Branching cross-examination

This is a MedVellum formative viva. It is not an official RACP, MRCPCH, ABP, ACGME or RCPSC station, mark scheme, duration or pass standard. Release each update only after the candidate states the diagnosis, the urgency tier, and the immediate action. [1] [3]

Candidate brief

You are the senior paediatric clinician in a regional emergency department. Speak as you would at the bedside. State the urgency tier and the time target for each decision, name the object confidently from the radiograph, and apply the NASPGHAN and ESPGHAN urgency categories. This is one continuous case with three branches. [3]

Question 1 — The doorway radiograph

Stimulus update. A three-year-old is brought in forty minutes after the family found a remote-control battery missing and the child drooling and refusing drinks. The anteroposterior chest radiograph shows a 20-millimetre disc-shaped radiopacity at the thoracic inlet with a concentric ring appearance. Question: What is the diagnosis, what single additional view do you want, and what is your urgency tier? [10]

Consultant-level model answer. "The concentric ring or halo sign on the anteroposterior view is diagnostic of a button battery lodged in the upper oesophagus. I want a lateral view to confirm the step-off at the rim and to exclude a second object. This is an emergency within two hours: endoscopic removal is required regardless of symptoms because alkaline necrosis is already underway. I call anaesthetics, the paediatric endoscopist, and retrieval if two-hour removal is not possible locally." [10] [3]

Probing follow-up. "Why a lateral view?" A strong answer is: "The lateral view confirms the battery by showing the step-off at its rim, distinguishes a battery from a stacked pair of coins, locates the object in three dimensions, and may reveal a second magnet behind the first. A single anteroposterior view is not enough to characterise the object confidently." [10]

Common weak answer. "It is probably a coin, so I will observe the child and repeat the radiograph in the morning." This misreads the halo sign as a solid disc, ignores the drooling and refusal of drinks, and chooses the wrong urgency tier for an oesophageal object. [1]

Escalation branch. If the candidate identifies the battery and the two-hour target, release Question 2. If they call it a coin, ask which radiographic sign distinguishes a battery from a coin and how the urgency tier changes. [3]

Question 2 — Delayed endoscopy and the demulcent

Stimulus update. The on-call endoscopist is ninety minutes away in theatre. The child is over twelve months, alert, drooling but protecting the airway, with no evidence of perforation. Question: Do you give a temporising agent? If so, name it, give the dose, and list the contraindications. [3]

Consultant-level model answer. "Yes. The child is over twelve months, has no evidence of perforation, and the battery is confirmed in the oesophagus with removal delayed beyond two hours, so I give honey by mouth at approximately 10 millilitres every ten minutes up to six doses while awaiting endoscopy, as endorsed by the ESPGHAN 2021 position paper. The contraindications are infants under twelve months because of botulism risk, suspected oesophageal perforation, inability to protect the airway, and a battery that has already passed the pylorus. Honey is an adjunct to, never a substitute for, emergency endoscopy." [3]

Probing follow-up. "Why is honey contraindicated under twelve months?" A strong answer is: "Infant botulism, caused by Clostridium botulinum spores that can be present in honey, is a recognised risk in the first year of life. The demulcent benefit does not justify that risk in an infant, in whom I would prioritise fastest possible endoscopy without the adjunct." [3]

Common weak answer. "I would induce vomiting or try to retrieve the battery with a finger sweep." Both are dangerous: induced vomiting can lodge the battery in the airway or worsen oesophageal injury, and a blind finger sweep can push the object deeper or perforate. [1]

Escalation branch. If the candidate gives the correct dose and contraindications, release Question 3 on the delayed complication. If they propose an unsafe manoeuvre, ask them to name three things they must not do at the bedside. [3]

Question 3 — The sentinel bleed

Stimulus update. Two weeks later the child is brought back with a small amount of fresh haematemesis, having apparently recovered well after an uneventful endoscopic removal of the battery. Question: What is your primary concern, and what do you do now? [5]

Consultant-level model answer. "My primary concern is an aorto-oesophageal fistula herald bleed. Injury from a button battery does not end at removal; delayed perforation and fistula can declare days to weeks later. This is a maximal emergency. I keep the child nil by mouth, establish two large-bore intravenous cannulae, take a full blood count, group and hold, cross-match and coagulation studies, mobilise paediatric intensive care, paediatric surgery, interventional radiology and cardiothoracic surgery, and arrange urgent endoscopy and imaging. A sentinel herald bleed can be followed by exsanguinating haemorrhage within minutes." [5] [3]

Probing follow-up. "Why can this happen so late?" A strong answer is: "The initial alkaline burn initiates a transmural inflammatory cascade that can erode into the aorta over days to weeks, even after the battery is removed. The negative-pole injury drives necrosis that the original endoscopy may not fully reveal. Delayed fistula is a recognised and feared complication of oesophageal battery injury." [5]

Common weak answer. "The child had the battery removed and has been well, so this is likely a minor gastric irritation and I will arrange an outpatient review." This dismisses the cardinal sentinel sign and risks catastrophic delay. [5]

Escalation branch. If the candidate recognises the fistula risk, move to Question 4 on the contrasting cases. If they reassure, ask what symptom would change their mind and what time window applies. [3]

Question 4 — The asymptomatic gastric coin and the magnets

Stimulus update. Two further children present in the same week. The first is a four-year-old who swallowed a coin four hours ago, is eating normally, and has a single solid disc in the stomach on radiograph. The second is a six-year-old who swallowed several small magnets from a toy and is asymptomatic, with two spherical radiopacities seen in the stomach. Question: Contrast the management of these two children. [1] [6]

Consultant-level model answer. "The first child has a gastric coin and is managed by observation at home, because he is asymptomatic, eating normally, and has no prior gastrointestinal surgery. I ask the family to check the stools, to return urgently with abdominal pain, vomiting, fever or bleeding, and to attend for a repeat radiograph at four weeks if the coin has not passed. The second child has multiple magnets in the stomach, which are high-risk anywhere in the tract because they can attract across bowel walls and cause pressure necrosis, fistula, volvulus and perforation. I involve paediatric gastroenterology and surgery, arrange endoscopic removal if reachable, and never observe multiple magnets. The contrast is that an asymptomatic blunt gastric coin below threshold size is observed, while multiple magnets are removed or referred for surgery regardless of symptoms." [1] [6]

Probing follow-up. "What threshold sizes would change your management of the coin?" A strong answer is: "Objects longer than 6 centimetres or wider than 2.5 centimetres are unlikely to pass the duodenum and warrant endoscopic or surgical removal. A button battery in the stomach would also lower the threshold for removal, particularly if large or in a younger child." [1]

Common weak answer. "I will observe both children at home." This treats the multiple magnets as if they were a single blunt object and ignores the inter-loop injury risk. [6]

Escalation branch. If the candidate distinguishes the two correctly, conclude the viva. If they observe the magnets, ask them to describe the mechanism by which two magnets in adjacent bowel loops cause perforation. [6]