Aim. To evaluate the efficacy and safety of prolonged sequential therapy with mexidol in the acute and early recovery stages of hemispheric ischemic stroke (IS). Material and methods. A randomized double blind multicenter placebo-controlled, in parallel groups trial included 151 patients (62 men and 89 women) with hemispheric IS. Using a method of simple randomization, 150 patients (62 men and 88 women), aged 40—79 years, were randomized into two groups. Patients of Group I were treated with mexidol: 500 mg/day IV infusion for 10 days, followed by 125 mg tid (375 mg/day) PO for 8 weeks. Patients of Group II received the placebo according to the same scheme. The total duration of patients’ participation in trial ranged from 67 to 71 days. Results. By the end of treatment, the mean score on the modified Rankin scale (mRS) was lower in Group I compared to Group II (p=0.04). In Group I, the decrease in mRS mean score (Visit 1—5) was more prominent (p=0.023), percentage of patients with 0—2 scores by mRS scale (Visit 5) was higher (p=0.039), mean NIHSS score lower (p=0.035) in Visit 5 compared to group II. By the end of treatment, the decrease in mean NIHSS score in patients with diabetes mellitus was more prominent in Group I in comparison with Group II (p=0.038). In Group I, the dynamic of improvement of quality of life was more prominent and started from Visit 2 in general population and subpopulation of patients with diabetes mellitus. The share of patients with no problems with movement in space was higher in Group I (p=0.022). There were no statistically significant differences in frequency of side effects in patients of both groups. Conclusion. It is recommended to include mexidol in therapy of patients with IS in the acute and early rehabilitation stages.