Cardiac arrest sits squarely in SLO3 (resuscitation) and turns up in some form almost every sitting. But the exam is written at the level of a day-one consultant — the person the resus room turns to when the standard loop isn’t working. So the testable question is rarely “what is the ALS algorithm?” It is the modification: the cause that changes the drugs, the threshold that switches a treatment on or off, the device you reach for when manual CPR is failing, and the judgement to keep going or to stop.

This guide follows the Resuscitation Council UK (RCUK) Special Circumstances framework and the “ALS and beyond” material, organised by the way the exam actually asks it. For each topic we give the modification, the doses and thresholds worth knowing, and the consultant-level point — the reason it matters and the trap it sets. It is a high-yield working set, not the entire syllabus, and clinical doses should always be checked against the BNF and your local protocols.

It drops you at a decision point and asks what the algorithm does here. The cause is handed to you in the stem; the mark is the next step. A few worked examples set the pattern:

• Core temperature 28°C, three shocks given, still VF → stop shocking until the core is above 30°C, and switch to antero-posterior pads.
• Massive PE thrombolysed, no ROSC at 30 minutes → keep going — continue CPR for 60–90 minutes.
• 37 weeks pregnant, no output, four minutes in → resuscitative hysterotomy, at the bedside, now.
• Anaphylaxis, two IM adrenaline doses, still shocked → start an IV adrenaline infusion.

None of those is the standard algorithm. Each is a deliberate, defensible deviation from it — and that is the level the exam is pitched at.

Often the cause is handed to you as an image. Be ready for an ultrasound — a common format shows you a POCUS still or clip, the patient arrests, and you match the finding to the action: a pericardial effusion with tamponade → pericardiocentesis or thoracotomy; a dilated, strained right ventricle → suspect PE and thrombolyse; absent lung sliding → tension pneumothorax, decompress; cardiac standstill → factor it into the prognosis. Whatever the image, scanning must not interrupt compressions.

An ECG works the same way. The exam shows you a tracing, the patient arrests, and you read off the likely cause and adapt the algorithm to it: a sine wave or tall tented T waves → hyperkalaemia; S1Q3T3 with right-heart strain → massive PE (thrombolyse, CPR 60–90 min); marked ST-elevation → coronary occlusion (PCI); a long QT → torsades (magnesium); Osborn (J) waves → hypothermia (the temperature-adapted algorithm); a very broad QRS in an overdose → sodium-channel blockade (bicarbonate). For the patterns themselves, see our guide to the ECGs the FRCEM Final tests.

Refractory VF is VF that persists after three shocks. The first job is to keep delivering high-quality, minimally-interrupted compressions and shocks while you escalate — not to abandon the basics for the exotic.

Drugs, with the timing right:

• Adrenaline 1 mg after the third shock, then every 3–5 minutes.
• Amiodarone 300 mg after three shocks, with a further 150 mg after five (lidocaine 100 mg, then 50 mg, if amiodarone is unavailable).

Change the vector before you reach for a second machine. If VF persists after three shocks, confirm the antero-lateral pads are correctly placed (the lateral pad is commonly too anterior), then change the defibrillation vector by moving to antero-posterior pad placement at the next rhythm check. RCUK favours this vector-change defibrillation, and is explicit that dual (double) sequential defibrillation is not recommended for routine use outside a research setting — it needs two machines, risks equipment damage, and the evidence for added benefit over a simple vector change is weak.

ECPR — which patients? Extracorporeal CPR is a rescue therapy for selected patients when conventional CPR is failing and the cause is reversible. The features that define the salvageable candidate are a witnessed arrest with bystander CPR, a refractory shockable rhythm, a short low-flow time, a reversible cause, and few comorbidities (typically younger). Selection is protocol-dependent (ELSO / local ECPR service). Think of it as a bridge to a definitive fix, activated early — not a last resort.

Electrical storm is three or more sustained episodes of VT, VF or appropriate ICD therapies within 24 hours. The key to the whole topic is the pathophysiology: sympathetic drive sustains the storm, which gives the guiding principle — beta-blockers help and adrenaline harms. That single idea drives most of the right answers. (The caveat: this applies to the recurrent storm in a patient with a perfusing rhythm — in a single asystolic or PEA arrest you still follow standard ALS, adrenaline included.)

The approach:

• Find and treat the trigger — acute ischaemia, electrolyte disturbance (especially potassium and magnesium), QT-prolonging drugs, and decompensated heart failure.
• Beta-blockade over more adrenaline — esmolol (500 mcg/kg load, then 50 mcg/kg/min titrated) is the short-acting choice when haemodynamically borderline; propranolol (non-selective and CNS-penetrant) has the best outcome evidence. Give cautiously in a poor ventricle.
• Magnesium 2 g for polymorphic VT, with amiodarone 300 mg (then 150 mg) or lidocaine 1–1.5 mg/kg as antiarrhythmic adjuncts.
• Torsades (polymorphic VT with a long QT): magnesium 2 g regardless of the serum level, correct K⁺ and Mg²⁺, stop QT-prolonging drugs; for pause-dependent torsades raise the rate with overdrive pacing or isoprenaline — but isoprenaline is contraindicated in congenital long QT.
• Deep sedation and intubation reduce the sympathetic drive directly — an underused but high-yield manoeuvre.
• Ultrasound-guided stellate ganglion block is a rescue option for the refractory storm, interrupting cardiac sympathetic outflow.
• Definitive care: survivors usually go to the cath lab post-ROSC; catheter ablation and mechanical support / ECMO are the escalation for the truly refractory.

The reversible causes drive the whole algorithm — identify and treat them without delay. The exam assumes you can list them; it tests the specific treatment and the threshold. The detail for each follows in its own section, but the framework is worth holding in one place.

In a special-circumstances stem the H or T is effectively the diagnosis — the cause, not the rhythm, dictates the next step.

Hypothermic arrest is one to know thoroughly — its clear, temperature-driven decision points are exactly what a question can be built around. The management changes with the core temperature, so the temperature itself tells you what to do. Measure it with a low-reading thermometer, and check for signs of life for up to a minute before confirming arrest.

Below 28°C, delayed or intermittent CPR is acceptable if continuous compressions are not feasible (for example during a difficult extrication).

How you rewarm depends on how sick they are — and the methods differ markedly in speed (rates are approximate and vary with the patient):

The best option in a hypothermic cardiac arrest is veno-arterial ECMO (ECLS) — it is both the fastest rewarming method and the only one that supports the circulation while it works. So a hypothermic arrest with risk factors (heart rate < 45, systolic BP < 90, VF, or a core below 30°C) is transferred to an ECPR centre, and in-hospital prognosis is judged with the HOPE score (Hypothermia Outcome Prediction after ECLS — it integrates age, the mechanism of cooling, CPR duration, serum potassium and core temperature into a survival probability). If an ECLS centre cannot be reached within a reasonable time (RCUK gives roughly 6 hours), begin active non-extracorporeal rewarming in the meantime.

Severe hyperkalaemia is a potassium above 6.5 mmol/L, and the ECG evolves predictably: peaked T waves, then P-wave flattening and QRS widening, then a sine-wave pattern and arrest. A broad-complex arrest in a dialysis patient should be treated empirically as hyperkalaemia before the result returns.

The given — assume it is already done:

• Calcium chloride 10 mL of 10% IV — protects the myocardium (does not lower potassium).
• Insulin 10 units + 25 g glucose and sodium bicarbonate 50 mmol — shift potassium intracellularly.

What the exam usually asks — what is next? Calcium and insulin–glucose are common knowledge, so a good stem hands you those and asks for the next step:

• Nebulised salbutamol 10–20 mg — further potassium shift.
• Urgent dialysis — the route that removes potassium fast enough to matter in the arrest; arrange it early.
• ECPR to bridge to dialysis in refractory or recurrent hyperkalaemic arrest.

A gut potassium binder — sodium zirconium cyclosilicate 10 g orally — also removes potassium, but its onset is far too slow to rely on during the arrest itself. Think of it as a peri-arrest / post-ROSC measure (in a patient who can take oral medication), not a resuscitation drug.

Divide hyperthermia into malignant hyperthermia and everything else. Malignant hyperthermia is the one with a specific drug: stop the trigger (volatile agent or suxamethonium), give dantrolene 2.5 mg/kg IV repeated to control (up to roughly 10 mg/kg per the AAGBI crisis guideline), cool actively, and treat the hyperkalaemia and acidosis. Heat stroke and the toxidromes (serotonin syndrome, neuroleptic malignant syndrome) are managed by active cooling and supportive care — dantrolene does not work in heat stroke.

Cooling methods, ranked by how fast they cool — an examinable list:

1. Cold-water / ice immersion — fastest; first-line for exertional heat stroke.
2. Evaporative cooling — tepid water spray plus fans; best tolerated in the elderly / classic heat stroke.
3. Ice packs to neck, axillae and groin, with cold IV fluids.
4. Endovascular or surface cooling devices — in-hospital.

Stop active cooling before the core reaches normal — the Wilderness Medical Society guidelines give a target of 38.3–38.8°C — to avoid overshoot into hypothermia, after which the body’s own thermoregulation resumes. And antipyretics do not work in any of these — heat stroke, serotonin syndrome or NMS — because the heat comes from excess production, not a raised hypothalamic set-point.

Suspected PE: give a fibrinolytic (the established agent is alteplase, commonly 50 mg IV during CPR), and then commit to 60–90 minutes of CPR before stopping — thrombolysis needs time to work, and stopping early is the trap. Surgical embolectomy, catheter thrombectomy and ECPR are the options for the refractory case.

Coronary thrombosis: after ROSC with persistent ST-elevation, or with haemodynamic / electrical instability, go to immediate coronary angiography and PCI (within 120 minutes). Without ST-elevation, angiography is generally delayed unless there is a high suspicion of acute occlusion. Thrombolysis is the fallback for STEMI only where timely PCI is not achievable.

The alteplase dose is the ESC / UK standard-practice figure; the RCUK Special Circumstances guideline states “use fibrinolytic drugs” without printing a specific dose — check your local thrombolysis protocol.

Two rules govern the poisoned arrest: resuscitate for far longer (drug levels fall as the toxin is metabolised or excreted), and protect yourself (skin contact can transmit some agents). Call TOXBASE or your poisons centre early. Then match the toxidrome or ECG to the antidote:

Toxic bradycardia or asystole may respond to atropine and early transcutaneous pacing, and local-anaesthetic toxicity is a recognised indication for prolonged resuscitation and ECPR.

Traumatic arrest inverts the usual priority: fixing the reversible cause comes before chest compressions, because compressing an empty heart achieves little. Work the reversible causes (the “HOTT” set) in parallel:

• Hypovolaemia — control catastrophic haemorrhage; give blood / activate the major haemorrhage protocol; tranexamic acid.
• Hypoxia — secure the airway and oxygenate.
• Tension pneumothorax — bilateral open (finger) thoracostomies, not needle decompression, converting to drains once perfusing.
• Tamponade — in penetrating arrest, resuscitative (clamshell) thoracotomy, the decision driven by the 4 Es: Expertise, Equipment, Environment and a short Elapsed time.

Point-of-care ultrasound guides the whole sequence. For the wider trauma framework this sits within, see our ATLS 11 guide for the FRCEM Final.

In the patient who still has a pulse, treat the anaphylaxis to prevent the arrest. First line is IM adrenaline 500 mcg (0.5 mL of 1 mg/mL) into the anterolateral thigh, repeated after five minutes if there is no improvement, with a rapid 500 mL IV crystalloid bolus (severe shock may need 3–5 litres).

After two IM doses without response — refractory anaphylaxis — start an IV adrenaline infusion (1 mg in 100 mL, roughly 0.5–1 mL/kg/h, titrated). This is an infusion, not IV boluses — bolus IV adrenaline is hazardous in a patient who still has output. Add glucagon if the patient is on a beta-blocker and not responding, and give high-flow oxygen and further fluids. Steroids are no longer part of routine emergency treatment.

In cardiac arrest, the dosing changes to standard ALS: IM absorption is now unreliable, so give IV adrenaline 1 mg boluses (not the titrated infusion used while there was a pulse), with large-volume IV fluids. Consider prolonged CPR and ECPR — these patients arrest from a sudden, reversible cause having been previously well, so they are exactly the group who can be salvaged.

By the time an asthmatic is peri-arrest this is near-fatal asthma, and the standard treatment should already be running — high-flow oxygen, back-to-back nebulised salbutamol 5 mg and ipratropium 500 mcg, and IV steroids (doses follow BTS/SIGN; RCUK defers asthma dosing to BTS). So the arrest-specific moves are to make sure magnesium sulphate 2 g IV over 20 minutes has been given if it hasn’t, escalate to IV bronchodilators, and prepare to intubate (ketamine is a useful induction agent for its bronchodilator effect).

In cardiac arrest, RCUK 2025 advises:

• Treat life-threatening hypoxia with 100% oxygen.
• Check for evidence of pneumothorax and tension pneumothorax.
• Exclude anaphylaxis as a precipitating cause.
• Provide endotracheal intubation (due to high inflation pressures).
• Consider manual decompression and disconnection from the ventilator to manage dynamic hyperinflation.
• Consider ECPR in accordance with local protocols if initial resuscitation efforts are unsuccessful.

Two of those carry the most weight. Dynamic hyperinflation (breath-stacking) is the trap: obstructed lungs cannot empty, so intrathoracic pressure climbs, venous return falls, and the patient drops into PEA — the fix is to disconnect from the ventilator and manually compress the chest to let the trapped air out. And the ventilator strategy matters as much as the drugs: ventilate to allow full exhalation and accept the trade-off (permissive hypercapnia):

• Low tidal volume (~6 mL/kg ideal body weight)
• Low respiratory rate (~6–10/min)
• Prolonged expiratory time — a low I:E ratio (~1:4 to 1:5)
• Tolerate a high CO₂ and a pH down to ~7.2
• Keep plateau pressure < 30 cmH₂O; use minimal PEEP

From 20 weeks (uterus at or above the umbilicus), relieve aortocaval compression with manual left uterine displacement, which is preferred over lateral tilt because it keeps the patient supine for effective compressions, and maintain it throughout.

Resuscitative hysterotomy is the time-critical intervention, performed at the site of the arrest. The target is delivery within 5 minutes of arrest — so if there is no ROSC, start the procedure by about 4 minutes. It is done primarily to resuscitate the mother: emptying the uterus relieves aortocaval compression and restores venous return, and many of the maternal causes improve once the baby is delivered (haemodynamics, oxygen demand and the gravid-uterus physiology all ease) — and in pre-eclampsia and eclampsia, delivery is itself the definitive treatment. Fetal survival is a secondary benefit. Do not move her to theatre — the transfer wastes the only minutes that matter, and CPR continues throughout.

The technique is examined too (it sits across SLO3 and SLO6), so know the practical detail:

• Who: the most experienced clinician present — ideally an obstetrician, but do not wait for one to arrive; an emergency physician or surgeon performs it.
• Equipment: essentially just a scalpel (a large, e.g. size-10, blade). No anaesthesia (she is in arrest) and asepsis is not the priority.
• Incision: a vertical midline abdominal incision from symphysis pubis to umbilicus, then a small entry into the lower uterus extended vertically up the uterus (a hand beneath the blade to protect the fetus).

Add the pregnancy-specific causes — the 4 Ps — to the standard 4 Hs and 4 Ts: Pre-eclampsia / eclampsia (magnesium sulphate, with a 4 g IV loading dose then 1 g/h per NICE NG133), Puerperal sepsis, Placental and uterine causes (haemorrhage), and Peripartum cardiomyopathy.

Drowning is a hypoxic, out-of-hospital arrest, and the one modification reflects that: start with 5 ventilations using 100% oxygen, then continue standard CPR — airway and breathing come first. In-water spinal immobilisation must not delay getting the patient out, and you should expect regurgitation.

Because most drownings are out-of-hospital arrests, the work that earns marks often comes after ROSC: aspiration drives acute lung injury / ARDS, so think lung-protective ventilation and PEEP, watch for delayed deterioration, and remember many of these patients are also hypothermic and should be rewarmed and prognosticated accordingly.

Waveform capnography runs through the whole arrest, and reading the trace is examinable in its own right. It answers four questions:

• Is the tube in? A present, square waveform confirms the tracheal tube is in the trachea; a flat trace means it is not (oesophageal or displaced) — the first thing capnography proves.
• Is the CPR any good? End-tidal CO₂ reflects the cardiac output your compressions generate. The RCUK physiology-guided target is ETCO₂ ≥ 3.3 kPa (25 mmHg) (with a diastolic BP ≥ 30 mmHg). A low value means push harder and faster, or swap a tiring compressor.
• Has there been ROSC? A sudden, sustained rise in ETCO₂ during compressions is often the first sign of return of spontaneous circulation — note it and confirm at the next rhythm check rather than stopping compressions to feel for a pulse. (A sudden drop means the opposite: lost output, or a displaced tube.)
• What is the prognosis? An ETCO₂ that stays below ~1.3 kPa (10 mmHg) after 20 minutes of ALS, in an intubated patient receiving good-quality compressions, is associated with a very poor outcome — but it is only one strand of the “when to stop” judgement, never used on its own.

One more pattern worth knowing: a persistently low ETCO₂ despite good compressions points to very low pulmonary blood flow — think massive pulmonary embolism — or inadequate CPR.

Return of spontaneous circulation is the start of the post-cardiac-arrest syndrome — hypoxic brain injury, myocardial dysfunction, a systemic inflammatory (vasoplegic) response, and the persisting precipitant. The examinable targets:

• Oxygenation: titrate after the initial high-flow phase to SpO₂ 94–98% (PaO₂ ~10–13 kPa) — avoid hyperoxia.
• Ventilation: aim for normocapnia, guided by end-tidal CO₂ and blood gases.
• Haemodynamics: target a mean arterial pressure ≥ 65 mmHg; an arterial line is recommended.
• Temperature: targeted temperature management — actively prevent fever / aim for normothermia (routine therapeutic hypothermia has fallen away in recent trials).
• Seizures: treat aggressively; use short-acting sedation so neurological assessment is possible.
• Coronary angiography as above for ST-elevation or a high suspicion of coronary occlusion.

Questions often go a step further — ROSC achieved, now hypotensive or bradycardic, what next? — so know the management, not just the target:

• Hypotension → cautious fluid boluses and a vasopressor: noradrenaline is first-line for the vasoplegic, low-resistance picture. Add an inotrope (dobutamine, or adrenaline) where there is post-arrest myocardial dysfunction with a low cardiac output. Aim for that MAP ≥ 65 mmHg.
• Bradycardia with adverse features → the bradycardia algorithm: atropine 500 mcg IV, repeated to a maximum of 3 mg, then an adrenaline (2–10 mcg/min) or isoprenaline (5 mcg/min) infusion, and/or transcutaneous pacing as a bridge to transvenous pacing.
• Tachyarrhythmia → treat per the peri-arrest tachycardia algorithm (synchronised cardioversion if unstable).
• Re-arrest → go straight back to the reversible causes — the precipitant has not gone away.

Some stops are clear-cut, not a judgement call. CPR is not started — or stopped if one comes to light mid-resuscitation — when there is a valid DNACPR decision, or a valid and applicable advance decision to refuse treatment (ADRT). It is also withheld where there are signs unequivocally incompatible with life (decapitation, massive cranial destruction, rigor mortis, hypostasis). And it stops, of course, on ROSC.

Away from those, this is one of the most reliably tested judgements in the exam, because there is no single number that ends a resuscitation. Stopping a resuscitation that isn’t working is a team decision, taken by the leader, that weighs the whole picture rather than one variable. The validated termination-of-resuscitation rules are prehospital tools — RCUK is explicit that a TOR rule should not be used as the sole basis for stopping an in-hospital attempt.

Features that point towards stopping (each reduces the likelihood of a good neurological outcome):

• Unwitnessed arrest, with no bystander CPR.
• An initial non-shockable rhythm (asystole / PEA).
• No signs of life at any point, and no ROSC despite prolonged, good-quality ALS with reversible causes addressed.
• A persistently very low end-tidal CO₂ — below ~1.3 kPa (10 mmHg) after 20 minutes in an intubated patient receiving good compressions — but never used on its own to call it.
• Sonographic cardiac standstill — no cardiac activity on point-of-care echo, a strong predictor of failure to achieve ROSC — but again not the sole reason to stop, and never at the cost of interrupting compressions.
• Significant frailty and comorbidity.

If a question asks for the single most important factor, the safest answer is usually that no single factor is decisive — the decision integrates the whole picture. The strongest individual pointers, though, are an unwitnessed arrest with no bystander CPR, an initial non-shockable rhythm, and no ROSC or signs of life after prolonged ALS, supported intra-arrest by a flat ETCO₂ and cardiac standstill.

Features that demand you keep going — the deliberate exceptions the exam loves:

• Hypothermia — not dead until warm and dead.
• Poisoning — levels fall with time; resuscitate for longer.
• Thrombolysis given for PE — continue CPR for 60–90 minutes.
• A young patient with a reversible cause, or intermittent ROSC / signs of life during CPR.

Standard ALS is the assumed baseline. The consultant answer is how the algorithm bends — and the exact dose, threshold or device that bends it. If you can attach the right modification to each cause, recall the handful of numbers (no adrenaline below 30°C; CPR 60–90 minutes after thrombolysis for PE; IM adrenaline 500 mcg; deliver by five minutes; salbutamol then dialysis then ECPR for hyperkalaemia), and make the judgement on when to escalate and when to stop, you have the topic.

For the bigger map of what the exam prioritises, see our FRCEM Final high-yield topics for 2026, and the companion ECG guide for the rhythms behind refractory arrest and electrical storm.

Sources: Resuscitation Council UK 2025 Special Circumstances and Adult Advanced Life Support Guidelines; RCEMLearning, Cardiac Arrest — the ALS algorithm and beyond; LITFL, Electrical Storm; NICE NG133 (eclampsia); AAGBI guidance (malignant hyperthermia, local-anaesthetic toxicity). Drug doses are a high-yield guide for revision — always confirm against the BNF and local protocols before clinical use.