%0 Journal Article %A Helen Daley %A Hilary Smith %A Samantha McEvedy %A Rachel King %A Edward Andrews %A Faye Hawkins %A Nicole Guppy %A Todorka Kiryazova %A Rebecca Macleod %A Emma Blake %A Rachael Harrison %T Intracranial injuries on computed tomography head scans in infants investigated for suspected physical abuse: a retrospective review %D 2021 %R 10.1136/archdischild-2020-319762 %J Archives of Disease in Childhood %P 456-460 %V 106 %N 5 %X Background UK national guidelines recommend that investigation of infants (aged <12 months) with suspected physical abuse should always include CT head scans. Such imaging carries small but recognised risks from radiation exposure. Studies report a range of yields for occult intracranial injuries in suspected physical abuse.Aims To report the yield of intracranial injuries on CT head scans carried out for suspected physical abuse in infants, compare yields for those presenting with or without signs of head injury and to describe selected clinical and radiological features.Methods A retrospective cross-sectional review of case records of infants undergoing skeletal survey for suspected physical abuse in Wessex, England. The main outcome measure was yield of intracranial injuries on CT head scan.Results In total, n=363 CT head scans were included (n=275 aged <6 months). The overall yield of intracranial injury was 37 (10%). Among 68 infants presenting with neurological signs or skull fractures, yield was 36 (53%) compared with just 1 (0.34%) of 295 without neurological signs or skull fractures. This one intracranial injury was found to be consistent with an accidental fall. Scalp injury was the only additional clinical feature associated with intracranial injury.Conclusion In suspected physical abuse, CT head scans should be carried out in infants who present with neurological signs, skull fractures or scalp injuries. However, in balancing potential risks and benefits, we question the value of performing a CT head scan in every infant investigated for suspected physical abuse.All data relevant to the study are included in the article or uploaded as supplementary information. %U https://adc.bmj.com/content/archdischild/106/5/456.full.pdf