Refractory cardiac arrest, candidate selection, low-flow time, and emergent VA-ECMO cannulation — a time-critical activation pathway
The overwhelming majority of patients who suffer cardiac arrest achieve return of spontaneous circulation (ROSC) with high-quality conventional CPR, defibrillation, and standard advanced life support — or they do not survive despite maximal effort. A smaller group falls into neither category: their circulation cannot be restored by chest compressions and drugs alone, yet their arrest may still be survivable if flow can be restored by other means. Recognizing this "refractory" state early, and doing so within a narrow window, is the first decision point in extracorporeal CPR (eCPR).
Refractory cardiac arrest describes a patient who remains in cardiac arrest despite an adequate trial of conventional resuscitation: correct compression depth and rate, timely defibrillation for shockable rhythms, airway management, and standard resuscitation drugs, continued for a defined period without achieving ROSC.
The exact duration used to define "refractory" varies between institutions and eCPR programs, but the underlying logic is the same in every protocol: beyond a certain point, the probability that continued conventional CPR alone will produce a good outcome falls steeply, while the physiological cost of ongoing arrest — evolving global ischemia — keeps accumulating. eCPR is conceived as a rescue option for exactly this window: a bridge that can restore perfusion mechanically while the underlying cause is identified and treated.
This simulator represents that judgment with an illustrative threshold: below the threshold, conventional CPR continues; once it is crossed, eCPR activation moves onto the table for discussion — always as one input among several, never as an automatic trigger on its own.
The decision to consider eCPR is not simply "conventional CPR failed, therefore proceed." It is a time-sensitive judgment that has to be made while the resuscitation is still in progress, because eCPR itself takes time to mobilize: assembling a cannulation team, preparing equipment, and performing the procedure all add further minutes of ongoing low-flow or no-flow state before extracorporeal circulation actually begins.
Because of this built-in lag, effective eCPR programs try to recognize a potentially refractory arrest as early as possible — often triggering team activation and transport decisions well before the formal refractory threshold is reached — rather than waiting until the threshold has already passed and only then beginning to mobilize resources.
Recognizing refractoriness is necessary but not sufficient. eCPR is intended for patients whose arrest stems from a plausibly reversible cause — for example a large but treatable coronary event, a massive pulmonary embolism, severe hypothermia, or a toxicological insult — where restoring circulation mechanically buys time for a definitive intervention to fix the underlying problem.
Equally important is the judgment that the patient has a reasonable likelihood of meaningful neurologic recovery if perfusion is restored. eCPR does not reverse the biology of prolonged global ischemia; it provides flow. If the interval before flow can be re-established is already very long, or the underlying process is not one that extracorporeal support can bridge to a fix, the physiological rationale for activation weakens considerably.
Refractory arrest recognition, a plausibly reversible cause, and a reasonable chance of neurologic recovery together form the entry gate to the eCPR decision pathway. None of these alone is sufficient — they are evaluated as a whole, continuously, as the resuscitation unfolds.
Not every refractory arrest is an equally good candidate for extracorporeal support. Because eCPR is resource-intensive, invasive, and time-critical, selection criteria are used to identify the patients most likely to benefit — concentrating a scarce, high-effort intervention where it has the best chance of translating into survival with intact neurologic function.
While formal eCPR selection protocols vary across institutions, most converge on a similar set of favorable features that correlate with better outcomes when extracorporeal support is deployed:
• Witnessed arrest — someone observed the collapse, so the no-flow interval before any CPR began is known and typically short, rather than an unknown or presumed-long delay.
• Bystander CPR — chest compressions were started immediately by someone at the scene, maintaining some coronary and cerebral perfusion pressure during the interval before professional responders arrived.
• Shockable initial rhythm — ventricular fibrillation or pulseless ventricular tachycardia as the first documented rhythm is generally associated with a more favorable underlying pathophysiology (often an acute coronary event) than an initial rhythm of asystole or pulseless electrical activity.
• Short expected low-flow time — a realistic anticipated interval from arrest to establishment of extracorporeal flow that is not excessively long, based on transport distance, team availability, and case complexity.
In this simulator, each of these is represented as one of four features scored on a simple present/absent basis, and the count of favorable features present is used to illustrate overall candidacy.
Each favorable feature is, in essence, a proxy for how much viable tissue — myocardium and brain — is likely to remain by the time flow can be mechanically restored. Witnessed collapse and immediate bystander CPR both limit the duration and severity of the no-flow and low-flow intervals prior to professional care. A shockable initial rhythm suggests an electrically excitable, still-perfusable myocardium at the time of collapse, rather than a heart that has already progressed to a more advanced state of arrest.
A short anticipated low-flow time to cannulation reflects the practical reality that extracorporeal support restores circulation, but it cannot reverse the cumulative damage of a very prolonged period without adequate flow. Selecting patients where that interval can realistically be kept short maximizes the chance that the tissue eCPR is trying to save is still salvageable when flow finally begins.
Beyond the arrest-specific features, candidate selection also weighs the patient's baseline health status and overall trajectory: severe pre-existing illness, advanced age with significant frailty, or a terminal underlying condition all reduce the expected benefit of an invasive, resource-intensive rescue therapy, even when the arrest-specific features look favorable.
Conversely, a previously well patient with a witnessed, bystander-CPR-treated, shockable-rhythm arrest and a short anticipated low-flow time represents close to the ideal eCPR candidate profile — the scenario in which mechanical circulatory support has the best chance of translating into meaningful survival.
Candidate selection is a composite judgment, not a single checkbox. In this simulator, the "favorable features present" slider (0–4) is an illustrative simplification of a multi-factorial clinical assessment that in practice also includes comorbidity, baseline function, and case-specific context.
Of all the variables that influence eCPR outcomes, few are as consistently and strongly associated with neurologic recovery as the low-flow time — the interval from the onset of arrest to the point at which extracorporeal flow is actually established. This single number compresses the effect of recognition speed, team mobilization, transport, and procedural efficiency into one outcome-defining metric.
Low-flow time refers to the period during which a patient is receiving CPR-generated circulation — mechanical or manual chest compressions producing a fraction of normal cardiac output — as opposed to "no-flow time," the interval before any CPR is started at all. In the context of eCPR, the low-flow time of greatest interest is the total interval from arrest onset to the moment extracorporeal circulation is established and delivering adequate flow.
This interval is the sum of several sequential steps: time to recognize the arrest, time to begin CPR, time to reach the refractory threshold, time to activate the eCPR team, time to transport the patient (or the team) to the cannulation location, and the time required to actually perform cannulation and initiate extracorporeal flow. Each step adds to the same clock.
Even during well-performed conventional CPR, cardiac output is only a fraction of normal — enough to provide some perfusion, but not enough to prevent progressive ischemic injury, particularly to the brain, over an extended period. As the low-flow interval lengthens, the cumulative ischemic burden grows, and the probability that extracorporeal support — once finally established — will be able to reverse the trajectory declines accordingly.
This is why eCPR programs invest heavily not just in the decision criteria for who should receive extracorporeal support, but in system-level efficiency: pre-positioned equipment, dedicated cannulation teams, streamlined activation protocols, and rehearsed workflows, all aimed at compressing every link in the chain between recognition and flow.
This simulator represents low-flow time as the sum of the minutes of conventional CPR already elapsed (set by the first slider) and an illustrative fixed setup delay representing team activation and the cannulation procedure itself. This is a simplification: real setup delays vary considerably depending on institutional readiness, whether the team comes to the patient or the patient is transported, and case-specific technical difficulty.
The purpose of showing this number prominently is conceptual: it reinforces that the clock does not stop when the decision to activate eCPR is made — it keeps running through every subsequent step, and the cumulative total, not any single step in isolation, is what ultimately matters for outcome.
A short anticipated low-flow time is one of the four favorable candidate features precisely because this single interval is such a powerful, consistent predictor of eCPR outcome. Systems that minimize recognition-to-flow time systematically outperform those that do not, independent of other factors.
Once the decision to proceed has been made, the eCPR team must accomplish something technically demanding under maximal time pressure: establishing large-bore vascular access for veno-arterial extracorporeal membrane oxygenation (VA-ECMO) while chest compressions — mechanical or manual — continue uninterrupted, in a patient with no native circulation of their own.
The defining technical challenge of eCPR cannulation is that it must be performed on a patient who is receiving ongoing chest compressions — meaning the torso and, to some degree, the femoral vessels themselves are subject to continuous mechanical motion. Whenever feasible, a mechanical chest compression device is used at this stage precisely because it delivers a more consistent, reproducible compression pattern than manual compressions, making vascular access easier to obtain safely and reducing operator fatigue during a procedure that may take several minutes under pressure.
Access is most commonly obtained percutaneously in the common femoral vein and common femoral artery, guided by ultrasound and often fluoroscopy or landmark technique when imaging is not immediately available, using a modified Seldinger technique to place progressively larger dilators and finally the venous drainage and arterial return cannulae.
eCPR uses a veno-arterial (VA) ECMO configuration: deoxygenated blood is drained from the venous system (typically via a cannula advanced toward the right atrium), pumped through an oxygenator that adds oxygen and removes carbon dioxide, and then returned under pressure into the arterial system (typically via the femoral artery, sometimes with a distal perfusion cannula to protect the limb).
Because the arterial return provides forward flow independent of any native cardiac contraction, VA-ECMO can fully substitute for cardiac output during ongoing arrest — this is precisely why it is chosen for eCPR rather than a veno-venous configuration, which only supports gas exchange and depends on adequate native cardiac output to circulate the oxygenated blood.
Several practical strategies are used to make cannulation feasible on a patient receiving active compressions:
• Mechanical compression devices provide a fixed, predictable compression point and rhythm, unlike manual compressions which vary between rescuers and fatigue over time.
• Ultrasound-guided vascular access allows real-time visualization of the vessel despite the motion imparted by compressions, improving first-pass success and reducing vascular complications.
• A dedicated, rehearsed cannulation team — separate from the team managing the airway, drugs, and compressions — allows the procedure to proceed in parallel with, rather than as an interruption to, ongoing resuscitation efforts.
• Anticoagulation is typically initiated once vascular access is secured, balancing the need to prevent circuit thrombosis against ongoing bleeding risk in a patient who has just undergone, or is undergoing, an invasive procedure.
The procedure ends, and the patient formally transitions onto extracorporeal support, at the moment adequate VA-ECMO flow is confirmed and compressions can finally be stopped.
The moment extracorporeal flow is confirmed and chest compressions can be safely discontinued marks the end of the low-flow interval tracked throughout this simulator — and the beginning of a fundamentally different phase of care.
Establishing extracorporeal flow is not the endpoint of eCPR — it is the beginning of the phase where the underlying, presumed-reversible cause of arrest is actively addressed, while a comprehensive post-cardiac-arrest care bundle works to limit secondary injury and support recovery of every organ system affected by the arrest.
With circulation now mechanically supported, the clinical team can pursue definitive treatment of the presumed cause of arrest without the extreme time pressure of an unsupported low-flow state. When an acute coronary event is suspected or confirmed, this typically means urgent transfer to the cardiac catheterization laboratory for coronary angiography and percutaneous coronary intervention where indicated.
Other reversible causes identified or suspected during the initial workup — massive pulmonary embolism, severe electrolyte disturbance, tension pneumothorax, cardiac tamponade, toxicological ingestion — are likewise addressed with their own definitive interventions once the patient is hemodynamically stabilized on extracorporeal support, which removes the immediate lethality of ongoing untreated cardiac arrest from the equation.
In parallel with treating the underlying cause, standard post-cardiac-arrest care principles apply, now delivered to a patient whose circulation is mechanically supported:
• Temperature management — avoiding fever and, per institutional protocol, targeting a controlled temperature range to help limit secondary neurologic injury.
• Hemodynamic optimization — titrating ECMO flow, vasoactive medications, and volume status to achieve adequate end-organ perfusion pressure, now largely decoupled from native cardiac function.
• Ventilation and gas exchange — protective lung ventilation strategies, with the ECMO circuit itself contributing to gas exchange and allowing more conservative ventilator settings.
• Neurologic monitoring — serial neurologic assessment and, where available, continuous EEG or other neuromonitoring, while recognizing that meaningful prognostication is deliberately deferred until sufficient time has passed and confounding factors (sedation, temperature, metabolic derangement) have resolved.
• Circuit and limb management — ongoing anticoagulation monitoring to balance thrombosis and bleeding risk, and surveillance for limb ischemia distal to the arterial cannula, with a distal perfusion cannula placed if needed.
From this point, the clinical trajectory branches according to how the patient responds: some patients recover sufficient native cardiac function to be weaned from VA-ECMO over subsequent days; others may be bridged to a longer-term mechanical circulatory support device or transplantation if the underlying cardiac injury is severe and irrecoverable; and, unfortunately, some patients do not survive despite full mechanical support, particularly when the ischemic burden accumulated before flow was established proves too great.
This is precisely why every stage of this simulator — recognition, candidate selection, and above all the minimization of low-flow time — matters so much: the quality of the decisions made in the first minutes of the resuscitation shapes the range of outcomes that remain possible once the patient reaches this final, definitive-treatment phase.
eCPR does not replace treating the cause of arrest — it buys the time needed to treat it. The value of the entire pathway depends on how efficiently the earlier stages were executed, which is why time-to-cannulation is tracked as a headline metric throughout this simulator.