THE LINK between INFLAMMATION and THROMBOSIS: CLINICAL and IMMUNOHISTOCHEMICAL CHARACTERIZATION of PULMONARY ARTERIAL THROMBOSIS in COVID-19 PATIENTS

ABSTRACT

Background: Coronavirus-19 disease (COVID-19) has been responsible, to date, for more than 5 million of deaths. Immunothrombosis may be a major factor contributing to mortality in COVID-19 and pulmonary arterial tree involvement that mimics multiple pulmonary embolism could be a major contributor to disease course. Immunomodulatory drugs are of some beneft but mechanism not completely clear. We investigated pulmonary arterial tree clots to better appreciate their immunothrombotic nature, in contrast to the pathological characteristics of non-infammatory thrombi (1). Objectives: The primary objective was to study in depth the arterial thrombosis in COVID-19, by characterizing the immunohistochemical nature of thrombi, performing macroscopic and microscopic analyses, and by comparing clinical, laboratory and anatomical-pathological data of these patients with other patients died for COVID-19 but without evidence of pulmonary arterial thrombosis. Methods: Autopsies were performed in patients (cases) who died for COVID-19 with evidence of pulmonary arterial thrombosis at autopsy fnding but without pathological signs of bronchopneumonia or peripheral venous thrombosis. COVID-19 positive patients without pulmonary arterial thrombosis were selected as control group. Hematoxylin and eosin stained slides were reviewed choosing those with visible pulmonary thrombi. Further histochemical and immunohisto-chemical staining were performed in selected paraffin blocks. Each component of the thrombus was evaluated with the software application QuPath in terms of fbrin, red blood cells, platelets and immune cells percentage after scanning the slides with Aperio System. Laboratory tests were recorded at 2 points at hospital admission and at Intensive Care Unit transfer. Results: We included 13 patients (cases) and 14 controls, matched for age, gender and time from diagnosis to death. Twenty arterial thrombi were studied. By immuno-histochemistry, arterial thrombi were composed by white blood cells (WBC) [median, IQR range 10% (5-12.25)], mainly neutrophils [58% (35.2-64.5)], red blood cells [12%, (6-34.25)], fbrin [19% (14.5-42.25)], platelets [39%, (31.75-48)] (Figure 1). Three cases had a history of previous thrombosis. All cases had received anticoagulant treatment during hospitalization, low molecular weight heparin in 12/13 (therapeutic regimen in 4/12, prophylactic in 8/12) while 1/13 continued oral anticoagulants for comorbidity. By comparing laboratory fndings between cases and controls, cases showed signifcantly higher levels of platelet count [median, IQR range 195000/mmc (157750-274500) vs 143500 (113000-175250), p=0.011], LDH [854 U/l (731-1315) vs 539 (391.5-660), p=0.003)] at hospital admission, and D-dimer at ICU transfer [25072 FEU (6951-50531) vs 1024 (620-5501), p=0.003)]. Conclusion: Pulmonary arterial thrombosis in COVID-19 is a type of immune-mediated infammatory thrombosis, since the amount of WBC is 6-times more than normal value seen in non-infammatory thrombi. Some markers of infammation, necrosis and coagulation are much more increased in this subset of patients. Chest CT angiography rather than simple CT scan at hospital admission could be more useful in this setting, and treatments with antiplatelet agents or anticoagulants, eventually in combination with immunotherapy, might positively affect the outcome.