Discrepancies in blood oxygen analyses by the methods of Van Slyke and Henderson-Smith. 1

Abstract

Loosely bound oxygen is liberated from the hemoglobin in blood by the addition of potassium ferricyanide. In the Van Slyke method, 2 all the gases are exhausted by means of a Toricellian vacuum from a laked blood-ferricyanide mixture and measured directly. In the Henderson-Smith method, 3 the oxygen is evolved into a fixed volume of air, a portion of which is analyzed directly for oxygen by absorption with alkaline pyrogallate. After application of all the corrections suggested by the authors of these methods, the results are not identical,—analyses by the Van Slyke method yielding 4 to 10 volumes per cent. more oxygen than those by the Henderson-Smith method. The divergence between the results from each method may represent a variation of 17 to 64 per cent. A few typical figures are cited in Table 1.

It is clear that there is a constant factor or factors at play, inherent in the methods of analysis employed, which ought to account for this discrepancy.

The gas evacuated by the Van Slyke procedure is not all oxygen but probably contains in addition to nitrogen, minute amounts of carbon monoxide, hydrogen, methane and the rare atmospheric gases. Bohr 1 states that blood contains I .23 volumes per cent. of nitrogen (incorrectly quoted by Van Slyke as 0.9 vol. per cent.) and 0.22 volumes per cent. of the other gases-a total of about 1.45 volumes per cent. of gas not oxygen. We have absorbed with alkaline pyrogallate, the oxygen from the gas extracted in the Van Slyke procedure and have found in all cases a residue of 0.055 to 0.082 C.C. from 2 C.C. of blood-an average of about 3.3 volumes per cent. This residue does not contain CO₂, and we have reason to believe that it is practically all nitrogen.