 Coronary oxygen delivery

Coronary oxygen delivery is the rate at which oxygen is brought to a specified myocardial territory in its arterial blood supply. If volume flow is $Q$ and arterial oxygen content is $C_a$, thenFlow and content must use compatible volume units and refer to the same territory and interval. Delivery is an inflow quantity; coronary oxygen uptake is the net amount removed from the blood by the represented tissue.

For an original example, $Q=250\mathrm{mL}\mathrm{mi}{\mathrm{n}}^{-1}$ and $C_a=0.20\mathrm{mL}{\mathrm{O}}_2\mathrm{m}{\mathrm{L}}^{-1}$ give $D_{{\mathrm{O}}_2}=50\mathrm{mL}{\mathrm{O}}_2\mathrm{mi}{\mathrm{n}}^{-1}$. The same delivery can arise from $Q=200$ with $C_a=0.20$, or $Q=250$ with $C_a=0.16$: both give $40\mathrm{mL}{\mathrm{O}}_2\mathrm{mi}{\mathrm{n}}^{-1}$. Delivery alone cannot identify which factor changed.

Oxygen content is not the same observable as oxygen partial pressure or oxygen saturation. Much of blood's oxygen is carried by hemoglobin; the amount of carrier therefore matters as well as its saturation.^[ref] Substituting a pressure value for $C_a$ would also fail dimensionally.

An original independent-error approximation givesRelative standard uncertainties $5\%$ and $2\%$ combine to about $5.39\%$ under the stated assumptions. A covariance term is required when the input errors are correlated.^[ref]

The measurement apparatus determines which input is actually observed. A primary coronary-flow study used continuous saline infusion and a pressure/temperature-sensing guidewire, testing volumetric flow estimation in five dogs and 35 patients.^[ref] Such a temperature-dilution observation does not itself measure oxygen content; a delivery estimate must obtain that input separately.

Mass-specific myocardial oxygen delivery adds a tissue-mass denominator. Total delivery $40\mathrm{mL}{\mathrm{O}}_2\mathrm{mi}{\mathrm{n}}^{-1}$ to $200\mathrm{g}$ corresponds to $20\mathrm{mL}{\mathrm{O}}_2\mathrm{mi}{\mathrm{n}}^{-1}$ per $100\mathrm{g}$. A bulk value, a mass-normalized value and a regional value answer different questions, and normalizing a total does not reveal uneven distribution within the territory.