Objectives. To assess the bacteriophage activity in localized and systemic infections caused by Staphylococcus aureus resistant to methicilin (MRSA). Materials and methods. An experimental study was performed in 45 mice of the Balb/c strain divided in nine groups of five individuals. Ten naive bacteriophages were isolated through clinical samples and hospital effluents. Lytic capacity and spectrum activity was evaluated on the basis of which six phages were selected for phagotherapy trials. Additionally, a commercial bacteriophage was used. The phagotherapy was evaluated through prophylaxis, and therapy of localized and systemic infections caused by MRSA by subcutaneous and intravenous inoculation, respectively. The effectiveness of three therapeutic schemes was tested: monotherapy, phage cocktail in multiple doses and phage cocktail in a single dose. The therapeutic activity of the phages was also compared with vancomycin and clindamycin. Results. The phage cocktail and the diverse dose therapy were effective in preventing and controlling MRSA localized infections; its activity was similar to the vancomycin and clindamycin activity. The single dose phage cocktail failed to control localized infection and phagotherapy was not effective in systemic infections. Conclusions. Phagotherapy could be a viable alternative for infections caused by MRSA. Further studies that assess related aspects to phages and patient safety are required.