Oxygen saturation by pulse oximeter (SpO2), as compared with the "true" value obtained by arterial puncture (SaO2), are generally within 2 or 3 percent down to about 80%. This is pretty good, but pulse oximeters become progressively unreliable below 80%. We will usually do an arterial puncture on a patient when the SpO2 is in the 80s or lower, so we can get the true SaO2, and PO2 which is actually more meaningful.
It is worth pointing out that SpO2 needs to be put into context. It only measures Hgb saturation, not the amount of hemoglobin or oxygen carrying capacity. You can be severely anemic (low Hgb), yet still be fully saturated. Oxygen delivery to tissues is DO2 = SaO2 x Hgb x blood flow.
Physiologically it is PO2 (not SaO2) that determines gas unloading at the tissues, symptoms of hypoxia, and loss of consciousness. Free divers tend to blackout at PO2 around 25-35 mm Hg, but it also depends on CO2 level, and other things we haven't figured out yet. There is a lot of variability here though.
As a rough rule of thumb, you can think "90=60 and 60=30", meaning SaO2 90% = PO2 60 mm Hg and SaO2 60% = PO2 30 mm Hg, but many factors can shift the curve. The relationship between PO2 and SaO2 also depends on the Bohr shift related to CO2 and pH, so you can't really predict PO2 based on an O2 sat. For example, SpO2 of 80% can mean PO2 is anywhere from 45 to 60 mm Hg, depending on CO2, pH, and temperature.
It is true that it can be hard to get a reading from a sat probe on patients in shock, due to peripheral vasoconstriction, but if you can get a decent signal, and it is over 80%, it is probably reasonably good. The same should be true of divers.
As for using pulse oximeters in BH diving or apnea, they are probably good for general comparison between dives, and studying rates of drop and recovery, but the numbers become more inaccurate as they drop into hypoxia. Anything less than 60% is probably meaningless, except to say that you are very hypoxic.
Pulse oximeters are really great instruments. I don't know how we could survive without them nowadays in clinical practice. We call O2 sat the fifth vital sign (along with pulse, resp, BP, and temp). I will sometimes make decisions on admitting vs discharging patients in the emergency department based on their SpO2. There is definitely a place for these devices in simple apnea research studies, but their limitations need to be appreciated.
John Fitz-Clarke, M.D., Ph.D.
Dalhousie University
Halifax, Canada