Coronavirus disease (COVID-19) exhibits a wide spectrum of clinical severity and has been associated with specific biomarkers linked to disease progression and outcomes. This retrospective study analyzed sera from 1222 adult COVID-19 patients hospitalized at the University Hospital for Infectious Diseases in Croatia. We examined the association between several laboratory biomarker levels measured at patient admission and disease severity, fatal outcomes, viral variants and clinical parameters. Deceased patients and surviving patients with severe COVID-19 exhibited significantly elevated levels of several biomarkers on admission, including hs-troponin T, N-terminal pro-brain natriuretic peptide, creatine kinase, C-reactive protein, procalcitonin, interleukin-6, lactate dehydrogenase, lactate, urea and creatinine. Random forest models identified lymphocyte percentage, D-dimers, and hs-troponin T as the most important biomarkers for fatal outcome prediction, achieving 84.1% accuracy. Patients infected with the Delta SARS-CoV-2 variant exhibited significantly higher levels of proinflammatory, cardiac and renal biomarkers. Vaccination correlated with reduced proinflammatory parameters and higher lymphocyte proportions. Hypertension, chronic renal disease and diabetes were associated with increased cardiac, renal and metabolic biomarker levels, respectively. These findings highlight the association of several laboratory biomarkers with COVID-19 severity, viral variants, vaccination status and comorbidities, potentially offering prognostic insights into COVID-19 outcomes.

Severe COVID-19 is characterized by immune dysregulation, yet reliable biomarkers associated with adverse outcomes remain limited. Immunoglobulin G (IgG) N-glycosylation modulates antibody effector function and reflects the immune system’s inflammatory state. In this study, we investigated whether specific IgG glycosylation profiles are associated with COVID-19 severity, mortality, and clinical course in patients without known prior SARS-CoV-2 infection. Of the 806 admitted COVID-19 patients included in the study, 377 hospitalized patients were longitudinally monitored, and 2001 serum samples were analyzed. Significant alterations in IgG N-glycome composition were observed in patients with severe disease and those who died. The most prominent changes included increased agalactosylation and decreased mono- and digalactosylation, features associated with a more pro-inflammatory IgG glycosylation profile. Nosocomial infections were associated with a more pro-inflammatory IgG glycosylation profile. No statistically significant differences in IgG N-glycome composition were observed between vaccinated and unvaccinated patients, although the small number of vaccinated patients limited this analysis. The observed glycomic profiles likely reflect immune responses characteristic of the early phase of the COVID-19 pandemic. Whether similar glycosylation changes occur in other infectious or inflammatory diseases remains to be established. IgG glycosylation analysis holds promise for personalized risk stratification and early therapeutic intervention in COVID-19 and other severe infectious diseases.