Chest CT in the evaluation of child abuse – When is it useful?

Abstract

BackgroundIndications for chest CT in evaluation of child abuse are unknown. ObjectiveDetermine which groups of children can best benefit from chest CT. Participants and setting10-year (1/2010 to 12/2019) retrospective study of children <3 years who had chest CT within 3 days of the initial skeletal survey. MethodsDemographic and clinical information were obtained from medical records. Two pediatric radiologists reviewed, independently and blinded to clinical information, anonymized rib X-rays (initial and follow up when available) and chest CT. Disagreements were resolved by a third pediatric radiologist. Agreement was evaluated using kappa statistics. Number and percentage of fractures were analyzed by negative binomial models and chi-square tests, respectively. Results50 children (21 females) with average age of 9.7 months, 27 of whom had follow-up radiography. Agreement on initial and follow-up X-rays was substantial (k = 0.72) and perfect (k = 1.00), respectively, and almost perfect (k = 0.82) for CT scans. Chest CT demonstrated more fractures than X-ray, both initially (112 vs. 42, p < 0.0001) and at follow-up (93 vs. 49, p < 0.0001). Significantly more additional fractures were found at CT (11/13, 84.6 %) in patients with positive than in those with negative initial surveys (7/37, 18.9 %, p < 0.001). Ten initial surveys had only indeterminate fractures; four of them had fractures and six had no fractures on CT. Chest CT missed one patient (1/27, 3.7 %) with acute nondisplaced anterior rib fractures. ConclusionChest CT can be considered in children with negative skeletal survey and high clinical suspicion for child abuse, and when the diagnosis of rib fractures is indeterminate.