Purpose or Case Report: The purpose is to evaluate how well non-pediatric radiologists interpret skeletal surveys in the cases of possible child abuse.
Methods & Materials: This is a retrospective study (Jan-Sept 2016). Fellowship trained pediatric radiologists reviewed outside skeletal surveys for child abuse in which our child protective team was consulted. We documented whether or not there was agreement with the outside interpretations and the type of discrepancies.
Results: 117 (23 different hospitals) studies of 112 children, age 5 days to 6 years (mean age 11 months) were reviewed. In 23/117 (19.7%) studies there was disagreement with the outside read (all of which were interpreted by non-pediatric radiologists). In 16/23 (69.6%) of the discrepant studies, fractures were identified which were not originally reported. The most common missed fractures were ribs (n=27). Five classic metaphyseal lesions were missed in three patients. In 7/23 (30.4%) of the discrepant studies, reported abnormalities were felt to represent normal findings or variants, which included two classic metaphyseal lesions thought to represent normal variants and one rib fracture which was a pseudoarthrosis.
Conclusions: We found discrepancies in 20% of the outside skeletal survey reports (all by non-pediatric radiologists). Due to the ramification of missing or over interpreting a fracture in children suspected for abuse, the ACR may consider establishment of specific requirements that radiologists must meet when interpreting skeletal surveys for possible child abuse.