- Email: [email protected]
Evaluation of the national immunisation programme in the Netherlands: immunity to diphtheria, tetanus, poliomyelitis, measles, mumps, rubella and Haemophilus influenzae type b
Vaccine 21 (2003) 716–720
Evaluation of the national immunisation programme in the Netherlands: immunity to diphtheria, tetanus, poliomyelitis, measles, mumps, rubella and Haemophilus influenzae type b夽 H.E. de Melker a,∗ , S. van den Hof a , G.A.M. Berbers b , M.A.E. Conyn-van Spaendonck a a
b
Department of Infectious Diseases Epidemiology, National Institute of Public Health and the Environment, P.O. Box 1, Bilthoven 3720 BA, The Netherlands Laboratory of Clinical Vaccine Research, National Institute of Public Health and the Environment, P.O. Box 1, Bilthoven 3720 BA, The Netherlands
Abstract The immunity to vaccine-preventable diseases included in the Dutch immunisation programme in the general population and among orthodox reformed individuals who refuse vaccination was assessed. The programme induces good protection. However, a large proportion of adults lacks diphtheria and tetanus immunity. Measles, mumps and rubella seroprevalence was somewhat lower among vaccinated compared to unvaccinated cohorts. The prevalence of HibPS antibodies declined during 2.5 years after the fourth vaccination. However, protection occurs also by memory immunity. Herd immunity is sufficient among the general population, but not among orthodox reformed individuals. Immunosurveillance is an efficient way to evaluate the effects of immunisation programmes and identify risk groups for infection. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Vaccine-preventable diseases; Immunisation programmes; Serumbank; Seroepidemiology
1. Introduction Post-licensing evaluation is of equal importance for the success of a vaccination programme as establishment of safety and efficacy of the vaccine in controlled trials before licensing of a vaccine. Epidemiological methods play an important role in this continuous evaluation to assess the safety of the vaccine, vaccination coverage, disease incidence, disease severity and the immunity of the population. With smaller number of cases as a result of mass vaccination with high coverage in The Netherlands, the role of serological surveillance as an epidemiological method becomes more evident. Mass vaccination can change the epidemiological dynamics of infectious diseases. It may result in a limited persistence of natural and vaccine-induced immunity and a higher mean age of infection that may lead to greater risk of complications. Assessment of specific antibodies offers the opportunity to identify groups with little immunity that might require changes in vaccination strategy to prevent (re)emergence of the disease. In order to estimate the immunity of the Dutch population reliably, a large-scale, population-based collection of serum samples was established in 1995–1996 (8359 sera in a na夽 ∗
Based on references [1–6]. Corresponding author. Tel.: +31-30-274-3958; fax: +31-30-274-4409.
tionwide sampling and 1589 sera from municipalities with low vaccine coverage) [7]. The sera were tested for all disease that were included in the national immunisation programme at that time (since September 2002 meningococcal C vaccination is included in the programme), i.e. diphtheria, tetanus, poliomyelitis, pertussis, Heamophilus influenzae type b, measles, mumps and rubella. This paper gives an overview of the immunity in the Dutch population for these vaccine-preventable diseases (with the exception of pertussis). It is based on disease-specific papers that contain more detailed results [1–6]. Since serological data on pertussis are more difficult to interpret regarding population protection they were excluded from this overview. Data on antibody distribution of pertussis toxin that were used in the improvement of serodiagnosis of pertussis are reported elsewhere [8].
2. Materials and Methods 2.1. Subjects and study population From October 1995 to December 1996, a population-based serumbank was established of 9948 individuals. Details of the study design have been published elsewhere [7,9]. Our objective was to select 40 municipalities with sampling
0264-410X/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 4 - 4 1 0 X ( 0 2 ) 0 0 5 8 7 - X
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
probabilities proportional to population size. To ensure that all geographical regions were presented, the Netherlands was first divided into five geographical regions of approximately equal population size. Within each of these five regions, eight municipalities were then sampled with probability proportional to size. Within each of these 40 municipalities an age-stratified sample of 380 individuals was drawn from the population register. The age strata were 0, 1–4, 5–9, . . . , 75–79 years. In each of the first two strata, 40 individuals were sampled while in each of the following strata, 20 individuals were sampled. Subjects were requested to give a blood sample, fill out a questionnaire and bring vaccination certificates. Similarly, individuals were selected from eight additional municipalities with low vaccine coverage to assess the immunity in geographically clustered orthodox reformed groups that refuse vaccination. The participation rates were 55 and 52.5% in the nationwide sample (Nnw = 8359) and the low vaccine coverage sample (Nlv = 1589), respectively. 2.2. Antibody assays Sera were stored at −86 ◦ C. Sera were tested with the Toxin Binding Inhibition assay for tetanus and diphtheria antibodies (Nnw = 7715, Nlv = 1419; protective titre ≥ 0.01 IU/ml), with the neutralization test for poliovirus types 1, 2 and 3 antibodies (Nnw = 7773, Nlv = 1501; titre ≥ 1:8), with an ELISA for antibodies against measles (Nnw = 8303, Nlv = 1582; ≥0.2 IU/ml), mumps (Nnw = 8298, Nlv = 1583; ≥45 IU/ml), rubella (Nnw = 8295, Nlv = 1582; ≥10 IU/ml) and the capsular polysaccharide of Haemophilus influenzae type b (HibPS) (Nnw = 7864, Nlv = not tested; >0.15 g/ml) [1–6]. 2.3. Statistical analysis Frequencies and geometric mean titres within each municipality were weighted by the proportion of the age group in
717
the population. To produce national estimates, the weighted frequencies and geometric mean titres were averaged over the 40 municipalities. For the low vaccine coverage sample, they were averaged weighting by the population size of the municipality. 3. Results 3.1. Diphtheria, tetanus and poliomyelitis 3.1.1. Nationwide sample and orthodox reformed individuals The overall seroprevalence for participants to the nationwide samples amounted to 88.1, 86.0, 96.6, 93.4 and 89.7% for diphtheria, tetanus, poliovirus types 1, 2 and 3, respectively. The age-specific prevalence for diphtheria, tetanus and poliomyelitis was high among those younger than 50 years, and decreased with older age for diphtheria and tetanus, but not for polio (Fig. 1). Both for persons younger and for persons of 50 years and above, the seroprevalence was lower among orthodox reformed individuals compared to the nationwide sample, i.e. the overall seroprevalence was 39.1, 62.6, 65.0, 59.0 and 68.7% for diphtheria, tetanus, poliovirus types 1, 2 and 3, respectively. 3.1.2. Persistence of antibodies after vaccination For diphtheria, tetanus and poliomyelitis, the GMTs for persons aged 10–34 years who were completely vaccinated according to the national immunisation programme (3, 4, 5 and 11 months, 4 and 9 years), the GMTs decreased with increasing age. For diphtheria the GMT amounted to 0.30 IU/ml (seroprevalence: 100%) for 10–14-year-olds compared to 0.09 IU/ml (96%) for 30–34-year-olds and for tetanus 1.38 IU/ml (100%) compared to 0.44 IU/ml (100%). The GMT in 2 logtiter for poliovirus type 1 was 10.5 (99.4%) for 10–14-year-olds compared to 8.0 (100%) for 30–34-year-olds. For the same agegroups, the GMT for poliovirus type 2 was 8.6 (99.4%) compared to 5.2 (95.4%)
Fig. 1. Age-specific seroprevalence for diphtheria, tetanus and poliomyelitis in the general population.
718
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
Fig. 2. Seroprevalence of measles, mumps and rubella in the general population.
and for poliovirus type 3 8.2 (98.6%) compared to 4.8 (87.4%). The GMTs for 20–34-year-olds with documented revaccination were similar to the GMT for 10–14-year-olds who were not revaccinated. 3.2. Measles, mumps and rubella 3.2.1. Nationwide sample and orthodox reformed individuals The age-specific seroprevalence for participants to the nationwide sample and among orthodox reformed individuals from the low vaccine coverage sample are given in Figs. 2 and 3, respectively. The overall seroprevalence for measles, mumps and rubella was 94–96% in the nationwide sample. For orthodox reformed individuals the seroprevalence was 2–4% lower. The difference was due to lower seroprevalence among 1–4-year-olds among the orthodox reformed individuals (Fig. 3). The seroprevalence among (mostly) vaccinated agegroups was slightly lower than among (mostly) natural infected agegroups; the difference was highest for measles (Fig. 2).
3.2.2. Persistence of antibodies after vaccination MMR vaccination is offered at ages 14 months and 9 years. GMT and seroprevalence decreased from 2.10 IU/ml (97.8%) at age 2 years to 0.81 IU/ml (92.4%) at age 8 years for measles, and from 75 (98.3%) to 30 IU/ml (96.3%) for rubella. For mumps, it increased to 157 RU/ml (93.2%) at 3 years, and subsequently decreased to 112 RU/ml (88.9%) at 7 years. The effect of the second vaccination on GMT and seroprevalence was already observed at age 8; when only 15% had received their second vaccination. 3.3. Haemophilus influenzae type b 3.3.1. Unvaccinated cohorts The GMT and seroprevalence declined with increasing age in the adult groups from 1.46 g/ml (94%) for 20–24-year-olds to 0.73 g/ml (83.7%) for 75–79-year-olds (Fig. 4). Among 3–7-year-olds the GMT and prevalence of antibodies were both lower than they were among children who were younger or older.
Fig. 3. Seroprevalence of measles, mumps and rubella among orthodox reformed individuals.
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
719
Fig. 4. Seroprevalence of HibPS in the general population. Note the different scale compared to Figs. 1–3.
3.3.2. Cohorts eligible for vaccination Longitudinal interpretation of the data for participants born after 1 April 1993 (eligible for vaccination) showed that both GMT and seroprevalence increased from 0.1 g/ml (43%) at 2 months to 4.1 g/ml (92%) at 6 months, and decreased to 1.4 g/ml (89%) at 11 months and increased to 7.6 g/ml (96%) in those aged 12–17 months. Afterwards it decreased from 1.7 g/ml (100%) for 24–29 months to 0.9 g/ml (82%) among 36–43-month-olds.
have stabilized at 0.8 g/ml. It was not possible to predict antibody levels for the long term after vaccination because of the relative newness of the conjugate Hib vaccine. Since conjugated vaccines induce a T-cell dependent immune response low or undetected antibody levels do not necessarily indicate a lack of protection against invasive Hib infections. The low incidence of Hib infections as well as the low number of Hib indicate good protection at present [10]. The most important question is how long memory immunity will persist after vaccination without re-exposure [6].
4. Discussion
4.2. Herd immunity
4.1. Effect of vaccination
4.2.1. General population While adults are well protected against poliomyelitis, measles, mumps, rubella and Haemophilus influenzae type b great number of adults lack diphtheria or tetanus antibodies. Since tetanus is not communicable, herd immunity concept does not apply. The decrease in tetanus antibodies from 50 years of age onwards is associated with higher risk at tetanus infections. Surveillance data show that most tetanus patients occur among those born before routine vaccination, i.e. before 1945. Therefore, offering a primary tetanus vaccination to cohorts born before introduction of vaccination would probably more effective in preventing tetanus than routine revaccination. Despite the lack of diphtheria antibodies in many adults as a result of waning immunity herd immunity seems sufficient. No diphtheria cases occurred in The Netherlands related to the diphtheria epidemic in the former Soviet Union. Studying whether adults with low or undetectable antibody levels show an adequate (memory) response after (re)vaccination would be relevant to obtain insight into the protection of these adults.
The immunisation programme induces (very) good protection against the target diseases diphtheria, tetanus, poliomyelitis, mumps, measles, rubella and Haemophilus influenzae type b. Vaccination against diphtheria, tetanus and poliomyelitis was introduced several decades ago. The seroprevalence studies showed that despite a slight decrease in geometric mean titres for these diseases, vaccination according to the national immunisation programme has resulted in long-term protection [1–6]. Seroprevalence of measles, mumps, and rubella was higher in those age groups that had received two vaccinations than in those under the age of 9 years, but due to the fact that only a few birth cohorts had two doses at the time of sampling (mumps and rubella) or they were not comparable with respect to exposure to wild virus (measles), the persistence of antibodies after the second dose could not be studied. We observed a sharp decline in HibPS antibody titer within 6 months after infants had received the fourth vaccination as schedules. After 2–2.5 years, the GMT seemed to
720
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
Epidemiological data of the polio outbreaks in The Netherlands among orthodox reformed individuals without spread to general population confirm the existence of adequate herd immunity in the general population. This is reassuring considering the global initiative for polio eradication. The seroprevalence for measles, mumps and rubella in vaccinated cohorts—particularly those with only one dose— was lower than in the unvaccinated, natural immune cohorts. While for mumps and rubella the prevalence was still above the level needed to prevent epidemics, for measles the prevalence was borderline sufficient for elimination. A recent measles epidemic (1999–2000) that started among orthodox reformed individuals hardly led to cases among the general population [11]. This shows that in contrast to orthodox reformed individuals, the herd immunity among the general population is sufficient for measles to prevent circulation. The prevalence of HibPS antibodies is high in unvaccinated adult cohorts, but it did show a decrease with increasing age. A key question would be whether Hib and cross-reacting organism will circulate sufficiently enough to provide periodic natural re-exposure to maintain the present existence of herd immunity. 4.2.2. Orthodox reformed individuals As a result of refusing vaccination by sociogeographlically clustered orthodox reformed groups spreading of various diseases included in the national immunisation programme constitute a danger of spreading. Examples are the measles epidemic in 1999–2000 and poliomyelitis outbreak in 1992–1993 [11,12]. Furthermore, a large part of the orthodox reformed children will reach adult age without mumps or rubella immunity. Future introduction of rubella virus in these groups could increase the number of infants with congenital rubella syndrome. This phenomena has already been observed among Amish in United States who refuse vaccination and in Greece were vaccine coverage was below 50% [13,14]. 5. Conclusion The seroprevalence studies confirm the great success of the vaccination programme in The Netherlands. The challenge is, first of all, to preserve and enhance this progress by sustaining high acceptance of vaccination in childhood. Furthermore, our studies show that we have to anticipate some long-term effects of mass vaccination. As a result of decreased circulation of the pathogens, gaps in immunity among orthodox reformed individuals were and will be observed. Waning of immunity—as already observed for diphtheria in adults—as a result of lack of boosting has to be anticipated. Finally, we showed that seroepidemiology plays an important role in evaluation of a vaccination programme. Assessment of the immunity in the population will have to continue to evaluate the (long-term) epidemiological effects of the existing vaccines and new vaccines.
Acknowledgements We acknowledge the Public Health Services, the Pienterproject Team, C.J.P. van Limpt, H.A.T. Kuijken, D.R. Jut, T.A.M. Antonolio and A. Schakelaar for their contribution to the realization of the Pienter-project and N.J.D. Nagelkerke, H.C. Rümke, F. Abbink, T.G. Kimman, A.M. van Loon, M.T.A. Beaumont, N. Elzinga-Gholizadea, P.H. van der Kraak and R. de Haas for their contribution to (one of) the original papers. References [1] de Melker HE, Berbers GAM, Nagelkerke NJD, Conyn-van Spaendonck MAE. Diphtheria antitoxin levels in The Netherlands: a population-based study. Emerg Infect Dis 1999;5:694–700. [2] de Melker HE, van de Hof S, Berbers GAM, Nagelkerke NJD, Rümke HC, Conyn-van Spaendonck MAE. A population-based study on tetanus immunity in The Netherlands. Vaccine 1999;18:100–8. [3] Conyn-van Spaendonck MAE, de Melker HE, Abbink F, ElzingaGholizadea N, Kimman TG, van Loon AM. Immunity against poliomyelitis in The Netherlands. Am J Epidemiol 2001;153:207– 14. [4] van den Hof S, Berbers GAM, de Melker HE, Conyn-van Spaendonck MAE. Seroepidemiology of measles antibodies in The Netherlands, a cross-sectional study in a national sample and in municipalities with low vaccine coverage. Vaccine 1999;18:931–40. [5] de Haas R, van den Hof S, Berbers GAM, de Melker HE, Conyn-van Spaendonck MAE. Prevalence of antibodies against rubella virus in The Netherlands 9 years after changing from selective to mass vaccination. Epidemiol Infect 1999;123:263–70. [6] van den Hof S, de Melker HE, Berbers GAM, van der Kraak PH, Conyn-van Spaendonck MAE. Antibodies to Haemophilus influenzae type b a few years after the introduction of routine vaccination. Clin Infect Dis 2001;32:2–8. [7] de Melker HE, Conyn-van Spaendonck MAE. Immunosurveillance and the evaluation of national immunisation programmes: a population-based approach. Epidemiol Infect 1998;121:637–43. [8] de Melker HE, Versteegh FGA, Conyn-van Spaendonck MAE, Elvers LH, Berbers GAM, van der Zee A, et al. Specificity and sensitivity of high levels of IgG antibodies against pertussis toxin in a single serum for diagnosis of infection with Bordetella pertussis. J Clin Microbiol 2000;38:800–6. [9] de Melker HE, Nagelkerke NJD, Conyn-van Spaendonck MAE. Non-participation in a population-based seroprevalence study of vaccine-preventable diseases. Epidemiol Infect 2000;124:255–62. [10] Breukels MA, Spanjaard L, Sanders LA, Rijkers GT. Immunological characterization of conjugated Haemophilus influenzae type b vaccine failure in infants. Clin Infect Dis 2001;32:1700–5. [11] van den Hof S, Meffre CME, Conyn-van Spaendonck MAE, Woonink F, de Melker HE, van Binnendijk RS. Measles outbreak in a community with very low vaccine coverage, The Netherlands. Emerg Infect Dis 2001;6:593–7. [12] Oostvogel PM, van Wijngaarden JK, van der Avoort HGAM, Mulders MN, Conyn-van Spaendonck MAE, Rümke HC, et al. Poliomyelitis outbreak in an unvaccinated community in The Netherlands, 1992–1993. Lancet 1994;344:665–70. [13] Mellinger AK, Cragan JD, Atkinson W, Williams WW, Kleger B, Kimber RG, et al. High incidence of congenital rubella syndrome after a rubella outbreak. Pediatr Infect Dis J 1995;14:573–8. [14] Panagiotopoulos T, Antoniadou I, Valassi-Adam E. Increase in congenital rubella occurrence after immunisation in Greece: retrospective survey and systematic review. BMJ 1999;319:1462–7.
Evaluation of the national immunisation programme in the Netherlands: immunity to diphtheria, tetanus, poliomyelitis, measles, mumps, rubella and Haemophilus influenzae type b夽 H.E. de Melker a,∗ , S. van den Hof a , G.A.M. Berbers b , M.A.E. Conyn-van Spaendonck a a
b
Department of Infectious Diseases Epidemiology, National Institute of Public Health and the Environment, P.O. Box 1, Bilthoven 3720 BA, The Netherlands Laboratory of Clinical Vaccine Research, National Institute of Public Health and the Environment, P.O. Box 1, Bilthoven 3720 BA, The Netherlands
Abstract The immunity to vaccine-preventable diseases included in the Dutch immunisation programme in the general population and among orthodox reformed individuals who refuse vaccination was assessed. The programme induces good protection. However, a large proportion of adults lacks diphtheria and tetanus immunity. Measles, mumps and rubella seroprevalence was somewhat lower among vaccinated compared to unvaccinated cohorts. The prevalence of HibPS antibodies declined during 2.5 years after the fourth vaccination. However, protection occurs also by memory immunity. Herd immunity is sufficient among the general population, but not among orthodox reformed individuals. Immunosurveillance is an efficient way to evaluate the effects of immunisation programmes and identify risk groups for infection. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Vaccine-preventable diseases; Immunisation programmes; Serumbank; Seroepidemiology
1. Introduction Post-licensing evaluation is of equal importance for the success of a vaccination programme as establishment of safety and efficacy of the vaccine in controlled trials before licensing of a vaccine. Epidemiological methods play an important role in this continuous evaluation to assess the safety of the vaccine, vaccination coverage, disease incidence, disease severity and the immunity of the population. With smaller number of cases as a result of mass vaccination with high coverage in The Netherlands, the role of serological surveillance as an epidemiological method becomes more evident. Mass vaccination can change the epidemiological dynamics of infectious diseases. It may result in a limited persistence of natural and vaccine-induced immunity and a higher mean age of infection that may lead to greater risk of complications. Assessment of specific antibodies offers the opportunity to identify groups with little immunity that might require changes in vaccination strategy to prevent (re)emergence of the disease. In order to estimate the immunity of the Dutch population reliably, a large-scale, population-based collection of serum samples was established in 1995–1996 (8359 sera in a na夽 ∗
Based on references [1–6]. Corresponding author. Tel.: +31-30-274-3958; fax: +31-30-274-4409.
tionwide sampling and 1589 sera from municipalities with low vaccine coverage) [7]. The sera were tested for all disease that were included in the national immunisation programme at that time (since September 2002 meningococcal C vaccination is included in the programme), i.e. diphtheria, tetanus, poliomyelitis, pertussis, Heamophilus influenzae type b, measles, mumps and rubella. This paper gives an overview of the immunity in the Dutch population for these vaccine-preventable diseases (with the exception of pertussis). It is based on disease-specific papers that contain more detailed results [1–6]. Since serological data on pertussis are more difficult to interpret regarding population protection they were excluded from this overview. Data on antibody distribution of pertussis toxin that were used in the improvement of serodiagnosis of pertussis are reported elsewhere [8].
2. Materials and Methods 2.1. Subjects and study population From October 1995 to December 1996, a population-based serumbank was established of 9948 individuals. Details of the study design have been published elsewhere [7,9]. Our objective was to select 40 municipalities with sampling
0264-410X/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 4 - 4 1 0 X ( 0 2 ) 0 0 5 8 7 - X
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
probabilities proportional to population size. To ensure that all geographical regions were presented, the Netherlands was first divided into five geographical regions of approximately equal population size. Within each of these five regions, eight municipalities were then sampled with probability proportional to size. Within each of these 40 municipalities an age-stratified sample of 380 individuals was drawn from the population register. The age strata were 0, 1–4, 5–9, . . . , 75–79 years. In each of the first two strata, 40 individuals were sampled while in each of the following strata, 20 individuals were sampled. Subjects were requested to give a blood sample, fill out a questionnaire and bring vaccination certificates. Similarly, individuals were selected from eight additional municipalities with low vaccine coverage to assess the immunity in geographically clustered orthodox reformed groups that refuse vaccination. The participation rates were 55 and 52.5% in the nationwide sample (Nnw = 8359) and the low vaccine coverage sample (Nlv = 1589), respectively. 2.2. Antibody assays Sera were stored at −86 ◦ C. Sera were tested with the Toxin Binding Inhibition assay for tetanus and diphtheria antibodies (Nnw = 7715, Nlv = 1419; protective titre ≥ 0.01 IU/ml), with the neutralization test for poliovirus types 1, 2 and 3 antibodies (Nnw = 7773, Nlv = 1501; titre ≥ 1:8), with an ELISA for antibodies against measles (Nnw = 8303, Nlv = 1582; ≥0.2 IU/ml), mumps (Nnw = 8298, Nlv = 1583; ≥45 IU/ml), rubella (Nnw = 8295, Nlv = 1582; ≥10 IU/ml) and the capsular polysaccharide of Haemophilus influenzae type b (HibPS) (Nnw = 7864, Nlv = not tested; >0.15 g/ml) [1–6]. 2.3. Statistical analysis Frequencies and geometric mean titres within each municipality were weighted by the proportion of the age group in
717
the population. To produce national estimates, the weighted frequencies and geometric mean titres were averaged over the 40 municipalities. For the low vaccine coverage sample, they were averaged weighting by the population size of the municipality. 3. Results 3.1. Diphtheria, tetanus and poliomyelitis 3.1.1. Nationwide sample and orthodox reformed individuals The overall seroprevalence for participants to the nationwide samples amounted to 88.1, 86.0, 96.6, 93.4 and 89.7% for diphtheria, tetanus, poliovirus types 1, 2 and 3, respectively. The age-specific prevalence for diphtheria, tetanus and poliomyelitis was high among those younger than 50 years, and decreased with older age for diphtheria and tetanus, but not for polio (Fig. 1). Both for persons younger and for persons of 50 years and above, the seroprevalence was lower among orthodox reformed individuals compared to the nationwide sample, i.e. the overall seroprevalence was 39.1, 62.6, 65.0, 59.0 and 68.7% for diphtheria, tetanus, poliovirus types 1, 2 and 3, respectively. 3.1.2. Persistence of antibodies after vaccination For diphtheria, tetanus and poliomyelitis, the GMTs for persons aged 10–34 years who were completely vaccinated according to the national immunisation programme (3, 4, 5 and 11 months, 4 and 9 years), the GMTs decreased with increasing age. For diphtheria the GMT amounted to 0.30 IU/ml (seroprevalence: 100%) for 10–14-year-olds compared to 0.09 IU/ml (96%) for 30–34-year-olds and for tetanus 1.38 IU/ml (100%) compared to 0.44 IU/ml (100%). The GMT in 2 logtiter for poliovirus type 1 was 10.5 (99.4%) for 10–14-year-olds compared to 8.0 (100%) for 30–34-year-olds. For the same agegroups, the GMT for poliovirus type 2 was 8.6 (99.4%) compared to 5.2 (95.4%)
Fig. 1. Age-specific seroprevalence for diphtheria, tetanus and poliomyelitis in the general population.
718
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
Fig. 2. Seroprevalence of measles, mumps and rubella in the general population.
and for poliovirus type 3 8.2 (98.6%) compared to 4.8 (87.4%). The GMTs for 20–34-year-olds with documented revaccination were similar to the GMT for 10–14-year-olds who were not revaccinated. 3.2. Measles, mumps and rubella 3.2.1. Nationwide sample and orthodox reformed individuals The age-specific seroprevalence for participants to the nationwide sample and among orthodox reformed individuals from the low vaccine coverage sample are given in Figs. 2 and 3, respectively. The overall seroprevalence for measles, mumps and rubella was 94–96% in the nationwide sample. For orthodox reformed individuals the seroprevalence was 2–4% lower. The difference was due to lower seroprevalence among 1–4-year-olds among the orthodox reformed individuals (Fig. 3). The seroprevalence among (mostly) vaccinated agegroups was slightly lower than among (mostly) natural infected agegroups; the difference was highest for measles (Fig. 2).
3.2.2. Persistence of antibodies after vaccination MMR vaccination is offered at ages 14 months and 9 years. GMT and seroprevalence decreased from 2.10 IU/ml (97.8%) at age 2 years to 0.81 IU/ml (92.4%) at age 8 years for measles, and from 75 (98.3%) to 30 IU/ml (96.3%) for rubella. For mumps, it increased to 157 RU/ml (93.2%) at 3 years, and subsequently decreased to 112 RU/ml (88.9%) at 7 years. The effect of the second vaccination on GMT and seroprevalence was already observed at age 8; when only 15% had received their second vaccination. 3.3. Haemophilus influenzae type b 3.3.1. Unvaccinated cohorts The GMT and seroprevalence declined with increasing age in the adult groups from 1.46 g/ml (94%) for 20–24-year-olds to 0.73 g/ml (83.7%) for 75–79-year-olds (Fig. 4). Among 3–7-year-olds the GMT and prevalence of antibodies were both lower than they were among children who were younger or older.
Fig. 3. Seroprevalence of measles, mumps and rubella among orthodox reformed individuals.
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
719
Fig. 4. Seroprevalence of HibPS in the general population. Note the different scale compared to Figs. 1–3.
3.3.2. Cohorts eligible for vaccination Longitudinal interpretation of the data for participants born after 1 April 1993 (eligible for vaccination) showed that both GMT and seroprevalence increased from 0.1 g/ml (43%) at 2 months to 4.1 g/ml (92%) at 6 months, and decreased to 1.4 g/ml (89%) at 11 months and increased to 7.6 g/ml (96%) in those aged 12–17 months. Afterwards it decreased from 1.7 g/ml (100%) for 24–29 months to 0.9 g/ml (82%) among 36–43-month-olds.
have stabilized at 0.8 g/ml. It was not possible to predict antibody levels for the long term after vaccination because of the relative newness of the conjugate Hib vaccine. Since conjugated vaccines induce a T-cell dependent immune response low or undetected antibody levels do not necessarily indicate a lack of protection against invasive Hib infections. The low incidence of Hib infections as well as the low number of Hib indicate good protection at present [10]. The most important question is how long memory immunity will persist after vaccination without re-exposure [6].
4. Discussion
4.2. Herd immunity
4.1. Effect of vaccination
4.2.1. General population While adults are well protected against poliomyelitis, measles, mumps, rubella and Haemophilus influenzae type b great number of adults lack diphtheria or tetanus antibodies. Since tetanus is not communicable, herd immunity concept does not apply. The decrease in tetanus antibodies from 50 years of age onwards is associated with higher risk at tetanus infections. Surveillance data show that most tetanus patients occur among those born before routine vaccination, i.e. before 1945. Therefore, offering a primary tetanus vaccination to cohorts born before introduction of vaccination would probably more effective in preventing tetanus than routine revaccination. Despite the lack of diphtheria antibodies in many adults as a result of waning immunity herd immunity seems sufficient. No diphtheria cases occurred in The Netherlands related to the diphtheria epidemic in the former Soviet Union. Studying whether adults with low or undetectable antibody levels show an adequate (memory) response after (re)vaccination would be relevant to obtain insight into the protection of these adults.
The immunisation programme induces (very) good protection against the target diseases diphtheria, tetanus, poliomyelitis, mumps, measles, rubella and Haemophilus influenzae type b. Vaccination against diphtheria, tetanus and poliomyelitis was introduced several decades ago. The seroprevalence studies showed that despite a slight decrease in geometric mean titres for these diseases, vaccination according to the national immunisation programme has resulted in long-term protection [1–6]. Seroprevalence of measles, mumps, and rubella was higher in those age groups that had received two vaccinations than in those under the age of 9 years, but due to the fact that only a few birth cohorts had two doses at the time of sampling (mumps and rubella) or they were not comparable with respect to exposure to wild virus (measles), the persistence of antibodies after the second dose could not be studied. We observed a sharp decline in HibPS antibody titer within 6 months after infants had received the fourth vaccination as schedules. After 2–2.5 years, the GMT seemed to
720
H.E. de Melker et al. / Vaccine 21 (2003) 716–720
Epidemiological data of the polio outbreaks in The Netherlands among orthodox reformed individuals without spread to general population confirm the existence of adequate herd immunity in the general population. This is reassuring considering the global initiative for polio eradication. The seroprevalence for measles, mumps and rubella in vaccinated cohorts—particularly those with only one dose— was lower than in the unvaccinated, natural immune cohorts. While for mumps and rubella the prevalence was still above the level needed to prevent epidemics, for measles the prevalence was borderline sufficient for elimination. A recent measles epidemic (1999–2000) that started among orthodox reformed individuals hardly led to cases among the general population [11]. This shows that in contrast to orthodox reformed individuals, the herd immunity among the general population is sufficient for measles to prevent circulation. The prevalence of HibPS antibodies is high in unvaccinated adult cohorts, but it did show a decrease with increasing age. A key question would be whether Hib and cross-reacting organism will circulate sufficiently enough to provide periodic natural re-exposure to maintain the present existence of herd immunity. 4.2.2. Orthodox reformed individuals As a result of refusing vaccination by sociogeographlically clustered orthodox reformed groups spreading of various diseases included in the national immunisation programme constitute a danger of spreading. Examples are the measles epidemic in 1999–2000 and poliomyelitis outbreak in 1992–1993 [11,12]. Furthermore, a large part of the orthodox reformed children will reach adult age without mumps or rubella immunity. Future introduction of rubella virus in these groups could increase the number of infants with congenital rubella syndrome. This phenomena has already been observed among Amish in United States who refuse vaccination and in Greece were vaccine coverage was below 50% [13,14]. 5. Conclusion The seroprevalence studies confirm the great success of the vaccination programme in The Netherlands. The challenge is, first of all, to preserve and enhance this progress by sustaining high acceptance of vaccination in childhood. Furthermore, our studies show that we have to anticipate some long-term effects of mass vaccination. As a result of decreased circulation of the pathogens, gaps in immunity among orthodox reformed individuals were and will be observed. Waning of immunity—as already observed for diphtheria in adults—as a result of lack of boosting has to be anticipated. Finally, we showed that seroepidemiology plays an important role in evaluation of a vaccination programme. Assessment of the immunity in the population will have to continue to evaluate the (long-term) epidemiological effects of the existing vaccines and new vaccines.
Acknowledgements We acknowledge the Public Health Services, the Pienterproject Team, C.J.P. van Limpt, H.A.T. Kuijken, D.R. Jut, T.A.M. Antonolio and A. Schakelaar for their contribution to the realization of the Pienter-project and N.J.D. Nagelkerke, H.C. Rümke, F. Abbink, T.G. Kimman, A.M. van Loon, M.T.A. Beaumont, N. Elzinga-Gholizadea, P.H. van der Kraak and R. de Haas for their contribution to (one of) the original papers. References [1] de Melker HE, Berbers GAM, Nagelkerke NJD, Conyn-van Spaendonck MAE. Diphtheria antitoxin levels in The Netherlands: a population-based study. Emerg Infect Dis 1999;5:694–700. [2] de Melker HE, van de Hof S, Berbers GAM, Nagelkerke NJD, Rümke HC, Conyn-van Spaendonck MAE. A population-based study on tetanus immunity in The Netherlands. Vaccine 1999;18:100–8. [3] Conyn-van Spaendonck MAE, de Melker HE, Abbink F, ElzingaGholizadea N, Kimman TG, van Loon AM. Immunity against poliomyelitis in The Netherlands. Am J Epidemiol 2001;153:207– 14. [4] van den Hof S, Berbers GAM, de Melker HE, Conyn-van Spaendonck MAE. Seroepidemiology of measles antibodies in The Netherlands, a cross-sectional study in a national sample and in municipalities with low vaccine coverage. Vaccine 1999;18:931–40. [5] de Haas R, van den Hof S, Berbers GAM, de Melker HE, Conyn-van Spaendonck MAE. Prevalence of antibodies against rubella virus in The Netherlands 9 years after changing from selective to mass vaccination. Epidemiol Infect 1999;123:263–70. [6] van den Hof S, de Melker HE, Berbers GAM, van der Kraak PH, Conyn-van Spaendonck MAE. Antibodies to Haemophilus influenzae type b a few years after the introduction of routine vaccination. Clin Infect Dis 2001;32:2–8. [7] de Melker HE, Conyn-van Spaendonck MAE. Immunosurveillance and the evaluation of national immunisation programmes: a population-based approach. Epidemiol Infect 1998;121:637–43. [8] de Melker HE, Versteegh FGA, Conyn-van Spaendonck MAE, Elvers LH, Berbers GAM, van der Zee A, et al. Specificity and sensitivity of high levels of IgG antibodies against pertussis toxin in a single serum for diagnosis of infection with Bordetella pertussis. J Clin Microbiol 2000;38:800–6. [9] de Melker HE, Nagelkerke NJD, Conyn-van Spaendonck MAE. Non-participation in a population-based seroprevalence study of vaccine-preventable diseases. Epidemiol Infect 2000;124:255–62. [10] Breukels MA, Spanjaard L, Sanders LA, Rijkers GT. Immunological characterization of conjugated Haemophilus influenzae type b vaccine failure in infants. Clin Infect Dis 2001;32:1700–5. [11] van den Hof S, Meffre CME, Conyn-van Spaendonck MAE, Woonink F, de Melker HE, van Binnendijk RS. Measles outbreak in a community with very low vaccine coverage, The Netherlands. Emerg Infect Dis 2001;6:593–7. [12] Oostvogel PM, van Wijngaarden JK, van der Avoort HGAM, Mulders MN, Conyn-van Spaendonck MAE, Rümke HC, et al. Poliomyelitis outbreak in an unvaccinated community in The Netherlands, 1992–1993. Lancet 1994;344:665–70. [13] Mellinger AK, Cragan JD, Atkinson W, Williams WW, Kleger B, Kimber RG, et al. High incidence of congenital rubella syndrome after a rubella outbreak. Pediatr Infect Dis J 1995;14:573–8. [14] Panagiotopoulos T, Antoniadou I, Valassi-Adam E. Increase in congenital rubella occurrence after immunisation in Greece: retrospective survey and systematic review. BMJ 1999;319:1462–7.