Introduction Ozone therapy can be used as a monotherapy or as an adjunctive treatment to standard COVID-19 treatment protocols. Current evidence indicates that this approach may improve clinical outcomes, paraclinical markers, and reduce radiological signs of inflammation, with no side effects. Material and methods he study included 100 consecutive patients aged 18 and older with COVID-19, admitted to the Intensive Care Unit at the Institute of Emergency Medicine. Patients were randomly divided into two groups: 50 patients underwent treatment according to the National Clinical Protocol along with major ozonated autohemotherapy (the study group), while the other 50 patients were treated only according to the National Clinical Protocol (the control group). Results Although the initial oxygenation index (PaO2/FiO2) values were similar in both study groups, a dynamic analysis revealed a clear efficacy of ozone therapy. By the end of the first-week treatment, the mean oxygenation index in the ozone-treated group was significantly higher than in the standard treatment group: 296.8±105.1 mm Hg versus 232.8±110.6 mm Hg (p<0.01). The use of oxygen therapy (70.0% vs. 78.0%), non-invasive ventilation (70.0% vs. 76.0%), and invasive mechanical ventilation (22.0% vs. 38.0%) tended to be lower in the ozone group, though this difference was not statistically significant (p>0.05). Both treatment groups showed a significant clinical improvement, with 54.0% of COVID-19 patients in the ozone group and 50.0% in the conventional treatment group achieving a two-point reduction in clinical severity score (p>0.05). Conclusions The mean oxygenation index significantly increased in the study patient group (246.86±30.3 mm Hg on day 1 and 296.75±105.1 mm Hg on day 7 of treatment; p<0.01) and remained unchanged in the control group (235.86±33.4 mm Hg on day 1 and 232.82±110.6 mm Hg on day 7 of treatment; p>0.05). Although the mortality rate was lower among COVID-19 patients treated with ozone therapy (24.0%) compared to those receiving standard treatment (34.0%), this difference did not achieve any statistical significance.