Predictive Immunological, Virological, and Routine Laboratory Markers for Critical COVID-19 on Admission.

Abstract

Background Early identification of COVID-19 patients at risk of critical illness is a challenging endeavor for clinicians. We aimed to establish immunological, virological, and routine laboratory markers, which, in combination with clinical information, may allow identifying such patients. Methods Blood tests to measure neutrophil/lymphocyte ratio (NLR) and levels of ferritin, CRP, D-dimer, complement components (C3 and C4), cytokines, and lymphocyte subsets, as well as SARS-Cov-2 RT-PCR tests, were performed in COVID-19-confirmed cases within 48 hours of admission. RT-PCR cycle threshold (Ct) values from oropharyngeal or nasopharyngeal swabs were determined on the day of admission. Symptom severity was categorized as mild (grade 1), severe (grade 2), or critical (grade 3). Results Of 120 patients who were included, 49 had mild, 32 severe, and 39 critical COVID-19. Levels of ferritin >370 ng/mL (OR 16.4, 95% CI 5.3–50.8), D-dimer >440 ng/mL (OR 5.45, 95% CI 2.36–12.61), CRP >7.65 mg/dL (OR 11.54, 95% CI 4.3–30.8), NLR >3.77 (OR 13.4, 95% CI 4.3–41.1), IL-6 >142.5 pg/mL (OR 8.76, 95% CI 3.56–21.54), IL-10 >10.8 pg/mL (OR 16.45, 95% CI 5.32–50.81), sIL-2rα (sCD25) >804.5 pg/mL (OR 14.06, 95% CI 4.56–43.28), IL-1Ra >88.4 pg/mL (OR 4.54, 95% CI 2.03–10.17), and IL-18 >144 pg/mL (OR 17.85, 95% CI 6.54–48.78) were associated with critical COVID-19 in the univariate age-adjusted analysis. This association was confirmed in the multivariate age-adjusted analysis only for ferritin, CRP, NLR, IL-10, sIL-2rα, and IL-18. T, B, and NK cells were significantly decreased in critical patients. SARS-CoV-2 was not detected in blood except in 3 patients who had indeterminate results. RT-PCR Ct values from oropharyngeal or nasopharyngeal swabs on admission were not related to symptom severity. Conclusion Ferritin, D-dimer, CRP, NLR, cytokine (IL-18 and IL-10), and cytokine receptor (IL-6, IL1-Ra, and sCD25) test results combined with clinical data can contribute to the early identification of critical COVID-19 patients.