A diagnosis of abusive head trauma (AHT) reflects any inflicted injury to the head, brain, and/or spine of an infant or young child, whether by inertial injury, contact injury, or some combination of the two.1 AHT is most common in infants in the first year of life and has an estimated incidence of 20–30 cases per 100,000 infants under one year of age.2 With a mortality rate of nearly 25 %, AHT is the leading cause of fatal head injury in children less than 2 years and is the most lethal form of child physical abuse.1,11 AHT may present with a range of clinical findings, from nonspecific such as vomiting or irritability to life threatening such as seizures or cardiac arrest. Recognition and prompt diagnosis of AHT is crucial, as missed diagnoses can result in further harm to a child.3 Even when a child survives, there is significant morbidity including developmental delays, vision problems, and seizure disorders, among other sequelae.

The evaluation of AHT has multiple components, including comprehensive history taking and physical exam, neuroimaging, skeletal imaging, laboratory studies, and ophthalmologic exam. Please refer to the corresponding article detailing ocular findings in child abuse, which includes a review of retinal hemorrhages (insert reference and/or hyperlink). While all of these components are important, accurate interpretation of neuroimaging findings is critical. Neuroimaging modalities utilized in the evaluation of AHT include plain radiographs and cranial ultrasound (US), though computed tomography (CT) and magnetic resonance imaging (MRI) are preferred for evaluation of the extra-axial spaces and brain parenchyma, respectively. Three-dimensional (3D) CT is superior to plain radiographs for assessing skull fractures, especially more complex fractures and can help delineate whether a provided accidental trauma history is plausible. With respect to MRI, protocols should include conventional T1, T2, diffusion weighted images (DWI) with apparent diffusion coefficient (ADC), susceptibility weighted images (SWI) and fluid attenuated inversion recovery (FLAIR).6 Diffusion sequences are useful for identifying cytotoxic edema and hypoxic ischemic injury (HII), which are important prognostic indicators.4 Gradient echo (GRE) and SWI sequences offer enhanced detection of hemorrhage, particularly those that are small. Additionally, short tau inversion recovery (STIR) fat-saturation sequences can aid in the identification of spinal ligamentous and soft tissue injury.4 Although fast MRI with motion tolerant imaging has been proposed to obviate the need for sedation in patients, sequences offer decreased sensitivity and sedation may be needed for MR acquisition.5,6

While there is not a singular finding on neuroimaging that is pathognomonic for AHT, there are certain findings, particularly when co-occurring with other injuries, that are more likely than others to be due to non-accidental trauma. Conversely, it is important to recognize findings on neuroimaging that are suggestive of accidental injuries, or other medical causes. In this article, we aim to discuss the various findings on neuroimaging that have been associated with AHT, compared to those that are more consistent with accidental injuries or with underlying medical causes that may also be on the differential.