Wednesday, April 22, 2020

Applying Applied Physiology to COVID: Silent Hypoxia and The Work of Breathing

Enough time has passed and enough clinical experience with the behavior of COVID gained that we can now apply some applied physiology to our understanding of this aweful illness.  I don't have the answers any more than the next person, but I do know that whatever the behavior of COIVD, it can be viewed through the lens of what we do know about applied lung physiology, and some inferences can be made.  This post will be about that physiology and those inferences, using the frameworks in these lectures about applied physiology.

One of the striking early observations about patients presenting with COVID was their apparent lack of distress despite very low oxygen saturations.  This was termed "silent hypoxia" at some point, and I leave it to internet sleuths to trace its origins.  It led to a policy of early intubation of some COVID patients:
“Never in my life have I had to ask a patient to get off the telephone because it was time to put in a breathing tube,” said Dr. Richard Levitan, who recently spent 10 days at Bellevue Hospital Center in Manhattan.
There are two simple explanations for the observed phenomenon of silent hypoxia: first is that hypoxia is not nearly as potent a stimulus of respiration as is hypercarbia, as I took great pains to explain in the Applied Physiology lectures, using the example of shallow water breath hold divers who do not pre-oxygenate and emerge with sats on the order of 50%, the same as climbers on Mt Everest.  (See also this post on the boy in the wheel well, unpressurized, from California to Hawaii.)  I thank Matt Wong (@EM_phile on Twitter) for pointing me to this video which provides a striking illustration of lack of dyspnea with sudden and severe hypoxemia.  This video is compulsory viewing:

Dyspnea is driven far more by ventilation demands and the work of breathing than it is by hypoxia alone.  This is underrecognized.  I wager this stems in part from the fact that in most diseases that cause hypoxia, there is significantly elevated work of breathing because of parenchymal disease/infiltration (elastic loads) and airway secretions (resistive loads), combined with minute ventilation loads from high CO2 production, dead space ventilation (Vd/Vt) and metabolic acidosis.  If COVID does not have all of those loads, patients may not experience substantial subjective dyspnea despite low oxygen.

Notice key word "if".  Here, I wish to avoid fostering a debate about the specific pathophysiology of COVID because it remains uncertain.  But we are hearing reports about (and seeing patients) with better compliance (the inverse of elastance) than we were accustomed to seeing with severe ARDS before COVID. Likewise, we are not seeing patients or hearing reports about purulent or copious airway secretions (until later in the course, inspissated in ETTs) and we are not hearing about bronchospasm either.  The patients are very tachypneic, reflecting a high CO2 production (from cytokines, fever etc.) as well as high Vd/Vt fraction (of course a presumption, but a reasonable one).  They are also not having metabolic acidosis early in the course.

If these circumstances are true (and there is fair reason to believe they are), then COVID is an illness where the only loads imposed on the respiratory system are high minute ventilation loads.  (This is ignoring obesity, which also imposes an inertial mass load, and this may partly explain why the obese are not faring well with COVID - they have a second load imposition.)  People can handle very very high minute ventilations for prolonged periods of time.

To wit: consider that a trained athlete at VO2 max has a minute ventilation (Ve) of approximately 200 liters per minute!  Now, that is not sustainable, but it reflects Vt (tidal volumes) of 4000cc and a respiratory rate of 50 breaths per minute.  These values have been confirmed time and again over many years.  An untrained person, say a 70 kg 30-year-old male of average height has a MVV (maximal voluntary ventilation) of about 140 liters per minute.  It has been estimated that an otherwise healthy person can sustain 50% of the 15-second MVV indefinitely.  You read that right, indefinitely - meaning that 30-year-old can breathe upwards of 70 liters per minute (say, 1.5 liter Vt at 45 breaths per minute) for a long time.  So, it should not be surprising that people breathing 40 times per minute are not "tiring out" if they do not have excessive loads on the respiratory system excepting the high minute ventilation.

My experience with COVID is limited, especially compared to many others who have been battling the beast for weeks.  However, among the patients I have seen, many of these observations are borne out, and this applied physiology seems to apply to COVID just as it does to all other types of respiratory illness.  It just appears that COVID may have some unique nuances that make the physiology more difficult to appreciate, especially given the uncertainties of a novel illness.

Through the lens of applied physiology, it is not surprising that noninvasive correction of hypoxemia - even if that means HFNC along with a NRB mask - allows people to be adequately supported even with very high respiratory rates and for prolonged periods of time.  Of course, some patients will not be able to sustain spontaneous ventilation with severe COVID.  Patients with obesity (inertial mass loads), pre-existing weakness (inability to sustain the high loads), worsening parenchymal disease and compliance (increased elastic loads), advanced age (lower MVV), or marked increases in CO2 production or Vd/Vt may experience workload imbalance and frank respiratory failure.  It is our job to discern them from the patients who can sustain the loads and who will be spared all of the iatrogenesis that results from prolonged sedation and mechanical ventilation.  I posit that applied respiratory physiology is the Rosetta Stone for that differentiation.


  1. Here is a link for the 50% MVV sustainable “indefinitely” contention

    And note also that the congenital heart docs also talk about well tolerated hypoxemia without respiratory distress. Why? Because vascular or intracranial shunt (or PAVMs) do not impose workloads on the respiratory system.


Note: Only a member of this blog may post a comment.