What should we add to the intensivists mask to relieve “hypoxic happiness” in COVID-19 patients?

Lessons from integrated systems physiology

Julian FR Paton, Igor Felippe, David J Paterson¹, and Joseph Donnelly²
Department of Physiology, University of Auckland, New Zealand
¹Department of Physiology Anatomy & Genetics, University of Oxford, Oxford, UK
²Department of Anaesthesiology, University of Auckland, New Zealand.

Response to the Guardian article: ‘Happy hypoxia’: unusual coronavirus effect baffles doctors
https://www.theguardian.com/world/2020/may/03/happy-hypoxia-unusual-coronavirus-effect-baffles-doctors

We all know the safety announcement given prior to take off by airlines “…..in the event of a loss of cabin pressure, oxygen masks will fall automatically from the ceiling; only help others after fitting the mask to yourself first”. Have you ever asked yourself, why can’t I help others first? Loss of cabin pressure is comparable to parachuting to the top of Mount Everest, where the air is so thin that within seconds you would fall unconscious. Indeed, the reason you may need to help your neighbour on losing cabin pressure is because unconsciousness would ensue within seconds. The repeated reports of “happy hypoxia” or “silent hypoxia” observed in many COVID-19 patients defies all prior presentations of patients with acute respiratory distress syndrome and viral pneumonias. We wish to consider possible reasons why there is an apparent lack of response to clinically severe hypoxia in patients with SARS-CoV-2 infection.

Classically the early signs of hypoxia are anxiety, confusion, and restlessness, yet COVID-19 patients remain calm, composed and follow directions even though they are obviously cyanotic. Normally, continual exposure to severe hypoxia causes pronounced hyperventilation resulting in hypocapnia. However, although COVID-19 patients can have blood oxygen saturations as low as 60% (PaO2 ~70 mmHg) they maintain a sub-normal PaCO2 of ~35 mmHg and are poorly responsive to supplemental oxygen. Consciousness is typically lost at an oxygen saturation of around 70%, but COVID-19 patients are coherent. So how can these mysteries be disentangled?

Carbon dioxide (CO₂) could be the key to many of these apparent conundrums.

Loss of consciousness during severe hypoxia is caused in part by reduced cerebral blood flow that occurs because cerebral arteries vasoconstrict in the presence of low CO₂ thereby diminishing the delivery of oxygen to the brain. With respectable PaCO2’s in COVID-19 patients brain blood flow is maintained. The tachycardia seen in these patients acts to support cardiac output and as a result arterial pressure is near normal, thus maintaining good cerebral perfusion pressure. Within the brain, which is highly metabolically active, the presence of severe hypoxia will reduce aerobic metabolism thereby lowering locally generated CO₂. Couple this with near normal cerebral blood flow and CO₂ will be rapidly washed out from the brain thereby lowering brain tissue PCO₂ well below PaCO₂. The relevance of this is that it may explain the “happiness” in “happy hypoxia”.

Dyspnoea (breathlessness) and anxiety are triggered by central chemoreceptors, which are located in the ventrolateral medulla oblongata, and are exquisitely sensitive to CO₂ and protons that are generated when CO₂ is hydrated. At normocapnia, central chemoreceptors are active and critical for driving normal breathing. An example of their function is the breaking point reached after a breath hold caused by the increasing hypercapnia within the brain. Central chemoreceptor activation plays a major role in triggering the sensation of dyspnoea and anxiety via their neuronal connections to the reticular activating and limbic systems. Thus, it is quite possible that COVID-19 patients have partially or fully inactivated central chemoreceptors due to low brain tissue CO₂ and/or depressed by hypoxia (see Fig).

 

Whilst the modest ventilatory response to the severe hypoxaemia in “happy hypoxic” COVID-19 patients may reflect depressed central chemoreceptor function, there may be an additional reason. Peripheral arterial chemoreceptors such as the carotid bodies are activated by both low oxygen tensions and hypercapnia. With PaO₂ around 70 mmHg these receptors would be buzzing with activity that would normally stimulate breathing. From our reading, patients have a raised respiratory frequency (~30 breaths/min) but minute ventilation is rarely quantified with some reports suggesting shallow breaths in “happy hypoxic” patients, resulting from lung oedema and J-receptor stimulation. There is evidence that patients can increase their breathing. We hypothesise that the peripheral chemoreceptors may be hijacked by SARS-CoV-2 (see Fig). First, SARS-CoV-2 may down regulate hypoxia inducible factor, which is essential for oxygen sensing in the carotid body, as has been shown by some respiratory viruses. Second, SARS-CoV-2 is neurotoxic as evidenced by the loss of smell and taste, which occurs early in the disease process in some patients. Taste is mediated by sensory nerves connecting the tongue to the brainstem that have their soma located within the petrosal ganglion. It is this ganglion that also relays carotid body signals to the brain. Dysfunctional peripheral chemoreceptor input remains to be validated, but it would be consistent with the mediocre hyperventilatory response and maintenance of PaCO₂ observed in COVID-19 patients.

Blood oxygen is dependent on its ability to diffuse across the alveolar membrane as well as the affinity of haemoglobin for oxygen. The mild hypocapnia would shift the oxyhaemoglobin curve to the left increasing haemoglobin’s affinity for oxygen. However, this classic physiological mechanism may be futile in face of ground glass patterning and some pulmonary consolidation seen at time of presentation of the “happy hypoxic” COVID-19 patient. These non-ventilated portions of the lung can cause a shunting of deoxygenated blood and worsening of hypoxaemia. Initial experience indicates that lying the patient on their belly, and opening up more of lung for business, can improve oxygenation. Moreover, the emerging evidence of microfibrinous clotting within alveolar capillaries, endotheliitis and possibility of reduced affinity of haemogloblin for oxygen could all contribute to the severe systemic hypoxaemia.

So, what else could be in the oxygen mask that falls automatically from the intensivist to the “happy hypoxic” COVID-19 patient? We return to CO₂ with the notion to re-engage central chemoreceptors and increase central respiratory drive to enhance ventilation while the lung remains compliant, thereby avoiding potential issues of a ventilator. To enhance respiratory drive, re-breathing or adding some CO₂ to the inspired air, doxapram (Dopram, Stimulex or Respiram), which is well known to stimulate peripheral and central chemoreceptors could be infused intra-venously, and oral benzolamide or acetalzolamide could be used to dam-up C0₂ in tissue, thereby enhancing arterial CO₂. This will have the advantage of maintaining good brain blood flow.

SARS-CoV-2 has challenged clinicians and academics globally because of the unique patterns of symptoms presented. Only through continued integrated thinking across specialties and disciplines and applying what we already know about systems physiology will we discover the secrets of treating SARS-CoV-2 in intensive care medicine.