The ECG and echocardiogram should immediately alter treatment rather than merely classify the arrest. Regional left-ventricular dysfunction may support an acute coronary syndrome; severe global dysfunction suggests post-arrest myocardial stunning; right-ventricular dilation may raise concern for pulmonary embolism; and tamponade or profound hypovolemia demands cause-specific intervention. At the same time, the neurologic pathway divides according to whether the patient is reliably awake and following commands. A patient who wakes, protects the airway, and ventilates adequately does not need intubation or temperature-control therapy solely because cardiac arrest occurred. A patient who remains unable to follow verbal commands needs a controlled physiologic platform: reliable oxygenation and ventilation, stable cerebral perfusion, deliberate temperature management, prompt electroencephalographic surveillance, and repeated examination without premature prognostic conclusions.

Immediately after return of spontaneous circulation, 100% inspired oxygen remains appropriate until oxygenation can be measured reliably. During low-flow shock, pulse oximetry may lag, lose its signal, or overestimate arterial saturation, with an additional risk of occult hypoxemia in patients with darker skin pigmentation. Once a dependable waveform and preferably an arterial blood gas are available, oxygen should be titrated rather than left at an inspired fraction of 1.0. The 2025 American Heart Association guideline defines a reasonable target as an arterial saturation of 90% to 98% or a PaO₂ of 60 to 105 mm Hg, whereas the 2025 European guideline generally targets 94% to 98% or a PaO₂ of 75 to 100 mm Hg. In practice, the narrower 94% to 98% range is a defensible early bedside target when it can be achieved without instability. The lower AHA boundary is best understood as the edge of an acceptable range, not an instruction to aim for 90% in a newly resuscitated patient (Hirsch et al., 2025; Nolan et al., 2025).

This distinction matters because hypoxemia is an immediate, proven threat, whereas the clinical importance of modest hyperoxemia is less certain. In the EXACT trial, targeting a saturation of 90% to 94% during prehospital and emergency care did not improve outcomes and produced more hypoxemic episodes than targeting 98% to 100%; the prematurely terminated trial also showed a concerning numerical reduction in survival in the lower-target group. It therefore argues strongly against aggressive early oxygen weaning when the saturation signal is uncertain or fluctuating (Bernard et al., 2022). Conversely, the BOX trial found no outcome difference between PaO₂ targets of approximately 68 to 75 and 98 to 105 mm Hg among stabilized comatose survivors, suggesting that either moderate normoxemic strategy is acceptable once monitoring and circulation are reliable (Schmidt et al., 2022). The practical sequence is therefore simple: eliminate hypoxemia first, obtain an arterial measurement, and then reduce unnecessary oxygen without chasing the lower edge of the acceptable range.

Carbon dioxide requires the same deliberate control. PaCO₂ is not merely a respiratory number; it is a potent regulator of cerebral vascular tone. Hypocapnia constricts cerebral vessels and can reduce blood flow through already injured tissue. Hypercapnia dilates cerebral vessels and may improve flow when autoregulation is impaired, but it can also increase cerebral blood volume, worsen intracranial pressure, and aggravate acidosis or right-ventricular stress. Current guidelines therefore favor normocapnia, generally a PaCO₂ of 35 to 45 mm Hg, unless a specific physiologic reason supports another target. The TAME trial tested deliberate mild hypercapnia, with a target of 50 to 55 mm Hg, against normocapnia for the first 24 hours after out-of-hospital arrest. Mild hypercapnia did not improve six-month neurologic outcome, mortality, or functional recovery, so it should not be adopted as routine neuroprotection (Eastwood et al., 2023).

Post-arrest patients are particularly vulnerable to unrecognized hypocapnia. They may be manually hyperventilated during transport, placed on an excessive ventilator rate, or develop spontaneous respiratory alkalosis as sedation is lightened. End-tidal CO₂ cannot be assumed to equal arterial CO₂: low cardiac output, pulmonary embolism, aspiration, atelectasis, and increased dead space may produce a substantial PaCO₂–end-tidal gradient. An arterial blood gas should therefore establish the gradient soon after ventilation is controlled and again when hemodynamics or lung mechanics change. A rising end-tidal value can indicate improving pulmonary blood flow rather than worsening ventilation, while a very low end-tidal value can reflect shock rather than a truly low PaCO₂.

Ventilation should otherwise remain lung protective, generally using a tidal volume of 6 to 8 mL/kg predicted—not actual—body weight, enough positive end-expiratory pressure to prevent derecruitment, and plateau-pressure limitation. Aspiration, cardiogenic pulmonary edema, chest-compression injuries, and evolving acute respiratory distress syndrome are common. Positive end-expiratory pressure can reduce venous return and affect cerebral venous drainage, but abandoning it and accepting hypoxemia is usually more harmful. The response should be individualized using oxygenation, compliance, blood pressure, right- and left-ventricular function, and, where relevant, intracranial physiology. Routine hyperventilation has no role in post-arrest brain protection; it should be reserved, if needed at all, as a brief rescue measure for impending cerebral herniation while definitive treatment is organized.

The hemodynamic goal is not simply to obtain a cuff pressure that looks normal. Cerebral perfusion pressure is approximately mean arterial pressure minus intracranial pressure. Following global ischemia and reperfusion, cerebrovascular autoregulation may be impaired, narrowed, or shifted toward higher pressures, particularly in a patient with chronic hypertension. A mean arterial pressure that is adequate for one patient may therefore leave another patient’s brain, kidneys, and myocardium underperfused. Nonetheless, randomized trials have not established a universal benefit from routinely pushing every post-arrest patient to a high pressure. The 2025 AHA guideline recommends avoiding hypotension and maintaining a mean arterial pressure of at least 65 mm Hg; the European guideline uses a similar floor of greater than 60 to 65 mm Hg (Hirsch et al., 2025; Nolan et al., 2025).

In the BOX blood-pressure trial, targeting a mean pressure of approximately 77 mm Hg rather than 63 mm Hg did not reduce death, severe disability, or coma among comatose survivors of out-of-hospital arrest (Kjaergaard et al., 2022). That finding does not prove that 63 mm Hg is sufficient for every brain, nor that 77 mm Hg is harmful. It means that routinely exposing the entire population to a higher target has not demonstrated benefit. The sensible bedside approach is to use 65 mm Hg as a floor and then ask whether organ perfusion is actually adequate. Persistent oliguria, cool or mottled skin, delayed capillary refill, rising lactate, low central venous oxygen saturation, worsening acidosis, recurrent ventricular ectopy, or echocardiographic evidence of low forward flow may justify a higher individualized target. Conversely, a patient with warm extremities, falling lactate, adequate urine output, and satisfactory cardiac output may gain little from escalating vasopressors solely to reach an arbitrary number.

Post-arrest shock often changes phenotype during the first day. Early myocardial stunning can produce severe but reversible biventricular dysfunction and low output; vasoplegia driven by ischemia-reperfusion inflammation may subsequently become dominant. Small, reassessed crystalloid challenges are reasonable when preload responsiveness is plausible, but blind administration of liters can worsen pulmonary edema, right-ventricular failure, and cerebral venous congestion. Norepinephrine is commonly used when vasodilation or mixed shock predominates, although no vasopressor has proved uniquely superior specifically after cardiac arrest. If pressure is adequate but forward flow remains poor, an inotrope may be more rational than further vasoconstriction. Serial focused echocardiography, lactate trajectory, urine output, perfusion examination, and an arterial line are usually more informative than any isolated pressure value. Persistent shock should continually reopen the causal differential: ongoing coronary occlusion, pulmonary embolism, tamponade, hemorrhage, sepsis, tension pneumothorax, dynamic obstruction, or recurrent seizure should not be mislabeled as generic post-arrest syndrome.

Temperature management should be deliberate in every adult who remains unresponsive to verbal commands. The central question is no longer whether every comatose survivor must be cooled to 33°C. It is whether the team will actively prevent fever and control temperature rather than allowing an injured brain to drift through recurrent hyperthermia. The 2025 AHA guideline permits a maintained target between 32°C and 37.5°C and considers at least 36 hours of protocolized temperature control reasonable. The European guideline emphasizes active fever prevention at 37.5°C or below for 36 to 72 hours. Both reject passive observation and require reliable core-temperature measurement with a protocol capable of maintaining the selected target (Hirsch et al., 2025; Nolan et al., 2025).

The biologic case for hypothermia remains compelling: cooling reduces metabolic demand, excitotoxic signaling, free-radical generation, inflammatory activation, and several pathways of delayed cell death. Yet plausible mechanisms have not translated into a consistent clinical advantage over well-executed normothermia. TTM2 randomized approximately 1,900 comatose out-of-hospital arrest patients to 33°C hypothermia or controlled normothermia with early treatment of fever. It showed no significant difference in six-month mortality or functional outcome, while hemodynamically significant arrhythmias were more frequent with hypothermia (Dankiewicz et al., 2021). Importantly, the control arm was not neglect: fever was actively detected and treated. TTM2 therefore supports controlled normothermia, not permissive hyperthermia.

The uncertainty is greatest in nonshockable arrest. HYPERION reported a small increase in favorable 90-day neurologic outcome with 33°C after nonshockable arrest, but the absolute difference was modest, the result was statistically fragile, and mortality was unchanged (Lascarrou et al., 2019). A subsequent individual-patient meta-analysis combining HYPERION with the nonshockable subgroup from TTM2 found no improvement in mortality or functional outcome from 33°C hypothermia among out-of-hospital nonshockable arrests, including the examined clinical subgroups (Taccone et al., 2024). Hypothermia at 33°C consequently remains an allowable protocol choice rather than a proven phenotype-specific treatment. Active normothermia and fever prevention are a reasonable default for most centers, while local experience, arrest characteristics, ongoing trials, and patient-specific risks may still influence target selection.

The newest evidence further narrows what should be inferred from deep cooling. ICECAP, published in August 2026, adaptively compared different durations of 33°C hypothermia in 1,158 rapidly cooled patients after out-of-hospital arrest. Extending deep cooling beyond six hours did not improve 90-day disability or survival in either shockable or nonshockable cohorts (Meurer et al., 2026). ICECAP does not establish that six hours of cooling is superior, because it did not compare hypothermia with controlled normothermia. Nor does it justify ending temperature management after six hours: patients underwent controlled rewarming and subsequent fever prevention. Its strongest message is that longer exposure to 33°C should not be presumed to deliver more neuroprotection, especially when deeper cooling adds arrhythmia, electrolyte, infection, sedation, and shivering burdens.

A patient who arrives spontaneously between 32°C and 36°C should not be rapidly warmed simply to reach normothermia. Temperature should be brought toward the chosen target in a controlled fashion. Rapid rewarming—generally faster than 0.5°C per hour—should be avoided because vasodilation, hypotension, electrolyte shifts, rebound hyperthermia, and increased metabolic demand may all follow. Large-volume prehospital infusions of cold intravenous fluid are not recommended; they can worsen pulmonary edema and rearrest without improving neurologic outcome. Surface or intravascular devices with feedback control provide more dependable temperature stability than intermittent antipyretics alone.

Shivering is not a cosmetic problem. It increases systemic and cerebral oxygen consumption, raises catecholamine output, impedes cooling, generates electromyographic artifact on EEG, and can conceal subtle seizures. Management usually begins with counterwarming of the hands, feet, and face, adequate analgesia and sedation, and a protocolized antishivering regimen. Short-acting agents are preferable when feasible because their accumulation delays examination and prognostication. Neuromuscular blockade may be necessary for refractory shivering or ventilator dyssynchrony, but routine continuous paralysis removes clinically visible seizure activity and makes continuous EEG particularly important. Sedation and paralysis should be recorded precisely; later examiners must know the agents, doses, organ-function context, and time since discontinuation before interpreting an absent motor response or brainstem reflex (Hirsch et al., 2024).

Cooling and rewarming also redistribute electrolytes. Potassium commonly moves intracellularly during cooling and returns to the extracellular compartment during rewarming. Aggressively replacing every modest fall in potassium near the end of cooling may therefore precipitate hyperkalemia during rewarming. Magnesium and phosphate should also be followed, particularly when arrhythmia, weakness, or difficulty controlling shivering develops. Sinus bradycardia at 33°C can be an expected temperature effect and does not automatically require treatment when blood pressure, perfusion, and lactate are satisfactory. Treatment should be directed at impaired perfusion, not at normalizing the monitor.

Glucose management follows the same principle of avoiding extremes. Both hypoglycemia and marked hyperglycemia are associated with worse neurologic outcomes, but intensive insulin therapy adds dangerous glucose variability and hypoglycemia without proven neurologic benefit. Current guidance recommends avoiding glucose below 70 mg/dL and above 180 mg/dL rather than pursuing tight euglycemia. Frequent measurement is necessary during cooling, rewarming, vasopressor escalation, corticosteroid exposure, and changes in nutrition because insulin sensitivity can shift quickly. Sodium, calcium, magnesium, phosphate, acid-base status, renal function, and hepatic function should be corrected sufficiently to protect the brain and permit meaningful drug clearance, without turning minor laboratory deviations into destabilizing interventions.

Lactate should be trended as a marker of whole-body oxygen debt and the adequacy of resuscitation, not treated as an isolated neurologic prognostic test. A falling lactate is reassuring about systemic reperfusion; a persistently rising value should trigger a search for low cardiac output, vasoplegia, regional ischemia, recurrent arrest, seizures, severe shivering, impaired hepatic clearance, or an uncorrected cause of the original arrest. Neither the initial lactate nor the early neurologic examination should be used alone to conclude that recovery is impossible.

Finally, a controlled airway, normal PaCO₂, adequate perfusion, stable temperature, and corrected metabolic derangements create the conditions in which cerebral electrical activity can be interpreted. Any patient who remains unable to follow commands should undergo prompt EEG monitoring, particularly when myoclonus, unexplained autonomic episodes, gaze deviation, or episodic motor activity occurs. Clinical and electrographic seizures should be treated, but routine antiseizure prophylaxis is not supported. Paralysis may abolish the visible component of status epilepticus while electrical injury continues. The next task is therefore not to prognosticate from coma or myoclonus, but to determine whether the brain is seizing, recovering, or displaying an evolving background pattern—and to do so only after the physiologic platform has been made as stable and interpretable as possible.

References

“`

Bernard, S. A., Bray, J. E., Smith, K., Stephenson, M., Finn, J., Grantham, H., Hein, C., Masters, S., Stub, D., Perkins, G. D., Dodge, N., Martin, C., Hopkins, S., Cameron, P., & EXACT Investigators. (2022). 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, 328(18), 1818–1826. https://doi.org/10.1001/jama.2022.17701

Dankiewicz, J., Cronberg, T., Lilja, G., Jakobsen, J. C., Levin, H., Ullén, S., Rylander, C., Wise, M. P., Oddo, M., Cariou, A., Bělohlávek, J., Hovdenes, J., Saxena, M., Kirkegaard, H., Young, P. J., Pelosi, P., Storm, C., Taccone, F. S., Joannidis, M., … Nielsen, N. (2021). Hypothermia versus normothermia after out-of-hospital cardiac arrest. The New England Journal of Medicine, 384(24), 2283–2294. https://doi.org/10.1056/NEJMoa2100591

Eastwood, G., Nichol, A. D., Hodgson, C., Parke, R. L., McGuinness, S., Nielsen, N., Bernard, S., Skrifvars, M. B., Stub, D., Taccone, F. S., Archer, J., Kutsogiannis, D., Dankiewicz, J., Lilja, G., Cronberg, T., Kirkegaard, H., Capellier, G., Landoni, G., Horn, J., … Bellomo, R. (2023). Mild hypercapnia or normocapnia after out-of-hospital cardiac arrest. The New England Journal of Medicine, 389(1), 45–57. https://doi.org/10.1056/NEJMoa2214552

Hirsch, K. G., Abella, B. S., Amorim, E., Bader, M. K., Barletta, J. F., Berg, K., Callaway, C. W., Friberg, H., Gilmore, E. J., Greer, D. M., Kern, K. B., Livesay, S., May, T. L., Neumar, R. W., Nolan, J. P., Oddo, M., Peberdy, M. A., Poloyac, S. M., Seder, D., … Geocadin, R. G. (2024). Critical care management of patients after cardiac arrest: A scientific statement from the American Heart Association and Neurocritical Care Society. Circulation, 149(2), e168–e200. https://doi.org/10.1161/CIR.0000000000001163

Hirsch, K. G., Amorim, E., Coppler, P. J., Drennan, I. R., Elliott, A., Gordon, A. J., Jentzer, J. C., Johnson, N. J., Moskowitz, A., Mumma, B. E., Presciutti, A. M., Rodriguez, A. J., Yen, A., & Rittenberger, J. C. (2025). Part 11: Post–cardiac arrest care: 2025 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation, 152(Suppl. 2), S673–S718. https://doi.org/10.1161/CIR.0000000000001375

Kjaergaard, J., Møller, J. E., Schmidt, H., Grand, J., Mølstrøm, S., Borregaard, B., Venø, S., Sarkisian, L., Mamaev, D., Jensen, L. O., Nyholm, B., Høfsten, D. E., Josiassen, J., Thomsen, J. H., Thune, J. J., Obling, L. E. R., Lindholm, M. G., Frydland, M., Meyer, M. A. S., … Hassager, C. (2022). Blood-pressure targets in comatose survivors of cardiac arrest. The New England Journal of Medicine, 387(16), 1456–1466. https://doi.org/10.1056/NEJMoa2208687

Lascarrou, J.-B., Merdji, H., Le Gouge, A., Colin, G., Grillet, G., Girardie, P., Coupez, E., Dequin, P.-F., Cariou, A., Boulain, T., Brulé, N., Frat, J.-P., Asfar, P., Pichon, N., Landais, M., Plantefève, G., Quenot, J.-P., Chakarian, J.-C., Sirodot, M., … Reignier, J. (2019). Targeted temperature management for cardiac arrest with nonshockable rhythm. The New England Journal of Medicine, 381(24), 2327–2337. https://doi.org/10.1056/NEJMoa1906661

Meurer, W. J., Yeatts, S. D., Geocadin, R. G., Roy, A., Callaway, C. W., Ramakrishnan, V., Berry, S. M., Caveney, A. F., Meyer, S., Bozeman, N., Harney, D. M., Speers, M., Stevenson, V. L. W., Sozener, C. B., Daya, M. R., Beiser, D. G., Elmer, J., Khan, I. R., Hirsch, K. G., … Silbergleit, R. (2026). Duration of therapeutic hypothermia after out-of-hospital cardiac arrest: The ICECAP randomized clinical trial. JAMA. Advance online publication. https://doi.org/10.1001/jama.2026.10247

Nolan, J. P., Sandroni, C., Cariou, A., Cronberg, T., D’Arrigo, S., Haywood, K., Hoedemaekers, A., Lilja, G., Nikolaou, N., Olasveengen, T. M., Robba, C., Skrifvars, M. B., Swindell, P., & Soar, J. (2025). European Resuscitation Council and European Society of Intensive Care Medicine guidelines 2025: Post-resuscitation care. Intensive Care Medicine, 51(12), 2213–2288. https://doi.org/10.1007/s00134-025-08117-3

Schmidt, H., Kjaergaard, J., Hassager, C., Mølstrøm, S., Grand, J., Borregaard, B., Obling, L. E. R., Venø, S., Sarkisian, L., Mamaev, D., Jensen, L. O., Nyholm, B., Høfsten, D. E., Josiassen, J., Thomsen, J. H., Thune, J. J., Lindholm, M. G., Meyer, M. A. S., Winther-Jensen, M., … Møller, J. E. (2022). Oxygen targets in comatose survivors of cardiac arrest. The New England Journal of Medicine, 387(16), 1467–1476. https://doi.org/10.1056/NEJMoa2208686

Taccone, F. S., Dankiewicz, J., Cariou, A., Lilja, G., Asfar, P., Bělohlávek, J., Boulain, T., Colin, G., Cronberg, T., Frat, J.-P., Friberg, H., Grejs, A. M., Grillet, G., Girardie, P., Haenggi, M., Hovdenes, J., Jakobsen, J. C., Levin, H., Merdji, H., … Lascarrou, J.-B. (2024). Hypothermia vs normothermia in patients with cardiac arrest and nonshockable rhythm: A meta-analysis of individual patient data. JAMA Neurology, 81(2), 126–133. https://doi.org/10.1001/jamaneurol.2023.4820

“`