MSE Logo

Adult ECMO Simulator

The effect of a decrease in cardiac output on arterial PO2.

CO and Saturation

The effect of a reduction in cardiac output (CO) is shown in the figure above. The data shown above are from a simulated patient with severe respiratory failure and an oxygen consumption of ~200 ml/min on stable VV ECMO at 5.0 lpm. Initially, the patient's cardiac output (CO) is 8.7 lpm, but, at the black arrow, the CO is reduced to 3.3 lpm. This change in output is not accompanied by any change in ECMO system flow (which remains at 5.0 lpm) - or in the patient's metabolic rate.

Note how the PaO2 (which had been steady at about 64 mm Hg) initially rises (to over 120 mm Hg) and then progressively falls to about 72 mm Hg.

The explanation of this observation is very similar to that advanced in the section entitled 'The effect of an increase in cardiac output on arterial saturation.'.

There are three competing influences on PaO2 when cardiac output decreases in a patient on VV ECMO.

First, PaO2 tends to fall because the decreased output leads to an decrease in mixed venous saturation which, in turn, magnifies the contribution which the shunted blood makes to the arterial PO2. (Refer to the recording entitled 'The effect of cardiac output on SaO2 when Qs/Qt remains constant.' for a more complete explanation of this effect.)

Second, the decreased output tends to increase the fraction of the total venous return which passes through the oxygenator and, as a result, mixed venous PO2 rises.

Finally, there is a more-or-less linear relationship between CO and Qs/Qt. As the output falls, so there is a commensurate decrease in Qs/Qt.

The effect of these mechanisms can be seen in the figure above. Initially, when CO decreases, the PaO2 rises as less of the mixed venous blood 'bypasses' the ECMO system. However, because CO has been decreased, (provided that the metabolic rate remains constant,) the arterial and mixed venous oxygen saturations now start to fall ('The effect of cardiac output on SaO2 when Qs/Qt remains constant.') and this decrease in saturation partially counteracts the rise in saturation which occurs as a result of the ECMO system processing a greater proportion of the total CO.

It should be noted that the relationship between changes in CO and Qs/Qt was only added in version 5.7 of the simulator. The author is grateful to Dr Roger Pye (of St Vincent's Hospital, Sydney) for advice on this matter.