Hypoxia—An overlooked trigger for thrombosis in COVID‐19 and other critically ill patients

In a recent publication, the GEMELLI against COVID‐19 group from Italy demonstrated higher incidence of deep vein thrombosis among non‐ICU patients despite pharmacological thromboprophylaxis.¹ They identified traditional risk factors like underlying malignancy and immobility as being significantly more prevent in the thrombosis group in addition to the need for high‐flow oxygenation. This finding of hypoxia as a trigger for thrombosis in COVID‐19 has so far not been explored in detail.

Hypoxia as a risk factor for venous thromboembolism (VTE) has been suggested based on studies in mountain dwellers who had disproportionately higher incidence compared to low‐altitude dwellers and also the rare congenital disorders of hypoxia‐sensing pathways.2., 3. High altitude itself, however, may not be the major risk factor in this situation in keeping with the rarity of the so‐called economy class syndrome and thrombosis in those who perform mountaineering activities. Because the likelihood of developing VTE is greater among people who stayed at moderate to high altitude for prolonged periods of time, chronic hypoxia is likely the unique risk factor among these populations.³ When translating this to patients with COVID‐19, prolonged hypoxia prior to hospitalization or even during hospitalization may significantly worsen the thrombotic complication (immunothrombosis) initiated by the virus. The importance of hypoxia in VTE was recently demonstrated in the clinical Tromsø Study in which the risk of mortality was two‐fold higher in patients with chronic obstructive pulmonary disease (COPD) compared to those with normal airflow with the risk being five‐fold higher if the patients had severe COPD (stage 3/4).⁴ The same group demonstrated that oxygen saturations less than 96% had a 1.5‐fold higher risk of VTE in patients with COPD.⁵ A similar hypoxic state could be one of the explanations for the higher incidence of thrombosis in COVID‐19 patients who require mechanical ventilation. Although difficult to prove due to the presence of several confounding factors, the significant improvement in outcomes noted by prone positioning of the ventilated COVID‐19 patient may be due to improved oxygenation and thus modulating the “hypoxithrombosis” trigger.

There have been both proponents and opponents for oxygen deprivation being a procoagulant phenomenon.2., 6. The contradictors may have focused on otherwise healthy individuals with pristine vessel walls, who are unlikely to develop thrombosis in the presence of acute hypoxia. On the other hand, in those with underlying health conditions, who are at higher risk of thrombosis in the COVID‐19 setting, hypoxia may indeed be procoagulant.⁶ Yan and colleagues have shown decreased oxygen levels to activate the transcription factor early growth response‐1, which upregulates tissue factor expression in mononuclear phagocytes.⁶ Alterations in fibrinolytic systems, especially increased expression of plasminogen activator inhibitor‐1, along with this increased tissue factor expression on the most abundant lung macrophages, can lead to pulmonary thrombi. 6., 7.

One of the notable findings in COVID‐19 thrombosis management is the failure of standard dose anticoagulation with heparin. No clear explanations have come forth so far for this anticoagulation inefficiency and trials are examining higher dose and even therapeutic doses of anticoagulation in such patients. Once again, hypoxia may be one of the culprits for this antithrombotic inadequacy. Hypoxia has been shown to induce heparanase activity in the tumor micro‐environment, which helps in cancer progression by facilitating tumor cell dissemination.⁸ Heparanase can degrade both unfractionated heparin and low molecular weight heparins and is one of the reasons described for heparin failure in cancer‐associated thrombosis.⁹ Increased heparanase activity induced by hypoxia may similarly lead to reduced heparin anticoagulant effect. The “hypoxia‐heparanse” concept may also explain the common occurrence of heparin resistance in critically ill patients (with and without COVID‐19) especially those on extracorporeal membrane oxygenators, even in the absence of explanations like antithrombin deficiency or high fibrinogen levels.

Another interesting aspect of “hypoxithrombosis” concept is the role of reoxygenation after a period of hypoxia as contributing to VTE development.¹⁰ Elegant studies from the Wagner lab showed that reoxygenation following exposure to hypoxia for 24 hours significantly increased release of von Willebrand factor (VWF) levels from endothelial Weibel–Palade bodies and could stimulate thrombosis through glycoprotein Ibα and VWF interaction.¹⁰ This pathophysiology model may be one of the explanations for the very high levels of VWF noted in COVID‐19 patients, higher mortality of patients who had respiratory compromise after several days being in hospital (reoxygenation), and once again anticoagulant failure (because anticoagulant drugs do not impact glycoprotein Ibα and VWF interaction).

In summary, hypoxia that occurs in moderate to severe COVID‐19 is a likely contributor to the thrombotic complications, often seen in these patients. Early intervention including prophylactic anticoagulation is important in patients who develop hypoxic symptoms rather than shielding at home without an urgent health care professional assessment. If the patients present after several days of symptoms related to hypoxia, prophylactic anticoagulation may be inadequate because the thrombotic process is likely to have spiralled and more aggressive and multimodal therapeutic approaches may be required for a good outcome. In the post‐COVID era, hypoxia should be considered as an additional risk factor in all patients who require hospitalization for cardiorespiratory illnesses; especially those who require respiratory support in critical care units. In these patients, hypoxia should also be considered as a mechanism for heparin resistance.

CONFLICTS OF INTEREST

None.