[Skip to Navigation]
[Skip to Navigation Landing]
March 1980

To the Editor.

Author Affiliations

Jamaica Plain, Mass

Arch Intern Med. 1980;140(3):440. doi:10.1001/archinte.1980.00330150154040

Perhaps my ignorance of statistical terms and methods handicaps my understanding of the conclusions drawn by the authors of the article "Relationship Between Arterial and Venous Bicarbonate Values." In their text, the authors state, "Indeed, if arterial pH and Paco, values are available, we can think of no situation in which venous CO2 levels would have clinical relevance." This would imply that they found little correlation between arterial and venous bicarbonate values. Yet, if "the linear correlation between measured venous HCO-3 and actual calculated arterial HCO-3 values was significant (P <.001), with an r2 of.7553" means what I think it does, then the comparative values were actually very close.

Knowledge of venous HCO-3 is necessary for calculating anion gap, an essential step in determining the various causes of metabolic acidosis.1 And when venous HCO-3 is elevated in association with a low serum chloride and/or low serum potassium level, I doubt