Air Care Series: No Heart, No Problem
/Veno-Arterial Extracorporeal Membrane Oxygenation Basics for Critical Care Transport Medicine
Background
Extracorporeal life support (ECLS) or extracorporeal membrane oxygenation (ECMO) includes veno-arterial (VA) and veno-venous (VV) ECMO configurations. VA ECMO is used primarily for patients suffering from cardiogenic shock and cardiac arrest, whereas the previously discussed VV ECMO is used in patients with acute respiratory distress syndrome (ARDS), asthma, and other pulmonary disease processes. For consideration of ECMO, the causative pathology should be reversible and unresponsive to conventional management. VA ECMO is the most advanced form of mechanical circulatory support, allowing for complete cardiopulmonary support with a greater level of support than that of other mechanical circulatory support devices, such as an intra-aortic balloon pump or an Impella (® Abiomed; Danvers, MA). Unlike other methods of support, VA ECMO is not reliant on the native function of the heart and lungs. As such, VA ECMO is less frequently limited by comorbid conditions seen in patients with cardiogenic shock, such as right ventricular failure, ARDS, or arrhythmias. VA ECMO provides temporary cardiac support as a bridge for recovery, to receive a durable mechanical support device (i.e., ventricular assist device or VAD), or to undergo cardiac transplant.
VA ECMO has been utilized for cardiopulmonary support since the 1970s. The Extracorporeal Life Support Organization (ELSO) registry has recorded 2167 cases of VA ECMO support since 2015, with the most common indication being cardiogenic shock. Overall, survival to hospital discharge was 42%, with patients suffering from congenital heart disease exhibiting the lowest survival to hospital discharge (37%) while patients with myocarditis had the highest survival to hospital discharge (65%).
Indications
VA ECMO is indicated for refractory cardiogenic shock, which is defined as a state where cardiac output is unable to sustain perfusion of vital organs despite implementation of conventional therapies. VA ECMO remains off-label and is considered a salvage therapy offered only to patients with a high risk of mortality despite aggressive conventional management.
Indications of VA ECMO are below in Table 1.
Primary cardiac failure: In patients suffering from cardiac failure, VA ECMO is considered in patients with cardiogenic shock and inadequate perfusion despite:
Adequate intravascular volume
Inotropic support and vasopressors
Mechanical support including impella and intra-aortic balloon pump when indicated
Cardiogenic shock requiring VA ECMO is most frequently the result of an acute myocardial infarction, but also utilized in myocarditis, peripartum cardiomyopathy and decompensated chronic heart failure.
Cardiac surgery with an inability to wean from cardiopulmonary bypass (post-cardiotomy syndrome)
ECPR: The American Heart Association Guidelines for Advanced Life Support recommend consideration of extracorporeal cardiopulmonary resuscitation (ECPR) for select patients in cardiac arrest refractory to conventional management when ECPR can be expeditiously implemented and supported by skilled providers. There are no universal guidelines for the use of ECPR for either in-hospital or out-of-hospital cardiac arrest, and varying inclusion criteria are used in the available literature. Although criteria are heterogenous, most studies use some variation of the following: (1)
< 75 years old
An initial shockable rhythm
Witnessed arrest with expeditious bystander CPR (with a no-flow interval < 5 minutes, i.e., a short time to bystander CPR)
Refractory arrest (sustained ventricular fibrillation or ventricular tachycardia following three rounds of ACLS, including three defibrillator shocks)
Contraindications
Contraindications are specific to individual patients based on a risk/benefit assessment by the provider. ELSO guidelines provide relative and absolute contraindications to consider. Frequently considered contraindications are in Table 2.
ECPR
A recent increase in ECPR has been observed, likely secondary to poor outcomes with conventional cardiopulmonary resuscitation, improved technology, and mobility of ECLS equipment. Fourteen percent of all adult ECLS patients reported to the ELSO registry received ECPR, including 2885 adult cases with a 29% survival rate. (4) As such, it is prudent to review the data and outcomes for this burgeoning technology.
ECPR provides cardiopulmonary support for patients in refractory cardiac arrest, preserving the heart and other organs' viability during this period and after return of spontaneous circulation (ROSC). Two prospective observational studies using propensity matching have compared conventional CPR with ECPR. (5,6) The study by Blumenstein et al evaluated 353 patients and showed a significant difference in short term (27% vs.17%; P=0.01) and long term survival (23.1% vs. 11.5%; P=0.008) with a 1:1 propensity score match. ECPR was the only significant and independent predictor of decreased mortality in a Cox regression analysis (HR 0.57, 95% CI 0.35–0.90; P=0.02). (5) The study by Chen et al, with 172 patients - 59 in the ECPR group and 113 in the CPR group - showed improved in-hospital (HR 0.51; 95% CI 0.35-0.74), 30 day (HR 0.47; 95% CI 0.28-0.77) and 1 year mortality (HR 0.53; 95% CI 0.33-0.83) after propensity matching. (6)
A systematic review of ECPR in out-of-hospital cardiac arrest (OHCA) included twenty studies with 833 patients where ECPR was performed in refractory cardiac arrest patients less than 75 years old, with an initial shockable rhythm, a witnessed arrest (with time to bystander CPR < 5 minutes), a presumably reversible cause of the cardiac arrest, and refractory arrest. (7) The systematic review found the overall survival to be 22%, with 13% having a good neurologic outcome (Cerebral Performance Category of 1–2 or Glasgow Outcome Scale of 4–5). Commonly cited exclusion criteria included the presence of a “do not resuscitate” order, severe baseline disability, severe comorbidities, and a non-cardiac cause of the arrest. (7) It should be noted that this review consisted of observational studies and case series, and as such there was significant variability in both patient selection criteria and ECPR processes.
A meta-analysis of eleven observational studies including 135 patients, 40% of whom survived to hospital discharge, found increased survival with early initiation of ECPR (< 30 min) compared to prolonged conventional CPR (> 30 min; OR 1.9). (8) Of note, this meta-analysis also included patients who obtained ROSC prior to cannulation, but were ultimately placed on ECMO for shock refractory to medical management following ROSC. The survival was higher at 48% when only evaluating patients receiving ECPR. Further, mortality was lowest in the younger cohort [OR for mortality 2.9 (17– 41 vs. 41–56 years old) and 3.4 (95% CL, 1.–9.7) compared to patients > 67 years old] and those who spent the least amount of time on the ECMO circuit (OR 0.2 for being on circuit < 2.3 days). (8)
VA ECMO Circuit/Configurations
The VA ECMO circuit is similar in many regards to a VV circuit, covered previously (http://www.tamingthesru.com/blog/air-care-series/vv-ecmo), and consists of: two cannulas (a larger drainage cannula and smaller return cannula); tubing; a pump (with an emergency manual pump); and a membrane oxygenator. A medical gas source, oxygen blender (often omitted in transport), and gas flowmeter or regulator are also incorporated into the circuit.
Follow the course of blood through the VA ECMO circuit:
Drainage or “venous” cannula - pulls blood from the venous system under a negative pressure
Pump - drives the flow of blood through the ECMO circuit
RPMs - the rotations per minute (RPM) of the ECMO pump is the only adjustable setting on the ECMO pump.
Flow rate - amount of blood moved through the ECMO circuit, which is related to RPM, intravascular volume, and cannula size.
Oxygenator - functions as the native lung, using the process of diffusion to oxygenate the blood and remove carbon dioxide (CO2).
Sweep gas - the flow rate of the sweep gas delivered through the oxygenator affects CO2 removal; increasing the sweep gas rate clears more CO2.
FdO2 - The fraction of delivered oxygen in the sweep gas can affect oxygenation, as does the ECMO flow. Because FdO2 is usually set at 100% after ECLS initiation and during transport, the best way to affect oxygenation is by adjusting ECMO flow rate.
Return or “arterial” cannula - returns oxygenated blood back to the arterial system under positive pressure
Reperfusion cannula - in peripheral VA ECMO setups, it is common to have a reperfusion cannula inserted into the femoral artery just distal to the return cannula to prevent limb ischemia from developing due to the relatively large diameter of the return cannula potentially occluding flow to the limb
Because VA ECMO returns oxygenated blood in a retrograde fashion into the aorta, cardiac afterload is greatly increased in an already failing heart. As such, LV distension and subsequent pulmonary congestion are common complications, and many centers often place a preemptive mechanism to offload the LV, called an “LV vent”. This could be a cannula placed directly into the LV for centrally cannulated patients, or a second mechanical device, such as an Impella or intra-aortic balloon pump, in patients cannulated peripherally.
VA ECMO Cannulation configurations
Unlike VV ECMO, in which all cannulation is done peripherally, VA ECMO can be performed through either peripheral or central access. (9)
Peripheral VA ECMO is typically performed with a drainage cannula in the femoral vein or inferior vena cava and a return cannula in the femoral or, less commonly, axillary artery.
Central VA ECMO is typically seen in patients following a failure to wean from cardiopulmonary bypass, with drainage typically from the right atrium and return into the ascending aorta. Central VA ECMO is performed via sternotomy, and these patients will often have an open chest at the time of transport.
Post-ECLS Management and Monitoring
In VA ECMO, the ECMO circuit accounts for a majority of the cardiac output, often resulting in a lack of pulsatility or arterial waveform.
Blood pressure should be monitored continuously through an arterial line with mean blood pressure (MAP) goals of 65-75 mmHg, with less consideration for systolic and diastolic blood pressure given the potential for decreased pulsatility. Placement of the arterial line should be in the right upper extremity (in patients cannulated peripherally). This allows for sampling of arterial blood from the site which is furthest from the source of oxygenated blood (i.e. the return cannula), ensuring adequate oxygen delivery to the brain. Transport teams must have a monitor capable of transducing an arterial line.
Ensuring adequate oxygenation and ventilation may require following non-conventional indicators including mixed venous saturation (SvO2), partial pressures of oxygen (PaO2) and carbon dioxide (PCO2), and serum lactate levels. If peripheral pulse oximetry and end-tidal CO2 (EtCO2) are inconsistent, the transport team should request an arterial blood gas be sent, or run a blood gas on a portable lab unit, prior to departure.
Pulse Oximetry: often difficult to maintain given critical illness, poor perfusion, and loss of pulsatility. Pulse oximetry should be measured on the right side of patients with VA ECMO.
End-Tidal CO2: Represents perfusion and ventilation. EtCO2 will be lower than normal because of decreased native cardiac function and a large portion of cardiac output bypassing the lungs through the circuit.
SvO2: When drawn from the distal pulmonary artery port of the catheter, as is traditionally done with other patients, is often unreliable. If sampled, SvO2 should be drawn just prior to the membrane oxygenation in the ECMO circuit. SvO2 is also measured and displayed on the circuit. An approximate goal SvO2 is 70%, indicating adequate oxygen delivery and consumption. If values remain below 70% despite medical optimization an increase in the flow should be considered.
Lactate levels correlate with outcomes in patients cannulated for both VV and VA ECMO. (10-12). They can initially be significantly elevated in patients post-arrest or recently cannulated for VA ECMO. However, a persistent lactic acidosis may reflect global hypoperfusion, limb ischemia, hepatic dysfunction or other metabolic derangements. These should be managed medically. A respiratory acidosis can be managed by increasing the sweep on the oxygenator, or, in some situations, making changes to the patient’s ventilator.
ECMO monitor values:
Blood flow through the circuit (lpm)
deltaP: Pre- and post-oxygenator pressure measurements (difference is typically 10 times the flow or less than 50 mmHg); if elevated or an increasing trend, evaluate for a clot in the oxygenator
Pre-pump drainage line pressure: prevents excess suction termed a “suction event” (should be greater than negative 300 mmHg)
Oxygenation can be difficult to monitor in patients on ECMO. If oxygenation cannot be monitored noninvasively, the clinical team should choose other parameters to monitor, such as arterial blood gases. If the patient is found to have poor oxygenation it should be determined if the poor oxygenation is secondary to medical or mechanical shortcomings. Gross determination can be made by evaluating the drainage and return cannulas. If blood in the return cannula is similar in color (dark) to the drainage cannula, a mechanical issue should be suspected as the return cannula should be brighter than the drainage cannula. If a mechanical issue is suspected, check the oxygen source to ensure a sufficient supply as well as the tubing to ensure no accidental discontinuity.
Blood in the return cannula appearing brighter than the drainage cannula indicates proper mechanical oxygenation, suggesting a medical (often termed patient centered) cause of hypoxemia. There are multiple methods to increase oxygen delivery that should be considered (Table 3). Infection is often under appreciated in these patients, especially as their temperature is regulated by the circuit thus preventing fever generation. There should be a high suspicion for sepsis in patients on circuit, with a low threshold for aggressive sepsis treatment.
Differential hypoxia refers to hypoxia of the proximal branches of the ascending aorta, due to poorly oxygenated blood natively ejected from the heart.This can be difficult to detect in transport. Therefore, it is recommended that these patients are transported on 100% FiO2 to maximize the oxygen content in the natively ejected blood from the left ventricle. Other interventions such as recruitment maneuvers, increasing positive end-expiratory pressure, or decreasing inotrope dosage can be used if needed when differential hypoxia is detected.
Hypotension, defined as a MAP < 65 mmHg, is caused by similar causes as in other critically ill patients: volume depletion and vasoplegia. Within the first 24 hours of being placed on circuit, patients can experience a profound inflammatory response, leading to capillary leak and hypotension, requiring fluid resuscitation and vasopressor support. When patients on VA ECMO become volume depleted, chatter or chugging - defined as intermittent suction events of the drainage cannula against the right atrium or inferior vena cava - may occur. Chatter is a result of the normal negative pressure generated by the drainage cannula in the context of hypovolemia, producing periodic collapse of the venous circulation against the cannula. If chatter is observed, the RPMs should be decreased while the patient is volume resuscitated. Prolonged chatter can lead to hemolysis and further complicate the patient’s clinical course if not addressed quickly. In patients with abrupt onset of hypotension, the patient should be evaluated for bleeding or tamponade.
Hypertension, defined as MAP greater than 80 mmHg (critical being greater than 110 mmHg), is often overlooked and can lead to devastating consequences, including intracranial hemorrhage. Analgesia and sedation should be addressed first in patients found to be hypertensive, with adjustment of vasopressor agents to follow. If needed, antihypertensives such as nicardipine and nitroglycerin can be used when all vasopressors have been weaned off.
The introduction of an ECMO circuit can lead to a prothrombotic state causing many undesired complications from thrombus formation. Despite significant advances in technology, such as heparin coated membranes, anticoagulation is still routinely needed in VA ECMO. If possible, baseline labs should be sent prior to initiation of anticoagulation to give providers baseline parameters later in the patient's course (Table 4). In the transport environment, these can be sent at the referring facility prior to transport. Unfractionated heparin is the most widely used anticoagulant in VA ECMO; however, some centers are implementing other anticoagulation strategies such as direct thrombin inhibitors. Prior to transport, most patients will be given a bolus of unfractionated heparin with additional boluses given in transport depending on the patient’s activated clotting time (ACT), with a goal ACT > 200. If the transport team is arriving prior to cannulation, a 50-80 U/kg bolus of unfractionated heparin should be considered prior to cannuation. However, anticoagulation is not benign. Hemorrhage as a result of anticoagulation can occur in any part of the body, with intracranial bleeding representing the most dreaded complication.
Complications
In cardiac ECLS, complications are common, requiring skilled ECMO management by trained providers. In adult cardiac ECLS, the ELSO registry reports a 2.2% incidence of central nervous system (CNS) hemorrhage, a 3.8% incidence of CNS infarction, a 12.3% incidence of renal failure, a 13% incidence of infection, an 18.5% incidence of cannula site hemorrhage, and a 6.6% mechanical oxygenator failure. (4) Transport on ECMO is associated with adverse events, simply by nature of the high acuity, let alone the logistical challenges.(13-16)
Pump Failure, caused by motor failure, power failure, or flow sensor failure, requires immediate action to prevent loss of flow. If the cause of pump failure cannot be addressed immediately a hand pump must be used. The possibility of mechanical failure necessitates the immediate availability of a hand pump at all times. Power failure should not happen immediately, as a backup battery should be activated if the generated power source is disrupted. However, a flow sensor failure will result in immediate disruption of flow and requires replacement of the sensor.
Accidental decannulation is often fatal. Therefore protecting cannulas throughout transport should be of the highest priority. A crew member should be dedicated to the cannulas throughout transfers and the cannulas should be secured at multiple sites, including the insertion site, prior to transfer of the patient.
Decreased pulsatility may be secondary to the initial cardiac failure requiring cannulation or from cardiac stunning from increased afterload (due to retrograde flow of the VA circuit) following cannulation. This often leads to left ventricular overload, LV distension, and a decreased ejection fraction. LV distention can be complicated by left atrial hypertension, pulmonary edema, thrombosis, and decreased cardiac recovery. This chain of events can be mitigated by the placement of a LV vent, including an Impella or intra-aortic balloon pump. (9) If pulsatility is lost in transport the team should ensure adequate MAP, continued ECMO flow, and assess for stable central venous pressure. (17) Inotropes and calcium can be considered for the treatment of loss of pulsatility throughout transport.
Bleeding is difficult to evaluate during transport given the configuration and cover protecting the patient in flight. Bleeding is common in ECLS, occurring in approximately 50% of cases. (18) The cannulation site is the most common source of bleeding. Each cannulation site and other surgical sites should be assessed prior to departure and with any clinical change. Complete blood count and coagulation studies should be checked prior to departure and transfusion should be considered if the patient’s hemoglobin is less than 8 g/dL. Disseminated intravascular coagulation should also be considered in these patients. Active bleeding with platelets less than 100 x 103 should be treated with a transfusion of platelets. Active bleeding with an INR greater than 1.6 should be treated with a transfusion of fresh frozen plasma. For bleeding from the cannula, consider applying a thrombotic surgical dressing to the cannula site. If bleeding persists, pausing anticoagulation can be considered if no clots are discovered after an inspection of the circuit. If massive transfusion is required, calcium replacement is key, as is a balanced resuscitation with a 1:1:1 ratio of packed red blood cells, fresh frozen plasma, and platelets.
Hypothermia is a concern when transporting patients on ECMO as the heating system does not operate on battery power. Continuous temperature monitoring should be used throughout transport, and patients should be warmed externally as indicated to keep central temperatures above 34 degrees Celsius.
Transport Considerations
Transferring patients for evaluation of mechanical support to a ECMO facility is associated with decreased mortality than continued management at a non-ECMO facility (19,20) and can be accomplished safely by highly specialized teams. (2,21-23)
Electricity: A plan to assure adequate access to electricity in transport should be clearly delineated and include both generated electricity, battery powered electricity, and a back-up source if additional energy is needed. The battery-life of the system and other battery dependent devices (Impella, Intra-aortic balloon pump, etc.) should be determined before unplugging from a secure, generated, electrical source. Unplugging the power source is the last step prior to departure.
Depth of the cannula: The depth of the cannula should be determined and marked prior to transferring the patient in order to monitor for migration of the cannula given the devastating consequence of decannulation. The marking of the cannula can be completed with a piece of tape or a marker.
EC-145 Helicopter Configurations
ECMO Rotor Diagram with Impella
ECMO Rotor Diagram without Impella
Recommended ECMO Education resources:
AUTHORED BY ADAM GOTTULA, MD (@laertezz); Christopher Shaw, MD (@cshaw1026); Kari Gorder, MD (@karigorder); Paige Barger, DNP (@CritCATNP); Jacob Miller, ACNP (@JacobMillerACNP); Elizabeth Powell, MD (@SpaceForceMD)
Dr. Gottula is a third-year Emergency Medicine resident at the University of Cincinnati and future Anesthesia Critical Care fellow at the University of Michigan with an interest in critical care and HEMS.
Dr. Shaw is a third-year Emergency Medicine resident at the University of Cincinnati and with an interest in critical care.
Dr. Gorder is a second-year Anesthesia Critical Care fellow at the University of Cincinnati.
Paige Barger is a doctor of nurse practitioner in the Cardiac Intensive Care Unit at the University of Cincinnati.
Jacob Miller is a critical care nurse practitioner at the University of Cincinnati Air Care & Mobile Care.
Dr. Powell is a second-year Anesthesia Critical Care fellow at the University of Cincinnati.
FACULTY EDITORS Gretchen Lemmink, MD; Jennifer Lakeberg, DNP (@JenniferLakebe2)
Dr. Lemmink is an attending in the Cardiac Intensive Care Unit at the University of Cincinnati.
Jennifer Lakeberg is a doctor of nurse practitioner at the University of Cincinnati Air Care & Mobile Care.
References:
Panchal, Ashish R., et al. “2019 American Heart Association Focused Update on Advanced Cardiovascular Life Support: Use of Advanced Airways, Vasopressors, and Extracorporeal Cardiopulmonary Resuscitation During Cardiac Arrest: An Update to the American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.” Circulation, 14 Nov. 2019, www.ahajournals.org/doi/10.1161/CIR.0000000000000732.
“Extracorporeal Life Support Organization (ELSO) Guidelines for Adult Cardiac Failure.” Extracorporeal Life Support Organization (ELSO), 2020, www.elso.org/Portals/0/IGD/Archive/FileManager/e76ef78eabcusersshyerdocumentselsoguidelinesforadultcardiacfailure1.3.pdf
“Extracorporeal Life Support Organization (ELSO) General Guidelines for All ECLS Cases August, 2017.” Extracorporeal Life Support Organization (ELSO), 2020, www.elso.org/Portals/0/ELSO%20Guidelines%20General%20All%20ECLS%20Version%201_4.pdf.
Thiagarajan, R R, et al. “Extracorporeal Life Support Organization Registry International Report 2016.” ASAIO Journal, vol. 63, no. 3, 2017, p. 355., doi:10.1097/mat.0000000000000579.
Blumenstein, Johannes, et al. “Extracorporeal Life Support in Cardiovascular Patients with Observed Refractory in-Hospital Cardiac Arrest Is Associated with Favourable Short and Long-Term Outcomes: A Propensity-Matched Analysis.” European Heart Journal: Acute Cardiovascular Care, vol. 5, no. 7, 2016, pp. 13–22., doi:10.1177/2048872615612454.
Chen, Yih-Sharng, et al. “Cardiopulmonary Resuscitation with Assisted Extracorporeal Life-Support versus Conventional Cardiopulmonary Resuscitation in Adults with in-Hospital Cardiac Arrest: an Observational Study and Propensity Analysis.” The Lancet, vol. 372, no. 9638, 2008, pp. 554–561., doi:10.1016/s0140-6736(08)60958-7.
Cardarelli MG, Young AJ, Griffith B. Use of extracorporeal membrane oxygenation for adults in cardiac arrest (E‑CPR): A meta‑analysis of observational studies. ASAIO J 2009;55:581‑6
Vieira, Jennifer, et al. “Extracorporeal Membrane Oxygenation in Transport Part 1:Extracorporeal Membrane Oxygenation Configurations and Physiology.” Air Medical Journal, vol. 39, no. 1, 2020, pp. 56–63., doi:10.1016/j.amj.2019.09.008
Lazzeri C, Bonizzoli M, Cianchi G, et al. Lactate and echocardiography before venovenous extracorporeal membrane oxygenation support. Heart Lung Circ. 2018;27:99–103.
Bonizzoli M, Lazzeri C, Cianchi G, et al. Serial lactate measurements as a prognostic tool in venovenous extracorporeal membrane oxygenation Support. Ann Thorac Surg. 2017;103:812–818.
Slottosch I, Liakopoulos O, Kuhn E, et al. Lactate and lactate clearance as valuable tool to evaluate ECMO therapy in cardiogenic shock. J Crit Care. 2017;42:35–41.
Mendes PV, de Albuquerque Gallo C, Besen BA, et al. Transportation of patients on extracorporeal membrane oxygenation: a tertiary medical center experience and systematic review of the literature. Ann Intensive Care. 2017;7:14.
Austin DE, Burns B, Lowe D, et al. Retrieval of critically ill adults using extracorporeal membrane oxygenation: the nine-year experience in New South Wales. Anaesth Intensive Care. 2018;46:579–588.
Biscotti M, Agerstrand C, Abrams D, et al. One hundred transports on extracorporeal support to an extracorporeal membrane oxygenation center. Ann Thorac Surg. 2015;100:34–40.
Ranney DN, Bonadonna D, Yerokun BA, et al. Extracorporeal membrane oxygenation and interfacility transfer: a regional referral experience. Ann Thorac Surg. 2017;104:1471–1478.
Mazzeffi M, Greenwood J, Tanaka K, et al. Bleeding, transfusion, and mortality on extracorporeal life support: ECLS Working Group on Thrombosis and Hemostasis. AnnThorac Surg. 2016;101: 682–689.
Vieira, Jennifer, et al. “Extracorporeal Membrane Oxygenation in Transport Part 2: Complications and Troubleshooting.” Air Medical Journal, vol. 39, no. 2, 2020, pp. 124–132., doi:10.1016/j.amj.2019.09.009.
Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet. 2009;374:1351–1363.
Noah MA, Peek GJ, Finney SJ, et al. Referral to an extracorporeal membrane oxygenation center and mortality among patients with severe 2009 influenza A (H1N1). JAMA. 2011;306:1659–1668.
Biscotti M, Agerstrand C, Abrams D, et al. One hundred transports on extracorporeal support to an extracorporeal membrane oxygenation center. Ann Thorac Surg. 2015;100:34–40.
Broman LM, Frenckner B. Transportation of critically ill patients on extracorporeal membrane oxygenation. Front Pediatr. 2016;4:63.
Broman LM, Holzgraefe B, Palmer K, Frenckner B. The Stockholm experience: interhospital transports on extracorporeal membrane oxygenation. Crit Care. 2015;19:278.
