Blood oxygen content estimated from PO2, PCO2, and pH

Oxygen content of blood (265 samples, 25 dogs) from two experimental groups was estimated from measurements of blood PO2, PCO2, pH, hemoglobin, and body temperature with the computational subroutines of Olszowka and Farhi (6). Each computed value was compared with the corresponding oxygen content measured directly by a galvanometric oxygen fuel cell (Lex-O2-Con analyzer). The computed values had consistent errors when compared with the corresponding fuel cell determinations. Linear regression of fuel cell determinations on the corresponding computed values was used to obtain a corrected estimate of oxygen content. When the linear regression correction procedure was performed within each of 25 dogs, the resultant estimates were substantially improved over a lumped correction. The mean absolute value of the difference between the corrected calculated value within an individual dog and the corresponding oxygen fuel cell content determination was 0.38 +/- 0.28 (6 anesthetized dogs, 76 samples) and 0.17 +/- 0.15 (19 awake dogs, 189 samples) (SD) vol% for the two experimental groups. It is concluded that blood oxygen content may be accurately estimated from PO2, PCO2, and pH when appropriately corrected. The advantage is that only a small subset of the total number of blood samples needs to be analyzed with the oxygen fuel cell.