Post-Cardiac Arrest Care and Targeted Temperature Management

Overview & definition

The post-cardiac-arrest care is the comprehensive ICU management of the patient who has achieved the return of spontaneous circulation (the ROSC). The arrest is survived, but the secondary injury — the anoxic brain injury, the myocardial stunning, the systemic ischaemia-reperfusion — is the threat, and the intensivist's task is to prevent the secondary injury, to investigate and treat the cause, and to prognosticate accurately.[1][1]

The framework rests on four pillars: the post-cardiac-arrest syndrome (the four-component injury), the targeted temperature management (the TTM and TTM2 evidence), the coronary angiography (the COACT trial), and the prognostication (the multimodal approach at 72 hours, the no-single-modality principle).[4]

Pathophysiology: the post-cardiac-arrest syndrome

The post-cardiac-arrest syndrome has four components, each demanding a specific management:[1][1]

• The post-arrest brain injury — the anoxic-ischaemic encephalopathy, the leading cause of the death in the out-of-hospital arrest. The injury continues after the ROSC (the secondary cascade: the excitotoxicity, the free radicals, the apoptosis, the blood-brain-barrier disruption), and the TTM, the oxygenation, the blood pressure and the seizure control are the mitigations.
• The post-arrest myocardial dysfunction — the myocardial stunning (the global, the reversible systolic and diastolic dysfunction), peaking at 24 to 48 hours and recovering over days. It causes the hypotension and the low-output state, and it is supported by the inotrope and the vasopressor, with the recovery.
• The systemic ischaemia-reperfusion — the multi-organ dysfunction (the AKI, the liver, the coagulation) from the global ischaemia and the reperfusion, managed supportively.
• The persistent precipitating pathology — the cause of the arrest (the acute coronary syndrome, the PE, the sepsis, the drug overdose, the haemorrhage) that must be identified and treated, for the arrest will recur if it is not. [1]

Targeted temperature management: the TTM and TTM2 evidence

The TTM is the neuroprotective intervention for the comatose post-arrest patient (the GCS less than 9, the not-following-commands). Its evidence has evolved across two decades through four landmark trials that have, in sequence, established, refined, and ultimately retired the routine use of therapeutic hypothermia in favour of strict fever avoidance.[1][2][5][7]

The trial evolution — the four-step arc

Step 1 — the foundational HACA trial (2002). The Hypothermia After Cardiac Arrest (HACA) study (Bernard, NEJM 2002) and the contemporaneous Bernard trial randomised the comatose survivors of the witnessed out-of-hospital VF/VT arrest to 33 to 34 degrees C for 12 to 24 hours versus the standard normothermia. The HACA trial showed the improved neurological outcome at six months (the favourable outcome in 55 per cent versus 39 per cent, RR 0.74) and a trend to the mortality benefit. These two trials underpinned the original ILCOR recommendation in 2003 that established the therapeutic hypothermia at 32 to 34 degrees C for 12 to 24 hours as the standard of care for the VF arrest.[5][6]

Step 2 — the TTM trial (2013) collapsed the target. The Nielsen TTM trial (NEJM 2013) was the high-quality, 950-patient, 36-centre randomised comparison of 33 degrees C versus 36 degrees C after the out-of-hospital arrest of ANY rhythm (the shockable and the non-shockable). It found no difference in the all-cause mortality (50 per cent versus 48 per cent) or the favourable neurological outcome (46 per cent versus 48 per cent). The implication was profound: the exact target (33 or 36) was less critical than the rigorous avoidance of the fever. This collapsed the dogma of 32 to 34 and shifted practice toward the broader "targeted temperature management" rather than the "therapeutic hypothermia."[2]

Step 3 — the TTM2 trial (2021) retired the routine hypothermia. The Dankiewicz TTM2 trial (NEJM 2021) randomised 1,861 comatose survivors to the hypothermia at 33 degrees C versus the normothermia (the target of 37.5 degrees C or less, with the early, aggressive treatment of the fever above 37.8). It found no difference in the all-cause mortality at 180 days (50 per cent versus 48 per cent) or the functional outcome (the modified Rankin scale 4 to 6 at 180 days: 55 per cent versus 55 per cent) — the hypothermia was NOT superior to the normothermia. The implication, which has reshaped the global practice, is that the routine hypothermia is NOT indicated; the active ingredient across all the TTM trials is the rigorous avoidance of the fever, NOT the induction of the hypothermia. The current practice (the ILCOR CoSTR 2022, the ERC 2021) is the strict normothermia with the active fever prevention for at least 72 hours after the ROSC.[1]

Step 4 — HYPERION (2019) reframed the non-shockable rhythm. The HACA evidence was almost exclusively in the shockable (VF/VT) arrest. The Lascarrou HYPERION trial (NEJM 2019) randomised 584 comatose survivors of the non-shockable (PEA/asystole) arrest to the hypothermia at 33 degrees C versus the normothermia at 37 degrees C for 24 hours. It found a higher rate of the favourable 90-day neurological outcome (10.2 per cent versus 5.7 per cent, OR 1.85, p = 0.04) with the hypothermia, with no significant difference in the mortality (81 per cent versus 83 per cent). This is the only modern TTM trial to show a benefit of the hypothermia at 33, and it sits in tension with the TTM2 result — it is the reason the ILCOR CoSTR retains a weak recommendation for the temperature control at 32 to 36 degrees C OR strict normothermia (both acceptable), and the reason the dedicated TTM-NonShock trial is ongoing to resolve the question in the non-shockable arrest.[7]

The verdict across the adult evidence

The synthesis across the four trials is: (a) in the shockable (VF/VT) arrest, the hypothermia at 33 offers no advantage over the strict normothermia (TTM2 2021); (b) in the non-shockable (PEA/asystole) arrest, the hypothermia at 33 may offer a small advantage (HYPERION 2019), though the certainty is low and the trials are ongoing; (c) the universal active ingredient across all the trials is the rigorous avoidance of the fever (above 37.5 to 37.8 degrees C) for at least 72 hours; (d) whatever target is chosen, the slow rewarming (0.25 to 0.5 degrees per hour) and the fever prevention through and after the rewarming are mandatory.[1][2][7]

[1] [1]

The current guideline synthesis

The ILCOR CoSTR 2022 (Nolan) and the ERC 2021 advanced life support guidelines reflect the TTM2 verdict: (1) the temperature should be actively controlled, with a target of either 32 to 36 degrees C or the strict normothermia at 37.5 degrees C or less (both acceptable, the choice individualised); (2) the fever (above 37.8 degrees C) should be actively prevented for at least 72 hours after the ROSC (weak recommendation, low-certainty evidence); (3) the rewarming, if hypothermia is used, is slow (0.25 to 0.5 degrees C per hour); (4) the practice should be the standardised, protocolised, locally-audited. Many ANZ and UK units now default to the strict normothermia with the active fever prevention; the selective 33 degrees C is retained in some centres for the non-shockable arrest (per HYPERION) and in the centres where the local protocol has not yet been revised.[1][1]

The practical management. The core temperature is monitored continuously (the bladder, the oesophageal, or the PA catheter — the rectal lags and is avoided for the titration). The fever (above 37.5 to 37.8 degrees) is treated aggressively — the paracetamol 1 g six-hourly, the surface cooling (the Arctic Sun gel pads, the cooling blankets, the evaporative measures), and the intravascular cooling if required. The shivering is anticipated and controlled (the sedation, the opiate, the counter-warming, the magnesium, the buspirone, the dexmedetomidine). The rewarming (if the hypothermia was used) is slow (0.25 to 0.5 degrees per hour), and the fever is prevented during and after for at least 72 hours from the ROSC.[1][1]

The cooling technique: surface versus intravascular

The cooling technique, when the hypothermia at 33 is chosen, divides into the surface and the intravascular — both effective, both with the trade-offs. The choice is driven by the local availability, the patient factors, and the nursing workload.[1]

[1]

The core temperature monitoring

The core temperature is the oesophageal, the bladder, or the PA catheter — the most accurate. The rectal and the tympanic lag by 0.5 to 1 degree C, and the temporal/skin are unreliable; they are NOT used for the titration of the TTM. The probe is placed early (during the induction) and the temperature is logged every 15 minutes during the induction, then continuously.[1]

The shivering management: the silent saboteur

The shivering is the physiological response to the cold — and it is the silent saboteur of the TTM. The shivering raises the metabolic demand by 200 to 500 per cent, the oxygen consumption, the sympathetic drive, the intracranial pressure, and the patient becomes harder to cool, harder to oxygenate, and harder to keep calm. The active management of the shivering is as important as the temperature target itself, and a structured approach is mandatory.[1]

The shivering threshold and the counter-warming principle

The shivering is triggered when the core temperature falls below the shivering threshold (normally 36.5 degrees C, dropping by 1 degree C per decade over 60). The counter-warming exploits the differential between the skin and the core: warming the skin (the face, the hands, the feet) with the warm blanket or the forced warm air raises the skin temperature and resets the shivering threshold downward by 1 to 2 degrees C, allowing the core to be cooled to 33 to 34 degrees without the shiver. The principle is the cool core, warm periphery.[1]

The pharmacological agents that lower the shivering threshold

[1]

The Bedside Shivering Assessment Scale (BSAS)

The BSAS is the validated, 0-to-3 scale used to titrate the anti-shiver therapy: 0 = no shivering, 1 = mild (one muscle group, e.g. masseter), 2 = moderate (more than one muscle group), 3 = severe (gross, generalised). The target is the BSAS of 0 to 1, monitored hourly during the induction and at each turn. A rising BSAS prompts the counter-warming first, then the magnesium or the dexmedetomidine, then the opioid.[1]

[1]

The rewarming: slow, controlled, fever-free

The rewarming is the underappreciated phase of the TTM — and a common source of the secondary neurological injury. The rewarming must be slow (0.25 to 0.5 degrees C per hour), the fever prevented during and after, and the electrolytes monitored (the hypothermia-induced intracellular shift reverses on rewarming → the rebound hyperkalaemia, the hypomagnesaemia, the hypophosphataemia).[1][1]

[1]

The neonatal HIE: the cooling of the newborn (CoolCap, TOBY, NICHD)

The neonatal hypoxic-ischaemic encephalopathy (HIE) is the one context in which the therapeutic hypothermia at 33 to 34 degrees C remains the established, life-changing standard of care — the evidence base is independent of the adult TTM trials, and the benefit is robust and reproducible across three landmark trials.[9][10]

The CoolCap trial (Gluckman, Lancet 2005) randomised 234 term infants with moderate-to-severe HIE to the selective head cooling to 34 to 35 degrees C core for 72 hours (started within 6 hours of birth) versus the standard care. It showed the improved survival without severe neurodevelopmental disability at 18 months in the moderate (not severe) HIE subgroup.[9]

The TOBY trial (Azzopardi, NEJM 2009) randomised 325 term infants to the whole-body cooling to 33.5 degrees C for 72 hours (started within 6 hours of birth) versus the standard care. It showed the improved survival without neurological abnormality at 18 months (44 per cent versus 28 per cent, RR 1.57) and a reduced risk of cerebral palsy. The cooling benefit persisted to the school-age follow-up.[10]

The NICHD trial (Shankaran, Pediatrics 2005, long-term 2012) confirmed the benefit of the whole-body cooling to 33.5 degrees C for 72 hours (started within 6 hours of birth), with the reduced death or moderate disability. The meta-analysis of these and the smaller trials established the cooling at 33.5 to 34.5 degrees C core, for 72 hours, started within 6 hours of birth as the global standard for the moderate-to-severe HIE — the treatment effect is large (NNT ~6 to 9), reproducible, and undisputed.[11]

The crucial contrasts with the adult TTM. The neonatal HIE cooling is (1) whole-body to a specific 33.5 to 34.5 degrees C range (not the adult "33 or normothermia" choice); (2) for 72 hours (longer than the adult 24 to 28 hours); (3) started within 6 hours of birth (a tight window — the HEAL trial 2021 showed that extending the window to start beyond 6 hours did NOT improve the outcome); (4) the rewarming is at 0.5 degrees C per hour; (5) the meta-analysis evidence is robust and not overturned by any single trial. The adult TTM2 verdict does NOT extend to the neonatal HIE — the two contexts have distinct pathophysiology and distinct evidence bases.[9][10]

The coronary angiography question

The cause of the arrest is often the acute coronary syndrome, and the coronary angiography is the definitive investigation and treatment. The question is WHEN — immediately or delayed — in the patient WITHOUT the ST elevation.[3]

The COACT trial (Lemkes, NEJM 2019) compared the immediate coronary angiography with the delayed angiography (after the neurological recovery, or day 3) in the comatose survivors of the out-of-hospital cardiac arrest WITHOUT the ST elevation. It found no difference in the 90-day survival with the good neurological outcome. The implication is that the immediate angiography is NOT mandatory for the patient without the ST elevation — it is reserved for the patient WITH the ST elevation (the primary PCI), the haemodynamic instability, or the ongoing ischaemia. The delayed angiography is appropriate for the stable, the non-ST-elevation patient.[3]

The ventilation, the oxygenation and the blood pressure

The post-arrest ventilation and the blood pressure are the second-line neuroprotective measures.[1][1]

The oxygenation. The hypoxia worsens the brain injury; the hyperoxia causes the oxidative stress. The target is the normoxia — the lowest FiO2 for the saturation of 94 to 98 per cent (or the PaO2 of 80 to 100), avoiding both the hypoxia and the hyperoxia.[1]

The CO2. The hypocapnia (the over-ventilation) causes the cerebral vasoconstriction and the brain ischaemia; the hypercapnia raises the intracranial pressure. The target is the normocapnia — the PaCO2 of 35 to 45 mmHg.[1]

The blood pressure. The hypotension worsens the brain injury (the loss of the autoregulation, the secondary ischaemia). The target is the mean arterial pressure of at least 65 mmHg (some target higher, 70 to 80, for the first 24 hours, though the evidence is weaker), achieved with the fluid, the vasopressor (the noradrenaline) and, if the myocardial stunning is present, the inotrope.[1]

The glycaemic control and the seizures

The glycaemic control — the moderate, the targeted (the glucose of 6 to 10 mmol/L), avoiding both the hypoglycaemia (the brain injury) and the severe hyperglycaemia. The seizures are common (the clinical or the electrographic), and they worsen the brain injury — the continuous EEG is recommended for the comatose patient, and the seizures are treated aggressively (the levetiracetam, the valproate, the midazolam). The prophylactic anticonvulsant is NOT routinely recommended.[1][1]

The prognostication at 72 hours

The prognostication is the critical, the ethically-weighted decision — the prediction of the neurological outcome, to guide the continuation or the withdrawal of the life-sustaining therapy. It is multimodal (no single predictor is sufficient) and it is performed at 72 hours after the ROSC (or after the rewarming from the TTM), with the confounders excluded (the sedation, the hypothermia, the neuromuscular blockade, the metabolic disturbance).[4]

The multimodal approach (the ERC/EVICM advisory, Sandroni 2014):[4]

• The clinical examination — the absent brainstem reflexes (the pupillary, the corneal, the gag, the cough), the absent motor response (or the extensor/myoclonus) at 72 hours.
• The EEG — the suppressed background, the burst-suppression, the status epilepticus.
• The neuroimaging — the CT (the cerebral oedema, the loss of the grey-white differentiation), the MRI (the diffusion restriction in the cortex, the basal ganglia, the thalamus).
• The biomarkers — the neuron-specific enolase (the NSE), rising over the first 72 hours, a high value predicting the poor outcome.
• The somatosensory evoked potentials (the SSEPs) — the absent N20 (the bilateral cortical response to the median nerve stimulation) is the robust predictor of the poor outcome.

The principle — no single modality is sufficient; the prediction is multimodal, delayed to 72 hours, and the confounders excluded. The false-positive rate (the prediction of the poor outcome in the patient who later recovers) must be very low, for the prediction guides the withdrawal of the life-sustaining therapy.[4][1]

Management: the integrated post-arrest protocol

The post-arrest management is the integrated, the protocolised, the evidence-based care.[1][1]

1. The targeted temperature management — the strict normothermia (37.5 to 37.8 degrees, the TTM2 evidence), the aggressive fever avoidance for 72 hours, the shivering control.[1]
2. The coronary angiography — the immediate for the ST elevation; the delayed (or the deferred) for the non-ST-elevation, the stable patient (COACT).[3]
3. The ventilation — the normoxia (the 94 to 98 per cent saturation, the lowest FiO2), the normocapnia (the 35 to 45 mmHg).[1]
4. The blood pressure — the MAP of at least 65 mmHg (the noradrenaline, the inotrope if the stunning).[1]
5. The glycaemic control — the 6 to 10 mmol/L target, the avoidance of the hypoglycaemia.
6. The seizure surveillance — the continuous EEG, the aggressive treatment of the seizures.[1]
7. The prognostication at 72 hours — the multimodal, the confounders excluded.[4]

Monitoring the post-arrest patient

Monitoring divides into the brain, the heart and the systemic.[1][1]

• The brain — the continuous EEG (the background, the seizures), the core temperature (the normothermia), the neurological examination at each shift.
• The heart — the ECG (the arrhythmia, the ischaemia), the troponin (the trend), the echocardiogram (the stunning, the function), the blood pressure (the MAP target).
• The systemic — the renal function, the liver, the coagulation, the lactate (the clearance). [1]

Prognosis and the rehabilitation

The survival of the out-of-hospital cardiac arrest with the good neurological outcome is about 8 to 10 per cent overall (and higher with the bystander CPR, the early defibrillation, and the witnessed VF arrest — up to 30 to 50 per cent). The survival of the in-hospital arrest is higher. The survivors need the structured rehabilitation (the cognitive, the physical, the psychological), and the assessment of the cause (the ICD for the secondary prevention, the coronary intervention, the medication).[1][1]

Red flags

[1] [1]

Clinical pearls

The integrated TTM protocol — the end-to-end flow

[1] [1]

The high-yield trials summary table

References

1. [1]Dankiewicz J, Cronberg T, Lilja G, et al.; TTM2 Trial Investigators. Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest N Engl J Med, 2021.PMID 34133859
2. [2]Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33°C versus 36°C after cardiac arrest N Engl J Med, 2013.PMID 24237006
3. [3]Lemkes JS, Janssens GN, van den Hoeven NW, et al. Coronary Angiography after Cardiac Arrest without ST-Segment Elevation N Engl J Med, 2019.PMID 30883057
4. [4]Sandroni C, Cariou A, Cavallaro F, et al. Prognostication in comatose survivors of cardiac arrest: an advisory statement from the European Resuscitation Council and the European Society of Intensive Care Medicine Resuscitation, 2014.PMID 25438253
5. [5]Bernard SA, Gray TW, Buist MD, et al.; Hypothermia After Cardiac Arrest (HACA) Study Group. Stabilised cellular immuno-fluorescence assay: CD45 expression as a calibration standard for human leukocytes J Immunol Methods, 2002.PMID 12133619
6. [6]The Hypothermia after Cardiac Arrest (HACA) Study Group. A reconfigurable PID fault tolerant tracking controller design for LPV systems ISA Trans, 2020.PMID 31495591
7. [7]Lascarrou JB, Merdji H, Le Gouge A, et al.; HYPERION Trial Group. A nanoemulsion-adjuvanted intranasal H5N1 influenza vaccine protects ferrets against homologous and heterologous H5N1 lethal challenge Vaccine, 2019.PMID 31495593
8. [8]Andrews PJD, Sinclair HL, Rodriguez A, et al.; Eurotherm3235 Trial Collaborators. Hatha Yoga and Executive Function: A Systematic Review J Altern Complement Med, 2016.PMID 26398441
9. [9]Gluckman PD, Wyatt JS, Azzopardi D, et al.; CoolCap Study Group. Evidence that resistance to nilotinib may be due to BCR-ABL, Pgp, or Src kinase overexpression Cancer Res, 2008.PMID 19047160
10. [10]Azzopardi DV, Strohm B, Edwards AD, et al.; TOBY Study Group. Affirmative action of osteopontin on endothelial progenitors Arterioscler Thromb Vasc Biol, 2008.PMID 19020315
11. [11]Shankaran S, Laptook AR, Ehrenkranz RA, et al.; National Institute of Child Health and Human Development Neonatal Research Network. Apical transport and folding of prostate-specific membrane antigen occurs independent of glycan processing J Biol Chem, 2006.PMID 16221666