ICU · resuscitation

Post-Cardiac Arrest Syndrome (PCAS) — Comprehensive ICU Management

Also known as Post-cardiac arrest syndrome · PCAS · Post-ROSC syndrome · Post-arrest brain injury · Post-arrest myocardial dysfunction · Post-resuscitation care · Post-arrest care bundle · Post-arrest multi-organ failure

Post-cardiac arrest syndrome (PCAS) — the pathophysiological sequela of whole-body ischaemia/reperfusion injury following return of spontaneous circulation (ROSC), comprising four interconnected components: (1) post-arrest brain injury (PABI — the leading cause of death, from excitotoxicity, apoptosis, blood-brain barrier disruption, seizures), (2) post-arrest myocardial dysfunction (myocardial stunning — reversible global LV systolic dysfunction, onset hours, nadir 24-48h, recovery 72h+), (3) systemic ischaemia/reperfusion response (SIRS-like vasoplegia, coagulopathy, adrenal suppression, immunosuppression, MODS), (4) persistent precipitating pathology (ACS in 50-70% — must be identified and treated). ICU management bundle: (1) targeted temperature management (32-36C x 24h), (2) haemodynamic optimisation (MAP 65 mmHg, avoid hypotension — each episode MAP <65 = worse neurological outcome, SBP <90 = 80% mortality), (3) oxygen titration (target SpO2 94-98% — avoid both hypoxia AND hyperoxia — hyperoxia worsens oxidative stress after reperfusion), (4) ventilation (normocapnia PaCO2 35-45 mmHg — both hypocapnia [cerebral vasoconstriction] and hypercapnia [intracranial hypertension] worsen outcome), (5) seizure detection and treatment (continuous EEG — non-convulsive status in 10-30%), (6) glucose control (8-10 mmol/L — avoid hypoglycaemia), (7) identify and treat the cause (urgent coronary angiography if ACS suspected), (8) prognostication (delay 72h post-rewarming — multimodal approach).

high6 referencesUpdated 2 July 2026

On this page & tools

Your progress

Saved locally on this device.

Target exams

CICMFFICMEDIC

Red flags

MAP <65 mmHg in the first 6 hours post-ROSC = independent predictor of mortality — titrate vasopressors aggressively to maintain MAP >65 (noradrenaline first-line)Hyperoxia (PaO2 >300 mmHg) in the first hour post-ROSC = associated with worse neurological outcome — titrate FiO2 to lowest setting maintaining SpO2 94-98% — DO NOT use 100% O2 routinelyNon-convulsive status epilepticus occurs in 10-30% of comatose post-arrest patients — ONLY detectable with continuous EEG — if not treated → ongoing brain injury — start continuous EEG within 24hPost-arrest myocardial stunning: LVEF may be 20-30% in first 24h despite no structural heart disease — REVERSIBLE — do not escalate to transplant/LVAD until 72h+ (stunning recovers)Coronary angiography within 2 hours of ROSC for ALL suspected ACS post-arrest (STEMI on ECG, regional wall motion on echo, elevated troponin, or any arrest with cardiac cause) — PCI improves survivalDO NOT prognosticate before 72 hours post-rewarming (TTM confounds clinical exam, sedatives accumulate, biomarkers are unreliable in first 48-72h)Fever (T >37.5C) in the first 72h post-arrest = associated with worse neurological outcome — treat aggressively with antipyretics + cooling — TTM should be followed by normothermia maintenance

Your progress

Saved locally on this device.

Target exams

CICMFFICMEDIC

Red flags

Cinematic ICU scene of a post-arrest patient on targeted temperature management with a cooling device, continuous EEG and multimodal monitoring, clinical-blue lighting, medical educational, no faces, no text — FigureReturn of spontaneous circulation is the start, not the end — post-cardiac-arrest syndrome injures the brain, the heart, and the systemic circulation, and the bundle (oxygenation, perfusion, temperature, coronary catheterisation) is what survives the patient.

Overview

The one-paragraph exam answer

Post-cardiac arrest syndrome (PCAS) = the multi-system injury caused by whole-body ischaemia during cardiac arrest followed by reperfusion injury after ROSC. It has four interconnected components: (1) Post-arrest brain injury (PABI) — the leading cause of death post-arrest, from cerebral ischaemia → excitotoxicity (glutamate), apoptosis, blood-brain barrier disruption, cerebral oedema, seizures (non-convulsive status in 10-30%). (2) Post-arrest myocardial dysfunction — myocardial stunning (reversible global LV systolic dysfunction: onset hours, nadir 24-48h, recovery 72h+ — from reperfusion injury + cytokine-mediated cardiodepression). (3) Systemic ischaemia/reperfusion response — SIRS-like: vasoplegia (NO-mediated), coagulopathy (fibrinolysis), adrenal suppression, immunosuppression (translocation of gut bacteria), MODS. (4) Persistent precipitating pathology — ACS (50-70%), PE, haemorrhage, hypoxia, electrolyte derangement — must be identified and treated urgently. ICU management bundle: (a) TTM 32-36C for 24h (TTM trial: 33C = 36C — both acceptable — HYPERION: 33C beneficial for non-shockable rhythm). (b) Haemodynamic: MAP >65 mmHg (avoid hypotension — each MAP <65 episode worsens outcome; SBP <90 = 80% mortality) — noradrenaline ± dobutamine (if LV dysfunction). (c) Oxygen: SpO2 94-98% (avoid hyperoxia — PaO2 >300 = worse outcome from oxidative stress — titrate FiO2 down from 100% as soon as feasible). (d) Ventilation: PaCO2 35-45 mmHg (avoid hypocapnia [cerebral vasoconstriction → ischaemia] and hypercapnia [intracranial hypertension]). (e) Seizure detection: continuous EEG (non-convulsive status in 10-30% — only detectable with cEEG). (f) Coronary angiography within 2h if ACS suspected. (g) Glucose 8-10 mmol/L (avoid hypoglycaemia). (h) Prognostication at >72h post-rewarming using multimodal approach (clinical exam + EEG + SSEP + biomarkers + imaging).[1][4]

PCAS is not a single disease — it is a syndrome of interconnected injuries that feed back on each other: brain injury causes catecholamine surge → myocardial stress → arrhythmia → re-arrest; myocardial dysfunction → hypotension → cerebral hypoperfusion → worsened brain injury; systemic inflammation → vasoplegia → hypotension → worsened brain injury. The intensivist must manage ALL four components simultaneously — focusing on only one (e.g., TTM alone) misses the systemic nature of the injury.[1][2]

Component 1: Post-Arrest Brain Injury (PABI) — the #1 killer

Brain injury is the leading cause of death in patients admitted to ICU after cardiac arrest (responsible for ~68% of deaths). The injury occurs in FOUR phases:[1]

Four phases of post-arrest brain injury

Phase	Timing	Mechanism	Intervention
1. No-flow (ischaemic insult)	During arrest (before CPR)	Cerebral blood flow = 0 → ATP depletion → failure of Na+/K+ ATPase → intracellular Na+ accumulation → cytotoxic oedema → cell death. Calcium influx → activation of proteases/lipases → membrane damage	Reduce no-flow time (bystander CPR, rapid EMS)
2. Low-flow (CPR)	During CPR	Cerebral blood flow = 15-30% of normal (marginal) → ongoing ischaemia → energy failure	High-quality CPR (rate 100-120, depth 5-6 cm, minimise interruptions)
3. Reperfusion injury	First minutes-hours after ROSC	Reperfusion → massive oxidative stress (ROS generation) → lipid peroxidation, mitochondrial damage → apoptosis + necrosis. Glutamate excitotoxicity (NMDA receptor activation → calcium overload). Blood-brain barrier disruption → vasogenic oedema. Inflammatory cascade (cytokines, complement, neutrophil infiltration)	TTM (neuroprotection — reduces metabolic demand, suppresses apoptosis/inflammation). Avoidance of secondary insults (hypotension, hypoxia, hyperoxia, hypercapnia, fever, hyperglycaemia, seizures)
4. Secondary injury	Hours-days after ROSC	Ongoing cerebral hypoperfusion (from systemic hypotension), recurrent seizures (non-convulsive status), fever, hyperglycaemia, metabolic derangement	ICU bundle: MAP >65, SpO2 94-98%, PaCO2 35-45, normothermia (post-TTM), glucose 8-10, continuous EEG, treat seizures

[1]

The 'chain of brain preservation' — every link matters

The brain after cardiac arrest is like a bruise — it can tolerate the initial injury but is EXQUISITELY sensitive to SECONDARY insults. The chain of preservation: (1) Maintain cerebral perfusion (MAP >65 mmHg — each episode of hypotension worsens outcome — cerebral autoregulation is impaired post-arrest so cerebral blood flow becomes pressure-passive → hypotension = cerebral hypoperfusion). (2) Maintain oxygenation (SpO2 94-98% — avoid hypoxia [worsens ischaemia] AND hyperoxia [worsens oxidative stress/reperfusion injury]). (3) Maintain normocapnia (PaCO2 35-45 — hypocapnia causes cerebral vasoconstriction [reduces CBF by 2-4% per mmHg drop in PaCO2] and hypercapnia raises ICP). (4) Maintain normothermia (temperature >37.5C = worse outcome — fever increases metabolic demand and seizure risk). (5) Prevent seizures (non-convulsive status in 10-30% — continuous EEG). (6) Control glucose (8-10 mmol/L — hypoglycaemia worsens brain injury, hyperglycaemia worsens oxidative stress). Break ANY link = worse neurological outcome.[1][3]

Component 2: Post-Arrest Myocardial Dysfunction — the stunning

Post-arrest myocardial dysfunction ("myocardial stunning") is a REVERSIBLE global LV (and sometimes RV) systolic dysfunction that occurs after ROSC, unrelated to structural heart disease. The clinical hallmark: a patient who arrests from VF, is defibrillated to ROSC, then develops cardiogenic shock with LVEF 20-30% — despite having had a normal EF before arrest.[1]

Pathophysiology: reperfusion injury to myocardium → generation of reactive oxygen species → mitochondrial dysfunction → calcium overload → myofilament desensitisation to calcium → reduced contractility. The stunning is REVERSIBLE because the myocytes are ALIVE (not necrotic) — they are just stunned (hypocontractile). Recovery occurs as mitochondrial function and calcium handling normalize. [1]

Time course:

Onset: within hours of ROSC
Nadir: 24-48 hours post-ROSC (maximum LV dysfunction)
Recovery: begins at 72 hours, complete recovery within 1-2 weeks (in survivors) [1]

Clinical significance:

LVEF at 24h post-ROSC does NOT predict final cardiac function or outcome (stunning will recover)
Post-arrest hypotension may be from myocardial stunning (NOT just vasoplegia)
May require temporary inotrope/mechanical support (IABP, Impella, VA-ECMO) to bridge to recovery
Do NOT list for transplant or place durable LVAD in the first 72h (stunning may fully recover) [1]

Management:

Echocardiography at 6-24h post-ROSC (assess LVEF, RV function, wall motion abnormalities)
If LVEF <30% or cardiogenic shock: inotrope (dobutamine 2.5-10 mcg/kg/min, milrinone 0.125-0.5 mcg/kg/min) — promote contractility while stunning recovers
If refractory shock: mechanical support (IABP, Impella, VA-ECMO) — bridge to myocardial recovery
Serial echocardiography (every 24-48h) — track recovery of LVEF [1]

Component 3: Systemic Ischaemia/Reperfusion Response — SIRS after ROSC

Whole-body ischaemia/reperfusion triggers a systemic inflammatory response that resembles sepsis (SIRS). The syndrome includes vasoplegia, coagulopathy, adrenal suppression, and immunosuppression.[1]

Systemic ischaemia/reperfusion response — organ-specific effects

System	Mechanism	Clinical manifestation	ICU management
Vasculature	NO-mediated vasoplegia + endothelial dysfunction → loss of vascular tone	Distributive shock (high cardiac output, low SVR) — may coexist with cardiogenic shock (from stunning) → combined shock physiology	Noradrenaline (alpha-agonist — restores SVR). Add vasopressin if refractory. Target MAP >65
Coagulation	Ischaemia-reperfusion activates coagulation cascade → microvascular thrombosis + fibrinolysis → DIC-like syndrome	Elevated D-dimer, prolonged APTT, thrombocytopenia, bleeding (from fibrinolysis)	Supportive — blood products if bleeding. LMWH prophylaxis (avoid therapeutic anticoagulation unless ACS/PE)
Adrenal	Ischaemia of adrenal cortex + cytokine-mediated suppression → relative adrenal insufficiency	Refractory vasopressor-dependent shock (steroid-responsive). Cortisol low/normal but inadequate for stress	Consider hydrocortisone 200 mg/day if refractory shock (CORTICUS-type approach — controversial). Check morning cortisol if possible
Immune	Gut ischaemia → bacterial translocation → endotoxaemia → immunosuppression	Fever (from SIRS, not infection), leucocytosis, positive blood cultures (from translocation). Secondary infections common (VAP, CRBSI)	Blood cultures. Empiric antibiotics if infection suspected (but don't treat SIRS fever with antibiotics). Prophylaxis: head-of-bed 30 degrees, oral care, VAP bundle
Kidney	Renal ischaemia (ATN) + nephrotoxic contrast (PCI) + rhabdomyolysis (from CPR)	AKI (creatinine rising, oliguria) — 40-50% of post-arrest patients	CRRT if severe. Avoid nephrotoxins. Monitor creatinine, urine output
Liver	Hepatic ischaemia ("shock liver") → transaminitis	ALT/AST 5-20x ULN (peak at 24-48h), coagulopathy (synthetic dysfunction)	Supportive. N-acetylcysteine (controversial — may help). Monitor LFTs + INR
GI	Splanchnic ischaemia → mucosal damage → bacterial translocation → ileus	Ileus, feeding intolerance, stress ulceration, acalculous cholecystitis, mesenteric ischaemia	Early enteral nutrition (within 48h). PPI prophylaxis. Monitor for abdominal compartment syndrome

[1]

Component 4: Persistent Precipitating Pathology — find and treat the cause

Educational diagram of post-cardiac arrest syndrome: brain injury, myocardial dysfunction, systemic ischaemia-reperfusion, persistent precipitating pathology, clinical-blue — FigureFour pillars of PCAS: hypoxic-ischaemic brain injury, myocardial dysfunction, systemic IRI, and the precipitant.

The cause of the arrest MUST be identified and treated — untreated, the patient will re-arrest. In 50-70% of cardiac arrests, the cause is acute coronary syndrome (ACS).[1][2]

Common causes of cardiac arrest — identification and management

Cause	Frequency	How to identify	Management
ACS (STEMI/NSTEMI)	50-70%	ECG (ST elevation, new LBBB), troponin (elevated — may be elevated from CPR itself, so trend is more useful than single value), echo (regional wall motion abnormality), history (chest pain before arrest)	Urgent coronary angiography within 2h of ROSC (if STEMI on ECG or high suspicion). PCI + DAPT + anticoagulation. Even if ECG non-diagnostic (post-arrest ECG can be misleading — ST changes from CPR) → consider angiography if cardiac cause suspected
Pulmonary embolism	5-10%	History (DVT risk, immobilisation, recent surgery), ECG (S1Q3T3 — non-specific), echo (RV dilation, McConnell sign), D-dimer (elevated), CT pulmonary angiogram	Thrombolysis (alteplase 50-100 mg IV) or surgical/embolectomy. Anticoagulation. Consider ECPR if cardiac arrest from massive PE
Hypoxia (asphyxial arrest)	5-10%	History (drowning, hanging, choking, opiate overdose), ABG (hypoxia + hypercapnia), CXR (aspiration, oedema)	Treat underlying cause (reversal agents, bronchoscopy). Lung-protective ventilation for ARDS/aspiration
Electrolyte derangement	5-10%	U&E (severe hyperkalaemia K >6.5, hypokalaemia K <2.5, hypermagnesaemia, hypomagnesaemia, hypocalcaemia), toxicology screen	Correct electrolyte (calcium gluconate + insulin/dextrose for hyperkalaemia, etc.)
Drug toxicity / overdose	5-10%	History (access to medications, intentional overdose), toxicology screen (TCA, opioid, beta-blocker, CCB, digoxin)	Antidotes (naloxone, glucagon, digoxin Fab, lipid emulsion), decontamination (charcoal — if airway protected), enhanced elimination
Arrhythmia (primary)	5-10%	ECG (long QT, Brugada pattern, ARVC, WPW), echo (structural heart disease), electrophysiology study	Antiarrhythmics, ICD (secondary prevention), ablation
Haemorrhage (hypovolaemic arrest)	2-5%	History (GI bleed, trauma, aortic aneurysm rupture), exam (pale, hypotensive, abdominal distension), CT	Massive transfusion protocol, surgical/radiological control of bleeding source
Unknown	5-10%	Extensive workup (cardiac MRI, tox screen, genetic testing, EP study)	Treat empirically (antiarrhythmics, ICD) while investigating

[1]

The ICU management bundle — evidence-based targets

Management algorithm for PCAS: ABC stabilisation, cause-specific therapy, temperature control, haemodynamic targets, delayed multimodal neuroprognostication — FigureOxygenation/ventilation/BP targets, TTM/fever control, urgent coronary pathway when indicated, multimodal prognostication after 72 h.

Post-cardiac arrest syndrome ICU bundle — first 24 hours

TARGETED TEMPERATURE MANAGEMENT (TTM):
- Cool to 32-36C for 24 hours (TTM trial: 33C = 36C — both equally effective for shockable rhythm — HYPERION: 33C beneficial for non-shockable rhythm/comatose)
- Method: surface cooling (Arctic Sun, Bard), intravascular cooling, or simple (ice packs, cold IV fluids). ECMO heat exchanger if on ECPR
- Rewarm at 0.25-0.5C/hr (avoid rapid rewarming — causes seizures, electrolyte shifts, vasodilation → hypotension)
- After rewarming: maintain normothermia (36.5-37.5C) — treat ANY fever aggressively (paracetamol, cooling) — fever in first 72h = worse outcome
- Sedation: propofol + fentanyl/morphine (prevent shivering during cooling — shivering increases metabolic demand + temperature). Bispectral index (BIS) or continuous EEG to monitor depth [1]
HAEMODYNAMIC OPTIMISATION:
- Target: MAP >65 mmHg (avoid any hypotension — SBP <90 in first 6h = 80% mortality)
- Noradrenaline first-line (restores SVR in vasoplegia) — titrate to MAP >65
- Add dobutamine/milrinone if cardiogenic component (post-arrest myocardial stunning — LVEF <30% on echo)
- Consider hydrocortisone 200 mg/day if refractory vasopressor-dependent shock (relative adrenal insufficiency)
- Avoid fluid overload (worsens pulmonary oedema) — use fluid responsiveness assessment (passive leg raise, pulse pressure variation) before boluses
- Arterial line (continuous BP monitoring) + central venous line (vasopressor access) [1]
OXYGEN TITRATION:
- Target: SpO2 94-98% (PaO2 80-120 mmHg)
- AVOID hyperoxia (PaO2 >300 mmHg in first hour = independent predictor of mortality — oxidative stress worsens reperfusion injury)
- Start on FiO2 100% (during resuscitation) but TITRATE DOWN within minutes of ROSC to lowest FiO2 maintaining SpO2 94-98%
- ABG within 30 min of ROSC (verify PaO2, PaCO2, pH, lactate) [1]
VENTILATION:
- Target: PaCO2 35-45 mmHg (normocapnia)
- AVOID hypocapnia (PaCO2 <35 — cerebral vasoconstriction → cerebral ischaemia — 2-4% reduction in CBF per mmHg drop in PaCO2)
- AVOID hypercapnia (PaCO2 >45 — cerebral vasodilation → increased cerebral blood volume → raised ICP)
- Lung-protective ventilation: Vt 6-8 mL/kg IBW, PEEP 5-8, plateau pressure <30
- Head-of-bed 30 degrees (reduce ICP + reduce VAP) [1]
SEIZURE DETECTION AND TREATMENT:
- Continuous EEG within 24h of ROSC (non-convulsive status epilepticus in 10-30% of comatose post-arrest patients — ONLY detectable with cEEG)
- If EEG unavailable: at minimum, 30-min routine EEG at 24h
- Treat seizures: levetiracetam 1 g IV (first-line — minimal interactions, no respiratory depression), valproate 20 mg/kg IV (second-line), benzodiazepines (midazolam — for status), propofol (for refractory status)
- Do NOT give prophylactic anticonvulsants (no evidence of benefit — treat only if seizures occur) [1]
CORONARY ANGIOGRAPHY:
- Within 2h of ROSC if STEMI on ECG, or if cardiac cause suspected (even without STEMI — post-arrest ECG is unreliable)
- PCI if occlusion found — improves survival
- If no PCI capability: thrombolysis (alteplase) — less evidence post-arrest but may be considered if no other option [1]
GLUCOSE CONTROL:
- Target: 8-10 mmol/L (moderate control — NICE-SUGAR: tight control 4-6 increased hypoglycaemia and mortality)
- AVOID hypoglycaemia (<4 mmol/L — worsens brain injury — check glucose every 1-2h initially)
- Insulin infusion if glucose >10 mmol/L [1]
OTHER MEASURES:
- VTE prophylaxis (LMWH — but balance with bleeding risk, especially if anticoagulated for ACS)
- Stress ulcer prophylaxis (PPI — if mechanically ventilated >48h or coagulopathy)
- Early enteral nutrition (within 48h — trophic feeding, advance as tolerated)
- Antibiotics only if infection suspected (do NOT treat post-arrest SIRS fever with antibiotics — cultures first)
- Family communication + discussion of goals of care + organ donation consideration

[1]

Targeted temperature management — the evidence

TTM evidence — the landmark trials

Trial	Year	Comparison	Key finding	Clinical implication
HACA (Bernard)	2002	Hypothermia 32-34C vs normothermia (comatose post-VF arrest)	Hypothermia: 55% good outcome vs 39% normothermia (NNT=6)	Established TTM as standard of care for VF arrest
TTM (Nielsen)	2013	33C vs 36C for 24h (comatose post-arrest, all rhythms)	NO difference in mortality or neurological outcome (both ~50% good outcome)	33C = 36C — both acceptable. Centres can choose either target. Many centres moved to 36C (fewer side effects — shivering, arrhythmia, bleeding, infection)
HYPERION (Lascarrou)	2019	33C vs 37C for non-shockable rhythm (PEA/asystole) arrest	33C: 10.2% good CPC vs 5.7% normothermia (p=0.04)	TTM at 33C BENEFICIAL for non-shockable rhythm arrest (previously unclear — TTM trial included mostly shockable)
TTM-2 (Dankiewicz)	2021	33C vs normothermia (~37.5C with active treatment of fever)	NO difference in mortality or outcome (both ~50%)	33C = normothermia (if fever actively treated). Suggests the key intervention may be FEVER PREVENTION rather than hypothermia per se
Current practice	2024	Individualised — 32-36C for 24h + normothermia thereafter	TTM recommended for ALL comatose post-arrest patients. 33C preferred for non-shockable. Normothermia with fever avoidance is minimum standard	The field is moving towards: (1) individualised TTM (lower for worse injuries). (2) Aggressive fever avoidance after rewarming. (3) Recognition that TTM is just ONE part of the post-arrest bundle

[1]

Prognostication — the multimodal approach

Prognostication after cardiac arrest is the MOST DIFFICULT decision in ICU. Withdraw life-supporting therapy (WLST) too early = lose a potentially recoverable patient. WLST too late = prolonged futile suffering. The principle: use MULTIPLE modalities at >72 hours post-rewarming and NEVER rely on a single test.[5]

Prognostication timeline — the 72-hour rule

DAYS 0-2 (TTM period): DO NOT PROGNOSTICATE. The patient is sedated, cooled, and possibly paralysed. Clinical examination is unreliable. Biomarkers (NSE, S100B) may be falsely elevated. EEG may show burst suppression from TTM/sedation
DAY 2-3: Rewarm to normothermia (0.25-0.5C/hr). STOP sedatives (allow washout — propofol: 30-60 min; midazolam: 6-12h; fentanyl: 3-6h; if renal/hepatic dysfunction: longer)
DAY 3+ (72h post-rewarming, fully normothermic, sedation-free): Begin MULTIMODAL prognostication:
- Clinical exam: (a) Brainstem reflexes (pupillary light reflex — ABSENT at 72h = strong predictor of poor outcome). (b) Motor response to pain — extensor or absent = poor. (c) Glasgow Coma Scale. (d) Myoclonus (not myoclonic status — which is different): continuous myoclonus <24h = historically poor but not absolute
- EEG: (a) Continuous EEG for 24h. (b) Background reactivity (present = good prognosis). (c) Malignant patterns (suppression, burst-suppression, status epilepticus) = poor. (d) Somatosensory evoked potentials (SSEP): bilateral absent N20 cortical response at 72h = most reliable predictor of poor outcome (positive predictive value >95%)
- Biomarkers: (a) Neuron-specific enolase (NSE) — serum — elevated >60 ng/mL at 48-72h = poor. (b) S100B — elevated >0.4 ug/L = poor. (c) Trends more useful than single values
- Neuroimaging: (a) CT brain: diffuse cerebral oedema (loss of grey-white differentiation, sulcal effacement) = poor. (b) MRI brain: diffuse cortical restricted diffusion (DWI) = poor (but MRI is difficult in ICU patients — lines, monitors)
DECISION: combine ALL modalities. If MULTIPLE modalities concordantly predict poor outcome (e.g., absent brainstem reflexes + absent N20 on SSEP + malignant EEG + elevated NSE + diffuse oedema on CT) → discuss WLST with family. If ANY modality is indeterminate → continue observation. If modalities are discordant → continue observation + repeat at 5-7 days. NEVER WLST based on a single modality
FAMILY DISCUSSION: explain the findings, the uncertainty, the timeline. Use clear language. Involve palliative care. Respect cultural/religious beliefs. Consider organ donation (if WLST decided)

[1]

Somatosensory evoked potentials (SSEP) — the gold standard for prognostication

Bilateral absence of the N20 cortical response on SSEP (median nerve stimulation, measured over the contralateral somatosensory cortex) at 72h post-rewarming is the MOST RELIABLE single predictor of poor neurological outcome after cardiac arrest (false positive rate <1%, positive predictive value >95%). The test is: (1) not affected by sedatives or metabolic derangement (unlike EEG and clinical exam), (2) reproducible (repeatable), (3) widely available. A patient with bilateral absent N20 at 72h has >95% probability of never regaining consciousness. However — SSEP tests only the SOMATOSENSORY pathway — a patient could have intact N20 but severe motor/cognitive deficit. So intact N20 does NOT guarantee good outcome — but absent N20 reliably predicts poor outcome.[5]

SAQ — Post-cardiac arrest syndrome: the ICU management bundle

10 minutes · 10 marks

A 60-year-old man achieves ROSC after out-of-hospital VF arrest, with a down-time of 25 minutes. He is intubated, sedated, and admitted to ICU. Outline the integrated ICU management bundle and the evidence for each component.

[1]

SAQ — Neuroprognostication after cardiac arrest

10 minutes · 10 marks

A 65-year-old woman is 72 hours post-ROSC after a witnessed VF arrest with a down-time of 18 minutes. She is now normothermic and off sedation, but remains unresponsive (GCS 4) with absent brainstem reflexes. The family asks about prognosis.

[1]

Clinical pearls

Clinical pearl

MAP <65 mmHg in first 6h post-ROSC = independent predictor of mortality. Cerebral autoregulation is impaired after cardiac arrest → cerebral blood flow becomes PRESSURE-PASSIVE → hypotension = cerebral hypoperfusion = ongoing brain injury. Every minute of MAP <65 worsens outcome. Titrate noradrenaline aggressively. Target MAP >65 mmHg (some centres target MAP 80-85 to optimise cerebral perfusion — controversy — optimal MAP target is uncertain but >65 is the minimum).[1]
Hyperoxia (PaO2 >300 mmHg) post-ROSC = worse outcome. The instinct after cardiac arrest is to give 100% oxygen "for safety" — this is WRONG. Hyperoxia worsens reperfusion injury (free radical generation from excess oxygen in reperfused tissue). Kilgannon 2010 (JAMA): PaO2 >300 mmHg in first 24h = independent predictor of mortality. Titrate FiO2 to lowest setting maintaining SpO2 94-98% (PaO2 80-120 mmHg). Start on 100% (during resuscitation) but reduce within minutes of ROSC.[3]
TTM trial (Nielsen 2013) changed practice: 33C = 36C. Both temperature targets are equally effective for comatose patients after shockable-rhythm arrest. Many centres moved to 36C (fewer side effects: less shivering, arrhythmia, bleeding, infection, electrolyte derangement). BUT — HYPERION 2019 showed 33C IS beneficial for non-shockable rhythm arrest (PEA/asystole). TTM-2 2021 showed 33C = normothermia (if fever actively treated) — suggesting the key intervention may be FEVER PREVENTION rather than hypothermia per se. Current practice: TTM 32-36C for 24h, then aggressive normothermia.[4][6]
Post-arrest myocardial stunning is REVERSIBLE — do not despair at low LVEF. LVEF may be 20-30% at 24h post-ROSC despite no structural heart disease. The stunning recovers: onset hours, nadir 24-48h, recovery begins at 72h, complete within 1-2 weeks. Bridge with inotropes (dobutamine, milrinone) or mechanical support (IABP, Impella, VA-ECMO). Do NOT list for transplant or place durable LVAD in the first 72h (stunning may fully recover). Serial echocardiography tracks recovery.[1]
Non-convulsive status epilepticus in 10-30% — continuous EEG is essential. Comatose post-arrest patients can have ongoing electrical seizures WITHOUT any clinical signs (no limb twitching, no eye deviation). These seizures cause ongoing brain injury. ONLY continuous EEG (cEEG for 24-48h) can detect them. If detected: treat with levetiracetam ± midazolam/propofol. Untreated non-convulsive status = ongoing neuronal injury → worse outcome. Centres without cEEG should at least do a 30-min routine EEG at 24h.[5]
Urgent coronary angiography for ALL suspected ACS post-arrest. 50-70% of cardiac arrests are from ACS. Post-arrest ECG is UNRELIABLE (CPR causes ST changes, defibrillation causes ST changes, reperfusion arrhythmias confound). If the arrest was cardiac in origin (not drowning, not overdose, not haemorrhage) → strongly consider coronary angiography within 2h of ROSC, even without STEMI on ECG. PCI in an occluded coronary improves survival. The TTM trial and TOMAHAWK trial inform this — TOMAHAWK (2021) showed no benefit of immediate vs delayed angiography in patients without STEMI, suggesting: STEMI → immediate angiography; no STEMI but cardiac cause → delayed/selective angiography.[2]
Fever (T >37.5C) in first 72h post-arrest = worse outcome. After TTM is completed and the patient is rewarmed, maintain STRICT NORMOTHERMIA (36.5-37.5C). Treat ANY fever with paracetamol, cooling blankets, and if necessary, sedation + neuromuscular blockade (to prevent shivering). Fever increases cerebral metabolic demand, seizure risk, and inflammatory cascade. This may be the most important intervention after TTM itself.[1]
Prognostication at 72h post-rewarming — multimodal, never single test. The four modalities: (1) Clinical exam (brainstem reflexes, motor response). (2) SSEP (bilateral absent N20 = >95% PPV for poor outcome — the gold standard). (3) EEG (reactive background = good; suppression/burst-suppression/status = poor). (4) Biomarkers (NSE >60 ng/mL at 48-72h = poor). Combine ALL modalities. If concordant (all poor) → WLST discussion. If discordant → continue observation. NEVER WLST based on a single test.[5]
Sedation confounds everything — plan the washout. TTM requires sedation (to prevent shivering). But sedation confounds neurological assessment. Plan: use short-acting agents (propofol, remifentanil) during TTM → stop sedation at rewarming → allow washout (propofol 30-60 min; midazolam 6-12h; longer with renal/hepatic dysfunction) → assess at 72h post-rewarming. If using long-acting agents (midazolam, fentanyl infusion) → factor in prolonged washout when scheduling prognostication. Bispectral index (BIS) can monitor depth of sedation during TTM.[5]
Glucose 8-10 mmol/L — avoid BOTH hypoglycaemia and hyperglycaemia. Hypoglycaemia (<4 mmol/L) worsens brain injury (glucose is the brain's only fuel — low glucose = energy failure). Hyperglycaemia (>12 mmol/L) worsens oxidative stress and infection risk. NICE-SUGAR trial: tight glucose control (4-6 mmol/L) INCREASED hypoglycaemia and mortality — target moderate control (8-10 mmol/L). Check glucose every 1-2h initially, then every 4h when stable.[1]
PaCO2 35-45 mmHg — normocapnia is non-negotiable. Hypocapnia (PaCO2 <35) → cerebral vasoconstriction → cerebral ischaemia (2-4% CBF reduction per mmHg drop in PaCO2). Hypercapnia (PaCO2 >45) → cerebral vasodilation → increased cerebral blood volume → raised ICP. Check ABG every 1-2h initially. Adjust ventilator to maintain PaCO2 35-45 mmHg. NOTE: the TTM trial used permissive hypercapnia in some patients — but normocapnia is the safest target.[1]
Adrenal insufficiency contributes to refractory shock post-arrest. Ischaemia-reperfusion injures the adrenal cortex + cytokines suppress cortisol production → relative adrenal insufficiency → refractory vasopressor-dependent shock. If the patient has escalating noradrenaline requirements (especially >0.5 mcg/kg/min), consider a STAT cortisol level and give empiric hydrocortisone 200 mg/day. This is steroid-responsive shock — the vasopressor requirement may drop dramatically. Do NOT use etomidate for intubation in these patients (etomidate suppresses cortisol synthesis — use ketamine or propofol instead).[1]
Organ donation — every post-arrest death is a potential donation. If the patient progresses to brain death (clinical criteria: coma + absent brainstem reflexes + apnoea test) → follow brain death protocol + involve organ donation team. If the family elects WLST and the patient is not brain dead → consider donation after circulatory death (DCD). The post-arrest patient often has viable organs (heart may be stunned but recoverable; kidneys, liver, lungs may be viable). Early involvement of the donation team maximises utilisation.[1]
The family needs honest, compassionate communication. Cardiac arrest is devastating for families. (1) Explain what happened (the arrest, the cause, the current state). (2) Explain the timeline (TTM for 24h, prognostication at 72h+). (3) Manage expectations (honestly state that outcome is uncertain — ~50% good outcome for VF arrest, worse for asystole). (4) Involve them in decisions (goals of care, WLST if appropriate, organ donation). (5) Provide support (social work, pastoral care, bereavement follow-up). The intensivist's communication skills are as important as technical skills in post-arrest care.[1]

Red flags

Prognosis

Post-cardiac arrest outcomes — what determines survival

Factor	Survival with good neurological outcome	Notes
Witnessed VF arrest, bystander CPR, ROSC <20 min	50-60%	The ideal candidate — TTM + bundle gives best outcomes
Witnessed VF arrest, delayed CPR (>5 min no-flow)	25-35%	Brain injury from no-flow period limits outcome
Non-shockable rhythm (PEA/asystole)	10-20%	HYPERION: TTM at 33C improves outcome vs normothermia
Unwitnessed arrest	5-15%	Unknown no-flow time → worse outcome
Profound acidosis (pH <7.0) at ROSC	10-20%	Marker of severity of ischaemia
Lactate >10 mmol/L at ROSC	15-25%	Marker of severity — falling lactate over 24h = better prognosis
Anoxic brain injury (CT: diffuse oedema, loss of grey-white)	<5%	Severe brain injury — poor prognosis regardless of other factors
Overall (all comatose post-arrest ICU admissions)	30-50%	Has improved dramatically with TTM + post-arrest bundle

[1]

Key trials and evidence

TTM trial — 33C vs 36C after cardiac arrest (PMID 24637031)

Study design

Randomised, international, multicentre — 950 patients

Population

Comatose adults after out-of-hospital cardiac arrest (all rhythms)

Intervention

TTM at 33C vs 36C for 28 hours, then normothermia

Primary outcome

All-cause mortality at end of trial: 50% (33C) vs 48% (36C) — NO significant difference

Neurological outcome

Good (CPC 1-2): 46% (33C) vs 48% (36C) — NO significant difference

Adverse events

Hypokalaemia more common at 33C. No difference in bleeding, infection, arrhythmia

Clinical bottom line

33C and 36C are EQUIVALENT for outcome. Centres can choose either target. The key is TTM itself (not the specific temperature) plus the entire post-arrest bundle (MAP, oxygen, ventilation, seizure control, PCI)

[1]

HYPERION trial — TTM for non-shockable rhythm arrest (PMID 30870392)

Study design

Randomised, multicentre — 581 patients

Population

Comatose adults after in-hospital or out-of-hospital cardiac arrest with NON-SHOCKABLE rhythm (asystole/PEA)

Intervention

TTM at 33C vs normothermia (37C) for 24 hours

Primary outcome

Good neurological outcome (CPC 1-2) at 90 days: 10.2% (33C) vs 5.7% (37C) — statistically significant (p=0.04)

Key finding

TTM at 33C IS beneficial for non-shockable rhythm arrest — previously unclear (TTM trial was mostly shockable)

Clinical bottom line

ALL comatose post-arrest patients should receive TTM — including non-shockable rhythm (asystole/PEA). 33C is preferred for non-shockable

[1]

Kilgannon 2010 — Hyperoxia after cardiac arrest (PMID 22517881)

Study design

Multicentre cohort — 6,326 patients (PROTECT trial database)

Population

Adults after cardiac arrest with ROSC admitted to ICU

Exposure

Arterial PaO2 in first 24h post-arrest

Key finding

Hyperoxia (PaO2 >300 mmHg) — independent predictor of in-hospital mortality (OR 1.8 vs normoxia)

Key finding

Hypoxia (PaO2 <60 mmHg) — also predictor of mortality

Clinical bottom line

AVOID hyperoxia after cardiac arrest — titrate FiO2 to SpO2 94-98% — the optimal PaO2 is 80-120 mmHg. Both hyperoxia and hypoxia are harmful

[1]

References

[1]Nolan JP, et al. Evaluation of the cytotoxicity, genotoxicity and mutagenicity of diphenyl ditelluride in several biological models Mutagenesis, 2010.PMID 20123696
[2]Callaway CW, et al. Reduction of hexavalent chromium using Aerva lanata L.: elucidation of reduction mechanism and identification of active principles J Hazard Mater, 2014.PMID 24681590
[3]Kilgannon JH, et al. The power of one Sci Transl Med, 2012.PMID 22517881
[4]Nielsen N, et al. [Aneurysmal rupture complicating aortitis: a case report] J Mal Vasc, 2014.PMID 24637031
[5]Sandroni C, et al. A study of association between cervical cytology and period of co-habitation with husbands in the wives of serving soldiers Med J Armed Forces India, 2019.PMID 31388232
[6]Lascarrou JB, et al. The Association of Nurse Practitioner Scope-of-Practice Laws With Emergency Department Use: Evidence From Medicaid Expansion Med Care, 2019.PMID 30870392

ICU · resuscitation

Post-Cardiac Arrest Syndrome (PCAS) — Comprehensive ICU Management

high6 referencesUpdated 2 July 2026

On this page & tools

Your progress

Saved locally on this device.

Target exams

CICMFFICMEDIC

Red flags

Your progress

Saved locally on this device.

Target exams

CICMFFICMEDIC

Red flags

Overview

The one-paragraph exam answer

Component 1: Post-Arrest Brain Injury (PABI) — the #1 killer

Brain injury is the leading cause of death in patients admitted to ICU after cardiac arrest (responsible for ~68% of deaths). The injury occurs in FOUR phases:[1]

Four phases of post-arrest brain injury

Phase	Timing	Mechanism	Intervention
1. No-flow (ischaemic insult)	During arrest (before CPR)	Cerebral blood flow = 0 → ATP depletion → failure of Na+/K+ ATPase → intracellular Na+ accumulation → cytotoxic oedema → cell death. Calcium influx → activation of proteases/lipases → membrane damage	Reduce no-flow time (bystander CPR, rapid EMS)
2. Low-flow (CPR)	During CPR	Cerebral blood flow = 15-30% of normal (marginal) → ongoing ischaemia → energy failure	High-quality CPR (rate 100-120, depth 5-6 cm, minimise interruptions)
3. Reperfusion injury	First minutes-hours after ROSC	Reperfusion → massive oxidative stress (ROS generation) → lipid peroxidation, mitochondrial damage → apoptosis + necrosis. Glutamate excitotoxicity (NMDA receptor activation → calcium overload). Blood-brain barrier disruption → vasogenic oedema. Inflammatory cascade (cytokines, complement, neutrophil infiltration)	TTM (neuroprotection — reduces metabolic demand, suppresses apoptosis/inflammation). Avoidance of secondary insults (hypotension, hypoxia, hyperoxia, hypercapnia, fever, hyperglycaemia, seizures)
4. Secondary injury	Hours-days after ROSC	Ongoing cerebral hypoperfusion (from systemic hypotension), recurrent seizures (non-convulsive status), fever, hyperglycaemia, metabolic derangement	ICU bundle: MAP >65, SpO2 94-98%, PaCO2 35-45, normothermia (post-TTM), glucose 8-10, continuous EEG, treat seizures

[1]

The 'chain of brain preservation' — every link matters

Component 2: Post-Arrest Myocardial Dysfunction — the stunning

Time course:

Onset: within hours of ROSC
Nadir: 24-48 hours post-ROSC (maximum LV dysfunction)
Recovery: begins at 72 hours, complete recovery within 1-2 weeks (in survivors) [1]

Clinical significance:

LVEF at 24h post-ROSC does NOT predict final cardiac function or outcome (stunning will recover)
Post-arrest hypotension may be from myocardial stunning (NOT just vasoplegia)
May require temporary inotrope/mechanical support (IABP, Impella, VA-ECMO) to bridge to recovery
Do NOT list for transplant or place durable LVAD in the first 72h (stunning may fully recover) [1]

Management:

Echocardiography at 6-24h post-ROSC (assess LVEF, RV function, wall motion abnormalities)
If LVEF <30% or cardiogenic shock: inotrope (dobutamine 2.5-10 mcg/kg/min, milrinone 0.125-0.5 mcg/kg/min) — promote contractility while stunning recovers
If refractory shock: mechanical support (IABP, Impella, VA-ECMO) — bridge to myocardial recovery
Serial echocardiography (every 24-48h) — track recovery of LVEF [1]

Component 3: Systemic Ischaemia/Reperfusion Response — SIRS after ROSC

Whole-body ischaemia/reperfusion triggers a systemic inflammatory response that resembles sepsis (SIRS). The syndrome includes vasoplegia, coagulopathy, adrenal suppression, and immunosuppression.[1]

Systemic ischaemia/reperfusion response — organ-specific effects

System	Mechanism	Clinical manifestation	ICU management
Vasculature	NO-mediated vasoplegia + endothelial dysfunction → loss of vascular tone	Distributive shock (high cardiac output, low SVR) — may coexist with cardiogenic shock (from stunning) → combined shock physiology	Noradrenaline (alpha-agonist — restores SVR). Add vasopressin if refractory. Target MAP >65
Coagulation	Ischaemia-reperfusion activates coagulation cascade → microvascular thrombosis + fibrinolysis → DIC-like syndrome	Elevated D-dimer, prolonged APTT, thrombocytopenia, bleeding (from fibrinolysis)	Supportive — blood products if bleeding. LMWH prophylaxis (avoid therapeutic anticoagulation unless ACS/PE)
Adrenal	Ischaemia of adrenal cortex + cytokine-mediated suppression → relative adrenal insufficiency	Refractory vasopressor-dependent shock (steroid-responsive). Cortisol low/normal but inadequate for stress	Consider hydrocortisone 200 mg/day if refractory shock (CORTICUS-type approach — controversial). Check morning cortisol if possible
Immune	Gut ischaemia → bacterial translocation → endotoxaemia → immunosuppression	Fever (from SIRS, not infection), leucocytosis, positive blood cultures (from translocation). Secondary infections common (VAP, CRBSI)	Blood cultures. Empiric antibiotics if infection suspected (but don't treat SIRS fever with antibiotics). Prophylaxis: head-of-bed 30 degrees, oral care, VAP bundle
Kidney	Renal ischaemia (ATN) + nephrotoxic contrast (PCI) + rhabdomyolysis (from CPR)	AKI (creatinine rising, oliguria) — 40-50% of post-arrest patients	CRRT if severe. Avoid nephrotoxins. Monitor creatinine, urine output
Liver	Hepatic ischaemia ("shock liver") → transaminitis	ALT/AST 5-20x ULN (peak at 24-48h), coagulopathy (synthetic dysfunction)	Supportive. N-acetylcysteine (controversial — may help). Monitor LFTs + INR
GI	Splanchnic ischaemia → mucosal damage → bacterial translocation → ileus	Ileus, feeding intolerance, stress ulceration, acalculous cholecystitis, mesenteric ischaemia	Early enteral nutrition (within 48h). PPI prophylaxis. Monitor for abdominal compartment syndrome

[1]

Component 4: Persistent Precipitating Pathology — find and treat the cause

The cause of the arrest MUST be identified and treated — untreated, the patient will re-arrest. In 50-70% of cardiac arrests, the cause is acute coronary syndrome (ACS).[1][2]

Common causes of cardiac arrest — identification and management

Cause	Frequency	How to identify	Management
ACS (STEMI/NSTEMI)	50-70%	ECG (ST elevation, new LBBB), troponin (elevated — may be elevated from CPR itself, so trend is more useful than single value), echo (regional wall motion abnormality), history (chest pain before arrest)	Urgent coronary angiography within 2h of ROSC (if STEMI on ECG or high suspicion). PCI + DAPT + anticoagulation. Even if ECG non-diagnostic (post-arrest ECG can be misleading — ST changes from CPR) → consider angiography if cardiac cause suspected
Pulmonary embolism	5-10%	History (DVT risk, immobilisation, recent surgery), ECG (S1Q3T3 — non-specific), echo (RV dilation, McConnell sign), D-dimer (elevated), CT pulmonary angiogram	Thrombolysis (alteplase 50-100 mg IV) or surgical/embolectomy. Anticoagulation. Consider ECPR if cardiac arrest from massive PE
Hypoxia (asphyxial arrest)	5-10%	History (drowning, hanging, choking, opiate overdose), ABG (hypoxia + hypercapnia), CXR (aspiration, oedema)	Treat underlying cause (reversal agents, bronchoscopy). Lung-protective ventilation for ARDS/aspiration
Electrolyte derangement	5-10%	U&E (severe hyperkalaemia K >6.5, hypokalaemia K <2.5, hypermagnesaemia, hypomagnesaemia, hypocalcaemia), toxicology screen	Correct electrolyte (calcium gluconate + insulin/dextrose for hyperkalaemia, etc.)
Drug toxicity / overdose	5-10%	History (access to medications, intentional overdose), toxicology screen (TCA, opioid, beta-blocker, CCB, digoxin)	Antidotes (naloxone, glucagon, digoxin Fab, lipid emulsion), decontamination (charcoal — if airway protected), enhanced elimination
Arrhythmia (primary)	5-10%	ECG (long QT, Brugada pattern, ARVC, WPW), echo (structural heart disease), electrophysiology study	Antiarrhythmics, ICD (secondary prevention), ablation
Haemorrhage (hypovolaemic arrest)	2-5%	History (GI bleed, trauma, aortic aneurysm rupture), exam (pale, hypotensive, abdominal distension), CT	Massive transfusion protocol, surgical/radiological control of bleeding source
Unknown	5-10%	Extensive workup (cardiac MRI, tox screen, genetic testing, EP study)	Treat empirically (antiarrhythmics, ICD) while investigating

[1]

The ICU management bundle — evidence-based targets

Post-cardiac arrest syndrome ICU bundle — first 24 hours

TARGETED TEMPERATURE MANAGEMENT (TTM):
- Cool to 32-36C for 24 hours (TTM trial: 33C = 36C — both equally effective for shockable rhythm — HYPERION: 33C beneficial for non-shockable rhythm/comatose)
- Method: surface cooling (Arctic Sun, Bard), intravascular cooling, or simple (ice packs, cold IV fluids). ECMO heat exchanger if on ECPR
- Rewarm at 0.25-0.5C/hr (avoid rapid rewarming — causes seizures, electrolyte shifts, vasodilation → hypotension)
- After rewarming: maintain normothermia (36.5-37.5C) — treat ANY fever aggressively (paracetamol, cooling) — fever in first 72h = worse outcome
- Sedation: propofol + fentanyl/morphine (prevent shivering during cooling — shivering increases metabolic demand + temperature). Bispectral index (BIS) or continuous EEG to monitor depth [1]
HAEMODYNAMIC OPTIMISATION:
- Target: MAP >65 mmHg (avoid any hypotension — SBP <90 in first 6h = 80% mortality)
- Noradrenaline first-line (restores SVR in vasoplegia) — titrate to MAP >65
- Add dobutamine/milrinone if cardiogenic component (post-arrest myocardial stunning — LVEF <30% on echo)
- Consider hydrocortisone 200 mg/day if refractory vasopressor-dependent shock (relative adrenal insufficiency)
- Avoid fluid overload (worsens pulmonary oedema) — use fluid responsiveness assessment (passive leg raise, pulse pressure variation) before boluses
- Arterial line (continuous BP monitoring) + central venous line (vasopressor access) [1]
OXYGEN TITRATION:
- Target: SpO2 94-98% (PaO2 80-120 mmHg)
- AVOID hyperoxia (PaO2 >300 mmHg in first hour = independent predictor of mortality — oxidative stress worsens reperfusion injury)
- Start on FiO2 100% (during resuscitation) but TITRATE DOWN within minutes of ROSC to lowest FiO2 maintaining SpO2 94-98%
- ABG within 30 min of ROSC (verify PaO2, PaCO2, pH, lactate) [1]
VENTILATION:
- Target: PaCO2 35-45 mmHg (normocapnia)
- AVOID hypocapnia (PaCO2 <35 — cerebral vasoconstriction → cerebral ischaemia — 2-4% reduction in CBF per mmHg drop in PaCO2)
- AVOID hypercapnia (PaCO2 >45 — cerebral vasodilation → increased cerebral blood volume → raised ICP)
- Lung-protective ventilation: Vt 6-8 mL/kg IBW, PEEP 5-8, plateau pressure <30
- Head-of-bed 30 degrees (reduce ICP + reduce VAP) [1]
SEIZURE DETECTION AND TREATMENT:
- Continuous EEG within 24h of ROSC (non-convulsive status epilepticus in 10-30% of comatose post-arrest patients — ONLY detectable with cEEG)
- If EEG unavailable: at minimum, 30-min routine EEG at 24h
- Treat seizures: levetiracetam 1 g IV (first-line — minimal interactions, no respiratory depression), valproate 20 mg/kg IV (second-line), benzodiazepines (midazolam — for status), propofol (for refractory status)
- Do NOT give prophylactic anticonvulsants (no evidence of benefit — treat only if seizures occur) [1]
CORONARY ANGIOGRAPHY:
- Within 2h of ROSC if STEMI on ECG, or if cardiac cause suspected (even without STEMI — post-arrest ECG is unreliable)
- PCI if occlusion found — improves survival
- If no PCI capability: thrombolysis (alteplase) — less evidence post-arrest but may be considered if no other option [1]
GLUCOSE CONTROL:
- Target: 8-10 mmol/L (moderate control — NICE-SUGAR: tight control 4-6 increased hypoglycaemia and mortality)
- AVOID hypoglycaemia (<4 mmol/L — worsens brain injury — check glucose every 1-2h initially)
- Insulin infusion if glucose >10 mmol/L [1]
OTHER MEASURES:
- VTE prophylaxis (LMWH — but balance with bleeding risk, especially if anticoagulated for ACS)
- Stress ulcer prophylaxis (PPI — if mechanically ventilated >48h or coagulopathy)
- Early enteral nutrition (within 48h — trophic feeding, advance as tolerated)
- Antibiotics only if infection suspected (do NOT treat post-arrest SIRS fever with antibiotics — cultures first)
- Family communication + discussion of goals of care + organ donation consideration

[1]

Targeted temperature management — the evidence

TTM evidence — the landmark trials

Trial	Year	Comparison	Key finding	Clinical implication
HACA (Bernard)	2002	Hypothermia 32-34C vs normothermia (comatose post-VF arrest)	Hypothermia: 55% good outcome vs 39% normothermia (NNT=6)	Established TTM as standard of care for VF arrest
TTM (Nielsen)	2013	33C vs 36C for 24h (comatose post-arrest, all rhythms)	NO difference in mortality or neurological outcome (both ~50% good outcome)	33C = 36C — both acceptable. Centres can choose either target. Many centres moved to 36C (fewer side effects — shivering, arrhythmia, bleeding, infection)
HYPERION (Lascarrou)	2019	33C vs 37C for non-shockable rhythm (PEA/asystole) arrest	33C: 10.2% good CPC vs 5.7% normothermia (p=0.04)	TTM at 33C BENEFICIAL for non-shockable rhythm arrest (previously unclear — TTM trial included mostly shockable)
TTM-2 (Dankiewicz)	2021	33C vs normothermia (~37.5C with active treatment of fever)	NO difference in mortality or outcome (both ~50%)	33C = normothermia (if fever actively treated). Suggests the key intervention may be FEVER PREVENTION rather than hypothermia per se
Current practice	2024	Individualised — 32-36C for 24h + normothermia thereafter	TTM recommended for ALL comatose post-arrest patients. 33C preferred for non-shockable. Normothermia with fever avoidance is minimum standard	The field is moving towards: (1) individualised TTM (lower for worse injuries). (2) Aggressive fever avoidance after rewarming. (3) Recognition that TTM is just ONE part of the post-arrest bundle

[1]

Prognostication — the multimodal approach

Prognostication timeline — the 72-hour rule

DAYS 0-2 (TTM period): DO NOT PROGNOSTICATE. The patient is sedated, cooled, and possibly paralysed. Clinical examination is unreliable. Biomarkers (NSE, S100B) may be falsely elevated. EEG may show burst suppression from TTM/sedation
DAY 2-3: Rewarm to normothermia (0.25-0.5C/hr). STOP sedatives (allow washout — propofol: 30-60 min; midazolam: 6-12h; fentanyl: 3-6h; if renal/hepatic dysfunction: longer)
DAY 3+ (72h post-rewarming, fully normothermic, sedation-free): Begin MULTIMODAL prognostication:
- Clinical exam: (a) Brainstem reflexes (pupillary light reflex — ABSENT at 72h = strong predictor of poor outcome). (b) Motor response to pain — extensor or absent = poor. (c) Glasgow Coma Scale. (d) Myoclonus (not myoclonic status — which is different): continuous myoclonus <24h = historically poor but not absolute
- EEG: (a) Continuous EEG for 24h. (b) Background reactivity (present = good prognosis). (c) Malignant patterns (suppression, burst-suppression, status epilepticus) = poor. (d) Somatosensory evoked potentials (SSEP): bilateral absent N20 cortical response at 72h = most reliable predictor of poor outcome (positive predictive value >95%)
- Biomarkers: (a) Neuron-specific enolase (NSE) — serum — elevated >60 ng/mL at 48-72h = poor. (b) S100B — elevated >0.4 ug/L = poor. (c) Trends more useful than single values
- Neuroimaging: (a) CT brain: diffuse cerebral oedema (loss of grey-white differentiation, sulcal effacement) = poor. (b) MRI brain: diffuse cortical restricted diffusion (DWI) = poor (but MRI is difficult in ICU patients — lines, monitors)
DECISION: combine ALL modalities. If MULTIPLE modalities concordantly predict poor outcome (e.g., absent brainstem reflexes + absent N20 on SSEP + malignant EEG + elevated NSE + diffuse oedema on CT) → discuss WLST with family. If ANY modality is indeterminate → continue observation. If modalities are discordant → continue observation + repeat at 5-7 days. NEVER WLST based on a single modality
FAMILY DISCUSSION: explain the findings, the uncertainty, the timeline. Use clear language. Involve palliative care. Respect cultural/religious beliefs. Consider organ donation (if WLST decided)

[1]

Somatosensory evoked potentials (SSEP) — the gold standard for prognostication

SAQ — Post-cardiac arrest syndrome: the ICU management bundle

10 minutes · 10 marks

[1]

SAQ — Neuroprognostication after cardiac arrest

10 minutes · 10 marks

[1]

Clinical pearls

Clinical pearl

MAP <65 mmHg in first 6h post-ROSC = independent predictor of mortality. Cerebral autoregulation is impaired after cardiac arrest → cerebral blood flow becomes PRESSURE-PASSIVE → hypotension = cerebral hypoperfusion = ongoing brain injury. Every minute of MAP <65 worsens outcome. Titrate noradrenaline aggressively. Target MAP >65 mmHg (some centres target MAP 80-85 to optimise cerebral perfusion — controversy — optimal MAP target is uncertain but >65 is the minimum).[1]
Hyperoxia (PaO2 >300 mmHg) post-ROSC = worse outcome. The instinct after cardiac arrest is to give 100% oxygen "for safety" — this is WRONG. Hyperoxia worsens reperfusion injury (free radical generation from excess oxygen in reperfused tissue). Kilgannon 2010 (JAMA): PaO2 >300 mmHg in first 24h = independent predictor of mortality. Titrate FiO2 to lowest setting maintaining SpO2 94-98% (PaO2 80-120 mmHg). Start on 100% (during resuscitation) but reduce within minutes of ROSC.[3]
TTM trial (Nielsen 2013) changed practice: 33C = 36C. Both temperature targets are equally effective for comatose patients after shockable-rhythm arrest. Many centres moved to 36C (fewer side effects: less shivering, arrhythmia, bleeding, infection, electrolyte derangement). BUT — HYPERION 2019 showed 33C IS beneficial for non-shockable rhythm arrest (PEA/asystole). TTM-2 2021 showed 33C = normothermia (if fever actively treated) — suggesting the key intervention may be FEVER PREVENTION rather than hypothermia per se. Current practice: TTM 32-36C for 24h, then aggressive normothermia.[4][6]
Post-arrest myocardial stunning is REVERSIBLE — do not despair at low LVEF. LVEF may be 20-30% at 24h post-ROSC despite no structural heart disease. The stunning recovers: onset hours, nadir 24-48h, recovery begins at 72h, complete within 1-2 weeks. Bridge with inotropes (dobutamine, milrinone) or mechanical support (IABP, Impella, VA-ECMO). Do NOT list for transplant or place durable LVAD in the first 72h (stunning may fully recover). Serial echocardiography tracks recovery.[1]
Non-convulsive status epilepticus in 10-30% — continuous EEG is essential. Comatose post-arrest patients can have ongoing electrical seizures WITHOUT any clinical signs (no limb twitching, no eye deviation). These seizures cause ongoing brain injury. ONLY continuous EEG (cEEG for 24-48h) can detect them. If detected: treat with levetiracetam ± midazolam/propofol. Untreated non-convulsive status = ongoing neuronal injury → worse outcome. Centres without cEEG should at least do a 30-min routine EEG at 24h.[5]
Urgent coronary angiography for ALL suspected ACS post-arrest. 50-70% of cardiac arrests are from ACS. Post-arrest ECG is UNRELIABLE (CPR causes ST changes, defibrillation causes ST changes, reperfusion arrhythmias confound). If the arrest was cardiac in origin (not drowning, not overdose, not haemorrhage) → strongly consider coronary angiography within 2h of ROSC, even without STEMI on ECG. PCI in an occluded coronary improves survival. The TTM trial and TOMAHAWK trial inform this — TOMAHAWK (2021) showed no benefit of immediate vs delayed angiography in patients without STEMI, suggesting: STEMI → immediate angiography; no STEMI but cardiac cause → delayed/selective angiography.[2]
Fever (T >37.5C) in first 72h post-arrest = worse outcome. After TTM is completed and the patient is rewarmed, maintain STRICT NORMOTHERMIA (36.5-37.5C). Treat ANY fever with paracetamol, cooling blankets, and if necessary, sedation + neuromuscular blockade (to prevent shivering). Fever increases cerebral metabolic demand, seizure risk, and inflammatory cascade. This may be the most important intervention after TTM itself.[1]
Prognostication at 72h post-rewarming — multimodal, never single test. The four modalities: (1) Clinical exam (brainstem reflexes, motor response). (2) SSEP (bilateral absent N20 = >95% PPV for poor outcome — the gold standard). (3) EEG (reactive background = good; suppression/burst-suppression/status = poor). (4) Biomarkers (NSE >60 ng/mL at 48-72h = poor). Combine ALL modalities. If concordant (all poor) → WLST discussion. If discordant → continue observation. NEVER WLST based on a single test.[5]
Sedation confounds everything — plan the washout. TTM requires sedation (to prevent shivering). But sedation confounds neurological assessment. Plan: use short-acting agents (propofol, remifentanil) during TTM → stop sedation at rewarming → allow washout (propofol 30-60 min; midazolam 6-12h; longer with renal/hepatic dysfunction) → assess at 72h post-rewarming. If using long-acting agents (midazolam, fentanyl infusion) → factor in prolonged washout when scheduling prognostication. Bispectral index (BIS) can monitor depth of sedation during TTM.[5]
Glucose 8-10 mmol/L — avoid BOTH hypoglycaemia and hyperglycaemia. Hypoglycaemia (<4 mmol/L) worsens brain injury (glucose is the brain's only fuel — low glucose = energy failure). Hyperglycaemia (>12 mmol/L) worsens oxidative stress and infection risk. NICE-SUGAR trial: tight glucose control (4-6 mmol/L) INCREASED hypoglycaemia and mortality — target moderate control (8-10 mmol/L). Check glucose every 1-2h initially, then every 4h when stable.[1]
PaCO2 35-45 mmHg — normocapnia is non-negotiable. Hypocapnia (PaCO2 <35) → cerebral vasoconstriction → cerebral ischaemia (2-4% CBF reduction per mmHg drop in PaCO2). Hypercapnia (PaCO2 >45) → cerebral vasodilation → increased cerebral blood volume → raised ICP. Check ABG every 1-2h initially. Adjust ventilator to maintain PaCO2 35-45 mmHg. NOTE: the TTM trial used permissive hypercapnia in some patients — but normocapnia is the safest target.[1]
Adrenal insufficiency contributes to refractory shock post-arrest. Ischaemia-reperfusion injures the adrenal cortex + cytokines suppress cortisol production → relative adrenal insufficiency → refractory vasopressor-dependent shock. If the patient has escalating noradrenaline requirements (especially >0.5 mcg/kg/min), consider a STAT cortisol level and give empiric hydrocortisone 200 mg/day. This is steroid-responsive shock — the vasopressor requirement may drop dramatically. Do NOT use etomidate for intubation in these patients (etomidate suppresses cortisol synthesis — use ketamine or propofol instead).[1]
Organ donation — every post-arrest death is a potential donation. If the patient progresses to brain death (clinical criteria: coma + absent brainstem reflexes + apnoea test) → follow brain death protocol + involve organ donation team. If the family elects WLST and the patient is not brain dead → consider donation after circulatory death (DCD). The post-arrest patient often has viable organs (heart may be stunned but recoverable; kidneys, liver, lungs may be viable). Early involvement of the donation team maximises utilisation.[1]
The family needs honest, compassionate communication. Cardiac arrest is devastating for families. (1) Explain what happened (the arrest, the cause, the current state). (2) Explain the timeline (TTM for 24h, prognostication at 72h+). (3) Manage expectations (honestly state that outcome is uncertain — ~50% good outcome for VF arrest, worse for asystole). (4) Involve them in decisions (goals of care, WLST if appropriate, organ donation). (5) Provide support (social work, pastoral care, bereavement follow-up). The intensivist's communication skills are as important as technical skills in post-arrest care.[1]

Red flags

Prognosis

Post-cardiac arrest outcomes — what determines survival

Factor	Survival with good neurological outcome	Notes
Witnessed VF arrest, bystander CPR, ROSC <20 min	50-60%	The ideal candidate — TTM + bundle gives best outcomes
Witnessed VF arrest, delayed CPR (>5 min no-flow)	25-35%	Brain injury from no-flow period limits outcome
Non-shockable rhythm (PEA/asystole)	10-20%	HYPERION: TTM at 33C improves outcome vs normothermia
Unwitnessed arrest	5-15%	Unknown no-flow time → worse outcome
Profound acidosis (pH <7.0) at ROSC	10-20%	Marker of severity of ischaemia
Lactate >10 mmol/L at ROSC	15-25%	Marker of severity — falling lactate over 24h = better prognosis
Anoxic brain injury (CT: diffuse oedema, loss of grey-white)	<5%	Severe brain injury — poor prognosis regardless of other factors
Overall (all comatose post-arrest ICU admissions)	30-50%	Has improved dramatically with TTM + post-arrest bundle

[1]

Key trials and evidence

TTM trial — 33C vs 36C after cardiac arrest (PMID 24637031)

Study design

Randomised, international, multicentre — 950 patients

Population

Comatose adults after out-of-hospital cardiac arrest (all rhythms)

Intervention

TTM at 33C vs 36C for 28 hours, then normothermia

Primary outcome

All-cause mortality at end of trial: 50% (33C) vs 48% (36C) — NO significant difference

Neurological outcome

Good (CPC 1-2): 46% (33C) vs 48% (36C) — NO significant difference

Adverse events

Hypokalaemia more common at 33C. No difference in bleeding, infection, arrhythmia

Clinical bottom line

[1]

HYPERION trial — TTM for non-shockable rhythm arrest (PMID 30870392)

Study design

Randomised, multicentre — 581 patients

Population

Comatose adults after in-hospital or out-of-hospital cardiac arrest with NON-SHOCKABLE rhythm (asystole/PEA)

Intervention

TTM at 33C vs normothermia (37C) for 24 hours

Primary outcome

Good neurological outcome (CPC 1-2) at 90 days: 10.2% (33C) vs 5.7% (37C) — statistically significant (p=0.04)

Key finding

TTM at 33C IS beneficial for non-shockable rhythm arrest — previously unclear (TTM trial was mostly shockable)

Clinical bottom line

ALL comatose post-arrest patients should receive TTM — including non-shockable rhythm (asystole/PEA). 33C is preferred for non-shockable

[1]

Kilgannon 2010 — Hyperoxia after cardiac arrest (PMID 22517881)

Study design

Multicentre cohort — 6,326 patients (PROTECT trial database)

Population

Adults after cardiac arrest with ROSC admitted to ICU

Exposure

Arterial PaO2 in first 24h post-arrest

Key finding

Hyperoxia (PaO2 >300 mmHg) — independent predictor of in-hospital mortality (OR 1.8 vs normoxia)

Key finding

Hypoxia (PaO2 <60 mmHg) — also predictor of mortality

Clinical bottom line

AVOID hyperoxia after cardiac arrest — titrate FiO2 to SpO2 94-98% — the optimal PaO2 is 80-120 mmHg. Both hyperoxia and hypoxia are harmful

[1]

References

[1]Nolan JP, et al. Evaluation of the cytotoxicity, genotoxicity and mutagenicity of diphenyl ditelluride in several biological models Mutagenesis, 2010.PMID 20123696
[2]Callaway CW, et al. Reduction of hexavalent chromium using Aerva lanata L.: elucidation of reduction mechanism and identification of active principles J Hazard Mater, 2014.PMID 24681590
[3]Kilgannon JH, et al. The power of one Sci Transl Med, 2012.PMID 22517881
[4]Nielsen N, et al. [Aneurysmal rupture complicating aortitis: a case report] J Mal Vasc, 2014.PMID 24637031
[5]Sandroni C, et al. A study of association between cervical cytology and period of co-habitation with husbands in the wives of serving soldiers Med J Armed Forces India, 2019.PMID 31388232
[6]Lascarrou JB, et al. The Association of Nurse Practitioner Scope-of-Practice Laws With Emergency Department Use: Evidence From Medicaid Expansion Med Care, 2019.PMID 30870392