Friday, April 16, 2021

Widespread misunderstanding of hypoxemia, hypoxia and pulse oximetry


A recent article in the American Journal of Respiratory and Critical Care Medicine about hypoxemia and covid-19 concludes with this:

In conclusion, COVID-19 has engendered many surprises, but features that baffle physicians are less strange when contemplated through the lens of long-established principles of respiratory physiology.  

Read this paper when you're well rested and well fed and you will find it a great exercise in the physiology of hypoxemia. It's much more about that, and the misconceptions derived from our current obsession with pulse oximetry, than it is about covid.

First let's list what I think are some of the main ideas in the paper.

There was widespread over-reliance and misunderstanding of pulse oximetry.

There is not a simple correlation between dyspnea and hypoxemia.

The threat of hypoxemia is poorly understood by clinicians.

The seeming paradox of asymptomatic hypoxemia is not unique to covid-19 but is explained by well established principles of respiratory physiology.

Dyspnea is mediated by hypercapnia,  afferent signals produced by inflammatory stimuli, mechanical properties of the lungs and, least of all, hypoxemia.  As will be brought out in the physiology below, multiple  conditions seem to have a permissive effect on the dyspnea produced by hypoxemia.

Hypercapnia is a very important cause of dyspnea.   It has a permissive effect on hypoxemia as a cause of dyspnea. Hypercapnia causes a drop in pH in the blood perfusing the CNS respiratory control center.  An acute increase in pCO2 of 10 mm Hg quickly causes profound dyspnea. The situation is different for hypoxemia.  As pO2 falls there appears to be a threshold of 60 mm Hg  below which stimulation of ventilation and dyspnea occur. (The so called hypoxic drive). In terms of mechanism, hypoxemia stimulates the carotid bodies which send messages to the respiratory control center. From there impulses are relayed to the cortex cruising the sensation of dyspnea. The correlation between dyspnea and the ventilatory response to hypoxemia is poor. This response to hypoxemia is blunted if the pCO2 is 39 or below.  These responses are blunted in individuals with diabetes and individuals over 65 which constitute a high portion patients presenting clinically with covid.

Could covid have effects on the brain that blunt the dyspnea response? A similar question has been asked regarding the symptom of anosmia. ACE2, the receptor for covid-19, is expressed both in the carotid bodies and the nasal mucosa so mucosal effects rather than brain involvement could account for these manifestations. This is a question yet to be answered.

The authors imply that our usual concerns about low pulse oximeter  readings are misdirected.  Another quote from the article:

Physicians are fearful of hypoxemia, and many view saturations between 80% and 85% as life threatening. We served as volunteers in an experiment probing the effect of hypoxemia on breathing patterns; our pulse oximeter displayed an SpO2 of 80% for over an hour, and we were not able to sense differences between an SpO2 of 80% and an SpO2 of 90% (24). In investigations on control of breathing and oximeter accuracy, subjects experience an SpO2 of 75% (12), or briefly 45% (25), without serious harm. Tourists on drives to the top of Mount Evans near Denver experience oxygen saturations of 65% for prolonged periods; many are comfortable, whereas some sense dyspnea (25).

The finding of a low pulse oximetry reading does not enable a complete physiological assessment. Instead it should lead the clinician to ask: what's going on? Pitfalls in the interpretation of pulse oximetry were cited in the article. Correlation with blood gas readings deteriorates at lower levels of saturation. Accuracy of pulse oximetry is less in critically ill patients than in normal volunteers. The oxyhemoglobin dissociation curve should factor into any interpretation of pulse oximetry readings but this is seldom the topic of bedside discussions. Fever and low pH, common in critically ill patients, cause a shift of the curve to the right. This leads to lower saturation readings at a given pO2. It is an adaptive mechanism by which hemoglobin unloads oxygen more readily. Herein lies another reason why oxygen saturation correlates poorly with dyspnea: the carotid bodies respond to changes in pO2 but  not to oxygen saturation.

Further complicating the discussion is the definition of terms. Though not addressed in the article, there is the common confusion between hypoxemia and hypoxia. Low pO2 or saturation readings indicate hypoxemia. However, to diagnose hypoxia, which is a reduction in oxygen delivery to the tissues, one must apply the oxygen delivery equation. This equation takes into account hemoglobin bound oxygen, oxygen dissolved in plasma, hemoglobin concentration and cardiac output. As to  the definition of hypoxemia the authors point out that it has been an evolving concept. The definition of hypoxemia is not essential, but rather a matter of usage and convention. For example, in the 1990s it was often defined as the raw number without regard to the FiO2. Recently hypoxemia is more often referred to in terms of the oxygen requirement. Both are important: the former for estimating oxygen delivery and the latter for making an assessment of ventilation and gas exchange.

After reading this article I have the following concluding thoughts:

We over rely on pulse oximetry. Blood gases are underutilized.

Misunderstanding of pulse oximetry readings is widespread.

The hypoxemia of covid-19 is not as unique as popularly believed.