key: cord-1030503-whfguaf8
authors: McBane, Robert D.
title: Arterial thrombosis and COVID-19.
date: 2020-12-23
journal: Mayo Clin Proc
DOI: 10.1016/j.mayocp.2020.12.009
sha: 62ea8375020fbebf8597d4b08cf2c984d50d40b0
doc_id: 1030503
cord_uid: whfguaf8

nan

Thrombotic complications from SARS COV2 infections (COVID 19) have generated considerable interest and concern due to the adverse impact on outcomes including survival. While macrovascular thrombosis involving named vessels have been reported with varying incidence, microvascular thrombosis particularly involving the pulmonary circulation appears to be widespread particularly in patients with severe and critical disease (1) (2) (3) . There is a growing understanding of the mechanisms of thrombosis which appear to be unique and engage various contributions from the immune system such that a recently coined term "immunothrombosis" is aptly applied (4) . Following inhalation, viral entry into the host alveolar epithelium occurs through endocytosis facilitated by engagement of the viral capsid spike protein and the host angiotensin converting enzyme 2 (ACE-2) receptor (5) (6) (7) . Once infected, the intimate relationship between alveolar epithelium and capillary endothelium promotes rapid inflammation and injury of both such that loss of the epithelial -endothelial cell barrier with diffuse alveolar damage (ARDS), increased lung stiffness, and impaired gas transfer ensues. On the capillary side, endothelial activation leads to expression of cellular adhesion molecules (ICAM, VCAM, P-Selectin, E-Selectin), release of ultra large von Willebrand factor multimers, and recruitment of neutrophils and platelets.

Neutrophil -macrophage feedback mechanisms result in a "storm" of cytokine release including TNFα, interleukin 1β, IL6, IL8, and CRP, each of which has been associated with disease progression and mortality outcomes. Complement activation may be unique to this thrombotic pathophysiology. While viral opsonization, lysis of infected cells, and inflammation amplification may be beneficial, bystander host cellular damage with associated thrombosis can occur as a consequence. Beyond direct endothelial injury, biochemical changes occur which promote local vasoconstriction through loss of nitric oxide synthesis and endothelin 1 release. Tissue factor expression, loss of endogenous anticoagulant pathways, and fibrinolysis inhibition promote microvascular thrombosis further worsening pulmonary perfusion. These combined host responses explain autopsy findings of a high incidence of diffuse alveolar damage and pulmonary microvascular thrombosis. This latter finding appears to be particularly prevalent amongst patients dying from COVID 19 pneumonia. Compared to patients dying from influenza A (H1N1), those with SARS COV2 have a 9 fold increased prevalence of J o u r n a l P r e -p r o o f pulmonary microthrombosis implying unique pathophysiology beyond simple diffuse lung injury (8) . While SARS COV2 has also been shown to infect endothelial cells from other vascular beds including heart, liver, and kidney, widespread microthrombosis has not been consistently found when complete autopsies have been performed. This suggests that the thrombotic response requires more than endothelial cell infectivity but also damage from adjacent tissues, more prevalent in the lungs compared to other organ systems.

Macrovascular thrombosis of named arteries and veins have also been well described. Most reports have focused on venous thromboembolism with much higher event rates recorded for the pulmonary circulation compared to arm or leg veins. Some have speculated that this finding argues for pulmonary artery thrombus in situ as opposed to more traditional embolic event from leg veins. One could envision a thrombus originating in the pulmonary arterioles, capillaries, or venules with propagation into main pulmonary arteries as a consequence. Alternatively, a thrombus could originate in a large pulmonary artery following sufficient endothelial infection and injury.

Unraveling this distinction is complicated. Some have suggested that finding occlusive thrombi within the macro-circulation of pulmonary arteries suggests thrombus in situ.

Neither this finding nor absence of deep vein thrombosis on arm or leg imaging is sufficient proof for this hypothesis.

Arterial thrombotic events have received less attention. Many recent epidemiology studies have failed to report an incidence of myocardial infarction, stroke or peripheral embolism. The few studies which have provided numbers reported incidence varying between 2 -5%. The current report in the Proceedings is therefore a welcomed addition to this effort (9) . In this issue of the proceedings, Fournier and 

Disease severity was assigned based on the percentage of lung involvement by CT imaging. Results were compared to a control cohort of patients admitted with arterial thrombosis in the absence of COVID-19 infection. Of the 531 COVID-19 patients admitted, 5.6% suffered either a myocardial infarction (n=9), stroke (n=8), or acute back/subacute limb ischemia (n=6). These patients had similar cardiovascular risk profiles compared to non COVID patients hospitalized for ischemic events. In contrast to the control group a minority of COVID-19 patient has developed thrombi at atypical locations including the aorta, splenic and renal arteries, and small vessels of the brain.

Mortality rates were nearly 3 fold higher compared to COVID patients without arterial thrombosis and more than 10 fold higher compared to other thrombosis patients without COVID. Elevated fibrin D-dimer levels were an independent predictor of arterial thrombosis in COVID patients and the authors provide a D-dimer threshold of 1250 ng/mL above which the receiver operator curves show reasonable sensitivity and specificity for arterial thrombotic events.

While microvascular thrombotic mechanisms appear to be well explained by direct infection and injury of pulmonary endothelial cells, the macrovascular arterial thrombotic mechanisms are less clear. Although arterial endothelial injury in this context is plausible, it is also possible that these events represent a response to critical illness in the context of native arterial disease and associated risk factors. For example, a troponin leak in the setting of hypotension and coronary artery stenosis could generate demand ischemia, the so-called Type II NSTEMI. Stress induced cardiomyopathy (Takotsubo cardiomyopathy), paroxysmal atrial fibrillation, or paradoxic embolization through a patent foramen ovale in the setting of pulmonary hypertension related to acute lung injury could each provide a source of embolism to the brain, splanchnic circulation, or peripheral arteries. Vasospasm related to pressor therapy, iatrogenic arterial injury, or spontaneous dissection are also plausible mechanisms in these critically ill patients. While it may be tempting to simply attribute the arterial occlusion to COVID-19, it is imperative to explore these potentially treatable entities in the diagnostic evaluation.

J o u r n a l P r e -p r o o f Management of these patients is clearly complicated. Adding an arterial occlusion to a patient already tenuous from a respiratory insult clearly adds to both morbidity and mortality. When identified, the underlying status of the patient often precludes a surgical intervention whereby the patient is in the ICU, on a ventilator and requiring pressor support. The experience with thrombolytic therapy is limited to small case series and cannot be advocated under general circumstances. Lost in this discussion is the limited reporting of major hemorrhage outcomes, an area which requires further study and emphasis. Without thorough understanding of bleeding outcomes, designing prevention and treatment strategies for this potentially devastating illness is challenging.

J o u r n a l P r e -p r o o f

Pulmonary Arterial Thrombosis in COVID-19 With Fatal Outcome: Results From a Prospective

Postmortem examination of COVID-19 patients reveals diffuse alveolar damage with severe capillary congestion and variegated findings in lungs and other organs suggesting vascular dysfunction

Autopsy Findings and Venous Thromboembolism in Patients With COVID-19: A Prospective Cohort Study

Anticoagulation in Hospitalized Patients with Covid-19

The coagulopathy, endotheliopathy, and vasculitis of COVID-19

Coagulation Abnormalities and Thrombosis in Patients Infected With SARS-CoV-2 and Other Pandemic Viruses

COVID-19-associated coagulopathy: An exploration of mechanisms

Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19

Arterial Thrombotic Events in Adult Inpatients With COVID-19