NACB - Scientific Shorts
NACB - Scientific Shorts (formerly NACB Blog)
By William E. Winter, MD, DABCC, FACB

Assume that a patient suffers from carbon monoxide exposure and that their arterial hemoglobin saturation is reduced to 75% as measured by co-oximetry. According to the oxy-hemoglobin saturation curve, the arterial pO2 should be ~40 mm Hg when the hemoglobin saturation is 75% (figure 1). However when you actually measure the pO2, it is normal at 95 mm Hg. How can this be true (that the arterial pO2 is within the reference interval and is not reduced)?


Using multiple wavelengths the co-oximeter measures the saturation of hemoglobin calculated as follows: the oxy-hemoglobin divided by the oxy-hemoglobin plus the deoxy-hemoglobin plus the carboxyhemoglobin and the methemoglobin. Since carboxyhemoglobin does not function in oxygen transport (and because methemoglobin levels are normally negligible), even if the remaining hemoglobin is 100% saturated, this only represents 75% of the total hemoglobin being saturated which was correctly reported by the co-oximeter. Recall that total hemoglobin measurements (in g/dL) includes all forms of hemoglobin (e.g., functional and non-functional: carboxyhemoglobin and methemoglobin).

Now here is the zing --- > because the oxy-hemoglobin dissociation curve only applies to oxy- and deoxyhemoglobin and in the absence of lung disease, hypoventilation and environmental hypoxia, the "functional" hemoglobin (the normal hemoglobin which excludes carboxyhemoglobin and methemoglobin) will be nearly 100% saturated. Therefore figure 1 cannot be used as depicted to predict the pO2. Because the functional hemoglobin is nearly 100% saturated, we do expect a pO2 of near 100 mm Hg(figure 2). Indeed if the hemoglobin saturation is 98%, the predicted pO2 is 95 mm Hg.

As a side note, in the absence of lung disease, hypoventilation or environmental hypoxia, the pO2 is also expected to be normal in cases of methemoglobinemia.


Please log in to leave a comment!

Posted by
On 12/20/2011

My comment about "oxygen gap" concerns the specificity of constelation of low Hb saturation and normal pO2. Wojciech Gernand, MD, PhD LOMA Medical Laboratory Olesno, Poland

Posted by
On 12/16/2011

How do you think, is it possible to make a suspicion of CO exposure when such "oxygen gap" is observed? This comment was approved by the NACBLOG editorial board. Please remember to add your name and affiliation!

Posted by
On 12/9/2011

The key to understanding this is that pO2 represents pressure, not amount, of oxygen. In a patient with CO poisoning, the amount of oxygen is reduced, but the pressure is not. Thus blood gases (pH, pCO2, pO2) without oximetry are misleading. This is a point I make regularly with pathology residents rotating through our lab. Phil Bach, PhD Primary Children's Medical Center Salt Lake City, UT

About the Author
William E. Winter, MD, DABCC, FACB
William E. Winter, MD, DABCC, FACB 

Additional Contributors to this blog post:

Neil S. Harris, MD, Associate Professor of Pathology

Lindsay Bazydlo, PhD, Assistant Professor Pathology