Abusive head trauma — branching viva

Opening

Examiner: A three-month-old infant arrives in the emergency department with altered responsiveness and a seizure. The CT shows a thin inter-hemispheric subdural bleed and cerebral oedema. The father says he rolled off a low bed an hour ago. How do you frame this? [1]

Candidate: I would treat this as a critically brain-injured infant and as suspected abusive head trauma in parallel. I would resuscitate the airway, breathing and circulation and treat the seizures and raised intracranial pressure with neuroprotection and urgent neurosurgical and intensive-care input, while informing my consultant and the child-protection team from the first minute. The history of a short fall does not fit a thin inter-hemispheric subdural bleed with cerebral oedema, so I would arrange the full inflicted-injury workup and keep the child safe while it completes. [1]

Branch 1 — building the probability

Examiner: What makes you think this is inflicted rather than accidental? [2]

Candidate: Four things raise the probability. First, the history is developmentally implausible — a three-month-old does not reliably roll, and a short fall onto a flat surface rarely causes cerebral oedema. Second, the injury pattern of an inter-hemispheric subdural bleed with oedema is characteristic of abusive rather than accidental head trauma. Third, I would look for retinal haemorrhages, skeletal injury and cutaneous injury across the other domains. Fourth, the Maguire pooled analysis and the Piteau radiographic systematic review define the discriminating features, and the PediBIRN screen — here, suspected seizure, severe head injury and, if present, ear or torso bruising — flags likely AHT. [2] [3]

Examiner (probe): So is AHT a diagnosis of a "triad"? [5]

Candidate: No. The triad of subdural haemorrhage, retinal haemorrhage and encephalopathy is supportive when mimics are excluded and the history is implausible, but no single finding and no fixed triad diagnoses AHT alone. The AAP renamed shaken baby syndrome to abusive head trauma in 2009 precisely to move away from mechanism- and label-based diagnosis, and the RCPCH critique reinforces that we present a probability built on the whole picture. [5]

Branch 2 — the workup

Examiner: Walk me through your investigation plan. [1]

Candidate: Neuroimaging first — the CT is done; I would arrange an MRI brain within days to map and date the injury, looking for diffuse axonal injury and hypoxic-ischaemic change. Then three parallel streams: a full skeletal survey to protocol with a repeat film at about eleven to fourteen days to surface occult fractures; dilated ophthalmology by an ophthalmologist with imaging to document the number, layer and extent of retinal haemorrhages; and a laboratory occult-injury screen — coagulation, full blood count, vitamin-K history, liver and pancreatic markers, urinalysis, and infection and toxicology screens. [1]

Examiner (probe): Why a repeat skeletal survey? [1]

Candidate: Because some fractures, particularly classical metaphyseal lesions and rib fractures, are occult on the initial films and become visible only as they heal and callus forms over the following two weeks. The repeat survey raises the yield of high-specificity inflicted-injury markers. [1]

Branch 3 — safeguarding and contacts

Examiner: He has a two-year-old sister at home. What do you do about her? [4]

Candidate: Screen her for occult injury through the child-protection team — a careful history, a full examination including the skin and a developmental assessment, and imaging as age-appropriate, typically a skeletal survey. The Lindberg study showed a real prevalence of abusive injuries in siblings and household contacts of physically abused children, so contact screening is standard, not optional. [4]

Examiner (probe): What is your mandatory-reporting duty? [1]

Candidate: In my jurisdiction a registered clinician must report a reasonable suspicion of child abuse to child protection. I would inform the consultant and the child-protection team, make the report, document it, and ensure the child is not discharged to the care of a suspected perpetrator while the assessment and the safety plan are completed. [1]

Branch 4 — the mimics and the court

Examiner: The defence raises a bleeding disorder. How do you respond? [5]

Candidate: I would have excluded reasonable mimics with the coagulation and full-blood-count screen and a vitamin-K history, and I would present the result honestly. A bleeding disorder can mimic the imaging and can also coexist with, or be a consequence of, brain injury — so a positive screen does not exclude abuse. The point is that I present a probability on the whole picture, not a verdict, and I document the mimics I considered and excluded. [5]

Examiner (final corner): And the long-term outlook? [6]

Candidate: AHT carries a high mortality — around a fifth of recognised cases — and a heavy burden of lifelong disability: cognitive impairment, epilepsy, visual impairment, motor deficits and behavioural disorders, often emerging only over years. I would plan neurodevelopmental follow-up, early-intervention services and rehabilitation from the outset, and a family-support plan that addresses the triggers and secures a safe disposition. [6]