We appreciate Gravel and colleagues’ efforts to develop a clinical decision rule that successfully identifies skull fractures among young children with mild head trauma and no indication for head computerized tomography (CT).1 We wonder, however, if these skull fractures warrant diagnosis and, specifically, if affected children benefit from their detection. Isolated skull fractures have been suggested as an example of overdiagnosis, the accurate detection of an abnormality from which a patient does not experience net benefit.2,3
Follow-up outcome data, such as receipt of surgical repair, are needed to assess the possibility of patient benefit, but they are not included in the present study. Other studies have found that clinical deterioration and surgical intervention are rare among well-appearing children with isolated skull fractures.4,5 Even when repair is performed among this cohort, the impetus is generally cosmetic. If growing skull fractures are the concern (which, as the authors concede, are exceedingly uncommon), then the important research question becomes how to best predict these specific fractures rather than how to predict skull fractures in general.
Faced with an unclear benefit of testing, we must consider the potential harms. How often did skull fracture findings trigger CT scans, for which there is an added risk of malignancy? Though isolated skull fractures do not necessarily warrant routine hospitalization, studies have demonstrated that most children with this finding are indeed admitted to hospital.5 Parental anxiety and guilt resulting from the news that their young child has a skull fracture is an additional concern.
Improving the means to detect abnormities is a timeless objective in medicine, but we must pair this work with efforts to determine whether children receive more benefit than harm as a result of increased or improved diagnosis.