Role of Serum E-Selectin as a Biomarker of Infection Severity in Coronavirus Disease 2019
Alessandra Oliva, Emanuele Rando, Dania Al Ismail, Massimiliano De Angelis, Francesca Cancelli, Maria Claudia Miele, Raissa Aronica, Vera Mauro, Federica Di Timoteo, Lorenzo Loffredo, Claudio Maria Mastroianni
Abstract
Introduction: E-selectin is a recognized marker of endothelial activation; however, its place in Coronavirus Disease 2019 (COVID-19) has not been fully explored. Aims of the study are to compare sE-selectin values among the Intensive Care Unit (ICU)-admitted and non-admitted, survived and non-survived patients and those with or without thrombosis. Methods: A single-center study of patients with COVID-19 hospitalized at Policlinico Umberto I (Rome) from March to May 2020 was performed. Simple and multiple logistic regression models were developed. Results: One hundred patients were included, with a median age (IQR) of 65 years (58–78). Twenty-nine (29%) were admitted to ICU, twenty-eight (28%) died and nineteen (19%) had a thrombotic event. The median value (IQR) of sE-selectin was 26.1 ng/mL (18.1–35). sE-selectin values did not differ between deceased and survivors (p = 0.06) and among patients with or without a thrombotic event (p = 0.22). Compared with patients who did not receive ICU treatments, patients requiring ICU care had higher levels of sE-selectin (36.6 vs. 24.1 ng/mL; p < 0.001). In the multiple logistic regression model, sE-selectin levels > 33 ng/mL, PaO2/FiO2 < 200 and PaO2/FiO2 200–300 were significantly associated with an increased risk of ICU admission. sE-selectin values significantly correlated with a neutrophil count (R = 0.32 (p = 0.001)) and the number of days from the symptoms onset to hospitalization (R = 0.28 (p = 0.004)). Conclusions: sE-selectin levels are predictive of ICU admission in COVID-19 patients. Since data on the relation between sE-selectin and COVID-19 are scarce, this study aims to contribute toward the comprehension of the pathogenic aspects of COVID-19 disease, giving a possible clinical marker able to predict its severity.