Background/Objective: The aim of our study is to comprehensively assess the diagnostic usefulness of serum hyaluronic acid (HA) determination in COVID-19 patients. Methods: The study group included 87 patients with COVID-19 disease and 45 healthy subjects. The HA concentration was measured using the immunochemical method. Results: The serum HA concentration was significantly higher in the COVID-19 patients before admission to hospital than that in the controls (p < 0.001). Differences were found in HA levels between the groups categorized according to disease severity (p = 002), being significantly higher in patients with critical as compared to moderate disease severity (p < 0.001). The HA concentration varied depending on the type of oxygen therapy (p = 0.004). It was significantly higher in patients on a ventilator than in those without oxygen therapy (p = 0.002). In patients who qualified for the steroid treatment and immunotherapy, the HA levels were significantly higher compared to those who did not qualify for such therapies (p = 0.043, p = 0.049, respectively). The HA levels were significantly higher in patients with cytokine storm compared to those without it (p < 0.001) and were significantly more elevated in non-survivors than in survivors (p < 0.001). HA had an excellent diagnostic power (AUC = 0.994) with sensitivity (83.3%) and specificity (97.8%) in identifying patients with critical disease severity and an excellent diagnostic power (AUC = 0.932) with sensitivity (88.2%) and specificity (95.6%) in identifying non-surviving patients. Conclusions: In summary, the results of our study indicate that HA is closely associated with severe SARS-CoV-2 infection and could be used as a novel serum biomarker to predict the risk of disease progression and as a predictor of COVID-19 mortality.