CO2, pCO2, and bicarbonate: What do we really measure?

Carbon dioxide (CO₂) is a byproduct of several human metabolic processes, most notably the Citric Acid (Krebs) Cycle, which produces two molecules of CO₂ per cycle. The fate of metabolically produced CO₂ is determined partly by chemistry and partly by respiration. Three measures are used clinically to assess CO₂ status: total CO₂, partial pressure of CO₂ gas (pCO₂), and bicarbonate (HCO₃-). The specific parameter that is measured can be a source of confusion.

In aqueous solution, CO₂ can exist in 4 chemical forms. As a gas at room temperature, CO₂ has a dissolution coefficient of 1.45 g per liter of water under 1 atm (100 kPa) pressure, so it can exist as dissolved CO₂ gas. CO₂(g) reacts with water to form carbonic acid, H₂CO₃, a weak acid with a pKa of 6.35 for the first proton, and 10.33 for the second proton. Loss of the first proton produces the monovalent bicarbonate anion, HCO₃-, and loss of the second proton produces the divalent carbonate anion, CO₃=. Because carbonic acid (H₂CO₃) is an intermediate in the carbonic anhydrase-catalyzed reaction of CO₂(g) with H₂O to produce bicarbonate (HCO₃-) and H⁺, its existence in solution can be considered equivalent to the concentration of CO²(g). Since the pKa for the deprotonation of bicarbonate to produce carbonate is >10, its concentration in aqueous solution at physiological pH is negligible. Hence, virtually all CO² in blood is either in the form of dissolved CO²(g) or bicarbonate. At physiological pH, the ratio of CO²(g) to bicarbonate is approximately 1:20.

All the major automated chemistry platforms (Roche, Beckman, Abbott, Siemens, Ortho Vitros) measure total CO₂ by the enzymatic reaction of bicarbonate with phosphoenolpyruvate (PEP) to form oxaloacetate, followed by the malate dehydrogenase-catalyzed conversion of oxaloacetate to malate with the concomitant oxidation of NADH to NAD⁺, which is monitored spectrophotometrically by the decrease in absorbance at 340 nm. Because bicarbonate is in equilibrium with dissolved CO₂(g), Le Chatelier’s principle requires that consumption of bicarbonate in the PEP reaction results in conversion of CO₂(g) to bicarbonate to re-establish equilibrium. Therefore, these methods measure all the CO₂ present, or Total CO₂.

The concentration of a dissolved gas ordinarily is expressed as the pressure (or partial pressure) it exerts. CO₂(g) can be measured by allowing it to diffuse across a gas-permeable membrane into an aqueous solution and monitoring the change in pH as the CO₂(g) reacts with water to form bicarbonate and hydrogen ions. This does not significantly disrupt the equilibrium between CO₂(g) and bicarbonate in the blood. Hence, the concentration of CO₂(g) in whole blood can be measured by direct potentiometry using a flow cell, a gas-permeable membrane, and a pH electrode. 

This brings us to bicarbonate: how do we measure it? Although the PEP reaction described above measures bicarbonate, it does not measure its physiological concentration because the reaction disrupts the equilibrium with CO₂(g), converting the gas to bicarbonate. Direct potentiometry using a bicarbonate ion-selective electrode is one way to measure the physiological bicarbonate concentration, but interferences from hydroxide and chloride ions limit the use of these devices in clinical applications. The most common way to determine the bicarbonate concentration is to calculate it using the Henderson-Hasselbalch equation:

A blood gas instrument measures the pH and pCO₂ by potentiometric methods. The CO₂(g) concentration can be calculated from the pCO₂ using a conversion factor of 0.03 mmol/L/mmHg, and pKa for carbonic acid, expressed as CO₂(g) is 6.35, providing everything needed to solve the equation for the bicarbonate concentration. The 19th century chemist and Nobel laureate Jacobus Van’t Hoff derived the relationship between the equilibrium constant for a chemical reaction and temperature, so the calculated bicarbonate concentration will depend on the temperature at which the pH and pCO₂ are measured and assumes the patient’s temperature is the same. A correction can be made if the patient is hypo- or hyperthermic, although there is debate over whether this is a necessary or advisable practice.

Additional Reading

1. Sanghavi SF, Albert TJ, Swenson ER. Will the real bicarbonate please stand up? Annals of the American Thoracic Society. Jul 2022;19(7):1226-1229.
2. Hansen JE, Sue DY. Should blood gas measurements be corrected for the patient’s temperature? New England Journal of Medicine. Aug 7 1980;303(6):341
3. Higgins C. Temperature correction of blood gas and pH measurement—an unresolved controversy. https://acutecaretesting.org (2012). Accessed 10 Sept 2025.