Since 1990, Japanese encephalitis (JE) vaccine has been part of EPI in northern Thailand, where there is a high prevalence of JE and HIV infection. To evaluate the immunogenicity and safety of JE vaccine among HIV-infected children, we conducted a retrospective study of HIV-infected and uninfected children who received 2 doses of JE vaccine at 12 months of age. Pre- and post-immunization plasma specimens were tested by plaque reduction neutralization for antibody levels to JE and dengue(1-4) viruses; titers of > or =10 were considered positive. Excluding 5 children with preimmunization antibodies, 5 of 14 (36%) HIV-infected children and 18 of 27 (67%) uninfected children had positive JE antibody titers after immunization [odds ratio (OR) 0.3, p=0.06]; 31% absolute difference [95% confidence interval (CI) 0-61.7%). The geometric mean titer of HIV-infected children with positive titers was lower than that of control children (15.1 vs, 23.8; p=0.17). No significant vaccine-associated adverse events were noted. We conclude that primary antibody response to JE vaccine was low among HIV-infected children and was approximately half of that seen among uninfected children. In endemic areas, HIV-infected children are likely to be at risk of acquiring JE despite routine immunization with 2 doses.