Differentiating COVID-19 Pneumonia from Acute Respiratory Distress Syndrome (ARDS) and High Altitude Pulmonary Edema (HAPE): Therapeutic implications

COVID-19 is an acute respiratory illness caused by a droplet-borne coronavirus, SARS-CoV-2. By May1st 2020 the pandemic had resulted in ~3.3 million infections, over 235,000 deaths and global disruption of trade. While 80% of people with COVID-19 suffer a minor, acute respiratory infection, the mortality ranges from 2–7%. Patients with COVID-19 pneumonia may decompensate due to hypoxemic respiratory failure. Autopsy data show inflammation, diffuse alveolar damage (DAD), alveolar fluid accumulation, and occasional hyaline membranes, consistent with acute respiratory distress syndrome (ARDS). Understanding the causes of hypoxemia in COVID-19 is complicated by a paucity of hemodynamic and autopsy data; however the presentation of COVID-19 patients is atypical of ARDS in that the hypoxemia is often profound without appropriate dyspnea, occurs despite relatively preserved lung compliance and is associated with a large intrapulmonary shunt. These traits suggest a failure of the body’s homeostatic O₂-sensing system (HOSS), which includes the pulmonary circulation, carotid body, adrenomedullary cells, and neuroepithelial bodies. The HOSS optimizes oxygen uptake and systemic oxygen delivery. Hypoxic pulmonary vasoconstriction (HPV) is the pulmonary circulation’s homeostatic response to airway hypoxia in lung diseases, such as pneumonia. HPV constricts pulmonary arteries (PA) serving hypoxic lung segments, diverting blood to better-ventilated alveoli, optimizing ventilation/perfusion (V/Q) matching. The carotid body senses hypoxemia, increasing respiratory drive. COVID-19 hypoxemia is variably attributed to ARDS, impaired HPV and a high altitude pulmonary edema (HAPE)-physiology (Figure 1). We propose that the best explanation is profound impairment of HPV and carotid body function, sometimes combined with virally-induced, ARDS.

Figure: Contrasting pulmonary hemodynamics in ARDS, COVID-19 pneumonia and HAPE.

The figure is read from top left to bottom right in clockwise direction. The hallmarks of each syndrome are summarized in the green insert table.

Top panel: HPV diverts blood away from an hypoxic alveolar segment, induced by a mucous plug in the bronchus. Blood from this segment is diverted toward a well-ventilated alveolar segment. There is no alveolar injury and segmental HPV does not elevate PA pressure or cause dyspnea.

Second panel: In ARDS the primary insult is inflammatory alveolar injury with associated pulmonary hypertension. There is also impaired HPV which increases V/Q mismatch and worsens hypoxemia. The lungs become stiff and fibrotic (green fibrous tissue and fibroblasts in alveolar wall) and hyaline membranes in the alveoli (pink crescent) are a prominent feature. Dyspnea is profound.

Third panel: In COVID-19 pneumonia the primary injury is airway inflammation due to SARS-CoV-2 infection. The resulting hypoxemia is exacerbated by an almost complete loss of HPV, resulting in a large V/Q mismatch. Unlike ARDS there is no fibrosis and hyaline membranes are less prominent. Dyspnea is variable and may be absent due to impaired oxygen sensing.

Bottom left panel: In HAPE the primary pathology is vascular, namely increased, heterogenous HPV and pulmonary venoconstriction, resulting in capillary damage and leak. Dyspnea is profound.

In pulmonary artery smooth muscle cells (PASMC) a mitochondria-based, redox sensor involving NDUFS2, regulates oxygen-sensitive potassium and voltage-gated calcium channels initiating HPV¹. HPV is inhibited by systemic vasodilators, like calcium channel blockers (CCB), and diseases, notably endotoxemia/sepsis. A similar mitochondrial mechanism accounts for O₂-sensing in the carotid body¹. Impaired carotid body function would impair respiratory drive and reduce dyspnea. SARS-CoV-2 infection of human cells changes expression of many proteins. Interestingly, 36% of all upregulated proteins and 19% of all downregulated proteins in SARS-CoV-2-infected cells are mitochondrial proteins². These include apoptosis mediators (apoptosis inducing factor and cytochrome C) and proteins involved in aerobic metabolism (relevant to the mechanism of O₂-sensing). These findings are associations, not proofs; however, they offer biologic plausibility that SARS-CoV-2 may interfere with mitochondrial O₂-sensing and cause mitochondrial-induced injury.

Several features of COVID-19 pneumonia distinguish it from typical ARDS. Patients often display little breathlessness, despite profound hypoxemia, a presentation euphemistically referred to as “happy hypoxemia”. Guan reported dyspnea in only 18.7% of 1099 hospitalized COVID-19 pneumonia patients, despite low PaO₂:FiO₂ ratios, abnormal CT scans (86%) and common requirement for supplemental oxygen (41%)³.Second, lung compliance is well preserved in COVID-19, relative to ARDS. Third, profound hypoxemia is associated with a large intrapulmonary shunt⁴. Fourth, the benefits of prone ventilation are larger than for typical ARDS. HPV is regionally variable, resulting in heterogenous V/Q matching. Prone ventilation minimizes this heterogeneity and allows HPV to divert blood flow to the lung’s dorsal/caudal regions. Although HPV is postulated to be weak in COVID-19, residual HPV may be optimized when prone. Thus, the clinical features of COVID-19 pneumonia may be explained by malfunction of O2-sensing mechanisms including impaired carotid body function or loss of HPV (Figure), although comprehensive hemodynamic assessment, including right heart catheterization, would be required to prove this. Impaired HPV may be further complicated by high dead space ventilation or pulmonary microemboli.

Recently, HAPE physiology was proposed to explain the edema and hypoxemia in COVID-19 pneumonia. HAPE is caused by exaggerated, regionally heterogenous HPV and pulmonary venoconstriction (Figure). These forces increase PA and capillary pressure, causing hydrostatic capillary damage and alveolar leak of blood and protein. We believe that in COVID-19, HPV is likely reduced, alveolar infiltrates result from the viral inflammation and PA pressures are low. Fried et al reported a 64-year old female with COVID-19 pneumonia, an O₂-saturation of 88% and patchy airspace disease⁵. She decompensated after receiving intravenous nitroglycerine. We postulate that the nitroglycerine may have reduced her already attenuated HPV, worsening V/Q matching and leading to decompensation. Her PA pressure was minimally elevated (45/20 mmHg), a finding more consistent with attenuated HPV than with HAPE (which is characterized by very high PA pressures) (Figure-table insert). In ARDS pulmonary hypertension is common but it is due to thrombosis and endogenous vasoconstrictors, whilst HPV is actually depressed. While HAPE and ARDS are both noncardiogenic forms of pulmonary edema, the initiators of the edema are very different. HAPE is a primary hemodynamic problem in the pulmonary arteries and veins, (Fig), whilst in ARDS the edema primarily results from inflammation and alveolar epithelial dysfunction. In severe COVID-19 pneumonia there are features of ARDS, but the consequences are exacerbated by loss of HPV, whether because of the viruses’ effects on mitochondria or the ability of endotoxin and inflammatory stimuli to obliterate HPV. The differentiation between HAPE and impaired HPV in COVID-19 has critical therapeutic implications. The HAPE hypothesis was accompanied by a proposal to inhibit HPV using acetazolamide, CCBs or phosphodiesterase-5 inhibitors. We predict that these drugs would exacerbate hypoxemia in COVID-19 pneumonia.

Many patients with COVID-19 have systemic hypertension (15%) and avoidance of systemic vasodilators particularly CCBs would preserve HPV. We call for research into the clinical physiology of COVID-19 pneumonia, including right heart catheterization/echocardiography. Were the impaired HPV phenotype confirmed, several drugs to reinforce HPV might merit study, including administration of almitrine (which increases HPV and activates the carotid body), and/or inhibitors of endogenous vasodilator pathways, such as the cyclooxygenase pathway (indomethacin or meclofenamate), the nitric oxide (NO) synthase pathway (methyl arginine analogs) or guanylate cyclase (methylene blue). The value of enhancing HPV is evident in single lung anesthesia which relies on HPV to create a dry operative field for lung surgery. Almitrine increases HPV and improves systemic oxygenation. Although almitrine can cause peripheral neuropathy, its ability to enhance HPV and augment carotid body function could be beneficial in COVID-19 pneumonia. Any of these strategies should be performed only in the context of an approved clinical trial.