ECMO Overview

Extracorporeal Membrane Oxygenation is the use of an oxygenator and pump to provide mechanical pulmonary or cardiopulmonary support in the setting of severe, acute, and reversible respiratory or cardiac failure.

ECMO is classified into modes by the location of the return cannula, and *configuration** by the vessels used.

Configurations are covered in more detail in the specific section.

Components

Necessary pieces of the circuit include

  • Access cannula
    Multistage cannula with side holes along the distal section, which drains venous blood.
  • Return cannula
    Returns circuit blood to the patient via a single hole in the tip.
  • Oxygenator
    Facilitates:
    • O2 provision
      Up to 7L/min, though diminishing returns with both modes means these flows are rarely seen in practice.
    • CO2 offloading
      Amount of removal is a function of the sweep gas flow (FGF in L/min). Usually this is set 1:1 with circuit flows on commencement of ECMO.
    • Warming
  • Pumps
    • Centrifugal pump
      Magnetically coupled pump that generates rotary force with a small area of blood-pump contact. These:
      • Minimise pump-head thrombosis risk
      • Minimise haemolysis
      • Can be clamped
        Circuit can be safely clamped with the pump running, as the contiguous blood column means that the tubing downstream from the pump will not be pressurised.
      • Afterload sensitive
        Flow rate ↓ with an ↑ in afterload for any given RPM.
    • Roller pump
      Pump compresses tubing, moving blood though.
      • Afterload insensitive
      • Cannot be clamped whilst running
        Downstream pressure will continue to rise and tube rupture will result.
  • Tubing
    Albumin and (often) heparin-bonded circuits reduce coagulation factor activation.
  • Controller
    Used to set:
    • Pump speed
      Via setting RPM (flow may vary) or setting flow.
    • Alarms

Concepts

Key concepts include:

  • Access insufficiency
  • Fresh Gas Flow
  • Transmembrane pressure

Access Insufficiency

Access insufficiency describes ‘suck-down’ of the vena cava onto the access cannula, resulting in partial or total occlusion of the cannula access holes and a ↓ in circuit flow. Access insufficiency:

  • Presents as
    • ↓ ECMO flows
      Alarms.
    • Circuit chatter
      Circuit shudders violently, and often audibly.
  • Can cause significant adverse effects
    • Hypoxia Severe ↓ in ECMO support in VV ECMO.
    • Haemodynamic instability
      Severe ↓ in ECMO support in VA ECMO, or secondary to hypoxia in VV ECMO.
    • Haemolysis
    • Vascular trauma
  • Occurs due to combination of:
    • Small IVC/SVC volume
      • Intravascular hypovolaemia
      • Extrinsic compression
      • ↓ VR
        • ↑ intrathoracic/intraabdominal pressure
          • Coughing
          • Movement
    • High ECMO circuit flows
    • Poor cannula position
      • Multistage cannula have multiple access holes along their length and are preferred for this reason
      • Ideal position depends on circuit configuration, however access is improved the closer the cannula is to the RA
  • Managment
    • Reduce ECMO flows to a level that maintains flow and gives tolerable parameters
    • Correct precipitant

Fresh Gas Flow

The FGF to the oxygenator should be adjusted to:

  • Maintain oxygenation
    Only a small threshold FGF (usually 0.5-1L/min) is required to adequately oxygenate ECMO blood.
  • CO2 clearance
    FGF is predomiantly titrated to achieve desired PaCO2, as ↑ FGF will ↓ PaCO2. The amount of FGF required is inversely proportional to:
    • Native lung function in VV ECMO
    • Native lung function and native CO in VA ECMO

Transmembrane Pressure:

The lowest pressure part of the ECMO circuit is directly before the pump, the highest pressure part is directly after the pump.

Access ports (if included) should be placed after the pump so that any opening or disconnection results in blood leaking rather than air entrainment.

The transmembrane pressure is the pressure differential across the oxygenator. The TMP:

  • Indicates the degree of thrombus burden
  • Should be <10mmHg per L/min of circuit flow

Complications

Major complications of ECMO support include:

  • Bleeding
  • Air Embolism
  • Haemolysis

Bleeding

Bleeding is a major complication and cause of death of patients on ECMO support. Bleeding can occur from:

Management of bleeding is covered under Bleeding.

  • Cannulation sites
    • Common
    • Difficult to manage
    • ↓ with percutaneous cannulation technique without surgical dissection
  • Pulmonary
    • Common in patients with associated lung injury

Air Embolism

Air embolism can occur if there is a breach in any negative pressure part of the circuit. Air embolism:

  • Commonly occurs from:
    • 3-way taps
      ↓ Risk by minimising number of unnecessary connectors.
    • Connectors
    • Venous access sites
  • Leads to:
    • Pump head air lock
      In a centrifugal pump will partially or completely ↓ circuit flow due to air trapped above the disk.
    • Distal air embolism

Haemolysis

Severe haemolysis may occur with oxygenator or pump head thrombosis. Severe haemolysis:

  • Is suggested by Plasma Free Hb >1.0g/L
  • Requires urgent circuit change

Key Studies

Trials in ECMO have been limited by challenges in ethically conducting placebo-control trials, as well as logistic difficulties. However, key trials include:

  • CESAR (2009)
    • Britons aged 18-65 with potentially reversible severe respiratory failure without contraindications to heparin or Pip >30
      • Multi-centre (92 tertiary, 11 conventional, 1 ECMO, assessor-blinded RCT
      • 80% power to detect 18% ↓ ARR in death or severe disability from control group of 73%
    • Transfer to ECMO centre vs. conventional ventilation
      • Transfer ECMO centre for consideration of ECMO
        • Standardised low-volume, low-pressure ventilation in majority
        • If no response to ventilation in 12 hours, then VV ECMO initiated (68/85, 75%)
          • Percutaneous cannulation
          • Roller pumps
          • PMP oxygenators
      • Conventional ventilation
        • Unprotocolised PCV or HFOV
    • Significant ↑ in survival without severe disability in ECMO group (63% vs. 47%, RR 0.69 (CI 0.05 - 0.97))
    • Secondary outcomes: ECMO group had significantly:
      • ↓ Proning (36%. vs 42%)
      • ↑ Steroids (84% vs. 64%)
      • ↑ MARS (17% vs. 0%)
      • ICU and hospital length of stay
    • Significant ↑ in survival with transfer to an ECMO centre, even if you didn’t get ECMO, compared to management in a non-ECMO centre
    • No protocolised ventilation in conventional group
    • Significant loss to follow up

CESAR was not so much a study of ECMO, as to whether transfer to an ECMO centre results in better management of respiratory failure.

  • EOLIA (2018)
    • 249 non-pregnant Europeans with ARDS with severe hypoxia despite adjunctive therapies and no ECMO contraindications
      Severe hypoxia including
      • P/F <50mmHg for 3 hours
      • P/F <80mmHg for 6 hours
      • pH <7.25 with PaC2 >60mmHg for >6 hours with lung protective ventilation
    • Prospective, multicenter, industry-supported RCT
    • 331 patients powers for 20% ARR in ECMO group, assuming 60% control mortality
    • VV ECMO vs. standard care
      • ECMO group
        • Percutaneous femoral/jugular configuration
        • FiO2 and pump flow adjusted for SpO2 >90%
        • Sweep adjusted for PaCO2 <45mmHg
      • Standard care
        • Lung protective ventilation
        • Encouraged:
          • NMBA
          • Proning
          • Recruitment
        • Inhaled pulmonary vasodilators permitted if above ineffective
        • Crossover to VV ECMO if SpO2 <80% despite the above
          28% crossed over.
    • Stopped early for futility
      Underpowered due to early stoppage and lower than expected control mortality.
    • No difference in mortality at day 60 (35% vs. 46%, RR 0.76 (CI 0.55-1.04)))
      This is clouded by the high proportion of control patients crossing over to receive ECMO.
      • If none of them got ECMO and as few as 30% survived, there would be a significant mortality benefit from ECMO
    • Secondary outcomes showed the ECMO group had:
      • ↓ “Treatment failure” (RR 0.62, CI 0.47-0.82)
        Death or crossover to ECMO.
      • ↓ RRT
      • ↑ Thrombocytopenia
      • ↑ Bleeding
    • Slow (6 years) recruitment

References

  1. Combes A, Hajage D, Capellier G, et al. Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome. New England Journal of Medicine. 2018;378(21):1965-1975. doi:10.1056/NEJMoa1800385