Secondary Logo

Journal Logo

Management of COVID-19 Patients

Extracorporeal Membrane Oxygenation Circuits in Parallel for Refractory Hypoxemia in COVID-19: A Case Series

Patel, Yatrik J.*; Stokes, John W.*; Gannon, Whitney D.; Francois, Sean A.*; Wu, Wei Kelly*; Rice, Todd W.; Bacchetta, Matthew*,‡

Author Information
ASAIO Journal: May 05, 2022 - Volume - Issue - 10.1097/MAT.0000000000001706
doi: 10.1097/MAT.0000000000001706


The use of extracorporeal membrane oxygenation (ECMO) for coronavirus disease 2019 (COVID-19)–related acute respiratory distress syndrome (ARDS) has introduced distinct clinical challenges, and outcomes widely vary.1–3 One such challenge includes persistent hypoxemia despite the use of ECMO support, requiring concomitant use of therapies such as prone positioning, neuromuscular blocking agents, inhaled vasodilators, and ventilator settings that exceed criteria considered lung protective.1,3 Further strategies to temporize refractory hypoxemia for patients receiving ECMO include optimizing cannula positioning to reduce recirculation, reducing metabolic demand using antipyretics or beta-blocking agents, blood transfusion, and placement of an additional drainage cannula.4 However, these interventions may not be safe in all clinical circumstances and have varying effects on hypoxemia. For example, the additional blood flow achieved with a second drainage cannula remains limited by cannula size and the maximum rated flow of an oxygenator. To address severe, refractory hypoxemia despite maximum ECMO support and other adjunct measures, we elected to introduce a second ECMO circuit in parallel to substantially increase the blood flow rate in select patients with COVID-19–related ARDS. Additionally, we describe a novel technique for dual circuit support using a single return cannula, obviating the need for two return cannulas in two upper body central veins.5 This is the largest case series of patients receiving ECMO circuits in parallel and the first to report this approach in COVID-19.

Case 1

A 50 year old male with a body mass index (BMI) of 39.5 kg/m2 (136 kg) and no known medical history was admitted to the hospital for hypoxia secondary to COVID-19. He was endotracheally intubated 2 days after admission for respiratory failure and required high-pressure mechanical ventilation, deep sedation, continuous neuromuscular blockade, and inhaled epoprostenol (Flolan, GlaxoSmithKline, Research Triangle Park, NC). He remained hypoxemic despite these strategies and was cannulated for venovenous (VV)-ECMO 3 days after intubation (Table 1). Ultrasound guidance was used to place a 25Fr drainage cannula in the right common femoral vein and a 20Fr return cannula in the right internal jugular (IJ) vein. An ECMO blood flow rate of approximately 5.8 L/min was achieved. The patient had an initial improvement in oxygenation yet continued to require 80–100% fraction of inspired oxygen (FiO2) on the ventilator and continuous neuromuscular blockade. On ECMO day 5, a second 21Fr drainage cannula was placed in the left common femoral vein and a blood flow rate up to 6.6 L/min was achieved. Despite maximum ECMO support and high-pressure ventilation, arterial partial pressure of oxygen (PaO2) fell to 49 mm Hg. Recirculation was ruled out. Hypotension precluded the use of beta-blockers and diuresis. Adequate oxygenator function was confirmed. Given the presence of sepsis, we suspected the patient’s cardiac output (CO) was substantially higher than the flow we could achieve using a single ECMO circuit. Therefore, we added a second ECMO circuit in parallel to achieve a blood flow rate to better match the patient’s metabolic demand. A 22Fr return cannula was placed in the left IJ vein and the previously placed 21Fr left femoral vein cannula was used as the drainage source for the second circuit (Figure 1). Subsequently, a blood flow rate up to 8 L/min was achieved and allowed reduction of ventilator pressures, FiO2, sedation, and discontinuation of neuromuscular blockade (Table 2). Blood flow and sweep gas flow was partitioned equally between the two circuits. As the lung injury, pulmonary compliance and gas exchange improved, we were able to wean ECMO support on both circuits. During this period, lung-protective mechanical ventilation was utilized and FiO2 was weaned to <60% before weaning ECMO support. Extracorporeal membrane oxygenation flow and sweep were first weaned on both circuits to maintain appropriate oxygen saturation and normal pH. Once the patient was consistently supported with a total ECMO flow of <5 L/min, the left IJ return cannula was removed and the left femoral drainage cannula served as an additional drainage cannula for the single circuit. The patient remained on parallel circuits for 9 days. The patient was decannulated from ECMO after 25 days. Mechanical ventilation was ultimately weaned and the patient was discharged from the hospital on 2 L of supplemental oxygen after 65 days.

Table 1. - Baseline Patient Characteristics
Case Patient 1 Patient 2 Patient 3 Patient 4 Patient 5
Age (years) 50 47 37 47 46
Sex Male Male Male Male Male
Body mass index (kg/m2) 39.5 32.6 38 36.9 39.7
Weight (kg) 136 99.7 127 133.8 132.9
Body surface area (m2) 2.56 2.15 2.46 2.59 2.51
Days from COVID-19 diagnosis to intubation 12 12 17 18 17
Days from intubation to ECMO initiation 3 2 1 3 1
Mechanical ventilation settings before ECMO initiation
 Mode Pressure control Volume control Volume control Pressure control Pressure control
 FiO2 (%) 100% 100% 100% 100% 100%
 Driving pressure (cmH2O) 14 22 20
 Tidal volume (ml) 350 450 400 430 350
 Plateau pressure (cmH2O) 32 34
 Respiratory rate (breaths per minute) 30 32 24 34 35
 Positive end-expiratory pressure, (cmH2O) 14 15 15 15 16
Arterial blood gas before ECMO
 pH 7.32 7.24 7.33 7.36 7.36
 PaCO2 (mmHg) 68 75 64 61 65
 PaO2 (mmHg) 62 63 61 63 53
 SOFA score before ECMO initiation 5 4 4 4 4
COVID-19, coronavirus disease 2019; ECMO, extracorporeal membrane oxygenation; FiO2, fraction of inspired oxygen; PaCO2, partial pressure of carbon dioxide; PaO2, partial pressure of oxygen; SOFA, sequential organ failure assessment.

Table 2. - ECMO Characteristics
Case Patient 1 Patient 2 Patient 3 Patient 4 Patient 5
Initial configuration (drainage-reinfusion) RFV-RIJ RFV-RIJ RFV-RIJ RFV-RIJ RFV-RIJ
Initial cannula sizes (Fr) 25–20 25–22 27–22 27–25 27–25
ECMO settings 24 hours after cannulation
 Blood flow rate (LPM) 5.3 4.2 4.5 4.4 5.4
 FdO2 (%) 100 100 100 100 100
 Sweep gas flow (LPM) 2 3 2 5 4
ECMO settings before second circuit
 Blood flow rate (LPM) 6.6 6 5.9 5.5 5
 FdO2 (%) 100 100 100 100 100
 Sweep gas flow (LPM) 8 7 9 7 5
 ECMO days before second circuit support 11 23 10 4 8
Mechanical ventilation settings before  second circuit
Mode Volume Controlled Pressure Controlled Pressure Controlled Pressure Controlled Pressure Controlled
 FiO2 (%) 100 100 100 100 100
 Driving pressure (cmH2O) 18 16 16 14
 Tidal volume (ml) 140 120 80 250 240
 Respiratory rate (breaths per minute) 20 18 15 14 22
 Positive end-expiratory pressure (cmH2O) 18 14 14 14 12
Arterial blood gas before second circuit
 pH 7.32 7.45 7.43 7.39 7.32
 PaCO2 (mmHg) 55 59 62 66 75
 PaO2 (mmHg) 58 67 65 65 65
 Neuromuscular blockade before second circuit  support Yes Yes Yes No No
 Inhaled pulmonary vasodilators before  second circuit support Yes Yes Yes Yes No
ECMO settings 24 hours after second circuit
 Flow 8 7.1 7.4 7.8 6.8
 FdO2 (%) 100 100 100 100 100
 Sweep gas flow (LPM) 7 4 6 6 8
Mechanical ventilation settings 24 hours  after second circuit
 Mode Pressure control Pressure control Pressure control Pressure control Pressure control
 FiO2 (%) 50 50 50 50 50
 Driving pressure (cmH2O) 12 16 16 16 14
 Tidal volume (ml) 160 140 150 150 270
 Respiratory rate (breaths per minute) 12 18 15 14 22
 Positive end-expiratory pressure (cmH2O) 14 12 14 14 14
Arterial blood gas 24 hours after second circuit
 pH 7.35 7.46 7.41 7.45 7.37
 PaCO2 (mmHg) 48 51 49 51 43
 PaO2 (mmHg) 71 117 104 116 122
ECMO, extracorporeal membrane oxygenation; FdO2, fraction of delivered oxygen via ECMO; FiO2, fraction of inspired oxygen; LPM, liters per minute; RFV, right femoral vein; RIJ, right internal jugular; PaCO2, partial pressure of arterial carbon dioxide; PaO2, partial pressure of arterial oxygen.

Figure 1.:
Venovenous-ECMO with parallel circuits with cannulation of bilateral femoral veins and bilateral internal jugular veins. ECMO, extracorporeal membrane oxygenation.

Case 2

A 47 year old male (BMI 32.5 kg/m2, 99.7 kg) with no known medical history was admitted with respiratory failure from COVID-19 requiring endotracheal intubation. His gas exchange remained poor despite maximized ventilator settings, deep sedation, neuromuscular blockade, and prone positioning. He was cannulated for VV-ECMO 2 days after intubation using a 25Fr right femoral vein drainage cannula and a 22Fr right IJ vein return cannula. His oxygenation improved with approximately 5 L/min of ECMO blood flow. Over the next several days, the patient’s pulmonary compliance worsened. Despite 5.8 L/min of ECMO flow, the patient continued to require 70–100% FiO2, deep sedation, and neuromuscular blockade. He had refractory hypoxemia with PaO2 decreasing to 52 mmHg despite maximum ECMO support and ventilator pressures exceeding those considered lung protective.6 There was no evidence of meaningful recirculation and beta blockade was deferred due to hypotension and septic shock. To better match the patient’s elevated metabolic demand, we inserted a second ECMO circuit in parallel using a 23Fr left femoral vein drainage cannula and a 20Fr left IJ vein return cannula. With two circuits, we were able to achieve up to 10 L/min of total ECMO blood flow. His arterial PaO2 improved to 124 mmHg, which allowed the ventilator FiO2 to be weaned to 40% and driving pressure to be reduced. Neuromuscular blockade was stopped. His lung compliance continued to slowly improve. The patient underwent tracheostomy placement on ECMO day 37 in an effort to reduce sedation requirements and improve pulmonary toilet. His hospital course was complicated by a right pneumothorax requiring chest tube placement and he died from uncontrolled bleeding after iatrogenic hepatic injury on day 38 of ECMO support despite improving lung compliance and function.

Case 3

A 37 year old male (BMI 38 kg/m2, 127 kg) with no significant medical history was cannulated for VV-ECMO the day after intubation for refractory hypoxemia secondary to severe COVID-19 ARDS. He was cannulated with a 27Fr right femoral vein drainage cannula and a 22Fr right IJ vein return cannula. His lung compliance continued to worsen; tidal volumes were approximately 120 ml on pressure-control ventilation with a driving pressure of 18 cmH2O. Despite 5.5 L/min of ECMO blood flow, he continued to require 100% FiO2, deep sedation, and continuous neuromuscular blockade. Septic shock precluded the use of beta blockade to attenuate his tachycardia and elevated CO. His arterial PaO2 was 56 mmHg despite maximum ECMO support. We elected to add a second circuit in parallel to achieve increased ECMO flows to match the patient’s elevated CO and to reduce excessive ventilator support and avoid long-term neuromuscular blockade. A 23Fr left femoral vein drainage cannula and a 20Fr left IJ vein return cannula were placed. With two circuits, we were able to achieve flows up to 10 L/min. After improved oxygenation, ventilator settings and neuromuscular blockade were weaned. Over time, sedation was weaned. While receiving dual circuit support, he was able to participate in physical therapy on a stationary bike while in bed. After 62 days on ECMO, his lung compliance remained poor with little evidence of improvement. He was evaluated and listed for lung transplant. He continued to participate in physical therapy while on two parallel circuits. After 72 days on parallel circuits and a total of 96 days on ECMO, he underwent double lung transplantation. His postoperative course was unremarkable, and he was discharged on room air oxygen after a 125 day hospitalization and returned home.

Case 4

A 47 year old male (BMI 36.8 kg/m2, 133.8 kg) presented with severe ARDS and hypoxia despite optimized mechanical ventilation, deep sedation, inhaled vasodilators, and neuromuscular blockade. He was cannulated for VV-ECMO 3 days after intubation using a 27Fr right femoral vein drainage cannula and a 25Fr right IJ vein return cannula. He was initially supported with up to 6 L/min of blood flow allowing for weaning of neuromuscular blockade and inhaled vasodilators. Six days into his ECMO course, his pulmonary compliance and oxygenation worsened requiring 100% FiO2 and deep sedation. The patient required more ECMO blood flow than what could be achieved with the addition of a second drainage cannula. Therefore, a circuit in parallel was added by inserting a 25Fr drainage cannula into the left femoral vein and the return line of the second circuit was joined with a Y connector to the existing 25Fr right IJ vein return line of the first circuit (Figure 2). The 25Fr return cannula was large enough to accommodate 10 L/min of total flow with outflow pressures consistently less than 200 mmHg. He required dual circuit support for 47 days and was weaned from total ECMO support after 98 days. After 154 hospital days, the patient was discharged to a long-term acute care facility where he tolerated slow weaning of mechanical ventilation.

Figure 2.:
Venovenous-ECMO with parallel circuits with cannulation of bilateral femoral veins and reinfusion via a 25Fr right internal jugular vein return cannula. ECMO, extracorporeal membrane oxygenation.

Case 5

A 46 year old male (BMI 40.9 kg/m2, 132.9 kg) was cannulated for VV-ECMO 1 day after intubation for refractory hypoxemia due to COVID-19 despite maximal medical management. He was cannulated with a 27Fr right femoral vein drainage cannula and a 25Fr right IJ vein return cannula. He continued to require 80–100% FiO2 and deep sedation despite maximum ECMO support with up to 6 L/min of flow. In an effort to achieve lung-protective ventilation6 and reduce sedation, a second circuit was added for increased support. A 25Fr left femoral vein drainage cannula served as the inflow of the second circuit and the outflow was connected to the existing 25Fr right IJ cannula with a Y connection. With a second circuit, we were able to achieve up to 10 L/min of ECMO flow allowing us to maintain lung-protective ventilation6 and dramatically reduce the patient’s sedation requirements. He required dual circuit support for 13 days. His hospital course was complicated by renal failure requiring renal replacement therapy and numerous infections, including Candida albicans fungemia. He was decannulated after 53 days on ECMO in an attempt to clear the fungemia. He required recannulation after 48 hours for hypercapnic respiratory failure; a 28Fr single-site dual-lumen cannula was placed. The patient was decannulated after 71 total days on ECMO. After a prolonged hospitalization, the patient was discharged from the hospital to a rehabilitation facility with evidence of both renal and pulmonary recovery. Mechanical ventilation was weaned and he returned home on 2 L of supplemental oxygen.


Extracorporeal membrane oxygenation is a life-saving therapy for select patients with COVID-19 ARDS. Refractory hypoxemia despite ECMO support can occur requiring the use of injurious ventilator settings and concomitant rescue therapies. We demonstrate that refractory hypoxemia can be ameliorated with a second, parallel ECMO circuit allowing for lung protective ventilation, awakening and physical rehabilitation. Outcomes were excellent; four of five (80%) of patients survived to hospital discharge (Table 3). This is the first case series reporting the use of two circuits for VV-ECMO to support a single patient.

Table 3. - Outcomes
Case Patient 1 Patient 2 Patient 3 Patient 4 Patient 5
Hospital length of stay (days) 64 37 124 154 95
ICU length of stay (days) 56 37 115 154 95
Total ECMO duration (days) 25 37 95 98 69
Duration of two circuit support (days) 9 14 72 47 13
ECMO-associated bleeding complications None None Yes, LIJ cannula dislodgement requiring replacement None None
Thromboembolic complications Cannula-associated DVT None None None None
Requirement of renal replacement therapy during ECMO Yes No No No Yes
Evidence of renal recovery at time of discharge Yes N/A N/A Yes Yes
Survival to decannulation Yes No Yes* Yes Yes
Survival to discharge Yes No Yes Yes Yes
Discharge disposition Skilled nursing facility N/A Rehabilitation hospital Long-term acute care hospital Long-term acute care hospital
*Patient required double lung transplantation.
ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; LIJ, left internal jugular.

COVID-19–induced inflammatory syndrome driven by hyperactivation of the immune system results in severe lung injury7 leading to marked impairment of gas exchange by the native lungs. On presentation, patients can have atypical hemodynamic profiles and an inflammation-driven hyperdynamic circulation contributing to lung stiffness creating a vicious cycle between the heart and lungs.8 We did not obtain CO data using a Swan-Ganz catheter as routine use of these catheters is not part of our institution’s standard of care for ARDS patients. However, to obtain an estimation of the CO in these patients, we used the LiDCOrapid monitor (LiDCO Ltd, Cambridge, United Kingdom). The LiDCOrapid uses the PulseCO (LiDCO Ltd) algorithm to calculate continuous beat-to-beat CO by analyzing the arterial blood pressure tracing from an existing arterial line.9Figure 3A shows the CO data over a 24 hour period immediately before reconfiguration to parallel circuits for patients 2–5; Figure 3B shows the calculated CO of the patients based on body surface area, ECMO flow with one circuit, and estimated CO based on the LiDCOrapid monitor. The estimated CO ranged from 9.8 to 19.2 L/min. Anemia, hypoxia, and sepsis further contribute to this hyperdynamic high CO state. With an additional circuit in parallel, we were able to significantly increase the total ECMO blood flow, thereby decreasing the total undrained systemic venous return and increasing the ECMO flow:patient CO ratio.

Figure 3.:
Cardiac output data obtained by LiDCO. A: Estimated cardiac output over a 24-hour period for patients 2–5 immediately prior to addition of a second parallel ECMO circuit. B: Comparison of each patient’s calculated cardiac output with the estimated mean cardiac output over a 24-hour period using the LiDCO device. ECMO, extracorporeal membrane oxygenation.

Mechanisms of refractory hypoxemia during VV-ECMO include a 1) high recirculation fraction, 2) high pulmonary shunt fraction, 3) low ECMO blood flow to patient CO ratio, and 4) oxygenator dysfunction.10 Chest radiography confirmed proper positioning of drainage and reinfusion cannulas in all patients, and there was no evidence of meaningful recirculation. The drainage cannula was inserted deep enough to lie in the intrahepatic portion of the inferior vena cava (IVC) close to the right atrial-IVC junction. The intrahepatic portion of the IVC is least collapsible because of the surrounding hepatic parenchyma and less likely to collapse with negative pressure from the drainage cannula. The reinfusion cannula was inserted with the tip in the superior vena cava (SVC) close to the right atrial-SVC junction. Oxygenator dysfunction was ruled out in all cases. In our cohort of patients, efforts to reduce the intrapulmonary shunt included increasing the ventilator FiO2, peak end expiratory pressure, and initiating inhaled epoprostenol. These interventions were marginally and transiently effective and compromised lung-protective ventilation. Methods of reducing the patient’s CO such as beta-blockers11 and hypothermia were used to decrease native CO but improvement in oxygenation was temporary, and use was limited due to clinical circumstances. Extracorporeal membrane oxygenation circuit temperature was maintained between 36°C and 37°C in all patients. To optimize venous drainage, we used the largest available femoral drainage cannula (27Fr) we have available at our institution. Despite using large drainage cannulas and ensuring adequate patient preload, we found it difficult to achieve ECMO flows greater than 6 L/min with a single circuit due to excessive negative drainage pressures. With the addition of a second drainage cannula to a single circuit, the maximum flow we were able to achieve was 6.5 L/min. Only by increasing ECMO blood flow above the rated limits of available oxygenators with parallel circuits were we able to better match the patients’ high CO state and consistently wean ventilator settings and discontinue rescue therapies.

The Extracorporeal Life Support Organization (ELSO) guidelines advocate for transfusion to achieve a hematocrit level greater than 40% to improve oxygen delivery.12 Targeting higher hematocrit thresholds may have improved oxygen delivery in our patients and could have been considered before providing dual circuit support although there are no data that provide evidence for maintaining higher hematocrit targets in adult patients. Furthermore, blood conservation protocols may reduce blood transfusions without increasing end-organ dysfunction and mortality.13 Prone positioning during VV-ECMO could also have been considered and may have led to improved oxygenation14 but we do not routinely place patients on ECMO in the prone position at our institution. Furthermore, benefits of prone positioning remain uncertain in patients with the most severe form of COVID ARDS requiring VV-ECMO15 and prone positioning of patients on ECMO complicates nursing care and increases risk for complications.

There were no complications associated with cannulation for dual circuit support. There were no thrombotic complications resulting from placing two large-bore drainage cannulas in the IVC. All patients were anticoagulated for a target activated partial thromboplastin time (aPTT) of 40–60 seconds unless indications for therapeutic anticoagulation were present. There was one major ECMO-associated bleeding complication in patient 3 secondary to a left IJ return cannula dislodgement attributed to skin breakdown surrounding the cannula entry site and compromised suture integrity due to long-term ECMO support. We modified our cannulation strategy after the first three patients to avoid cannulation of both IJ veins. For addition of a parallel circuit, we placed a drainage cannula in the contralateral femoral vein and the reinfusion line of the circuit no. 2 was connected to the reinfusion line of circuit no. 1 using a Y connector. Reinfusion from both circuits occurred through the single large cannula placed in the right IJ vein (Figure 2). With the addition of a second circuit in parallel, circuit pressures typically decreased because we were able to achieve approximately 3.5–4 L/min of flow through each circuit rather than 5–6 L/min through a single circuit. There was no increase in postartificial lung membrane pressures even when using a single reinfusion cannula for both circuits. We saw no evidence of increased hemolysis from this modified cannulation strategy. We began placing larger 23–25Fr outflow cannulas at the time of initial cannulation for all COVID-19 patients to obviate the requirement for a second reinfusion cannula should the patient need parallel ECMO support later in their course. We found no differences in the thrombotic risk between the two configurations in our patients, but the risk of thrombus formation with larger cannulas should be recognized. Details regarding the types of ECMO devices and cannulas used are provided in Table 1, Supplemental Digital Content,

There are important factors to consider when selecting patients for a second circuit in parallel. Extracorporeal membrane oxygenation is a limited and resource-consuming therapy. Triaging scarce resources during a pandemic has practical and ethical implications. Choosing to offer dual circuit support to a patient could potentially deprive another patient of a life-saving therapy in the setting of resource limitations. Further, while outcomes were promising, no conclusions can be made about the safety or efficacy of this strategy. Other considerations include the increased burden of care on bedside staff and cost of disposables. We initiate ECMO for all patients with the intent of managing the patient on a single circuit and employ a second circuit based on physiologic need. Our ECMO and critical care teams maintained daily communication with the patients’ family regarding the plan of care, and risks and benefits of each therapy were extensively discussed before any intervention.

Venovenous-ECMO management in patients with COVID-19 ARDS is challenging given the severity of lung injury and inflammatory and metabolic derangements. A single ECMO circuit may not provide adequate support in the setting of an elevated CO and severely impaired gas exchange. In select patients, addition of a second ECMO circuit in parallel can result in increased ECMO blood flow and improved oxygenation in the setting of refractory hypoxemia.


1. Lebreton G, Schmidt M, Ponnaiah M, et al.: Extracorporeal membrane oxygenation network organisation and clinical outcomes during the COVID-19 pandemic in Greater Paris, France: A multicentre cohort study. Lancet Respir Med 9: P851–862, 2021.
2. Badulak J, Antonini MV, Stead CM, et al.; ELSO COVID-19 Working Group Members: Extracorporeal Membrane Oxygenation for COVID-19: Updated 2021 Guidelines from the Extracorporeal Life Support Organization. ASAIO J 67: 485–495, 2021.
3. Akoumianaki E, Jonkman A, Sklar MC, Georgopoulos D, Brochard L: A rational approach on the use of extracorporeal membrane oxygenation in severe hypoxemia: Advanced technology is not a panacea. Ann Intensive Care 11: 107, 2021.
4. Montisci A, Maj G, Zangrillo A, Winterton D, Pappalardo F: Management of refractory hypoxemia during venovenous extracorporeal membrane oxygenation for ARDS. ASAIO J 61: 227–236, 2015.
5. Malik A, Shears LL, Zubkus D, Kaczorowski DJ: Parallel circuits for refractory hypoxemia on venovenous extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 153: e49–e51, 2017.
6. Abrams D, Schmidt M, Pham T, et al.: Mechanical ventilation for acute respiratory distress syndrome during extracorporeal life support. Research and practice. Am J Respir Crit Care Med 201: 514–525, 2020.
7. Rendeiro AF, Ravichandran H, Bram Y, et al.: The spatial landscape of lung pathology during COVID-19 progression. Nature 593: 564–569, 2021.
8. Caravita S, Baratto C, Di Marco F, et al.: Haemodynamic characteristics of COVID-19 patients with acute respiratory distress syndrome requiring mechanical ventilation. An invasive assessment using right heart catheterization. Eur J Heart Fail 22: 2228–2237, 2020.
9. Pearse RM, Ikram K, Barry J: Equipment review: An appraisal of the LiDCO plus method of measuring cardiac output. Crit Care 8: 190–195, 2004.
10. Nunes LB, Mendes PV, Hirota AS, et al.; ECMO Group: Severe hypoxemia during veno-venous extracorporeal membrane oxygenation: Exploring the limits of extracorporeal respiratory support. Clinics (Sao Paulo) 69: 173–178, 2014.
11. Guarracino F, Zangrillo A, Ruggeri L, et al.: β-Blockers to optimize peripheral oxygenation during extracorporeal membrane oxygenation: A case series. J Cardiothorac Vasc Anesth 26: 58–63, 2012.
12. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support Extracorporeal Life Support Organization. Version 1.4. 2017. Ann Arbor, MI, USA. Available at: Accessed August 5, 2021.
13. Agerstrand CL, Burkart KM, Abrams DC, Bacchetta MD, Brodie D: Blood conservation in extracorporeal membrane oxygenation for acute respiratory distress syndrome. Ann Thorac Surg 99: 590–595, 2015.
14. Giani M, Martucci G, Madotto F, et al.: Prone positioning during venovenous extracorporeal membrane oxygenation in acute respiratory distress syndrome. A multicenter cohort study and propensity-matched analysis. Ann Am Thorac Soc 18: 495–501, 2021.
15. Garcia B, Cousin N, Bourel C, Jourdain M, Poissy J, Duburcq T; Lille Intensive Care COVID-19 group: Prone positioning under VV-ECMO in SARS-CoV-2-induced acute respiratory distress syndrome. Crit Care 24: 428, 2020.

extracorporeal membrane oxygenation; acute respiratory distress syndrome; ARDS; COVID-19; refractory hypoxemia; parallel circuits

Copyright © ASAIO 2022