Article Text

Download PDFPDF

Child protection and radiology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


L.J. Fender, N. Broderick, J. Somers, T. Jaspan, N. McConchie, K.H. Halliday.

Queen’s Medical Centre, Nottingham

Aims: To determine the detection rate of occult fractures in patients undergoing a skeletal survey for suspected NAI, the value of routine neurological imaging and the impact of follow up radiographs.

Materials and Methods: Skeletal Surveys of 100 consecutive patients with suspected NAI were retrospectively reviewed. The routine follow up chest X-ray performed at 10 to 14 days was reviewed to assess the detection of further rib fractures not evident on the initial film and to determine if further information to aid the dating of rib fractures became apparent. The findings of neurological imaging were correlated with clinical indicators of intracranial injury.

Results: Clinically unsuspected fractures were detected in 34 patients. Of the 100 patients follow up chest radiographs were performed in 29 patients. In 3 cases (14%) further useful information about dating was obtained and in 1 case rib fractures not visible on the initial film were revealed. 31 pts had an intracranial abnormality demonstrated on cranial Computed Tomography. 23% (7 patients) of these had no neurological symptoms or signs. Of 16 patients with no skeletal fractures or clinical features suggesting neurological injury, 1 had a subdural haematomata of differing ages highly suggestive of abuse.

Conclusions: Occult injuries were detected in 34%. Routine cranial imaging is advocated to detect occult intracranial injuries, which maybe the only radiological manifestation of Non Accidental Injury. The follow up chest radiograph did not alter the diagnosis in any patient but was helpful in 14% of cases.


P.M. Barnes, J.R. Sibert, C.A. Norton, A.M. Kemp.

Department of Community Child Health, Llandough Hospital, Cardiff

Introduction: Internal abdominal injury is a recognised consequence of physical child abuse, but there is little published literature concerning this condition. The paediatrician may face difficulties when analysing such cases in a child protection investigation. This study aims to determine the incidence, pattern and outcome of internal abdominal injury due to child abuse as compared with accidental injury.

Methods: Cases of internal abdominal injury due to abuse in children aged 0–14 years were collected over a one year period through notification to the BPSU. Information concerning the demographic details of each case, type of injury, presentation, management and outcome details were collected from each respondent. Information concerning accidental internal abdominal injury in childhood was obtained through the ‘Major Trauma Outcome Study’.

Results: Fourteen cases were notified (mean age 4 years; 50% male), including five fatalities (36%). Preceding child protection concerns were evident in eight cases (57%). Five cases (36%) had no external bruising to the abdominal wall despite internal injury. Although nine cases (64%) had suffered perforation of small bowel/ duodenum, this pattern of injury was seen in only 8% of internal abdominal injuries following road traffic accidents (and for accidental falls >2 metres, in only 4%).

Conclusions: Internal abdominal injury is an uncommon but serious form of child abuse with a high risk of death. External abdominal bruising may be absent. The pattern of small bowel/duodenal injury in cases is not commonly seen following major accidents, and when encountered should raise suspicion of abuse. Case ascertainment has continued for a second year, and additional information will be complete and available in March 2003.


S.A. Maguire, A.M. Kemp, M.M. Mann, J.R. Sibert, Welsh Child Protection Systematic Review Group.

University of Wales College of Medicine, Cardiff, Wales, UK

Aims: A frequent query in the assessment of child protection cases is the age of a bruise. Bruises are often defined as “fresh”, “old”, or “recent”, however we question the scientific evidence behind this classification. We conducted a comprehensive systematic review to establish if it is possible to age bruises accurately in clinical practise. This is the first in a series of systematic reviews of the medical aspects of physical abuse.

Methods: We conducted an all language literature search of original articles and conference abstracts for the period 1970–2002. We used the ASSIA 1987–, Caredata (1980–2002), Medline (1966–2002), Child Data (1958–2002), CINAHL (1982–2002), Embase (1980–2002), PsychINFO (1887–2002), Social Science Citation Index (1981–2002), and TRIP databases. Total records found, 6117. Browsing titles and abstracts the final total was 142 full text papers and 6 papers found through references, 12 required translation. 15 reviewers carried out 2 independent reviews, and if necessary a third review, using standardised criteria for study definition, and specially devised Data Extraction, and Critical appraisal forms.

Results: Only three papers met our criteria and were robust enough to be included: one case control, one case series and one cross-sectional study. Based on these, out of 148 reviewed, the only clear finding is yellow bruising does not occur <24 hours, but it’s absence does not imply that the bruise is less than 24 hours, as it may never appear in a bruise. Red colour does not occur after one week, but may never be present.

Discussion: This first Systematic Review of the ageing of bruises in children with the naked eye shows that, there is no sound scientific evidence to support any such assessment. Yellow bruising is not seen before 24 hours, a finding which is weakened by the paucity and quality of evidence behind it.


M. Govindshenoy, N. Spencer.

Gulson Hospital, Coventry and University of Warwick

Aim: To systematically review all the observational studies from the period 1966 to Jan 2002, which look at the relationship between pre- existing childhood disability and the incidence of child abuse and neglect.

Design: Systematic review of the literature with disability as the exposure and abuse and neglect as the outcome.

Data sources: Medline, Embase, Cinahl, Cochrane Library, National Research Register, Social Sciences Database, and PsychInfo.

Included studies: Observational studies that reported the relationship between pre-existing childhood disability and child abuse and neglect. A quality protocol with a scoring system was applied to the studies. Odds ratios were calculated where possible.

Results: 32 studies met the criteria for inclusion in the review. 14 studies were found to satisfy quality criteria for inclusion in the final analysis. There were 7 case-controlled studies, 2 cohorts and 5 cross sectional surveys. Statistical analysis was not possible for all the included studies, thus making meta-analysis difficult. Six studies showed odds ratios ranging from 1.24 to 8.5 suggesting that a disabled child is at higher risk of abuse. Higher risk may be associated with psychiatric illnesses, learning disability and speech and language problems. Specific disabilities may induce specific forms of abuse. The majority of studies were based on a retrospective review of abused or disabled children. We found no population based studies.

Conclusions: There is some evidence to suggest that the disabled child is more likely to be abused and abused children are more likely to have pre-existing disabilities. However, the highly selected population on which the majority of the studies were based significantly weakens these conclusions.


J. Mok, B. Daniel, on behalf of the Review Team.

Scottish Executive Social Work Services Inspectorate

Introduction and Aims: Following an inquiry into the death of a child in Scotland, a multi-agency team was established by the Scottish Executive to review child protection practices. Amongst its tasks, an audit was performed into the practice of medical, nursing, social work, police, education, Children’s Reporters and voluntary sector staff, to assess how well the agencies worked separately and together to meet children’s needs. The principal focus was on the outcome for the child.

Methods: Agencies were asked to identify children in whom there were concerns about abuse and neglect in the week beginning 3.9.01. From a sample of 5045, 188 children were selected for detailed evaluation of case notes; and in 103, structured interviews were also conducted with 438 key professionals involved in the care. In addition, the Team met with 11 children and parent(s) of a further 17 children. The fieldwork was conducted between January to March 2002.

Results: The sample consisted of 100 girls and 88 boys, ranging in age from 1 month to 17 years. The majority (85.6%) were white. Only 7 children referred themselves. Although 95 (50.5%) were on the Child Protection Register, other concerns documented by professionals included neglect (85; 45.2%), risk of physical injury (43; 22.9%), parental substance misuse (76; 40.4%) and domestic abuse (56; 29.8%). Following intervention by the agencies, there was insufficient information to assess outcome in 9 children. In the remaining 179, 40 (22.3%) were either not protected or their needs not met; 62 (34.6%) had their needs partially met while 77 (43%) were judged to have been adequately protected.

Conclusions: This audit showed that in Scotland, many children are currently living in conditions and under threats that are not tolerable in a civilised society. Children and families do not always get the help they need when they need it. There was duplication of effort and energies were diverted into meeting system requirements rather than the real needs of children. Despite state intervention, a quarter of the children in this sample remained unprotected or their needs were not met.


P.D. Sidebotham, J. Heron, ALSPAC Study Team.

University of Bristol

Objectives: To analyse the multiple factors affecting risk of abuse in young children within a comprehensive theoretical framework.

Setting and design: The Avon Longitudinal Study of Parents and Children (ALSPAC) is a cohort study investigating factors influencing the health and development of 14 000 children born in Avon in 1991–2. Data on the environment and health of these children have been collected at regular intervals throughout pregnancy and childhood. Children placed on the child protection register have been compared to the overall cohort using logistic regression analysis to explore risk factors within the parents’ backgrounds, the family and home environment and the wider socio-economic context.

Results: 115 cohort children were placed on the child protection register by age 6. Significant risk factors included various indicators of deprivation and poor social networks, young parents with lower educational achievement; and a parental history of psychiatric illness. Significant relationships were found between low birthweight, unintended pregnancies, poor health and developmental problems in infancy, and subsequent maltreatment.

Discussion: This study, one of the largest in this field, and the first of its kind in the UK, provides empirical data on a range of factors related to the risk of child abuse. These factors are combined in a comprehensive theoretical framework which allows an understanding of their relative importance and potential interactions.


N.K.S. Thalange, R.J. Reading.

Jenny Lind Children’s Department, Norfolk & Norwich University Hospital, Norwich

Aim: To determine health care expenditure and resource use on children.

Methods: Hospital and Community Health Services (HCHS) activity was determined from routine Health Episode Statistics supplied to Norfolk Health Authority for the calendar year 1999. Financial data for HCHS was derived from the Minimum Data-Set and contracts with tertiary providers, based on 1999–2000 financial year. Expenditure on Primary Care prescribing and General Medical Services was estimated by applying national Age-Cost Curves. Social Services expenditure and activity for the financial year 1998–9 was supplied by Norfolk Social Services. The mid-year ONS population estimate for Norfolk children was used to calculate activity rates.

Results: Total health expenditure amounted to £34.1m/100 000 child population. Acute services cost £13.9m/100 000, of which neonatology was £2.6m (10.8% of live births admitted) and general paediatrics £3.2m (5,341 admissions/100 000). There were 38 226 outpatient attendances (35% first attendances), and 10 493 hospital admissions (Emergency 5 588), of which general paediatrics was 5341 (50.9%) /100 000 population. 4,741/100 000 children underwent an operative procedure (ENT 27.2%, Orthopaedics 18.2%). Community services cost £7.9m/100 000 of which nursing and health-visiting costs were 53.3%. Learning disability and child and adolescent mental health services cost £3.2m/100 000. Primary care expenditure was estimated at £9.1m/100 000, of which prescribing was 29.7%. Social services expenditure was £13.6m/100 000. 0.6% of the Norfolk child population was referred to social services in 1999–9, of which 32% were referred with child protection concerns.

Conclusions: There is a tension between high-volume, low-cost services such as general and community paediatrics, and high-cost, low-volume work such as neonatology and services for looked-after children, which often command a higher public and political profile. Public policy-makers and commissioners must take account of the needs of all children in deciding priorities and targeting resources.


S. Karandikar, L. Coles, A.M. Kemp.

Department of Child Health and University of Wales College of Medicine

Introduction: Shaken baby syndrome is the commonest cause of non-accidental head injury (NAHI) in infants. Patients often have subdural haemorrhages (SDH), retinal haemorrhages, extra cranial injuries and intracerebral damage. Data on long-term outcome of the survivors are not widely reported.

Aims: Describe the physical and educational outcomes of children <2years diagnosed with NAHI and formulate guidelines on follow-up.

Methods: We identified 65 children <2 years with NAHI and SDH diagnosed on CT/MRI/ post mortem. Data was extracted from the hospital and community notes. The Kings Outcome Scale for Childhood Head Injury (KOSCHI) was used.

Results: 16 children died (24.6%), 13 within 6 weeks and 3, 5 years after the insult. Of the 43 children followed up (follow up time 21.7–103 months.), 25 had good outcome (KOSCHI 5), 6 moderately disabled (KOSCHI 4), 11 severely disabled (KOSCHI 3), while 1 was vegetative (KOSCHI 2). Problems identified included cerebral palsy(14), ongoing seizures(5), visual problems(11), speech and language problems(17), and behavioural problems (13).

Of the 27 children of school age, 10 children had a statement of educational needs.

Conclusion: Young children who survive NAHI can have significant physical, cognitive and behavioural problems. Regular follow-up by a multidisciplinary team led by a community paediatrician involving therapists, school health personnel, and ophthalmologic services until the child is established in school may facilitate early identification of emerging problems and optimise management. Care must be taken not to discharge children with good outcome as behavioural and cognitive problems can emerge with time.