Purpose or Case Report: This study aimed to elucidate the imaging appearance of costochondral junction (CCJ) fractures in young children, determine the association with nonaccidental trauma (NAT), and suggest a mechanism for injury.
Methods & Materials: This IRB-approved, retrospective study identified 19 children ages 0-3 years with CCJ fractures over a 10 year period. Radiographs, CT, MRI, and ultrasound studies were reviewed by a pediatric radiologist to determine number, location, and imaging appearance of the CCJ fractures. Medical records were reviewed for concurrent injuries, predisposing risk factors for fractures, and conclusions of the child abuse team.
Results: A total of 92 CCJ fractures (36 right and 56 left) were found in 6 females and 13 males. The child abuse team concluded 17 cases (94%) were NAT. Most cases had additional injuries (78.9%) and all had multiple CCJ fractures. Fractures in 10 cases (53%) were bilateral, 1 (5%) was unilateral right, and 8 (42%) were unilateral left. Eleven of the 14 cases (79%) with both CT and skeletal survey showed 36 additional CCJ fractures on CT compared to skeletal survey. Key imaging features included: bucket handle fracture (n=28), corner fracture (n=45), corner irregularity (n=14), cortical buckle (n=35), longitudinal separation of the cortex from the shaft (n=51) and longitudinal bone growth along the costal cartilage (n=29). The cortical break was identified on the inner cortex (n=45), outer cortex (n=2), or both cortices (n=14).
Conclusions: This study reveals that CCJ fractures are highly associated with NAT and specific for NAT in this age group. Patients with CCJ fractures likely have additional injuries; therefore, laboratory, physical, and imaging tests should be conducted looking for visceral and head injuries. The inner cortex of the CCJ is most commonly fractured, supporting the previously proposed injury mechanism as compressive force on the anterior chest. This study also shows the range of imaging appearances of CJJ fractures. The bucket handle and corner fracture appearance suggest that CCJ fractures are analogous to classic metaphyseal lesions (CMLs) of long bones. The longitudinal cortical separation not seen in CMLs is likely related to costal cartilage extension into the fractures during healing as reported previously. Finally, CT is superior to radiographs in identifying CCJ fractures. Given the high specificity of CCJ fractures for NAT, consider acquiring CT chest examinations in NAT cases to identify radiographically occult CCJ fractures.