This study aimed to quantify the animate source provided by the patients using the concept of "absolute carriage" by multiplying the carrier rate by the level of carriage; and to compare the impact of a low and high dose of an oropharyngeal vancomycin gel on the absolute MRSA carriage and infection. In all, 265 patients were included, 126 were MRSA positive. Fifty-five patients received 2% vancomycin gel during the first year whilst 4% vancomycin gel was given to 50 patients during the second year. Surveillance swabs of throat and rectum were obtained from all eligible patients on admission and then twice weekly. The vancomycin protocol was started as soon as the surveillance cultures were positive for MRSA. Those patients received one gram of enteral vancomycin daily, divided into four doses. During the first year 2% vancomycin gel 4 ml (80 mg) was applied in the oropharynx in four doses in addition to the enteral solution (Group A). During the second year 4% vancomycin gel 4 ml (160 mg) was used (Group B). The absolute carriage was high during both periods: 3.6 for Group A, and 3.2 for Group B. The 4% vancomycin protocol significantly reduced the absolute carriage, compared to the 2% vancomycin protocol: 2.6 versus 1.5 (P < 0.01). Significant reduction in secondary endogenous infections was found in the second year: seven versus 15 patients (P < 0.05). A total of 3,588 microbiological samples were processed. Neither Staphylococcus aureus with intermediate sensitivity to vancomycin (VISA) nor vancomycin-resistant enterococci (VRE) were detected.