An old saying goes: The lack of oxygen not only stops the machinery, it wrecks it. Hypoxia of the brain first causes confusion, then coma, and eventually irreversible brain damage. Other organs follow that pattern, even though most can survive hypoxia longer than the brain. Thus, knowing whether arterial blood carries oxy-gen to the organs assumes great importance. Because oxyhemoglobin is red and reduced hemoglobin bluish, this color difference can be exploited to assess the oxygenation of blood. Clinically, we recognize cyanosis, but we cannot well grade the degree of bluishness.
Enter pulse oximetry. The concept is what you might call “elegant.” A probe sends light impulses into a finger (or earlobe or nose or toe) and then collects the light that has passed through the tissue. The light comprises two wavelengths: one (infra-red) more likely to be absorbed by oxyhemoglobin, the other (red) by reduced hemoglobin. By rapidly (too rapid for the eye to recognize) alternating the two wavelengths with no light at all, the unit is able to estimate the proportion of oxyhemoglobin to reduced hemoglobin. This is called “functional saturation.” Some instruments estimate (not measure) the other species of hemoglobin in blood (methemoglobin, carboxyhemoglobin) and compare the oxyhemoglobin as a percentage of the sum of all known hemoglobins. This is called “fractional saturation,” which will be a little lower than functional saturation.
We want to know the percentage of oxyhemoglobin saturation in arterial blood (rather than in the tissue or in arterial plus venous blood), therefore we need to catch the saturation reading in the artery, rather than in the whole finger. To accomplish this, the unit functions as a plethysmograph assessing the thickness of the finger (or earlobe or nose or toe). Because the tissue swells a little with each arterial pulsation, the unit can discard data arising during diastole and report on data only recorded during systole, which represent arterial blood. The saturation is reported as SpO2, the p referring to the fact that the measurement is based on pulse oximetry rather than on a direct in vitro measurement of oxygen saturation from an arterial blood sample, which would be SaO2. A healthy person breathing room air at sea level (at least not at Mount Everest) should have an SpO2 of about 98% +/− 2%. Here is a rough correlation of SpO2to arterial partial pressure of oxygen(PaO2):
SpO2 : PaO2
100% : 100 mmHg or higher
90% : 60 mmHg
60% : 30 mmHg
In patients with normal lungs and nothing more than a small physiologic shunt (2% to 4%), the PaO2 should be within spitting range of inspired oxygen pressure. If it is substantially different, a shunt is likely to exist.
There is more to pulse oximetry than outlined here. But we will not dwell on issues of other dyes interfering with the measurements, on the amount of pulsa-tion required, on the influence of venous pulsation, or on the confounding effects of external light. For all of these issues, we refer you to one of many exhaustive texts on monitoring or pulse oximetry.