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): A68–A70.
PMCID: PMC2066123


General paediatrics

G/THUR/GEN1 Chronic fatigue syndrome in primary school aged children

S. Davies1, J. Griffin1, E. Crawley2. 1North Bristol NHS Trust, Bristol, UK; 2Bath Royal United Hospital, Bath, UK

AimTo describe the presentation and outcome of a cohort of children diagnosed with chronic fatigue syndrome (CFS/ME) while in primary school.

MethodsData were collected for patients diagnosed with CF/ME under the age of 12 by the Bath/Bristol CFS/ME service between September 2004 and September 2006. Data collected at time zero and 12 months were analysed and included the Chalder fatigue scale, symptom checklist, Pain VAS, SF36, anxiety and depression inventory and school attendance. The notes were reviewed to check whether alternative diagnosis had been made at follow‐up.

ResultsTwenty six children under 12 were diagnosed as having CFS/ME in this period which is 14% of the total number of children seen. The age range was 3–12 years (mean 9 years). 19/26 children were girls (73%). At diagnosis 2 children were under 6 (8%), 7 children between 5–8 years (27%) and 17 children 9–12 years (65%). The age range at onset of disease was from 2 years 4 months to 12 year (mean 7 year 8 months). At disease onset 6 children were under 6 (23%). The length of time to diagnosis ranged between 3–60 months (median 11 m, mean 8 m). 25/26 children fulfilled the CDC and RCPCH criteria for CFS/ME. 25/26 children had excessive fatigue according to the Chalder fatigue scale. After fatigue the most common feature was unrefreshing sleep which was found in 20/20 (100%) of children. The second and third most common feature were headache in 19/22 (86%) of children and memory impairment in 18/22 (82%). The average number of symptoms at diagnosis was 10 (range 4–14) out of a maximum total of 17. 13/23 (56%) had high levels of pain (>0.5 on VAS). 12/26 (46%) had significant levels of anxiety (Spence anxiety scale). Disability was high with none of the children attending full‐time school. Average school attendance was just over 40% at diagnosis. Outcome data at one year were available in 9/26 children; 6/9 showed improvement. One child was shown to have coeliac disease when screening bloods were repeated at one year.

ConclusionWe believe this is the first study to describe a cohort of children with CFS/ME presenting under the age of 12. The clinical pattern and outcome is similar to that described in older children. Disability in this age group was high with considerable amount of school missed, which may suggest that we are seeing the severe end of the spectrum.

G/THUR/GEN2 What happens to children with severe chronic fatigue syndrome/ME?

J. Griffin, E. Crawley. University of Bristol, Bristol, UK

Aims(1) Describe all children with severe chronic fatigue syndrome (CFS/ME) in a defined geographical area who accessed a CFS/ME service. (2) Analyse outcome for children accessing the domiciliary service.

IntroductionIn a specified area, all children who are housebound with CFS/ME are offered domiciliary support. At the first assessment, 12 months and discharge children complete: Chalder fatigue scale, pain VAS, SF‐36, anxiety and depression measures, hours of home tutoring, symptom checklist, etc. There is no evidence on what is effective treatment for this patient group. This is the first follow‐up study of severely affected children with CFS/ME in the UK.

MethodCase notes and follow‐up data for all children who received a domiciliary visit between November 2004 and August 2006 were reviewed. Data were analysed with a primary outcome of discharge from the service.

ResultsTwenty patients (14 female) received domiciliary visits. This is 10% of all children seen by the service within the same time period. Age range 4–19 (mean 13.95) years at initial assessment with mean age of onset 11 years (range 10 months to 16 years). All but two met the CDC criteria for diagnosis. The most frequent symptoms after fatigue was postexertional malaise (93%) followed by unrefreshing sleep (88%). 80% also had joint and/or muscle pain, headaches and abdominal pain. Weight changes and hypersensitivities were the least common symptoms (50%). Only 10 (50%) had uncomplicated CFS/ME. Two had pervasive refusal syndrome and 8 (40%) had comorbid anxiety or depression. Absence of significant comorbidities and time from onset to assessment improved outcome (see table).

Table thumbnail
Abstract G/THUR/GEN2

ConclusionsOur data are consistent with patient surveys suggesting 10% of children with CFS/ME are severely affected and suggests that outcome may be considerably better then previously thought. Prognosis appears to depend on the presence or absence of comorbid conditions and on the time to assessment. It seems likely that delayed diagnosis may increase the presence of comorbidity. We strongly recommend that domiciliary services are provided for children with severe CFS/ME as soon as possible after diagnosis.

G/THUR/GEN3 The Paediatric Early Warning Tool: can its use predict and shorten time of referral to paediatric intensive care?

T. Ralph, B. Bradbrook, S. Ganguly, A. Mayer. Sheffield Children's NHS Foundation NHS Trust, Sheffield, UK

IntroductionEarly warning tools are intended to highlight a clinical deterioration in children using clearly defined physiological criteria so triggering medical review and instigation of a management pathway. The Sheffield Paediatric Early Warning Tool (SPEWT) was adapted from the Bristol Paediatric Early Warning Tool, and introduced in November 2005. To monitor its effectiveness at identifying potentially critically ill children, an audit was undertaken of all PICU emergency admissions from within the Trust over a 5‐month period from December 2005 to April 2006 inclusive.

AimThe aim of the study was to examine whether children admitted to PICU not using the PEW referral process would have triggered the tool in the previous 24 h.

MethodAfter excluding PICU admissions from theatre and external hospitals, 61 patients' notes were reviewed. Data were collected about the referral and admission process, clinical and physiological parameters in the previous 24 h and demographic variables. Data are expressed as whole numbers, median and interquartile ranges.

ResultsFifty six (92%) admissions to ICU were without triggering a PEW score, of these only 8% (5/61) would not have triggered the PEW in the previous 24 h on review. The time to ICU admission was shorter in those children who were admitted following a PEW. Six (10%) patients were admitted to ICU following cardiac arrest all of these would have triggered PEW in the previous 24 h. Of those admitted to ICU there were 2 deaths; neither of these would have triggered the PEW prior to ward review.

Table thumbnail
Abstract G/THUR/GEN3

ConclusionPEW is an effective tool for identifying children at risk of acute clinical deterioration and so alerting ward‐based medical teams to make appropriate referral for ICU. The incorporation into clinical practice reduces ICU referral and admission times. Further evaluation of the PEW score is required especially with regard to early recognition of critically ill children and preventing cardiopulmonary arrest.

G/THUR/GEN4 Nurse‐led inpatient management of infants with bronchiolitis during the seasonal epidemics 2004–5 and 2005–6

S. Pate, R. Hague, N. Gibson. Royal Hospital for Sick Children, Yorkhill, Glasgow, UK

AimsTo study the impact of introduction of a nurse‐led care pathway for infants with bronchiolitis.

MethodsInfants with a clinical diagnosis of bronchiolitis requiring admission were reviewed once by the receiving medical team, and thence referred to the nurse‐led pathway, unless they had: history of apnoea; uncertainty regarding diagnosis; complications of prematurity or any other significant comorbidity. Nursing intervention included observation using a respiratory score; adjustment of supplemental oxygen; nutritional assessment and fluid balance; determination of appropriate feeding and fluid regimen: all within set parameters with triggers for requesting medical review. In 2004–5, medical review was required before discharge. In 2005–6, discharge was carried out by trained senior nurses. The criteria for discharge were set at the initial medical review.

ResultsOver the 2 seasons, 251 infants were referred to the pathway. Mean age was 5.9 months (range 8 days–24 months). Mean length of stay was 2.7 days. 191 had a definitive diagnosis of RSV. 15 infants were transferred back to medical care. 5 were subsequently admitted to intensive care. 10 developed complicated disease or other medical problems. Readmission rate was 5/143 in the first year, and 3/93 in the second year. 754 routine medical reviews (3.2/child) were avoided, with a high degree of parental satisfaction.

ConclusionA nurse‐led care pathway is a safe and effective means of managing infants with bronchiolitis which has a significant impact on the demands on medical time during the peak winter period.

G/THUR/GEN5 Progression from first symptom to diagnosis in childhood brain tumours: a multicentre study

S. Wilne1, J. Collier1, C. Kennedy2, A. Jenkins3, J. Grout4, S. Mackie4, K. Koller1, R. Grundy1, D. Walker1. 1University of Nottingham, Nottingham, UK; 2University of Southampton, Southampton, UK; 3Sheffield Children's Hospital, Sheffield, UK; 4Southampton General Hospital, Southampton, UK

AimsTo investigate symptom and sign progression from symptom onset to diagnosis in children with brain tumours and to determine predictors of symptom interval (SI).

MethodsPatients newly diagnosed with a brain tumour between January 2004 and March 2006 were recruited from four regional paediatric neuro‐oncology centres. The relation between initial symptoms or signs and SI was explored with Cox regression analysis. For symptoms and signs with an unknown date of onset Fisher's exact test was used to explore the relation between a long (greater than median) or short (less than or equal to median) SI.

Results139 children were recruited. At symptom onset children had a median of one symptom or sign (range 1–8); by diagnosis this had increased to 6 (range 1–16). 76 children had a single symptom or sign at symptom onset; by diagnosis only three children had a single symptom or sign and no child had only headache or vomiting. By diagnosis 95% of children had symptoms and or signs in one or more of the following categories: headache, nausea and vomiting, motor system abnormalities and visual system abnormalities. The median symptom interval was 3.3 months (range 0 days to 6.9 years). A shorter symptom interval was associated with initial presentation with nausea and vomiting (p = 0.003), abnormal gait (p = 0.001), coordination difficulties (p = 0.006), focal motor weakness (p = 0.002) and any motor system abnormality (p = 0.001). A prolonged SI was associated with initial presentation with head tilt (p = 0.006) and cranial nerve palsies (0.025) and with reduced visual acuity (p = 0.028) and endocrine and growth abnormalities (p = 0.018) occurring at any time.

ConclusionsChildren with brain tumours develop multiple symptoms and signs. Children with symptoms and signs that may result from a CNS tumour should undergo assessment of growth (including head circumference and pubertal staging) motor and visual systems. CNS imaging should be considered in any child in whom these assessments reveal abnormalities as well as in those with focal seizures or obvious symptoms of raised intracranial pressure.

G/THUR/GEN6 The pathways project: developing guidelines to aid rapid diagnosis of childhood brain tumours

S. Wilne1, J. Collier1, C. Kennedy2, K. Koller1, A. Jenkins3, J. Grout4, S. Mackie4, R. Grundy1, D. Walker1. 1University of Nottingham, Nottingham, UK; 2University of Southampton, Southampton, UK; 3Sheffield Children's Hospital, Sheffield, UK; 4Southampton General Hospital, Southampton, UK

AimsUK children with brain tumours experience a long symptom interval in comparison with both other malignancies and other national groups. Concern regarding this led to the Pathways project to devise enhanced guidance to facilitate rapid diagnosis of paediatric brain tumours.

MethodsA systematic review and meta‐analysis of published data on paediatric brain tumour presentation and analysis of presentation, referral pathways and symptom interval for children newly diagnosed with a brain tumour at four paediatric neuro‐oncology centres was undertaken. The results informed a professional consensus process undertaken with the objectives of producing diagnostic and referral guidelines suitable for professional training, clinical decision support and public dissemination.

Results1417 patients were included in the meta‐analysis. 56 symptoms and signs were identified at diagnosis, ranked by frequency and clustered according to patient age and anatomical location. For all brain tumours the most frequent symptoms and signs were: headache (33%), vomiting (32%), abnormalities of gait and coordination (27%), papilloedema (13%) and seizures (13%). The 139 recruited patients had a median of 1.0 symptom or sign at symptom onset and 6.0 by diagnosis. The most frequent symptoms and signs at diagnosis were: visual system abnormalities (70%), motor system abnormalities (67%), nausea and/or vomiting (63%), headache (58%) and cranial nerve palsies (54%). 95% of children had one or more of the following: headache, nausea and vomiting, motor system abnormalities and visual system abnormalities. A multidisciplinary team devised 77 statements detailing the presentation of paediatric brain tumours, assessment of a child who may have a brain tumour, and indications for CNS imaging. Statements were reviewed by three rounds of a Delphi panel comprising 150 healthcare professionals. 64 statements achieved consensus and form the guideline.

ConclusionsWe have developed an evidence‐based, professionally reviewed guideline to facilitate the diagnosis of childhood brain tumours.

G/THUR/GEN7 Do adolescent inpatient wards make a difference? Findings from a national young patient survey

R. Viner. Institute of Child Health, University College London, London, UK

AimsThere is little evidence to support the effectiveness of adolescent inpatient wards. These analyses test the hypotheses that nursing young people in adolescent wards improves quality of care outcomes and patient satisfaction compared with children's or adult wards.

MethodsSecondary analyses of the English national Young Patient Survey 2004, weighted to take account of variations in hospital size and response rate. Participants: 8855 subjects aged 12–17 years from 150 acute or specialist Trusts. Measures: Ward type (adolescent, children's or adult) and quality of care indicators: rating by young people of overall care, respect, safety, confidentiality, communication, team working, noise, leisure facilities. Logistic regression models were adjusted for sex, disability and previous hospital admissions.

Results12% of 12–14‐year‐olds and 18% of 15–17‐year‐olds were nursed on an adolescent ward, 0.4% of 12–15‐year‐olds and 16% of 15–17‐year‐olds on an adult ward, with the remainder on a children's ward. Compared with being on an adolescent ward, 15–17‐year‐olds were less likely to report excellent overall care on an adult ward (OR 0.6; 95% CI 0.5 to 0.7; p<0.0001) and less likely to report feeling secure, having confidentiality maintained, feeling treated with respect, confidence in staff, appropriate information transmission, appropriate involvement in own care and appropriate leisure facilities (all p<0.001). Compared with being on an adolescent ward, 12–14‐year‐olds were less likely to report excellent overall care on an children's ward (OR 0.8; 95% CI 0.6 to 1.0; p<0.05) and less likely to report feeling involved in their own care (p = 0.03).

ConclusionsDedicated adolescent inpatient wards improve quality of care for young people compared with children's or adult wards, particularly for older adolescents. Despite this, few young people in the UK are currently managed on adolescent wards. These data support the further development of adolescent wards in larger general hospitals and children's hospitals.

G/THUR/GEN8 Trends in primary care drug prescribing in children and adolescents: changing epidemiology or evolving practice?

J. Shetty1, S. Greene1, M. McGilchrist1, M. Kirkpatrick2. 1University of Dundee, Dundee, UK; 2NHS Tayside, Dundee, UK

BackgroundLimited data exist on primary care paediatric prescribing practice for the UK. European studies demonstrate that 50–60% of all children receive a prescribed drug each year. No longitudinal data are reported on overall paediatric prescribing trends.

Aims and SettingTo measure the primary care prescription rate for individual drugs for children and adolescents in the Tayside Region of Scotland (<16 years age; population c 70 000) over a 10‐year period. To compare trends in prescribing rates for the 10 most commonly prescribed drugs categorised by BNF classification and stratified by age.

MethodsData on all prescriptions “encashed” at any community pharmacy in Tayside Region have been recorded onto a database since 1993 (MEMO). This unique database documents drug name, date of encashment and specific patient identifier. Paediatric prescriptions were divided into four age bands (0–1, 2–4, 5–10, 11–16 years). Tayside population numbers were extracted from the General Register Office for Scotland.

ResultsA total of 1 090 985 prescriptions were collated over 54 months in three epochs (1996/7, 2000/1 and 2005). The proportion of children receiving a drug prescription has risen from 69% in 1996 to 88% in 2005, a 19% rise over 10 years. Over a third of prescribed drugs in 2005 were for children >10 years of age. Annualised prescription rates over the 10‐year period demonstrate a marked upward trend in the older child, with no change in the under 5s. The five most commonly prescribed drugs were amoxicillin, inhaled beta agonists, topical steroid creams, inhaled steroids and analgesics. Amoxicillin prescription fell by 37% across all age bands over 10 years, whereas skin emollient and topical steroids increased by 26%. Antihistamine prescription in children aged 11–16 years has increased by 50%.

ConclusionsAlmost 90% of children receive drug prescriptions from primary care over a single year. Over the last 10 years the increase has been exclusively in the older child. However, for individual drugs trends vary both up and down, across all ages. These data reflect either changing patterns of disease or changes in prescribing practice by primary care physicians.

G/THUR/GEN9 Surveillance of adverse drug reactions in Scottish children: a pilot study

R. Lynn2, J. Ellinghaus3, K. Riding4, N. McIntosh1. 1University of Edinburgh, Edinburgh, UK; 2Institute of Child Health, London, UK; 3Royal College of Paediatrics & Child Health, London, UK; 4Lothian Universities NHS Trust, Lothian, UK

BackgroundMedicines for children are frequently provided off‐label or off‐licence. Because of physiological changes in childhood, it is likely that children have more than average adverse drug reactions (ADRs) but concerns that MHRA yellow card system underreports ADRs in children have prompted alternative reporting schemes to be considered.

Aims(1) To investigate the feasibility of establishing an active surveillance system for ADRs in children; (2) to compare the effectiveness of electronic and postal reporting schemes.

MethodA green card modelled on the BPSU orange card was sent by post each month for 14 months from January 2006 to 150 paediatricians and 50 pharmacists. The card asked participants to report ADRs occurring with four specific drugs (ibuprofen; ceftriaxone; fluticasone; sodium valporate) or with any sedative prescribed for children. In addition participants were encouraged to identify any other drug they believed had caused an ADR. Reported ADRs had to be severe enough to precipitate admission to hospital, occur in hospital or because of an outpatient medicine prescription. Non‐respondents were re‐contacted after 3 months. Respondents reporting ADR were sent a proforma seeking minimal identifier information to discount duplication and obtain details of the type of ADR. After 8 months' surveillance the respondents were randomly assigned to one of two groups. Half continued to receive the card mailing while the other half received an email with an electronic card attached for a second 6‐month period. A research nurse visited hospitals to advertise the scheme; posters were placed on wards and in the pharmacy.

Results55% of cards were initially returned rising to 84% by the ninth month, 88% by paediatricians and 81% by pharmacists. Mean monthly response rates with electronic reporting were 66%. Data on 66 ADRs in 45 children have been supplied. Fever, vomiting or diarrhoea was the commonest reported ADR. More severe reactions, for example renal failure, thrombocytopenia, cardiac and respiratory arrest were also reported. There were 2 reports of ibuprofen, sodium valporate, ceftriaxone and 1 of fluticasone and sedatives. There were 36 “Other” drug reports, for example Linezoid, vancomycin, and ciprofloxacin.

ConclusionsClinicians accepted both postal and electronic reporting. Transferred nationally we could expect 1500 reports per annum. (DH Funded).

G/THUR/GEN10 Working harder, working faster: general paediatric outpatient referrals 1988–2006

E. Porter, P. Todd, C. N. Brolchain. Wirral NHS Trust, Wirral, UK

AimsTo re‐audit outpatient referrals and outcomes to the general paediatric service in a UK district general hospital in 2006. To compare these with 1988.

MethodsA prospective analysis of general paediatric outpatient referrals was performed from 1 Feb–31 May 2006. The same four‐month period had been analysed in 1988 in the same DGH using the same methodology. Due to the geographical characteristics of the area studied, this is likely to represent population‐based information. In the absence of detailed official statistics, these two studies allow a unique comparison of changes over time in general paediatric outpatients in the same locality. Community patients were not included in the original study and are therefore excluded. We compared referrals, management and outcomes for general paediatric patients.

ResultsReferrals for asthma, the commonest condition referred in 1988, had decreased from 48 to 18 (p = 0.02). Heart murmurs, the next most common condition in 1988, had also dropped from 45 to 20 (p = 0.12). Recurrent abdominal pain and headache had not changed much (29 in 1988 v 25 in 2006).

Table thumbnail
Abstract G/THUR/GEN10

ConclusionsGeneral paediatric outpatient clinics have changed significantly since 1988 with a 40% increase in referrals, better attendance, less need for investigations (as more were investigated by GPs), earlier treatment, more discharges and more reliance on a multidisciplinary team (eg more heart murmurs were referred to a paediatric cardiologist; fewer children saw a consultant). Case mix had also changed and details of case mix presenting in 2006 will be described separately.

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