Search tips
Search criteria 


Logo of archdischArchives of Disease in ChildhoodVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Arch Dis Child. 2007 April; 92(Suppl1): A56–A62.
PMCID: PMC2066115


Neurology and palliative medicine joint session

G/WEDS/PAN1 Evidence for a lack of early structural facial encoding in a child with Prosopagnosia acquired in infancy

A. Basu1, T. Kelly2, N. Greening1, S. Mafeld1, J. Eyre1. 1Child Health, University of Newcastle upon Tyne, Newcastle, UK; 2Neuropsychology Department, Newcastle General Hospital, Newcastle, UK

IntroductionProsopagnosia is impaired or absent face recognition despite normal visual processing of other objects. In healthy adults and children, face recognition evokes an early EEG potential, the N170, which is characteristically greater in amplitude to inverted faces and reflects early structural encoding of faces. Our aim was to determine whether a child with prosopagnosia had evidence of early facial encoding and might therefore be able to be taught to recognise faces.

MethodsThe subject (RC; female, 13 years) who was of normal intelligence, had focal damage to the occipital and temporal lobes from herpes simplex encephalitis, age 11 months. Control subjects were 10 young adults and 3 children aged 7, 9 and 12 years. RC had standardised neuropsychological assessments of visual processing. EEG was recorded from all subjects while they viewed 64 randomly ordered, sex‐matched, presentations of their own face, famous persons' faces and unknown faces, either in upright or inverted orientations. A control image of a black square was also presented. The amplitude and latency of the N170 were compared (1) using averaged event related potentials and an ANOVA for adults (2) using independent component analysis and a z‐test for all subjects.

ResultsNeuropsychological testing: RC demonstrated normal object recognition despite some object naming difficulties. There was no evidence of facial recognition. EEG: There was no difference between RC and normal subjects for the potentials evoked by the control stimulus. All normal subjects demonstrated an N170 to faces, which was larger (p<0.001) and delayed (p = 0.004) to face inversion, but not affected by face familiarity. RC did not demonstrate an N170, even to her own face.

ConclusionThe results highlight the importance of very early facial structural encoding for normal face recognition, even for one's own face. It is unlikely that RC could learn to recognise faces.

G/WEDS/PAN2 Continuous subcutaneous infusion of phenobarbitone in the management of status epilepticus in paediatric palliative care patients: a descriptive case series

D. Rajapakse, F. Craig, M. Koh. Great Ormond Street Hospital, London, UK

IntroductionContinuous subcutaneous infusion (CSCI) of phenobarbitone is unlicensed for use in status epilepticus (SE) in children.

AimsTo describe the demographics and diagnoses of five paediatric palliative care patients who received CSCI of phenobarbitone for SE. To describe the effectiveness and outcome of phenobarbitone via CSCI in these patients.

MethodA retrospective analysis of the case records of five patients treated with CSCI of phenobarbitone for SE, referred to a tertiary palliative care team was carried out.

ResultsThe five children were aged 3–13 years, and four were boys. Diagnoses were undiagnosed progressive epileptic encephalopathic disorders (2), adrenoleukodystrophy (1), undiagnosed Batten's‐like disorder (1) and progressive cerebral HLH (1). In all the cases, seizures were not controlled by combinations of enteral and intravenous anti‐convulsant therapies. The seizures ranged from focal to generalised attacks. All the children had received up to 4 doses of buccal midazolam or rectal diazepam, and up to 2 doses of rectal paraldehyde after onset of SE. Three children had been commenced on CSCI of midazolam. In three cases CSCI of phenobarbitone was added as seizures had continued for over 24 h despite increasing the dose of midazolam infusion, and in two cases phenobarbitone by CSCI was started without midazolam infusion as the seizures had not responded to buccal midazolam. In all cases seizure control was established within 12 h of starting the phenobarbitone infusion, and required 1–3 dose increments of 20%. In all cases there were no further episodes of SE, and the only side effect was increased drowsiness in all children. The period on the phenobarbitone CSCI ranged from 5 days to 6 months. All the children died from non‐seizure related terminal events, whilst on this treatment. Two children died at home, 2 in a children's hospice and 1 in hospital. The infusions were changed by the parents (1) or by community and hospice nurses (4).

DiscussionThis study suggests the efficacy of phenobarbitone via CSCI as an adjunct to other agents or as the sole infusion agent in the management of SE in paediatric palliative care patients with a variety of progressive neurological conditions. There is rapid and sustained establishment of seizure control, with minimal side effects. The CSCI can be administered either at home or in hospices, by a range of trained personnel.

G/WEDS/PAN3 Clinical usefulness of personal resuscitation plans in children with neurodisability and life‐limiting conditions

T. Wolff1, S. Hollingsworth2, W. Whitehouse3. 1Nottingham University Hospitals, Nottingham, UK; 2Broxtoe & Hucknall PCT, Nottingham, UK; 3University of Nottingham, Nottingham, UK

IntroductionWe have previously presented the template for family held personal resuscitation/emergency care plans for children with neurodisability and life‐limiting conditions. We are now looking at the effect on patient care from a parent/carer and nursing perspective.

AimsTo audit the usefulness of personal resuscitation plans (PRPs) in a population of children with neurodisability and life‐limiting (LLC) or life‐threatening (LTC) conditions.

MethodsThe community children's nursing team identified all children on their case load at the end stage of a LLC/having life‐threatening events. This team serves children with neurodisabilites and nursing needs in a population of 600 000 (150 000 children). Nurses completed an audit tool after supporting a life threatening event. Case notes were reviewed. Parents were sent a postal questionnaire.

ResultsIn the 12 months, Nov 05–Nov 06, 19/24 children with LLC/LTC had a PRP (9 male): severe CP = 7, degenerative disease = 12, median age 12 years (range 3.3–19). Seven children died (4 at home). PRPs were followed and found very useful (7) by professionals. In 3 cases the PRP changed to less intervention in final 24–48 h and supported that discussion/process. Four children recovered from life threatening episodes. PRP available, followed and very useful (3) or quite useful (1). One event revealed disagreement with plan by respite care nurses. In one event (again in respite care) ambulance directive could not be found. 6/10 parent satisfaction questionnaires returned. Quite easy to complete (4), quite difficult (1), very difficult (1). Would recommend to other families (6). Only 1 life‐threatening event reported, in which PRP was “quite useful”. None of the children was admitted to PICU.

ConclusionsFamily‐held personal resuscitation plans for this client group are acceptable and useful to families and professionals.

G/WEDS/PAN4 Cerebral venous thrombosis in children: a case series from two centres

R. Belderbos1, S. Joseph1, V. Ganesan1, E. Wraige1. 1Great Ormond Street Hospital, London, UK; 2Evelina Children's Hospital, London, UK

AimsTo determine clinical features of cerebral venous thrombosis including symptoms and signs; predisposing factors; treatment; radiological findings and outcome and whether those with infection require different treatment and investigations.

MethodsCase note review: 38 cases.

ResultsPresenting features: headache 39%; reduced consciousness 55%; vomiting 45%; confusion 29%; fits 45% (most focal, multiple or prolonged); fever 47%; hemiplegia 16%; ataxia 10%; opthalmoplegia 37%; reduced acuity 12%; papilloedema 21%; hypernatraemic dehydration 10%. Risk factors: infection 68% (2/3 additional risk factor; 42% positive thrombophilia screen); no infection 31% (58% additional risk factor; 50% positive thrombophilia screen); dehydration 32%; anaemia 29%; polycythaemia 8% (neonates 30%); thrombocythaemia 16%. Positive thrombophilia screens: 5% protein C def; 8% antithrombin III def; 8% DRVVT positive; 8% antiphospholipid antibodies; 3% homozygous, 16% heterozygous MTHFR mutation; 3% heterozygous factor V leiden mutation; 21% not done. Radiology: CT 38% normal; MRI 8% normal initially; infarcts 46%: haemorrhagic 25%. Follow‐up scan: not done 50%; resolved/improved 66%; not improved 4. Treatment: IV heparin 5%; SC Heparin 26%; IV then SC heparin 8%. Long‐term treatment: 47%: warfarin 24%; heparin 5%; aspirin 18%.

figure ac07abs18.f1
Abstract G/WEDS/PAN4.

ConclusionCerebral venous thrombosis is a potentially treatable cause of mortality and morbidity in children. The diagnosis is easily missed due to non‐specific presenting features. Recommendations exist for treatment and investigation but the evidence for this is lacking. Questions regarding optimum treatment, investigation and prophylaxis remain.

G/WEDS/PAN5 The role of a tertiary centre paediatric palliative care team within a managed network

F. Craig, D. Rajapakse, M. Koh, A. K. Anderson. Great Ormond Street Hospital for Children NHS Trust, London, UK

AimTo determine the referral pattern of patients to a tertiary centre palliative care team and demonstrate the need for specialist teams as part of managed palliative care networks.

MethodsData were collected on all children referred to a tertiary centre palliative care team over 4 years (2003–6). Children treated for cancer at the tertiary centre were not included in the study as there was already an established network for palliative care referral. Data were analysed to identify the main referrers to the service and to determine if there were specific reasons for referral. Place of death was also recorded to determine if the team was supporting end‐of‐life care outside the tertiary centre.

Results330 children were referred to the palliative care service, with a 40% increase in referrals between the first and 4th years. Although the majority of referrals were from tertiary centre consultants, the proportion of referrals made by community children's nursing teams, Big Lottery Fund (BLF) teams and local hospitals increased from 17% to 31% over the 4 years. Only 7% of children were referred for management of specific symptoms, with 74% of referrals being for “general palliative care support”. There were 171 deaths, 88% of which occurred outside the tertiary centre. Over the 4 years there was an increase in the proportion of children receiving end‐of‐life care in their local hospice, from 9% to 25%.

ConclusionsThe specialist palliative care team receives a significant proportion of referrals from Community Children's Nurses, BLF teams and local paediatricians. The main reason for referral is for “general palliative care support” rather than for specific issues. Although based in a tertiary centre, the team is able to support end‐of‐life care in the community. This use of the tertiary centre palliative care team within the community setting confirms an essential role for the specialist palliative care team as part of a managed palliative care network.

G/WEDS/PAN6 Opsoclonus myoclonus syndrome in childhood: guideline for diagnosis, treatment and follow‐up

A. Brunklaus, C. de Sousa. Great Ormond Street Hospital, London, UK

BackgroundOpsoclonus myoclonus syndrome (OMS), also called “dancing eye syndrome”, is a rare but serious neurological condition presenting with opsoclonus, cerebellar ataxia, myoclonus and behavioural change in early childhood. The underlying pathology is thought to be immune mediated, either as paraneoplastic manifestation of occult neuroblastoma or as molecular mimicry post viral infection. Children can present with severely disabling neurological and behavioural symptoms and often continue to have long‐term motor as well as cognitive impairment. Treatment strategies are mainly based on immune‐suppressive and immune‐modulating agents and have been shown to be effective in the short term. However the treatment course is complicated by frequent relapses and there is a lack of evidence for long‐term benefit from drug therapy.

AimsWe propose a guideline as an aid in assessment and diagnosis of OMS, which contains an evidence‐based treatment protocol and a structured follow‐up, including regular neurological, behavioural and cognitive assessment.

MethodsWe performed a literature search via PubMed from 1980 to 2006 using the key words “opsoclonus myoclonus syndrome”, “dancing eye syndrome”, and “developmental sequelae” focussing on best evidence diagnostic tools and treatment protocols.

ResultsThe guideline consists of 3 parts: (A) Diagnosis, (B) Treatment and (C) Follow‐up. Part A defines diagnostic criteria for OMS and proposes assessment tools such as laboratory investigations, imaging and neurodevelopmental testing. Part B suggests a standardised treatment protocol based on latest advances in research. Part C proposes a structured follow‐up including regular grading of symptoms and developmental/behavioural assessment via rating scales, as well as intermittent cognitive testing.

ConclusionThe use of an evidence‐based guideline for OMS optimises current treatment, enables an accurate evaluation of long‐term outcome and will ultimately improve future patient care.

G/WEDS/PAN7 Are paediatric trainees gaining experience and knowledge of palliative care through their current training programme?

A. Anderson1, F. Craig2. 1Royal London Hospital, London, UK; 2Great Ormond Street Hospital, London, UK

AimThe RCPCH has clearly defined a palliative care knowledge base and competencies to be achieved by the end of paediatric training. The aim of this study was to determine if Paediatric trainees are achieving these targets.

MethodUsing RCPCH guidelines a questionnaire was developed asking 6 questions to identify knowledge of local palliative care services and national guidelines. This was distributed to all email accessible paediatricians and paediatric trainees at one paediatric teaching hospital.

ResultsThe response rate was 72% (18 doctors) across all grades. Trainees and consultants showed limited awareness as to which children could be referred to palliative care services and 41% did not know when referral would be appropriate. Consultants and trainees demonstrated good knowledge of children's hospices and specialist teams, but only a few recognised the importance of the GP and community nurses. None considered his or her own role or that of the multidisciplinary team as important. Trainees were not aware of national and local guidelines for withholding and withdrawing treatment, or of guidelines on brainstem death testing. Almost half (47%) did not know what bereavement services were available for families. Most doctors did not know where to seek support for themselves.

ConclusionBoth trainee and consultant paediatricians fail to identify children who could be referred to palliative care services, are unaware that referral should be made before death is imminent and lack knowledge of bereavement services. Trainees are not familiar with important clinical guidelines both nationally and locally. This study demonstrates very basic yet significant deficiencies in palliative care knowledge and competencies amongst Paediatricians and trainees. It highlights the need for improved training which we believe should be integrated into local teaching programmes, so that both consultants and trainees can benefit.

G/WEDS/PAN8 Normal values for and feasibility of the Gower and 10 metre running tests in young children

A. Priestman1, P. Baxter2, A. Rigby3. 1University of Sheffield, Sheffield, UK; 2Sheffield Children's NHS Trust, Sheffield, UK; 3Department of Cardiology, University of Hull, Hull, UK

AimsTo define normal values and feasibility for two clinical tests of strength in young children, standing from supine and running 10 metres, and to assess the effect of age, gender and BMI.

MethodEighty children were recruited from local nurseries, 42 girls, 38 boys, 4 aged 2–3, 48 aged 3–4 and 28 aged 4–5 years; mean age 3.8 years, range 3–4.6 years. After standardisation each test was carried out twice, timed and videoed.

ResultsThe mean time taken to stand from supine was 2.4 seconds (range 1.1–5.1). There was a negative correlation between age and time to stand in both sexes, which was statistically significant in girls (p = 0.006) but not boys (p = 0.19). There was no gender difference in the time taken to stand. Neither group showed a correlation between BMI and time to stand (girls p = 0.92, boys p = 0.72). A variety of methods were adopted to stand from supine, which included sitting and then standing straight up, rolling or turning to one side before standing and turning over onto all fours. Comparing age groups and the method used to stand, children aged over 4 years were more likely to roll to a side before standing than children under 4 (95% CI 1.6–13.3 p<0.05), but less likely to turn over onto all fours, although the latter was statistically insignificant (95% CI 0.05–4.6 p>0.05). Comparing gender and the method used to stand, boys are 3.2× as likely as girls to roll to the side, when compared to standing straight up, (95% CI 1.2–8.9 p<0.05). Similarly boys are 11.6× as likely to turn over onto all fours compared to girls (95% CI 1.3–105 p<0.05). The mean time taken to run 10 metres was 4 seconds (range 2.7–5.9). This showed a significant negative correlation with age (p = 0.001). There was no gender difference in the time taken. However nursery environments were too inconsistent to allow standardisation.

ConclusionStanding from supine (the Gower test) can be used as a clinical test in all nursery settings, but not the 10 metre run test. Standing from supine requires minimal equipment and can be used in children as young as two. The time taken to stand is affected by age, but not by gender or BMI. The method of standing is affected by both age and gender.

G/WEDS/PAN9 Paediatric palliative care drug boxes: facilitating safe and effective symptom management at home at end of life

L. Brook, J. Vickers, C. Osbourne. Alder Hey Children's Hospital, Liverpool, UK

BackgroundEight years ago a system of palliative care drug boxes was developed in order to avoid delays in initiating subcutaneous or intravenous infusions for symptom management at end of life. The boxes contain necessary medication for symptom management pre‐prescribed to be administered via continuous intravenous or subcutaneous infusion. The boxes are prescribed a few days before they are expected to be needed and remain in the home until after the child has died.

MethodsRetrospective review of the palliative care drug box prescriptions, medication use and outcomes during the period July 2001 to June 2006.

ResultsSeventy four boxes (34 intravenous and 40 subcutaneous) were prescribed for 69 children: 50 with cancer and 19 with other life‐limiting conditions. Two children each had 3 box prescriptions at different times reflecting difficulty in identifying end of life. 21 palliative care drug box prescriptions were not used (8 oncology and 13 non‐oncology). Eighty infusions were commenced via Graseby MS26 syringe driver. Most common combinations were; diamorphine, midazolam and leveomepromazine (n = 13); diamorphine, midazolam and cyclizine (n = 11); diamorphine and cyclizine (n = 9). Contents of the syringe were renewed every 24 h and continued for a median of 75 h (interquartile range 17–256 h). 78% of symptoms were controlled with a combination of one or more of the following: a strong opiate, (morphine or diamorphine) cyclizine, haloperidol, levomepromazine, midazolam and hyoscine hydrobromide. Where medication other than these 6 “essential drugs” was required to control symptoms this had usually been started before end of life care. Drug boxes remained in the house a median of 4 days (range <1 to 106 days). Despite several families with known substance abusers all medication was accounted for except 1 instance when the morphine “disappeared” from the unused box after the child's death.

ConclusionsPaediatric palliative care drug boxes containing 6 “essential drugs” are effective in controlling the majority of symptoms at end of life.

G/WEDS/PAN10 Excess baggage: bilateral brachial plexus palsy in a teenager after carrying a heavy rucksack – an unusually severe first manifestation of hereditary neuropathy with liability to pressure palsies (HNPP)

P. Rao1, T. Fenton1, M. Lim2, S. Mansour3. 1Mayday University Hospital NHS Trust, Croydon, London, UK; 2Evelina Children's Hospital, London, UK; 3St George's Hospital, London, UK

IntroductionHereditary neuropathy with liability to pressure palsy (HNPP) is characterised by recurrent mononeuropathies following minor trauma. We report a patient with a fulminant presentation following modest physical activity.

Case synopsisA 14‐year‐old boy developed weakness of both arms while carrying a 15 kg rucksack during a walk for his Duke of Edinburgh Award. The weakness remained, and he was referred to clinic where he was seen 3 weeks later. Examination showed bilateral asymmetric brachial plexus palsy, with upper brachial predominence on the right and lower brachial plexus predominence on the left.

Our patient's father and paternal grandfather had a clinical diagnosis of Charcot Marie Tooth (CMT) disease. However, father had experienced at least three self‐resolving episodes.

Our patient had a normal magnetic resonance imaging study of the brain, cervical spine, nerve roots and brachial plexus. Nerve conduction studies showed a motor and sensory demyelinating neuropathy with an emphasis on pressure sites, supporting the clinical suspicion of HNPP. Electromyographically, there was also evidence of acute denervation.

Genetic study revealed deletion of all 5 exons at 17p11.2, confirming the diagnosis HNPP. Our patient made steady functional recovery over the next 4 months, with support by the physiotherapists and occupational therapists.

DiscussionAlthough some of the clinical signs of HNPP can mimic CMT, careful history often reveals the liability to pressure palsies as in our patient's father, whose diagnosis has since been revised. CMT is often due to a duplication in PMP‐22, and individuals with HNPP often later develop features of CMT. HNPP is likely to be more common and its presentation more varied than currently thought.

G/WEDS/PAN11 “Epilepsy 12” – A pilot of standardised clinical performance indicators for regional services

C. Dunkley1, K. Martin2, R. Sunley2, S. Gough3, C. Ferrie4, W. Whitehouse5. 1King's Mill Hospital, Sutton in Ashfield, UK; 2Nottingham University Hospitals NHS Trust, Nottingham, UK; 3Lincoln County Hospital, Lincoln, UK; 4Department of Paediatric Neurology, Leeds General Infirmary, Leeds, UK; 5School of Human Development, University of Nottingham, Nottingham, UK

Meaningful audit of epilepsies in children is important but difficult to achieve in practice.

AimsTo develop and pilot practical clinical audit tools which could provide standardised and nationally applicable measures of quality of services for children with suspected epilepsies.

MethodChildren presenting to 4 hospital services; acutely or non‐acutely; with paroxysmal event(s); in whom an epileptic basis was suspected; assessed for the first time by a paediatric service; within a defined 10‐week period (2004) were ascertained. A standardised audit proforma was applied retrospectively to casenotes. This was anonymised and then inputted centrally into a spreadsheet template by the coordinating clinician. Methods of ascertainment, demographics, waiting times, 12 clinical performance indicators and descriptors of service were defined for the whole cohort and each cohort individually.

Results198 children seen with paroxysmal episodes were identified of which 40 met inclusion criteria. At first assessment, 23 (57.5%) episode(s) were diagnosed as epileptic, non‐epileptic 7 (17.5%), and uncertain 10 (25%). Diagnosis at 1 year was recurrent epileptic seizures (epilepsy) 15 (37.5%), of which 8 were commenced on anti‐epileptic drug treatment (AEDs); single epileptic seizures 8 (20%); non‐epileptic 10 (25%); and uncertain 7 (17.5%). 12 performance indicators were calculated. For example 45% of children assessed met criteria for evidence of appropriate history and examination; 100% of children prescribed AEDS at any time within the first year maintained a diagnosis of epilepsy at 1 year.

ConclusionsThe target population was difficult to ascertain and community based services were not represented. Retrospective casenote analysis has a number of methodological limitations. Performance indicators can be determined and act as broad ‘feedback' for local services. This could prompt service development or more detailed analysis of audit data acquired. “Epilepsy 12” and supporting audit tools can be coordinated on a regional basis and are proposed as a useful tool for regional epilepsy networks.

G/WEDS/PAN12 Can clinical features, seizure semiology and initial electroencephalogram (EEG) findings predict outcome in childhood absence epilepsy?

K. Kneen2, R. Kneen1, R. Appleton1. 1Littlewoods Neurosciences Foundation, Royal Liverpool Childrens NHS Trust, Liverpool, Merseyside, UK; 2Countess of Chester Hospital NHS Foundation Trust, Chester, UK

AimsChildhood absence epilepsy (CAE) is a common paediatric epilepsy syndrome but most studies looking at prognosis include patients with absence seizures who have other epilepsy syndromes. We aim to describe the clinical and EEG findings in children with CAE (as defined by The International League Against Epilepsy criteria) to try and identify factors at diagnosis that may predict seizure control and remission.

MethodsCasenotes of children identified with CAE from our region were reviewed and findings were recorded on a proforma. A single assessor reported EEG findings.

ResultsFifty one children [35 (69%) girls] were identified. Median (range) age at diagnosis was 7 (4–11) years. Median length of follow‐up was 38 (12–156) months. Forty two (84%) achieved seizure control after a median of 4 (1–49) months; the remaining 9 did not achieve control despite treatment with up to 7 drugs. Twenty nine children were followed up for more than 36 months and 18 (62%) of these children achieved remission. Two of the children who did not achieve remission went on to have a generalised tonic clonic seizure during the period of follow‐up. Six (12%) of children had a family history of epilepsy in a first‐degree relative, 5 (10%) had a history of behaviour problems, 10 (20%) had mild learning difficulties (mainstream school with help), 7 (14%) had a history of absences longer than 20 seconds and 28 (55%) reported automatisms. Specific features of the EEG were asymmetry of discharge onset [5 (10%)], slow background [8 (16%)], polyspike and slow wave [8 (16%)], eyelid myoclonia noted [4 (8%)] and photosensitivity [2 (4%)]. None of the features of the history, seizure semiology or EEG had a statistically significant relationship to seizure freedom at 12 months or remission. Seizure freedom at 3 months predicted remission (relative risk 1.93, 95% confidence interval 1.10 to 3.77, p = 0.05)

ConclusionNo factors were identified that could predict seizure outcome in CAE at diagnosis. However, if seizure‐free at 3 months after diagnosis, a child is twice as likely to achieve remission as a child who is not.

G/WEDS/PAN13 A prospective observational study of the use and effects of Levetiracitam in children with epilepsies

R. Wheway1, G. Chow2, D. Peake3, S. Philip3, S. Mordekar4, C. Rittey4, N. Hussain5, L. Mewasingh5, J. Gosalakkal5, W. Whitehouse1. 1University of Nottingham, Nottingham, UK; 2Nottingham University Hospitals, Nottingham, UK; 3Birmingham Children's Hospital, Birmingham, UK; 4Sheffield Children's Hospital, Sheffield, UK; 5Leicester Royal Infirmary, Leicester, UK

Aimsto systematically assess the effects of Levetiracitam (LEV) in children with intractable epilepsies, including prospectively recorded seizure frequency and quality of life, following the previously reported retrospective study.

MethodsA multicentre, open, observational study in paediatric neurology departments in the Midlands, UK. Following Research Ethics Committee and Research and Development department approvals, and informed consent, participants and parents/guardians kept a 2 week baseline diary then started LEV as clinically indicated. Quality of life (QoL), epilepsy impact and seizure diaries were used pre‐treatment and at intervals on treatment. Simple descriptive statistics have been used preliminarily.

ResultsSo far data on 11 children, 5 female, aged 2–15 years (mean 8.5 years) have been collected. Most had intractable epilepsies and had failed to respond to 2 to 5 other antiepileptic drugs (AEDs). 8/11 have had follow‐up for 2 <6 months and 1/11 beyond 6 months. Maintenance/maximum doses ranged from 14 to 55 mg/kg/day (mean 41 mg/kg/day). 3/9 were still on LEV at last observation and had a greater than 50% reduction in seizure frequency. All reported additional benefits. 5/11 reported possibly or probably related adverse events, but there have been no serious adverse events so far.

ConclusionsThe preliminary data suggest that LEV will prove effective in a significant minority of this population. QoL, further efficacy, and tolerability data will be available in the next 3 months.

AcknowledgementsWe thank all the participants and their parents/guardians for their help with the study and UCB Pharma for unrestricted financial support for the part‐time research nurse investigator.

G/WEDS/PAN14 Can a paediatric care pathway be used for the management of children with febrile convulsions?

J. Mammas, A. Hughes, L. Breen. Arrowe Park Hospital, Wirral, Merseyside, UK

IntroductionThe role of clinical care pathways is to improve the quality of treatment, the coordination of care and the satisfaction of patients treated in hospitals. Since to date there is no published evidence of a clinical care pathway for children with febrile convulsions (FC), the most common type of convulsions in childhood. The aim of our study is to report our experience from using a paediatric care pathway in children with FC in our department and to evaluate its acceptance to the paediatric staff.

MethodsWe reviewed retrospectively the case notes of all children with a discharge diagnosis of FC who were admitted to the paediatric department over a 12‐month period (15/2/2005–15/2/2006). Data were extracted based on the parameters of the completion of the pathway and the management of the admitted children. A self‐administered questionnaire was also used to assess the acceptance of the pathway by the paediatric staff.

ResultsOut of 50 cases of children with FC, the pathway was followed in 41 children. The pathway was not used in 2 children aged less than 6 months, in 3 children with a floppy/vacant episode without an initial suspicion of pyrexia on admission, in 2 children with pyrexia and not well described FC episodes on admission, in 1 child with chicken pox and in 1 case of prolonged FC more than 15 minutes. Parental education and information sheets were given to all parents/guardians in the cases where the pathway was followed, while only in 2 out of 9 cases where the pathway wasn't followed parental education was documented. No differences were observed in the investigations requested, treatment, duration of admission and the outcome of children who followed or not the pathway. A total of 33 out of 36 (94%) paediatric staff that completed the personal questionnaire considered the paediatric care pathway helpful for data collection, 92% useful in terms of medical and nursing co‐ordination, 86% useful in terms of parental education and satisfaction, while only 53% considered it helpful for the decision on patient's treatment.

ConclusionOur study indicates the role of a care pathway as a useful tool for the management of children with FC. Further review and feedback of the pathway is going to improve its usage in clinical practice.

G/WEDS/PAN15 The diagnostic value of paediatric EEG‐video telemetry in identifying Paroxysmal non‐epileptic events in children with epilepsy

N. Hussain, J. Gosalakkal. Leicester Royal Infirmarmy, Leicester , UK

AimTo report on the usefulness of adding video telemetry to routine EEG studies of infants and children with frequent atypical paroxysmal events.

MethodsWe analysed the efficacy of this diagnostic means during a 2‐year period. The decision whether to add video recording was made by the paediatric neurologist, when there was a diagnostic doubt about atypical events, which included: paroxysmal eye movements, absence like (staring) episodes, cyanotic episodes, suspected psychogenic non‐epileptic events etc. Video EEG recordings were done for 1–5 days. Outcomes were classified as: ‘useful‐epileptic' (successful classification of epilepsy), ‘useful‐nonepileptic' (demonstration of non‐epileptic habitual events), ‘uneventful' (normal EEG without habitual events captured) and ‘inconclusive' (inability to clarify the nature of habitual events with abnormal inter‐ictal EEG findings).

Results40 children (17 male, 23 female) had video telemetry. Age of the patient ranged from 1 to 15 years (mean: 7 years; median age: 8.5 years). The mean length of stay was 2.5 days (range: 1–5 days). The major subgroups of non‐epileptic events were: staring/daydreams (32%), sleep phenomena – night terrors (15%), syncope (20%), motor tics (11%), ritualistic movements (15%) and shuddering (7%). Developmental delay (60%) was common in children who presented with staring spells. A diagnosis of a specific non‐epileptic event was reached in 55% of cases. Based on identification of non‐epileptic events, antiepileptic drugs (AEDs) were discontinued completely in eight patients (20%) and the total number of AEDs was reduced in thirteen others (33%).

ConclusionParoxysmal non‐epileptic events can cause diagnostic confusion, particularly in children with developmental delay, epilepsy (especially refractory epilepsy) or those with previous epileptiform EEG. A diagnosis of non‐epileptic events should be considered in all children with refractory seizures or multiple seizure types. Accurate diagnosis can be reached in the majority of cases using prolonged video EEG monitoring.

G/WEDS/PAN16 Arnold Chiari I malformation in children: a potentially life threatening emergency

N. Hussain, J. Gosalakkal. Leicester Royal Infirmary, Leicester, UK

IntroductionThe Arnold Chiari malformation (ACM) is a congenital brainstem abnormality characterised by caudal herniation of the cerebellar tonsils through the foramen magnum, resulting in crowding at the cranio‐cervical junction. Exact incidence in children is not known. Typically it presents in adults. It can result in sudden death and long‐term neurological deficits. The availability of MR scanning has lead to an increasingly earlier diagnosis of the Chiari Malformation before the onset of frank neurological dysfunction especially in children.

AimTo study the clinical presentation of ACM type I in children presenting to a tertiary neurology centre.

MethodAll children with a diagnosis of ACM type I admitted to our centre over the past 1 year were studied. Results: five children had been diagnosed with ACM type I. They presented with different symptoms: the first child with snoring and sleep disordered breathing, the second child with headache, neck pain, difficulty in lifting his arm above the shoulders, the third child presented with ataxia and down‐beating nystagmus, the fourth with headache especially in the occipital region with no focal neurological signs and in the fifth child, ACM was diagnosed following a head injury sustained after a fall from a height. All had ACM type 1 and required urgent referral to neurosurgeons for decompression. Their symptoms improved after decompression surgery.

ConclusionSigns and symptoms of ACM can be subtle and evolve slowly. Paediatricians need to be aware of the varied presentation of ACM. Good history and clinical examination is always helpful. Children with signs and symptoms suggestive of ACM should have an urgent MRI brain and referral to the neurologist.

G/WEDS/PAN17 Paediatric multiple sclerosis: A clinical profile

R. Mittal, A. Curran. Royal Liverpool Children's Hospital, Alder hey, Liverpool, UK

Multiple sclerosis (MS) is a rare condition in children. But recent literature estimates that up to 15% may present before 18 years of age. MS in children commonly presents with systemic, polysymptomatic disease and can be difficult to differentiate from acute disseminated encephalomyelitis (ADEM). We present our experience of MS in a tertiary care centre over a period of ten years.

MethodsRetrospective case note review was carried out of all patients of multiple sclerosis over a period of ten years. Data were recorded on a structured proforma.

ResultsA total of 6 patients were managed. The age of presentation varied from 3 to 15 years. Male to female ratio was 1:2. Two patients presented with systemic symptoms and were initially diagnosed as ADEM and Miller Fischer syndrome respectively. Two presented with unilateral optic neuritis. The rest had other visual and pyramidal symptoms. Visual symptoms were present in all, either at onset or developing later. Two patients developed epilepsy. The oligoclonal bands were detected in 4 (67%) cases. The diagnosis was made by characteristic brain MRI findings in all. The pattern of disease was relapsing remitting type (RRMS) in all the children. Glatiramer acetate (copaxone) has been used with good result in these patients.

DiscussionMS is an uncommon disease in children. However it can present at any age including infancy. Two of our patients were 3 and 4 years old respectively. It diagnosis is difficult because presentation can be similar to ADEM or other localised or disseminated CNS inflammation. Children are noted to have higher relapse rate but lower disability compared to adults. Also cognitive disability is an important issue.

ConclusionPaediatric MS is an unfamiliar entity for paediatricians. The spectrum of presentation is very wide. The general paediatrician and the paediatric neurologist should be alert to the possibility of MS in children.

G/WEDS/PAN18 Health care professionals' perceptions of pain and pain assessment in children with severe to profound neurological impairments

A. Hunt1, S. Robertson2, A. Thompson3, K. Seers2. 1University of Central Lancashire, Preston, UK; 2Royal College of Nursing Institute, Oxford, UK; 3North Warwickshire PCT, Warwick, UK

Children with severe to profound neurological impairments may be unable to self‐report their pain. The Paediatric Pain Profile (PPP), which incorporates a 20‐item behaviour rating scale, has been developed specifically for this population of children and validated across a number of settings.

AimsThe aims of this ongoing study are to facilitate the implementation of the PPP as a parent‐held document across a number of primary care trusts and evaluate its acceptability and feasibility for parents and professionals involved in the child's care. The results from a pre‐implementation survey of professionals are reported.

MethodsPre‐implementation questionnaires and invitations to take part in the evaluation of the PPP were sent to parents of children with a severe neurological impairment unable to self‐report their pain. Prior to a visit to the parents by one of the health care team to set up the PPP with them, parents were visited by the study co‐ordinator who took informed consent, collected demographic data and the names of professionals involved in the child's care. These professionals were notified of the family's entrance to the study and asked if they would complete a questionnaire on their general perceptions of pain in this group of children.

ResultsParents of 56 children agreed to use the PPP and nominated 386 professionals, 29% of whom (n = 110) completed a pre‐implementation questionnaire. These included 36% doctors, 36% nurses and 9% physiotherapists, who had seen from one to 500 (median 30) children with severe physical and learning disability in the previous 12 months. Sixty‐seven percent professionals thought pain was ‘moderately' to a ‘great deal' a problem for this group of children. The proportion of the children who they thought had a significant pain problem varied from 0 to 100% (median 30%). There was a negative correlation between the number of children seen and the proportion believed to have pain (rs = −0.280, p<0.008). Muscle spasms (26%), constipation (26%) and gastro‐oesophageal reflux pain (26%) were reported to be the most common pain sources.

ConclusionsProfessional respondents recognise difficulty in assessing and treating pains in this group of children and are receptive to implementation of a pain assessment tool.

G/WEDS/PAN19 An alternative approach to managing abdominal pain and distension in children with complex needs

B. Piel, G. Debelle, L. Cuddeford. South Birmingham PCT, Birmingham, UK

We present a 7 year old girl (OM) with chromosome 22q deletion without the full phenotypic picture of DiGeorge Syndrome. Her early life was complicated by multiple significant apnoeic episodes and difficult to control epilepsy. She had a previous left nephrectomy for hydronephrosis and bilateral squints.

OM is gastrostomy fed via a Mickey button due to poor swallow and from the age of 2 had suffered from intermittent episodes of abdominal distension with severe pain. This had required frequent venting of her gastrostomy, which was forceful in nature. Her symptoms were further complicated by alternating constipation and diarrhoea. Despite extensive assessment and investigations by the paediatric gastroenterology team the cause for her symptoms has remained unclear. Symptomatic treatment in the form of regular analgesia and antispasmodics in association with treatment of constipation and diarrhoea was unsuccessful.

OM's ability to engage in normal childhood activities was compromised by her symptoms. As a consequence she was missing significant time from school.

A trial of use of the Farrell valve (enteral decompression valve) was embarked upon. The aim of this treatment was to provide constant decompression of the stomach and hence elimination of pain and discomfort associated with gas and fluid build up. Since the introduction of the Farrell valve there has been a remarkable increase in OM's school attendance and her parents report a significant reduction in episodes of pain and discomfort.

We feel that the Farrell valve provides another potential therapeutic treatment option for symptom control in children fed by gastrostomy with abdominal pain and distension not responding to conventional therapy.

G/WEDS/PAN20 A review of deaths of children with MPS Types I, II and III

D. Box, S. Browning. Martin House Hospice, Leeds, UK

AimsTo review the demographics, causes, symptoms and care around the deaths of children with Hurler, Hunter and Sanfilippo syndromes.

MethodsReview of the case notes of all the children who died with MPS types I, II and III under the care of a children's hospice over the last 20 years. Data looked at included: diagnosis, terminal symptoms and their management, cause of death, age of death, place of death. These data were analysed for any patterns within the MPS subtypes and the variability in the patterns of death over time.


Total number of deaths: 28. Deaths from Hurler syndrome: 5. Deaths from Hunter syndrome: 5. Deaths from Sanfilippo syndrome: 18. Terminal symptoms included apnoeas, agitation, presumed GI pain (possibly bowel spasm), respiratory distress, reduction in conscious level, peritonitis, respiratory infections, hydrocephalus, central respiratory failure, and sudden unexpected death.

The medical management of the terminal stage of the illness ranges from nothing to diamorphine, phenobarbital, hyoscine, cyclizine, midazolam, diazepam, haloperidol, oxygen and merbentyl.

Causes of death: pneumonia ‐ 11; central respiratory failure ‐ 3; peritonitis related to a gastrostomy ‐ 2; sudden unexpected death ‐ 2; hydrocephalus causing IVH ‐ 1; not documented (died at home or in hospital) – 2. Interestingly, the causes of death have changed over time with the increasing numbers of interventions offered to these children (gastrostomies, tracheostomies and neurosurgery). The commonest place of death was at home (12 children) with 10 children dying at the hospice and 6 dying at home. These proportions have stayed fairly constant over the 20 years.

ConclusionThe patterns of end‐of‐life symptoms and management are varied and changing with advances in medicine. Parents and professionals have a need for up‐to‐date knowledge of what to expect at the end of the child's life.

G/WEDS/PAN21 Paediatric Consultants' experiences of decision making leading up to and following a patient's death

A. Baverstock, F. Finlay. Community Child Health Dept, Bath, UK

AimsA study to establish current practice around the decision making process in withdrawing and withholding life‐sustaining treatment. In particular, factors that make these decisions difficult, strategies to help with prognostic uncertainty and experiences of dealing with conflict and differing opinions.

BackgroundWe know that dealing with the death of a child or baby can be a stressful experience. ‘Over time the resources of the individual to deal with dying infants and children may be depleted'.1 We often spend much time with families leading up to and following a child's death. Consultants have the added burden of discussing decision making with parents and staff during these often difficult times.

MethodsFollowing ethical approval, a self‐administered questionnaire was sent to 100 paediatric consultants within the South West region.

Results and discussionMost consultants had experienced differing opinions on whether to withhold or withdraw life sustaining treatment – ‘father wanted to let infant die, mother could not allow herself to say yes'; ‘often conflict – we don't all come to the same conclusion at the same time'. Consultants highlighted factors that made decision making difficult – ‘unclear prognosis'; ‘where there is no time to properly prepare parents for impending decision'. Some had developed strategies to deal with prognostic uncertainty – ‘seek advice from tertiary specialist team'; ‘be honest with the parents'; ‘discuss with the hospital ethics committee'. Respondents were asked if there was a patient/book/lecture or experience that has changed their practice – ‘hearing a talk given by a mother whose son had died – reiterated that many are fearful of death, have never seen a dead body, and don't know what to do. I learned that I need to discuss things in more detail in advance'. ‘Being on the receiving end – must remember that everyone is an individual and not just another patient.'

What can we learn from the experience of other consultants? ‘Caring for a dying child must be a shared professional experience…..the burden is too great for any one person…'2

ConclusionDecision making regarding withdrawing or withholding life sustaining treatment is rarely straightforward. Consultants need ongoing support to enable them to cope with the challenges involved in looking after dying children.

Ped Nursing199117 pp 103-105 Hospice care for children. 184–97

G/WEDS/PAN22 Junior doctors' self‐assessment of skills in paediatric palliative care: undergraduate training and clinical experience is not enough

L. Brook. Alder Hey Children's Hospital, Liverpool, UK

AimsTo evaluate junior doctors self‐perceived skills and confidence in paediatric palliative care and to correlate this with clinical experience and previous palliative care education and training.

MethodsOver the period August 2003–July 2006 junior doctors attending paediatric palliative care training as part of the regional postgraduate training programme were invited to fill in an anonymous self‐assessment questionnaire prior to the teaching session. The previously validated questionnaire requested information on trainees' clinical experience including caring for a dying child and previous palliative care education and training. Respondents were then invited to rate their confidence in undertaking a number of tasks related to paediatric palliative care using a 10 cm linear analogue scale (LAS).

Results52 junior doctors completed questionnaires over the 3‐year period. Overall 50% had looked after a dying child in the last 6 months but 12 doctors had never looked after a dying child. Of those who had never looked after a dying child all were relatively junior; only one had MRCPCH part 1 and none part 2. Respondents with MRCPCH part 1 or 2 (n = 34) reported looking after a dying child a median of 2.5 times a year (range 0–7). Of these, 14 reported previous palliative care training but only 5 at postgraduate level. Respondents reported their palliative care training moderately useful (median 6.5 cm). However trainees reported low levels of confidence undertaking key tasks in end‐of‐life care including pain and symptom management, discussing resuscitation and breaking bad news (mean 3.4 cm, standard deviation 2.8 cm). There was no correlation between reported palliative care training, recent care of a dying child or more frequent care of a dying child with perceived skills in breaking bad news, pain and symptom management or discussing resuscitation.

ConclusionOn average trainee paediatricians are likely to care for a dying child at least twice a year. Undergraduate training and clinical experience alone appear insufficient to ensure that paediatricians have the necessary skills for end‐of‐life care and formal training is needed.

G/WEDS/PAN23 Children with life limiting and life threatening disorders: The impact on fathers

N. Eaton1, A. Emond2, M. Lewis1, A. Berenger1, A. O'Brien1. 1University of the West of England Bristol, Bristol, UK; 2University of Bristol, Bristol, UK

A diagnosis in a child of a life limiting or life threatening (LL) condition has an impact on each individual member of the family. Fathers particularly are a neglected group in the research literature on parental stress and coping.

AimsA preliminary study to explore fathers' responses, coping styles and modes of adaptation, identify any unmet need and develop a tool for a future national study on fathers needs.

MethodsSemi structured interviews were conducted with 13 fathers. All the fathers had children receiving care and support from a local children's palliative care service. The Parenting Stress Index (PSI) and the Impact on Family Scale (IFS) were completed during the interview.

ResultsThe children's ages ranged from 3 to 20 years with a mean age of 9 years. There were 8 girls and 5 boys with a range of diagnoses. Most fathers were aged 35–44 years (n = 10). The interviews generated the following consistent themes: limitations on their contribution to care; impact of uncertainty of life expectancy; relinquishing responsibility; childs appearance; social isolation; financial constraints; employment issues; attendance at medical consultations. Themes were substantiated by the PSI and IF scores which indicate that this group of fathers are under significant stress. PSI scores (out of 180): range 58–117, mean 87.8 (>90 indicates a parent is experiencing a clinically significant level of stress). Seven of the fathers scored above 90. IFS (out of 96): range 43–77, mean 57.75. The higher the score the more the impact on the family. These results indicate that having a child with a LL condition in the family has a significant impact on the father in a number of domains. There is little normative data for fathers using these tools; however, for mothers, higher scores on the IFS have been shown to be associated with significant psychiatric symptoms.

ConclusionsFathers are a group often receiving little attention in the research literature. The part they play in supporting the sick child, and the rest of the family, and their contribution to their childs' care needs to be recognised and supported.

G/WEDS/PAN24 An exploratory qualitative study of milestones in the trajectory of life‐limiting illness in childhood

R. Hain, S. Simpson, F. Wood, E. Barnes. Cardiff University, Cardiff, UK

AimsTo describe the ‘illness journey' of young people with life‐limiting conditions and their families. Streamline models used for care provision funding in those families.

MethodQualitative semi‐structured interviews were carried out with family members of children with life‐limiting conditions (LLC). Focus groups were also carried out with professionals who provide social, educational, or medical care to children with LLC and their families. Participants were presented with four vignettes describing different conditions and asked to discuss events they considered to be milestones for each condition and their timing. The data were transcribed and thematically analysed using NUD*IST qualitative analysis software.


Three main themes emerged: social – isolation, lack of privacy and identity change; emotional – response to diagnosis, hiding emotions and exhaustion; practical issues – home modifications and changes in living conditions and daily routine.

ConclusionChildren with LLCs and their families require holistic care and multi‐agency involvement throughout their illness. These data provide the basis for developing a prospective mathematical model for funding.

Articles from Archives of Disease in Childhood are provided here courtesy of BMJ Publishing Group