resulting in DMD phenotype, or a partly functional truncated pro-
tein, resulting in the milder Becker muscular dystrophy (BMD).1
Dystrophin links the cytoskeleton to the extracellular matrix.1
Lack of dystrophin compromises this link, leading to muscle fibre
degeneration. Dystrophin also appears to play a role in signal
transduction, although this has a limited role in the pathogenesis
of the muscle degeneration.
The common presentation of DMD is with abnormal gait and
difficulty in rising from the floor, which becomes clinically evi-
dent between the ages of 3 and 5 years. Progression of muscle
weakness and contractures of the tendon Achilles (TA) leads to
loss of walking at a mean age of 9.5 years. Wheelchair depen-
dence is associated with progressive scoliosis. In untreated indi-
viduals, the leading cause of death is respiratory insufficiency
in the late teens or early 20s. Approximately 80% of deaths are
related to respiratory complications, while 20% succumb to
dilated cardiomyopathy (DCM). Feeding difficulties and weight
loss are common in the late stages of the disease. No curative
treatment for DMD is known, but the quality of life and comfort
of the patient can be improved by symptomatic physiotherapeutic
and medical treatments.
Symptomatic treatment modalities
Physiotherapy
Physiotherapy to promote walking and prevent joint deformities
remains the mainstay of treatment. Prevention/control of lower
limb contractures is achieved by a regular passive stretching of
the TA and the use of night splints once the ankles cannot be
dorsiflexed beyond the neutral position. Exercise programmes,
such as swimming, are recommended, while activity against
resistance is not. Detailed recommendations for exercise and
physiotherapy are available.2
Prolongation of walking with orthoses
Rehabilitation in knee–ankle–foot orthoses (KAFOs) is effective
in prolonging walking for an average of 18 months to 2 years.
Prolongation of walking beyond 13 years of age protects against
rapidly progressive scoliosis.3 KAFOs are offered to boys with
DMD at the end of independent ambulation, i.e. when they can
take only a few steps with/without assistance. The technique
involves custom-built KAFOs and surgical release of the TA, or
serial casting to reduce ankle contracture and allow fitting of
KAFOs. Regular review at a centre with rehabilitation facilities
has the advantages of psychological preparation of the child and
family for loss of walking, and choosing the appropriate timing
for intervention.
22 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Pharmacotherapy
Corticosteroids: of the pharmacological agents tried in DMD, glu-
cocorticoid corticosteroids are the most effective. The Cochrane
systematic review of the glucocorticoids in DMD4 and European
Neuromuscular Centre Steroids in DMD workshop report5 review
the evidence and summarize the benefits, risks, research and
clinical aspects of this treatment.
Randomized controlled trials (RCT) have shown that treatment
with prednisolone can stabilize strength and function for 6 months
to 2 years.4 Prednisolone has been the most widely used medica-
tion and the starting dose is 0.75 mg/kg/day. Non-­randomized
studies with prednisolone or deflazacort have documented pro-
longation of walking ability, preservation of respiratory function
as indicated by forced vital capacity (FVC) and reduction in the
incidence of scoliosis and cardiomyopathy in boys with DMD who
tolerated long term daily doses of corticosteroids. The daily steroid
regimens have significant side effects, notably vertebral fractures
in approximately a third of patients treated long term. In order
to lessen the adverse effects, Dubowitz recommended an inter-
mittent regimen of prednisolone; a 6-month RCT of prednisolone
0.75 mg/kg/day for the first 10 days of every month demonstrated
slowing of functional deterioration.
Despite the uncertainties regarding the long term safety of cor-
ticosteroids in DMD, they have been shown to alter the natural
history in relation to ambulation, cardiomyopathy, respiratory
function and scoliosis. Corticosteroids should be preferably com-
menced in all early ambulant cases (4–7 years) and in most older
ambulant children, unless contraindicated. Treatment needs to be
monitored for benefit, adverse effects and dose adjustment. The
optimal starting dose of prednisone (0.75 mg/kg/day) is often
not tolerated in the long term and, over the course of years, the
dose often has to be decreased. Follow-up in a specialist centre
allows for appropriate monitoring, dosing and management of
adverse effects. Optimizing bone health in corticosteroid-treated
patients includes dietary advice regarding calcium and vitamin D,
and supplementation if plasma vitamin D levels are low.6 There
is currently no consensus on the prophylactic use of bisphospho-
nates, but they are indicated for treatment of vertebral fractures.
Creatine: a dietary component in meat eaters, is also synthesized
endogenously and is stored primarily in skeletal muscle. Creatine
in the muscle is converted to phosphocreatine to build and pro-
vide energy in the form of adenosine triphosphate. Creatine may
enhance the performance of high-­intensity, short-duration exer-
cise but is not useful in endurance sports. Creatine pre-­treatment
of mdx mice muscle cell cultures increased phosphocreatine
­levels, myotube formation and survival. A RCT of creatine sup-
plementation in DMD has failed, however, to show any signifi-
cant improvement in strength and function.7
Coenzyme Q10 and antioxidants: coenzyme Q10 (CoQ10) is
­synthesized endogenously, absorbed by the gastrointestinal sys-
tem and then bound in the inner mitochondrial membrane regu-
lating respiratory chain effects. Previous double blind trials have
shown low efficacy, making it difficult to rule out any disease
modifying effect. A phase III clinical trial is currently assessing
the effect of CoQ10 (serum level greater than 2.5 μg/ml) and
prednisolone (0.75 mg/kg/day) in wheelchair-dependent patients
with DMD (http://www.clinicaltrials.gov/).
PAEDIATRICS AND CHILD HEALTH 18:1
Beta-2 agonists: beta-2 adrenergic agonists have been shown to
induce muscle hypertrophy and prevent atrophy after physical and
biochemical insults in animal studies. A 12-week RCT showed that
albuterol increased muscle strength in knee extensors and flexors
in nine ambulant patients with DMD.8 These results need confirma-
tion in a larger RCT. The safety of albuterol on increasing heart rate,
especially in those with concurrent DCM, remains to be addressed.
Management of respiratory complications
The teenage years in DMD are marked by decreased respiratory
reserve and sleep hypoventilation, which is a sequel of respi-
ratory muscle weakness and rapid eye movement (REM) sleep
related hypoxemic dips. The spectrum of symptoms includes
morning drowsiness, poor appetite, headaches, nausea, fatigue,
tiredness, dysthymia, poor concentration at school, failure to
thrive, reduced coughing ability or overt respiratory failure in
the course of ‘minor respiratory infections’. In untreated patients
who have become hypercapnic, the survival is less than a year.9
Until recent decades, this signified imminent demise, as the
only option to prolong life was the use of a mechanical ventilator
through tracheostomy; this was limited by the complex ethical
issues of invasive ventilation of patients with totally incapacitat-
ing and incurable disease with no option for a ‘normal’ death.
In recent years, domiciliary non-invasive ventilation (NIV) has
proven effective in symptom relief and prolonging survival. The
patient’s breathing at night is augmented with breaths delivered
by a compact, portable ventilator with a snugly fitting facial or
nose mask. NIV corrects sleep hypoventilation without significant
noise, encroachment on living space or restriction of travel, and
together with the use of cough-assist devices, can extend survival
to the average mid-20s and in rare cases to the fourth decade.9
Because of these findings, many physicians have argued that
denying NIV to hypercapnic patients with DMD is unethical.
Regular monitoring for symptoms of sleep hypoventilation, FVC
and overnight sleep studies when the FVC falls below 60% allow
for timely initiation of NIV. Gradual initiation of NIV in individuals
with nocturnal hypercapnia but daytime normocapnia is advanta-
geous, as waiting for daytime ventilatory failure exposes patients
to minor chest infections and uncontrolled decompensation.10
Management of cardiac complications
DCM occurs in up to 90% of boys with DMD aged ≥18 years.
DCM becomes symptomatic only in a minority and is the cause of
death in 20%. It is likely that DCM may manifest in a larger pro-
portion of individuals in whom NIV prevents respiratory-related
morbidity. The optimal timing of introducing therapy for DCM
remains an unresolved issue. Duboc et al11 have shown that early
treatment with perindopril delayed the onset and progression of
left ventricular dysfunction. There is a debate in the literature
as to whether to treat a complication that is often asymptomatic
for a long time before deteriorating into clear-cut cardiac failure.
Current evidence on other forms of DCM shows early treatment
to be clearly superior to late therapy.12
While awaiting the results of RCTs, several consensus
documents have been prepared which recommend the use of
­angiotensin converting enzyme (ACE) inhibitors, beta blockers
and diuretics in patients with early cardiomyopathy.13 Aggrava-
tion of cardiac failure by nocturnal hypoventilation should be
looked for and treated.
23 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
The risk of cardiac involvement in carriers of DMD is approxi-
mately 10%, and this may occur in the absence of muscle ­weakness.
Genetic counselling should include informing carriers of the car-
diac risks, and their surveillance and treatment planned for.
Management of scoliosis
Scoliosis usually manifests after loss of walking, shows rapid
progression coincident with the pubertal growth spurt and
has a negative impact on respiratory function, feeding, seating
and comfort. Scoliosis and age to which walking is preserved
are interrelated, with much lower incidence in boys who have
walked for longer, having used steroids and/or KAFOs.14 A spi-
nal brace does not prevent progression of scoliosis, but may be
useful in postural management.
Surgical spinal fusion is indicated when the spinal curve
is progressing, and the optimum time for making the decision
is when the range of the curve’s Cobb angle is 20–40°.15 The
decision to offer surgery needs input from the multidisciplinary
team, and pre-operative assessment is essential to ensure that the
operation is safe and a time chosen for surgery when the FVC is
greater than 30% predicted for height and the cardiac function, as
­demonstrated by echocardiogram, is good. Spinal surgery can be
performed when the FVC is between 20% and 30%, but the risks
are greater and this should be undertaken in specialized centres.
Nutritional aspects
The spectrum of nutritional difficulties includes initial presen-
tation with failure to thrive, obesity during the late ambulant
phase, especially in corticosteroid treated individuals, and wast-
ing in the spinal surgery postoperative period and the late teen-
age years. All patients with DMD should be given dietary advice
to avoid obesity, which has negative implications for mobility,
especially when the patient is treated with daily ­steroids.
Young adults with DMD may have chewing and swallowing
difficulties, choking and fear of choking, and failure to thrive.
Appropriate facilities for weighing wheelchair-dependent ado-
lescent patients should be available in the clinics to allow for
regular weight monitoring. Patients with failure to thrive and/or
swallowing difficulties benefit from a dietetic and a speech and
language therapist’s assessment as nutritional supplementation
and observation of mealtimes and swallowing videofluoroscopy
allow further advice about postural management, feeding aids or
gastrostomy insertion.16
Survival and transition of care
It was common practice for the paediatrician to stay in charge
of the medical care of patients with DMD because the limited
survival into the third decade of life in only a minority of patients
was considered not to justify the transfer of care to the adult med-
ical teams with the attendant psychological impact on the young
person and family. This resulted in lack of formal arrangements
for transition of care to the adult teams. The improvements in
general care in the 1970s and the frequent provision of NIV from
the 1990s, has improved the mean survival in the UK to 27 years17
and further prolongation of survival is anticipated as a result of
corticosteroid treatment. This underlines the need for develop-
ment of good transition protocols and, in particular, improvement
in rehabilitation and social services for the adult with DMD.18
PAEDIATRICS AND CHILD HEALTH 18:1
Experimental therapies
The last decade has seen major advances in understanding of
the molecular genetics and pathogenesis of the muscular dystro-
phies, and this in turn has raised expectation of a curative treat-
ment with gene therapy.
Animal models
Two naturally occurring animal models for DMD have been cru-
cial in testing the safety and efficacy of gene therapy and other
new pharmacotherapies. The dystrophic golden retriever dog
(GRMD) has a point mutation, which induces exon skipping,
muscle wasting and weakness, contractures and decreased sur-
vival akin to the human disease. The mdx mouse has a stop
codon in exon 23, resulting in dystrophin-deficient muscle fibres,
but phenotypically the mouse is not weak and its survival is only
minimally limited compared to wild type.
Drugs affecting fibrosis: blocking endogenous muscle calcium
proteases and transforming growth factor β
Transforming growth factor β (TGF β) inhibits terminal differen-
tiation of human cultured myoblasts in vitro. Increased activity
of TGF β leads to failed regeneration in mdx mice.19 Losartan
(an angiotensin II type 1 receptor antagonist) antagonizes TGF β
and has been shown to ameliorate the signs of dystrophy in mdx
mice. This has inspired hope that use of a common drug such
as losartan could potentially be effective for all cases of DMD; a
RCT of losartan is currently planned in DMD.
Genetic approaches
Currently, adeno-associated virus (AAV) is the vector of choice for
gene therapy of muscular dystrophy. AAV is a small, non-envel-
oped, single-stranded DNA virus, which requires co-infection
with a second ‘helper’ virus for replication. The main advantage
of AAV resides in its efficiency in transducing skeletal and cardiac
muscles after intravenous injection. Wild-type AAV does not cause
any human disease and only induces a mild immune response.
There are several challenges of the AAV gene therapy
approach. The small genome size carried by the AAV is a draw-
back for the insertion of large transgenes such as dystrophin.
Removal of the non-essential portion of the dystrophin gene
has allowed ‘minigenes’ and smaller ‘microdystrophins’ to be
engineered but their effectiveness remains under investigation,
particularly in larger animal models. Stimulation of an undesir-
able immune response and toxicity can be minimized by using
muscle-specific promoters to restrict expression to target mus-
cle cells. As high viral load is necessary, substantial advances
in vector production and muscle transduction efficiency with
recently developed serotypes to lower AAV doses are impor-
tant20 before use in future clinical trials. Other hurdles are that
viral vectors rarely integrate into chromosomes and are lost from
cells contributing to muscle regeneration following multiple cell
divisions, and the concern of potential immune response to the
vector and/or transgene. Recombinant AAV vectors do not con-
tain any viral genes, but the vector capsid proteins are capable
of inducing humoral immune responses in animals, a major
obstacle to subsequent treatments. Transient immune suppres-
sion strategies to prevent any cell-mediated response to the vec-
tor itself are being explored,21 in addition to the possibility of
24 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
using different serotypes for the subsequent administrations. It
remains to be seen whether human dystrophin will provoke an
immune response in patients with DMD.
Utrophin upregulation
Utrophin, a protein homologous to dystrophin,22 binds to the
dystrophin–glycoprotein complex. In normal adults, utrophin is
localized at the post-synaptic membranes of the neuromuscular
junction, peripheral nerves and vascular bed of their skeletal
muscles. Utrophin is virtually absent at the sarcolemma of mature
muscle fibres but is highly expressed in the sarcolemma of fetal
and regenerating fibres and represents another indirect marker
of dystrophy. Utrophin can prevent dystrophic pathology in the
transgenic mdx mouse, leading to the hypothesis that utrophin
might be capable of supplanting dystrophin, without provoking
an immune response and possibly reducing the disease sever-
ity.22 The UK company VASTox recently discovered a drug com-
pound VOXC1100, which upregulates utrophin, and clinical trials
are planned. The wide applicability of such a technique holds
promise for different muscular dystrophies; however, potential
response differences between humans and the mdx mouse will
have to be addressed first.
Cell transplantation
Transplantation of myoblasts has demonstrated negative or very
modest results in DMD. The proof of principle came from studies
in the mdx mouse, which resulted in dystrophin expression.
The main limitation of this approach is that transplanted cells
are unable to migrate following direct injection, so any effect is
limited to the area of the injection. Moreover, transplanted cells
undergo a rapid and massive death (80%) soon after the injection,
though this can be prevented by immunosuppressive therapy.23
The need for direct injection of a large number of cells at high den-
sity in all muscles of the body makes this approach impractical.
The recent identification of subpopulations of stem cell
which can reach skeletal muscle with arterial injection has
boosted new hopes in the cellular approach. Stem cells have the
ability of self renewal and give rise to progenitor or precursor
cells before differentiating into other cells. A promising study
on the mdx mouse has shown that human-derived ­ pericytes,
usually embedded in the basal membrane of microvessels, can
differentiate into satellite stem cells.24 These studies have been
paralleled by the identification of similar cells in the GRMD
and their transplantation resulted in some functional success.25
Clinical trials using pericytes are now being planned in DMD.
Read-through of stop codon mutations
This technique is applicable to ∼7–10% of patients with DMD
who have nonsense mutations in the dystrophin gene. Aminogly-
cosides can cause misreading of the RNA code, allowing inser-
tion of alternative amino acids at the site of the mutated codon,
transcription and protein formation.26 This was originally shown
with gentamycin in the mdx mouse,26 although the results were
not replicated in human trials. More recently, PTC-124, a novel
orally administered molecule with a similar mechanism of action
as gentamycin, has been shown to be more efficient and less toxic
in two human clinical trials on healthy volunteers.27 A phase II
clinical trial in DMD is underway and preliminary manufacturer’s
data (October 2006) suggest that it is safe and well tolerated.
PAEDIATRICS AND CHILD HEALTH 18:1
Modification of dystrophin mRNA splicing: antisense oligomers
and exon skipping trials
Antisense oligonucleotides or oligomers (AOs) are research tools
used to knock down target gene expression. More recently, AOs
have also been used to redirect splicing and induce exon skipping.
The dystrophin gene is an ideal target for such a strategy: most
individuals with DMD have out-of-frame deletions or duplications,
which result in non-functional dystrophin proteins, while deletions
that maintain the open reading frame give rise to semi-functional
dystrophins and milder phenotypes (BMD). Internally-truncated
dystrophins are also expressed in so-called ‘revertant fibres’ – indi-
vidual dystrophin-positive fibres found in 50% of DMD patients
and in mdx mice.1 These revertant fibres arise via aberrant splicing
and unexpected skipping of exons, which flank the deleted exons;
this results in restoration of the open reading frame. In DMD it is
possible to target an exon which flanks an out-of-frame deletion so
that the reading frame can be restored and dystrophin production
allowed. This can be achieved by AOs which prevent the normal
splicing of genes by masking crucial areas of the messenger RNA
during the splicing process. The early proof of concept studies
were done in cell cultures of the mdx mouse and subsequently
in DMD cells. More recently, the direct injection of AOs into the
muscle of mdx mice resulted in robust restoration of dystrophin
expression at the sarcolemma.28 Systemic administration of AOs
in the mdx mouse also resulted in appreciable induction of exon
skipping, and repeated intravenous administration of AOs resulted
in functional levels of dystrophin expression in body-wide skel-
etal muscles of the mdx mouse, with corresponding improvement
in muscle function.29 Despite these encouraging results, there are
several limitations of AOs. First, different deletions will require
different AOs and, second, the treatment is not permanent but
limited to the period in which the AO persists in the tissue. AO
treatment will therefore require repeated administrations for the
entire life of the patient with DMD, and whether this will be asso-
ciated with any toxicity is not known. AOs nevertheless have a
fairly good safety profile from data available on human trials. Two
European consortia in the Netherlands (http://prosensa.eu/news/
news_may10_06.pdf) and the UK (http://clinicaltrials.gov/ct/gui/
show/NCT00159250?order=1) are currently testing the safety and
local efficacy of intramuscularly administered AOs and results are
expected in late 2007. These results will inform the feasibility of
future systemic delivery studies, which are planned for 2008.
Disclosures
Two of the authors (FM and MK) are involved in a phase I/II
trial using morpholino antisense oligomers in DMD. The study
is funded by the Department of Health, and Imperial College
­London is the study sponsor. ◆
References
1 Muntoni F, Torelli S, Ferlini A. Dystrophin and mutations: one gene,
several proteins, multiple phenotypes. Lancet Neurol 2003; 2:
731–740.
2 Eagle M. Report on the muscular dystrophy campaign workshop:
exercise in neuromuscular diseases Newcastle, January 2002.
Neuromuscul Disord 2002; 12: 975–983.
25 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
3 Rodillo EB, Fernandez-Bermejo E, Heckmatt JZ, et al. Prevention of
rapidly progressive scoliosis in Duchenne muscular dystrophy by
prolongation of walking with orthoses. J Child Neurol 1988; 3: 269–274.
4 Manzur AY, Kuntzer T, Pike M, et al. Glucocorticoid corticosteroids
for Duchenne muscular dystrophy. Cochrane Database Syst Rev
2004(2): CD003725.
5 Bushby K, Muntoni F, Urtizberea A, et al. Report on the 124th
ENMC International Workshop. Treatment of Duchenne muscular
dystrophy; defining the gold standards of management in the
use of corticosteroids. 2–4 April 2004, Naarden, The Netherlands.
Neuromuscul Disord 2004; 14: 526–534.
6 Quinlivan R, Roper H, Davie M, et al. Report of a Muscular Dystrophy
Campaign funded workshop Birmingham, UK, January 16th 2004.
Osteoporosis in Duchenne muscular dystrophy; its prevalence,
treatment and prevention. Neuromuscul Disord 2005; 15: 72–79.
7 Escolar DM, Buyse G, Henricson E, et al. CINRG randomized
controlled trial of creatine and glutamine in Duchenne muscular
dystrophy. Ann Neurol 2005; 58: 151–155.
8 Fowler EG, Graves MC, Wetzel GT, et al. Pilot trial of albuterol in
Duchenne and Becker muscular dystrophy. Neurology 2004; 62:
1006–1008.
9 Simonds AK. Recent advances in respiratory care for neuromuscular
disease. Chest 2006; 130: 1879–1886.
10 Ward S, Chatwin M, Heather S, et al. Randomised controlled trial
of non-invasive ventilation (NIV) for nocturnal hypoventilation
in neuromuscular and chest wall disease patients with daytime
normocapnia. Thorax 2005; 60: 1019–1024.
11 Duboc D, Meune C, Lerebours G, et al. Effect of perindopril on the
onset and progression of left ventricular dysfunction in Duchenne
muscular dystrophy. J Am Coll Cardiol 2005; 45: 855–857.
12 Bourke JP. Cardiac monitoring and treatment for children and
adolescents with neuromuscular disorders. Dev Med Child Neurol
2006; 48: 164.
13 Bushby K, Muntoni F, Bourke JP. 107th ENMC international
workshop: the management of cardiac involvement in muscular
dystrophy and myotonic dystrophy. 7th–9th June 2002, Naarden, the
Netherlands. Neuromuscul Disord 2003; 13: 166–172.
14 Kinali M, Main M, Eliahoo J, et al. Predictive factors for the
development of scoliosis in Duchenne muscular dystrophy. Eur J
Paediatr Neurol 2007; 11: 160–166.
15 Muntoni F, Bushby K, Manzur AY. Muscular Dystrophy Campaign
Funded Workshop on Management of Scoliosis in Duchenne
Muscular Dystrophy 24 January 2005, London, UK. Neuromuscul
Disord 2006; 16: 210–219.
16 Pane M, Vasta I, Messina S, et al. Feeding problems and weight gain
in Duchenne muscular dystrophy. Eur J Paediatr Neurol 2006; 10:
231–236.
17 Eagle M, Bourke J, Bullock R, et al. Managing Duchenne muscular
dystrophy – the additive effect of spinal surgery and home nocturnal
ventilation in improving survival. Neuromuscul Disord 2007; 17:
470–475.
18 Parker AE, Robb SA, Chambers J, et al. Analysis of an adult
Duchenne muscular dystrophy population. QJM 2005; 98: 729–736.
19 Friedman KJ, Kole J, Cohn JA, et al. Correction of aberrant splicing of
the cystic fibrosis transmembrane conductance regulator (CFTR) gene
by antisense oligonucleotides. J Biol Chem 1999; 274: 36193–36199.
20 Gregorevic P, Allen JM, Minami E, et al. rAAV6-microdystrophin
preserves muscle function and extends lifespan in severely
dystrophic mice. Nat Med 2006; 12: 787–789.
PAEDIATRICS AND CHILD HEALTH 18:1
21 Wang Z, Kuhr CS, Allen JM, et al. Sustained AAV-mediated dystrophin
expression in a canine model of Duchenne muscular dystrophy with a
brief course of immunosuppression. Mol Ther 2007; 15: 1160–1166.
22 Tinsley JM, Davies KE. Utrophin: a potential replacement for
dystrophin? Neuromuscul Disord 1993; 3: 537–539.
23 Skuk D, Goulet M, Roy B, et al. First test of a “high-density injection”
protocol for myogenic cell transplantation throughout large volumes
of muscles in a Duchenne muscular dystrophy patient: eighteen
months follow-up. Neuromuscul Disord 2007; 17: 38–46.
24 Dellavalle A, Sampaolesi M, Tonlorenzi R, et al. Pericytes of human
skeletal muscle are myogenic precursors distinct from satellite cells.
Nat Cell Biol 2007; 9: 255–267.
25 Sampaolesi M, Blot S, D’Antona G, et al. Mesoangioblast stem cells
ameliorate muscle function in dystrophic dogs. Nature 2006; 444:
574–579.
26 Barton-Davis ER, Cordier L, Shoturma DI, et al. Aminoglycoside
antibiotics restore dystrophin function to skeletal muscles of mdx
mice. J Clin Invest 1999; 104: 375–381.
27 Hirawat S, Welch EM, Elfring GL, et al. Safety, tolerability, and
pharmacokinetics of PTC124, a nonaminoglycoside nonsense
mutation suppressor, following single- and multiple-dose
administration to healthy male and female adult volunteers.
J Clin Pharmacol 2007; 47: 430–444.
28 Wells DJ. Therapeutic restoration of dystrophin expression in Duchenne
muscular dystrophy. J Muscle Res Cell Motil 2006; 27: 387–398.
29 Alter J, Lou F, Rabinowitz A, et al. Systemic delivery of morpholino
oligonucleotide restores dystrophin expression bodywide and
improves dystrophic pathology. Nat Med 2006; 12: 175–177.
Practice points
• Multidisciplinary team input with follow-up in specialized
centres is the key to optimal management
• Regular physiotherapy by parents and use of night ankle
splints in the ambulant stages prevents/controls lower limbs
contractures
• Rehabilitation in knee–ankle–foot orthoses (KAFOs) is
effective in prolonging walking for an average of 2 years
• Glucocorticoid corticosteroids should be offered to all
ambulant patients from the age of 4 years onwards
• Close monitoring for corticosteroid dose adjustment, based
on response and adverse effects, is essential
• Regular echocardiographic monitoring (every 2 years
until age 10 and then annually) allows early treatment of
cardiomyopathy
• Scoliosis management is individualized, depending upon age,
rate of scoliosis progression and cardiorespiratory function
• Annual sleep studies, once FVC is ≤ 60%, and surveillance for
symptoms of sleep hypoventilation, allow timely introduction
of non-invasive ventilatory support
• Night-time non-invasive ventilation is well tolerated, relieves
symptoms and improves survival
• Every unit should develop robust protocols for transition of
care to adult medical teams
26 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Management of neonatal
hydrocephalus
Neil Buxton
Abstract
Neonatal hydrocephalus is a complex disorder due to many different
causes. This review seeks to encapsulate the management of neonatal
hydrocephalus in the term neonate. The current treatments are explored
and explained.
Keywords hydrocephalus; intraventricular haemorrhage; neonatal
­meningitis; neonates; shunts
Introduction
Neonatal hydrocephalus (NHC) is increasingly becoming the
most difficult management problem in paediatric neurosurgery
but survival from the hydrocephalus has improved.1 There are
many problems associated with aetiology, body weight and
immaturity, including unfused sutures, relating to risks of infec-­
tion and controversies with actual treatment protocols. This
review is intended to give an overview of current thoughts on the
management of hydrocephalus in the term neonate. The manage-­
ment of hydrocephalus (HC) in the premature child will not be
covered.
Neonatal hydrocephalus occurs in approximately 1 in 1000
live births. It is secondary to full term intraventricular haem-­
orrhage (IVH), infection or congenital causes such as tumours,
aqueduct stenosis, Dandy–Walker syndrome and its variants or,
of course, it can be truly idiopathic.
Where possible, treatment of the cause is the first priority but
in many cases the treatment of the hydrocephalus takes prece-­
dence. For example, in meningitis with HC, draining an enlarged
ventricular system may be necessary even before the infection
has been completely cleared.
Birth weight influences treatment choices as well. There is
a reluctance to introduce any permanent shunt systems into a
child less than 2 kg in weight because below this weight there is
a substantially increased risk of shunt failure due to infection.2
Fortunately, most term babies exceed this weight.
Treatment choices
All authorities in Western countries agree that, except in the most
devastating of circumstances, the HC should be treated. ­Difficulty
Neil Buxton MB ChB DMCC FRCS(Ed) FRCS(Neuro Surg) is a Consultant Paediatric
Neurosurgeon, Royal Liverpool Children’s Hospital, Liverpool, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
arises in choosing the best treatment option for a ­particular aeti-­
ology and is discussed below.
Intraventricular haemorrhage
Full term IVH probably occurs more often than we think but is
more common at lower birth weights.3–5 IVH does not invariably
cause HC. There is controversy as to how HC follows IVH. Some
have suggested that proteinaceous material blocks the arach-­
noid granulations, thus resulting in excess cerebrospinal fluid
(CSF) due to reduced reabsorption.6 In some cases there must
be some kind of block to CSF passage at the exit foraminae of
the fourth ventricle, as this would explain why endoscopic third
ventriculostomy (ETV) works in some patients with IVH (18%).7
However, the majority need to have shunts inserted and, despite
many recently advocating the use of ETV,8–11 this is still consid-­
ered by most to be the treatment of first option.
ETV has also been used to washout heavy protein loaded CSF
and to enable shunting to be implemented somewhat earlier or
even avoided.12 In heavy blood/protein loaded CSF, drainage
may be needed for some time before shunting to reduce the risk
of shunt blockage, although lumbar drainage has been successful
in avoiding shunting in IVH (see below).13
Neonatal meningitis
Meningitis can also lead to HC. This may be due to a heavy pro-­
tein load causing problems at the arachnoid granulations or dis-­
crete blockages of exit foraminae by debris or membranes. Hence,
again, ETV has been shown to be effective in some post-­meningitic
HC cases but the majority will require shunt insertion.
During the acute, infective phase, and whilst there is heavy
protein load in the CSF, it may be necessary to drain the HC with
an external ventricular drain (EVD) in order to reduce the intra-­
cranial hypertension. Obviously, such a device can be used to
drain excess CSF, but is also useful for obtaining CSF samples for
serial cultures and for the administration of intraventricular anti-­
biotics (a technique restricted to specialist neurosurgical units).
Draining the CSF in this way will allow it to return to its normal
constituent levels, so allowing shunting to be implemented. It is
generally believed that the higher the protein load of the CSF the
more likely it is that the shunt will fail due to blockage by debris;
this is not the case with HC secondary to tuberculous meningitis.
In these circumstances, the protein load is much less important
and shunting can take place earlier.14,15
Non-communicating hydrocephalus
The terminology communicating and non-communicating hydro-­
cephalus is becoming controversial, partly because of issues
with post-infectious and post-haemorrhagic HC, as briefly men-­
tioned above. It is clear that tumours, aqueduct stenosis and
Dandy–Walker syndrome and its variants can have physical
blocks to the passage of the CSF, and thus can lead to truly non-
­communicating HC. In these types of HC, seemingly in all ages,
ETV is the treatment of choice.16,17
Unfortunately, ETV in the truly non-communicating hydro-­
cephalus in some younger children will still fail. There are no
accepted theories for this but it may be that the pressure of CSF
required to initiate CSF reabsorption via the arachnoid granu-­
lations exceeds the pressure needed to expand the cranium in
those with unfused sutures; in such a situation, ETV is certain to
27 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
fail and a shunt is required. Whilst seemingly very simple, shunt-­
ing is controversial too and in order to address this small number
of children and their requirements an international randomized
trial is looking at the efficacy of shunting versus ETV.
Idiopathic
So-called idiopathic HC is best treated by treating the underlying
anatomical precedence. If there is evidence of flow obstruction,
then ETV may well work; otherwise it is likely that shunting will
be required.
Tumours
In the presence of third or fourth ventricular tumours or tectal plate
or brainstem tumours causing HC, the HC can easily be treated by
ETV and in some biopsies obtained.11,18 In successful resection of
a cerebellar tumour, for example, CSF flow may be restored and no
longer requires diversion. ETV or shunt is often needed, however,
although temporary EVD may ‘buy enough time’ for definitive
tumour treatment. Arachnoid cysts in or near the third ventricle
behave like tumours and can be successfully managed by ETV.19
Low birth weight
If the birth weight is less than 2 kg, then shunting tends not to be
­recommended.2 This is because of concerns about the anaesthetic,
neonatal care, risk of infection, operating on someone so small,
etc, with infection of the shunt being the most worrisome. The HC
can be ameliorated until the baby gains weight by serial lumbar
punctures, serial ventricular taps, an EVD (which can be used up
to 3 weeks without changing, with care) or a more permanent
Ommaya reservoir (an implanted ventricular tube with an injection
part).20 Intuitively there is concern about introducing a permanent
or semi-permanent foreign body into the child, just as there is con-­
cern definitive shunting; however, this seems to be safe in cases of
low birth weight and post-haemorrhagic hydrocephalus.21
Which shunt?
The type of shunt largely depends on the individual surgeon,
their experience with a particular model and their own biases.
Whilst this approach is not scientifically sound, a surgeon will
get ‘used’ to a particular model, understand its idiosyncrasies
and become confident with its use. This is perhaps more impor-­
tant than any other consideration such as cost, ‘newness’, etc, as
the surgeon uses their own experience to decide what is best for
a particular patient. This is where experience counts and, dare it
be said, some of the art in the science remains. Notwithstanding
all of the above, most would agree that the smaller the child the
less bulky the shunt and the quicker it should be to insert, with
fewer components to increase surgery time (hence infection risk)
and in the long run with fewer options for malfunction. Unfor-­
tunately, the answer to the problem of deranged physiology is
difficult to identify when we have mechanical devices with fixed
tolerances for, literally, a fluid system.
Where does the distal end go? Obviously in treating HC we
are inserting the upper end into the lateral ventricle. This tends
to be on the right (the non-dominant hemisphere). This is con-­
nected to a one-way valve device to the distal catheter. The dis-­
tal end is placed into the peritoneal cavity by choice, although
PAEDIATRICS AND CHILD HEALTH 18:1
c­ ommon alternatives include the right atrium via a neck vein
and the superior vena cava (the historical site of choice) and also
the pleural space. The pleura is used as a last resort in those in
whom there has been extensive abdominal surgery, peritonitis or
necrotizing enterocolitis, and whose neck veins have been dam-­
aged by central lines. Pleural shunts always cause effusions and,
if there is a significant CSF volume, then the effusion resulting
may embarrass lung function. A balance must be struck. Simi-­
larly, in the abdomen there can be a bulging tense abdominal
wall and the development of CSF hydrocoeles.
Shunt failure
This is almost inevitable in the lifetime of a patient with a shunt,
with the greatest number, approximately 20%, occurring in the
first year after insertion. Failure manifests itself in many ways, e.g.
increased head circumference, tense fontanel, drowsiness, vomit-­
ing, squint, CSF tracking alongside the shunt tubing, signs of infec-­
tion, and banging the head with the hands or against something,
which can indicate headache. In these circumstances shunt revi-­
sion is usually required. It is a neurosurgical rule that if the primary
carer says that the child is ‘not right’ and that they ‘think it’s the
shunt’, it is a brave and foolhardy person to ignore the warning.
In some in whom a shunt subsequently fails, an ETV may well
work and paediatric neurosurgeons will always assess a ‘new’
shunt failure for anatomical suitability for the procedure.9,22,23
Conclusion
The numbers of children surviving to term in the West with HC
are increasing as better obstetric care, earlier antenatal diagnosis
and better awareness lead to more informed decisions. In the
last 20 years, paediatric neurosurgery has evolved into a distinct
subspecialty on a par, for example, with spinal neurosurgery.
More aggressive, better targeted treatment for paediatric HC, no
longer in isolation from other children’s specialists, provided by
surgeons with expertise and training in the management of these
difficult clinical scenarios is improving the situation for these
patients. Treatment in the West has moved out of the hands of
paediatric general surgeons but their historical contribution can-­
not be underestimated as without their skills and expertise there
would be no paediatric neurosurgery at all. With the develop-­
ment of paediatric neurosurgical centres there is no longer an
excuse to dabble in the management of such complex problems
and such an approach is to be discouraged. ◆
References
1 Chi JH, Fullerton HJ, Gupta N. Time trends and demographics of
deaths from congenital hydrocephalus in children in the United
States: National Center for Health Statistics data, 1979 to 1998.
J Neurosurg 2005; 103(suppl 2): 113–118.
2 Bruinsma N, Stobberingh EE, Herpers MJ, et al. Subcutaneous
ventricular catheter reservoir and ventriculoperitoneal drain related
infections in preterm infants and young children. Clin Microbiol
Infect 2000; 6: 202–206.
3 Fanaroff AA, Stoll BJ, Wright LL, et al. Trends in neonatal morbidity
and mortality for very low birth weight infants. Am J Obstet Gynecol
2007; 196: 147 e1–8.
28 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
4 Fink S. Intraventricular haemorrhage in the term infant. Neonatal
Netw 2000; 19: 13–18.
5 Prat Puig M, Campistol Plana J, Muniz Llama F, et al. Intraventricular
haemorrhage in healthy newborn infants at term. An Esp Pediatr
1987; 27: 107–111.
6 Cherian S, Whitelaw A, Thoresen M, et al. The pathogenesis of
neonatal posthemorrhagic hydrocephalus. Brain Pathol 2004; 14:
305–311.
7 O’Brien DF, Seghedoni A, Collins DR, et al. Is there an indication for
ETV in young infants in aetiologies other than isolated aqueduct
stenosis? Childs Nerv Syst 2006; 22: 1565–1572.
8 Scavarda D, Bednarak N, Litre F, et al. Acquired aqueductal
stenosis in preterm infants: an indication for neuroendoscopic third
ventriculostomy. Childs Nerv Syst 2003; 19: 756–759.
9 Giomin V, Cinalli G, Grotenhuis A, et al. Endoscopic third
ventriculostomy in patients with CSF infection and/or hemorrhage.
J Neurosurg 2002; 97: 519–524.
10 Scarrow AM, Levy EI, Pascucci L, et al. Outcome analysis of endoscopic
third ventriculostomy. Childs Nerv Syst 2000; 16: 442–444.
11 Macarthur DC, Buxton N, Punt J, et al. The role of neuroendoscopy
in the management of brain tumours. Br J Neurosurg 2002; 16:
465–470.
12 Kamikawa S, Inui A, Kobayashi N, et al. Intraventricular hemorrhage
in neonates: endoscopic findings and treatment by the use of our
newly developed Yamadori type 8 ventriculoscope. Minim Invasive
Neurosurg 2001; 44: 74–78.
13 Huttner HB, Schwab B, Bardutzky J. Lumbar drainage for
communicating hydrocephalus after ICH with ventricular
haemorrhage. Neurocrit Care 2006; 5: 193–196.
14 Kemaloglu S, Ozkan U, Bukte M, et al. Timing of shunt surgery
in childhood tuberculous meningitis with hydrocephalus. Pediatr
Neurosurg 2002; 37: 194–198.
15 Palur R, Rajshekhar V, Chandy MJU, et al. Shunt surgery for
hydrocephalus in tuberculous meningitis: a long term follow-up
study. J Neurosurg 1991; 74: 64–69.
PAEDIATRICS AND CHILD HEALTH 18:1
16 Baldauf J, Oertal J, Gaab MR, et al. Endoscopic third ventriculostomy
in children younger than 2 years of age. Child Nerv Syst 2007; 23:
623–626.
17 Koch D, Wagner W. Endoscopic third ventriculostomy in infants less
than 1 year of age: which factors influence the outcome? Child Nerv
Syst 2004; 20: 405–411.
18 Javadpour M, Mallucci CL. The role of neuroendoscopy in the
management of tectal gliomas. Childs Nerv Syst 2004; 20: 852–857.
19 Kirollos RW, Javadpour M, May P, et al. Endoscopic treatment of
suprasellar and third ventricle related arachnoid cysts. Child Nerv
Syst 2001; 17: 713–718.
20 Khalil BA, Sarsam Z, Buxton N. External ventricular drains: Is there a
time limit in children? Childs Nerv Syst 2005; 21: 355–357.
21 Peretta P, Ragazzi P, Carlino CF, et al. The role of Ommaya reservoir
and endoscopic third ventriculostomy in the management of
posthaemorrhagic hydrocephalus of prematurity. Child Nerv Syst
2007; 23: 765–771.
22 O’Brien DF, Javadpour M, Collins DR, et al. Endoscopic third
ventriculostomy: An outcome analysis of primary cases and
procedures performed after ventriculoperitoneal shunt malfunction.
J Neurosurg 2005; 103(suppl 5): 393–400.
23 Buxton N, Macarthur D, Robertson I, et al. Neuroendoscopic third
ventriculostomy for failed shunts. Surg Neurol 2003; 60: 201–203.
Practice points
• Neonatal hydrocephalus has varied aetiology
• Choice of method of treatment for the hydrocephalus can
depend heavily on the aetiology
• Early consultation with a paediatric neurosurgeon is essential
29 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Tuberous sclerosis complex
Finbar O’Callaghan
Abstract
Tuberous sclerosis complex (TSC) is an autosomal dominant genetic dis-
order characterized by the formation of hamartomas in multiple ­organs.
It is caused by mutations in either the TSC1 gene on ­chromosome 9 or
the TSC2 gene on chromosome 16. Both genes are tumour ­suppressor
genes and the protein products of the two genes, hamartin and
tuberin, co-localize in the cell and help regulate cell growth by inhibit-
ing the mammalian target of rapamycin (mTOR) in the akt-mTOR-S6
kinase cell growth pathway. The discovery of the TSC proteins’ role in
this intracellular pathway has recently lead to investigation of chemo-
therapeutic agents, such as rapamycin (sirolimus), that also influence
this pathway and that may partly substitute for their role in regulating
cell growth. Clinically, the disease commonly causes epilepsy, learning
difficulties and behavioural problems (autism, hyperactivity and sleep
disturbance), although as many as half of affected individuals may
have a normal IQ. Life-threatening hamartomas may develop during
life in the kidneys (renal angiomyolipomas) and brain (subependymal
giant cell astrocytomas). Clinicians need to watch carefully for these
complications. Patients with TSC will require multidisciplinary clinical
involvement and, preferably, this should be coordinated through a
­specialist TSC clinic.
Keywords angiomyolipoma; epilepsy; giant cell astrocytoma; ­ mTOR;
tuberous sclerosis
Introduction
Clinicians have been aware of tuberous sclerosis complex (TSC)
as a distinct clinical entity for approximately 125 years since
Desiree Magloire Bourneville described the first case in 1880. For
most of that time they believed that the disease inevitably led to
severe learning difficulties and intractable epilepsy, but since the
clinical and epidemiological work in the late 20th century they
now realize that the disease manifests in virtually every organ
and that often, perhaps in the majority cases, it may be associated
with normal intelligence levels.1,2 The disease is characterized by
the formation of hamartomas in multiple organs. In the last 20
years major advances have been made in our ­understanding of
Finbar O’Callaghan MA MB ChB MSc PhD FRCPCH is a Consultant Senior
Lecturer in Paediatric Neurology, Institute of Child Life & Health,
Department of Clinical Sciences @ South Bristol, University of Bristol,
Bristol, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
the genetics and the mechanism of the disease, such that poten-
tial new therapeutic avenues are opening.
Epidemiology
By the standards of many genetic diseases, TSC is common. The
best epidemiological studies suggest it has a prevalence in the
region of 4–5 per 100 000 of the population. It affects all ethnic
groups and is found equally commonly in both sexes.3
Genetics
TSC is inherited in an autosomal dominant fashion, i.e. if an
individual has the disease, they have a 1 in 2 chance of pass-
ing the disease on to each of their offspring. However, approxi-
mately 70% of cases of TSC will be new mutations, such that
neither the child’s father nor mother is affected with the disease.
Occasionally, apparently unaffected parents may have more than
one affected child and this may be because of the phenomenon
of gonadal mosaicism, i.e. the situation where an apparently
unaffected parent has a population of affected cells confined to
their gonads and who is therefore liable to have more than one
affected child (Figure 1).4,5
Linkage studies in multigenerational families demonstrated
that two genes were responsible for TSC: one gene (TSC1) on
chromosome 9 and one on chromosome 16 (TSC2).6–9 Mutations
in either gene can cause all the clinical manifestations of TSC,
although TSC1 appears to be associated with a slightly milder
clinical phenotype.10 The frequency of mutations described is
considerably higher in TSC2 than in TSC1.11 This disparity may be
because there is an ascertainment bias in that TSC2 is ­associated
with more severe disease or it may reflect an increased level of
germline or somatic mutations in TSC2.
Figure 1 The concept of gonadal mosaicism i.e. where an apparently
unaffected parent can have more than one affected child because they
have a population of affected cells within their gonads.
30 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology

We are now able to look for mutations in TSC1 and TSC2 and proliferation (Figure 2). The hamartin–tuberin complex acts
in apparently affected individuals. However, even in the best through a molecule called Rheb (RAS-homologue expressed in
­laboratories, mutations are only being described in ­approximately brain) to inhibit the mammalian target of rapamycin (mTOR)
80% of cases. This may be because of somatic mosaicism which, without this inhibition, activates S6 kinase and therefore
whereby an affected individual has two populations of cells: one cell growth and proliferation.18,19
population has a mutation in one of the TSC genes and one does
not have a mutation in the TSC genes.12 The fact that we can now Rapamycin and inhibition of mTOR
screen for mutations in TSC1 and TSC2 raises the possibility of There has recently been much interest in the possibility of using
antenatal screening for TSC but it should be remembered that the a drug called rapamycin (sirolimus) in patients with TSC. Rapa-
phenomenon of somatic mosaicism means that a negative result mycin is a licensed drug that has been used for some time as an
on chorionic villous sampling does not completely exclude the immunosuppressant in transplant patients. It is also an inhibitor
possibility that a fetus may develop the disease. of mTOR and therefore would theoretically inhibit the activity of
It is thought that TSC1 and TSC2 are tumour suppressor S6 kinase in patients with TSC in whom the hamartin–tuberin
genes, i.e. that their function is to help regulate cell growth and complex is not working effectively. There is clear proof that rapa-
differentiation. When altered, by mutation, control of cell growth mycin has a biological effect in TSC patients. It has been shown,
is disturbed and tumours form throughout the body. Some of for instance, to cause regression in the size of subependymal
the strongest evidence supporting the idea that the TSC genes giant cell astrocytomas.20 However, there are still problems with
are tumour suppressor genes has been the demonstration that its use in TSC patients. Rapamycin is a relatively toxic drug and
some of the hamartomas in tuberous sclerosis patients show
loss of heterozygosity (LOH) either in the chromosomal region
9q34 or in 16p13.13,14 The loss of heterozygosity implies that an
individual with tuberous sclerosis inherits or acquires through
mutation a deletion in one copy of the gene but only develops
a lesion when there is a somatic mutation in the other copy.
These second mutations are usually large and involve the loss of
surrounding loci. This two-hit mechanism was first proposed by
Knudson in the early 1970s to explain the pathogenesis of retino-
blastoma. LOH has been demonstrated consistently in many TSC
lesions such as cardiac rhabdomyomas, renal angiomyolipomas
and subependymal giant cell tumours.
The two-hit mechanism does not, however, appear to operate
in all TSC lesions. LOH is rarely demonstrated in cerebral tubers.
It has been suggested that haploinsufficiency is sufficient to lead
to the development of cerebral tubers even in the presence of a
‘normal’ copy of the TSC gene on the second allele.

Disease mechanism at the molecular level

The protein products of the TSC1 and TSC2 genes are named
hamartin and tuberin, respectively. Hamartin is a 140 kDa, 1164
amino acid protein that is expressed in most adult tissues. It has
a coiled-coil domain. It probably has a role in regulating cell
adhesion. Hamartin interacts with the Ezrin–Radixin–Moiesin
family of actin-binding proteins. Inhibition of hamartin function
in cells results in loss of cell substrate adhesion and it has been
suggested that this may have a role in initiating the development
of TSC hamartomas. Tuberin is a 200 kDa, 1807 amino acid pro-
tein with a region near the carboxy terminal that is homologous
with GTPase activating protein (GAP). The GAP activity of TSC2
is thought to be essential for its physiological function.15–17
Substantial progress in our understanding of the molecular
function of tuberin and hamartin has come from experiments
in the fruit fly, drosophila. It was found that mutations in the
drosophila homologues of the TSC1 and TSC2 genes resulted
in increased cell and organ size.18 It has been known for some Figure 2 A schematic diagram of the P13 kinase – Akt – S6 kinase
time that hamartin and tuberin co-localize in the cell.15,16 It has pathway. TSC1-TSC2 inhibit the activity of mTOR via the molecule Rheb.
recently been discovered that they act together in the P13 kinase- Rapamycin, a commercially available drug, also inhibits mTOR. The net
akt-S6 kinase pathway. This pathway is stimulated by insulin and affect of both TSC1-TSC2 complex and rapamycin is to inhibit S6 kinase
insulin-like growth factors, and it helps to regulate cell growth and thereby help regulate cell growth.

PAEDIATRICS AND CHILD HEALTH 18:1 31 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology

is known to cause apthous ulceration, interstitial pneumonitis

and hyperlipidaemia. Its therapeutic effects in TSC patients may
only last for as long as a patient is undergoing treatment, i.e.
when treatment stops the hamartomas may re-grow. If life-long
treatment with a relatively toxic drug is going to be necessary
to maintain a therapeutic benefit, then many patients may opt
for other treatment options, e.g. surgery. Much work still needs
to be done to investigate the role of mTOR inhibitors in TSC
and, currently, they should not be used outside the context of a
­clinical trial.

Clinical manifestations
Brain
Three lesions can arise in the brain in individuals with TSC:
tubers, subependymal nodules and subependymal giant cell
astrocytomas (SEGA). Tubers are the lesions that give the dis-
ease its name (Figure 3). They are developmental abnormali-
ties of the cerebral cortex. The lesions have lost the normal
six-layered laminar architecture of the cerebral cortex and con-
tain ­ dysplastic neurones, astrocytes and characteristic giant
cells (Figure 4). Tubers can be identified in fetal life and per-
sist throughout life. It is thought that tubers do not increase in
number after birth, although they may become more visible on
magnetic resonance imaging (MRI) as the brain myelinates in
the first 2–3 years of life.
The tubers are thought to underlie the epilepsy that occurs
in approximately 75% of TSC individuals. Epilepsy may begin
in the first year of life with infantile spasms. These epileptic
spasms are probably most effectively treated with the GABA
inhibitor, vigabatrin.21 In later life, individuals with TSC may
develop multiple different types of seizures (e.g. focal seizures,
atonic seizures, atypical absences) that are often refractory to
anticonvulsant medication. Some individuals will develop the
Figure 4 Histological slide of a tuber showing characteristic giant cel
(arrowed).
epileptic encephalopathy, Lennox–Gastaut syndrome. Some-
times it is possible to identify the ‘epileptogenic tuber or tubers’
and in these patients epilepsy surgery should be considered
early in order to prevent the damage that can result from epilep-
tic ­encephalopathies such as West syndrome or Lennox–Gastaut
syndrome. A recent meta-analysis of epilepsy surgery in TSC has
shown that 57% of patients achieve seizure freedom following
surgery.22
Subependymal nodules are small hamartomas that are found
on the walls of the lateral ventricles. They are usually multiple
and are sometimes calcified. There is no evidence that, per se,
they cause any neurological symptoms. They are the second
most common cerebral feature of TSC. The nodules are easily
visualized by computerized tomography (CT) but they may also
be seen on MRI.
Subependymal giant cell astrocytomas (SEGAs) occur in
approximately 5% of patients with TSC (Figure 5). It is possi-
ble that they arise from pre-existing subependymal nodules but

Figure 3 Magnetic Resonance Image (axial cut, FLAIR sequence) showing Figure 5 CT scan of brain of an individual with TSC who has a
multiple cerebral tubers in Tuberous Sclerosis Complex patient. subependymal giant cell astrocytoma at the foramen of Monro (arrowed).

PAEDIATRICS AND CHILD HEALTH 18:1 32 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
this has not been conclusively proven. They are histologically
indistinguishable from subependymal nodules. They arise most
often in the region of the foramen of Monro and cause clini-
cal problems by blocking the flow of cerebrospinal fluid in the
ventricular system. Untreated they can cause raised intracranial
pressure leading to hydrocephalus, blindness and, even, death.
Clinicians should be vigilant for signs and symptoms of raised
intracranial pressure and symptomatic lesions should be surgi-
cally removed.
Cognition and psychopathology: there is a bi-modal distribu-
tion of intelligence in individuals with TSC. There is a population
of individuals with severe learning difficulties and a population
with a normal distribution of IQ around a mean of approximately
90. More than half of individuals with TSC will have an IQ that
will fall within the range of normal for the ­general population
(i.e. greater than 70). Both the infantile encephalopathy and the
number of cerebral tubers are thought to influence the level of
intelligence in the TSC population, although, in an individual
case, it should be remembered that normal intelligence can be
achieved even in the presence of many tubers and a history of
infantile spasms.23
Autism and hyperactivity are also common in TSC. Again, the
autism may be related to tuber number and location. Tubers in
the temporal lobe may be associated with an increased risk of
autism. Psychopathologies, such as depression and anxiety, have
also been found to be more common in TSC individuals than
unaffected controls.3,24,25
Cardiac lesions
Cardiac rhabdomyomas are the commonest cardiac tumours of
childhood and they are often associated with TSC. They can be
detected on antenatal ultrasound scanning. They occur within any
of the four heart chambers as intracavity or intramural tumours
but they are most commonly seen in the left ventricle. Clinically
they can cause problems in the postnatal period because they
obstruct ventricular function and outflow, and they may need to
be removed surgically. They can also cause rhythm disturbances
when the rhabdomyomatous tissue acts as an accessory conduct-
ing pathway predisposing the patient to pre-excitation syndromes
such as Wolff–Parkinson–White syndrome. The rhabdomyomas
regress in size after birth and although they may still be vis-
ible on echocardiograms, they rarely cause problems outside the
postnatal period.26–28
Renal lesions
Renal angiomyolipomas, hamartomas comprised of smooth
muscle, fat and abnormal blood vessels, occur in up to 75% of
patients with TSC. They do not usually appear before the age of
5. In most patients they will remain asymptomatic but in a small
proportion they will cause clinical problems secondary to haem-
orrhage. The lesions are easily visualized by ultrasound scanning
when they are visible as bright hyperechoic lesions. Patients with
large angiomyolipomas (i.e. greater than 3.5–4 cm in diameter)
are at the greatest risk of suffering from haemorrhage. Bleed-
ing angiomyolipomas can be treated successfully with selective
arterial embolization. This method of treatment can stop haem-
orrhaging whilst preserving the maximum amount of function-
ing renal tissue. Sometimes the process of ­embolization causes
PAEDIATRICS AND CHILD HEALTH 18:1
­transitory but significant pain and fever and this can be con-
trolled with analgesia and steroids.29
Cysts are the second most common renal manifestation of
TSC. They arise from anywhere within the nephron. Occasion-
ally, the mass effect from cysts can compromise renal function.
Cysts can be associated with either TSC1 or TSC2 disease. Rarely,
a particularly severe phenotype occurs when there is a contigu-
ous deletion of the TSC2 gene and the polycystic kidney disease
gene (PKD1) on chromosome 16. In these patients, polycystic
kidneys develop in early childhood and usually result in renal
failure in adolescence or early ­adulthood.30
Pulmonary lesions
An interstitial lung disease, known as lymphangioleiomyoma-
tosis (LAM), occurs in female TSC patients. It affects some-
where between 26% and 39% of women with TSC. The onset
is usually in adolescence or adulthood. The disease is charac-
terized by the proliferation of abnormal smooth muscle cells
and cystic changes within the parenchyma of the lung. Clinical
symptoms are dyspnoea and pneumothorax. It is best diag-
nosed using high-resolution CT of the chest. There is no recog-
nised effective treatment that reverses the progression of the
lung disease and it is one of the causes of early mortality in
TSC patients.31
Concept of ‘benign metastases’: it has been found that women
who do not have germline mutations in their TSC1 or TSC2
genes can develop both lymphangioleiomyomatosis and renal
angiomyolipomas. In these patients, identical TSC2 mutations
have been identified in cells from their abnormal renal and
­pulmonary lesions that are not identifiable anywhere else (i.e.
in DNA from blood). This finding suggests that these women
have developed lung and kidney lesions from a single progeni-
tor cell. Consequently, investigators have developed a ‘benign
metastasis’ model that proposes that histologically ‘benign
metastases’ (i.e. not cancerous) with mutations in TSC1 or
TSC2 travel from renal angiomyolipomas to the lung (Figure 6).
It is postulated that cells with TSC1 or TSC2 mutations have
decreased cell adhesion and therefore an increased motility. It is
not understood why this phenomenon occurs almost exclusively
in women, although it is suggested that oestrogen, in some way,
must alter TSC signalling.32
Skin lesions
There are multiple skin lesions associated with TSC. None of the
lesions causes a significant clinical problem although because
they can sometimes be disfiguring there may be considerable
psychological distress. The cutaneous lesions characteristically
appear at different ages. Hypomelanic macules or ‘ash-leaf
patches’ are areas of white skin that can most easily be visual-
ized by an ultraviolet (Wood’s) light. They are often present at
birth or in early infancy. Facial angiofibromatosis, a red mac-
ulopapular rash, develops later in childhood and continues to
progress through adolescence and adulthood. This lesion can be
significantly disfiguring and may require laser treatment for cos-
metic reasons. The ‘shagreen patch’ is an area of raised rough-
ened skin that characteristically appears in the small of the back
but can occur anywhere on the trunk. It tends to appear in child-
hood. Ungual fibromas are small lesions that appear around the
33 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology

Tuberous sclerosis complex related lesions

Facial angiofibromatosis or forehead fibrous plaque

Periungual fibromas
Shagreen patch
Retinal astrocytoma
Cortical or subcortical tuber
Subependymal nodule
Subependymal giant cell astrocytoma
Renal angiomyolipoma or pulmonary lymphangioleiomyomatosis
Cardiac rhabdomyoma

The demonstration of two of these lesions in one individual constitutes

clinical proof of a diagnosis of TSC.
Note: Renal angiomyolipoma and pulmonary lymphangioleiomyomatosis
count as one lesion for the purposes of diagnosis.

Table 1

the disease in an individual, a full clinical examination, including

inspection with Wood’s light and fundoscopy, and cranial imag-
ing should be undertaken. Renal imaging may be helpful although
sometimes this can be difficult to interpret and may confuse the
issue. Echocardiography is only useful in young children as
­cardiac rhabdomyomas regress after birth. White patches (ash-
leaf patches, hypomelanic macules) do not count as hamartomas
but are suggestive of the diagnosis. Greater than five hypomelanic
macules on an individual are very rarely seen outside of the con-
text of TSC.
Once a diagnosis has been made, baseline assessment with
cranial imaging and renal ultrasound scanning should take
Figure 6 Diagram illustrating the concept of “benign metastasis” place if not already been done as part of the diagnostic work-
in linking renal and lung pathology in TSC patients i.e. cells up. An ECG should also be done to look for any evidence of
migrate from angiomyolipomas in the kidney that then cause cardiac arrhythmias as pre-excitation syndromes, ­ including
lymphangioleiomyomatosis in the lung. Wolf–­Parkinson–White syndrome, may occur in TSC. A neuro­
psychological assessment at the time of diagnosis is also
­appropriate.
nail beds in late adolescence and adulthood. The fibromas often The frequency and timing of follow-up examinations in TSC
cause a longitudinal ridge to occur in the nail growing beneath is controversial. Some clinicians advocate regular screening with
them.33 cranial imaging and renal ultrasound scanning. Others argue that
regular screening is not warranted as the natural history of TSC
lesions is not understood and therefore there is no guide as to
Management
how often follow-up imaging should be undertaken nor when an
Despite the advances in genetics and genetic diagnosis of TSC, asymptomatic lesion should be treated. These clinicians advo-
it is still important to be able to make a secure clinical diagno- cate close supervision of TSC patients and a low threshold for
sis. Genetic testing, even in the best centres, is still only approxi- repeat scanning if there is any significant clinical change in the
mately 80% sensitive. There have been various published clinical patient or the appearance of any worrying clinical symptom or
criteria on how to make the diagnosis of TSC. Essentially it is nec- sign. Currently the medical advisors of the Tuberous Sclerosis
essary to demonstrate the presence of TSC-related hamartomas Association (www.tuberous-sclerosis.org.) take the latter view
in two separate organs. These hamartomas occur rarely in the in their guidelines.
general population and so the discovery of two or more indepen- The management of TSC patients is often complex, requiring
dent hamartomas in an individual strongly supports a diagnosis the input of many disciplines, e.g. neurology, urology, ­psychiatry,
of TSC (Table 1). Individuals with a single hamartoma may have nephrology, psychology, neurosurgery, genetics, dermatology,
developed this through the random occurrence of two somatic plastic surgery, ophthalmology, cardiology, radiology and respi-
mutations in a cell in that organ but this scenario is unlikely to ratory medicine. Ideally, management should be coordinated
occur twice in different organs in the same individual without through one of the specialist TSC clinics distributed throughout
an inherited or early embryonic mutation. In practice, it is often the UK in Bath, Leeds, London (St Georges and Great Ormond
easy to find two hamartomas in affected individuals. To exclude Street), Cambridge, Northern Ireland and Scotland.

PAEDIATRICS AND CHILD HEALTH 18:1 34 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Conclusion
TSC is a significant and disabling genetic condition and relatively
common in comparison to the vast majority of genetic diseases. It
causes severe and sometimes life-threatening pathology in ­multiple
organs but particularly within the central nervous system, renal
and respiratory systems. In recent years major advances have
been made in our understanding of the genetics and mechanism
of the disease such that we now understand more clearly how the
products of the two TSC genes help regulate normal cell growth
and how, when they are not functioning properly, as in TSC, there
is a tendency to form hamartomas throughout the body. There is
still much that we do not understand. We do not understand why
some lesions only occur in women and why lesions appear at
different ages. We do not yet know how to manipulate the S6
kinase pathway for the clinical benefit of our patients. The next
decade promises to be both exciting and productive as we search
for answers to these and other questions. ◆
References
1 Osborne JP, Fryer A, Webb D. Epidemiology of tuberous sclerosis.
Ann N Y Acad Sci 1991; 615: 125–127.
2 Webb DW, Fryer AE, Osborne JP. Morbidity associated with tuberous
sclerosis: a population study. Dev Med Child Neurol 1996; 38:
146–155.
3 Joinson C, O’Callaghan FJ, Osborne JP, et al. Learning disability and
epilepsy in an epidemiological sample of individuals with tuberous
sclerosis complex. Psychol Med 2003; 33: 335–344.
4 Yates JR, van Bakel I, Sepp T, et al. Female germline mosaicism
in tuberous sclerosis confirmed by molecular genetic analysis.
Hum Mol Genet 1997; 6: 2265–2269.
5 Rose VM, Au KS, Pollom G, et al. Germ-line mosaicism in tuberous
sclerosis: how common? Am J Hum Genet 1999; 64: 986–992.
6 Fryer AE, Chalmers A, Connor JM, et al. Evidence that the gene for
tuberous sclerosis is on chromosome 9. Lancet 1987; 1: 659–661.
7 Kandt RS, Haines JL, Smith M, et al. Linkage of an important
gene locus for tuberous sclerosis to a chromosome 16 marker for
polycystic kidney disease. Nat Genet 1992; 2: 37–41.
8 van Slegtenhorst M, de Hoogt R, Hermans C, et al. Identification of
the tuberous sclerosis gene TSC1 on chromosome 9q34. Science
1997; 277: 805–808.
9 Identification and characterization of the tuberous sclerosis gene on
chromosome 16. Cell 1993; 75: 1305–1315.
10 Dabora SL, Jozwiak S, Franz DN, et al. Mutational analysis in a
cohort of 224 tuberous sclerosis patients indicates increased
severity of TSC2, compared with TSC1, disease in multiple organs.
Am J Hum Genet 2001; 68: 64–80.
11 Jones AC, Daniells CE, Snell RG, et al. Molecular genetic and
phenotypic analysis reveals differences between TSC1 and TSC2
associated familial and sporadic tuberous sclerosis. Hum Mol Genet
1997; 6: 2155–2161.
12 Kwiatkowska J, Wigowska-Sowinska J, Napierala D, et al. Mosaicism
in tuberous sclerosis as a potential cause of the failure of molecular
diagnosis. N Engl J Med 1999; 340: 703–707.
13 Sepp T, Yates JR, Green AJ. Loss of heterozygosity in tuberous
sclerosis hamartomas. J Med Genet 1996; 33: 962–964.
PAEDIATRICS AND CHILD HEALTH 18:1
14 Henske EP, Scheithauer BW, Short MP, et al. Allelic loss is frequent
in tuberous sclerosis kidney lesions but rare in brain lesions.
Am J Hum Genet 1996; 59: 400–406.
15 van Slegtenhorst M, Nellist M, Nagelkerken B, et al. Interaction
between hamartin and tuberin, the TSC1 and TSC2 gene products.
Hum Mol Genet 1998; 7: 1053–1057.
16 Nellist M, van Slegtenhorst MA, Goedbloed M, et al. Characterization
of the cytosolic tuberin-hamartin complex. Tuberin is a cytosolic
chaperone for hamartin. J Biol Chem 1999; 274: 35647–35652.
17 Maheshwar MM, Cheadle JP, Jones AC, et al. The GAP-related domain
of tuberin, the product of the TSC2 gene, is a target for missense
mutations in tuberous sclerosis. Hum Mol Genet 1997; 6: 1991–1996.
18 Kwiatkowski DJ. Rhebbing up mTOR: new insights on TSC1 and
TSC2, and the pathogenesis of tuberous sclerosis. Cancer Biol Ther
2003; 2: 471–476.
19 Tee AR, Fingar DC, Manning BD, et al. Tuberous sclerosis complex-1
and -2 gene products function together to inhibit mammalian target
of rapamycin (mTOR)-mediated downstream signaling. Proc Natl
Acad Sci U S A 2002; 99: 13571–13576.
20 Franz DN, Leonard J, Tudor C, et al. Rapamycin causes regression of
astrocytomas in tuberous sclerosis complex. Ann Neurol 2006; 59:
490–498.
21 Hancock E, Osborne JP, Milner P. The treatment of West syndrome:
a Cochrane review of the literature to December 2000. Brain Dev
2001; 23: 624–634.
22 Madhavan D, Schaffer S, Yankovsky A, et al. Surgical outcome in
tuberous sclerosis complex: A multicenter survey. Epilepsia 2007;
48: 1625–1628.
23 O’Callaghan FJ, Harris T, Joinson C, et al. The relation of infantile
spasms, tubers, and intelligence in tuberous sclerosis complex.
Arch Dis Child 2004; 89: 530–533.
24 Harrison JE, O’Callaghan FJ, Hancock E, et al. Cognitive deficits in
normally intelligent patients with tuberous sclerosis. Am J Med
Genet 1999; 88: 642–646.
25 Raznahan A, Joinson C, O’Callaghan F, et al. Psychopathology in
tuberous sclerosis: an overview and findings in a population-based
sample of adults with tuberous sclerosis. J Intellect Disabil Res
2006; 50: 561–569.
26 Webb DW, Thomas RD, Osborne JP. Cardiac rhabdomyomas and their
association with tuberous sclerosis. Arch Dis Child 1993; 68: 367–370.
27 Jozwiak S, Kotulska K, Kasprzyk-Obara J, et al. Clinical and genotype
studies of cardiac tumors in 154 patients with tuberous sclerosis
complex. Pediatrics 2006; 118: e1146–1151.
28 O’Callaghan FJ, Clarke AC, Joffe H, et al. Tuberous sclerosis complex
and Wolff-Parkinson-White syndrome. Arch Dis Child 1998; 78:
159–162.
29 O’Callaghan FJ, Noakes MJ, Martyn CN, et al. An epidemiological
study of renal pathology in tuberous sclerosis complex. BJU Int
2004; 94: 853–857.
30 Brook-Carter PT, Peral B, Ward CJ, et al. Deletion of the TSC2 and
PKD1 genes associated with severe infantile polycystic kidney
disease - a contiguous gene syndrome. Nat Genet 1994; 8: 328–332.
31 Hancock E, Osborne J. Lymphangioleiomyomatosis: a review of the
literature. Respir Med 2002; 96: 1–6.
32 Juvet SC, McCormack FX, Kwiatkowski DJ, et al. Molecular
pathogenesis of lymphangioleiomyomatosis: lessons learned from
orphans. Am J Respir Cell Mol Biol 2007; 36: 398–408.
33 Webb DW, Clarke A, Fryer A, et al. The cutaneous features of
tuberous sclerosis: a population study. Br J Dermatol 1996; 135: 1–5.
35 © 2007 Elsevier Ltd. All rights reserved.
 Symposium: neurology
Practice points
• Tuberous sclerosis complex is an autosomal dominant genetic
disease caused by mutations in either the TSC1 gene on
chromosome 9 or the TSC2 gene on chromosome 16
• The protein products of the two genes, hamartin and tuberin,
act together within cells to regulate cell growth
• The disease is characterized clinically by a tendency to form
hamartomas in multiple organs
• The hamartomas appear in different organs at different
stages of life
• Life-threatening hamartomas can develop in the kidney and
brain during life and clinicians looking after TSC patients
need to be aware of their development and the complications
PAEDIATRICS AND CHILD HEALTH 18:1
associated with them. Clinicians should have a low threshold
for repeat neuro- or renal imaging if there is any significant
clinical change in their patients
• Epilepsy occurs in approximately 75% of patients and
learning difficulties in approximately 50%
• Epilepsy surgery may provide a possible cure for epilepsy in TSC
• Behavioural problems such as autistic spectrum disorders,
hyperactivity and sleep disturbance may be the most
problematic aspect of the condition for many parents
• Potential treatments that mimic the effects of TSC proteins
on intracellular growth pathways of TSC are now being
investigated in clinical trials
• Care for TSC patients should preferably be coordinated
through specialist TSC clinics
36 © 2007 Elsevier Ltd. All rights reserved.
 Occasional Review
Long-term effects of
cannabis
Anna Boyce
Paul McArdle
Abstract
Cannabis use is common in young people and the drug is widely per-
ceived as being the least harmful of the numerous illicit substances
available. There is reasonable evidence, however, that long-term use
of cannabis is associated with adverse psychosocial outcomes, and
­accumulating evidence that cannabis use may have a causal association
with onset of psychosis. The role of cannabis as a ‘gateway’ to use of
other drugs such as heroin is considered. Other adverse associations
with cannabis use include impairment in driving skills, of which a long-
term consequence may be serious injury, and a possible association with
subtle cognitive deficits in children exposed to cannabis prenatally.
Keywords cannabis; adolescence; outcome
Introduction
As is evident from the media and the regular clinical experience
of paediatricians and child psychiatrists, the use of illicit drugs
and alcohol is widespread amongst adolescents and young adults
in the UK and has lost much of the stigma with which it was pre-
viously associated. Cannabis is widely perceived as a safe drug in
comparison with heroin or cocaine and until relatively recently
was not believed to induce dependence or lead to a withdrawal
syndrome. Its classification was downgraded from a Class B con-
trolled drug to a Class C controlled drug in 2004.
Data from the National Office for Statistics in 2004/5 show
that in the UK 33% of males and 21% of females aged 16–24
years used illicit substances in the previous year, of which the
majority were cannabis users – 30% of males and 18% of females
in this age group. Whilst these figures might seem high, they
represent a small decrease in cannabis use amongst 16–24 year
olds since 1998.1 In addition, illicit drug use is not confined to
middle or late adolescence; figures from 2000 suggest that 15%
of boys and 13% of girls aged 11–15 years had used drugs in the
previous year. Of those aged 11–15 years in England, 12% were
using cannabis.1
A study published in 1996 of 7722 young people aged 15 and
16 years, at school in the UK, found that greater than 40% had
Anna Boyce MRPCH MRCPsych is a Specialist Registrar in Child and
Adolescent Psychiatry, Young People’s Unit, Newcastle General
Hospital, Westgate Road, Newcastle upon Tyne, UK.
Paul McArdle MRCPI MRCPsych DCH is a Consultant in Child and Adolescent
Psychiatry, Fleming Nuffield Unit, Newcastle upon Tyne, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
tried illicit drugs at some time. Cannabis was the most frequently
tried drug and its use was strongly associated with cigarette
smoking; 7% of non-smokers had tried cannabis compared to
89% of young people who had smoked greater than 10 cigarettes
in the previous week.2
These figures represent a significant number of young people
who may present to health professionals incidentally or as a con-
sequence of difficulties encountered through substance misuse.
Awareness of the current evidence regarding the possible long-
term adverse consequences of regular cannabis use, in terms of
physical, mental and social effects on the young person, is impor-
tant for the health professionals working with and attempting
to understand young people. Another significant but (probably)
small subgroup is the offspring of women who have used can-
nabis during pregnancy.
Short-term effects of cannabis
The main psychoactive component of the cannabis plant is a
cannabinoid compound – Δ9- tetrahydrocannabinol (Δ9 THC),
although there are over 60 cannabinoids present in the body of
the plant and the resin produced by the female plant.
Cannabis is bought as resin (‘hash’) or marijuana, and can be
smoked with tobacco in a joint, burned and inhaled in a bong
or ‘bucket’, or cooked in food and eaten. The concentration and
rate of absorption varies with method of use. Smoking and inhal-
ing cannabis leads to a rapid rise in blood concentration of Δ9
THC. Ingesting cannabis in food results in a much more gradual
absorption with effects being noticeable over several hours. Vari-
eties of cannabis plants such as skunkweed are said to have a
higher Δ9 THC concentration than traditional varieties. Cannabis
is fat soluble and accumulates in adipose tissue with an elimina-
tion half-life of 7 days.3 Regular use of cannabis therefore leads
to persistently high blood concentrations of Δ9 THC.
Cannabis is thought to exert its central effects via the CNR1
cannabinoid receptor in the brain. These effects include euphoric
and relaxed mood, a sense of detached observation and increased
appetite, although anxiety symptoms may also be experienced.
Cannabis intoxication causes impairments in concentration, fine
motor skills and reaction times.4 Peripheral effects include con-
junctival hyperaemia (‘red eye’), raised heart rate and alterations
in blood pressure.
Cannabis use or intoxication may have significant long-term
consequences in terms of increased likelihood of injury follow-
ing road traffic accidents. For instance, Mura et al5 compared the
cannabis intake of 900 injured drivers with that of 900 matched
controls presenting to the same hospitals and found that 10% of
injured drivers had THC present in blood samples, with an odds
ratio of 2.0, drivers to controls, increasing to an odds ratio of 2.5
in those under 27 years old. By comparison 26% of the injured
drivers were over the legal limit for blood alcohol levels and the
odds ratio was 3.8. A detailed review of the effects of cannabis
use on driving ability can be found in Kalant.6
Tolerance to the effects of cannabis develops with use and
the existence of a dependence syndrome is recognized. A recent
study reported that cannabis users meeting criteria for depen-
dence compared with those users not deemed to be cannabis
dependent were more likely to have started using cannabis ear-
lier, to have had fewer years of schooling and to consume more
37 © 2007 Published by Elsevier Ltd.
 Occasional Review
alcohol. Scores of happiness and life satisfaction were reduced in
those who were cannabis dependent.7 Clear evidence of a can-
nabis withdrawal syndrome has been described with features of
irritability, aggression, restlessness and strange dreams.8
Issues complicating research into the long-term
effects of cannabis
Many of the studies from which the evidence relating to the long-
term consequences of cannabis use has emerged are longitudinal
studies recruited from the adult general population.9 The advan-
tages of these cohort studies are the large number of participants
and the capacity to detect adverse effects. Birth cohort studies, on
the other hand, have the advantages of being able to investigate
factors relating to early psychosocial- and neuro-­development
and early onset of cannabis use, but have the disadvantage of
being expensive, with lengthy intervals between commencement
and eventual outcome.
Ascertaining the causal role of cannabis use in adverse psy-
chosocial outcomes is complicated by the potential confound-
ing effects of multiple substance misuse, alcohol use, cigarette
smoking, psychiatric disorders such as depression and anxiety
and association with antisocial peer groupings. Reverse causa-
tion should also be considered, e.g. the use of cannabis as self-
medication for prodromal psychotic symptoms.10
Macleod et al9 set out systematically to appraise the evidence
from longitudinal studies for a causal relation between cannabis
use and psychosocial harm. They found 48 longitudinal studies,
of which only 16 met their criteria for methodology and lack of
significant bias. Three birth cohorts were included; the remaining
studies were of school age cohorts. Outcome measures included
antisocial behaviour, psychological functioning, subsequent illicit
drug use and educational attainment. The authors commented
that the variation in recruitment methods, methods of assessment
of exposure to illicit substances and how confounding factors were
taken into account precluded the use of meta-analysis. They con-
cluded that there was reasonable evidence for associations between
cannabis use and lower educational attainment and later use of
illicit drugs, but argued that the evidence for associations between
cannabis use and psychological health problems and behavioural
problems was less robust. For instance, although ­ cannabis use
may be ­associated with depression, third factors such as disturbed
family life may explain both the use and the distress.
Arseneault et al,10 in their review of the evidence for a specific
link between cannabis use and risk of later psychotic symptoms,
discuss the criteria by which causality can be inferred, such as
association, temporal priority and direction (see below).
Cannabis use and risk of psychosis
The findings of four major longitudinal studies, yielding much of
the evidence for a causal association between cannabis use and
psychotic illness, were reviewed in 2004 by Arseneault et al.10
One of these studies, the Swedish study of 50 000 army recruits,11
found a dose–response relationship between heavy cannabis use
at 18 years of age and a subsequent diagnosis of schizophrenia
within the 15 year follow-up period. This effect persisted after
the confounding effect of pre-existing psychiatric diagnosis was
controlled for, with a relative risk of 2.3.
PAEDIATRICS AND CHILD HEALTH 18:1
The Dunedin Multidisciplinary Health and Development Study,
a birth cohort study of 1037 individuals, collected data includ-
ing self-reported psychiatric symptoms at age 11 years, cannabis
use at 15 and 18 years and interview data leading to Diagnostic
Statistical Manual (DSM) IV diagnoses at 26 years.12,13 Amongst
those whose use of cannabis preceded the age of 15 years, there
was an odds ratio of developing adult schizophreniform disorder
of 3.1, after controlling for the presence of psychiatric symptoms
at the age of 11 years. The review authors concluded that at an
individual level cannabis usage doubled the relative risk for later
schizophrenia. In addition, they concluded that at a population
level, avoidance of cannabis use would lead to an 8% reduc-
tion in incidence of schizophrenia, and that whilst cannabis is
‘neither a sufficient nor a necessary cause for psychosis, it is a
component cause, part of a complex constellation of factors lead-
ing to psychosis’.10
Participants in the Dunedin study were further studied to
investigate the possibility that a gene–environment interac-
tion might provide an explanation for the observation that only
a minority of regular heavy cannabis users develop persistent
psychotic symptoms.14 The investigators found that the effect of
adolescent cannabis use was moderated by a functional polymor-
phism of the catechol-O-methyltransferase (COMT) gene, a gene
on chromosome 22q11 in a region implicated by genome scans
as having a possible aetiological role in schizophrenia. Using
hierarchical regression analysis, the authors reported that the
effect of genotype on psychotic symptoms was not significant,
but the effect of cannabis use and the interaction between can-
nabis use and genotype were significant. In line with the hypoth-
esis that cannabis use may have an effect on refinement of the
mechanisms of dopamine transmission occurring during adoles-
cence, the authors found that onset of cannabis use in adulthood
was not associated with subsequent development of psychotic
­symptoms.
Other psychiatric and cognitive sequelae
Cannabis use has been linked to symptoms of low mood and anx-
iety. Data from the Christchurch Health and Development Study
were examined to ascertain associations between frequency of
cannabis use and psychosocial outcomes, including depression,
assessed using DSM criteria. Across the four age groups, 14–15
years, 15–16 years, 17–18 years and 20–21 years, the odds ratio
for depression for weekly cannabis users relative to non-users
was 1.7. With the exception of the age range 20–21 years, can-
nabis use was also significantly associated with rate of suicide
attempts.15
A review of cohort studies,16 whose participants included
adolescents and adults, concluded that there was a modest
association between current heavy cannabis use and depres-
sive symptomatology, and a modest association between early
onset regular cannabis use and later depressive illness. No asso-
ciation was found between onset of depression and subsequent
­cannabis use.
The same review found that early cannabis use may signifi-
cantly increase the risk of leaving school early.16 Possible explana-
tions are discussed, with reference to reverse causality, a specific
‘problem behaviour’ syndrome (likely to include the DSM IV con-
cept of conduct disorder) of which substance misuse is one facet,
38 © 2007 Published by Elsevier Ltd.
 Occasional Review
and the role of confounding factors. The authors make the point
that it is unlikely that any single factor operates in isolation; rather
the true picture may be a combination of many such processes.
The relationship between educational attainment and canna-
bis use was also examined in the Christchurch Health and Devel-
opment study;17 35.6% of young people who used cannabis on
10 or more occasions in the year between ages 15 and 16 years
left school with no formal qualifications compared to 17.1% who
had not used cannabis.
The findings of this review are elaborated by those of a Cana-
dian longitudinal birth cohort, the Ottawa Prenatal Prospective
Study (OPPS).18 In this cohort 113 young people were tested at
9–12 years of age on measures of general intelligence, vocabu-
lary, memory and sustained attention. Of this sample, 10 children
had already tried cannabis, although none was using cannabis
regularly at the time of testing. At the time period 17–20 years the
cognitive tests were repeated. After exclusion of confounding fac-
tors, such as presence of psychiatric disorder and socioeconomic
status, the current heavy use group had lower scores in measures
of memory and IQ in comparison to the light use, former use and
no use groups. No significant differences were found between
the group of former cannabis users and the comparison group
who had not used cannabis, suggesting that the poor educational
attainment of young people using cannabis on a regular basis
may be a direct consequence of recent cannabis use.
Cannabis as a ‘gateway’ drug
An influential theory regarding the role of cannabis in the pro-
gression to further illicit substance misuse is the ‘stage theory’
of substance misuse, which supposes that young people proceed
from using legal drugs such as alcohol to use of illicit drugs with
cannabis use at an intermediate stage. Users of illicit substances
are predicted to have previously used legal substances but not all
of those who use legal substances will progress to use of illicit
substances. Arguments against this theory include the possibil-
ity of confounding factors such as a predisposition to substance
misuse or risk taking behaviour in general.
By the age of 21 years, 70% of the Christchurch Health
and Development Study cohort had used cannabis and 26% had
used other illicit drugs.19 The vast majority of those who had
used other illicit drugs had previously used cannabis. Using a
form of multivariate analysis the investigators found that those
who used cannabis on more than 50 occasions in a year had haz-
ards of other drug use that were 140 times higher than those who
did not use cannabis, after adjustment for confounding factors
such as frequency of alcohol consumption and childhood sexual
abuse. Inclusion of adverse life events and adolescent risk taking
behaviour as co-factors in the model led to a reduction in hazards
of other illicit drug use to 59 times higher than for non-users.
A study of twins discordant for cannabis use, derived from the
Netherlands Twin Register,20 examined subsequent progression
to use of ‘party’ drugs and ‘hard’ drug use. In a sample of 219
twin pairs, average age 27 years, among whom only one twin
had used cannabis before the age of 18 years and one subse-
quently, the odds ratio for ‘party’ drug use was 7.4 (confidence
interval 2.3–23.4) and for ‘hard’ drug use was 16.5 (2.4–111.3),
leading the authors to speculate that a factor directly related to
cannabis itself might be at work.
PAEDIATRICS AND CHILD HEALTH 18:1
Effects of prenatal cannabis exposure
Studies have failed to show an association between maternal can-
nabis use and low birth weight and low weight at 6 years.21,22
There is some evidence that prenatal exposure to cannabis
may be associated with impaired attention in children. A study
comparing the performance of 330 4-year olds with prenatal
exposure to cocaine, marijuana, alcohol and tobacco found an
association between heavy maternal marijuana use and omission
rates on the Continuous Performance Test, suggesting a possible
effect on sustained attention, although this finding did not reach
statistical significance.23 Another study found an effect of prena-
tal marijuana exposure on attention in adolescents in one of five
domains of attention.24
Exposure to prenatal illicit substances is a common compo-
nent of background history in children assessed for adoption.
It appears that, on the basis of available evidence, the effect of
maternal cannabis use on a child’s later presentation is statisti-
cally detectable but modest.
Prenatal cannabis exposure has also been implicated in the
aetiology of childhood acute myeloid leukaemia (AML). A case–
control study compared parental interview data of 517 children
(maternal data available) and 450 children (paternal data avail-
able) with AML, with similar data from 610 and 523 controls,
respectively. No positive association was found between parental
marijuana use and childhood AML.25
Long-term effects on physical health
The effects of long-term cannabis exposure on the human respira-
tory system has been the focus of considerable research. A recent
systematic review of clinical studies published between 1966 and
2005 found 34 studies meeting selection criteria.26 These could
be subdivided into studies examining the short-term conse-
quences of cannabis exposure on bronchodilation and studies
of long-term smoking of marijuana and pulmonary function or
respiratory symptoms. All 14 of the latter demonstrated associa-
tion between cannabis use and respiratory symptoms, including
cough, phlegm and wheeze.
A cohort study of 53 controls and 188 subjects, of whom
40 were regular smokers of marijuana alone, 44 smokers of
marijuana and tobacco and 31 smokers of tobacco only, who
underwent bronchoscopy and endobronchial biopsy, found
that smoking tobacco and marijuana were each associated
with significant bronchial mucosal histopathology. The authors
found that the effects of marijuana and tobacco on the bron-
chial mucosa were additive, including possible pre-malignant
changes.27
Adverse effects on animal cellular immunity have raised ques-
tion about the effects of cannabis use on the immune system.
However, a review of animal and human studies concluded that
there was no evidence of a significant effect in humans, with the
exception of the effect of marijuana smoking on broncho-alveolar
immunity.28
Substance misuse services for adolescents
The UK government has recognized the need for the provision of
substance misuse services for adolescents, and set out its agenda
39 © 2007 Published by Elsevier Ltd.
 Occasional Review
Figure 1 Cycle of change model.30
in the report Every Child Matters: Change for Children, Young
People and Drugs.29
A useful framework for identifying appropriate timing of
referral to specialist services, and as a method of engaging
­adolescents known to be involved in harmful use of illicit sub-
stances in the referral process, is provided by the cycle of change
model described by Prochaska and Diclemente (­Figure 1).30 In
this model an individual may be at any stage when they pres-
ent. It should become apparent which stage they are at after
an initial conversation about their drug use and their attitudes
towards it. Clearly, referral of an individual contemplating
change or already preparing to make changes in their drug use
to the local substance misuse service is appropriate and should
be done promptly. Individuals in the pre-contemplation stage
may also benefit from referral, if only for discussion of the
advantages and disadvantages of their drug use and education
about harm minimization. Adolescents can be reassured about
the confidentiality of such services (with the usual exceptions,
e.g. child protection), although involvement and support of
­parents is encouraged.
Conclusion
There is reasonable evidence for a link between cannabis use
and risk of overall adverse psychosocial functioning, lower edu-
cational attainment, development of psychosis, particularly in
individuals vulnerable by virtue of family history, and possibly
cognitive performance. Young people using cannabis on a regu-
lar basis should be warned of the possible consequences to their
mental health in addition to the physical health risks of smoking
cannabis. Health professionals working with adolescents should
be prepared to ask young people about their drug use and be
aware of local resources to support young people who wish to
address their drug use. ◆
PAEDIATRICS AND CHILD HEALTH 18:1
References
1 National Statistics website: www.statistics.gov.uk
2 Miller PMcC, Plant M. Drinking, smoking and illicit drug use among
15 and 16 year olds in the United Kingdom. BMJ 1996; 313: 394–397.
3 Ashton CH. Pharmacology and effects of cannabis: a brief review.
Br J Psychiatry 2001; 178: 101–106.
4 Hall W, Solowij N. Adverse effects of cannabis. Lancet 1998; 352:
1611–1616.
5 Mura P, Kintz P, Ludes B, et al. Comparison of the prevalence of
alcohol, cannabis and other drugs between 900 injured drivers
and 900 control subjects: results of a French collaborative study.
Forensic Sci Int 2003; 133: 79–85.
6 Kalant H. Adverse effects of cannabis on health: an update of the
literature since 1996. Prog Neuro-Psychopharmacol Biol Psychiatry
2004; 28: 849–863.
7 Looby A, Earleywine M. Negative consequences associated with
dependence in daily cannabis users. Substance Abuse Treat Prev
Policy 2007; 2: 3.
8 Wiesbeck GA, Schukit MA, Kalmijn JA, et al. An evaluation of the
history of a marijuana withdrawal syndrome in a large population.
Addiction 1996; 1: 1469–1478.
9 Macleod J, Oakes R, Copello A, et al. Psychological and social
sequelae of cannabis and other illicit drug use by young people:
a systematic review of longitudinal, general population studies.
Lancet 2004; 363: 1579–1588.
10 Arseneault L, Cannon M, Witton J, et al. Causal association
between cannabis and psychosis: examination of the evidence.
Br J Psychiatry 2004; 184: 110–117.
11 Zammit S, Allebeck P, Andreasson S, et al. Self reported cannabis
use as a risk factor for schizophrenia in Swedish conscripts of 1969:
a historical cohort study. BMJ 2002; 325: 1199–1201.
12 Arseneault L, Cannon M, Poulton R. Cannabis use in adolescence
and risk for adult psychosis: longitudinal prospective study. BMJ
2002; 325: 1212–1213.
13 The American Psychiatric Association. DSM-IV-TR: Diagnostic
Statistical Manual of Mental Disorders. American Psychiatric Press
Inc, 1994.
14 Caspi A, Moffitt TE, Cannon M, et al. Moderation of the effect of
adolescent-onset cannabis use on adult psychosis by a functional
polymorphism in the catechol-o-methyl transferase gene:
longitudinal evidence of a gene x environment interaction. Biol
Psychiatry 2005; 57: 1117–1127.
15 Fergusson DM, Horwood LJ, Swain–Campbell N. Cannabis use and
psychosocial adjustment in adolescence and young adulthood.
Addiction 2002; 97: 1123–1135.
16 Degenhardt L, Hall W, Lynskey M. Exploring the association between
cannabis use and depression. Addiction 2003; 98: 1493–1504.
17 Lynskey M, Hall W. The effects of adolescent cannabis use on
educational attainment: a review. Addiction 2000; 95: 1621–1630.
18 Fried PA, Watkinson B, Gray R. Neurocognitive consequences of
marihuana - a comparison with pre-drug performance. Neurotoxicol
Teratol 2005; 27: 231–239.
19 Fergusson DM, Horwood LJ. Does cannabis use encourage other
forms of illicit drug use? Addiction 2000; 95: 505–520.
20 Lynskey MT, Vink JM, Boomsma DI. Early onset cannabis use and
progression to other drug use in a sample of Dutch twins. Behav
Genet 2006; 36: 195–200.
21 English DR, Hulse GK, Milne E, et al. Maternal cannabis use and
birth weight: a meta-analysis. Addiction 1997; 92: 1553–1560.
40 © 2007 Published by Elsevier Ltd.
 Occasional Review
22 Day NL, Richardson GA, Geva D, et al. Alcohol, marijuana, and
tobacco: effects of prenatal exposure on offspring growth and
morphology at age six. Alcohol Clin Exp Res 1994; 18: 786–794.
23 Noland JS, Singer LT, Short EJ, et al. Prenatal drug exposure and
selective attention in preschoolers. Neurotoxicol Teratol 2005; 27:
429–438.
24 Fried PA, Watkinson B. Differential effects of facets of attention
in adolescents prenatally exposed to cigarettes and marihuana.
Neurotoxicol Teratol 2001; 23: 421–430.
25 Trivers KF, Mertens AC, Ross JA, et al. Parental marijuana use
and risk of childhood acute myeloid leukaemia: a report from
the Children’s Cancer Group (United States and Canada). Paediatr
Perinat Epidemiol 2006; 20: 110–118.
26 Tetrault JM, Crothers K, Moore BA, et al. Effects of marijuana
smoking on pulmonary function and respiratory complications.
Arch Intern Med 2007; 167: 221–228.
27 Fliegel SE, Roth MD, Kleerup EC, et al. Tracheobronchial
hisopathology in habitual smokers of cocaine, marijuana, and/or
tobacco. Chest 1997; 112: 319–326.
28 Kraft B, Kress HG. Cannabioids and the immune system. Schmerz
2004; 18: 203–210.
PAEDIATRICS AND CHILD HEALTH 18:1
29 UK Government Department. Every Child Matters: Change for
Children. Young People and Drugs. London: Department for
Education and Skills, 2005.
30 Prochaska JO, Diclemente CC. Towards a comprehensive model of
change. In: Miller WR, Heather N, eds. Treating addictive behaviours:
processes of change. New York: Plenum Press; 1986, p. 3–27.
Practice points
• Misuse of cannabis in adolescence may have long-term
consequences on psychosocial functioning and mental health
• Cannabis use is associated with later use of other illicit
substances, although the mechanism for this is unclear
• There is not enough evidence to enable prediction of
outcome for individuals exposed to cannabis prenatally
• Questions about substance misuse should form part of
routine assessment by health professionals working with
young people and referral to substance misuse services made
as appropriate
41 © 2007 Published by Elsevier Ltd.
 Personal practice
Headache in children
and adolescents
Ishaq Abu-Arafeh
Abstract
Headache is a common problem in children and adolescents. Migraine
and tension type headache are most common causes of headache and
only occasionally chronic headache is the main presenting feature of a
serious brain disease or a brain tumour. Full clinical history, physical
examination and headache diary recordings are essential requirement
in the assessment of children with headache and the assessment of the
impact of headache on child’s education, social and family life. Manage-
ment of headache includes the promotion of healthy life style with regu-
lar meals, sleep and exercise. Painkillers are used in appropriate dose,
as early as possible after the onset of headache and the use of most
appropriate route of administration.
Keywords adolescents; children; migraine; tension-type headache
What do you do?
Laura, a 12-year-old girl, attends the clinic with her mother com-
plaining of headache for the past year. Laura and her mother
describe two types of headache. The first type is a migraine
­occurring once a month. The attacks last for 24 h and the pain
is severe enough to stop all activities. Laura describes the head-
ache as throbbing with maximum intensity in her forehead. She
can identify no trigger factors and has no warning signs before
the onset of pain. The headache builds up in intensity over a
period of 30–60 min and is almost always associated with loss of
appetite, feeling sick, light intolerance, noise intolerance and pal-
lor. In most, but not all, attacks she also vomits. She feels better
after rest and sleep. Paracetamol helps a little, but she finds
codeine most helpful in relieving symptoms.
The second type is episodes of headache that occur at least
three times per week and last between 2 and 3 h each. This
headache does not stop normal activities and is described as ‘just
sore’ around the head. There are no other associated symptoms
and, in particular, she is able to have normal meals, has no feel-
ing of sickness and does not vomit. Neither light nor noise both-
ers her during attacks. After an episode of throat infection, the
headaches became daily for about 1 month and then reverted to
the normal rate of occurrence of three per week. She finds relief
from rest and she only occasionally treats these headaches with
paracetamol.
Ishaq Abu-Arafeh MBBS MD MRCP FRCPCH DCH is a Consultant Paediatrician,
Stirling Royal Infirmary, Stirling, Scotland, UK.
PAEDIATRICS AND CHILD HEALTH 18:1
Between headaches, she is back to her normal self and has
no other illnesses. Laura and her mother are concerned about the
headaches, their frequency and the number of school days lost
due to headache. Physical and neurological examinations were
normal.
Key points in the history
Children and their parents can often identify different types of
headache experienced by the child. Laura was able to make a dis-
tinction between the severe headache episodes that she rightly
described as migraine and other milder headache episodes that
occurred more frequently. Unfortunately, other children may not
be able to make this distinction. These children may describe the
most severe attack, as it has made most impression on them and
their family, or the most recent attack as it is still fresh in their
minds. A prospective diary recording symptoms is a powerful
tool in identifying the different types of headache, and may also
help to identify triggers.
Interpretation of clinical history
By making the distinction between the two types of headache, it
was possible, on applying a recognized definition and dia­gnostic
criteria, to establish the diagnosis of migraine without aura for
the first type of headache, and frequent tension-type headache
for the second. It should be remembered that migraine headache
is commonly not lateralized in children, as in Laura’s case. The
second edition of the International Classification of ­ Headache
Disorders (2004) is an excellent resource for the definition of
headaches in children and adults. Criteria for the diagnosis of
migraine headache without aura and tension type headache
are described in Tables 7 and 8, respectively, in the review by
­Mukhopadhyay and White elsewhere in this issue.
Physical examination
The negative findings in the physical examination and the
­neurological assessment are very important. They not only
exclude the possibility of serious underlying disorder, such as
a brain tumour, but also give the child and the parent the con-
fidence that the doctor has taken their complaint seriously and
looked for abnormal signs that may indicate a serious illness.
Physical examination should include measuring the blood pres-
sure and neurological assessment should include measuring head
circumference, ophthalmoscopy and cerebellar function (looking
for ataxia, nystagmus and intention tremor or point passing).
Investigations
In the absence of symptoms suggesting increased intracranial
pressure or signs suggesting cerebellar dysfunction, no further
investigations are needed. A lower threshold for neuroimaging
may be considered in:
• children under the age of 5 years;
• if in doubt about the physical findings;
• cases of inconsistent features.
42 © 2007 Elsevier Ltd. All rights reserved.
 Personal practice
Impact of headache on the child and family
Although it is true that anxiety about possible intracranial pathology
is an important reason for seeking medical advice, concerns about
missing school and worries about the impact on the child’s edu-
cation and enjoyment of family and social life are also important.
The whole family becomes embroiled with the child’s symptoms,
causing stress and anxiety to the rest of the family. Doctors need
to enquire from the child and family what worries them most and
to address these concerns in a direct and relevant manner, rather
than confining the discussion to general reassurances on the benign
nature of the disorder. In children with frequent headache, the help
of an experienced clinical psychologist may prove most useful.
Advice and education
The natural course of migraine is one of relapses and remissions.
It is common for children to describe well-defined periods of high
attack frequency separated by periods of relative remission. Good
and bad spells of headache are often well recognized by patients
and their parents. In many children it is not possible to identify
a clear trigger for the bad spells, but in some anxiety and stress
related to school exams and similar events are evident. It should
be remembered that ‘exciting’ events, such as birthday parties,
that may be seen in a positive light by the parents may be very
stressful to the child.
As they grow older children tend to have longer remissions
(good spells), but unfortunately they continue to be prone to
headaches and there is as yet no treatment that can be described
as a cure for migraine. Similarly tension headache can recur after
a period of remission.
Children can reduce the frequency of headaches (both
migraine and tension-type headache) and prolong the ‘good
spell’ by adopting a healthy lifestyle. There is no single measure
that is universally effective in migraine, but attention should be
paid to the following:
• regular meals;
• regular sleep;
• regular exercise and rest;
• avoidance or at least reduction of exposure to caffeine-
­containing drinks;
• avoidance of other xanthines such as chocolate;
• avoidance of taking analgesics more than twice a week to
­prevent the emergence of overmedication headache.
A carefully taken history may uncover triggers for the attacks
and, of course, these should be avoided. However, these triggers
are often rather vague and not easily avoided. Where stress is cited
as a trigger, the help of a clinical psychologist may be enlisted.
Managing migraine headache
Children should be given the opportunity to lie down for rest and
sleep if possible. Successful acute treatment of migraine attacks
depends on the recognition of early symptoms and the administra-
tion of the most appropriate painkiller, in an appropriate dose and
using the most appropriate route. There is a tendency for parents
and primary care medical practitioners to administer small doses of
analgesia, and also to delay administration of treatment until the
headache is established and shown to be severe enough to warrant
PAEDIATRICS AND CHILD HEALTH 18:1
treatment. For effective pain relief, analgesics should be given in
their optimum doses – 10–20 mg/kg for paracetamol and 10–15
mg/kg for ibuprofen. If simple analgesics are given in adequate dos-
age, there is seldom any additional benefit from using opiates such
as codeine. In Laura’s case, by ensuring that her paracetamol dos-
age was adequate, she was able to discontinue the use of codeine.
Oral administration of medications is the route preferred by
most children and their parents, but if nausea and vomiting are
early symptoms, oral medications are not effective. In such chil-
dren, early treatment with an anti-emetic, such as cyclizine or
metclopromide, may offer good relief of nausea and may improve
the response to painkillers. In other children, the nasal route may
offer a good alternative. Sumatriptan as a nasal spray (10 mg) is
licensed for use in children over the age of 12 years and has been
shown to be effective in many but not all patients.
Preventing migraine headache
Preventative treatment of migraine is indicated if attacks occur
on at least four occasions per month and are severe and long
enough to stop activities, and when simple lifestyle measures are
ineffective. There are no reliable medications to prevent head-
ache in all children all the time, but pizotifen, propranolol and
possibly topiramate are worth trying as they may offer relief to
some children. Medication should be taken regularly for at least
2 months in appropriate dosages before its success or failure can
be confidently decided.
Managing tension-type headache
The management of tension-type headache can be tailored to suit
the child. In many children the use of simple analgesics is safe and
effective, and should be used early and in the full recommended
dose. In other children, frequent tension-type headache can be
transformed to chronic daily headache and becomes resistant to
analgesia. In these cases, overmedication headache is a real risk
and analgesia should be withdrawn in order to achieve resolution of
the daily headache. The withdrawal of analgesia can cause appre-
hension and worry, and also a transient worsening of the headache.
If children and their parents are properly warned of possible wors-
ening of symptoms during the first week of withdrawal, the compli-
ance with advice is usually good and improvement follows. ◆
Practice points
• Headache and migraine are common in children and
adolescents
• Headache can cause disruption to the child’s schooling and
social life
• Different types of headache frequently co-exist in the
individual child
• Headache diaries are useful in making the diagnosis of
different types of headache
• Investigations are rarely needed in children with typical
history and normal examination
• Explanation of the diagnosis and education of the child and
their family make compliance with advice and treatment easier
43 © 2007 Elsevier Ltd. All rights reserved.
 self-assessment

Self-assessment
Questions Creatinine 107 μmol/L (20–74 μmol/L)
Bilirubin 17 μmol/L (0–17 μmol/L)
Case 1 Alkaline phosphatase 566 IU (110–270 IU)
A 9-year-old boy presented to A&E with a 2-day history of GGT 649 IU (5–32 IU)
double vision, squint (right eye) and headache. He had been ALT 150 IU (10–35 IU)
apyrexial with no other symptoms. Past medical history was Amylase 225 IU (28–100 IU)
unremarkable. The observations were as follows: HR 80/min, CRP 6 mg/dL
RR 20/min and BP 120/75 mmHg. Examination was normal
Calcium (corrected) 2.7 mmol/L (2.15–2.60 mmol/L)
other than isolated 6th nerve palsy on the right side.
Phosphate 1.46mmol/L (1.10–1.80 mmol/L)
Blood sugar 4.7 mmol/L
1. What is the likely diagnosis? Choose ONE answer ONLY
from the following: 3. What is your diagnosis? Choose ONE answer ONLY from
A. Diabetic neuropathy the following:
B. Malignancy A. Viral hepatitis
C. Multiple sclerosis B. Autoimmune hepatitis
D. Myasthenia gravis C. Lymphoma
E. Meningitis D. Hepatoblastoma
2. What investigation will you request? Choose ONE answer
ONLY from the following: Case 2
A. HbA1C A 4-year-old boy presented to his GP with a 5-day history of
B. Full blood count and blood film vomiting and was diagnosed with viral gastroenteritis. He
C. Lumbar puncture improved temporarily over the next 2 days but needed hos-
D. Tensilon test pital admission when the symptoms worsened. The child did
E. MRI scan of brain not have any fever, diarrhoea or skin rash. He had not passed
F. Bone marrow aspiration urine over the previous day. He had been in good health prior
to this illness. On examination he looked unwell, sleepy and
MRI of the brain was normal and the ophthalmologist sug- was dehydrated. His abdomen was soft and there was no
gested eye patching. Two weeks later the boy presented with organomegaly. Bowel sounds were normal. Observations were
nausea, vomiting and reduced appetite. He had seen the GP the as follows: temperature 37° C, HR 130/min, RR 22/min, SaO2
previous week because of abdominal pain and constipation. 100% in air, BP 89/57 mmHg and capillary refill time 3 s. He
On examination he looked unwell with a distended abdomen. was administered a bolus of normal saline (10 ml/kg) and
A slightly tender mass was felt in the right hypochondrium intravenous fluid was commenced to correct the deficit and
extending 7 cm below the costal margin with a firm consis- provide maintenance.
tency and smooth contour. Bilateral inguinal lymphadenopa- The blood results showed the following:
thy was also noted. The full blood count showed:
Hb 13.4 g/dL (9.5–13.5 g/dl)
Hb 10.9 g/dL (9.5–13.5 g/dL) WCC 6.3 × 109/L (4.5–13.0 × 109/L),
WCC 10.7 × 109/L (4.5–13.0 × 109/L), Neutrophils 4.1 × 109/L (1.8–8.0 × 109/L)
Neutrophils 7.17 × 109/L (1.8–8.0 × 109/L) Lymphocytes 1.9 × 109/L (1.0–5.0 × 109/L)
Lymphocytes 2.03 × 109/L (1.0–5.0 × 109/L). Platelets 369 × 109/L (170–400 × 109/L)
There were no blast cells on peripheral blood film CRP <5 mg/dL
Na 136 mmol/L (135–145 mmol/L)
Na 134 mmol/L (135–145 mmol/L) K 4.4 mmol/L (3.8–5.5 mmol/L),
K 4.6 mmol/L (3.8–5.5 mmol/L) Urea 6.4 mmol/L (1.8–6.4 mmol/L)
Urea 9.3 mmol/L (1.8–6.4 mmol/L) Creatinine 52 μmol/L (20–74 μmol/L)

Liver function tests:

Archana Joshi MBBS DCH MRCPCH is a Specialist Registrar in Paediatrics at Total bilirubin 6 μmol/L (0–17 μmol/L)
the Lister Hospital, Corey’s Mill Lane, Stevenage, Herts SG1 4AB, UK. ALT 5 IU (0–35 IU)
Basheer P Mohamed MBBS MRCPCH is a Specialist Registrar in Paediatrics Alkaline phosphatase 100 IU (110–270 IU)
at the Lister Hospital, Corey’s Mill Lane, Stevenage, Herts SG1 4AB, UK. Total protein 62 g/L (64–83 g/L)
Jonathan Kefas MB DTCH MRCPCH is a Consultant Neonatologist at the Glucose 5.3 mmol/L
Lister Hospital, Stevenage, Corey’s Mill Lane, Herts SG1 4AB, UK. Urine dipstick showed 2 + of ketones

Paediatrics and child health 18:1 44 © 2007 Elsevier Ltd. All rights reserved.
 self-assessment

1. What is the likely diagnosis? Choose ONE answer ONLY bilateral weakness of extraocular eye movements. Meningo-
from the following: encephalitis can cause cranial nerve palsy, but it is unlikely in
A. Viral gastroenteritis the absence of other clinical signs of infection, as in this case.
B. Intussusception Therefore, a malignant or an infiltrative process is the most
C. Appendicitis likely diagnosis.
D. Inflammatory bowel disease
2. E
He continued to vomit following admission and his obser- Isolated 6th nerve palsy (one or both) is a relatively common
vations overnight were as follows: HR 65/min, RR 26/min, feature of generalized increase of intracranial pressure. Non-
BP 125/80 mmHg, SaO2 89% in air, overnight urine output solid tumours such as leukaemia can impair the reabsorptive
2.3 ml/kg/h. mechanism in the subarachnoid space leading to increased
intracranial pressure and sometimes hydrocephalus. Therefore,
2. What is your diagnosis? neuroimaging should be performed urgently. MRI is better than
A. Haemolytic uraemic syndrome CT scan for infratentorial lesions. Lumbar puncture should not
B. Intracranial space occupying lesion be performed without ruling out raised intracranial tension.
C. Superior sagittal sinus thrombosis
D. Diabetic ketoacidosis 3. C
Acute lymphoblastic leukaemia and non-Hodgkin lymphoma
Case 3 (non-solid tumours) can present with neurological signs. Viral
A 10-year old girl was brought to A&E by grandparents late in and autoimmune hepatitis may rarely be associated with neu-
the evening, as they were unable to cope with the child’s loss rological features, especially isolated cranial nerve palsies.
of appetite and nausea. The child had recently lost her mother Hepatic tumours are rare in children. Metastasis from other
and the whole family was grieving. Further inquiry revealed a common childhood malignancies should always be ruled out.
history of occasional vomiting, headaches and weight loss. She Hepatoblastoma generally presents in the first 18 months of
had been a fit and healthy child previously. A brief examination life as an asymptomatic abdominal mass with other features
done by the doctor was normal. The girl was cooperative but like abdominal pain, vomiting, weight loss and metastasis
wanted to go home to her bed. Hence the doctor sent the child occurring later.
home but with a plan for admission to the ward the next day.
Case 2
1. What is the most likely diagnosis? Choose ONE answer 1. A
ONLY from the following: Intussusception is unlikely in the absence of abdominal pain
A. Depression and diarrhoea (note: the classic redcurrant jelly stool is usu-
B. Anorexia nervosa ally a late sign). Appendicitis is usually diagnosed from the
C. Malignancy right iliac fossa tenderness and guarding. Inflammatory bowel
D. Bulimia disease has a more insidious onset with diarrhoea, weight
loss, loss of appetite and raised inflammatory markers.
She was admitted to the ward the next day and a detailed
clinical examination revealed loss of pupillary light reflex 2. B
with upward gaze palsy and nystagmus on attempted upward This patient presented with classical features of Cushing’s
gaze. reflex – hypertension and bradycardia – which is seen in raised
intracranial pressure. Abnormal breathing pattern (Cheyne–
2. What is the clinical feature described? Choose ONE Answer Stokes breathing) is part of the triad of symptoms. Superior
ONLY from the following sagittal sinus thrombosis should be considered in any child
A. Horner syndrome with severe dehydration and evolving neurological symptoms.
B. Trochlear nerve palsy Absence of tachycardia and a good urine output indicates a
C. Parinaud syndrome well-hydrated child. Haemolytic uraemic syndrome presents
D. Oculomotor nerve palsy with anaemia, thrombocytopaenia and impaired renal func-
tion. Diabetic ketoacidosis is unlikely given the normal blood
glucose. In this case, an urgent MRI was performed, which
Answers revealed a medulloblastoma.

Case 1
1. B
Diabetic neuropathy and multiple sclerosis are unlikely to
present at this age, although they should be considered in
the differential diagnosis. Myasthenia gravis presents with
Case 3
1. A
Depression is often under diagnosed in children. It may be pre-
cipitated by changes in body, roles and relationships. Further-
more, bereavement is an important cause of major depression.

Paediatrics and child health 18:1 45 © 2007 Elsevier Ltd. All rights reserved.
 self-assessment
Depressed mood, weight loss or gain, insomnia or hypersom-
nia, fatigue, feeling of worthlessness or guilt are features of
depression. When these symptoms occur within 3 months of
an identifiable stressor, this is considered as adjustment dis-
order with depressed mood. Anorexia nervosa is a ­psychiatric
disorder characterized by intense fear of becoming obese,
altered self perception of body image and refusal to main-
tain body weight over a minimal normal weight for age and
height. Bulimia is defined as a separate clinical entity by DSM
IV ­criteria with the following features: recurrent episodes of
binge eating and fear of not being able to stop eating during a
binge, and regularly engaging in activities such as self-induced
vomiting, abuse of laxatives, rigorous exercise which counter-
acts the effects of binge eating. Eating disorders are commonly
seen in the 16–18 year age group and more commonly involve
girls than boys. The incidence is on the rise in all western
countries. Malignancy should be considered in the differential
diagnosis of weight loss in children and hence a full clinical
examination is vital, before a diagnosis of non-organic cause
is made.
2. C
The clinical features are consistent with a condition referred
to as Parinaud syndrome, which is due to mid-brain dysfunc-
tion resulting from pressure by pineal tumours. The clinical
Paediatrics and child health 18:1
features include supranuclear palsy of upward gaze with
preservation of downward gaze, loss of pupillary light reflex
and nystagmus on attempted upward gaze. This child was
diagnosed with pinealoma. A pineal tumour compressing the
anterior hypothalamus causes loss of vision, diabetes insipi-
dus, precocious puberty, emaciation (diencephalic syndrome)
and sometimes obesity from insatiable appetite.
Horner syndrome results from cervical sympathetic chain
injury and comprises ptosis, miosis, anhydrosis and enoph-
thalmos. Ptosis in Horner’s syndrome is often incomplete,
whereas in 3rd nerve palsy there is complete ptosis along with
mydriasis. There is inability to elevate and adduct the eye.
Trochlear nerve palsy, on the other hand, results in paralysis
of the superior oblique muscle and hence causes difficulty
with looking down and in. Children typically present with a
head tilt and face turned towards the unaffected side.
Further reading
Fenichel GM. Clinical Pediatric Neurology: A Sign and Symptoms
Approach, 5th edn. Philadelphia: WB Saunders, 2005.
Kleigman RM, Behrman RE, Jenson HB, Stanton BF eds. Nelson’s
Textbook of Pediatrics, 18th edn. Philadelphia: WB Saunders,
2007.
46 © 2007 Elsevier Ltd. All rights reserved.