= Alexander R. Opotowsky
{c}
{tag=Physiology and neuroscience author}
\b[Alexander R. Opotowsky] is a <physician> whose research includes adult congenital heart conditions, exercise physiology and evaluation of <cardiac output> measurements. His institutional profile identifies these scientific interests.<Alexander Opotowsky profile at Cincinnati Children's>[Research interests: exercise physiology, pulmonary vascular disease and right ventricular function]{ref}
In a 2017 primary study, Opotowsky and colleagues compared paired <thermodilution> and estimated-oxygen-uptake <Fick principle> measurements from 12,232 adults in a Veterans Affairs cohort and 3,391 adults at Vanderbilt. The investigators examined method agreement and subsequent mortality associations.<Thermodilution vs Estimated Fick Cardiac Output Measurement in Clinical Practice by Alexander R. Opotowsky and others>[Abstract, Design, Exposures and Main outcomes; JAMA Cardiology 2, 1090–1099, DOI 10.1001/jamacardio.2017.2945]{ref} Comparing two methods does not make either an error-free ground truth.
The measurement distinction can be explained by <oxygen-uptake assumptions in estimated Fick output>. At steady state, the oxygen <Fick principle> writes
$$
Q=\frac{\dot V_{\mathrm O_2}}{C_{a,\mathrm O_2}-C_{v,\mathrm O_2}},
$$
where $Q$ is <cardiac output>, the numerator is oxygen consumption per unit time and the denominator is the arterial-minus-mixed-venous oxygen content difference. Estimating the numerator from an external formula introduces an assumption that directly scales the inferred <cardiac output>; it is distinct from measuring oxygen consumption.
For stipulated oxygen uptake of 200 mL/min and a content difference of $50\,\mathrm{mL\,L^{-1}}$, $Q=4$ L/min. Substituting an estimated uptake of 250 mL/min gives 5 L/min, a 25% increase with unchanged content measurements. This is an original error-propagation example, not a patient result or a universal performance estimate. Errors in the content difference can also affect the ratio.
The primary study found that a small average difference between methods coexisted with wide limits of agreement.<Thermodilution vs Estimated Fick Cardiac Output Measurement in Clinical Practice by Alexander R. Opotowsky and others>[Abstract, Results; Figure 1, paired-method differences]{ref} Positive and negative discrepancies can cancel in a mean while remaining substantial for individuals. This separates average bias, individual agreement and prognostic association, rather than treating all three as one measure of accuracy.
\Image[https://cincinnatichildrens.widen.net/content/kg6stflb03/jpeg/20200110_Alexander-Rueben-Opotowsky_ml-4.jpg?crop=yes&h=480&keep=c&quality=80&w=320&x.app=portals]
{title=Alexander Opotowsky, explicitly named on his Cincinnati Children's institutional profile.}
Back to article page
