Abstract
BACKGROUND: The aim of this study was to elucidate the relationship between HA levels and clinical outcomes, including disease severity and prognosis. METHODS: A retrospective multicenter cohort analysis was conducted across three medical centers in China from September 2022 to February 2023. Patients with positive nucleic acid for the novel coronavirus were included in the study. The primary endpoint is 28-day mortality.The severity was assessed by the 'Diagnosis and Treatment Plan for Novel Coronavirus Pneumonia (Trial Version 9)'. Analyses assessed HA correlation with severity (Kendall), predictive value for mortality (logistic regression, ROC, Delong test), and survival differences (KM analysis using a Youden index-derived HA cutoff). RESULTS: A total of 862 patients (median age 72; 58.47% male) were included in the study, of whom 108 died. Non-survivors were older, more men, lower BMI, and more complicatinos. The overall median HA level was 121.65 ng/ml, significantly higher in non-survivors(p < 0.001). Kendall correlation analysis revealed a significant correlation between HA level and COVID-19 severity (correlation coefficient: 0.365, p < 0.001). Our preliminary analysis revealed positive correlations between HA and D-dimer and chest CT. In logistic regression models, HA was an independent predictor of 28-day mortality(p = 0.001). The AUROC for HA was 0.840 (95% CI: 0.806-0.974). The cut-off value was 141.65 ng/ml (sensitivity 61.27%, specificity 91.67%). KM survival analysis showed that patients with HA levels higher than 141.65 ng/ml had a significantly lower survival probability(p < 0.001). CONCLUSION: HA may potentially function as a biomarker for COVID-19 and could provide a vital reference for the clinical treatment.