Secondary Logo

Journal Logo

Management of COVID-19 Patients

Lean Ad hoc Extracorporeal Membrane Oxygenation Systems for COVID-19

Hunziker, Patrick*; Zenklusen, Urs

Author Information
doi: 10.1097/MAT.0000000000001278
  • Free


In coronavirus disease (COVID-19),1 bilateral viral pneumonia2 is a leading cause of death.3 In H1N1 influenza viral pneumonia patients with intractable hypoxemia, Extracorporeal Membrane Oxygenation (ECMO) therapy for specific groups with severe disease was beneficial,4–7 leading to a recommendation to use ECMO in select patients with severe COVID-19 by an international expert group while at the same time recommending rationing its use.8

Global healthcare systems have not been well prepared for a pandemic with such a rapid spread of a highly infectious virus. The large numbers of infected individuals and patients requiring hospitalization, and in particular, intensive care for respiratory support is overwhelming healthcare systems. Key consumables like face masks, key infrastructures like intensive care beds, and key systems for respiratory assist like ventilators are not everywhere available in sufficient numbers. An insufficient number of resources may lead,9 and already has led,10 to rationing of potentially life-saving activities like intensive care unit admission and ventilation, in some patients with high predicted mortality and reduced life expectation. This calls for rapid efforts for improving the availability of vital support equipment.11 Historically, rapid technological innovation is driven by urgent clinical demand during epidemics, for example, in the poliomyelitis epidemic, has led to long-lasting benefits for healthcare.12


The feasibility of ad hoc integrating Impella blood pumps that are already on-site in many hospitals with standard ECMO gas exchange cartridges, tubes, and connectors was tested using different connection techniques. Figure 1 shows the concept, the used parts, assembly process, and the results of such lean ad hoc ECMO systems. Prototype systems were built as a first simplified closed loop with an embedded Impella CP pump only, as a second closed-loop setup incorporating a venous suction cannula, a venous return cannula, an oxygenator and in-line Impella pump driven by a standard Impella AIC console (Abiomed, Danvers, MA), and a third system additionally containing a second gas exchange module flushed with nitrogen to simulate oxygen extraction in patients. Abiomed Impella CP and Impella 5.0 pumps, and a standard ECMO oxygenator (Hilite 7000, Stolberg, Germany) were used. Key steps are the coaxial introduction of the Impella pump into the tube through a Y tube connector, prevention of recirculation of the pumped blood around the Impella pump head by suited obstruction/constriction of the outer tube, suited tube diameter stepups/stepdowns to minimize flow resistance at pump inlet/outlet and the use of large enough venous drainage cannulas (17–29 F; 24–31 F dual-lumen cannulas) and sufficiently large (17–22 F) jugular venous return cannulas. Supplemental Digital Content 1 gives visual clues and instructions for easy assembly and Supplemental Digital Content 2 is a list of components used. Multiday tests were performed as described in Supplemental Digital Content 3

Figure 1.
Figure 1.:
Ad hoc ECMO. (A) Basic connectivity of ad hoc ECMO composed from Impella pumps and standard oxygenators. (B, C) Pieces and assembly of Impella CP-based ECMO system. (D, E) Pieces and assembly of Impella 5.0-based ad hoc ECMO system. (F) Closed-loop setup for pump testing was run for 48 hours and showed stable performance. (G) Venous access cannulas; the 25F and 22F cannula were used in F. (F) Closed-loop system including Impella pump, venous drainage and return cannulas, oxygenator module; to simulate oxygen extraction by a patient, an additional gas exchange module (flushed with N2) is added. 48 hours –testing of this setup with reconstituted blood showed stable flows and stable gas exchange. ECMO, extracorporeal membrane oxygenation.

Figure 2A shows predicted, pressure gradient-dependent pump performance of the Impella pumps used, based on available Impella performance charts and shows observed system flow performance in our CP-based systems. Figure 2B shows expected pressure drops across typical cannulas according to publicly available data13–16 and gives the measured pressure drops across the used oxygenator type when driven at various flows using standard ECMO consoles in patients with a hematocrit of approximately 30%, based on own historical observations.

Figure 2.
Figure 2.:
Flow-pressure relation of individual components and ad hoc ECMO. (A) Flow-pressure relation prediction of used Impella CP and 5.0 pumps (red, green lines) based on performance chart. Observed flow range in assembled ad hoc ECMO systems driven by Impella CP (blue). Observed flows were in the high range of the prediction, presumably benefiting from the improved flow characteristics of an axial pump positioned in a coaxial tube. (B) Flow-pressure relation estimates of typical cannulas and oxygenator. Diverse data sources, diverse measurement conditions. Dotted black line/Oxi: observed pressure drops across gas exchange module Hilite 7000LT in patients with ECMO. Dashed lines: estimated pressure drops in venous drainage cannulas (55–65 cm length). Red: Maquet 29 F; violet: Edwards QuickDraw 25 F, Medtronic Biomedicus 25 F; Green: Maquet Avalon 31 F bicaval (drainage lumen); orange: Medos Novaport dual lumen Twin24 F (drainage lumen). Solid lines, underlined labels: estimated pressure drops in venous return cannulas (15–23 cm length). Blue: Maquet 19 F. green: Edwards 22 F Red: Maquet/Getinge 23 F cannula, yellow: 17 F cannula. ECMO, extracorporeal membrane oxygenation.

Benefit/risk analysis was performed. This included patient risk and selection, probability of survival with/without added treatment, institutional setup, and known risks of blood pumps including device thrombosis, embolism, infection, bleeding, hemolysis, and air embolism.


Systems were successfully built using standard equipment with both, the Impella CP and the Impella 5.0 pump, although the different pump sizes require differently sized tubes and Y pieces. The Impella CP fits nicely through a standard 3/8" Y piece while the 5.0 Impella pump needs a larger Y piece and some special care to fit through the Y piece as described in more detail in the Supplemental Digital Content Once the procedure had been established, performing it in a sterile manner posed no obstacle, as compared to Impella and ECMO preparation, only the introduction of the pump head into the Y piece using a guidewire was an added step requiring sterile handling.

Using the Impella CP, blood flows of up to 3.6 L/min of saline (minimal circuit) and 3.5 L/min of blood (complete system) could be achieved at pump performance level 9: pressure gradients and flows across the system remained stable over 48 hours. Flows were at least as high as predicted by combining pump performance and component pressure drop charts. The combination of pump performance chart, cannula pressure drop chart and experimental observations indicates that when using large enough cannulas, the “sweet spot” for Impella CP-based systems is between 2.8 and 3.5 L/min, while for Impella 5.0, it is between 3.5 and 4.5 L/min.

Given an oxygen-binding capacity of 1.34 mL/g hemoglobin, this translates to an oxygen transport capacity at a normal hemoglobin content of ECMO blood of up to 200 mL/L × 3.5 L/min, that is, 700 mL oxygen/min. At a hypothetical venous oxygen saturation of 60% with a near-normal body oxygen consumption of approximately 125 mL/min/m2, such an Impella CP-based ad hoc ECMO will still result in a significant contribution to systemic oxygen delivery, even taking some venous/ECMO recirculation and some anemia into account.

The gas exchange module showed no visual evidence of clots and the pressure gradient across the module remained unchanged over the observation time. Hemolysis was not quantified because Impella pumps are known to exhibit very little hemolysis in mock loops when properly positioned (personal communication, C.Nix), and our use of expired packed red blood cells was not considered representative for predicting clinical hemolysis.

In the dual gas exchanger setup that uses a first gas exchange module in-line to partially deoxygenate the blood and then uses the second gas exchanger to re-oxygenate the blood, we observed a stable, near-maximal oxygen content after the second oxygenator over the observation time of 2 days, although the experiments performed at room temperatures and with N2 for O2 washout may not fully reflect physiologic conditions.

Results of the benefit/risk analysis are shown in Table 1. We determined that the benefit/risk relation for using such an off-label system is most acceptable in a scenario with a patient who cannot be ventilated conventionally because a ventilator is not available or who requires an ECMO because of disease severity but standard ECMO systems are no more available.

Table 1. - Benefit/risk analysis and minimization of risks
Potential patient benefit of “ad hoc ECMO” Benefit improbable Benefit possible
Good prognosis with available ventilatory support
Very high predicted mortality despite ECMO
Conventional ECMO devices available
Fallback option in ventilation failure: Need for ECMO with realistic survival probability, but ECMO unavailable
Primary therapy for lung failure: Need for ventilatory assist for survival, but ventilators unavailable
Potential Hazard Probability Risk Risk minimization
Thrombosis, embolism Low to moderate Low, preventable Maintain therapeutic anticoagulation. Visual control of tubes and gas exchange module. Change when necessary.
Air embolism Low to moderate Low, preventable Avoid or securely close access port on low-pressure side of pump. Install pump on low-pressure side of gas exchanger.
Hemolysis Moderate Low Drive pump at level necessary for reasonable oxygenation, not higher
Arterial complications Low Low, preventable Vascular access by experienced physician, ultrasound-guided vascular puncture.
Infection Low–moderate Low–moderate Monitor inflammation markers, repeated blood cultures. Replace system/antibiotics.
ECMO, extracorporeal membrane oxygenation.


Rapid assembly of an ad hoc ECMO system is feasible using an Impella blood pump, for example, the Impella CP or 5.0, a standard Abiomed Impella console, a standard ECMO gas exchange module and standard tubing/cannulas that are already widely distributed in hospitals, following the constructive approach and material delineated in this contribution. In an established ECMO center, assembling such ad hoc ECMO systems should pose no major problem for a trained perfusionist or a physician experienced in installing and using ECMO and Impella systems.

Such ad hoc ECMO systems might prove life-saving if used as a last-resort option in COVID-19 patients with bilateral pneumonia who urgently need gas exchange support when conventional ventilation fails but a conventional ECMO device is not rapidly available. Also, they offer hope as primary therapy in life-threatening lung failure where ventilators are no more available. A patient benefit is improbable if the predicted outcome is anyway very good or very bad. However, if standard ventilators and conventional ECMO devices are available, those established, proven and certified devices should be used preferentially. Experience in large-bore venous access and in handling blood pumps including ECMO and Impella is probably an important success factor for the ad hoc ECMO approach.

An advantage is the use of proven standard equipment like Impella pumps and consoles, of which thousands are already distributed in hospitals in the USA and Europe. In our institution, such Impella-based ad hoc ECMO systems almost double the number of ECMO treatments that could be delivered to patients.

In a pandemic, specialized personnel may become a relevant bottleneck. During the COVID-19 peak in Switzerland, we found that emergency, ICU, and infectious diseases physicians and nurses reached their limit, but elective interventional cardiology and cardiac surgery were largely shut down, thereby freeing cardiologists, surgeons, and perfusionists for COVID-specific tasks. Reorganization of the perfusionist work schedule and training on ad hoc ECMO in a core team was done and are recommended for sites where a COVID-19 peak is impending.

Beyond the COVID-19 pandemic, the very compact resulting system may also give design inspirations for future ECMO system developments.


Impella consoles and Impella pumps are certified for conventional hemodynamic support but not for such nonstandard use. While we saw that a trained perfusionist will easily master the technicalities of system composition, there will still be a learning curve in composition and handling such a system over time and questions remain regarding liability, formal training, etc. The system’s longevity has not been exhaustively tested but the flow and oxygenator parameters remained stable over days; the used components, including the Impella pumps, gas exchangers and tubes are routinely clinically used for days to weeks, adding confidence for their prolonged use in a “bridging the lung to recovery” scenario in COVID-19. Before establishing and using such last-resort setups in individual patients, the physicians, perfusionists, and institutions may thus benefit from consultation with ethical advisors and regulatory bodies. For a given patient, this may encompass a careful patient-centered assessment of potential benefit and risk, ethical and regulatory aspects, in particular examining the expanded access possibilities for medical devices in emergencies by the European Medical Agency,17,18 the US FDA,19 the Japanese regulatory agency20 and others.21

The system has not yet been applied to patients in our hospital as the ventilation and ECMO infrastructure at our site at the time of writing is not yet exhausted and does therefore not call for the application of last-resort measures, while the time window for formal, pivotal clinical trials is too narrow.


Composing ad hoc ECMO systems by integrating readily available Impella blood pumps is feasible using standard oxygenators and off-the-shelf cannulas and may offer another opportunity to oxygenate, to “recover the lungs” in viral pneumonia and potentially, save lives when standard equipment is no longer available because demand exceeds availability, for example, during the ongoing COVID-19 pandemic.


The authors thank Christoph Nix, Abiomed, Aachen, Germany, for helpful discussions and information about pressure/flow relations of Impella pumps.


1. Wu F, Zhao S, Yu B, et al. A new coronavirus associated with human respiratory disease in China. Nature. 2020; 579:265–269
2. Xu Z, Shi L, Wang Y, et al. Pathological findings of COVID-19 associated with acute respiratory distress syndrome. Lancet Respir Med. 2020; 8:420–422
3. Ruan Q, Yang K, Wang W, et al. Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China. Intensive Care Med. 2020.
4. Davies A, Jones D, Bailey M, et al. Australia and New Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO) influenza investigators. Extracorporeal membrane oxygenation for 2009 influenza A(H1N1) acute respiratory distress syndrome. JAMA. 2009; 302:1888–1895
5. 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
6. Peek GJ, Mugford M, Tiruvoipati R, et al.; CESAR trial collaboration. 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
7. Pham T1, Combes A, Rozé H, et al. REVA research network extracorporeal membrane oxygenation for pandemic influenza A(H1N1)-induced acute respiratory distress syndrome: a cohort study and propensity-matched analysis. Am J Respir Crit Care Med. 2013; 187:276–285
8. Alhazzani W, Møller MH, Arabi YM, et al. Surviving sepsis campaign: guidelines on the management of critically ill adults with coronavirus disease 2019 (COVID-19). Intensive Care Med. 2020; 28:1–34
9. Truog RD, Mitchell C, Daley GQ. The toughest triage—allocating ventilators in a pandemic. N Eng J Med. 2020; 382:1973–1975
10. Rosenbaum L. 2020. Facing Covid-19 in Italy—ethics, logistics, and therapeutics on the epidemic’s front line. N Eng J Med. 2020; 382:1873–1875
11. Rubin EJ, Baden LR, Morrissey S. Audio interview: emerging tools in the fight against Covid-19. N Engl J Med. 2020; 382:e39
12. Lassen HC. A preliminary report on the 1952 poliomyelitis epidemic. Lancet. 1953; 1:37
13. accessed on 2020-04-06
14. accessed on 2020-04-06
15. accessed on 2020-04-06
16. accessed on 2020-04-06
17. accessed on 2020-04-06
18. accessed on 2020-04-06
19. accessed on 2020-04-06
20. accessed on 2020-05-08
21. accessed on 2020-04-06

interventional cardiology; blood pumps; Impella; extracorporeal membrane oxygenator; COVID-19; pneumonia

Supplemental Digital Content

Copyright © ASAIO 2020