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 shock grade, the phenotype, the immediate action and the reassessment endpoint. [1] [6]

Candidate brief

You are the senior paediatric clinician in the emergency department. Speak as you would during resuscitation. Recognise shock from the whole child and the trend, never from a single blood pressure. State the change you expect from each action, and say what you will reassess. This is one continuous case. Each escalation branch leads to the next update. [9]

Question 1 — Doorway and recognising compensated shock

Stimulus update. A parent carries a three-year-old who has had three days of fever and now poor intake. Before you touch the child you see reduced interaction, cool mottled limbs, and fast laboured breathing. Question: What do you say and do now? [1]

Consultant-level model answer. "I am immediately concerned. The first impression shows abnormal appearance, increased work of breathing and poor circulation to skin, so I call the senior paediatric and resuscitation teams, name a leader and allocate roles. I begin the hands-on circulation assessment immediately. My working diagnosis is compensated shock until proven otherwise, because children hold their blood pressure until late and this child already shows failing perfusion at the doorway." [1] [6]

Probing follow-up. "Define shock." A strong answer is: "Shock is the failure of oxygen and substrate delivery, or of cellular use, to meet tissue demand, producing a shift to anaerobic metabolism. It is a clinical, whole-child diagnosis of poor end-organ perfusion, not a blood pressure number." [1] [9]

Common weak answer. "I will take a full history and order bloods and a chest X-ray first." This delays resuscitation for diagnostic completeness in a visibly shocked child. [6]

Escalation branch. If the candidate recognises compensated shock and calls for help, release the survey data in Question 2. If they anchor on the blood pressure, ask which signs already prove failing perfusion. [1]

Question 2 — The survey data and why blood pressure is normal

Stimulus update. Heart rate is 170, central pulses are felt but peripheral pulses are weak, capillary refill is 4 seconds on the sternum, skin is cool and mottled, blood pressure is normal for age, and the child responds to voice but cannot sustain interaction. Question: Is this child in shock despite the normal blood pressure, and why? [2]

Consultant-level model answer. "Yes, this is compensated shock. Cardiac output in young children is heart-rate dependent because stroke volume is relatively fixed, so the heart raises output by increasing heart rate, and the vasculature raises systemic vascular resistance to hold mean arterial pressure while perfusion to skin, gut and muscle fails. The blood pressure is normal precisely because compensation is still working; the weak peripheral pulses, prolonged capillary refill, cool mottled skin, reduced interaction and falling urine output prove failing tissue perfusion. Waiting for hypotension would mean diagnosing shock only at the cliff edge." [1] [2]

Probing follow-up. "What is the cliff edge?" A strong answer is: "The moment compensation breaks. The heart can no longer raise its rate and the vessels can no longer constrict, so blood pressure, consciousness and pulses fall together toward arrest. A falling or abnormally low heart rate in a deteriorating child is the hallmark that the cliff edge has arrived." [1] [9]

Common weak answer. "The blood pressure is normal so the child is stable; I will observe." This mistakes a maintained blood pressure for adequate perfusion and ignores the failing circulation. [3]

Escalation branch. If the candidate explains compensation correctly, release in Question 3 a fluid scenario. If they treat blood pressure as reassuring, ask what sign would prove shock. [2]

Question 3 — Phenotype and the fluid-as-a-ceiling principle

Stimulus update. You now suspect septic shock. The child is currently cold with a narrow pulse pressure. Question: Classify the shock phenotype and outline your fluid strategy. [5]

Consultant-level model answer. "This is septic shock, a distributive phenotype that is now cold and low-output with likely myocardial depression and relative hypovolaemia, so mixed. I give an isotonic crystalloid aliquot of 10 to 20 mL per kilogram, I state the response I expect before it goes in, and I reassess the whole child immediately afterwards. The first-hour total is a ceiling, not a target, so I repeat an aliquot only while the child remains in shock and is not overloaded. The Surviving Sepsis Campaign 2026 children's guideline supports up-front aliquots with careful reassessment and attention to fluid balance." [4] [5] [9]

Probing follow-up. "What did FEAST teach?" A strong answer is: "FEAST found increased early mortality with saline or albumin bolus compared with no bolus in African children with severe febrile illness. Its lesson is that a fluid algorithm cannot be transplanted across populations, shock types and rescue resources; it is not proof that every bolus is harmful. Here, with a fluid-responsive child in septic shock and critical care available, I still give aliquots but reassess rigorously and escalate rather than driving toward a fixed volume." [4]

Common weak answer. "I will give 60 mL per kilogram in the first hour." This turns the ceiling into a target and risks fluid overload. [4]

Escalation branch. If the candidate applies the ceiling principle, release in Question 4 that the child fails to respond. If they target a fixed volume, ask how they would recognise overload. [5]

Question 4 — No longer fluid-responsive: vasoactive support

Stimulus update. After two aliquots the child is no longer fluid-responsive: perfusion is unchanged, the respiratory rate is rising and there are basal crackles. Question: What now? [9]

Consultant-level model answer. "The child is no longer fluid-responsive and is showing overload, so I stop fluid and start vasoactive support early. The first agent follows the phenotype: a cold, low-output state leans toward an inotrope such as adrenaline, while a warm, vasodilated state leans toward a vasoconstrictor such as noradrenaline. I call critical care and retrieval in parallel, before local options are exhausted, and I keep reassessing the whole circulation." [5] [9]

Probing follow-up. "Why escalate to retrieval now rather than wait?" A strong answer is: "Decompensated shock that is no longer fluid-responsive is on the cliff edge toward arrest. The cost of waiting for local options to fail is an arrest that critical care could have prevented, so I call retrieval in parallel with the vasoactive decision." [6]

Common weak answer. "I will give a third and fourth aliquot." This continues fluid into a non-responsive, overloaded child and delays vasoactive support. [4]

Escalation branch. If the candidate starts vasoactive support, release in Question 5 that the child arrests. If they continue fluid, ask when fluid becomes harmful. [9]

Question 5 — When shock becomes arrest

Stimulus update. The child becomes unresponsive, with agonal breathing and no central pulse. Question: What now? [6]

Consultant-level model answer. "The child has arrested. I move directly into the paediatric arrest algorithm: high-quality chest compressions, ventilation with oxygen, attach a defibrillator, and establish intraosseous access. The arrest-dose anchors are intravenous or intraosseous adrenaline at 10 micrograms per kilogram every three to five minutes, and defibrillation at 4 joules per kilogram for a shockable rhythm. I do not finish a survey of an arrested child." [6]

Probing follow-up. "Why intraosseous?" A strong answer is: "Intraosseous access is rapid, reliable and equivalent to intravenous for drug delivery in arrest when intravenous access is delayed, so it is the recommended route when the clock is running." [6]

Common weak answer. "I will continue looking for a peripheral vein." This delays arrest drugs; intraosseous access is the answer in arrest. [6]

Escalation branch. If the candidate runs the arrest algorithm correctly, move to the handover. [6]

Question 6 — Structured handover

Stimulus update. Return of spontaneous circulation is achieved and the retrieval team arrives. Question: Give your handover. [9]

Consultant-level model answer. "My structured handover transfers identity and working weight, current physiology and trend, the shock grade and phenotype, each fluid aliquot and vasoactive agent with time, dose and response, the prioritised differential and pending tests, the local limits, family and safeguarding information, and the next contingency with a named owner. The one principle I want the team to carry forward is: recognise shock from the whole child and the trend, give fluid as a ceiling not a target, and escalate to vasoactive support and retrieval before local options fail." [9]

Probing follow-up. "What was the single most important early sign in this child?" A strong answer is: "Tachycardia out of keeping with the illness, integrated with the failing perfusion signs and the trend, because it is the first and most important compensatory response, and its fall heralds decompensation." [1] [2]

Common weak answer. "I will let the retrieval team work it out from the chart." This loses trend, timed actions and contingency. [9]