Background: Thrombosis is frequent during COVID-19 disease, and thus, identifying predictive factors of hemostasis associated with a poor prognosis is of interest. The objective was to explore coagulation disorders as early predictors of worsening critical conditions in the intensive care unit (ICU) using routine and more advanced explorations. Materials: Blood samples within 24 h of ICU admission for viscoelastic point-of-care testing, (VET), advanced laboratory tests: absolute immature platelet count (A-IPC), von Willebrand-GPIb activity (vWF-GpIb), prothrombin fragments 1 + 2 (F1 + 2), and the thrombin generation assay (TGA) were used. An association with worse outcomes was explored using univariable and multivariable analyses. Worsening was defined as death or the need for organ support. Results: An amount of 85 patients with 33 in critical condition were included. A-IPC were lower in worsening patients (9.6 [6.4–12.5] vs. 12.3 [8.3–20.7], p = 0.02) while fibrinogen (6.9 [6.1–7.7] vs. 6.2 [5.4–6.9], p = 0.03), vWF-GpIb (286 [265–389] vs. 268 [216–326], p = 0.03) and F1 + 2 (226 [151–578] vs. 155 [129–248], p = 0.01) were higher. There was no difference observed for D-dimer, TGA or VET. SAPS-II and A-IPC were independently associated with worsening (OR = 1.11 [1.06–1.17] and OR = 0.47 [0.25–0.76] respectively). The association of a SAPS-II ≥ 33 and an A-IPC ≤ 12.6 G/L predicted the worsening of patients (sensitivity 58%, specificity 89%). Conclusions: Immature platelets are early predictors of worsening in severe COVID-19 patients, suggesting a key role of thrombopoiesis in the adaption of an organism to SARS-CoV-2 infection.

Background: The different waves of SARS-CoV-2 infection have strained hospital resources and, notably, intensive care units (ICUs). Identifying patients at risk of developing a critical condition is essential to correctly refer patients to the appropriate structure and to spare limited resources. The soluble form of RAGE (sRAGE), the endoplasmic stress response and its surrogates, GRP78 and VEGF-A, may be interesting markers. Methods: This was a prospective monocenter cohort study of adult patients admitted to the ICU for severe COVID-19 pneumonia. The plasma levels of sRAGE, GRP78 and VEGF-A were measured within the first 24 h. Patients were classified as critical if they further needed vasopressor therapy, renal replacement therapy, or invasive mechanical ventilation, or died during their ICU stay, and were otherwise classified as not critical. Results: A total of 98 patients were included and 39 developed a critical condition. Critical patients presented higher sRAGE (626 [450–1043] vs. 227 [137–404] pg/mL, p < 0.0001), interleukin-6 (43 [15–112] vs. 11 [5–20] pg/mL, p < 0.0001), troponin T (17 [9–39] vs. 10 [6–18] pg/mL, p = 0.003) and NT-pro-BNP (321 [118–446] vs. 169 [63–366] pg/mL, p = 0.009) plasma levels. No difference was observed for VEGF-A and GRP78. The variables independently associated with worsening in the ICU were sRAGE (1.03 [1.01–1.05] per 10 pg/mL) and age (1.7 [1.2–2.4] per 5 years). An sRAGE value of 449.5 pg/mL predicted worsening with a sensitivity of 77% and a specificity of 80%. Conclusion: sRAGE may allow the identification of patients at risk of developing a critical form of COVID-19 pneumonia, and thus may be useful to correctly refer patients to the appropriate structure of care.