From 1992 to 1997, 140 Thai children 4-15 years of age received an investigational live attenuated tetravalent dengue vaccine (LATDV). These children were contacted 3-8 years later in 2001 to assess humoral immunity and investigate whether they were subsequently at higher risk of developing severe dengue. One hundred thirteen were successfully contacted and participated in this retrospective cohort study with two age- and address-matched controls per vaccinee. The number of vaccinated subjects with neutralizing antibodies increased compared with 3-8 years earlier, which was probably due to subsequent wild-type dengue infections. There were no excess hospitalizations for clinically suspected dengue fever (DF) or dengue hemorrhagic fever (DHF) in vaccinees (one with DF and three with DHF) compared with controls (14 with DHF). Results suggest that preexisting dengue antibodies induced by LATDV do not enhance dengue illness, and the use of the vaccine in a dengue-endemic area is safe.