In 1946 John Caffey described the association of subdural hematomas and skeletal fractures in infants and raised the possibility of maltreatment to explain these injuries.¹ Kempe and Silverman later coined the term “battered child syndrome” in a landmark article to the general medical community that described the characteristic radiologic features seen in abused children.² Child abuse is the second most common cause of brain injury and fractures in infants and young children.³ Diagnostic imaging is crucial in the evaluation of child abuse and its imitators.

Skeletal injures can be classified with regard to their relative specificity for abuse based on their imaging pattern and location (Table 41-1).^4-6 Highly specific fractures are usually identified in infants and are typically clinically occult.⁴ Most of these occur with indirect forces, rather than direct blows, explaining the usual absence of bruising overlying the fracture sites.⁷ Rib fractures near the costovertebral articulations occur with anteroposterior compression of the thorax that may be associated with violent shaking (Figure 41-1). The classic metaphyseal lesion (CML) results from torsional and tractional forces applied to the extremities (Figure 41-2); it may also occur with accelerational forces associated with infant shaking. Highly specific injuries are not caused by simple falls^8-11 or by two-finger cardiopulmonary resuscitation efforts.^12,13

Skull fractures have been reported in 10% of abused children and although usually linear, they may be multiple, diastatic, complex, bilateral, and depressed.^14-16 Skull fractures can occur with short falls to firm surfaces in young infants; therefore, simple linear skull fractures have a low specificity for abuse. Although long bone shaft fractures have a strong association with abuse in non-ambulatory infants, they can occur with household and playground accidents, particularly in toddlers and older children. Thus long bone shaft fracture must be viewed in conjunction with the clinical history.

Adequate imaging is critical for the detection of subtle fractures specific for abuse. Failure to perform an adequate skeletal survey may result in returning a child to a potentially dangerous environment. Jenny et al. found that nearly one-third of cases of abusive head trauma (AHT) had been previously seen by a physician where the diagnosis was missed.¹⁷ The detection of skeletal injuries depends on the technical quality and thoroughness of the skeletal survey. Digital imaging should be performed according to the guidelines of the American College of Radiology (ACR) and the American Academy of Pediatrics (AAP), and optimized to a high detail technique.^18-20 The skeletal survey protocol is shown in Table 41-2.²⁰ The study should be reviewed before completion by an experienced radiologist, and positive sites should be imaged in at least two projections.^19-21

Skeletal surveys are recommended in all infants and toddlers <2 years of age when there is suspicion of abuse. Clinically occult trauma in children older than 2 years is less common than in infants¹⁴; therefore skeletal surveys should be reserved for those older children when there is a strong suspicion of physical abuse. There is little value to global screening for inflicted skeletal injury beyond age 5.¹⁹