Routine care of the patient on ECMO

The process of having your blood sucked out of your body and forced through a gas-enriched sponge is neither effortlessly self-sustaining nor consequence-free. One can predict that some sort of complications might arise, and some sort of monitoring and maintenance might be required.

Idiosyncratic features of ICU management for the ECMO patient

• There will be no "routine" line changes.
• There will be no chest drain insertions.
• There will be no alcohol-based cleaning solutions applied to the circuit (it dissolves the circuit tubing!) - betadine is the only cleaning agent allowed.
• There will be a minimum of transfers
• Transport should be in-hours, if at all.
• The oxygenator must be positioned below the level of the patient: it is microporous, meaning that plasma cannot exit the circuit but gas can enter it: the circuit will entrain gas if ever the blood pressure is lower than gas pressure.

Ventilation and oxygenation management for VV ECMO

• The ventilator should be set to a low mandatory rate (4-6 breaths/min)
• Small tidal volume should be used, with low pressure
• PEEP may be maintained around 10
• FiO₂ should be minimised

Ventilation and oxygenation management for VA ECMO

• Ventilation should be as normal as pssible, to avoid the native cardiac output from accidentally perfusing the brain and heart with hypoxic "stale" blood from the sluggish pulmonary circulation

Circulatory management for VV ECMO

• The arterial waveform is unaffected
• The PA catheter is useless for measuring cardiac output by thermodilution

Circulatory management for VA ECMO

• The arterial waveform may be absent in the absence of cardiac activity.
• The PA catheter may be used to monitor native cardiac output by thermodilution

Dialysis

• This can be connected to the circuit, as vas cath insertion would be a bloody business.

Transfusion threshold

• Most intensivists set an arbitrary target for Hb: 100-120g/L

Anticoagulation

• APTT should be 50-70, corresponding to an ACT of 180-200

Routine nursing care

• Hourly neurovascular limb obs
• Tracheal suctioning no more frequently than two-hourly
• Pressure area care is performed once a day, and with a senion medical staff member present (in case of sudden circuit problems)

Drug adsorption by the circuit

• Sequestration by the ECMO circuit is a known thing. Drugs passing though the circuit can become adsorbed onto the membrane, and cleared from the circulation.
• There appears to be a correlation between the lipophilicity of the drug and the extent of its absortion: drugs which are most lipophilic are also the most "circuit-philic".
• Silicone membrane oxygenators adsorb substantially more drug molecules than polypropylene oxygenators.
• For the silicone circuit, the following fraction of drug remains in circulation after 180 minutes of exposure to the circuit:
  • Midazolam 0.62%
  • Morphine 23.9%,
  • Fentanyl 0.35%
  • Paracetamol 34.0%
  • Cefazolin 84.3%
  • Meropenem 82.9%
  • Vancomycin 67.8%
  (this was performed in a circuit primed with blood but not attached to any human organism; the drug loss was therefore inevitably due to circuit adsorption, rather than any sort of metabolism. Different studies get slightly different values, but all agree that fentanyl and midazolam adsorption is a major issue)
• According to the  Wildschut article,  the  silicone membrane roller pump circuits retain significantly more midazolam and fentanyl than the centrifugal pump hollow-fibre membrane oxygenator: midazolam 63.4 versus 0.62%, fentanyl 33.8 versus 0.35%.
• After a prolonged period of ECMO, the "circuit compartment " becomes saturated, and begins to exchange back into the patient. The oxygenator membrane can thus become a rich source of fentanyl and midazolam. Thus far, I am not aware of any sort of recreational abuse potential for discarded membranes.

Additionally, some other factors play a major role, even though they are not specific to the ECMO scene:

• Massive fluid shifts will be inevitable, and the volume of distribution for many substances will be increased, thereby decreasing the concentration of available drug.
• However the albumin level will be low (well, after all it's a "critical illness ") and the amount of free drug available will be increased, most prominently for highly protein-bound drugs.