Abstract
Background : COVID-19 disease severity markers include mostly molecules related to not only tissue perfusion, inflammation, and thrombosis, but also biomarkers of neural injury. Clinical and basic research has demonstrated that SARS-COV-2 affects the central nervous system. The aims of the present study were to investigate the role of neural injury biomarkers and to compare them with inflammatory markers in their predictive ability of mortality. Methods : We conducted a prospective observational study in critically ill patients with COVID-19 and in a cohort of patients with moderate/severe disease. S100b, neuron-specific enolase (NSE), and inflammatory markers, including soluble urokinase plasminogen activator receptor (suPAR), were measured on intensive care unit or ward admission, respectively. Statistical comparisons between patient groups were performed for all biomarkers under investigation. Correlations between different biomarkers were tested with Spearman correlation coefficient. Receiver operating characteristic curves were plotted using mortality as the classification variable and the biomarker levels on admission as the prognostic variables. Results : A total of 70 patients with COVID-19 were included in the final analysis. Of all studied biomarkers, s100b had the best predictive ability for death in the intensive care unit, with an area under the curve of 0.73 (0.61-0.83), P = 0.0003. S100b levels correlated with NSE, interleukin (IL)-8, and IL-10 (0.27 < rs < 0.37, P < 0.05), and tended to correlate with suPAR ( rs = 0.26, P = 0.05), but not with the vasopressor dose ( P = 0.62). Conclusion : Among the investigated biomarkers, s100b demonstrated the best predictive ability for death in COVID-19 patients. The overall biomarker profile of the patients implies direct involvement of the nervous system by the novel coronavirus.
