The anti-Fim response and long-term persistence after vaccination and infection may be of importance in understanding population immunity. Longitudinal serum samples (n = 1330) from 542 non-infected children related to a Swedish vaccine trial showed that the post vaccination (DTPa5) antibody decay curve for pertussis ELISA IgG anti-fimbriae2/3 (anti-Fim2/3) was bi-phasic. A slower one followed an initial rapid decay approximately 5-6 months after the third dose at 12 months of age. After 71 months, however, 60% still had concentrations above > or =5 EU/ml, a level that had been shown to correlate with decreased risk of disease. Booster responses after re-vaccination with DTPa5 at 4, 5 and 6 years of age were strong and appeared within 1 week after vaccination, indicating immune memory. Ninety-six young children with verified pertussis infection, for whom we had serum samples both before, during and after the infection, showed a high response if they had been primed with fimbriae (either DTPa5 or DTPwc). In contrast, 76% of infected children not primed with fimbriae (a DTPa2 or DT group) only had concentrations below the minimum level of detection in all samples taken during and after the infection. In two Swedish seroepidemiological surveys, one from 1997 just after reintroduction of universal childhood vaccination against pertussis and one from 2007, the proportion of children 2-3 years with anti-Fim2/3 concentrations <5 EU/ml was similar and above 90%. This reflects that the two- or three-component pertussis vaccines (DTPa2 and DTPa3) that were introduced in Sweden in 1996 do not induce anti-Fim2/3 antibodies. In previous studies it was shown in multivariate analyses that levels of IgG anti-Fim2/3 > or =5 EU/ml reduced short-term risk of pertussis in small children. As the antibody response to Fim2/3 after infection is poor in children who have not been primed earlier in life, inclusion of immunogenic Fim2/3 in future pertussis vaccines should be considered.