Conjugate vaccines for preventing Haemophilus influenzae type B infections (Review)

Transcription

1 Conjugate vaccines for preventing Haemophilus influenzae type B infections (Review) Swingler G, Fransman D, Hussey G This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2007, Issue 2 1

2 T A B L E O F C O N T E N T S ABSTRACT PLAIN LANGUAGE SUMMARY BACKGROUND OBJECTIVES CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW SEARCH METHODS FOR IDENTIFICATION OF STUDIES METHODS OF THE REVIEW DESCRIPTION OF STUDIES METHODOLOGICAL QUALITY RESULTS DISCUSSION AUTHORS CONCLUSIONS POTENTIAL CONFLICT OF INTEREST ACKNOWLEDGEMENTS SOURCES OF SUPPORT REFERENCES TABLES Characteristics of included studies Characteristics of excluded studies ADDITIONAL TABLES Table 01. Haemophilus influenzae type B conjugate vaccines Table 02. Baseline risk and the number of people to immunise to prevent one case of Hib ANALYSES Comparison 01. Conjugate Hib vaccine versus control Comparison 02. Subgroup analysis: Vaccine type Comparison 03. Subgroup analysis: Number of vaccine doses Comparison 04. Subrroup analysis: Age at first vaccination Comparison 05. Subgroup analysis: High income and low income countries Comparison 06. Sensitivity analysis - Fixed-effect model INDEX TERMS COVER SHEET GRAPHS AND OTHER TABLES Analysis Comparison 01 Conjugate Hib vaccine versus control, Outcome 01 All invasive Hib disease... Analysis Comparison 01 Conjugate Hib vaccine versus control, Outcome 02 Hib specific mortality.... Analysis Comparison 01 Conjugate Hib vaccine versus control, Outcome 03 All cause mortality..... Analysis Comparison 02 Subgroup analysis: Vaccine type, Outcome 01 All invasive Hib disease..... Analysis Comparison 02 Subgroup analysis: Vaccine type, Outcome 02 Hib specific mortality Analysis Comparison 03 Subgroup analysis: Number of vaccine doses, Outcome 01 All invasive Hib disease. Analysis Comparison 03 Subgroup analysis: Number of vaccine doses, Outcome 02 Hib specific mortality. Analysis Comparison 04 Subrroup analysis: Age at first vaccination, Outcome 01 All invasive Hib disease.. Analysis Comparison 04 Subrroup analysis: Age at first vaccination, Outcome 02 Hib specific mortality.. Analysis Comparison 05 Subgroup analysis: High income and low income countries, Outcome 01 All invasive Hib disease Analysis Comparison 05 Subgroup analysis: High income and low income countries, Outcome 02 Hib specific mortality Analysis Comparison 06 Sensitivity analysis - Fixed-effect model, Outcome 01 All invasive Hib disease... Analysis Comparison 06 Sensitivity analysis - Fixed-effect model, Outcome 02 Hib specific mortality... Analysis Comparison 06 Sensitivity analysis - Fixed-effect model, Outcome 03 All cause mortality i

3 Conjugate vaccines for preventing Haemophilus influenzae type B infections (Review) Swingler G, Fransman D, Hussey G Status: Updated This record should be cited as: Swingler G, Fransman D, Hussey G. Conjugate vaccines for preventing Haemophilus influenzae type B infections. Cochrane Database of Systematic Reviews 2007, Issue 2. Art. No.: CD DOI: / CD pub2. This version first published online: 18 April 2007 in Issue 2, Date of most recent substantive amendment: 08 February 2007 A B S T R A C T Background Haemophilus influenzae (H. influenzae) is an important cause of meningitis and pneumonia in children. Vaccine cost is a significant barrier to use in low income countries. Determining the size of the effects of the vaccine will enable cost-effectiveness comparisons with competing priorities in low income countries. Objectives 1. To determine the effects of conjugate Hib vaccine in preventing Hib disease or death in children under five years of age. 2. To determine any variation in effect with type of vaccine, number of doses, age at first dose, in children with known HIV infection, or in high and low income countries. 3. To determine any serious adverse outcomes. Search strategy We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 4, 2006); MEDLINE (January 1966 to December 2006); EMBASE (1990 to June 2006) and scanned reference lists and contacting of authors of trial reports. Reports in all languages were considered. Selection criteria All randomised controlled trials (RCTs) or quasi-rcts of conjugate H. influenzae type b vaccines compared with placebo or no treatment in children who were followed until at least two years of age. Data collection and analysis Two review authors independently selected eligible studies and extracted data. Main results Six studies were included in the review, and four in the meta-analyses. The overall quality of the trials was good. The relative risk for invasive Hib disease was 0.20 (95% confidence interval (CI) 0.07 to 0.54; random-effects model), but there was statistically significant unexplained variation (heterogeneity) in the effects of the four trials in the meta-analysis (P = 0.002). The size of the effects did not appear to differ consistently with different vaccine types, the number of vaccine doses, age at first vaccination or use in high income versus low income countries, but the CIs for the effect estimates were wide. Hib-related mortality data showed a non-significant trend towards benefit (relative risk was 0.29; 95% CI 0.07 to 1.20; random-effects model). The relative risk for all cause mortality in the two trials from which data were available were 1.01 (95% CI 0.38 to 2.67, random-effects model) and No serious adverse effects were reported in any of the trials. Authors conclusions Hib vaccine is safe and effective. In resource-poor settings, decisions to use the vaccine will depend on its cost, the local burden of Hib disease and competing priorities. 1

4 P L A I N L A N G U A G E S U M M A R Y Trials have shown that the Haemophilus influenzae type b (Hib) vaccine can prevent many children becoming severely ill from this disease Haemophilus influenzae (H. influenzae) type b (Hib) usually affects children under the age of five. It can cause life-threatening meningitis (inflammation of the membrane around the brain) or pneumonia (serious lung infection). Hib vaccine has been introduced in high income countries, but the cost of the vaccine has prevented it being introduced into routine childhood immunisation schedules in low income countries. The review found that a large number of children have now been involved in trials of the Hib vaccine. The vaccine can reduce Hib disease and no serious adverse effects have been reported in the trials. B A C K G R O U N D Haemophilus influenzae (H. influenzae) type b (Hib) is an important cause of meningitis and pneumonia in children. In addition, Hib is responsible for other invasive diseases, including epiglottitis, septicaemia, cellulitis, arthritis, osteomyelitis and pericarditis. Of all hospitalisations for H. influenzae infections, 90% are due to Hib (Asensi 1995). Although invasive Hib disease may occur in any age group, it is predominantly a disease of early childhood, being relatively uncommon after five years of age. It is also relatively uncommon in the first two months of life. Due to lack of adequate testing facilities and resources for laboratory diagnosis, especially in low income countries, the true burden of Hib disease is not known. It has been estimated before the introduction of conjugate vaccines to have caused two to three million cases of serious disease and more than half a million deaths annually worldwide, all but 5000 of the deaths occurring in developing countries (Peltola 2000). Several Hib conjugate vaccines have been developed. The first generation vaccines were developed in the early seventies and consisted of the type b capsular polysaccharide, polyribosylribitol phosphate (PRP). They were found to be ineffective in children less than 18 months of age, the age group in which most Hib disease burden occurred. The reason for the poor response in young children was thought to be because bacterial polysaccharides do not generate memory B cells. In an effort to improve the immunogenicity (stimulation of an antibody response) of these vaccines, the capsular polysaccharides were conjugated with protein carriers. Four such vaccines, conjugated with different carrier proteins, have been licensed for use in infants (PRP-T, Hb-OC, PRP-OMP and PRP- D: see Table 01). These conjugate vaccines differ in their immunogenicity, probably related to the choice of the protein carrier and the method of coupling. PRP-T is the most widely used of the four vaccines and PRP-D is no longer used in young infants because of its poor immunogenicity (stimulation of an antibody response). Hib vaccines can be administered separately or in combination with other vaccines. Hib conjugate vaccines have an impact on reducing carriage of the organism in vaccinated children and by reducing the pool of infectious children in the community, and are thus expected to confer a degree of herd immunity by providing protection for unvaccinated children. Since the introduction of conjugate Hib vaccines into routine immunisation schedules in high income countries, there has been a rapid decline in disease occurrence. However, many low income countries have not introduced Hib vaccine into their routine immunisation programmes due to the current cost constraints, lack of information on burden of disease and how best to provide the vaccine cost-effectively (Mahoney 1999). The size of the effect of Hib vaccine is important in determining its cost effectiveness in low income countries, particularly in competition with other priorities. There is a need to determine the size of the effects of the vaccine through a systematic review of the evidence. O B J E C T I V E S 1. To determine the effects of conjugate Hib vaccine in preventing Hib disease or death in children under five years of age. 2. To determine any variation in effect with type of vaccine, number of doses, age at first dose, in children with known HIV infection, or in high and low income countries. 3. To determine any serious adverse outcomes. C R I T E R I A F O R C O N S I D E R I N G S T U D I E S F O R T H I S R E V I E W Types of studies All randomised controlled trials (RCTs) or quasi-rcts in which children were followed up until at least two years of age. Types of participants Children less than five years old, irrespective of HIV status. Types of intervention Conjugate H. influenzae type b vaccines, compared with placebo or no treatment. 2

5 Types of outcome measures 1. All invasive H. influenzae disease (Hib or other), including meningitis, pneumonia, bacteraemia, cellulitis, epiglottitis, arthritis. 2. All definitive Hib disease (clinical and bacteriological or serological confirmation). 3. All-cause mortality. 4. Hib-specific mortality. 5. All-cause mortality from meningitis and pneumonia. 6. All reported serious adverse events of Hib vaccination, defined as any condition which results in any life threatening or disabling reaction. S E A R C H M E T H O D S F O R I D E N T I F I C A T I O N O F S T U D I E S See: Cochrane Acute Respiratory Infections Group methods used in reviews. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 4, 2006); MEDLINE (January 1966 to December 2006); EMBASE (1990 to June 2006) and scanned reference lists and contacting of authors of trial reports. Reports in all languages were considered. MEDLINE was searched using the Cochrane highly sensitive search strategy for RCTs (Higgins 2005) in combination with the following specific terms. MEDLINE (OVID) 1 exp Haemophilus Vaccines/ 2 ((hemophilus or haemophilus) adj vaccin$).mp. 3 ((hemophilus influenzae or haemophilus influenzae) adj vaccin$).mp. [mp=title, original title, abstract, name of substance word, subject heading word] 4 exp Haemophilus influenzae/ 5 exp Haemophilus influenzae type b/ 6 or/1-5 7 exp Vaccines, Conjugate/ 8 conjugate.mp. 9 or/ and 9 A similar search strategy was used for CENTRAL and EMBASE. Reports in all languages were included. We also scanned reference lists of identified articles and contacted trial authors in an attempt to locate additional published and unpublished trials. M E T H O D S O F T H E R E V I E W Two review authors independently screened the results of the literature search and selected eligible studies according to the preset criteria. Differences between authors were resolved by discussion. The following aspects of methodological quality were assessed individually, without using a score: method of allocation, adequacy of concealment of allocation, blinding, duration of follow up of subjects. Random allocation and allocation concealment were taken as the primary measures of study quality. Data extraction was performed independently by two review authors using a pre-piloted data extraction form. Disagreements were resolved by discussion. The aspects of methodological quality listed above and the following additional data were extracted from each report: type of conjugate vaccine (PRP-D, HbOC, PRP- OMP, PRP-T), the nature of the control, number of vaccine doses, target age at first vaccination, HIV status, country in which the trial was conducted, data on pre-specified outcomes listed above and serious adverse effects. When required information was missing, we attempted to contact the trial authors for the information. A quantitative analysis was performed using the intention to treat principle. The chi-squared test for heterogeneity was used to assess the likelihood of heterogeneity (significance set at P < 0.10). Meta-analysis was performed using a random-effects model, with a sensitivity analysis using the fixed-effect model. Relative risks (RR) with 95% CIs were calculated. Pre-specified subgroup analyses were planned to compare the effectiveness according to the type of vaccine used, the number of doses given, age at first vaccination, in children with known HIV infection status and in high income versus low income countries. In an attempt to assess the impact of herd immunity, a further subgroup analysis was planned comparing vaccine effectiveness in trials of individually randomised children with that in cluster randomised trials. D E S C R I P T I O N O F S T U D I E S After elimination of duplicates, 407 potentially eligible studies were identified from the original literature search. Forty-six studies did not deal with Hib; 131 dealt with safety and/or immunogenicity only; 70 with vaccine combinations; 40 were review articles; 24 dealt with the post-licensure impact of Hib; 23 with physiological aspects; and 62 with other issues not covered by the inclusion criteria. Eleven trials were considered more closely for inclusion. Six of these were excluded, and the remaining five studies were included in the review (Eskola 1990; Lagos 1996; Mulholland 1997; Santosham 1991; Ward 1990). Two out of six researchers replied to a request for information, without additional eligible trials being identified. The two researchers who responded also provided additional unpublished information from their studies (Lagos 1996; Santosham 1991). Of our pre-specified outcomes, data were available for three (invasive Hib disease, Hib-specific mortality and all-cause mortality). 3

6 In the 2006 update search, 470 potentially eligible studies were identified, and three were considered more closely for inclusion. One was included in the review (Gessner 2005). One trial assembled two sets of a total of 71 clinics and then randomised one set to receive immunisation and the other set no treatment other than standard immunisation (Lagos 1996). This effectively constituted block randomisation of only two units, and analysis by the number of individuals in the groups would have provided a spuriously narrow CI and given the trial undue weight in the meta-analysis. Information on intra-cluster correlation, to enable adjustment for the cluster effect, was not available from the trial authors. A second cluster randomised controlled trial studied 55,073 children in 818 hamlets. Findings were expressed as incidence per 100, 000 child-years of observation, after statistical adjustment for within cluster correlation. Information from the trial authors on intra-cluster correlation is awaited. M E T H O D O L O G I C A L Q U A L I T Y The overall quality of the trials was good. Five trials were randomised with adequate allocation concealment. The remaining trial was quasi-randomised in that allocation to treatment and control was on alternate days. Two trials were cluster randomised. Four of the trials were double-blind, placebo-controlled and one (quasi-randomised) was not blinded. The remaining trial was cluster randomised and not placebo-controlled, but assessment of outcome was blinded. Completeness of follow up could not be assessed because outcomes were measured by case detection using community surveillance. R E S U L T S Invasive Hib disease was reported in all six studies included in the meta-analysis. Four studies were included in the meta-analysis. Two cluster randomised trials (Gessner 2005; Lagos 1996) was not included because information on intra-cluster correlation, to enable adjustment for the cluster effect, was not available from the trial authors and inclusion without adjustment would have resulted in spuriously narrow CIs. One of the trials (Gessner 2005) will be included in the meta-analysis when the additional data become available. There was statistically significant heterogeneity in the outcomes of the four trials (P = 0.002). The heterogeneity was due to one trial (Ward 1990). Using a random-effects model the relative risk for invasive Hib disease was 0.20 (95% CI 0.07 to 0.54). A sensitivity analysis using the fixed-effect model produced very similar findings (relative risk 0.19, 95% Cl 0.13 to 0.28). No significant difference was demonstrated in the effect of different vaccine types, the number of vaccine doses, age at first vaccination or use in high versus low income countries, as judged by the overlapping CIs of sub-groups in each analysis. No data on HIV status were available for the subgroup analysis comparing the effect in HIV-infected and non-infected people. Hib-related mortality data were available from two trials included in the meta-analysis and showed a non-significant trend towards benefit (relative risk was 0.29; 95% CI 0.07 to 1.20; randomeffects model). The relative risk for all-cause mortality in the two trials from which data were available were 1.01 (95% CI 0.38 to 2.67, random-effects model) and One of the two cluster randomised trials (Lagos 1996) found an effect compatible with that of the other four trials for both invasive Hib disease (4/48080 with Hib versus 40/46948 with control, relative risk 0.10 for the single trial, compared with 0.20 (95% CI 0.07 to 0.54) for the meta-analysis) and for Hib-specific mortality (2/48080 with Hib versus 3/4698 with control, relative risk 0.65 for the Lagos trial compared with 0.29 (95% CI 0.07 to 1.20) for the meta-analysis). The second cluster randomised trial (Gessner 2005) found a relative risk of 0.97 for all cause mortality (absolute reduction in incidence rate 59 per 100,000 child years, 95% CI -249 to +367), similar to the relative risk of 1.01 (95% CI 0.38 to 2.67) for the only other trial reporting this outcome (Santosham 1991). No serious adverse effects were reported in any of the trials, involving a total of 257,000 infants. D I S C U S S I O N The findings of RCTs suggest an 80% reduction in invasive Hib disease as a result of vaccination, although the 95% CI suggests that the size of the effect could plausibly be anywhere between a 46% and a 93% reduction in Hib invasive disease, before the effect of herd immunity is taken into account. There was, however, significant variation (heterogeneity) in the estimates of effect in different trials. This heterogeneity does not affect the conclusion that the vaccine is effective in preventing invasive Hib disease, but does mean that the size of the summary estimate of effect should be regarded with caution. The effectiveness of the vaccine in preventing Hib disease has been confirmed in post-marketing surveillance. Since the introduction of the conjugate vaccines in the early 1990s, the incidence of Hib disease has declined dramatically in high income countries. In fact, the disease has been virtually eliminated in parts of the USA and Europe (Adams 1993; CDC 1996; Peltola 1992). Herd immunity - the protection of communities from infection that is brought about by the presence of immune individuals in the population, an effect that reduces the proportion of infectious individuals in the community, may play an important part in the effectiveness of Hib vaccines. Hib vaccines reduce carriage of the organism in vaccinated children and should thus provide protection for unvaccinated children since they are less likely to be ex- 4

7 posed to infection (Barbour 1995). The impact of the introduction of the vaccine into communities has also been greater than expected if only vaccinated children were protected (Moulton 2000). It was not possible to assess a herd effect in the included trials in this review. Four of the six included studies allocated individual children and not communities to vaccine or control. The other two allocated the vaccine by clinic or hamlet respectively, with a potential herd effect if the clinics had served circumscribed communities. The effect estimates in these trials were similar to the overall estimate, but the uncertainty around the estimates means that a herd effect was neither demonstrated nor excluded. We found no evidence from the included RCTs that the effectiveness of the vaccine was modified by the type of conjugate vaccine, the number of doses given (two, three or four), age at first vaccination (2 months, 42 to 90 days, 3 months) and whether the vaccine was tested in a high income or low income country. The small sample sizes and consequent wide CIs do not, however, preclude any such differences in effects. In the absence of such evidence, other factors such as differences in the magnitude and timing of the immune response promoted by the different vaccines may be important in deciding the choice of vaccine type. We were unable to explain the discordant finding of a smaller nonsignificant effect in one trial (Ward 1990). The subgroup analyses failed to identify a reason for the heterogeneity and there was no apparent difference in the design and conduct of the trial. Differences in the degree or type of Hib exposure, or differences in susceptibility to disease, appear unlikely explanations because a subsequent trial in a similar population to the Ward trial (Santosham 1991) found an effect similar to that of the other trials. The vaccine type also appears an unlikely explanation, because the other trial of PRP-D (Eskola 1990) found an effect similar to that of trials of other types. For Hib-related mortality there was a strong trend towards benefit, with an estimate of approximately a 70% reduction, but the wide CI does not exclude a harmful effect. The wide CI reflects the paucity of information available, with only two of five trials reporting this outcome. Two of the six included trials reported mortality from all Hib diseases, while authors of another two trials provided mortality data on request. Great effort is expended to perform vaccine field trials, to establish the efficacy of the vaccine in actually preventing disease. This includes setting up a surveillance system for a specific disease. Given this effort, it appears important to plan, in each case, to measure mortality at the same time. It is also important to report such mortality, if measured, to enable meta-analysis of findings of trials that may be too small to detect differences in rare events. Decisions regarding the use of the vaccine in poorer countries will depend not only on the effectiveness of the vaccine but also on factors such as vaccine cost and burden of disease, including baseline risk of individual populations, and information on how best to provide the vaccine cost-effectively. As illustrative examples, the number of children needing to be immunised to prevent one case of invasive Hib disease in each of the trials is presented in Table 02, with 95% CIs where they can be calculated. In the studies finding a statistically significant effect, the estimated number needed to immunise ranged from 124 (Alaska) to 1307 (Chile). In summary, the findings suggest a large beneficial effect of conjugate Hib vaccine on invasive Hib disease, but the size of the effect is uncertain because of variation of effect sizes in different trials. No statistically significant effect was found on Hib-related mortality, although there was a clinically meaningful trend towards benefit. Insufficient evidence was found of an effect on all cause mortality. No important adverse effects were reported. A U T H O R S Implications for practice C O N C L U S I O N S Hib vaccine is safe and effective in preventing invasive Hib disease. The findings of RCTs suggest an 80% reduction in Hib invasive disease, although the size of the effect could plausibly be anywhere between a 46% and 93%. In resource-poor settings, decisions regarding the use of the vaccine will depend on factors such as vaccine cost and local burden of disease, and information on how best to provide the vaccine cost-effectively (Mahoney 1999). Implications for research The cost-effectiveness of the vaccine in resource-poor settings needs investigation. P O T E N T I A L C O N F L I C T O F I N T E R E S T Greg Hussey and Des Michaels (Fransman) have been involved in contract research for conducting immunogenicity Hib vaccination trials. There was no financial gain from conducting the systematic review. A C K N O W L E D G E M E N T S The Cochrane Vaccines Field (Antonella Barale and Vittorio Demicheli) gave advice on cluster randomisation. Elizabeth Pienaar (South African Cochrane Centre) assisted with an earlier literature search. Paul Garner assisted with protocol design. We wish to thank the following people for commenting on the draft of this updated review: Rebecca Coghlan, Mahomed Patel, Mark Jones, and Peter Morris. 5

8 S O U R C E S O F S U P P O R T External sources of support No sources of support supplied Internal sources of support No sources of support supplied R E F E R E N C E S References to studies included in this review Eskola 1990 {published data only} Eskola J, Kayhty H, Takala A, Peltola H, Ronnberg P, et al. Randomized prospectiive field trial of a conjugate vaccine in the protection of infants and young children against invasive Haemophilus influenza type B disease. New England Journal of Medicine 1990;323 (20): Gessner 2005 {published data only} Gessner BD, Sutanto A, Liniehan M, Djelantik IG, Fletcher T, Gerudug IK, et al. Incidences of vaccine-preventable Haemophilus influenzae type b pneumonia and meningitis in Indonesian children: hamlet-randomised vaccine-probe trial. Lancet 2005;365(9453): Lagos 1996 {published and unpublished data} Lagos R, Horwitz I, Toro J, San Martin O, Abrego P, Bustamante C, et al. Large scale, postlicensure, selective vaccination of Chilean infants with PRP-T conjugate vaccine: practicality and effectiveness in preventing invasive Haemophilus influenzae type b infections. Pediatic Infectious Disease Journal 1996; Vol. 15, issue 3: Mulholland 1997 {published data only} Mulholland K, Hilton S, Adegbola R, Usen S, Oparaugo A, Omosigho C, et al. Randomised trial of Haemophilus influenzae type-b tetanus protein conjugate for prevention of pneumonia and meningitis in Gambian infants. Lancet 1997;349(9060): Santosham 1991 {published and unpublished data} Santosham M, Wolff M, Reid R, Hohenboken M, Bateman M, et al. The efficacy in Navajo infants of a conjugate vaccine consisting of Haemophilus influenza type B polysaccharide and Neisseria meningitidis outer-membrane protein complex. New England Journal of Medicine 1991;324(25): Ward 1990 {published data only} Ward J, Brenneman G, Letson GW, Heyward W L. Limited efficacy of a Haemophilus influenzae type b conjugate vaccine in Alaska Native infants. The Alaska H. influenzae Vaccine Study Group. New England Journal of Medicine 1990;323(20): References to studies excluded from this review Black 1991 Black SB, Shinefield HR, Fireman B, Hiatt R, Polen M, Vittinghoff E. Efficacy in infancy of oligosaccharide conjugate Haemophilus influenzae type b (HbOC) vaccine in a United States population of 61,080 children. The Northern California Kaiser Permanente Vaccine Study Center Pediatrics Group. Pediatic Infectious Disease Journal 1991;10: Booy 1994 Booy R, Hodgson L, Mayon-White RT, Slack MPE, Macfarlane JA, Haworth EA, et al. Efficacy of Haemophilus influenzae type b conjugate vaccine PRP-T. Lancet 1994;344: Eskola 1987a Eskola J, Peltola H, Takala AK, Kayhty H, Hakulinen M, Karanko V, et al. Efficacy of Haemophilus influenzae type b polysaccharidediphtheria toxoid conjugate vaccine in infancy. New England Journal of Medicine 1987;317(12): Eskola 1987b Eskola J, Peltola H, Takala AK, Kayhty H, Hakulinen M, Karanko V, et al. Antibody levels achieved in infants with haemophilus vaccine [Imev_isille annetun hemofilusrokotteen suojateho]. Suomen Laakarilehti 1987;42(16): Levine 1999 Levine OS, Lagos R, Munoz A, Villaroel J, Alvarez AM, Abrego P, et al. Defining the burden of pneumonia in children preventable by vaccination against Haemophilus influenzae type b. Pediatric Infectious Disease Journal 1999;18: Peltola 1994 Peltola H, Eskola J, Kayhty H, Takala AK, Makela H. Clinical comparison of the Haemophilus influenzae type B polysaccharide-diphtheria toxoid and the oligosaccharide-crm197 protein vaccines in infancy. Archives of Pediatrics and Adolescent Medicine 1994;148: Prymula 2006 Prymula R, Peeters P, Chrobok V, Kriz P, Novakova E, Kaliskova E, et al. Pneumococcal capsular polysaccharides conjugated to protein D for prevention of acute otitis media caused by both Streptococcus pneumoniae and non-typable Haemophilus influenzae: a randomised double-blind efficacy study. Lancet 2006;367: Vadheim 1993 Vadheim CM, Greenberg DP, Partridge S, Jing J, Ward JI. Effectiveness and safety of an Haemophilus influenzae type b conjugate vaccine (PRP-T) in young infants. Kaiser-UCLA Vaccine Study Group. Pediatrics 1993;92(2): Van Alphen 1995 Van Alphen L, Spanjaard L, van der Ende A, Dankert J. Absence of meningitis caused by Haemophilus influenzae type b in The Netherlands following twofold vaccination [Uitblijven van meningitis door Haemophilus influenzae type b in Nederland na tweevoudige vaccinatie]. Nederlands Tijdschrift voor Geneeskunde 1995;139:

9 Additional references Adams 1993 Adams WG, Deaver KA, Cochi SL, Plikaytis BD, Zell ER, Broome CV, et al. Decline in Haemophilus influenzae type b (Hib) diseases in the Hib vaccine era. JAMA 1993;269(2): Asensi 1995 Asensi F, Otero MC, Perez-Tamarit D, Miranda J, Pico L, Nieto A. Economic aspects of a general vaccination against invasive disease caused by Haemophilus inflkuenzae Type B (Hib) via the experience of the Children s Hospital La Fe, Valencia, Spain. Vaccine 1995;13 (16): Barbour 1995 Barbour ML, Mayon-White RT, Coles C, Crook DW, Moxon ER. The impact of conjugate vaccines on carriage of Haemophilus influenzae type b. Journal of Infectious Diseases 1995;171(1):93 8. CDC 1996 Center for Disease Control. Progress toward elimination of Haemophilus influenzae type b disease among infants and children - United States, MMWR 1996;45:901. Higgins 2005 Higgins JPT, Green S, editors. Highly sensitive search strategies for identifying reports of randomized controlled trials in MEDLINE. Cochrane Handbook for Systematic Reviews of Interventions [updated May 2005]; APPENDIX 5b. The Cochrane Library, Issue 3. Chichester, UK: John Wiley & Sons, Ltd, Mahoney 1999 Mahoney RT, Maynard JE. The introduction of new vaccines into developing countries. Vaccine 1999;17(7-8): Moulton 2000 Moulton LH, Chung S, Croll J, Reid R, Weatherholtz RC, Santosham M. Estimation of the indirect effect of Haemophilus influenzae type b conjugate vaccine in an American Indian population. International Journal of Epidemiology 2000;29(4): Peltola 1992 Peltola H, Kilpi T, Anttila M. Rapid disappearance of Haemophilus influenzae type b meningitis after routine childhood immunization with conjugate vaccines. Lancet 1992;340: Peltola 2000 Peltola H. Worldwide Haemophilus influenzae type b disease at the beginning of the 21st century: global analysis of the disease burden 25 years after the use of polysaccharide vaccine and a decade after the advent of conjugates. Clinical Microbiology Reviews 2000;13(2): References to other published versions of this review Swingler 2003 Swingler G, Fransman D, Hussey G. Conjugate vaccines for preventing Haemophilus influenzae type b infections. In: Cochrane Database of Systematic Reviews, 4, T A B L E S Characteristics of included studies Study Eskola 1990 Methods Participants Interventions Outcomes Notes Allocation concealment Quasi-randomised (alternate days). Unblinded, no placebo 114,000 infants in Finland PRP-D (ProhibiT) conjugate Hib at 3, 4, 6 and 14 to 18 months (4 doses) Control: conjugate Hib at 24 months All invasive HiB disease C Inadequate Study Gessner 2005 Methods Participants Interventions Outcomes Notes Allocation concealment Cluster randomised controlled trial. Placebo controlled. Outcome assessment blinded 55,073 infants in villages in Indonesia. Not tested for HIV PRP-T (Acthib) conjugate Hib at 6, 10 and 14 weeks (3 doses) Control: nothing All cause mortality No unpublished information yet from Gessner (author of the new included trial) A Adequate 7

10 Characteristics of included studies (Continued ) Study Lagos 1996 Methods Participants Interventions Outcomes Notes Allocation concealment Cluster randomised controlled trial. No placebo. Outcome assessment blinded 93,028 infants in Santiago, Chile. Not tested for HIV PRP-T (Acthib) conjugate Hib at 2, 4, 6 months (3 doses) Control: nothing All invasive Hib disease A Adequate Study Mulholland 1997 Methods Participants Interventions Outcomes Notes Allocation concealment Randomised controlled trial. Double blind placebo controlled 42,848 infants in Gambia PRP-T (Actihib) conjugate Hib vaccine at 2, 3, 4 months (3 doses) Control: nothing All invasive Hib disease Hib specific mortality A Adequate Study Santosham 1991 Methods Participants Interventions Outcomes Notes Allocation concealment Randomized controlled trial. Double blind placebo controlled 5190 Navajo infants in Alaska No cases with HIV infection PRP-OMP (PedvaxHib) Conjugate Hib vaccine at 42 to 90 days and 70 to 146 days (2 doses) Control: nothing All invasive Hib disease All cause mortality Hib specific mortality A Adequate Study Ward 1990 Methods Participants Interventions Outcomes Notes Allocation concealment Randomized controlled trial. Double blind placebo controlled 2102 Alaskan infants PRP-D (ProhibiT) Conjugate Hib vaccine at 2, 4, and 6 months (3 doses) Control: saline placebo All invasive Hib disease A Adequate 8

11 Characteristics of excluded studies Study Black 1991 Booy 1994 Eskola 1987a Eskola 1987b Levine 1999 Peltola 1994 Prymula 2006 Vadheim 1993 Van Alphen 1995 Reason for exclusion Allocation not random. (Those refusing consent were included in the control group). Data from an intention-totreat analysis were not reported Not random or quasi-random allocation. Prospective controlled trial Earlier report of the included trial Earlier report of the included trial Supplementary report of an included trial, without additional information pre-specified in the review Comparison of two Hib vaccines. No comparison with placebo Both intervention and control groups received conjugate Hib vaccine Hepatitis B vaccine as control. Follow up stopped before two years of age. Controls received Hib immunisation before the end of the surveillance period. Allocation not random (before-after study) A D D I T I O N A L T A B L E S Table 01. Haemophilus influenzae type B conjugate vaccines Abbreviation Carrier protein Trade name PRP-T Tetanus toxoid ActHIB Hb-OC CRM197 (non-toxic mutant diphtheria toxin) HIBTITER PRP-OMP Outer membrane protein of N. meningitidis Pedvax PRP-D Diphtheria toxoid Prohibit Table 02. Baseline risk and the number of people to immunise to prevent one case of Hib Country Relative risk Baseline / Number to treat 95% CI Chile (Lagos) Finland (Eskola) to 1145 Gambia (Mulholland) to 1315 Alaska (Santosham) to 223 Alaska (Ward)

12 Comparison 01. Conjugate Hib vaccine versus control A N A L Y S E S Outcome title No. of No. of studies participants Statistical method Effect size 01 All invasive Hib disease Relative Risk (Random) 95% CI 0.20 [0.07, 0.54] 02 Hib specific mortality Relative Risk (Random) 95% CI 0.29 [0.07, 1.20] 03 All cause mortality Relative Risk (Random) 95% CI Totals not selected Comparison 02. Subgroup analysis: Vaccine type Outcome title No. of No. of studies participants Statistical method Effect size 01 All invasive Hib disease Relative Risk (Random) 95% CI Subtotals only 02 Hib specific mortality Relative Risk (Random) 95% CI Totals not selected Comparison 03. Subgroup analysis: Number of vaccine doses Outcome title No. of No. of studies participants Statistical method Effect size 01 All invasive Hib disease Relative Risk (Random) 95% CI Subtotals only 02 Hib specific mortality Relative Risk (Random) 95% CI Totals not selected Comparison 04. Subrroup analysis: Age at first vaccination Outcome title No. of No. of studies participants Statistical method Effect size 01 All invasive Hib disease Relative Risk (Random) 95% CI Subtotals only 02 Hib specific mortality Relative Risk (Random) 95% CI Subtotals only Comparison 05. Subgroup analysis: High income and low income countries Outcome title No. of No. of studies participants Statistical method Effect size 01 All invasive Hib disease Relative Risk (Random) 95% CI Subtotals only 02 Hib specific mortality Relative Risk (Random) 95% CI Totals not selected Comparison 06. Sensitivity analysis - Fixed-effect model Outcome title No. of No. of studies participants Statistical method Effect size 01 All invasive Hib disease Relative Risk (Fixed) 95% CI 0.19 [0.13, 0.28] 02 Hib specific mortality Relative Risk (Fixed) 95% CI 0.29 [0.07, 1.20] 03 All cause mortality Relative Risk (Fixed) 95% CI Totals not selected Medical Subject Headings (MeSH) I N D E X T E R M S Developing Countries; Haemophilus Infections [ prevention & control]; Haemophilus influenzae type b; Haemophilus Vaccines [ therapeutic use]; Polysaccharides, Bacterial [ therapeutic use]; Randomized Controlled Trials; Vaccines, Conjugate [therapeutic use] MeSH check words Humans 10

13 C O V E R S H E E T Title Authors Contribution of author(s) Conjugate vaccines for preventing Haemophilus influenzae type B infections Swingler G, Fransman D, Hussey G George Swingler contributed to the protocol design and literature search, selected studies for inclusion and extracted data, assisted with data entry, performed the analysis and wrote the first draft of the full review. Des Michaels contributed to the design of the protocol and wrote the first draft of the protocol, located copies of study reports, performed data entry and assisted with the analysis. Gregory Hussey initiated the review, contributed to protocol design, selected studies for inclusion and extracted data. All authors participated in the update of the review and contributed to the writing of the report. Issue protocol first published 1999/3 Review first published 2003/4 Date of most recent amendment 16 February 2007 Date of most recent SUBSTANTIVE amendment What s New Date new studies sought but none found Date new studies found but not yet included/excluded Date new studies found and included/excluded Date authors conclusions section amended Contact address DOI Cochrane Library number Editorial group Editorial group code 08 February 2007 In this 2006 update we included one additional trial reporting all cause mortality. No change to conclusions. Information not supplied by author Information not supplied by author 07 December 2006 Information not supplied by author George Swingler Professor of Paediatrics and Child Health School of Child and Adolescent Health University of Cape Town, ICH Building, Red Cross Childlren s Hospital Klipfontein Road Rondebosch Cape Town 7700 SOUTH AFRICA George.Swingler@uct.ac.za Tel: Fax: / CD pub2 CD Cochrane Acute Respiratory Infections Group HM-ARI 11

14 G R A P H S A N D O T H E R T A B L E S Analysis Comparison 01 Conjugate Hib vaccine versus control, Outcome 01 All invasive Hib disease Review: Conjugate vaccines for preventing Haemophilus influenzae type B infections Comparison: 01 Conjugate Hib vaccine versus control Outcome: 01 All invasive Hib disease Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) n/n n/n 95% CI (%) 95% CI Eskola / / [ 0.05, 0.21 ] Mulholland / / [ 0.12, 0.50 ] Santosham / / [ 0.01, 0.34 ] Ward / / [ 0.30, 1.47 ] Total (95% CI) [ 0.07, 0.54 ] Total events: 29 (Vaccine), 149 (Control) Test for heterogeneity chi-square=14.70 df=3 p=0.002 I² =79.6% Test for overall effect z=3.17 p=0.002 Analysis Comparison 01 Conjugate Hib vaccine versus control, Outcome 02 Hib specific mortality Review: Conjugate vaccines for preventing Haemophilus influenzae type B infections Comparison: 01 Conjugate Hib vaccine versus control Outcome: 02 Hib specific mortality Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) n/n n/n 95% CI (%) 95% CI Mulholland / / [ 0.06, 1.37 ] Santosham /2588 1/ [ 0.01, 8.22 ] Total (95% CI) [ 0.07, 1.20 ] Total events: 2 (Vaccine), 8 (Control) Test for heterogeneity chi-square=0.01 df=1 p=0.93 I² =0.0% Test for overall effect z=1.70 p=

15 Analysis Comparison 01 Conjugate Hib vaccine versus control, Outcome 03 All cause mortality Review: Conjugate vaccines for preventing Haemophilus influenzae type B infections Comparison: 01 Conjugate Hib vaccine versus control Outcome: 03 All cause mortality Study Vaccine Control Relative Risk (Random) Relative Risk (Random) n/n n/n 95% CI 95% CI Santosham /2588 8/ [ 0.38, 2.67 ] Analysis Comparison 02 Subgroup analysis: Vaccine type, Outcome 01 All invasive Hib disease Review: Conjugate vaccines for preventing Haemophilus influenzae type B infections Comparison: 02 Subgroup analysis: Vaccine type Outcome: 01 All invasive Hib disease Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) n/n n/n 95% CI (%) 95% CI 01 PRP-D (ProhibiT) versus control Eskola / / [ 0.05, 0.21 ] Ward / / [ 0.30, 1.47 ] Subtotal (95% CI) [ 0.04, 1.68 ] Total events: 18 (Vaccine), 87 (Control) Test for heterogeneity chi-square=11.91 df=1 p= I² =91.6% Test for overall effect z=1.42 p= PRP-OMP (PedvaxHib) versus control Santosham / / [ 0.01, 0.34 ] Subtotal (95% CI) [ 0.01, 0.34 ] Total events: 1 (Vaccine), 22 (Control) Test for heterogeneity: not applicable Test for overall effect z=3.02 p= PRP-T (ActHib) versus control Mulholland / / [ 0.12, 0.50 ] Subtotal (95% CI) [ 0.12, 0.50 ] Total events: 10 (Vaccine), 40 (Control) Test for heterogeneity: not applicable Test for overall effect z=3.94 p=

16 Analysis Comparison 02 Subgroup analysis: Vaccine type, Outcome 02 Hib specific mortality Review: Conjugate vaccines for preventing Haemophilus influenzae type B infections Comparison: 02 Subgroup analysis: Vaccine type Outcome: 02 Hib specific mortality Study Vaccine Control Relative Risk (Random) Relative Risk (Random) n/n n/n 95% CI 95% CI 01 PRP-OMP (PedvaxHib) versus control Santosham /2588 1/ [ 0.01, 8.22 ] 02 PRP-T (ActHib) versus control Mulholland / / [ 0.06, 1.37 ] Analysis Review: Comparison: Outcome: Comparison 03 Subgroup analysis: Number of vaccine doses, Outcome 01 All invasive Hib disease Conjugate vaccines for preventing Haemophilus influenzae type B infections 03 Subgroup analysis: Number of vaccine doses 01 All invasive Hib disease Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) n/n n/n 95% CI (%) 95% CI 01 Two doses Santosham / / [ 0.01, 0.34 ] Subtotal (95% CI) [ 0.01, 0.34 ] Total events: 1 (Vaccine), 22 (Control) Test for heterogeneity: not applicable Test for overall effect z=3.02 p= Three doses Mulholland / / [ 0.12, 0.50 ] Ward / / [ 0.30, 1.47 ] Subtotal (95% CI) [ 0.15, 1.04 ] Total events: 20 (Vaccine), 55 (Control) Test for heterogeneity chi-square=3.36 df=1 p=0.07 I² =70.2% Test for overall effect z=1.87 p= Four doses Eskola / / [ 0.05, 0.21 ] Subtotal (95% CI) [ 0.05, 0.21 ] Total events: 8 (Vaccine), 72 (Control) Test for heterogeneity: not applicable Test for overall effect z=6.09 p<

17 Analysis Review: Comparison: Outcome: Comparison 03 Subgroup analysis: Number of vaccine doses, Outcome 02 Hib specific mortality Conjugate vaccines for preventing Haemophilus influenzae type B infections 03 Subgroup analysis: Number of vaccine doses 02 Hib specific mortality Study Vaccine Control Relative Risk (Random) Relative Risk (Random) n/n n/n 95% CI 95% CI 01 Two doses Santosham /2588 1/ [ 0.01, 8.22 ] 02 Three doses Mulholland / / [ 0.06, 1.37 ] Analysis Review: Comparison: Outcome: Comparison 04 Subrroup analysis: Age at first vaccination, Outcome 01 All invasive Hib disease Conjugate vaccines for preventing Haemophilus influenzae type B infections 04 Subrroup analysis: Age at first vaccination 01 All invasive Hib disease Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) n/n n/n 95% CI (%) 95% CI 01 2 months of age Mulholland / / [ 0.12, 0.50 ] Ward / / [ 0.30, 1.47 ] Subtotal (95% CI) [ 0.15, 1.04 ] Total events: 20 (Vaccine), 55 (Control) Test for heterogeneity chi-square=3.36 df=1 p=0.07 I² =70.2% Test for overall effect z=1.87 p= to 90 days of age Santosham / / [ 0.01, 0.34 ] Subtotal (95% CI) [ 0.01, 0.34 ] Total events: 1 (Vaccine), 22 (Control) Test for heterogeneity: not applicable Test for overall effect z=3.02 p= months of age Eskola / / [ 0.05, 0.21 ] Subtotal (95% CI) [ 0.05, 0.21 ] Total events: 8 (Vaccine), 72 (Control) Test for heterogeneity: not applicable Test for overall effect z=6.09 p<

18 Analysis Comparison 04 Subrroup analysis: Age at first vaccination, Outcome 02 Hib specific mortality Review: Conjugate vaccines for preventing Haemophilus influenzae type B infections Comparison: 04 Subrroup analysis: Age at first vaccination Outcome: 02 Hib specific mortality Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) n/n n/n 95% CI (%) 95% CI 01 2 months of age Lagos / / [ 0.11, 3.90 ] Mulholland / / [ 0.12, 0.50 ] Subtotal (95% CI) [ 0.15, 0.54 ] Total events: 12 (Vaccine), 43 (Control) Test for heterogeneity chi-square=0.97 df=1 p=0.32 I² =0.0% Test for overall effect z=3.84 p= to 90 days of age Santosham / / [ 0.01, 0.34 ] Subtotal (95% CI) [ 0.01, 0.34 ] Total events: 1 (Vaccine), 22 (Control) Test for heterogeneity: not applicable Test for overall effect z=3.02 p=0.003 Analysis Review: Comparison: Outcome: Comparison 05 Subgroup analysis: High income and low income countries, Outcome 01 All invasive Hib disease Conjugate vaccines for preventing Haemophilus influenzae type B infections 05 Subgroup analysis: High income and low income countries 01 All invasive Hib disease Study Vaccine Control Relative Risk (Random) Weight Relative Risk (Random) 01 High income n/n n/n 95% CI (%) 95% CI Eskola / / [ 0.05, 0.21 ] Santosham / / [ 0.01, 0.34 ] Ward / / [ 0.30, 1.47 ] Subtotal (95% CI) [ 0.03, 0.83 ] Total events: 19 (Vaccine), 109 (Control) Test for heterogeneity chi-square=14.77 df=2 p= I² =86.5% Test for overall effect z= Low income p=0.03 (Continued... ) 16