While reading an orthopaedic text you find a table that states rib fractures are highly specific for non-accidental injury in children. No papers are referenced and you wonder what evidence exists to support this statement.

Structured clinical question

In children undergoing chest radiography [patients] are rib fractures on plain radiographs [test] specific for non-accidental injury [outcome]?

Search strategy and outcome

Secondary sources

Cochrane: rib fractures and non-accidental injury; no relevant reviews found.

Primary sources

Medline (including Medline corrections) 1966–01/02/2004. [(Validated paediatric search filter (March 2003) for Ovid¹) AND (exp Child Abuse or non-accidental injury.mp or child abuse$.mp or deliberate injury.mp or exp. domestic violence or child abuse, sexual or exp. Munchausen syndrome by proxy or exp torture or domestic violence.mp or Munchausen syndrome by proxy.mp. or torture.mp or non-accidental injuries.mp) and (exp.rib fractures or rib fracture$.mp or posterior rib fracture$.mp or multiple rib fracture$.mp or bilateral rib fracture$.mp or exp. thoracic injuries or thoracic injury.mp or chostochondral junction injuries.mp)] limited to human and English Language.

A total of 113 papers were identified; 105 were of insufficient quality for inclusion or irrelevant. One paper was subsequently excluded on critical appraisal due to flaws in case selection.

Hand search of references

Three further papers of sufficient quality for inclusion.

Summary of papers

See table 1.

Commentary

A number of studies have sufficient data to derive a 2×2 table from which likelihood ratios could be calculated, if the data were alternatively presented. The paper (not appraised in this article) by Worlock and colleagues,¹² comparing patterns of injury in children with non-accidental and accidental injury is such an example. The rib fracture data are presented as total number of rib fractures, rather than absolute numbers of children in each sub-group. The presentation of data in this manner does not allow for the calculation of sensitivity, specificity, or likelihood ratios.

The small studies by Barsness and colleagues,² Bulloch and colleagues,⁴ and Cadzow and Armstrong⁵ support the premise that rib fractures in a child less than 3 years are predictive of non-accidental injury, while the data from Garcia and colleagues⁷ show rib fractures in unselected age groups are poor predictors of non-accidental injury. Further analysis of the data by Garcia and colleagues⁷ by age group may confirm the positive predictive value of rib fractures in non-accidental injury in young children. Thomas¹¹ also showed a lower predictive value for rib fractures in children who diagnoses at the time were felt not to be either pathological or accidental; however, on re-reading the paper these may well have been non-accidental in nature by current criteria. King and colleagues⁸ and Merten and colleagues¹⁰ reported data in a way that allows the calculation of test sensitivity. Their papers show rib fractures have low sensitivities in this setting, and as such the absence of a rib fracture does not rule out non-accidental injury.

In any study, patient selection affects the applicability of the results, a point of particular relevance to diagnostic test studies. A study carried out in ventilated children in a tertiary referral central is not applicable to a general practitioner presented with a report confirming a rib fracture in a 2 year old child. The largest UK data set is from Carty and Pierce;³ however, this personal case series, while large in number, is unreflective of the situation in the emergency department or paediatric assessment unit. This selection bias leads to an overestimation of the specificity of rib fractures in the paediatric population.

To the authors’ knowledge there is no nationally or globally agreed gold standard diagnostic tool for the diagnosis of non-accidental injury. As a result, the various studies use a variety of criteria against which the performance of radiological rib fracture in the diagnosis of non-accidental injury is assessed. A further weakness of the selected gold standards is the inclusion of the test being validated (chest x ray) within the gold standard test.

In all the reviewed papers there is a lack of explicit blinding between radiologists and investigating parties. The clinical information that accompanies a radiograph is an important factor in the interpretation of subtle differences in rib morphology. To minimise bias, partial blinding could be achieved by independent reporting of the chest radiograph without the accompanying components of a skeletal survey and clinical details.

The studies reviewed have obvious radiological and ethical limitations, which are difficult to circumvent in clinical practice. The study to answer the question asked in the title would need to look at all children undergoing chest radiography, with each child being investigated for non-accidental injury through a standardised protocol. This would allow the calculation of the sensitivity and specificity of rib fractures for non-accidental injury in the population undergoing chest radiography. Issues of consent for such a study may be difficult to resolve, as the consent form would explain the reason for the study (the diagnosis of non-accidental injury) and lead to an underestimation of the prevalence of non-accidental injury, as abusers may not give consent for their child’s inclusion.