Elsevier

Vaccine

Volume 25, Issues 37–38, 17 September 2007, Pages 6655-6660
Vaccine

Review
Metaanalysis of vaccine effectiveness in varicella outbreaks

https://doi.org/10.1016/j.vaccine.2007.07.010Get rights and content

Abstract

There is a number of reports on varicella outbreaks in populations where a one dose varicella immunization program has been implemented. We performed a metaanalysis to provide a summary vaccine effectiveness (VE) estimate and to assess the possible impact of waning immunity by means of a subgroup analysis by time since immunization. We found a VE of 72.5% (95% CI 68.5–76.0) derived from 3157 children in 14 publications. Immunization coverage of the respective population was unrelated to VE. All studies (n = 4) that allowed the computation of VE over time since immunization showed a substantial decrease. In total, waning immunity was assessed by nine studies. Two reported no relation between VE and time since immunization without specifying how this had been assessed. Seven studies calculated relative risks for contracting varicella after prolonged as compared to a shorter period since immunization and reported an increased relative risk for prolonged periods. In conclusion, this metaanalysis confirms a limited effectiveness of one dose of varicella vaccine and points to waning immunity as an important causal factor. Waning might also be an issue with the newly recommended two dose vaccination schedule. Sustained surveillance for varicella outbreaks in populations with varicella immunization programs therefore is mandatory.

Introduction

In the US, a universal varicella immunization program for children has been introduced in 1995 and an immunization rate of 73% in children 19–35 months of age had been reached by 2002 [1]. This immunization program [2] initially included one dose to be administered between 12 and 18 months of age with attenuated virus from the OKA-strain (Varivax®, Merck and Company, Westpoint, PA). Unexpectedly, several varicella outbreaks in nursing homes, kindergartens and schools have been observed [3], raising concerns about robustness of VE. These concerns were supported by immunological data, that showed incomplete seroconversion after one dose in >5% of vaccinees (range from 0 to 18%, depending on the test used, see [4]) and also pointed to potential waning immunity. The potential role of waning immunity over time since immunization is subject of controversial discussions [5], [6], [7], [8]. However, a second dose of varicella immunization to be given at 4 to 6 years of age has recently been recommended in the US [9]. In Europe, the vaccine is licensed in at least 13 countries [10]. In seven of these varicella vaccination for all children is currently considered. Varicella vaccination for all children was first introduced in Germany. At present a second dose is only recommended for the Varicella-MMR combined vaccine. In order to provide an overall assessment of the VE for one dose of immunization with the live-attenuated OKA strain we performed a metaanalysis of outbreaks published between 1 January 1995 and 31 December 2006. By means of subgroup analysis we attempted to assess the time course of waning immunity.

Section snippets

Identification of studies

PubMed and Embase were searched applying the terms ‘varicella OR chickenpox’ and ‘outbreak’ and ‘vaccination OR vaccine’. Publication time limits from 1995 (year of approval of Varivax® by the US food and drug administration) until 31 December 2006 were chosen.

Inclusion criteria

We considered any reports of outbreaks in at least partially vaccinated general populations available in English or German language. Outbreaks in restricted subgroups, e.g. immunocompromized children were not considered. To be included, a

Identified studies

The applied search strategy yielded 100 potentially relevant publications in Medline and Embase which were reduced to 82 by eliminating double hits from both sources. After reviewing title and abstract 63 publications were mostly comments or discussions on different topics such as vaccine coverage or recommendations, and outbreaks in special settings (e.g. immunocompromized children). Of the remaining 19 publications, five dealt with public health responses to outbreaks, vaccine coverage, were

Discussion

To our knowledge this is the first metaanalysis on varicella VE for one dose of attenuated OKA strain virus administered at an age of least 12 months in outbreak settings. Overall a VE of 72.5% was calculated. Our findings additionally pointed to waning immunity after immunization following an either linear or exponential course.

Outbreak situations offer a good opportunity to evaluate the effect of immunization under high infectious pressure in the field where it is most needed. The fact that

Acknowledgement

O. Bayer is supported by LMUinnovativ research priority project MCHealth (sub-project II).

References (25)

  • D.J. Krysan

    Effectiveness over time of varicella vaccine

    J Pediatr

    (2004)
  • Jumaan AO. Varicella Vaccination Program Netconference; 2004....
  • Centers for Disease Control and Prevention

    Recommended childhood immunization schedule—United States, January–June 1996

    MMWR Morb Mortal Wkly Rep

    (1996)
  • Centers for Disease Control and Prevention

    Public health response to varicella outbreaks—United States, 2003–2004

    MMWR Morb Mortal Wkly Rep

    (2006)
  • S. Hambleton et al.

    Preventing varicella-zoster disease

    Clin Microbiol Rev

    (2005)
  • M. Vázquez et al.

    Effectiveness over time of varicella vaccine

    JAMA

    (2004)
  • A.A. Gershon

    The current status of live attenuated varicella vaccine

    Arch Virol Suppl

    (2001)
  • K. Krause

    Reply to “Varicella vaccine revisited”

    Nat Med

    (2000)
  • Advisory Committee on Immunization Practices. Resolution No. 6/06-3; 2006....
  • A.P. de Moira et al.

    Varicella zoster virus vaccination policies and surveillance strategies in Europe

    Euro Surveill

    (2005)
  • J.J. Deeks et al.

    Statistical methods for examining heterogeneity and combining results from several studies in meta-analysis

  • A. Arnedo-Pena et al.

    Varicella vaccine effectiveness during an outbreak in a partially vaccinated population in Spain

    Pediatr Infect Dis J

    (2006)
  • Cited by (56)

    • Varicella seroepidemiology and immunization in a cohort of future healthcare workers in the pre-vaccination era

      2020, International Journal of Infectious Diseases
      Citation Excerpt :

      The critical vaccination coverage is estimated to be 90–92% and the basic reproduction rate (R0) is estimated to be 12–13 (Anderson and May, 1990); however, R0 was recently reduced to 4.64 in Italy (Melegaro et al., 2011), similar to that in the United Kingdom and lower than those in Belgium, Finland, and Poland. There are further questions regarding the following findings: (1) 20% of subjects vaccinated with one dose develop varicella (Chaves et al., 2008), because the effectiveness is approximately 72–81% (Bayer et al., 2007; Seward et al., 2008). Universal routine vaccination against varicella does not appear to increase susceptibility in adults (Tafuri et al., 2014; Baxter et al., 2014), with a decrease in morbidity from 150.7 cases per 100 000 (2001–2010) to 102.6 cases per 100 000 (2010) (Trucchi et al., 2015). (

    • Effectiveness of second-dose varicella vaccination as post-exposure prophylaxis: a prospective cohort study

      2019, Clinical Microbiology and Infection
      Citation Excerpt :

      Varicella outbreaks in school settings continue to occur, despite more than 80% of students receiving one-dose VarV in this study and other outbreaks [22–25]. This may be because the protection conferred by one-dose VarV had waned since the time of vaccination [26–28]. Our study showed that administration of 2nd-dose VarV as PEP is a good strategy for outbreak control.

    View all citing articles on Scopus
    View full text