The purpose of this study was to classify patients with severe COVID-19 into more detailed risk groups using coagulation/fibrinolysis, inflammation/immune response, and alveolar/myocardial damage biomarkers, as well as to identify prognostic markers for these patients. These biomarkers were measured every day for eight intensive care unit days in 54 adult patients with severe COVID-19. The patients were classified into survivor (n = 40) and non-survivor (n = 14) groups. Univariate and multivariate analyses showed that the combined measurement of platelet count and presepsin concentrations may be the most valuable for predicting in-hospital death, and receiver operating characteristic curve analysis further confirmed this result (area under the curve = 0.832). Patients were consequently classified into three groups (high-, medium-, and low-risk) on the basis of their cutoff values (platelet count 53 × 103/µL, presepsin 714 pg/mL). The Kaplan–Meier curve for 90-day survival by each group showed that the 90-day mortality rate significantly increased as risk level increased (P < 0.01 by the log-rank test). Daily combined measurement of platelet count and presepsin concentration may be useful for predicting in-hospital death and classifying patients with severe COVID-19 into more detailed risk groups.

No abstract

Background: coronavirus disease 2019 (COVID-19) patients are frequently complicated with COVID-19 associated coagulopathy. This time, we investigated whether there was a relationship between the CAC pattern and the risk of in-hospital death in severe COVID-19 patients.Methods: We enrolled 54 severe COVID-19 adult patients admitted to intensive care unit (ICU). Patients were divided into two groups whether in-hospital death (non-survivor) or not (survivor). We measured not only various coagulation/fibrinolysis markers, but inflammation/immunoresponse markers which were cytokines or chemokines, and alveolar/myocardial damage markers in peripheral blood every day for 8 ICU days. Univariate logistic regression analysis and receiver operating characteristic curve (ROC) analysis of the area under the curve (AUC) were used to identify the optimal combination of biomarkers to predict in-hospital death.Results: Among all 54 patients, the patients were classified into the survivor (n = 40) and non-survivor (n = 14) groups. We found that the platelet count decreased significantly and presepsin (P-SEP) increased significantly in non-survivor group compared to survivor group from the time of ICU admission. Using univariate logistic regression analysis, platelet count and P-SEP are found to be predictive of in-hospital mortality and ROC analysis using these two markers could further confirm this result (AUC=0.832; sensitivity: 64.3%, specificity: 90.0%). The cut-off values for in-hospital mortality for these markers were 153 x 103/µL in platelet count and 714 pg/mL in P-SEP. The patients were consequently classified into the following 3 groups: (1) High risk (n = 7), platelet count <153 x 103/µL and P-SEP >714 pg/mL; (2) Low risk (n = 27), platelet count >190x103/µL and P-SEP <714 pg/mL; (3) Medium risk (n = 20), areas other than (1) and (2). And the 90-day mortality increased significantly as the risk of the group increases (p<0.01). Additionally, the platelet count was lower and P-SEP level was higher in the non-survivor group than in the survivor group on ICU days throughout the 8 days.Conclusions: Simultaneous daily platelet counts and P-SEP measurements from the time of ICU admission would be a useful method for predicting in-hospital mortality.