Objective: To evaluate the humoral immune response to the pre-exposure schedule of human rabies vaccination through intradermal and intramuscular routes, as well as the need for serological monitoring., Methods: A randomized and controlled intervention study was carried out in São Paulo, Southeastern Brazil, from 2004-2005. There were 149 volunteers, of which 127 completed the vaccination schedule (65 intradermal and 62 intramuscular) and underwent humoral immune response evaluation at ten, 90 and 180 days post-vaccination. Two outcomes were considered for comparing the two routes of administration: the geometric average of neutralizing antibody titers and the proportion of individuals with satisfactory titers (> 0.5 IU/mL) at each evaluation point. The association of the humoral immune response with anthropometric and demographic data was analyzed through a normal distribution test and a chi-square test with a Yates correction. After completion of the vaccination schedule, the proportion of seropositive results was compared by the Kruskall Wallis test, and the average titers were compared by variance analysis., Results: the average antibody titers were higher in patients who were vaccinated intramuscularly. The percentage of volunteers with satisfactory titers (> 0.5% IU/mL) decreased over time in both groups. However, in the group vaccinated intradermally the rate of satisfactory titers on day 180 ranged from 20% to 25%, while the intramuscular route varied from 63% to 65%. An association between the humoral immune response and the demographic and anthropometric variables was not observed., Conclusions: Serology after the third dose can be considered unnecessary in unexposed patients, since 97% and 100% of volunteers respectively vaccinated by the intradermal and intramuscular route presented satisfactory antibody levels (> 0.5% IU/mL).