= Oxygen-uptake assumptions in estimated Fick output
\b[Oxygen-uptake assumptions in estimated Fick output] arise when a <Fick principle> calculation substitutes predicted <oxygen consumption> for a direct measurement. With an <arteriovenous oxygen content difference> $\Delta C=C_a-C_{\bar v}$,
$$
Q=\frac{\dot V_{\mathrm O_2}}{\Delta C},\qquad
\widehat Q=\frac{\widehat{\dot V}_{\mathrm O_2}}{\widehat{\Delta C}}.
$$
The measured <blood> samples and the predicted numerator are different sources of information. Precise <arterial oxygen content> and <mixed venous oxygen content> measurements cannot by themselves validate the assumed <oxygen consumption>.
A primary comparison explicitly examined estimated-oxygen-uptake <Fick principle> output against <thermodilution>. Its clinical observations therefore compare particular measurement procedures, not a conservation identity against an unrelated physical law.<Thermodilution vs Estimated Fick Cardiac Output Measurement in Clinical Practice by Alexander R. Opotowsky and others>[Abstract, Importance, Objectives and Exposures]{ref}
An original exact error decomposition is
$$
\frac{\widehat Q}{Q}
=\frac{\widehat{\dot V}_{\mathrm O_2}}{\dot V_{\mathrm O_2}}
\frac{\Delta C}{\widehat{\Delta C}}.
$$
Suppose actual <oxygen consumption> is $250$ mL/min and $\Delta C=5$ mL/dL. Actual <cardiac output> is then $50$ dL/min, or $5$ L/min. Assuming $200$ mL/min with the correct content difference produces $4$ L/min. If the measured content difference were also underestimated to $4$ mL/dL, the two errors would cancel numerically and again produce $5$ L/min. Agreement of the final answer would not prove either input accurate.
The numerator must also represent the state during sampling. A prediction for one physiological condition need not describe <oxygen consumption> after a change in activity or metabolism. The steady-state balance assumes compatible sampling and no continuing oxygen accumulation within the represented system. Near a small content difference, even a modest absolute error in $\Delta C$ can strongly change the inferred <blood flow>. Reporting estimated uptake as an assumed input keeps these model and sampling limitations visible.
Back to article page
