Description of the condition
Pertussis (whooping cough) is a vaccine-preventable, acute bacterial respiratory infection caused by the organism Bordetella pertussis (B. pertussis) and is an infection of global public health importance. It is characterised by a protracted coughing illness and can result in severe complications including death. In many countries, pertussis vaccine is offered as part of a routine childhood immunisation programme (e.g. two, three and four months in the UK; and two, four and six months in the USA. The youngest age of immunisation in any country for pertussis-containing vaccine is six weeks). However, protection against infection is not life-long and the disease continues to show cyclical peaks in activity every three to four years despite high vaccine coverage. Whilst adolescents and adults tend to display milder symptoms, young infants less than six months of age remain at a disproportionate risk of severe or complicated infections (Amirthalingam 2013a).
Many countries with long-standing vaccination programmes have reported a resurgence of pertussis, particularly in adolescents and adults (Quinn 2007; Skowronski 2002; Tan 2005) and young infants less than six months of age (Somerville 2007; Spokes 2010; Tanaka 2003), despite sustained high vaccine coverage. This has led to a growing international debate on the potential strategies to optimise global pertussis control. A 2010 review by the Strategic Group of Experts in Immunisation on pertussis control strategies recommended a booster dose for children aged one to six years, preferably during the second year of life, following completion of the primary infant schedule (WHO 2010). Although a number of countries, including France, the USA and Australia, have recommended adolescent boosters and cocooning (vaccinating close household contacts of young infants), data to support the introduction of neonatal pertussis immunisation remain inconclusive (WHO 2010). In 2011, the USA became the first country to advise that pertussis-containing vaccine can be safely administered to pregnant women who have not previously received the recommended adult dose (ACIP 2010). This advice was updated in October 2012 to recommend that pertussis-containing vaccine be routinely offered to women in every pregnancy (ACIP 2013).
In England and Wales, an increase in laboratory-confirmed cases of pertussis amongst those aged 15 years or older was observed during July to September of 2011, which differed from the usual age distribution of cases. The increase in cases has continued into 2012 and extended to affect very young infants with an unusually high number of infant deaths (14 infant deaths in 2012 compared with average of 4.4 infant deaths in the previous four years (2008 to 2011)). This has prompted a review of current vaccination strategies, with the aim of providing further protection for those infants who are too young to benefit from existing immunisation programmes (Amirthalingam 2013b).
Description of the intervention
Several vaccination strategies such as an adolescent or adult booster programme, and neonatal and maternal immunisation (post-partum and antenatal) have been considered, some of which have already been implemented in other countries (WHO 2010). A recent review, conducted by the World Health Organization's (WHO) Strategic Advisory Group of Experts on Immunization, concluded that there was insufficient evidence that the introduction of booster doses for adolescents provides indirect protection for young infants (WHO 2010). A cross-sectional study carried out in the USA showed little impact of post-partum vaccination on laboratory-diagnosed cases of pertussis infection in infants of six months of age and under, pre- and post-introduction of a maternal vaccination programme (Castagnini 2012). More targeted strategies including neonatal and antenatal maternal vaccination programmes are being considered for the protection of infants against pertussis.
How the intervention might work
Data from the pre-vaccine era suggest that maternal antibodies may provide at least short-term protection for newborn infants, with a lower proportion of deaths observed in children less than one month of age compared with those aged one to three months (Sako 1945). This pattern of mortality is no longer typically observed and likely reflects a reduction in maternal antibody following the introduction of mass vaccination programmes and decreased immunity from natural infection (Van Rie 2005). All subclasses of immunoglobulin (Ig) G are transferred from mother to infant across the placenta, primarily during the third trimester of pregnancy (Van Rie 2005). Vaccination in the third trimester of pregnancy, when concerns about fetal loss and development are reduced (Healy 2006a), may boost maternal antibody levels to a sufficiently high level to result in indirect infant protection following transplacental transfer. It may also offer indirect protection for newborn infants through reduction of maternal illness. Transplacental transfer of pertussis IgG antibody has already been demonstrated, with concentrations in newborn (Healy 2004; Van Savage 1990) or cord serum samples (Gonik 2005; Healy 2004; Healy 2006b) reflective of those in the mother. Indeed, higher concentrations of pertussis antibodies have been demonstrated in cord blood for newborn infants of vaccinated when compared with non-vaccinated mothers (Gall 2011; Leuridan 2011). Maternal antibodies are thought to have a half-life of approximately six weeks and so if boosted to sufficiently high levels, are likely to provide time-limited, passive protection for newborn infants prior to administration of the first childhood immunisation and during the period when they are most vulnerable to severe infection (Healy 2006a; Van Savage 1990). Indeed, antenatal maternal immunisation programmes for other vaccine-preventable diseases, such as tetanus, are well established (WHO 2010) and have been shown to be effective in terms of reducing neonatal mortality (Demicheli 2013). However, there is no clear immunological correlate of protection following pertussis vaccination and, as such, evidence regarding the clinical relevance of vaccination and any blunting of response to subsequent immunisation is more difficult to ascertain.
Some animal studies suggest that vaccination of pregnant sows results in elevated serological levels of pertussis IgG with evidence of transfer to newborn piglets (Elahi 2006). Those newborn piglets receiving passive immunity exhibit fewer clinical symptoms, suggesting this may result in clinical protection (Elahi 2006), although it is not clear whether such findings can be extrapolated to humans.
Why it is important to do this review
In 2011, the USA recommended that pregnant women who had not previously received a pertussis-containing vaccine, should be offered the Tdap vaccine at more than 20 weeks gestation (ACIP 2011). However, to date, evidence of infant protection is largely indirect. In October 2012, in response to the outbreak and continued increase in disease amongst infants less than three months of age in the UK, the Department of Health, with advice from the Joint Committee on Vaccination and Immunisation (JCVI), announced the introduction of a temporary programme of maternal antenatal vaccination (DoH 2012). The committee noted that whilst this was likely to be the most effective strategy to provide protection of infants from birth and there is good evidence of transplacental transfer of antibodies, the effectiveness in terms of providing infant protection remains uncertain, primarily consisting of single studies. Following this announcement, in October 2012, the USA revised their recommendation to include all pregnant women (ACIP 2013).
As with any maternal immunisation programme, there are also concerns about the safety of vaccination during pregnancy. A number of vaccines are already recommended for use in pregnant women and, to date, there is little evidence of any adverse outcomes for either mother or child (Mooi 2007). A randomised clinical trial carried out amongst pregnant women in Egypt, for example, demonstrated few adverse effects amongst mothers or their children following immunisation with tetanus in pregnancy (Salama 2009). Enhanced reactogenicity of acellular pertussis vaccines has been described with increasing numbers of doses in infants and young children and is associated with more pronounced local side effects (ACIP 2000) but further evidence of any adverse impact following maternal antenatal vaccination is required.
A Cochrane systematic review on this topic would enable evidence-based conclusions to be drawn to further inform this policy. This is a relevant priority topic that will fill an important gap in the current literature and help to guide international policy.
The primary objective of this review is to assess existing evidence for the effectiveness and safety of immunisation in pregnancy (in any trimester) in preventing pertussis infection in infants who are too young to be protected by the routine primary vaccinations compared to no vaccination, sham or placebo vaccination.
Criteria for considering studies for this review
Types of studies
Any primary (experimental or observational) or secondary study type reporting on the effectiveness of antenatal vaccination with a pertussis-containing vaccine, in any pregnant woman.
The most relevant study design to address the proposed review questions and that presenting the lowest risk of bias is a randomised controlled trial (RCT). However, given that antenatal vaccination for pertussis has only recently been considered in immunisation policy, we anticipate that the existing evidence base is likely to be limited. As such, we will also consider other experimental and observational study designs including non-RCTs, prospective and retrospective cohort studies and case-control studies. We will use systematic reviews to obtain further primary studies of relevance, but will not extract data from secondary analyses of these.
- Any study that does not consider immunisation with a pertussis-containing vaccine during pregnancy.
- Any study that considers vaccination of groups which do not include pregnant women.
- Animal studies.
- Any study which considers minor adverse events only (i.e. those that are self limiting).
Types of participants
Pregnant woman of all ages and gestations of pregnancy, with any pregnancy (single, multiple, complicated or uncomplicated) and their infants.
Types of interventions
We will consider pertussis-containing vaccines (acellular or whole cell) administered at any stage of pregnancy for the purposes of this review as compared with no vaccination, sham or placebo vaccination. In the UK the dTaP-IPV vaccine (Repevax and Boostrix-IPV) is the licensed acellular pertussis-containing vaccine suitable for use in adults and adolescents (HPA 2012). In other countries, additional vaccines are licensed for this, e.g. Tdap (ADACEL or BOOSTRIX) in the USA (ACIP 2011).
Types of outcome measures
- Evidence of infant infection as demonstrated by:
- cases of pertussis infection with clinical or laboratory confirmation (or both);
- admission to a healthcare facility or any other suggestion of severity of illness such as length of stay, admission to intensive or critical care, or death, where pertussis is specified as the primary diagnosis or a major contributing factor.
- Evidence of transplacental transfer of antibody, as demonstrated by cord, fetal or infant blood maternal antibody titres to pertussis vaccine component, at birth or prior to the first primary vaccination.
- Antibody titres to pertussis vaccine component in mother pre- and post-immunisation.
- Evidence of blunting of the response to routine, primary childhood immunisations as demonstrated by infant antibody titres to pertussis vaccine component and immunological correlates of protection (where these exist), pre- and post-immunisation.
- Any reported major side effect or adverse event in the mother (between 30 minutes post-administration and four months post-delivery) or in the fetus/infant(s) (from birth until 13 months of age), or in both.
- Evidence of protection against pertussis infection in infants, including frequency of clinically defined or laboratory-confirmed cases, admission to a healthcare facility or any other indication of severity of illness (e.g. requirement for intensive care or death) where pertussis is specified as the primary diagnosis or a major contributing factor.
- Evidence of the safety of vaccination in pregnancy, as demonstrated by major adverse events in the mother or in the fetus/infant(s), or both. Given that the profile of any associated serious adverse events is unknown, we will consider the time period between 30 minutes post-administration and four months post-delivery for mothers and from birth until 13 months of age for infants, in accordance with ongoing clinical trials (Dalhousie University 2013; NIAID 2013).
- Given that immunisation schedules are likely to vary, we will consider the time period between birth and 12 months of age, or six months post-completion of the primary immunisation programme.
- Antibody response in the mother following vaccination in pregnancy, as evidenced by maternal serum antibody titres to pertussis vaccine component pre- and post-immunisation.
- Transplacental transfer of antibody as evidenced by cord, fetal or infant blood titres of maternally derived antibody to any vaccine component, at birth or prior to the first primary vaccination.
- Blunting of the immunological response to primary childhood vaccinations as evidenced by antibody levels to any vaccine component pre- and post-administration of the vaccine and established correlates of protection where these exist.
- Impact on maternal pertussis infection as evidenced by the incidence of pertussis (clinically or laboratory defined). As described above, we will consider the period from two weeks after vaccination to six months post-completion of the infants primary immunisation programme in order to capture potential indirect effects of reduced mother to infant transmission during the time infants remain most susceptible to severe infection.
Search methods for identification of studies
We will search the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, current issue), which contains the Cochrane Acute Respiratory Infections (ARI) Group’s Specialised Register, MEDLINE (1946 to present), CINAHL (1981 to present), EMBASE (1974 to present), LILACS (1982 to present) and Web of Science (1985 to present).We will use the search strategy in Appendix 1 to search MEDLINE and CENTRAL. We will adapt the search strategy for the other databases. We will not use any language, date or publication restrictions.
Searching other resources
We will search the trials registries, ClinicalTrials.gov (www.clinicaltrials.gov) and WHO ICTRP (http://apps.who.int/trialsearch). In addition, snowballing technique such as reference list follow up, expert authors, advisory bodies (e.g. SAGE working group) and vaccine manufacturers will be contacted in order to identify any further potentially relevant, but unpublished studies. Reference and citation tracking will be carried out for all included papers (particularly to look for historical papers – pre-1950) and any links to additional sources of relevance identified during searching of conference proceedings and Internet sites followed in order to identify any additional studies of interest.
Where possible, results will be merged using reference management software in attempt to minimise duplication.
Data collection and analysis
Selection of studies
We will scan the titles and abstracts from the searches and obtain full-text articles when they appear to meet the eligibility criteria, or when there is insufficient information to assess the eligibility. We will assess the eligibility of the studies independently and document the reasons for exclusion. We will resolve any disagreements between the review authors by discussion. We will contact the trial authors if clarification is needed. We will translate papers in languages other than English with the help of the Cochrane ARI Group.
Please refer to the Criteria for considering studies for this review for inclusion and exclusion criteria.
Data extraction and management
The pairs of review authors (SG and SK; GD and NA; HC and GA) will independently extract data from the trials using pre-tested data extraction forms. We will extract the following data from included studies ( Table 1; Table 2; Table 3; Table 4; Table 5).
- Study characteristics (study design, setting, study sample size (allocated, received, analyzed), power calculation performed, methods of recruitment/selection, inclusion and exclusion criteria, type and sources of funding/sponsorship).
- Characteristics of participants (average age and measure of spread, clinical/socioeconomic characteristics, gestation of pregnancy).
- Characteristics of intervention (type of vaccine, timing of vaccination in relation to gestation of pregnancy, type of comparator intervention, cases exposed/total (for case-control studies only), controls exposed/total (for case-control studies only).
- Study outcomes (outcome measure such as definition, measurement tool, occurrence of disease in infants, clinical and laboratory case definitions used, methods used for serological diagnosis, timing of measurement, unit of measurement, analysis outcomes such as duration of follow-up, method of statistical analysis, losses to follow-up).
- Results (result of intervention and comparator, statistical output, control for selection bias and confounding, as well as limitations).
- Adverse effects (any reported major side effect or adverse event in the mother (between 30 minutes post-administration and four months post-delivery) or in the fetus/infant(s) (from the point of exposure until 13 months of age), or in both.
We will resolve any disagreements about data extraction by referring to the study report and by discussion. We will attempt to contact the study authors where data are insufficient or missing.
Assessment of risk of bias in included studies
We will assess the risk of bias ( Table 2) based on random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessors, completeness of outcome data, selectivity of outcome reporting and other bias, as discussed in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will resolve any disagreements by discussion.
We will use the GRADE approach (Duclos 2012) (as modified by the Strategic Advisory Group of Experts) to cater for non-randomised vaccine studies.
Measures of treatment effect
Where data are available, we will express effect sizes for dichotomous outcomes as summary risk ratios (RRs) with 95% confidence intervals (CIs), and for continuous variables as mean differences (MDs) if outcomes are measured in the same way with 95% CIs.
Unit of analysis issues
We will assume an intention-to-treat (ITT) policy for all outcomes such that all participants will be analyzed in the group to which they are allocated, regardless of whether or not they received the intervention.
Dealing with missing data
We will deal with missing data in accordance with the recommendations in Chapter 16 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). In the first instance, we will attempt to contact the trial authors to request missing data. If this is not possible and missing data are thought likely to have occurred at random, we will exclude them. Where missing data are not thought to have occurred randomly, we will use replacement values. We will record any instance of missing data in the data extraction form and will outline subsequent assumptions made and methods of dealing with this elsewhere in the methodological description. Where appropriate, we will use a sensitivity analysis to assess the impact of any assumptions made.
Assessment of heterogeneity
We will assess the presence of heterogeneity in two steps. First, we will assess obvious heterogeneity at face value by comparing populations, settings, interventions and outcomes. Second, we will assess statistical heterogeneity by means of the I
Assessment of reporting biases
If there are sufficient studies, we will use funnel plots to assess the possibility of reporting biases or small study biases, or both.
We would only consider a meta-analysis of studies where the intervention, design and end point were similar. Where this is not the case we will just summarise the results of different studies (giving descriptive results) but not produce pooled estimates. Any evidence from RCTs will be analyzed separately from observational evidence.
If the studies are considered appropriate for combining in meta-analysis, we plan to undertake meta-analyses using Review Manager 5.2 software (RevMan 2012) if appropriate, and will use either a fixed-effect or random-effects model to pool data, depending on the degree of significant clinical or statistical heterogeneity. Thresholds for the interpretation of the I
Subgroup analysis and investigation of heterogeneity
If possible, we will conduct the following subgroup analyses:
- gestation of pregnancy in which the intervention occurs;
- type of intervention (vaccine used: whole cell or acellular);
- antibody response by vaccine component;
- previous vaccination status; and
- clinically confirmed or laboratory-confirmed.
We will explore the potential sources when important heterogeneity is present and either not pool the studies or use a random-effects model.
If a meta-analysis is conducted we will conduct a sensitivity analysis in order to determine the potential impact of any assumptions made.
We wish to thank the following people for commenting on the draft protocol: Siddhartha Datta, Dee Shneiderman, Kerrie Wiley, Peter McIntryre, Daniel Floret, Sanjeev Kumar, Sree Nair and Mieke van Driel.
Appendix 1. MEDLINE AND CENTRAL search strategy
1 Whooping Cough/
3 Bordetella pertussis/
6 exp Vaccines/
7 exp Immunization/
8 (vaccin* or immuni*).tw.
10 5 and 9
11 exp Pertussis Vaccine/
12 (dtp or dtap or dtap-ipv or repevax or adacel or boostrix or infanrix or tripacel or quadracel).tw,nm.
13 10 or 11 or 12
14 exp Pregnancy/
16 Maternal Exposure/
17 (pregnan* or maternal* or mother* or antenatal*).tw.
18 exp Infant/
19 (infant* or infancy or newborn* or neonat* or fetus or fetal or foetus or foetal or prematur* or postmatur* or preterm* or baby or babies).tw.
21 13 and 20
Contributions of authors
Declarations of interest
The Immunisation, Hepatitis and Blood Safety Department has provided vaccine manufactures with post-marketing surveillance reports (not pertussis-containing vaccines to date) which the companies are required to submit to the UK Licensing authority in compliance with their Risk Management Strategy. In accordance with HPA policy a cost recovery charge is made for these reports. SG has no known personal declarations of interest.
The Immunisation, Hepatitis and Blood Safety Department has provided vaccine manufactures with post-marketing surveillance reports (not pertussis-containing vaccines to date) which the companies are required to submit to the UK Licensing authority in compliance with their Risk Management Strategy. In accordance with HPA policy a cost recovery charge is made for these reports. GA has received honoraria for consultancy and assistance from vaccine manufacturers to attend scientific meetings.
The Immunisation, Hepatitis and Blood Safety Department has provided vaccine manufactures with post-marketing surveillance reports (not pertussis-containing vaccines to date) which the companies are required to submit to the UK Licensing authority in compliance with their Risk Management Strategy. In accordance with HPA policy a cost recovery charge is made for these reports. HC has no known personal declarations of interest
The Immunisation, Hepatitis and Blood Safety Department has provided vaccine manufactures with post-marketing surveillance reports (not pertussis-containing vaccines to date) which the companies are required to submit to the UK Licensing authority in compliance with their Risk Management Strategy. In accordance with HPA policy a cost recovery charge is made for these reports. NA has no known personal declarations of interest
Whilst seconded to the University of Nottingham Health Protection Research Group, Gayle Dolan conducted a systematic review entitled 'Vaccination of Health Care Workers to Protect Patients at Increased Risk for Acute Respiratory Disease' commissioned and funded by the World Health Organization Global Influenza Programme. In relation to this she attended an expert consultation for the development of a standard guideline on clinical management of influenza virus infection, in June 2011, for which travel and accommodation was also funded by the WHO.
Sources of support
- Public Health England (previously called Health Protection Agency), UK.Five review authors are currently employed by Public Health England and were permitted to work on this project as part of their work time
- No sources of support supplied