Persistence of anti-HBs 5 years after the introduction of routine infant and adolescent vaccination in Italy

Vaccine 19 (2001) 2812– 2818 www.elsevier.com/locate/vaccine

Persistence of anti-HBs 5 years after the introduction of routine infant and adolescent vaccination in Italy A. Faustini a,*, E. Franco b, M. Sangalli a, T. Spadea a, R.M. Calabrese b, M. Cauletti b, C.A. Perucci a a

Department of Epidemiology, Lazio Region Health Authority, Via S. Costanza 53, 00198 Rome, Italy Department of Public Health, Tor Vergata Uni6ersity, Via Orazio Raimondo 18, 00173 Rome, Italy

b

Received 14 June 2000; received in revised form 30 November 2000; accepted 19 December 2000

Abstract A population survey was conducted to assess the duration of anti-HBs levels \ 10 IU/l in vaccinees living in Lazio Region (Italy) 5 years after the introduction (15 June 1991) of compulsory vaccination of new-borns and 11-year-old children. A random sample of 1192 (533 children born in 1991–92 and 659 adolescents born in 1979– 81) was selected. In 92.9% of children and 94.1% of adolescents anti-HBs titres were protective ( ]10 IU/l). These subjects with protective titres were divided into three categories: low responders (anti-HBs titres =10–500 IU/l), medium responders (anti-HBs titres = 501– 2000 IU/l) and high responders (anti-HBs titres \2000 IU/l). Factors associated with the level of response were analysed, using a multiple politomic logistic regression analysis. Greater age at first dose (11–12 years) was associated with higher titres (OR = 2.1, 95% CI =1.4–3.2 for medium responders and OR=3.0, 95% CI = 1.9–4.8 for high responders). Simultaneous administration of DT vaccine was associated with lower titres (OR =0.4, 95% CI = 0.2– 0.8 for medium responders and OR = 0.3, 95% CI = 0.1– 0.7 for high responders). © 2001 Elsevier Science Ltd. All rights reserved. Keywords: Long-term anti-HBs levels; Hepatitis B vaccine; Population survey

1. Introduction The long-term immunogenicity of vaccine against hepatitis B virus (HBV) is supported by many studies of up to 5 years [1–4] and over [5 – 7] in at-risk populations. Long-term persistence of anti-HBs has been studied as well in healthy adult volunteers [8 – 10] and in groups of healthy children [11,12]. No evidence is available about long-term immunogenicity in the general population after mass vaccination in a low-prevalence population. Persistence of antibodies against HBV surface antigen (anti-HBs) has been proved to be dependent on the level of anti-HBs concentration after basal immunisation, so that anti-HBs titres after HBV vaccination have made it possible to predict the duration of protection [3,8,9,13]. Other factors influencing the amount and the persistence of anti-HBs were identified

as age at vaccination, mother’s HBsAg positivity, certain diseases, such as renal failure necessitating dialysis, tumours, AIDS and blood and metabolic disorders, and low birthweight. The degree of protection closely parallels the persistence of anti-HBs, although low or undetectable levels of circulating anti-HBs are not necessarily indicative of loss of protection [2,5,14,15]. This study evaluates the persistence of anti-HBs for up to 5 years in new-born and child populations in the Lazio region (Italy) after the introduction in 1991 of mass vaccination. The factors related to the persistence of anti-HBs were analysed and the possibility of predicting the further duration of protection against hepatitis B in vaccine responders was considered.

2. Materials and methods

2.1. Subjects * Corresponding author. Tel.: + 39-6-51686486; fax: +39-651686463. E-mail address: [email protected] (A. Faustini).

Vaccination of new-born infants and 11-year-old children was introduced in the region on 15 June 1991

0264-410X/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 4 - 4 1 0 X ( 0 1 ) 0 0 0 0 5 - 6

A. Faustini et al. / Vaccine 19 (2001) 2812–2818

in accordance with a governmental ordinance [16], which provided for a vaccination schedule for new-borns of three paediatric doses (5 or 10 mg) of recombinant hepatitis B vaccine at 2 – 3, 4 – 5 and 11–12 months of age [17]. For 11-year-old children the protocol provided for three adult doses (10 or 20 mg) of recombinant HB vaccine at 0, 1 and 6 months. The vaccine was to be administered i.m. in the anterolateral side of the quadriceps of the new-born children, and in the deltoid muscle of the children [18]. The study population consists of two cohorts: subjects born between 15 June 1991 and 30 September 1992 who received their first dose of hepatitis B vaccine by 31 December 1992, and subjects born between 15 June 1979 and 14 June 1981 who started a course of vaccination by 31 December 1992. Data were collected in eight of the 51 Local Health Units geographically representative of the Lazio region. Demographic data about the 5–6-year-old children and the 15–17year-old adolescents who still resided in the region on 30 September 1996 were obtained from local Register Offices. Data about anti-hepatitis B vaccination (doses of vaccine and date of administration) were obtained from the public health services of the Local Health Units (LHUs). The sample size was determined by considering the immune state of the population as a binomial variable. The following constraints guided sample size determination: (1) proportion of vaccinees with a protective antibody level (]10 IU/l) [19] of 70% 5 years after vaccination; (2) confidence level of 95% and precision of 5%; (3) accounting for a proportion of untraceable or non-responding people of 50%. A total study population of 1192 subjects was formed, subdivided into two age groups of 533 children and 659 adolescents. These numbers moreover enabled one to make independent estimates for the two age groups and for each of the eight Local Health Units participating in the study. The subjects were drawn proportionally to the population of each birth cohort in the eight LHUs and randomly selected from Register Office lists.

2.2. Enrolment Parents were informed about the study by a letter. Another letter was sent directly to the adolescents. By means of subsequent telephone calls from a health care worker or, when no telephone number was available, a visit to the house, an appointment was made to complete the questionnaire and take a blood sample. The enrolment procedure was standardised throughout the eight Local Health Units. A subject was considered untraceable only if the third telephone call in the evening or the visit to his house met with no response.

2813

Refusal was defined as either explicit or as failure to appear at two successive appointments. The parents of all study participants gave informed written consent; participating adolescents also had to sign an informed consent.

2.3. Conduct of the sur6ey From November 1996 through October 1997 specifically trained health care workers administered a questionnaire to the parents, from which information about the subjects’ state of health and complete vaccination history as well as socio-economic factors was gathered. A blood sample was taken on the same occasion to screen for the presence of hepatitis B markers. Subjects who were negative for HBsAg and antiHBc, with anti-HBs titres below 10 IU/l, were offered a booster dose of HBV.

2.4. Laboratory procedures All serum samples taken from children and adolescents were tested for anti-HBs and anti-HBc. The samples with positive anti-HBc were tested also for HBsAg; the ones that tested positive were again tested for HBeAg and anti-HBe, using an enzyme-linked immunosorbent test (AUSAB-EIA, Corzyme and Auszyme ABBOTT Laboratories, Abbott Park, IL). Quantitative anti-HBs detection was performed with the same method, using a standard curve obtained from a panel of sera containing known amounts of anti-HBs. Antibody values were expressed in IU/l. The sensitivity of the test was 5 IU/l.

2.5. Data analysis The percentage of subjects with protective anti-HBs titres (] 10 IU/l) persisting after 5 years was calculated taking account of the stratification by Local Health Unit of our sample. Subjects with protective anti-HBs titres were compared with those without measurable antibodies or without a protective level of anti-HBs titres (B 10 IU/l) 5 years after vaccination for hepatitis B, in relation to the several potentially associated factors, by calculating the odds ratios (ORs) and the relative 95% confidence intervals (95% CI). Among the variables related to vaccination procedures, age at the initial dose (classified according to the two birth cohorts, as 0–13 months for children and 133 –160 months for adolescents), the number of doses administered, the interval since the last dose (a dichotomous variable above or below the mean) and the simultaneous administration, on the occasion of the hepatitis B vaccination, of oral polio (OPV),

2814

A. Faustini et al. / Vaccine 19 (2001) 2812–2818

diphtheria and tetanus (DT), or pertussis and measles vaccines were considered. Of the health conditions which are reported in the literature as associated with a reduced antibody response to hepatitis B vaccination, mental retardation [20], chronic kidney disease [21], being a dialysis patient [22], having a diagnosed tumour [23], immunodeficiency [24,25] and such chronic pathologies as anaemia [26] and diabetes [27] were analysed. A birthweight of less than 3 kg has also been considered a risk factor for infants vaccinated at birth [2]. Lastly, the father’s level of education [3], an indicator of socio-economic status, was considered as a proxy of the possibility of asymptomatic infection, working as a booster. The subjects with a protective titre were divided into three categories: low responders (anti-HBs titres= 10– 500 IU/l), medium responders (anti-HBs titres= 501 – 2000 IU/l) and high responders (anti-HBs titres\2000 IU/l) and the association between these three categories and the above-mentioned variables was evaluated. Crude ORs and the relative 95% confidence intervals were calculated, having low responders as a reference group. For those variables that showed statistically significant association, a multivariate analysis was carried out using a multiple politomic logistic regression model [28]. STATA software was used for data analysis [29].

3. Results Of the sample, 62.4% (744 out of 1192) took part in the study, 55.7% (297) of the children and 67.8% (447) of the adolescents. Of the 236 non-compliant children, 119 could not be located (12 had moved to another region), 68 refused to participate and 49 failed to appear at two consecutive appointments. Of the 212 non-compliant adolescents, 131 could not be located (five had moved to another region), 43 refused to participate and 38 failed to appear at two consecutive appointments. From six children and six adolescents it was impossible to take a blood sample. Of the 732 subjects who completed the questionnaire and gave a blood sample, a further 11 adolescents were not included in the analysis because they had received at least one dose of vaccine before June 1991. Five years after the beginning of the vaccination schedule (mean=4 years and 8 months; range= 4.0 – 5.7 years) the anti-HBs titres were higher than or equal to 10 IU/l in 92.9% (95% CI 92.3 – 93.6) of the children and 94.1% (95% CI 93.7 – 94.5) of the adolescents. Ten children and eight adolescents had an anti-HBs titre of 5 – 9 IU/l. Eleven children and 18 adolescents had an unmeasurable titre (less than 5 IU/l). Anti-HBc was positive in two children and five adolescents. Of the latter, one was HBsAg-positive, HBeAg-negative and

anti-HBe-positive. Chronic illness was reported in 21 cases, mental retardation in four (two children), kidney disease not involving dialysis in three (two children), immunodeficiency in one child and anaemia in 13 (six children). No cases of Down’s syndrome, tumours or diabetes were reported. Because the numbers were so low, the variables related to chronic pathologies were not further analysed. Among the 144 mothers who had reported being tested for HBsAg, three (2.1%) were positive. The vaccination schedules were followed, with average intervals between the first and second injections of 53.3 days (95% CI= 49.3 –57.3) for the children and 40.5 days (95% CI= 34.2 –46.7) for the adolescents. The intervals between the second and third injections were of 202.3 days (6.7 months) for the children and 177.3 days (5.9 months) for the adolescents. Of the variables considered, none proved to be associated with protective anti-HBs levels (] 10 IU/l), as opposed to titres of less than 10 IU/l, 5 years after vaccination (Table 1). Significant associations were observable between certain variables and the three groups of protective antibody levels (Table 2). In the vaccine recipients whose age at the first dose was greater (11 –12 years old) there was a higher probability of a medium antibody level (OR = 2.6, CI= 1.8 –3.7) or an high antibody level (OR = 4.1, CI= 2.7 –6.3). The number of doses of antihepatitis B vaccine, the interval since the last dose (mean= 4.1 years and range= 11 days –5.1 years) and having had a vaccination for pertussis and for measles did not show statistically significant association with antibody titres. Simultaneous administration of vaccines against hepatitis B and against diphtheria and tetanus (DT) appeared to be a risk factor for a lower antibody titre (OR= 0.2, 95% CI= 0.1 –0.5 for medium responders and OR= 0.1, 95% CI= 0.1 –0.3 for high responders). Simultaneous administration of polio vaccine was also associated with a lower HBs titre (OR= 0.3, 95% CI= 0.2 –0.5 for medium responders; OR= 0.2, 95% CI= 0.1 –0.3 for high responders). Birthweight was not associated with anti-HBs titres. The father’s level of education was associated, although not statistically significantly, with anti-HBs titres: among children of fathers with university degrees there was a higher probability of an antibody titre above 2000 IU/l (OR =1.9, 95% CI= 0.9 –4.1). Table 3 shows the results of the multivariate analysis. The age at the first dose and simultaneous DT vaccination were still associated with both categories of antibody titre (OR= 2.1, 95% CI= 1.4 –3.2 and OR=3.0, 95% CI= 1.4 –3.2 for age, OR= 0.4, 95% CI= 0.2 –0.8 and OR= 0.3, 95% CI= 0.1 –0.7 for the DT vaccination). Simultaneous anti-polio vaccination was still associated with a lower antibody titre only for the highest group (OR = 0.3, 95% CI= 0.1 –0.8); in this model one

A. Faustini et al. / Vaccine 19 (2001) 2812–2818

did not adjust for simultaneous DT because the two variables were too closely correlated. In the multivariate models, number of doses, interval since the last dose, vaccination for measles and pertussis and father’s educational level were not associated with anti-HBs titres.

2815

4. Discussion After a 5-year interval, 93% of the children and 94% of the adolescents vaccinated against hepatitis B at the beginning of the vaccination campaign had a protective antibody titre. These results concerning long-term im-

Table 1 Crude odds ratios of protective antibody titre 5 years after HBV vaccination, Lazio region, 1997 Anti-HBs titre (IU/l) B10 IU/l

]10 IU/l

Total

OR

95% CI

Age at first dose (months) 0–13 (children) 133–160 (adolescents)

21 26

270 404

291 430 721

1 1.2

0.7–2.2

No. doses 3 B3 \3

41 1 5

591 17 66

632 18 71 721

1 1.2 0.9

0.2–9.1 0.4–2.4

Inter6al since last dose (years) 54 \4

20 26

343 330

363 356 719

1 0.7

0.4–1.4

Vaccination Polio

42

614

656 0 656 0 165 382 547 159 444 603

Yes No Yes No Yes No

42

614

10 26

155 356

Yes No

6 31

153 413

Yes No

5 37

87 527

Yes No

4 38

72 542

Yes No

0 10

6 149

Yes No

0 6

2 151

Birthweight (g) B3000 ]3000

2 19

48 221

50 240 290

1 0.5

0.1–2.2

Father’s le6el of education Primary school Secondary school (lower) Secondary school (upper) University degree

9 16 20 1

97 292 215 65

106 308 235 66 715

1 1.7 1.0 5.9

0.7–3.9 0.4–2.2 0.7–47.3

DT Pertussis

Measles

Simultaneous 6accination Polio

DT

Pertussis

Measles

92 564 656 76 580 656 6 159 165 2 157 159

n.c. n.c. 1 0.9

0.4–1.9

1 0.5

0.2–1.3

1 0.8

0.3–2.1

1 0.8

0.3–2.3

n.c.

n.c.

A. Faustini et al. / Vaccine 19 (2001) 2812–2818

2816

Table 2 Association (crude odd ratios) between various factors and levels of immune response 5 years after HBV vaccination, Lazio region, 1997 Anti-HBs ranges (IU/l)

Age at first dose (months) 0–13 (children) 133–160 (adolescents) No. doses 3 B3 \3 Inter6al since last dose (years) 54 \4 Vaccination Pertussis Yes No Measles Yes No Simultaneous 6accination Polio No Yes DT No Yes Birthweight (g) B3000 ]3000 Father’s le6el of education Primary school Secondary school (lower) Secondary school (upper) University degree

\2000

10–500

501–2000

n

n

ORa

95% CI

n

ORa

95% CI

154 120

73 146

1 2.6

1.8–3.7

43 138

1 4.1

2.7–6.3

235 5 34

192 7 20

1 1.7 0.7

0.5–5.5 0.4–1.3

164 5 12

1 1.4 0.5

0.4–5.0 0.3–1.0

153 121

107 112

1 1.3

0.9–1.9

83 97

1 1.5

1.0–2.2

70 135 66 161

47 118 51 132

1 1.3 1 1.1

38 103 36 120

1 1.4 1 1.4

190 61 197 54

181 18 187 12

1 0.3 1 0.2

0.1–0.5

156 8 158 6

1 0.2 1 0.1

0.1–0.3

27 127

14 59

1 0.9

0.4–1.8

7 35

1 1.1

0.4–2.6

36 123 87 21

38 89 72 17

1 0.7 0.8 0.8

0.4–1.2 0.5–1.4 0.3–1.7

23 76 51 25

1 1.0 0.9 1.9

0.5–1.8 0.5–1.7 0.9–4.1

0.8–2.0 0.7–1.6

0.2–0.5

0.9–2.3 0.9–2.2

0.1–0.3

a Odds ratios were calculated by comparing subjects in the intermediate (501–2000 IU/l) and high (\2000 IU/l) titre categories with those in the lowest category.

munogenicity in a population with a low HBV endemicity are comparable to those reported in the literature [14] in studies of at-risk population groups and in follow-up studies of small groups. It was hypothesised that the immune response to the HBV vaccine was not influenced by change of residence on the part of the vaccinees, and hence one chose the population sample from the Register Office lists of residents as of 30 September 1996 so as to maximise feasibility. The interval before follow-up evaluation of antibody titres was defined as 5 years from the date of the first vaccination, as has been the practice in most follow-up studies [10,30]. Vaccination procedure in the public health services followed the national protocol [16]. The vaccine used was a recombinant DNA vaccine in the two commercially available forms, Engerix B and Recombinax HB, in combined sequences. None of the considered variables proved to be associated with the persistence of protective anti-HBs titres 5 years after vaccination, whereas certain variables were

associated with a persistence of different but still protective antibody levels. Age at vaccination was associated with the immune response, the adolescents having higher anti-HBs titres than the children. It has been reported that age at vaccination influences the immune response to vaccine, but inversely, in contrast to the findings. The anti-HBs response is reduced in people over 40 years old [31], and the percentage of vaccine recipients with post-vaccination protective antibodies is lower among adults than among children; there have been reports of 76% among health care workers [32] and 50% in homosexuals [33], as opposed to 96% and higher in healthy children and adolescents [14]. Five years after vaccination higher levels of antibodies have been described in the lower age groups. In children more than 93% have protective levels without booster shots [14], whereas the same percentage is reported in healthy adults only after booster shots [34]. Some authors have observed a persistence of antibodies 5 years after vaccination with titres that are inversely proportional to age at vaccination even among children of different ages: those who had been vaccinated at birth

A. Faustini et al. / Vaccine 19 (2001) 2812–2818

had titres with GMT (Geometric mean titre) of 105 IU/l, those who had received their vaccine as 2-year-old children had titres with GMT of 86 IU/l, and the 6-year-old vaccine recipients had titres with GMT of 58 IU/l [11]. The association between antibody level and age at vaccination does not change, in these data, even when adjustment is made for the other variables associated with antibody titres. Natural infections in adolescents, before or after vaccination, cannot explain the results since anti-HBc, a marker of infection, was detected only in two children and five adolescents. As for the number of doses of HBV vaccine administered, although significant associations were not observed, in the children who received more than three doses antibody titres are lower. In the literature there is complete agreement that number of doses and HBsAg concentration per dose of recombinant vaccine are posTable 3 Adjusteda odds ratios (OR) of immune response 5 years after HBV vaccination, Lazio region, 1997 Anti-HBs ranges (IU/l) \2000 IU/l

501–2000 ORb Age at first dose (months) 0–13 (children) 1 133–160 (adolescents) 2.1 No. doses 3 1 B3 1.6 \3 0.8 Inter6al since last dose (years) 54 1 \4 1.0 Vaccination Pertussis Yes 1 No 0.7 Measles Yes 1 No 0.7 Simultaneous 6accination Polioc No 1 Yes 0.5 DT No 1 Yes 0.4 Father’s le6el of education Primary school 1 Secondary school 1.0 (lower) Secondary school 1.1 (upper) University degree 0.8 a

95% CI

ORb

95% CI

1.4–3.2

1 3.0

1.9–4.8

0.4–6.5 0.4–1.4

1 0.8 0.5

0.2–4.1 0.2–1.0

0.7 –1.5

1 1.1

0.7–1.7

0.2–1.8 0.4–1.1

1 0.7 1 0.7

0.2–2.0 0.4–1.3

0.2–0.8

1 0.3 1 0.3

0.1–0.7

0.5–1.8

1 1.3

0.7–2.4

0.6–2.1

1.1

0.5–2.2

0.4–2.0

2.1

0.9–4.9

0.3–1.0

0.1–0.8

All estimates were adjusted for Local Health Unit, age at first dose and simultaneous DT vaccination. b Odds ratios were calculated by comparing subjects in the intermediate (501–2000 IU/l) and high (\2000 IU/l) titre categories with those in the lowest category. c In this model simultaneous DT vaccination was excluded because of the strong correlation.

2817

itively associated with antibody titres [35]. One explanation could be that the children may have received a fourth dose because their anti-HBs titres were lower than expected. The other variable associated with HBs antibody titres 5 years after vaccination is the simultaneous administration of other vaccines: children who received OPV and/or DT vaccines at the same time as HBV vaccine have lower antibody levels. The effect of simultaneous administration of other vaccines is still present after adjustment for age at vaccination. Earlier studies agree that simultaneous administration of OPV, DT and HBV vaccines is not associated with a reduced immune response to HBV vaccine, if one considers a protective anti-HBs titre immediately after completion of the HBV vaccine cycle as the outcome [36 –39]. A recent Italian study [12] reports a lower antibody response when acellular pertussis vaccine was given simultaneously. Another study [40] reports lower anti-HBs titres in the 6 months following simultaneous DTP-polio vaccination. It is possible that simultaneous administration causes a lower, although still protective, antibody level after the course of vaccinations is completed, or a more rapid reduction in antibodies over time. Since the observed effect of simultaneous administration one will persist in the population for some years, approximately 50% of the children who have titres in the range of 10 –500 IU/l anti-HBs, will not have protective anti-HBs titres when they will be 10–12 years old, assuming for these children a reduction in antibodies, over the 5 years following 1997, by a factor of 1.3, in accordance with Gesemann-Scheiermann’s formula [9]. While one now knows [3,14,15] that a decrease in the anti-HBs level to below 10 IU/l does not involve a loss of the ability to produce antibodies to hepatitis B surface antigen, it is not yet clear what the impact of this condition will be on the population in the case of exposure to risk. Therefore some authors [14] are attracted by the possibility of a booster vaccination at 10 years, in keeping with anti-tetanus vaccination schedules, and others [41] suggest, for vaccinees with antibody titres below 10 IU/l, a booster dose of HBV vaccine in addition to specific immunoglobuline after certain exposures. An Italian group [42] has proposed and tested, as a proxy marker of immunological memory, the serological response to a booster of HBV. A further follow-up of this population 10 years after the start of the vaccination campaign would be helpful to determining how long this vaccine can offer protection against infection from HBV virus and chronic hepatitis. References [1] Szmuness W, Stevens CE, Harley EJ, et al. Hepatitis B vaccine. N Engl J Med 1980;303:833– 41. [2] Beasley RP, Hwang LY, Lee GC, et al. Prevention of perinatally transmitted hepatitis B virus infections with hepatitis B immune globulin and hepatitis B vaccine. Lancet 1983;2:1099– 102.

2818

A. Faustini et al. / Vaccine 19 (2001) 2812–2818

[3] Hadler SC, Francis DP, Maynard MD, et al. Long-term immunogenicity and efficacy of hepatitis B vaccine in homosexual men. N Engl J Med 1986;315:209–14. [4] Mintai Z, Kezhou L, Lieming D, Smego RA, Jr. Duration and efficacy of immune response to hepatitis B vaccine in high-risk Chinese adolescents. Clin Infect Dis 1993;16:165–7. [5] Coursaget P, Chotard J, Vincelot P, et al. Seven-year study of hepatitis B vaccine efficacy in infants from an endemic area (Senegal). Lancet 1986;2:1143–4. [6] Lieming D, Mintai Z, Yinfu W, Shaochon Z, Weigin K, Smego RA, Jr. A 9-year follow-up study of the immunogenicity and long-term efficacy of plasma-derived hepatitis B vaccine in highrisk Chinese neonates. Clin Infect Dis 1993;17:475–9. [7] Lee P-I, Lee C-Y, Huang L-M, Chang M-H. Long-term efficacy of recombinant hepatitis B vaccine and risk of natural infection in infants born to mothers with hepatitis B e antigen. J Pediatr 1995;126:716– 21. [8] Jilg W, Schmidt M, Deinhardt F, Zachoval R. Hepatitis B vaccination: how long does protection last? Lancet 1984;2:458. [9] Gesemann M, Schroder S, Scheiermann N, Maurer C. Kinetics of anti-HBs after hepatitis B vaccination: a comparison of two recombinant and one plasma-derived vaccines. Arch Virol 1992;4S:154– 5. [10] Van Herck K, Van Damme P, Thoelen S, Meheus A. Long-term persistence of anti-HBs after vaccination with a recombinant DNA yeast-derived hepatitis B vaccine: 8-year results. Vaccine 1998;16:1933– 5. [11] Milne A, Krugman S, Waldon JA, et al. Hepatitis B vaccination in children: five-year booster study. NZ Med J 1992;105:336– 8. [12] Salmaso S, Piscitelli A, Rapicetta M, Chionne P, Madonna E, Argentini C. Immunogenicity of hepatitis B vaccines among infant recipients of acellular and whole cell pertussis DTP vaccines. Vaccine 1998;16:643–6. [13] Laplanche A, Courouce´ A-M, Jungers P, Benhamou E, Crosnier J. Hepatitis B vaccination: how long does protection last? Lancet 1984;2:866. [14] West DJ, Calandra GB. Vaccine induced immunologic memory for hepatitis B surface antigen: implications for policy on booster vaccination. Vaccine 1996;14:1019–27. [15] Van Hattum J, Maikoe T, Poel J, De Gast GC. In vitro anti-HBs production by individual B cells of responders to hepatitis B vaccine who subsequently lost antibody. In: Hollinger FB, Lemon SM, Margolis HS, editors. Viral hepatitis and viral diseases. Baltimore: Williams and Wilkins, 1991:774–6. [16] Gazzetta Ufficiale 1991. N° 127, legge 27, Maggio No. 165. [17] Piazza M, Da Villa G, Picciotto L, et al. Mass vaccination against hepatitis B in infants in Italy. Lancet 1988;2:1132. [18] Centers for Disease Control. Recommendation of the Immunization Practice Advisory Committee (ACIP). Recommendations for protection against viral hepatitis. MMWR 1985;34. [19] Goudeau A, Coursaget P, Barin F, et al. Prevention of hepatitis B by active and passive-active immunization. In: Szmuness W, Alter HJ, Maynard JE, editors. Viral hepatitis 1981, International Symposium. Philadelphia: Franklin Institute Press, 1982:509– 25. [20] Ueda K, Tokugawa K, Hashiguchi Y, et al. Prevention of horizontal transmission of hepatitis B: efficacy of hepatitis B immunoglobulin and vaccine in an institution for the handicapped. Vaccine 1988;6:54– 8. [21] Seaworth B, Drucker J, Starling J, Drucker R, Stevens C, Hamilton J. Hepatitis B vaccines in patients with chronic renal failure before dialysis. J Infect Dis 1988;157:332–7. [22] Steketee RW, Ziarnik ME, Davis JP. Seroresponse to hepatitis B vaccine in patients and staff of renal dialysis centers, Wisconsin. Am J Epidemiol 1988;127:772–82. [23] Berberoglu S, Buyukpamukcu M, Sarialioglu F, Akyuz C, Ilhan I. Hepatitis B vaccination in children with cancer. Pediatr Hematol Oncol 1995;12:171– 8.

[24] Drake JH, Parmley RT, Britton HA. Loss of hepatitis B antibody in human immunodeficiency virus-positive hemophilia patients. Pediatr Infect Dis J 1987;6:1051– 4. [25] Zuin G, Principi N, Tornaghi R, et al. Impaired response to hepatitis B vaccine in HIV infected children. Vaccine 1992;10: 857– 60. [26] Sarnaik SA, Merline JR, Bond S. Immunogenicity of hepatitis B vaccine in children with sickle cell anemia. J Pediatr 1988;112: 429– 30. [27] Pozzilli P, Arduini P, Visalli N, et al. Reduced protection against hepatitis B virus following vaccination in patients with type 1 (insulin-dependent) diabetes. Diabetologia 1987;30:817–9. [28] Hosmer DM, Lemeshow S. Applied logistic regression. New York: Wiley, 1989. [29] Stata Corporation 1997. Stata statistical software: release 5.0. College Station, TX: Stata Corporation. [30] Scheiermann N, Gesemann M, Maurer C, Just M, Berger R. Persistence of antibodies after immunization with a recombinant yeast-derived hepatitis B vaccine following two different schedules. Vaccine 1990;8S:44– 6. [31] Clements ML, Miskovsky E, Davidson M, et al. Effect of age on the immunogenicity of yeast recombinant hepatitis B vaccines containing surface antigen (S) or preS2 +S antigens. J Infect Dis 1994;170:510– 6. [32] Taylor PE, Stevens CE. Persistence of antibody to hepatitis B surface antigen after vaccination with hepatitis B vaccine. In: Zuckerman A, editor. Viral hepatitis and liver disease. New York: Alan R. Liss Inc., 1988:995– 7. [33] Gibas A, Watkins E, Hinkle C, Dienstag JL. Long-term persistence of protective antibody after hepatitis B vaccination of healthy adults. In: Zuckerman A, editor. Viral Hepatitis and Liver Disease. New York: Alan R. Liss Inc., 1988:998– 1001. [34] Courouce´ A-M, Laplanche A, Benhamou E, Jungers P. Long-term efficacy of hepatitis B vaccination in healthy adults. In: Zuckerman A, editor. Viral Hepatitis and Liver Disease. New York: Alan R. Liss Inc., 1988:1002– 5. [35] Yeoh EK, Lai CL, Chang WK, Lo HY. Comparison of the immunogenicity, efficacy and safety of 10 micrograms and 20 micrograms of a hepatitis B vaccine: a prospective randomized trial. J Hyg (Lond) 1986;96:491– 9. [36] Chiron JP, Coursaget P, Yvonnet B, et al. Simultaneous administration of hepatitis B and diphtheria/tetanus/polio vaccines. Lancet 1984;1:623– 4. [37] Giammanco G, Li Volti S, Mauro L, et al. Immune response to simultaneous administration of a recombinant DNA hepatitis B vaccine and multiple compulsory vaccines in infancy. Vaccine 1991;9:747– 50. [38] Da Villa G, Picciotto L, Ribera G, Bencivenga M, Cotugno M, Hartmann P. Effective antibody response in newborn babies living in Maldives to simultaneous vaccination against hepatitis B, poliomyelitis, diphtheria and tetanus. Vaccine 1995;13:795–8. [39] Aristegui J, Muniz J, Perez Legorburu A, et al. Newborn universal immunisation against hepatitis B: immunogenicity and reactogenicity of simultaneous administration of diphtheria/tetanus/ pertussis (DTP) and oral polio vaccines with hepatitis B vaccine at 0, 2 and 6 months of age. Vaccine 1995;13:973– 7. [40] Grosheide PM, del Canho R, Voogd M, Heijtink RA, Schalm SW and the Dutch Study Group on Prevention of Neonatal Hepatitis B. Anti-HBs levels in infants of hepatitis B carrier mothers after delayed active immunization with recombinant vaccine concomitant with DTP-polio vaccine: is there need for a second dose of HBIg? Vaccine 1994;12:1059– 63. [41] Stanley ML, Thomas DL. Vaccines to prevent viral hepatitis. N Engl J Med 1997;336:196– 204. [42] Resti M, Azzari C, Mannelli F, Rossi ME, Lionetti P, Vierucci A. Ten-year follow-up study of neonatal hepatitis B immunization: are booster injections indicated? Vaccine 1997;15: 1338– 40.