Persistent Hypoxemia and Biomarkers in Severe COVID-19 Pneumonia Patients Treated with Methylprednisolone

Abstract

Despite corticosteroids, biomarkers, SpO2/FiO2, and racial/ethnic groups have been associated with mortality in Severe COVID-19 Pneumonia. This is a secondary analysis of a cohort study of adults admitted from March to June 2020 in 13 hospitals in New Jersey, United States who received methylprednisolone. Out of 380 patients, those with persistent hypoxemia (SpO2/FiO2 < 315 on day 1 and day 7), mortality was 86 patients (43%) with an in-hospital-survival median 31 days [IQR: 22–48 days] whereas those without persistent hypoxemia (SpO2/FiO2 > 315 on day 1 and SpO2/FiO2 > 315 day 7), mortality was 8 patients (9.6%) with an In-hospital-survival median 60 days; IQR 25–60 days. (HR = 2.52 95% CI 1.23–5.15) P = 0.038. There was no statistically significant difference between racial/ethnic groups. At < 7 days from admission, the AUROC for SpO2/FiO2 < 315 for d-dimer was 0.644 (CI 0.485,0.802), IL-6 was 0.756 (CI 0.626,0.886), and CRP was 0.659(CI 0.510,0.809). At ≥ 7 days from admission, the AUROC for d-dimer was 0.729(CI 0.617,0.842), IL-6 was 0.697(CI 0.571,0.822) and CRP was 0.570(0.448,0.691). In our study, inflammatory markers (CRP and IL-6) have higher discriminatory ability for hypoxemia < 7 days from hospitalization, while thrombotic marker D-dimer have higher discriminatory ability ≥ 7 days.