In an attempt to evaluate the long-term immunogenicity and efficacy of plasma-derived hepatitis B vaccine in preventing hepatitis B virus infection, 199 infants born to hepatitis B e antigen-positive hepatitis B surface antigen-carrier mothers were found to be antibody to HBsAg-positive (greater than or equal to 10 mIU per ml) 2 months after the first booster of hepatitis B vaccination at age 1, and their serum HBsAg and anti-HBs were rechecked annually to ages 3 to 5. Of the nine infants whose initial anti-HBs were low (10 to 100 mIU per ml) in concentration, four (44%) were found to be anti-HBs seronegative at age 3, while none of the 127 vaccine responders with high anti-HBs levels (greater than 1,000 mIU per ml) lost their anti-HBs during the 4-year follow-up period. Also, in 63 infants whose initial anti-HBs titers were around 101 to 1,000 mIU per ml, only two lost their anti-HBs at age 4, and another two at age 5, respectively. Whether the vaccine responders lost their anti-HBs or not, no hepatitis B virus infection occurred in these vaccinees during the follow-up period. Thus, in the first 5 years of life, the protective efficacy in the high-risk infants who responded to plasma-derived hepatitis B vaccine was 100%. Because of the diversity of anti-HBs response in individuals, we suggest testing anti-HBs titer in all vaccinated infants after the first booster vaccination in order to calculate the time of next booster before the minimal protective level is reached.