Library › Emergencies and critical care › Post-cardiac arrest care

Post-cardiac arrest care

Care of adults after return of spontaneous circulation, from the first hour of stabilization through temperature control, coronary angiography, neurocritical care, prognostication, and recovery.

Post-cardiac arrest syndrome: definition, phases, and outcomes

Return of spontaneous circulation (ROSC) ends the arrest but begins a second illness. The 2008 International Liaison Committee on Resuscitation (ILCOR) consensus statement called it the post-cardiac arrest syndrome and defined four components that coexist in varying proportion: post-cardiac arrest brain injury, post-cardiac arrest myocardial dysfunction, the systemic ischemia-reperfusion response, and the persistent precipitating pathology that caused the arrest. The severity of each depends on the duration of no flow (collapse to the start of compressions) and low flow (compressions to ROSC), the cause of the arrest, and the patient's premorbid reserve. A brief witnessed ventricular fibrillation arrest terminated by an early shock can leave almost no syndrome, whereas a prolonged asphyxial arrest with pulseless electrical activity typically produces severe brain injury and multiorgan failure.

The same statement divided the course into phases that remain useful for organizing care. The immediate phase covers the first 20 minutes after ROSC. The early phase extends from 20 minutes to 6 to 12 hours, when interventions to limit secondary injury are thought to have the most effect. The intermediate phase runs from 6 to 12 hours to 72 hours, when injury pathways remain active and temperature control, hemodynamic support, and seizure surveillance continue. The recovery phase begins after about 3 days, when prognostication becomes reliable, and is followed by rehabilitation. The 2025 American Heart Association (AHA) algorithm uses a two-step version of the same sequence. Initial stabilization comprises airway, oxygenation and ventilation, hemodynamics, and early diagnostic testing. Continued management comprises treatment of the cause, emergency cardiac intervention, temperature control, seizure evaluation, and prognostication. Many steps proceed in parallel.

Survival and causes of death. Survival to hospital discharge after emergency medical services–treated out-of-hospital cardiac arrest (OHCA) in the United States remains approximately 10%, and survival after in-hospital cardiac arrest (IHCA) is approximately 24%, according to the 2025 AHA guideline summary. Among patients who reach the intensive care unit comatose after OHCA, most deaths are attributed to hypoxic-ischemic brain injury, and most of those deaths follow withdrawal of life-sustaining therapy for a perceived poor neurologic prognosis. Refractory shock and recurrent arrest account for most deaths in the first days, while brain injury dominates later deaths. The 2025 acute coronary syndrome (ACS) guideline from the American College of Cardiology (ACC) and AHA notes that survival to discharge for comatose patients after OHCA is below 10% regardless of cause. In contrast, patients with ST-elevation myocardial infarction (STEMI) who are awake after resuscitation have outcomes comparable to patients with STEMI who did not arrest.

Out-of-hospital and in-hospital arrest. The two populations differ. OHCA is more often cardiac in origin, is frequently due to ventricular fibrillation from acute or chronic coronary disease, and occurs in patients with less comorbidity, while IHCA more often follows respiratory failure, sepsis, hemorrhage, or drug effects, presents with a nonshockable rhythm, and occurs in patients with substantial illness before the arrest. Most randomized trials of temperature control, angiography, and physiologic targets enrolled comatose OHCA survivors; HYPERION enrolled both in-hospital and out-of-hospital arrests with nonshockable rhythms. Guidance for IHCA survivors is extrapolated from these trials and applies the same principles.

The resuscitation itself is covered on the adult cardiac arrest and advanced cardiovascular life support page: high-quality compressions, early defibrillation, epinephrine and antiarrhythmic drugs, treatment of reversible causes, extracorporeal cardiopulmonary resuscitation, and termination of resuscitation.

Pathophysiology of post-cardiac arrest brain injury, myocardial dysfunction, and reperfusion

Brain injury. The brain tolerates the least ischemia of any organ. Within seconds of circulatory arrest, cerebral oxygen stores are exhausted, and within minutes adenosine triphosphate is depleted, ion pumps fail, neurons depolarize, and glutamate floods the synapse. Calcium entry through glutamate receptors activates proteases, phospholipases, and nitric oxide synthase, and mitochondrial injury releases cytochrome c and triggers apoptotic cascades. Reperfusion restores oxygen but adds a second injury from oxygen free radicals, inflammation, and microvascular no-reflow. Neurons in the hippocampus (CA1 region), cerebellar Purkinje layer, neocortical layers 3, 5, and 6, thalamus, and basal ganglia are the most vulnerable. This distribution explains the pattern of injury on diffusion-weighted magnetic resonance imaging (MRI) and the clinical picture of survivors, in whom memory, attention, and executive function are most often affected.

Secondary brain injury continues for hours to days. Cerebral autoregulation is impaired or right-shifted in a substantial proportion of patients, particularly those with chronic hypertension, so cerebral blood flow becomes pressure-passive. Hypotension, hypoxemia, hypocapnia (cerebral vasoconstriction), hyperthermia, seizures, and hyperglycemia or hypoglycemia each worsen the balance of cerebral oxygen supply and demand. Cerebral edema, when it occurs, typically develops over the first 2 to 4 days and can progress to herniation and death by neurologic criteria. These mechanisms are the rationale for every physiologic target in post-arrest care: the injury cannot be reversed, but its extension can be limited.

Myocardial dysfunction. Global ventricular systolic and diastolic dysfunction is common after ROSC, even when coronary arteries are normal. It reflects ischemia-reperfusion injury (stunning), the effects of defibrillation and high-dose epinephrine, and, in many patients, an acute coronary occlusion that caused the arrest. Cardiac index is lowest in the first hours and typically improves over 24 to 72 hours when perfusion is supported and the cause is treated. A depressed ejection fraction on the first echocardiogram therefore does not establish chronic cardiomyopathy; reassessment after recovery, generally before decisions about a defibrillator, distinguishes stunning from fixed dysfunction.

Systemic ischemia-reperfusion response. Whole-body ischemia followed by reperfusion activates immune and coagulation pathways in a pattern that resembles sepsis. Plasma cytokines rise, endotoxin is detectable in some patients from gut translocation, endothelial injury increases capillary leak, coagulation is activated with impaired fibrinolysis, and adrenal function may be relatively insufficient. The clinical result is vasodilatory shock, fever, lactic acidosis, acute kidney injury, ischemic hepatitis, and ileus in the first 72 hours. Early fever in this period is frequently noninfectious, which complicates the diagnosis of aspiration pneumonia and catheter infection.

Persistent precipitating pathology. The condition that caused the arrest is still present after ROSC: an occluded coronary artery, a pulmonary embolism, a ruptured aortic dissection, intracranial hemorrhage, hyperkalemia, tension pneumothorax, a drug overdose, or the arrhythmic substrate of cardiomyopathy or a channelopathy. Its identification and treatment run in parallel with resuscitation from the first minutes, because an untreated cause produces re-arrest.

The four components of the post-cardiac arrest syndrome, after the 2008 ILCOR consensus statement.

Immediate stabilization in the first hour after return of circulation

Resuscitation continues after ROSC, and the four tasks of initial stabilization proceed concurrently. The patient is then assessed for the ability to follow commands, off sedation and neuromuscular blockade when that is possible, because the answer determines whether temperature control and neuroprognostication apply.

Airway. A patient who is comatose after ROSC needs a definitive airway if one is not already in place. A supraglottic device placed during resuscitation is commonly exchanged for an endotracheal tube. Tube position is confirmed with continuous waveform capnography, which also tracks ventilation, and with a chest radiograph. Intubation after ROSC is performed with hemodynamically conservative induction, since the post-arrest circulation is fragile and induction agents can precipitate hypotension or re-arrest.

Oxygenation and ventilation. The AHA algorithm specifies 100% inspired oxygen until oxygen saturation or arterial PaO₂ can be measured reliably, then titration to a saturation of 90% to 98% (PaO₂ 60 to 105 mm Hg) and ventilation to a PaCO₂ of 35 to 45 mm Hg unless severe acidemia is present. An arterial line provides both continuous pressure and serial blood gases.

Hemodynamics. Vasopressors, fluids, or both are initiated or adjusted to a mean arterial pressure (MAP) of at least 65 mm Hg. Intra-arterial monitoring is standard because cuff readings are unreliable in vasoconstricted, low-output states and because pressure targets must be held continuously.

Early diagnostic testing. A 12-lead ECG is obtained to identify ischemia and arrhythmia. CT of the head, chest, abdomen, and pelvis and point-of-care ultrasound or echocardiography are performed when they can identify the cause of arrest or injuries from resuscitation. Initial laboratory tests include an arterial blood gas, lactate, electrolytes (potassium, magnesium, ionized calcium, and phosphate), glucose, creatinine, hepatic panel, complete blood count, coagulation tests, troponin, and, when the history suggests it, toxicology.

The bundle below summarizes US targets and states where the 2025 European Resuscitation Council (ERC) and European Society of Intensive Care Medicine (ESICM) guideline differs.

Post-cardiac arrest care bundle. AHA values from the 2025 Part 11 adult post-cardiac arrest care algorithm; ERC-ESICM values from the 2025 post-resuscitation care guideline.

Oxygenation and ventilation targets after cardiac arrest

Hypoxemia and hyperoxia. Hypoxemia after ROSC extends brain injury and is avoided without qualification. Concern about the opposite extreme arose from observational data. In a registry of 6,326 patients admitted to intensive care after resuscitation, arterial hyperoxia (PaO₂ of 300 mm Hg or more on the first blood gas) was associated with in-hospital mortality of 63%, against 45% with normoxia, and with an adjusted odds ratio for death of 1.8. Reactive oxygen species generated during reperfusion provide a mechanism. Randomized trials have since tested how tightly oxygen needs to be restricted.

• EXACT randomized 428 unconscious OHCA survivors in the prehospital phase to titration to a saturation of 90% to 94% or 98% to 100% until arrival in intensive care. The trial stopped early during the COVID-19 pandemic; survival to discharge was 38.3% with the lower target and 47.9% with the higher target (P = .05), and hypoxic episodes before admission were more frequent with the lower target (31.3% against 16.1%). Early prehospital restriction is not supported.
• BOX randomized 789 comatose OHCA survivors in intensive care to a restrictive PaO₂ target of 68 to 75 mm Hg or a liberal target of 98 to 105 mm Hg. Death or severe disability or coma at 90 days occurred in 32.0% and 33.9% (hazard ratio 0.95). The ILCOR review that followed found no benefit from lower targets over higher targets after admission.

These results define the current position: 100% oxygen until a reliable measurement is available, then rapid titration to avoid both hypoxemia and sustained hyperoxia. The AHA range is a saturation of 90% to 98% (PaO₂ 60 to 105 mm Hg); the ERC-ESICM range is 94% to 98%. Pulse oximetry overestimates arterial saturation in patients with darker skin pigmentation: in a comparison of paired measurements, occult hypoxemia (arterial saturation below 88% with pulse oximetry of 92% to 96%) was detected nearly three times as often in Black patients as in White patients. Arterial blood gases confirm oxygenation whenever a saturation near the lower limit is being accepted, and the 2025 AHA guideline makes awareness of this source of error a recommendation (Class 2b).

Carbon dioxide. Hypocapnia causes cerebral vasoconstriction and is avoided. Mild hypercapnia increases cerebral blood flow and was tested as a neuroprotective strategy. TAME randomized 1,700 comatose OHCA survivors to a PaCO₂ target of 50 to 55 mm Hg or normocapnia (35 to 45 mm Hg); favorable neurologic outcome at 6 months was 43.5% and 44.6% (relative risk 0.98), and 6-month mortality was 48.2% and 45.9%. Normocapnia is therefore the target. During hypothermia, metabolic CO₂ production falls and the same minute ventilation produces hypocapnia, so ventilation is reduced as the patient is cooled and increased during rewarming. Blood gases are interpreted consistently by one method (alpha-stat, at 37°C, is most common).

Lung-protective ventilation. Aspiration, pulmonary contusion from compressions, and the inflammatory response predispose to acute respiratory distress syndrome. Tidal volumes of 6 to 8 mL/kg of predicted body weight with positive end-expiratory pressure titrated to oxygenation are standard in ERC-ESICM guidance, extrapolated from the general intensive care literature. Patients with severe hypoxemia receive the full management of acute respiratory distress syndrome, which sometimes requires neuromuscular blockade and accepts mild hypercapnia; in those cases the respiratory and cerebral goals are weighed against each other.

Blood pressure targets, vasopressors, and perfusion monitoring

Causes of hypotension after ROSC. Hypotension is common in the first hours after resuscitation and has three main sources: vasodilation from the ischemia-reperfusion response and from sedatives, low output from myocardial stunning or an acute infarct, and hypovolemia from capillary leak, hemorrhage, or diuresis. Persistent causes such as tamponade, tension pneumothorax, pulmonary embolism, and hemorrhage from resuscitation injuries are excluded with echocardiography and CT. Early hypotension is associated with lower survival and worse neurologic outcome in observational series.

Mean arterial pressure targets. Because autoregulation is impaired, higher pressures were proposed to protect the brain. Four randomized trials compared lower and higher MAP targets after OHCA, and none found better survival or neurologic outcome with the higher target. The largest, BOX, also randomized its 789 patients to a MAP target of 63 or 77 mm Hg; death or hospital discharge with severe disability or coma occurred in 32% and 34% (hazard ratio 1.08), and 90-day mortality was 29% and 31%. On this evidence, the 2025 AHA guideline recommends avoiding hypotension by maintaining a MAP of at least 65 mm Hg and finds insufficient evidence to recommend a specific vasopressor. The ERC-ESICM guideline accepts a MAP above 60 to 65 mm Hg as the default. It individualizes a higher target for patients with chronic hypertension or persistent signs of hypoperfusion, such as rising lactate or oliguria. Patient-specific targets based on autoregulation monitoring with near-infrared spectroscopy remain investigational, and the trials that tested higher pressures did not use them.

Vasopressors and inotropes. Norepinephrine is the usual first agent for vasodilatory or mixed hypotension and is the first-line vasopressor in the European guideline; epinephrine is an alternative, particularly in the prehospital setting when norepinephrine is unavailable. Dobutamine, or milrinone when beta-receptor response is blunted, is added when cardiac output remains low after the pressure target is reached and is commonly combined with norepinephrine because it lowers systemic vascular resistance. Doses, hemodynamic phenotypes, and the evidence comparing agents are on the cardiogenic shock page; the same principles apply after arrest, with closer attention to arrhythmia because the myocardium is irritable after defibrillation.

Fluids. Isotonic crystalloid is given for hypovolemia and in volume-responsive patients. Fluid overload worsens oxygenation and cerebral edema, so repeated assessment of volume responsiveness (dynamic indices, passive leg raising, echocardiography) is preferred to fixed volumes.

Perfusion monitoring. Pressure is a means, not the goal. Lactate clearance, urine output, mixed or central venous oxygen saturation, capillary refill, and serial echocardiography assess adequacy of output. Lactate is high immediately after ROSC in most patients and normally falls over hours; failure to clear indicates continuing hypoperfusion, an untreated cause, mesenteric ischemia, or seizures. A pulmonary artery catheter is reserved for persistent shock in which the phenotype or response to treatment is uncertain, as described on the right heart catheterization and hemodynamics page.

Arrhythmias after ROSC. Recurrent ventricular fibrillation or ventricular tachycardia in the first hours suggests ongoing ischemia, an electrolyte disturbance, a QT-prolonging drug, or an underlying substrate. Correction of potassium and magnesium, treatment of ischemia, and antiarrhythmic infusion address it; recurrent or refractory ventricular arrhythmia is one of the AHA indications for emergency cardiac intervention. There is no evidence that prophylactic antiarrhythmic infusion after ROSC improves outcome. Sinus bradycardia is expected during hypothermia and is usually tolerated without treatment when perfusion is adequate. Management of electrical storm is covered on the ventricular tachycardia and sudden cardiac death page.

Finding the cause: ECG, echocardiography, and CT after return of circulation

12-lead ECG. Every adult with ROSC has a 12-lead ECG as soon as feasible, and it is repeated. ST elevation recorded in the first minutes after ROSC can be transient, produced by the arrest, defibrillation, and high-dose epinephrine; a repeat tracing several minutes later improves specificity for acute occlusion. The European guideline bases the decision for emergency PCI on ischemic changes that persist on the repeat ECG. Conversely, the absence of ST elevation does not exclude acute occlusion. Left circumflex and posterior occlusions are often electrically silent. Left bundle branch block and ventricular pacing obscure ischemia, and diffuse ST depression with ST elevation in aVR can reflect global ischemia from low output. The ECG also shows the clues to a noncoronary substrate: QT prolongation (common after arrest and during hypothermia, and reassessed after recovery), a Brugada pattern, pre-excitation, epsilon waves, and early repolarization.

Echocardiography. Bedside echocardiography identifies regional wall motion abnormalities, global ventricular dysfunction, pericardial effusion and tamponade, right ventricular dilation from pulmonary embolism, severe valve disease, aortic root dilation or a dissection flap, hypertrophic cardiomyopathy, and hypovolemia. The 2025 AHA guideline states that echocardiography or point-of-care ultrasound may be reasonable after ROSC to identify diagnoses requiring intervention; the ERC-ESICM guideline recommends echocardiography as soon as possible in all patients. Serial studies over 24 to 72 hours document recovery from stunning.

CT from head to pelvis. About one-third of OHCA cases have a noncardiac cause. Whole-body CT detects intracranial hemorrhage, pulmonary embolism, aortic dissection, pneumothorax, aspiration, and abdominal catastrophes, and identifies injuries from resuscitation, which are frequent after prolonged mechanical or manual compressions: rib and sternal fractures, pneumothorax, hemothorax, mediastinal hematoma, and liver or spleen laceration. The 2025 AHA guideline states that head-to-pelvis CT may be reasonable to investigate the cause and resuscitation-related complications. The ERC-ESICM guideline supports CT when the cause is uncertain, with a CT pulmonary angiogram when the history suggests pulmonary embolism. A noncontrast head CT in the first hours is performed mainly to exclude hemorrhage; early CT has limited value for prognosis, which depends on later imaging.

Pulmonary embolism. Arrest from massive pulmonary embolism presents most often with pulseless electrical activity. After ROSC, persistent right ventricular failure is treated with systemic fibrinolysis, catheter-directed therapy, or surgical embolectomy according to bleeding risk and local resources, and anticoagulation is started; the options are discussed on the pulmonary embolism page.

Emergency coronary angiography after cardiac arrest

Coronary disease is the most common identifiable cause of OHCA in adults, particularly with a shockable rhythm. Angiographic series after OHCA find significant coronary lesions in roughly 40% to 60% of patients and an acute culprit in a smaller proportion of those without ST elevation. In the PEARL trial of patients without ST elevation, early angiography found a culprit vessel in 47% and an acutely occluded culprit in 14%. Randomized trials have defined which patients need angiography immediately.

ST elevation after ROSC. The 2025 ACC/AHA acute coronary syndrome guideline recommends primary PCI to improve survival in resuscitated patients with evidence of STEMI who are not comatose, or who are comatose with favorable prognostic features (Class 1). In comatose patients with STEMI and unfavorable prognostic features, primary PCI may be reasonable after individualized assessment (Class 2b). Resuscitated patients with STEMI are preferentially transported by emergency medical services to a PCI-capable center (Class 1). The 2025 AHA resuscitation guideline recommends emergency coronary angiography for all patients with a suspected cardiac cause of arrest and persistent ST elevation, regardless of coma (Class 1). Coma alone does not exclude angiography, because neurologic outcome cannot be predicted at the time of ROSC and restoration of coronary flow also restores cardiac output. The guideline lists features associated with poor outcome that are weighed before proceeding in a comatose patient: an unwitnessed arrest, no bystander cardiopulmonary resuscitation, a nonshockable rhythm, cardiopulmonary resuscitation for more than 30 minutes, time to ROSC longer than 30 minutes, arterial pH below 7.2, lactate above 7 mmol/L, age above 85 years, and end-stage kidney disease on dialysis.

No ST elevation. Six contemporary randomized trials compared immediate angiography with a delayed or selective strategy in comatose OHCA survivors without ST elevation. Each found no benefit from the immediate strategy. The 2025 ACC/AHA guideline states that immediate angiography is not recommended, because of lack of benefit, in resuscitated patients who are comatose, electrically and hemodynamically stable, and without evidence of STEMI (Class 3: No Benefit). The 2025 AHA resuscitation guideline states that emergency coronary angiography is reasonable without ST elevation, regardless of coma, in selected patients with a suspected cardiac cause who have cardiogenic shock, recurrent ventricular arrhythmias, or significant ongoing myocardial ischemia (Class 2a). Without these features, emergency angiography is not recommended over a delayed or selective strategy (Class 3: No Benefit). These features were exclusion criteria or crossover triggers in the trials.

Randomized trials of immediate angiography in OHCA survivors without ST elevation. DISCO and COUPE, also cited by the 2025 ACC/AHA guideline, were likewise neutral, and meta-analysis of the trials found no survival or neurologic benefit.

Coronary angiography decision table by ECG and presentation after resuscitated cardiac arrest.

Angiography before discharge. The 2025 AHA guideline recommends coronary angiography before discharge in adult survivors with a suspected cardiac cause, particularly with an initial shockable rhythm, unexplained left ventricular systolic dysfunction, or evidence of severe ischemia. Coronary CT angiography is an alternative in younger patients with a low likelihood of atherosclerosis and also detects anomalous coronary origin.

Conduct of PCI after arrest. Radial access is preferred when the pulse is adequate, because femoral access may later be needed for mechanical support. The culprit lesion is treated; in infarct-related cardiogenic shock with multivessel disease, routine PCI of nonculprit lesions at the index procedure is harmful (Class 3: Harm), as described on the STEMI page. Oral P2Y12 inhibitors are absorbed slowly and unpredictably in intubated patients receiving opioids, during hypothermia, and with impaired gut perfusion, which delays platelet inhibition after stenting. Angiography and temperature control proceed in parallel; in COACT, the immediate-angiography group reached target temperature later than the delayed group (5.4 against 4.7 hours). Angiography is not a reason to suspend temperature management.

The full management of STEMI and non–ST-elevation acute coronary syndrome, including antithrombotic therapy, is covered on the STEMI and non–ST-elevation acute coronary syndrome pages.

Temperature control: from therapeutic hypothermia to fever prevention

Rationale for temperature control. Cerebral metabolic rate falls with each degree of cooling, and cooling also suppresses excitotoxicity, free radical generation, inflammation, apoptosis, and blood-brain barrier disruption in experimental models. Fever does the opposite and is common after ROSC; in observational series it is associated with worse neurologic outcome. These observations led to trials of induced hypothermia, then to trials of how cold, how long, and whether hypothermia adds anything to strict prevention of fever. The terminology changed with the evidence: "therapeutic hypothermia" became "targeted temperature management" after 2013, and current guidelines use "temperature control," which includes both hypothermic targets (32°C to 34°C) and normothermic or fever-prevention targets (36°C to 37.5°C).

Randomized trials of temperature targets. The two 2002 trials established induced hypothermia. The Hypothermia after Cardiac Arrest (HACA) trial randomized 275 patients after witnessed ventricular fibrillation arrest to cooling to 32°C to 34°C for 24 hours or standard care: favorable neurologic outcome at 6 months was 55% against 39% (risk ratio 1.40), and mortality was 41% against 55%. An Australian trial of 77 patients cooled to 33°C for 12 hours found good outcome in 49% against 26%. Neither control group received temperature control, and fever was frequent. The Targeted Temperature Management (TTM) trial (2013) compared 33°C with 36°C in 939 unconscious OHCA survivors of presumed cardiac cause, with fever prevented in both groups; mortality was 50% and 48%, and neurologic outcome did not differ. TTM2 (2021) compared targeted hypothermia at 33°C, followed by controlled rewarming, with targeted normothermia in which a device was started if temperature reached 37.8°C. Among 1,850 patients, 6-month mortality was 50% and 48%, and moderately severe disability or worse occurred in 55% of each group. Arrhythmia with hemodynamic compromise was more frequent with hypothermia (24% against 17%). HYPERION (2019) is the principal positive modern trial: in 581 comatose patients after arrest with a nonshockable rhythm, 33°C for 24 hours produced survival with Cerebral Performance Category 1 or 2 at 90 days in 10.2% against 5.7% with targeted normothermia at 37°C. Mortality did not differ (81.3% against 83.2%); the lower limit of the 95% confidence interval for the difference was 0.1 percentage point.

Randomized trials of temperature control after cardiac arrest. VF, ventricular fibrillation; OHCA, out-of-hospital cardiac arrest.

Duration of temperature control. TTH48 found no advantage of 48 over 24 hours of hypothermia, and the BOX fever-duration comparison found no difference between 36 and 72 hours of device-based temperature control. ICECAP, published in 2026, randomized 1,158 comatose OHCA survivors who were cooled to 33°C to durations from 6 to 72 hours with response-adaptive allocation. The trial met a prespecified stopping rule at interim analysis, and increasing the duration of cooling did not improve neurologic outcome, secondary outcomes, or mortality. ICECAP did not include a normothermic arm, so it does not address whether hypothermia itself is beneficial.

Prehospital cooling. In a trial of 1,359 patients in Seattle and King County, rapid prehospital infusion of cold saline lowered core temperature by about 1.2°C to 1.3°C by hospital arrival but did not improve survival or neurologic status, and it increased re-arrest in the field (26% against 21%) and pulmonary edema on the first chest radiograph. Routine prehospital cooling with rapid infusion of large volumes of cold intravenous fluid is not recommended by either the AHA or the ERC-ESICM guideline.

In-hospital arrest. No randomized trial has tested temperature control exclusively after IHCA. In a propensity-matched registry analysis of 26,183 IHCA patients, of whom 6.0% received therapeutic hypothermia, hypothermia was associated with lower survival to discharge (27.4% against 29.2%) and lower favorable neurologic survival. These data are observational and confounded. Guidelines apply the same temperature recommendations to comatose survivors of IHCA, and HYPERION included in-hospital arrests.

Current recommendations. The 2025 AHA guideline recommends a deliberate, protocolized strategy of temperature control for all adults who do not follow commands after ROSC, irrespective of arrest location or presenting rhythm, with a temperature maintained between 32°C and 37.5°C (both Class 1). It also states that maintaining temperature control for at least 36 hours is reasonable (Class 2a). The 36-hour minimum applies to total temperature control, and hypothermic and normothermic strategies are both within the range. The AHA retains hypothermic targets as options, cites observational data suggesting benefit in patients with more severe illness and brain injury, and states that it is unclear whether a specific temperature improves outcomes in these subgroups (Class 2b). The 2025 ERC-ESICM guideline is more restrictive. It recommends active prevention of fever, defined as a temperature above 37.5°C, for 36 to 72 hours in patients who remain comatose, recommends against routine induction of hypothermia, and recommends against active rewarming of patients with mild spontaneous hypothermia after ROSC. Within the AHA range, a normothermic target is consistent with TTM2, and a hypothermic target for patients with a nonshockable rhythm draws on HYPERION. Temperature control is applied to patients who do not follow commands; a patient who follows commands after ROSC does not require it, although fever is still treated.

Methods and devices. Core temperature is measured continuously, preferably in the esophagus, bladder (unreliable with oliguria), or pulmonary artery; rectal and tympanic readings lag behind core temperature and are less accurate. Temperature-control devices with continuous feedback, either surface systems with water-circulating gel pads or intravascular catheters, hold a target more precisely than ice packs, fans, and blankets, which overshoot and undershoot. ILCOR systematic reviews found no difference in survival or neurologic outcome between surface and intravascular methods; intravascular catheters achieve target faster and with less overshoot but add the risks of central venous access and thrombosis. Antipyretics such as acetaminophen reduce fever but cannot maintain a precise target alone. Esophageal heat-exchange devices offer another option.

Phases of temperature control. Induction aims to reach the target quickly once the decision is made, using the device, sedation, and, for hypothermic targets, limited volumes of cold fluid in the hospital when the circulation tolerates it. Maintenance holds the target with fluctuations of no more than about 0.5°C, which requires active management of shivering. Rewarming from a hypothermic target is controlled, because rapid rewarming causes vasodilation, hypotension, hyperkalemia from potassium shifting out of cells, and rebound hyperthermia. The 2025 AHA guideline suggests avoiding rewarming faster than 0.5°C per hour after hypothermic temperature control or spontaneous hypothermia (Class 2b); recent trials used 0.25°C to 0.5°C per hour, and TTM2 used about 0.33°C per hour. After rewarming or after a normothermic period, fever is prevented, with the device kept in place in a normothermic mode.

Physiologic effects of induced hypothermia and their management. Effects are more marked at 32°C to 33°C than at normothermic targets.

Shivering. Shivering is the body's defense against cooling below the hypothalamic set point; it increases oxygen consumption, energy expenditure, and CO₂ production and makes the target harder to hold. The Bedside Shivering Assessment Scale grades it from 0 to 3: 0, none; 1, mild, localized to the neck or thorax; 2, moderate, with gross movement of the upper extremities; and 3, severe, with gross movement of the trunk and upper and lower extremities. Antishivering protocols escalate from nonsedating measures (acetaminophen, buspirone, magnesium replacement, and skin counterwarming of areas not covered by the cooling device) to dexmedetomidine, opioids such as meperidine or fentanyl, deeper propofol sedation, and, last, neuromuscular blockade. Shivering is less frequent and less severe with normothermic targets, which is one of their practical advantages.

Sedation, analgesia, and neuromuscular blockade

Comatose patients receive sedation and analgesia for ventilator synchrony, comfort during endotracheal intubation, control of shivering, and reduction of cerebral metabolic demand. No sedative regimen has been shown to improve neurologic outcome. Short-acting agents (propofol, dexmedetomidine, and remifentanil or fentanyl) are preferred, because accumulation of long-acting drugs delays awakening and confounds the neurologic examination. In a randomized comparison cited by guidelines, propofol with remifentanil produced faster awakening than midazolam with fentanyl, at the cost of higher norepinephrine requirements. Hypothermia, shock, and kidney and liver dysfunction reduce clearance, so midazolam and its active metabolite, and opioids, can persist for days after the infusion stops. Sedation is reduced or interrupted as soon as temperature control and respiratory status allow, to permit neurologic assessment.

Routine continuous neuromuscular blockade is not recommended by the ERC-ESICM guideline. Blockade abolishes the clinical examination and conceals convulsive seizures, so it is reserved for refractory shivering despite deep sedation, severe hypoxemic respiratory failure, and ventilator dyssynchrony that sedation does not control. A patient who receives neuromuscular blockade is monitored with continuous EEG for that period.

Seizures, myoclonus, and EEG monitoring after cardiac arrest

Incidence and recognition. Seizures, nonconvulsive status epilepticus, and other rhythmic or periodic EEG patterns occur in roughly 20% to 30% of comatose survivors. Many are nonconvulsive, and sedation, neuromuscular blockade, and temperature control conceal clinical signs. The 2025 AHA algorithm specifies assessment for clinical seizure and EEG in patients who do not follow commands. Continuous monitoring is preferred when available; the ERC-ESICM guideline recommends EEG from the first day after ROSC, for both seizure detection and prognosis. American Clinical Neurophysiology Society (ACNS) terminology defines an electrographic seizure as epileptiform discharges faster than 2.5 Hz for 10 seconds or more, or any pattern with definite evolution lasting 10 seconds or more. Electrographic status epilepticus is a seizure lasting 10 minutes or more, or occupying more than 20% of any 60-minute period.

Seizure prophylaxis. Routine seizure prophylaxis is not recommended in adults who do not follow commands after ROSC (AHA 2025), and the ERC-ESICM guideline also recommends against it. Antiseizure drugs add sedation, hypotension, and drug interactions without proven benefit.

Treatment of seizures. Clinically evident seizures and unequivocal electrographic seizures are treated. Benzodiazepines are the first step for convulsive seizures; second-line drugs are given as for status epilepticus in other settings. Propofol and midazolam infusions also suppress seizures and are titrated with EEG in refractory cases. Observational series report favorable outcomes in a minority of patients with post-arrest status epilepticus, especially when the EEG background is continuous and reactive and brainstem reflexes and somatosensory evoked potentials (SSEPs) are preserved, so status epilepticus by itself is not treated as evidence of futility.

Rhythmic and periodic patterns. Whether to suppress patterns on the ictal-interictal continuum was tested in TELSTAR. Among 172 comatose patients with rhythmic or periodic EEG activity, detected at a median of 35 hours, an antiseizure treatment strategy suppressed the activity for 48 hours in 56% against 2% of controls, but poor outcome at 3 months occurred in 90% and 92%, and mortality was 80% and 82%. Aggressive suppression of these patterns does not improve outcome on average; the ERC-ESICM guideline treats them case by case.

Antiseizure drugs used for seizures after cardiac arrest. Loading doses are those used in the Established Status Epilepticus Treatment Trial, in which seizure cessation was similar with the three drugs; weight-based doses were capped at 75 kg.

Myoclonus. Myoclonus is common in comatose survivors and has two clinically different forms. Early generalized myoclonus, particularly status myoclonus (continuous, generalized myoclonus lasting 30 minutes or more, beginning within the first 72 hours), usually reflects severe cortical injury and is associated with a poor outcome when it accompanies a highly malignant EEG. Myoclonus with a continuous, reactive EEG background can occur in patients who recover. Lance-Adams syndrome is chronic action myoclonus in patients who have regained consciousness, and it is compatible with good cognitive recovery. EEG is recorded during the jerks to distinguish these patterns. The 2025 AHA guideline added that treatment to suppress myoclonus without an EEG correlate is not recommended, because such myoclonus has no demonstrated role in secondary brain injury. The 2023 Neurocritical Care Society (NCS) guideline states that myoclonus within 48 hours of ROSC, without concomitant EEG evaluation, is not a reliable predictor of poor outcome.

Glucose, electrolytes, kidney protection, and general intensive care

Glucose. Hyperglycemia is frequent after ROSC because of catecholamines, stress hormones, and hypothermia, and is associated with worse outcome. Tight control does not improve outcome and increases hypoglycemia. In a trial of 90 patients after ventricular fibrillation arrest, strict control (72 to 108 mg/dL) and moderate control (108 to 144 mg/dL) produced 30-day mortality of 33% and 35%, with moderate hypoglycemia in 18% and 2%. The 2025 AHA algorithm specifies avoidance of hypoglycemia (below 70 mg/dL) and hyperglycemia (above 180 mg/dL). An insulin infusion with hourly glucose measurement during temperature control is usual, and hypoglycemia is corrected immediately with intravenous glucose.

Electrolytes. Potassium falls during cooling as it shifts into cells and rises during rewarming; potassium replacement is moderated before rewarming to avoid rebound hyperkalemia. Magnesium is kept in the normal range, which also helps prevent arrhythmia and shivering. Phosphate falls with insulin, cooling, and refeeding. Severe acidemia usually improves with restored perfusion; bicarbonate is reserved for specific indications such as hyperkalemia and sodium-channel blocker toxicity.

Kidney protection and renal replacement. Acute kidney injury is frequent after cardiac arrest because of the arrest itself, shock, rhabdomyolysis, nephrotoxic drugs, and iodinated contrast from angiography and CT. It is managed by maintaining perfusion, avoiding nephrotoxins, adjusting drug doses, and using the minimum contrast needed. Contrast exposure is not a reason to withhold indicated emergency angiography. Renal replacement therapy follows the usual indications (refractory hyperkalemia, acidemia, volume overload, and uremic complications); continuous therapy is better tolerated hemodynamically and avoids osmotic shifts that worsen cerebral edema.

Infection, antibiotics, and steroids. Aspiration during arrest and intubation makes early pneumonia common. In a French trial of 198 patients cooled to 32°C to 34°C, 2 days of amoxicillin-clavulanate reduced early ventilator-associated pneumonia (19% against 34%) but did not change ventilator-free days, intensive care stay, or 28-day mortality (41% against 37%). The 2025 AHA algorithm lists antibiotics as a consideration among the elements of ongoing care, and the benefit of routine prophylaxis is uncertain. The ERC-ESICM guideline recommends against routine prophylactic antibiotics, with a low threshold for treatment when pneumonia is suspected. Fever during and after temperature control has many noninfectious causes, and procalcitonin and C-reactive protein are raised by the arrest itself, which makes the diagnosis of infection depend on cultures, imaging, and clinical course. The benefit of routine corticosteroids after ROSC is uncertain in the AHA guideline, and the ERC-ESICM guideline recommends against their routine use; trials of corticosteroids after in-hospital arrest have not shown a consistent survival benefit.

Other intensive care measures. Stress ulcer prophylaxis is recommended by the ERC-ESICM guideline because ventilation, shock, and antithrombotic therapy after PCI raise the risk of gastrointestinal bleeding. Venous thromboembolism prophylaxis follows standard intensive care practice once bleeding risk allows; intravascular cooling catheters add catheter-related thrombosis. Enteral nutrition is started early at a low (trophic) rate and advanced as tolerated. The head of the bed is elevated to reduce aspiration, and cerebral venous drainage is not obstructed by tight tube ties or extreme neck rotation.

Temporary mechanical circulatory support after return of circulation

Cardiogenic shock is both a cause and a consequence of cardiac arrest. The 2025 AHA guideline added a recommendation that temporary mechanical circulatory support may be considered in highly selected adults with refractory cardiogenic shock after cardiac arrest and ROSC. The selection problem is the brain: support restores flow but cannot reverse established anoxic injury, and neurologic status is usually unknown when the decision is made.

The randomized evidence was generated largely in populations that excluded or under-represented comatose arrest survivors. DanGer Shock showed lower 180-day mortality with a microaxial flow pump in STEMI-related cardiogenic shock (45.8% against 58.5%) with more severe bleeding, limb ischemia, and renal replacement therapy, but it excluded patients who were comatose after OHCA. ECLS-SHOCK enrolled patients with infarct-related shock planned for early revascularization. Early venoarterial extracorporeal membrane oxygenation (VA-ECMO) did not reduce 30-day mortality (47.8% against 49.0%) and was accompanied by more moderate or severe bleeding (23.4% against 9.6%) and peripheral vascular complications. The benefit of a microaxial pump has therefore not been shown in comatose arrest survivors, and routine VA-ECMO is not indicated for infarct-related shock after arrest. Device choice, hemodynamic criteria, left ventricular unloading, and complications are described on the cardiogenic shock page.

Extracorporeal cardiopulmonary resuscitation (ECPR), VA-ECMO initiated during refractory arrest, was tested in the ARREST trial from Minnesota, the Prague OHCA trial, and INCEPTION, and is covered with resuscitation on the adult cardiac arrest and advanced cardiovascular life support page. Patients who survive ECPR enter the same post-arrest pathway, with additional problems: differential hypoxemia on femoral VA-ECMO, left ventricular distension, limb ischemia, bleeding, and a higher frequency of death by neurologic criteria. The ERC-ESICM guideline addresses this with active assessment for death by neurologic criteria in comatose ECPR survivors.

Multimodal neuroprognostication after cardiac arrest

Why timing and method matter. Prognostication determines whether life-sustaining therapy continues, and errors run in both directions. A false prediction of poor outcome leads to withdrawal of treatment and death in a patient who would have recovered, an error that is invisible afterward because the prediction causes the outcome (the self-fulfilling prophecy). A false prediction of good outcome prolongs treatment the patient would not have wanted. In a secondary analysis of a large North American OHCA trial, withdrawal of life-sustaining therapy for perceived neurologic prognosis before 72 hours occurred in one-third of patients who died in hospital; modeling predicted that 26% of them would have survived, and 16% with favorable function, had withdrawal not occurred early. Extrapolated nationally, early withdrawal was estimated to be associated with about 2,300 deaths each year in the United States, nearly 1,500 of them in patients who might have recovered function.

Timing of assessment. The 2025 AHA algorithm specifies a multimodal approach with delayed prognostic assessment, at 72 hours or later from ROSC or from achieving normothermia. The corresponding recommendation (Class 2a) places the interpretation of multimodal tests at a minimum of 72 hours after normothermia and discontinuation of sedatives, typically at least 5 days after ROSC when a hypothermic target was used. The 2023 NCS guideline recommends deferring assessment for at least 72 hours after ROSC in patients not treated with hypothermia (goal temperature below 36.5°C) and for at least 72 hours after rewarming in patients who were. It also states that persistence of coma beyond this period must not be equated with a poor prognosis. Some tests are valid earlier and are collected in the first days for later integration: EEG from the first day, SSEP from 24 hours in European guidance (48 hours in the AHA and NCS guidelines), serial neuron-specific enolase (NSE) at 24, 48, and 72 hours, and CT or MRI at the intervals described below.

Confounders of the neurologic examination. The examination is performed in the absence of sedation, neuromuscular blockade, hypothermia, severe hypotension, and major metabolic disturbance (uremia, hepatic encephalopathy, hyponatremia or hypernatremia, hypoglycemia), and after seizures are controlled. Hypothermia, shock, and kidney or liver injury prolong the effects of midazolam, opioids, and propofol, sometimes for days. Neuromuscular blockade is confirmed to have worn off with a train-of-four stimulus. The NCS guideline also separates overall prognosis (baseline function, life-limiting illness, multiorgan failure) from neurologic prognosis, and states that age, initial rhythm, and time to ROSC alone are not reliable predictors of poor functional outcome.

Principles common to US and European guidance. Assessment is multimodal and never based on a single variable. Prognosis is communicated as a likelihood with stated uncertainty. When no reliable predictor, or no combination of moderately reliable predictors, is present, the NCS guideline recommends counseling surrogates that the likelihood, extent, and time course of recovery are uncertain, that recovery may take days to months, and that an extended period of observation is appropriate if it is consistent with the goals of care.

Neuroprognostication timing and predictors. NCS classifications from the 2023 Neurocritical Care Society guideline, in which reliable predictors combine higher certainty of evidence with a false-positive rate of about 3% or less in most studies; timing for ERC-ESICM from the 2025 guideline. The 2025 AHA guideline also added a high serum neurofilament light chain concentration within 72 hours, in combination with other tests, as support for a prognosis of unfavorable outcome (Class 2b).

Clinical examination. The bilaterally absent pupillary light reflex at 72 hours or later is the most specific bedside sign, but it occurs in only a minority of patients with poor outcome. Automated pupillometry measures the percentage of constriction and the neurological pupil index and removes the variability of visual assessment; the ERC-ESICM guideline recommends it when available. The corneal reflex is affected by sedation and neuromuscular blockers. The motor response (Glasgow Coma Scale motor score) is the most sensitive and least specific finding. The European algorithm of 2021 selected patients with a score of 3 or less at 72 hours for prognostic testing. The 2025 update applies to all patients who remain comatose, with a motor score of 5 or less (that is, all who do not obey commands). It counts a score of 4 or 5 as a favorable sign.

Somatosensory evoked potentials. Stimulation of the median nerve normally produces a cortical N20 response. Its bilateral absence, with preserved peripheral and cervical responses to confirm the stimulus reached the cord, is the most specific electrophysiologic predictor of poor outcome, with a false-positive rate below 3% in the NCS evidence review. SSEPs are relatively resistant to sedatives but are degraded by muscle artifact, so a short-acting neuromuscular blocker is sometimes given during recording. A preserved N20 does not predict good outcome; a high N20 amplitude is associated with recovery.

Electroencephalography. EEG serves both seizure detection and prognosis. Standardized ACNS terminology classifies the background by continuity, voltage, and reactivity. Highly malignant patterns are a suppressed background (below 10 µV) without discharges, a suppressed background with continuous periodic discharges, and burst suppression (suppression occupying 50% to 99% of the record), with or without discharges. In a multicenter cohort using this classification, highly malignant patterns after 24 hours predicted poor outcome with high specificity. Malignant patterns (abundant periodic or rhythmic discharges, unequivocal seizures, discontinuous or low-voltage background, and an unreactive background) are less specific. A continuous, normal-voltage background without discharges is the principal favorable sign. The 2025 AHA guideline added that using a continuous, normal-voltage background within 72 hours, together with other tests, to support a prognosis of favorable outcome may be reasonable. Lack of EEG reactivity alone is not used to predict poor outcome, because interobserver agreement is limited.

Serum biomarkers. NSE is released from injured neurons. Values above 60 µg/L at 48 or 72 hours, or a rise between 24 and 48 or 72 hours, support a poor prognosis in the ERC-ESICM algorithm. A value of 17 µg/L or less within 72 hours is a favorable sign. NSE is also present in red cells and platelets, so hemolysis (common with extracorporeal circuits) falsely raises it, and assays differ. The NCS guideline therefore does not regard NSE as reliable alone. Neurofilament light chain, glial fibrillary acidic protein, and tau show high specificity in research cohorts but lack standardized thresholds and widely available assays.

Brain imaging. Generalized edema with loss of gray-white differentiation and sulcal effacement, visible from about 48 hours, indicates severe injury; the gray-white matter ratio quantifies it, but no threshold has been validated across scanners and timing. Diffusion-weighted MRI between 2 and 7 days shows the distribution of cytotoxic edema; diffuse restriction in the cortex and deep gray matter supports a poor prognosis, while normal or limited findings support recovery. Repeat CT at 72 to 96 hours is useful in patients who remain comatose after a normal initial study.

The European algorithm and favorable signs. The ERC-ESICM algorithm, introduced in 2021 and revised in 2025, applies to patients who remain comatose (Glasgow Coma Scale motor score of 5 or less) at 72 hours or later, after confounders are excluded. Poor outcome is considered likely when at least two of the following are present: bilaterally absent pupillary and corneal reflexes at 72 hours or later; bilaterally absent N20 SSEP waves; a highly malignant EEG after 24 hours; NSE above 60 µg/L at 48 or 72 hours; status myoclonus within 72 hours; and diffuse, extensive anoxic injury on CT or MRI. The 2025 guideline added favorable signs: a motor score of 4 or 5, NSE of 17 µg/L or less within 72 hours, a continuous, normal-voltage EEG background within 72 hours, and absent or limited diffusion restriction on MRI at 2 to 7 days. With fewer than two predictors, one or more favorable signs make a good outcome likely, and the likelihood rises with the number of favorable signs. With neither, the outcome is indeterminate, and observation continues with repeated assessment.

Communication with surrogates. Family meetings begin early, explain the syndrome and the expected timeline, and set expectations that prognosis will be addressed after 72 hours and after sedation has cleared. The prognosis is then presented with its basis and its uncertainty, and decisions about continuing or withdrawing treatment are made with surrogates on the basis of the patient's known values. Palliative care consultation supports these discussions and does not imply withdrawal.

Neuroprognostication predictor summary

Predictors of poor outcome and favorable signs from the 2025 ERC-ESICM post-resuscitation guideline; reliability classes from the 2023 Neurocritical Care Society guideline (pupillary reflex and N20 SSEP reliable; EEG background, CT, and MRI moderately reliable; NSE and status myoclonus not graded as reliable alone). A motor response of 4 or 5 counts as a favorable sign. The summary organizes findings for a multidisciplinary assessment; it does not determine goals of care.

Death by neurologic criteria, withdrawal of life-sustaining therapy, and organ donation

Death by neurologic criteria. A minority of comatose survivors progress to death by neurologic criteria, usually from cerebral edema and herniation over the first days; the proportion is higher after ECPR. Determination follows the 2023 consensus guideline of the American Academy of Neurology, American Academy of Pediatrics, Child Neurology Society, and Society of Critical Care Medicine, which added guidance for patients treated with temperature control and ECMO. The prerequisites include an established cause of catastrophic brain injury, exclusion of confounding drugs and metabolic disturbance, and a core temperature maintained at 36°C or higher. When core temperature has been 35.5°C or lower, as during hypothermic temperature control, the evaluation waits at least 24 hours after rewarming to 36°C or higher, and the clearance of sedating drugs is judged with allowance for slower elimination during hypothermia. The examination documents coma, absent brainstem reflexes, and apnea; ancillary tests are used when the examination or apnea test cannot be completed safely. Death by neurologic criteria is distinct from a poor neurologic prognosis and is not inferred from prognostic tests.

Withdrawal of life-sustaining therapy. Most deaths after admission for OHCA follow a decision to withdraw treatment. The decision rests on the multimodal prognosis, the patient's previously expressed wishes, and the overall medical condition, and it is documented with its basis. Withdrawal for reasons other than neurologic prognosis, such as refractory shock, multiorgan failure, or a life-limiting comorbidity, is distinguished from withdrawal for perceived neurologic prognosis.

Organ donation. Patients who regain circulation after cardiac arrest and then progress to death by neurologic criteria, or in whom life-sustaining treatment is withdrawn, are evaluated as potential organ donors, either after neurologic determination of death or after circulatory determination of death. Organs from donors who received cardiopulmonary resuscitation function comparably to organs from donors who did not. The ERC-ESICM guideline recommends evaluation of all such patients and reporting of donation outcomes in cardiac arrest registries. Donation discussions are separated from prognostic and withdrawal decisions and follow local legal and organ procurement organization requirements.

Finding the cause in survivors without coronary disease

When angiography or coronary CT excludes an acute coronary cause, the arrest is investigated for a structural, inflammatory, infiltrative, or primary electrical substrate before discharge. The sequence, set out in the 2017 AHA/ACC/HRS ventricular arrhythmia guideline and summarized on the ventricular tachycardia and sudden cardiac death page, includes repeated 12-lead ECGs after recovery, echocardiography, and cardiac MRI for scar, myocarditis, sarcoidosis, arrhythmogenic cardiomyopathy, and infiltration. It also includes exercise testing for catecholaminergic polymorphic ventricular tachycardia and long QT syndrome, sodium-channel blocker challenge for Brugada syndrome, ambulatory monitoring, and toxicology. Genetic testing is directed by the phenotype and accompanied by genetic counseling and evaluation of first-degree relatives, as described on the genetic testing and counseling in cardiovascular disease page. When the evaluation is negative, the arrest is classified as idiopathic ventricular fibrillation, and the possibility that a substrate will declare itself later is addressed with follow-up and family screening.

Secondary prevention, defibrillator decisions, and cardiac rehabilitation

Defibrillator. In survivors of cardiac arrest from ventricular fibrillation or hemodynamically unstable ventricular tachycardia without a completely reversible cause, a secondary-prevention implantable cardioverter-defibrillator (ICD) is recommended when meaningful survival of more than a year is expected (Class 1 in the 2017 AHA/ACC/HRS guideline). Ventricular fibrillation within the first 48 hours of an acute myocardial infarction is treated as a consequence of the infarct, and the defibrillator decision then follows the post-infarction ejection fraction pathway with reassessment of ventricular function. Electrolyte abnormalities and drugs are seldom accepted as the sole cause without careful evaluation. Implantation follows neurologic recovery and a clear understanding of the patient's goals. Device selection, the wearable defibrillator, programming, and driving restrictions are covered on the ventricular tachycardia and sudden cardiac death page.

Treatment of the underlying disease. Survivors with coronary disease receive full secondary prevention: antiplatelet therapy, high-intensity lipid lowering, beta-blockade after infarction, renin-angiotensin system inhibition, mineralocorticoid receptor antagonism, and other guideline-directed therapy for a reduced ejection fraction, and treatment of risk factors. Survivors with a cardiomyopathy or channelopathy receive disease-specific therapy. Cardiac rehabilitation is appropriate after arrest with an ischemic or structural cause and supports physical reconditioning and return to activity.

Survivorship: cognitive, emotional, and caregiver recovery

Recovery continues for months after discharge, and survival with a favorable score on a crude outcome scale does not mean recovery without deficits. The 2020 AHA scientific statement on survivorship describes a trajectory from acute care through rehabilitation, return home, and long-term follow-up, and lists the sequelae survivors experience: cognitive impairment, most often in memory, attention, processing speed, and executive function; fatigue; anxiety, depression, and post-traumatic stress; reduced participation in work and social roles; and physical deconditioning, critical illness myopathy, and injuries from resuscitation.

The 2025 AHA guideline recommends structured assessment of survivors and their caregivers for emotional distress, with treatment or referral, after medical stabilization and before discharge. About one in four survivors and caregivers experiences emotional distress, and studies of survivor-caregiver pairs found improvement with psychosocial interventions. AHA guidance also calls for assessment of physical, neurologic, cardiopulmonary, and cognitive impairments before discharge, rehabilitation according to the needs identified, and a multidisciplinary discharge plan. The ERC-ESICM guideline adds structured follow-up within 3 months of discharge, with screening for cognitive, emotional, and physical problems, inclusion of co-survivors (family and caregivers), and information on resuming driving, work, and relationships.

Functional outcome scales used in cardiac arrest trials. A favorable outcome is usually defined as CPC 1–2 or mRS 0–3; TTM2 defined poor functional outcome as mRS 4–6.

Special populations: in-hospital arrest, pregnancy, asphyxial and toxic arrest, and older adults

In-hospital arrest. Survivors of IHCA have more comorbidity, more nonshockable rhythms, and more noncardiac causes, so the search for the cause centers on respiratory failure, sepsis, hemorrhage, pulmonary embolism, and drugs. The same physiologic targets, temperature control, and prognostic standards apply. Coronary angiography is directed by the ECG and clinical picture, with the same restrictions in the stable patient without ST elevation.

Pregnancy. After ROSC in a pregnant patient who has not been delivered, left lateral uterine displacement continues to relieve aortocaval compression. The 2020 AHA adult advanced life support guideline recommends targeted temperature management for pregnant patients who remain comatose after resuscitation, with continuous fetal monitoring for bradycardia and obstetric and neonatal consultation during treatment (both Class 1). The causes specific to pregnancy (amniotic fluid embolism, hemorrhage, preeclampsia and eclampsia, magnesium toxicity, peripartum cardiomyopathy, and coronary artery dissection) are part of the search for the cause.

Asphyxial arrest. Arrest from drowning, hanging, airway obstruction, or severe asthma is preceded by a period of profound hypoxemia with continued perfusion, which produces more severe brain injury for a given duration of circulatory arrest and is often associated with a nonshockable rhythm. Acute respiratory distress syndrome is frequent after drowning.

Toxicologic arrest. Opioids, sedative-hypnotics, and other drugs cause arrest through respiratory depression or direct cardiotoxicity. After ROSC, specific antidotes and elimination are added to standard care, and prognostication is delayed until the drugs have cleared, because they mimic or exaggerate the clinical and EEG findings of brain injury.

Older adults. Age is associated with outcome, but the NCS guideline states that age alone is not a reliable predictor of poor functional outcome. In the ACS guideline, age above 85 years is one of several unfavorable features weighed before primary PCI in a comatose patient, not a contraindication.

Systems of care: cardiac arrest centers, protocols, and bundles

Post-arrest care involves emergency medicine, cardiology, critical care, neurology, neurophysiology, radiology, and rehabilitation, and outcomes depend on how reliably the elements are delivered. Cardiac arrest centers combine continuous access to a catheterization laboratory, intensive care with temperature control, EEG and neuroprognostication expertise, mechanical circulatory support, and survivorship follow-up. The Class 1 recommendation to transport resuscitated patients with STEMI to a PCI-capable center (2025 ACC/AHA ACS guideline) rests on registry data associating such transfer with improved survival. For patients without ST elevation, the London ARREST trial randomized 862 patients after OHCA without ST elevation to expedited transfer to a cardiac arrest center or to the nearest emergency department; 30-day mortality was 63% in both groups (risk ratio for survival 1.00). Transfer of all patients without ST elevation to specialized centers therefore did not reduce early mortality in an urban system with short distances.

Written protocols and bundles make delivery of these targets reliable: order sets for oxygen, ventilation, pressure, glucose, and temperature; a checklist for the search for the cause; a sedation and EEG plan; a prognostication protocol that fixes timing and required tests; and a discharge pathway covering the defibrillator decision, family screening, cognitive and emotional assessment, and follow-up. Registry participation, such as the Cardiac Arrest Registry to Enhance Survival (CARES) and Get With The Guidelines–Resuscitation in the United States, supports measurement of these processes.

Sources of error in post-cardiac arrest care

• Leaving the inspired oxygen at 100% after reliable saturation or PaO₂ measurement is available, or accepting a pulse oximetry value near 90% without arterial confirmation in a patient with dark skin.
• Hyperventilating a cooled patient, producing hypocapnia and cerebral vasoconstriction because CO₂ production falls with temperature.
• Accepting brief or recurrent hypotension in the first hours, when autoregulation is impaired.
• Diagnosing STEMI or excluding it from a single ECG recorded within minutes of ROSC, without a repeat tracing.
• Withholding angiography from a patient with ST elevation solely because of coma, or performing immediate angiography in a stable comatose patient without ST elevation, where trials found no benefit.
• Suspending temperature control during transport or catheterization, or allowing fever after the initial period of temperature control.
• Rewarming rapidly from a hypothermic target and causing hypotension and hyperkalemia, or continuing full potassium replacement during rewarming.
• Giving rapid large-volume cold fluid before hospital arrival, which increased re-arrest and pulmonary edema without benefit.
• Treating myoclonus or rhythmic EEG patterns as proof of futility, or suppressing myoclonus without an EEG correlate.
• Assigning a poor prognosis before 72 hours, while sedatives are still active after hypothermia or organ failure, or from a single test such as an absent motor response, a high NSE, or an unreactive EEG.
• Interpreting a long QT interval recorded during hypothermia or within the first day as long QT syndrome.
• Omitting the search for a noncoronary cause and family screening when coronary angiography is normal.
• Discharging a survivor without cognitive and emotional screening, a defibrillator decision, and a follow-up plan.

References

1. Hirsch KG, Amorim E, Coppler PJ, et al. Part 11: post–cardiac arrest care: 2025 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2025;152(16 suppl 2):S673–S718. doi:10.1161/CIR.0000000000001375
2. Nolan JP, Sandroni C, Cariou A, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines 2025: post-resuscitation care. Resuscitation. 2025;215(suppl 1):110809. doi:10.1016/j.resuscitation.2025.110809 (also published in Intensive Care Med. 2025;51:2213–2288; corrigendum in Resuscitation. 2026;228:111315, doi:10.1016/j.resuscitation.2026.111315)
3. Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines 2021: post-resuscitation care. Intensive Care Med. 2021;47:369–421. doi:10.1007/s00134-021-06368-4
4. Perman SM, Elmer J, Maciel CB, et al. 2023 American Heart Association focused update on adult advanced cardiovascular life support: an update to the American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2024;149:e254–e273. doi:10.1161/CIR.0000000000001194
5. Panchal AR, Bartos JA, Cabañas JG, et al. Part 3: adult basic and advanced life support: 2020 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2020;142(16 suppl 2):S366–S468. doi:10.1161/CIR.0000000000000916
6. Rao SV, O'Donoghue ML, Ruel M, et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI guideline for the management of patients with acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2025;151:e771–e862. doi:10.1161/CIR.0000000000001309 (corrections in Circulation. 2025;151:e865, 2025;151:e1098, and 2025;152:e402)
7. Rajajee V, Muehlschlegel S, Wartenberg KE, et al. Guidelines for neuroprognostication in comatose adult survivors of cardiac arrest. Neurocrit Care. 2023;38:533–563. doi:10.1007/s12028-023-01688-3
8. Neumar RW, Nolan JP, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication: a consensus statement from the International Liaison Committee on Resuscitation. Circulation. 2008;118:2452–2483. doi:10.1161/CIRCULATIONAHA.108.190652
9. Sawyer KN, Camp-Rogers TR, Kotini-Shah P, et al. Sudden cardiac arrest survivorship: a scientific statement from the American Heart Association. Circulation. 2020;141:e654–e685. doi:10.1161/CIR.0000000000000747
10. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med. 2002;346:549–556. doi:10.1056/NEJMoa012689
11. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346:557–563. doi:10.1056/NEJMoa003289
12. 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;369:2197–2206. doi:10.1056/NEJMoa1310519
13. Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus normothermia after out-of-hospital cardiac arrest (TTM2). N Engl J Med. 2021;384:2283–2294. doi:10.1056/NEJMoa2100591
14. Lascarrou JB, Merdji H, Le Gouge A, et al. Targeted temperature management for cardiac arrest with nonshockable rhythm (HYPERION). N Engl J Med. 2019;381:2327–2337. doi:10.1056/NEJMoa1906661
15. Kirkegaard H, Søreide E, de Haas I, et al. Targeted temperature management for 48 vs 24 hours and neurologic outcome after out-of-hospital cardiac arrest: a randomized clinical trial. JAMA. 2017;318:341–350. doi:10.1001/jama.2017.8978
16. Le May M, Osborne C, Russo J, et al. Effect of moderate vs mild therapeutic hypothermia on mortality and neurologic outcomes in comatose survivors of out-of-hospital cardiac arrest: the CAPITAL CHILL randomized clinical trial. JAMA. 2021;326:1494–1503. doi:10.1001/jama.2021.15703
17. Meurer WJ, Yeatts SD, Geocadin RG, et al. Duration of therapeutic hypothermia after out-of-hospital cardiac arrest: the ICECAP randomized clinical trial. JAMA. 2026;336:857–867. doi:10.1001/jama.2026.10247
18. Hassager C, Schmidt H, Møller JE, et al. Duration of device-based fever prevention after cardiac arrest. N Engl J Med. 2023;388:888–897. doi:10.1056/NEJMoa2212528
19. Kim F, Nichol G, Maynard C, et al. Effect of prehospital induction of mild hypothermia on survival and neurological status among adults with cardiac arrest: a randomized clinical trial. JAMA. 2014;311:45–52. doi:10.1001/jama.2013.282173
20. Chan PS, Berg RA, Tang Y, et al. Association between therapeutic hypothermia and survival after in-hospital cardiac arrest. JAMA. 2016;316:1375–1382. doi:10.1001/jama.2016.14380
21. Kjaergaard J, Møller JE, Schmidt H, et al. Blood-pressure targets in comatose survivors of cardiac arrest. N Engl J Med. 2022;387:1456–1466. doi:10.1056/NEJMoa2208687
22. Schmidt H, Kjaergaard J, Hassager C, et al. Oxygen targets in comatose survivors of cardiac arrest. N Engl J Med. 2022;387:1467–1476. doi:10.1056/NEJMoa2208686
23. Kilgannon JH, Jones AE, Shapiro NI, et al. Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA. 2010;303:2165–2171. doi:10.1001/jama.2010.707
24. Bernard SA, Bray JE, Smith K, et al. Effect of lower vs higher oxygen saturation targets on survival to hospital discharge among patients resuscitated after out-of-hospital cardiac arrest: the EXACT randomized clinical trial. JAMA. 2022;328:1818–1826. doi:10.1001/jama.2022.17701
25. Eastwood G, Nichol AD, Hodgson C, et al. Mild hypercapnia or normocapnia after out-of-hospital cardiac arrest (TAME). N Engl J Med. 2023;389:45–57. doi:10.1056/NEJMoa2214552
26. Sjoding MW, Dickson RP, Iwashyna TJ, et al. Racial bias in pulse oximetry measurement. N Engl J Med. 2020;383:2477–2478. doi:10.1056/NEJMc2029240
27. Lemkes JS, Janssens GN, van der Hoeven NW, et al. Coronary angiography after cardiac arrest without ST-segment elevation (COACT). N Engl J Med. 2019;380:1397–1407. doi:10.1056/NEJMoa1816897
28. Desch S, Freund A, Akin I, et al. Angiography after out-of-hospital cardiac arrest without ST-segment elevation (TOMAHAWK). N Engl J Med. 2021;385:2544–2553. doi:10.1056/NEJMoa2101909
29. Kern KB, Radsel P, Jentzer JC, et al. Randomized pilot clinical trial of early coronary angiography versus no early coronary angiography after cardiac arrest without ST-segment elevation: the PEARL study. Circulation. 2020;142:2002–2012. doi:10.1161/CIRCULATIONAHA.120.049569
30. Hauw-Berlemont C, Lamhaut L, Diehl JL, et al. Emergency vs delayed coronary angiogram in survivors of out-of-hospital cardiac arrest: results of the randomized, multicentric EMERGE trial. JAMA Cardiol. 2022;7:700–707. doi:10.1001/jamacardio.2022.1416
31. Patterson T, Perkins GD, Perkins A, et al. Expedited transfer to a cardiac arrest centre for non-ST-elevation out-of-hospital cardiac arrest (ARREST): a UK prospective, multicentre, parallel, randomised clinical trial. Lancet. 2023;402:1329–1337. doi:10.1016/S0140-6736(23)01351-X
32. Ruijter BJ, Keijzer HM, Tjepkema-Cloostermans MC, et al. Treating rhythmic and periodic EEG patterns in comatose survivors of cardiac arrest (TELSTAR). N Engl J Med. 2022;386:724–734. doi:10.1056/NEJMoa2115998
33. Kapur J, Elm J, Chamberlain JM, et al. Randomized trial of three anticonvulsant medications for status epilepticus (ESETT). N Engl J Med. 2019;381:2103–2113. doi:10.1056/NEJMoa1905795
34. Westhall E, Rossetti AO, van Rootselaar AF, et al. Standardized EEG interpretation accurately predicts prognosis after cardiac arrest. Neurology. 2016;86:1482–1490. doi:10.1212/WNL.0000000000002462
35. Elmer J, Torres C, Aufderheide TP, et al. Association of early withdrawal of life-sustaining therapy for perceived neurological prognosis with mortality after cardiac arrest. Resuscitation. 2016;102:127–135. doi:10.1016/j.resuscitation.2016.01.016
36. François B, Cariou A, Clere-Jehl R, et al. Prevention of early ventilator-associated pneumonia after cardiac arrest. N Engl J Med. 2019;381:1831–1842. doi:10.1056/NEJMoa1812379
37. Oksanen T, Skrifvars MB, Varpula T, et al. Strict versus moderate glucose control after resuscitation from ventricular fibrillation. Intensive Care Med. 2007;33:2093–2100. doi:10.1007/s00134-007-0876-8
38. Badjatia N, Strongilis E, Gordon E, et al. Metabolic impact of shivering during therapeutic temperature modulation: the Bedside Shivering Assessment Scale. Stroke. 2008;39:3242–3247. doi:10.1161/STROKEAHA.108.523654
39. Møller JE, Engstrøm T, Jensen LO, et al. Microaxial flow pump or standard care in infarct-related cardiogenic shock (DanGer Shock). N Engl J Med. 2024;390:1382–1393. doi:10.1056/NEJMoa2312572
40. Thiele H, Zeymer U, Akin I, et al. Extracorporeal life support in infarct-related cardiogenic shock (ECLS-SHOCK). N Engl J Med. 2023;389:1286–1297. doi:10.1056/NEJMoa2307227
41. Greer DM, Kirschen MP, Lewis A, et al. Pediatric and adult brain death/death by neurologic criteria consensus guideline. Neurology. 2023;101:1112–1132. doi:10.1212/WNL.0000000000207740 (erratum in Neurology. 2024;102:e208108)
42. Jeejeebhoy FM, Zelop CM, Lipman S, et al. Cardiac arrest in pregnancy: a scientific statement from the American Heart Association. Circulation. 2015;132:1747–1773. doi:10.1161/CIR.0000000000000300
43. Al-Khatib SM, Stevenson WG, Ackerman MJ, et al. 2017 AHA/ACC/HRS guideline for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Circulation. 2018;138:e272–e391. doi:10.1161/CIR.0000000000000549