The effect of disconnection of the gas supply during VA ECMO.
The data above were acquired from a 'normal', 75 kg patient on stable VA ECMO at 36.9 (C), a pump flow rate of 4.9 lpm, FiO2 of 1.0, and a fresh gas flow rate of 2.8 lpm. The supervisor had rendered the patient asystolic (so, in essence, the patient was on full cardio-pulmonary bypass).
At the start of the experiment, the PaO2 was 425 mm Hg and the PaCO2 38 mm Hg.
30 seconds into the experiment (at the black arrow), the fresh gas ('Sweep Gas') supply to the oxygenator has been set to zero.
Over the next two minutes, the PaO2 falls from 425 mm Hg to less than 50 mm Hg and thereafter declines more slowly. In contrast, the PaCO2 increases quite rapidly from 38 mm Hg to 45 mm Hg and then climbs more slowly.
Note that the inflexion in the PaCO2 curve occurs much earlier in the experiment (~60 seconds) than that in the PaO2 curve (~130 seconds).
Before we can interpret these results, we need to understand that the ‘Functional Residual Capacity’ (FRC) of a modern oxygenator is only in the order of 150 mls. Because the patient is on full cardio-pulmonary bypass, he has no pulmonary blood flow and, as a result, the patient’s own FRC is not available to act as an oxygen store when gas supply to the oxygenator fails.
In the early stages of gas supply failure, oxygen continues to be transferred from the oxygenator’s FRC into the blood. When the gas phase PO2 is equal to the PO2 of the incoming venous blood, further exchange ceases and thereafter the blood phase PO2 declines at a rate that is largely determined by the patient’s metabolic rate.
The situation regarding CO2 transfer is somewhat different. As soon as gas flow ceases, there is a rapid equalization of blood and gas phase PCO2 and for this reason the PaCO2 increases very quickly to that of the incoming venous blood. As with PaO2, the subsequent increase in PaCO2 occurs at a rate that is largely determined by the patient’s metabolic rate.