Durgical
EN
Open the app

Imaging when abuse is suspected: the skeletal survey

  • Reading

When a young child may have been hurt on purpose, a full set of bone X-rays and often a head scan look for hidden injuries, including old ones.

Updated

What it is

Imaging in suspected physical abuse looks for injuries that cannot be seen or felt, dates them roughly and checks whether they fit the story given. The core study is the skeletal survey: a standardised series of many separate radiographs of the whole skeleton (skull, spine, chest with ribs, pelvis and each long bone, hand and foot), not a single whole-body film. It is indicated in every child under about two years with suspected physical abuse and is considered in older children who cannot communicate. A follow-up survey about two weeks later finds fractures that were invisible at first and become visible as they heal. Infants with suspected abuse, especially under six months or with any neurological sign, also need head imaging: non-contrast CT in the acute setting and MRI of the brain and spine afterwards, to look for abusive head trauma. Imaging supports the diagnosis but does not make it alone: it is combined with the history, examination, eye examination and laboratory tests by a multidisciplinary child-protection team. A reasonable suspicion is enough to trigger the duty to report.

What to check

  • High specificity for abuse: classic metaphyseal lesions (corner or bucket-handle fractures at the end of long bones), posterior rib fractures, scapular, spinous process and sternal fractures.
  • Moderate specificity: multiple fractures, especially bilateral or of different ages, vertebral body fractures, finger fractures in non-mobile infants, complex skull fractures, epiphyseal separations.
  • Low specificity: clavicle fractures, single long-bone shaft fractures and linear skull fractures; these are common in accidents but suspicious when the history does not fit.
  • Any fracture in an infant who is not yet walking or cruising deserves an explanation.
  • Abusive head trauma: subdural haematomas, often along the falx or over both hemispheres and sometimes of different densities, with brain oedema or hypoxic injury; retinal haemorrhages are seen by the ophthalmologist, not on CT.
  • Salter-Harris types (I to V) describe fractures through the growth plate; they are common in accidental injury and help describe and follow any physeal fracture.

Reading it step by step (and why)

  1. Confirm the indication (child under two, or non-verbal) and the full protocol — a single whole-body film misses subtle fractures.
  2. Check that every view is present and of good quality, including both hands, feet and oblique rib views — ribs and hands hide fractures.
  3. Look at every metaphysis for corner or bucket-handle fractures — the classic metaphyseal lesion is highly specific for abuse.
  4. Trace each rib, front to back, for fractures and callus — posterior rib fractures come from squeezing the chest.
  5. Review skull, spine, pelvis, scapulae and sternum — unusual sites raise specificity.
  6. Date each fracture roughly: no periosteal reaction means recent; new bone and callus appear over one to two weeks — mixed ages suggest repeated injury.
  7. Look for mimics: osteopenia, rickets, osteogenesis imperfecta, birth injury, normal variants — mimics must be excluded before concluding.
  8. Correlate with the head imaging, the history given and the developmental stage — the key question is whether the injury fits the story.
  9. Plan a follow-up survey in about two weeks — healing makes occult fractures visible and helps dating.

Common mistakes

  • Fracture dating is approximate; report ranges, not exact days.
  • Physiological periosteal new bone in young infants is thin, symmetric and along the shafts; do not call it injury.
  • A normal skeletal survey does not exclude abuse; bruising, burns or head injury may be the only signs.
  • Siblings and other children under two in the same home may also need assessment.
  • Use neutral, factual words in the report; describe the findings and their specificity, and avoid accusing anyone.

Next step

  • Child under two with suspected abuse: skeletal survey, and head CT if under six months or with any neurological sign, irritability or bruising of the head.
  • Request brain and spine MRI after the acute phase when the CT is abnormal or the suspicion is high; eye examination by ophthalmology.
  • Abdominal injury suspected (raised liver enzymes, pancreatic enzymes, blood in urine or abdominal signs): contrast-enhanced abdominal CT.
  • Order a follow-up skeletal survey about two weeks later.
  • Call the radiologist to review the images and the child-protection team, and report the suspicion according to the legal duty in your country.

Test yourself

Think of the answer, then tap the question to see it.

Which two fracture types are most specific for physical abuse in infants?

Classic metaphyseal lesions (corner or bucket-handle fractures) and posterior rib fractures; scapular, spinous process and sternal fractures are also highly specific.

Why is a follow-up skeletal survey done about two weeks after the first?

Some fractures are invisible at first and only show once periosteal new bone and callus form; healing also helps date injuries.

Which infants with suspected abuse need head imaging even without neurological signs?

Infants under about six months, because abusive head trauma can be silent at that age; CT acutely, MRI later.

Which two conditions can mimic abusive fractures on radiographs?

Osteogenesis imperfecta and rickets; birth injury, osteopenia of prematurity and normal variants are other mimics.

Add to my cards (4)

Elsewhere in DurgiPedia

Connections

← All of radiology

Educational content; it does not replace a radiologist's report and is not a clinical instruction. Each study is ordered on the patient's assessment and your institution's protocols.

Sources: Child Abuse and Neglect (StatPearls (NCBI Bookshelf), 2025); Pediatric Abusive Head Trauma (StatPearls (NCBI Bookshelf), 2025); Salter-Harris Fracture (StatPearls (NCBI Bookshelf), 2023).

Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake

Representative model, for educational purposes only.