- Email: [email protected]
The effectiveness and safety of hepatitis A vaccine: a systematic review
Vaccine 21 (2003) 2242–2245
The effectiveness and safety of hepatitis A vaccine: a systematic review Vittorio Demicheli∗ , Donatella Tiberti Servizio Epidemiologico, ASL 20 Alessandria and Cochrane Vaccines Field, Via Venezia 6, 15100 Alessandria, Italy
Abstract We report on the conduct of a systematic review to assess the efficacy and the safety of hepatitis A vaccines in adults and children. We identified, retrieved, and assessed all trials evaluating the effects of hepatitis A vaccines on prevention of cases of hepatitis A, death from hepatitis A, and assessing nature and frequency of adverse events. We included eight randomised trials, four containing efficacy outcomes, three containing only safety outcomes and a single study containing efficacy and adverse events outcomes. Combined inactivated vaccine effectiveness was 86% (95% CI: 63–95%). Combined attenuated vaccine effectiveness was 95% (95% CI: 81–99%). Inactivated vaccine effectiveness in the prevention of HAV secondary cases, compared to non-intervention was 82% (95% CI: 23–96%). Safety profile of vaccines was similar to that of their comparators. Despite poor design and reporting of trials, we found convincing evidence of the effectiveness and safety of inactivated HAV vaccines. © 2003 Elsevier Science Ltd. All rights reserved. Keywords: Hepatitis A vaccine; Systematic review; Inactivated HAV vaccines
1. Background
2. Methods
Several inactivated and attenuated hepatitis A vaccines have been developed and evaluated in human clinical trials and in non-human primate models of HAV infection. Only inactivated vaccines have been evaluated in randomised clinical trials. Local reactions (soreness at the injection site) and systemic reactions (headache, malaise and feeding problems in children) have been reported for one type of vaccine. Local reactions (tenderness, pain, and warmth at the injection site) have been reported for another type of vaccine. Reviews of data from multiple sources with longer than 5 years of follow-up did not identify serious adverse events among children or adults that could be definitively attributed to hepatitis A vaccines. The rates of serious adverse events in vaccinated and unvaccinated populations are similar. We report on the conduct of a systematic review to assess the efficacy and the safety of hepatitis A vaccines in adults and children. We identified, retrieved, and assessed all trials evaluating the effects of hepatitis A vaccines on prevention of cases of hepatitis A, avoiding death from hepatitis A, and frequency of adverse events.
We included randomised, cluster randomised, and quasi-randomised trials (i.e. trials using alternation, date of birth, or case record number) irrespective of blinding, language, publication status, and study period carried out on any person, irrespective of age, sex, immune status, or risk category. We included trials assessing the effects of any type of hepatitis A vaccine (attenuated or inactivated), irrespective of route of administration, dosage, and schedule compared with placebo, no intervention, or any vaccine other than HAV vaccine. Attenuated and inactivated vaccines were analysed separately. We chose the following outcomes: Primary: • Numbers of clinically defined hepatitis A cases. • Numbers of laboratory confirmed (immunoglobulin M-HAV (IgM-HAV)) hepatitis A cases. • Numbers of deaths from hepatitis A. • All-cause mortality. Secondary:
∗ Corresponding author. Tel.: +39-0131-307821; fax: +39-0131-307847. E-mail address: [email protected] (V. Demicheli).
0264-410X/03/$ – see front matter © 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0264-410X(03)00135-X
• Adverse events—number and seriousness of adverse events (classified as local or systemic). Local adverse events include induration, soreness, redness, tenderness, pain, and warmth at the injection site. Systemic adverse events include cases of fever,
V. Demicheli, D. Tiberti / Vaccine 21 (2003) 2242–2245
headache, malaise, feeding problems in children, and more generalised, serious signs. Depending on availability of data, we attempted to classify the adverse events as serious or non-serious according to the 1997 International Conference on Harmonisation-Good Clinical Practice. We searched the Cochrane Library to identify reports of randomised, cluster randomised, and quasi-randomised trials and systematic reviews. The following subject search terms were combined: HEPATITIS-A-VACCINES or (HEPATITIS-A or hepatitis A) and (VACCINES-ATTENUATED or VACCINES-INACTIVATED or explode VACCINES or VACCINATION or explode IMMUNIZATION or vaccin∗ or immuni∗ or inoculat∗ ) (N.B.: Upper case denotes controlled vocabulary and lower case denotes free-text terms). This subject search strategy was adapted to search the following databases: • MEDLINE (for the years 1966–December 2000 inclusive). • EMBASE (for the years 1974–2000 inclusive). • SciSearch (Science Citation Index for the years 1974–2000 inclusive). • PubScience (for the years 1974–2000 inclusive). The methodological search filter for high sensitivity in identifying randomised controlled trials in MEDLINE, parts I and II (11a.15 Appendix B, Cochrane Reviewers’ Handbook 4.1, June 2000, p. 153) was added to the subject search strategy to identify reports of controlled trials in MEDLINE. This filter was adapted for use in searching the other databases listed above. The searches were conducted in any language. We read the bibliographies of relevant articles in order to identify additional trials. We used the results of journals searched by hand, e.g. Vaccine which are included in the CENTRAL/Cochrane Controlled Trials Register. To locate further published or unpublished trials, we wrote to vaccine manufacturers and to the first or corresponding authors of relevant studies. The review was performed and reported according to the rules contained in the QUOROM statement. Two reviewers (VD and DT) read all trials retrieved in the search and applied the inclusion and exclusion criteria. The two reviewers judged the trials separately and possible controversies were arbitrated by the co-ordinator of the Cochrane Vaccine Field. The methodological quality of randomised, cluster randomised, and quasi-randomised trials was assessed using the following definitions adapted from the Cochrane Handbook. Randomisation: • A = individual participants allocated to vaccine or control group. • B = groups of participants allocated to vaccine or control group.
2243
Generation of the allocation sequence: • A = adequate, e.g. table of random numbers or computer generated random numbers. • B = inadequate, e.g. alternation, date of birth, day of the week, or case record number. • C = not described. Allocation concealment: • A = adequate, e.g. numbered or coded identical containers administered sequentially, on-site computer system that can only be accessed after entering the characteristics of an enrolled participant, or serially numbered, opaque, sealed envelopes. • B = possibly adequate, e.g. sealed envelopes that are not sequentially numbered or opaque and trials in which the prognostic variables of participants allocated to vaccine and control intervention are significantly different although the report adequate allocation concealment seems adequate. • C = inadequate, e.g. open table of random numbers. • D = not described. Blinding: • A = adequate double blinding, e.g. placebo vaccine. • B = single blind, i.e. blinded outcome assessment. • C = no blinding. 2.1. Follow-up Average duration of follow-up and number of losses to follow-up. 2.2. Data extraction The following data were extracted, verified, and recorded for each included study. Characteristics of participants: • number of participants; • age and ethnic group. Characteristics of interventions: • type of vaccine, type of control, dose, and schedule; • length of follow-up. Characteristics of outcome measures • case definition(s) used by the study; • number of hepatitis A cases, number of deaths from hepatitis A, and all-cause mortality in vaccination and control groups; • occurrence and type of adverse events. Trial characteristics: • date; • location;
2244
V. Demicheli, D. Tiberti / Vaccine 21 (2003) 2242–2245
• sponsor of trial (known or unknown and type of sponsor); • publication status.
A total population of 404 healthy young and adults was analysed. Vaccine effectiveness was 82% (95% CI: 23–96%).
2.3. Data synthesis Primary and secondary outcomes were calculated as relative risk with its 95% CI and presented as vaccine effectiveness according to the following formula: VE = 1 − RR. A random effects model was used in case of significant heterogeneity (P < 0.1). We performed sensitivity analyses to compare trials with adequate and inadequate/unclear methodological quality and randomised trials, cluster randomised trials, and quasi-randomised trials. We attempted to perform subgroup analyses of attenuated and inactivated vaccines in general populations and in high-risk populations. The analyses were performed on a per protocol basis, due to the non-availability of the data in the primary studies.
3. Results
3.1.2. Studies included in the adverse events review A variety of secondary outcome measures were found, none following the recommendations of the International Conference on Harmonisation-Good Clinical Practice. Two studies [3,6] assessed the effect of HAV inactivated vaccine in the production of at least a local adverse event following the immunisation. No difference was found between the vaccine and the placebo arm (RR = 1; 95% CI: 0.68– 1.47). Two studies [7,8] assessed the effect of HAV inactivated vaccine compared to immunoglobulin in producing headache and any adverse event in the first 2 weeks after vaccination. None of the two outcomes showed differences between the vaccine and the immunoglobulin arm. RR for headache = 1.39 (95% CI: 0.69–2.99). RR for any adverse events within 2 weeks from the immunisation = 1.01 (95% CI: 0.80–1.27).
3.1. Description of studies Search strategies allowed the identification of 1042 potentially relevant studies. As a result of the analysis of the titles and of the available abstracts, 149 studies were retrieved, 115 of which were considered for inclusion. Eight studies met the inclusion criteria. Four studies contained primary outcomes, three studies contained only secondary outcomes and a single study contained contemporarily efficacy and adverse events outcomes. One hundred and seven studies were excluded (53% lacking of adequate comparator, 17% lacking outcome measures, 11% concerning interventions different from HAV vaccine and 19% for various other reasons). 3.1.1. Studies included in the efficacy review The primary outcome measure found was the number of clinically and contemporarily laboratory confirmed cases of hepatitis A. Three studies [1–3] considered the effectiveness of HAV inactivated vaccine in the prevention of cases of hepatitis A. Two studies [1,2] used HBV vaccine as control, and one [3] was placebo-controlled. A total population of 41,417 healthy children was considered by the three studies. Resulting vaccine effectiveness was 86% (95% CI: 63–95%). One study [4] regarded the effectiveness of HAV attenuated vaccine in the prevention of laboratory confirmed cases of hepatitis A and was placebo-controlled. Because of the different randomisation methods applied in the two phases of the study, results were introduced separately into the analysis. There were 495,352 children involved in total. Summary vaccine effectiveness was 95% (95%CI: 81–99%). Another single [5] study evaluated the effectiveness of HAV inactivated vaccine in the prevention of HAV secondary cases, with a no-intervention comparative arm.
4. Discussion The size of the included studies and the consistency of their results across different settings and population indicate convincing evidence of effectiveness of inactivated HAV vaccine in the prevention of both primary and secondary cases of hepatitis A. Nevertheless, the quality assessment of the available studies showed the presence of a large range of problems regarding the design, the execution and the reporting of the studies. Very little can be said on the effects of the attenuated HAV vaccine, because of the existence of a single poorly reported study. Results from this review indicate that the vaccine is effective, but no safety information were found as a result of the extensive searches performed. The results of the safety review were disappointing, as from more than 100 studies we found safety information from four studies. The data included related to very general and clinically irrelevant outcome measures. The main reasons for this wastage of data were the use of non-comparative study designs (53% had no comparative arm) and in the use of unclear and not standardised definition of adverse events that dramatically reduced the number of studies containing comparable outcome measures. Moreover, the use of population of volunteers and the poor methodological quality of the available studies probably undermine their representativeness and the transferability of their results. In conclusion, HAV inactivated vaccine is an effective intervention for the prevention of primary and secondary hepatitis A cases, even if incomplete information is available on
V. Demicheli, D. Tiberti / Vaccine 21 (2003) 2242–2245
its safety. Specific recommendations for the use of the vaccine in a general population should be developed taking into account the local burden of the disease and the cost effectiveness of the many possible immunisation strategies. Future studies on HAV vaccine should be better designed and reported. There is an urgent need for the process of standardising definitions of adverse event following immunisation to be completed and fully implemented in all future studies regardless of the vaccine assessed and of the study design applied.
Acknowledgements We would like to thank Tom Jefferson for arbitration in the review process and Carlo Di Pietrantonj for the statistical advice given. We would also like to thank Alfonso Mele for providing us suggestion based on his extensive knowledge of the subject and the editors of Cochrane Hepatobiliary Review Group and the anonymous referees for the help provided and their constructive comments.
2245
References [1] Innis BL, Snitbhan R, Kunasol P, et al. Protection against hepatitis A by an inactivated vaccine. JAMA 1994;271(17):1328–34. [2] Riedemann S, Reinhardt G, Frosner GG, et al. Placebo-controlled efficacy study of hepatitis A vaccine in Valdivia, Chile. Vaccine 1992;10(Suppl 1):S152–5. [3] Werzberger A, Mensch B, Kuter B, et al. A controlled trial of a formalin-inactivated hepatitis A vaccine in healthy children. N Engl J Med 1992;327(7):453–7. [4] Xu Z, Li R, Meng Z. Immunogenicity and efficacy trials of live attenuated hepatitis A vaccines. Zhonghua Yi Xue Za Zhi 1998;78(4):254–6 [in Chinese]. [5] Sagliocca L, Amoroso P, Stroffolini T. Efficacy of hepatitis A vaccine in prevention of secondary hepatitis A infection: a randomised trial. Lancet 1999;353(9159):1136–9. [6] Ellerbeck EF, Lewis JA, Nalin D, et al. Safety profile and immunogenicity of an inactivated vaccine derived from an attenuated strain of hepatitis A. Vaccine 1992;10(10):668–70. [7] Leentvaar-Kuijpers A, Coutinho RA, Brulein V, Safary A. Simultaneous passive and active immunization against hepatitis A. Vaccine 1992;10(Suppl 1):S138–41. [8] Walter EB, Hornick RB, Poland GA. Concurrent administration of inactivated hepatitis A vaccine with immune globulin in healthy adults. Vaccine 1999;17(11/12):1468–73.
The effectiveness and safety of hepatitis A vaccine: a systematic review Vittorio Demicheli∗ , Donatella Tiberti Servizio Epidemiologico, ASL 20 Alessandria and Cochrane Vaccines Field, Via Venezia 6, 15100 Alessandria, Italy
Abstract We report on the conduct of a systematic review to assess the efficacy and the safety of hepatitis A vaccines in adults and children. We identified, retrieved, and assessed all trials evaluating the effects of hepatitis A vaccines on prevention of cases of hepatitis A, death from hepatitis A, and assessing nature and frequency of adverse events. We included eight randomised trials, four containing efficacy outcomes, three containing only safety outcomes and a single study containing efficacy and adverse events outcomes. Combined inactivated vaccine effectiveness was 86% (95% CI: 63–95%). Combined attenuated vaccine effectiveness was 95% (95% CI: 81–99%). Inactivated vaccine effectiveness in the prevention of HAV secondary cases, compared to non-intervention was 82% (95% CI: 23–96%). Safety profile of vaccines was similar to that of their comparators. Despite poor design and reporting of trials, we found convincing evidence of the effectiveness and safety of inactivated HAV vaccines. © 2003 Elsevier Science Ltd. All rights reserved. Keywords: Hepatitis A vaccine; Systematic review; Inactivated HAV vaccines
1. Background
2. Methods
Several inactivated and attenuated hepatitis A vaccines have been developed and evaluated in human clinical trials and in non-human primate models of HAV infection. Only inactivated vaccines have been evaluated in randomised clinical trials. Local reactions (soreness at the injection site) and systemic reactions (headache, malaise and feeding problems in children) have been reported for one type of vaccine. Local reactions (tenderness, pain, and warmth at the injection site) have been reported for another type of vaccine. Reviews of data from multiple sources with longer than 5 years of follow-up did not identify serious adverse events among children or adults that could be definitively attributed to hepatitis A vaccines. The rates of serious adverse events in vaccinated and unvaccinated populations are similar. We report on the conduct of a systematic review to assess the efficacy and the safety of hepatitis A vaccines in adults and children. We identified, retrieved, and assessed all trials evaluating the effects of hepatitis A vaccines on prevention of cases of hepatitis A, avoiding death from hepatitis A, and frequency of adverse events.
We included randomised, cluster randomised, and quasi-randomised trials (i.e. trials using alternation, date of birth, or case record number) irrespective of blinding, language, publication status, and study period carried out on any person, irrespective of age, sex, immune status, or risk category. We included trials assessing the effects of any type of hepatitis A vaccine (attenuated or inactivated), irrespective of route of administration, dosage, and schedule compared with placebo, no intervention, or any vaccine other than HAV vaccine. Attenuated and inactivated vaccines were analysed separately. We chose the following outcomes: Primary: • Numbers of clinically defined hepatitis A cases. • Numbers of laboratory confirmed (immunoglobulin M-HAV (IgM-HAV)) hepatitis A cases. • Numbers of deaths from hepatitis A. • All-cause mortality. Secondary:
∗ Corresponding author. Tel.: +39-0131-307821; fax: +39-0131-307847. E-mail address: [email protected] (V. Demicheli).
0264-410X/03/$ – see front matter © 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0264-410X(03)00135-X
• Adverse events—number and seriousness of adverse events (classified as local or systemic). Local adverse events include induration, soreness, redness, tenderness, pain, and warmth at the injection site. Systemic adverse events include cases of fever,
V. Demicheli, D. Tiberti / Vaccine 21 (2003) 2242–2245
headache, malaise, feeding problems in children, and more generalised, serious signs. Depending on availability of data, we attempted to classify the adverse events as serious or non-serious according to the 1997 International Conference on Harmonisation-Good Clinical Practice. We searched the Cochrane Library to identify reports of randomised, cluster randomised, and quasi-randomised trials and systematic reviews. The following subject search terms were combined: HEPATITIS-A-VACCINES or (HEPATITIS-A or hepatitis A) and (VACCINES-ATTENUATED or VACCINES-INACTIVATED or explode VACCINES or VACCINATION or explode IMMUNIZATION or vaccin∗ or immuni∗ or inoculat∗ ) (N.B.: Upper case denotes controlled vocabulary and lower case denotes free-text terms). This subject search strategy was adapted to search the following databases: • MEDLINE (for the years 1966–December 2000 inclusive). • EMBASE (for the years 1974–2000 inclusive). • SciSearch (Science Citation Index for the years 1974–2000 inclusive). • PubScience (for the years 1974–2000 inclusive). The methodological search filter for high sensitivity in identifying randomised controlled trials in MEDLINE, parts I and II (11a.15 Appendix B, Cochrane Reviewers’ Handbook 4.1, June 2000, p. 153) was added to the subject search strategy to identify reports of controlled trials in MEDLINE. This filter was adapted for use in searching the other databases listed above. The searches were conducted in any language. We read the bibliographies of relevant articles in order to identify additional trials. We used the results of journals searched by hand, e.g. Vaccine which are included in the CENTRAL/Cochrane Controlled Trials Register. To locate further published or unpublished trials, we wrote to vaccine manufacturers and to the first or corresponding authors of relevant studies. The review was performed and reported according to the rules contained in the QUOROM statement. Two reviewers (VD and DT) read all trials retrieved in the search and applied the inclusion and exclusion criteria. The two reviewers judged the trials separately and possible controversies were arbitrated by the co-ordinator of the Cochrane Vaccine Field. The methodological quality of randomised, cluster randomised, and quasi-randomised trials was assessed using the following definitions adapted from the Cochrane Handbook. Randomisation: • A = individual participants allocated to vaccine or control group. • B = groups of participants allocated to vaccine or control group.
2243
Generation of the allocation sequence: • A = adequate, e.g. table of random numbers or computer generated random numbers. • B = inadequate, e.g. alternation, date of birth, day of the week, or case record number. • C = not described. Allocation concealment: • A = adequate, e.g. numbered or coded identical containers administered sequentially, on-site computer system that can only be accessed after entering the characteristics of an enrolled participant, or serially numbered, opaque, sealed envelopes. • B = possibly adequate, e.g. sealed envelopes that are not sequentially numbered or opaque and trials in which the prognostic variables of participants allocated to vaccine and control intervention are significantly different although the report adequate allocation concealment seems adequate. • C = inadequate, e.g. open table of random numbers. • D = not described. Blinding: • A = adequate double blinding, e.g. placebo vaccine. • B = single blind, i.e. blinded outcome assessment. • C = no blinding. 2.1. Follow-up Average duration of follow-up and number of losses to follow-up. 2.2. Data extraction The following data were extracted, verified, and recorded for each included study. Characteristics of participants: • number of participants; • age and ethnic group. Characteristics of interventions: • type of vaccine, type of control, dose, and schedule; • length of follow-up. Characteristics of outcome measures • case definition(s) used by the study; • number of hepatitis A cases, number of deaths from hepatitis A, and all-cause mortality in vaccination and control groups; • occurrence and type of adverse events. Trial characteristics: • date; • location;
2244
V. Demicheli, D. Tiberti / Vaccine 21 (2003) 2242–2245
• sponsor of trial (known or unknown and type of sponsor); • publication status.
A total population of 404 healthy young and adults was analysed. Vaccine effectiveness was 82% (95% CI: 23–96%).
2.3. Data synthesis Primary and secondary outcomes were calculated as relative risk with its 95% CI and presented as vaccine effectiveness according to the following formula: VE = 1 − RR. A random effects model was used in case of significant heterogeneity (P < 0.1). We performed sensitivity analyses to compare trials with adequate and inadequate/unclear methodological quality and randomised trials, cluster randomised trials, and quasi-randomised trials. We attempted to perform subgroup analyses of attenuated and inactivated vaccines in general populations and in high-risk populations. The analyses were performed on a per protocol basis, due to the non-availability of the data in the primary studies.
3. Results
3.1.2. Studies included in the adverse events review A variety of secondary outcome measures were found, none following the recommendations of the International Conference on Harmonisation-Good Clinical Practice. Two studies [3,6] assessed the effect of HAV inactivated vaccine in the production of at least a local adverse event following the immunisation. No difference was found between the vaccine and the placebo arm (RR = 1; 95% CI: 0.68– 1.47). Two studies [7,8] assessed the effect of HAV inactivated vaccine compared to immunoglobulin in producing headache and any adverse event in the first 2 weeks after vaccination. None of the two outcomes showed differences between the vaccine and the immunoglobulin arm. RR for headache = 1.39 (95% CI: 0.69–2.99). RR for any adverse events within 2 weeks from the immunisation = 1.01 (95% CI: 0.80–1.27).
3.1. Description of studies Search strategies allowed the identification of 1042 potentially relevant studies. As a result of the analysis of the titles and of the available abstracts, 149 studies were retrieved, 115 of which were considered for inclusion. Eight studies met the inclusion criteria. Four studies contained primary outcomes, three studies contained only secondary outcomes and a single study contained contemporarily efficacy and adverse events outcomes. One hundred and seven studies were excluded (53% lacking of adequate comparator, 17% lacking outcome measures, 11% concerning interventions different from HAV vaccine and 19% for various other reasons). 3.1.1. Studies included in the efficacy review The primary outcome measure found was the number of clinically and contemporarily laboratory confirmed cases of hepatitis A. Three studies [1–3] considered the effectiveness of HAV inactivated vaccine in the prevention of cases of hepatitis A. Two studies [1,2] used HBV vaccine as control, and one [3] was placebo-controlled. A total population of 41,417 healthy children was considered by the three studies. Resulting vaccine effectiveness was 86% (95% CI: 63–95%). One study [4] regarded the effectiveness of HAV attenuated vaccine in the prevention of laboratory confirmed cases of hepatitis A and was placebo-controlled. Because of the different randomisation methods applied in the two phases of the study, results were introduced separately into the analysis. There were 495,352 children involved in total. Summary vaccine effectiveness was 95% (95%CI: 81–99%). Another single [5] study evaluated the effectiveness of HAV inactivated vaccine in the prevention of HAV secondary cases, with a no-intervention comparative arm.
4. Discussion The size of the included studies and the consistency of their results across different settings and population indicate convincing evidence of effectiveness of inactivated HAV vaccine in the prevention of both primary and secondary cases of hepatitis A. Nevertheless, the quality assessment of the available studies showed the presence of a large range of problems regarding the design, the execution and the reporting of the studies. Very little can be said on the effects of the attenuated HAV vaccine, because of the existence of a single poorly reported study. Results from this review indicate that the vaccine is effective, but no safety information were found as a result of the extensive searches performed. The results of the safety review were disappointing, as from more than 100 studies we found safety information from four studies. The data included related to very general and clinically irrelevant outcome measures. The main reasons for this wastage of data were the use of non-comparative study designs (53% had no comparative arm) and in the use of unclear and not standardised definition of adverse events that dramatically reduced the number of studies containing comparable outcome measures. Moreover, the use of population of volunteers and the poor methodological quality of the available studies probably undermine their representativeness and the transferability of their results. In conclusion, HAV inactivated vaccine is an effective intervention for the prevention of primary and secondary hepatitis A cases, even if incomplete information is available on
V. Demicheli, D. Tiberti / Vaccine 21 (2003) 2242–2245
its safety. Specific recommendations for the use of the vaccine in a general population should be developed taking into account the local burden of the disease and the cost effectiveness of the many possible immunisation strategies. Future studies on HAV vaccine should be better designed and reported. There is an urgent need for the process of standardising definitions of adverse event following immunisation to be completed and fully implemented in all future studies regardless of the vaccine assessed and of the study design applied.
Acknowledgements We would like to thank Tom Jefferson for arbitration in the review process and Carlo Di Pietrantonj for the statistical advice given. We would also like to thank Alfonso Mele for providing us suggestion based on his extensive knowledge of the subject and the editors of Cochrane Hepatobiliary Review Group and the anonymous referees for the help provided and their constructive comments.
2245
References [1] Innis BL, Snitbhan R, Kunasol P, et al. Protection against hepatitis A by an inactivated vaccine. JAMA 1994;271(17):1328–34. [2] Riedemann S, Reinhardt G, Frosner GG, et al. Placebo-controlled efficacy study of hepatitis A vaccine in Valdivia, Chile. Vaccine 1992;10(Suppl 1):S152–5. [3] Werzberger A, Mensch B, Kuter B, et al. A controlled trial of a formalin-inactivated hepatitis A vaccine in healthy children. N Engl J Med 1992;327(7):453–7. [4] Xu Z, Li R, Meng Z. Immunogenicity and efficacy trials of live attenuated hepatitis A vaccines. Zhonghua Yi Xue Za Zhi 1998;78(4):254–6 [in Chinese]. [5] Sagliocca L, Amoroso P, Stroffolini T. Efficacy of hepatitis A vaccine in prevention of secondary hepatitis A infection: a randomised trial. Lancet 1999;353(9159):1136–9. [6] Ellerbeck EF, Lewis JA, Nalin D, et al. Safety profile and immunogenicity of an inactivated vaccine derived from an attenuated strain of hepatitis A. Vaccine 1992;10(10):668–70. [7] Leentvaar-Kuijpers A, Coutinho RA, Brulein V, Safary A. Simultaneous passive and active immunization against hepatitis A. Vaccine 1992;10(Suppl 1):S138–41. [8] Walter EB, Hornick RB, Poland GA. Concurrent administration of inactivated hepatitis A vaccine with immune globulin in healthy adults. Vaccine 1999;17(11/12):1468–73.