REVIEW ARTICLEPEDIATRICS Volume 137 , number 3 , March 2016 :e 20153741

Global Varicella Vaccine Effectiveness: A Meta-analysisMona Marin, MD,a Melanie Marti, MD,b Anita Kambhampati, MPH,a Stanley M. Jeram, MD, MSc,b Jane F. Seward, MBBS, MPHa

abstractCONTEXT: Several varicella vaccines are available worldwide. Countries with a varicella

vaccination program use 1- or 2-dose schedules.

OBJECTIVE: We examined postlicensure estimates of varicella vaccine effectiveness (VE)

among healthy children.

DATA SOURCES: Systematic review and descriptive and meta-analysis of Medline, Embase,

Cochrane libraries, and CINAHL databases for reports published during 1995–2014.

STUDY SELECTION: Publications that reported original data on dose-specific varicella VE among

immunocompetent children.

DATA EXTRACTION: We used random effects meta-analysis models to obtain pooled one dose

VE estimates by disease severity (all varicella and moderate/severe varicella). Within

each severity category, we assessed pooled VE by vaccine and by study design. We used

descriptive statistics to summarize 1-dose VE against severe disease. For 2-dose VE, we

calculated pooled estimates against all varicella and by study design.

RESULTS: The pooled 1-dose VE was 81% (95% confidence interval [CI]: 78%–84%) against

all varicella and 98% (95% CI: 97%–99%) against moderate/severe varicella with no

significant association between VE and vaccine type or study design (P > .1). For 1 dose,

median VE for prevention of severe disease was 100% (mean = 99.4%). The pooled 2-dose

VE against all varicella was 92% (95% CI: 88%–95%), with similar estimates by study

design.

LIMITATIONS: VE was assessed primarily during outbreak investigations and using clinically

diagnosed varicella.

CONCLUSIONS: One dose of varicella vaccine was moderately effective in preventing all varicella

and highly effective in preventing moderate/severe varicella, with no differences by vaccine.

The second dose adds improved protection against all varicella.

aNational Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia; and bDepartment of Immunization, Vaccines and Biologicals,

World Health Organization, Geneva, Switzerland

Dr Marin conceptualized and designed the study, reviewed the articles and collected the data, supervised the analysis, interpreted the data, drafted the initial

manuscript, and revised the manuscript; Dr Marti conceptualized the study, reviewed part of the articles, interpreted the data, and reviewed and revised the

manuscript; Ms Kambhampati carried out the meta-analyses, interpreted the data, and reviewed the manuscript; Dr Jeram conducted an initial literature review,

reviewed part of the articles and collected the data, and reviewed the manuscript; Dr Seward conceptualized and designed the study, reviewed part of the articles,

critically reviewed the manuscript, and interpreted the data; and all authors approved the fi nal manuscript as submitted.

The fi ndings and conclusions in this report are those of the authors and do not necessarily represent the offi cial position of the Centers for Disease Control and

Prevention. For authors affi liated with the World Health Organization, the authors alone are responsible for the views expressed in this publication and they do not

necessarily represent the views, decisions or policies of the World Health Organization.

To cite: Marin M, Marti M, Kambhampati A, et al. Global Varicella Vaccine Effectiveness: A Meta-analysis. Pediatrics. 2016;137(3):e20153741

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

MARIN et al

A varicella vaccine, based on the

attenuated live varicella zoster virus

(VZV) Oka strain, was developed by

Dr Takahashi in Japan in the mid-

1970s.1 Two decades later, the United

States became the first country

to implement a routine childhood

varicella vaccination program after

the vaccine was licensed in 1995.2

Varicella vaccines are now licensed

and available throughout the world;

however, they are recommended

for routine use only in a small

number of primarily industrialized

countries. Where coverage rates

are high, vaccination has resulted

in important declines in varicella-

related incidence, morbidity, and

mortality.3–12

Worldwide, there are several

formulations of varicella vaccines; all

contain live attenuated VZV, and all,

except the vaccine licensed in South

Korea, are based on the Oka strain

of VZV isolated in Japan. Currently,

varicella vaccines are licensed as

a monovalent or a combination

measles, mumps, rubella, varicella

vaccine (MMRV). Monovalent

vaccine is produced in the United

States (VARIVAX; Merck & Co., Inc),

Belgium (Varilrix; GlaxoSmithKline),

Japan (OKAVAX; Biken, distributed

by Sanofi Pasteur), China (4

manufacturers: Shanghai Institute

of Biologic Products, Changchun

Keygen Biological Products Co, Ltd,

Changchun BCHT Biotechnology Co

[Baike], Changchun Changsheng Life

Sciences Ltd), and Korea (Suduvax).

MMRV is produced in the United

States (ProQuad; Merck) and Belgium

(Priorix-Tetra; GlaxoSmithKline).

Vaccine safety and efficacy were

assessed in double-blind, placebo-

controlled, 1-dose studies for 3 of

the monovalent vaccines.13–15 MMRV

vaccines were licensed on the basis

of noninferior immunogenicity of

the antigenic components compared

with simultaneous administration

of MMR and varicella vaccines16,17;

an efficacy study for 2 doses of

MMRV vaccine was performed

after licensure.18 Efficacy studies

conducted used vaccines with

varying concentrations of the Oka

strain, making comparisons across

studies and inference of results

to licensed vaccine formulations

difficult; moreover, vaccine

performance may be different under

conditions of real-world use (vaccine

effectiveness [VE]). Therefore,

postlicensure evaluations of VE are

important to inform public health

programs and policies. We conducted

a systematic literature review and

descriptive and meta-analysis to

assess the effectiveness of varicella

vaccine among immunocompetent

children to prevent (1) varicella of

any severity (1 and 2 vaccine doses),

(2) combined moderate and severe

varicella (1 dose), and (3) severe

varicella (one dose).

METHODS

Identifi cation of Studies

The complete search strategy is

described in the Supplemental

Information (Supplemental Table

1). In brief, we searched for

reports published between 1995

(year the varicella vaccine was

first recommended for routine

vaccination) and December 15,

2014, in the following databases:

Medline, Embase, Cochrane

libraries, and CINAHL. We used

search terms including “varicella,”

“chickenpox,” “vaccine,” “vaccination,”

“effectiveness,” “effective,” and

“performance” to identify reports

that presented data on varicella

vaccine effectiveness. We did not

restrict our search by language.

Study Selection and Data Collection

Each article title and abstract was

reviewed by 2 authors, and relevant

publications were selected for

full-text review. The criteria for

assessment of eligibility for meta-

analysis included the following:

(1) original report on dose-specific

varicella VE and (2) the population

studied was immunocompetent

children. Clinical trials and studies

that reported VE for postexposure

prophylaxis were excluded. For each

study included in the analysis, we

abstracted information on study

setting, study design including

case definition and classification of

disease severity, vaccine studied,

number of vaccine doses, number of

study participants age ≥12 months,

and VE with confidence intervals.

Analysis

We used random effects meta-

analysis models to obtain pooled

1-dose VE estimates and 95%

confidence intervals (CIs) for

monovalent vaccines by disease

severity (against varicella of any

severity (all varicella) and against

combined moderate and severe

varicella). Within each of the 2

categories of disease severity we

assessed pooled VE by vaccine

and by study design. For studies in

countries where >1 varicella vaccine

is licensed, if publications did not

specify the vaccine studied, we

classified VE estimates as for “mixed/

multiple vaccines.” For 2-dose VE,

because most studies included

multiple vaccines, we calculated

pooled estimates and 95% CIs overall

against varicella of any severity and

stratified by study design.

To be included in the analysis, a

publication had to report VE and a

measure of its variance (CI) or the

raw numbers to allow us to calculate

CI. When CI or raw numbers were not

reported, we contacted the authors

to obtain 1 of them.19–22 For 1 study

that reported VE and CIs for 4 birth

cohorts and an overall VE without

CI, we pooled estimates from the

4 birth cohorts to calculate CI for

the overall VE.23 When publications

reported crude and adjusted VE,

we included the adjusted estimates.

VE was calculated as 1 – relative

risk (RR) or 1 – odds ratio (OR),

depending on the study design. To

calculate the pooled VE estimates, we

2 by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

PEDIATRICS Volume 137 , number 3 , March 2016

first obtained the RR / OR for each

study using the formula RR / OR =

(1 – VE) * 100. We then calculated

the pooled OR, and transformed the

pooled OR into a pooled VE estimate

([VE = 1 – OR] * 100). We assessed

residual heterogeneity by calculating

the I2 statistic. Publication bias was

assessed by using Egger’s regression

test. We conducted mixed-effects

meta-regression analyses by using

restricted maximum likelihood

estimation, among the studies that

reported VE on all varicella and

combined moderate and severe

varicella, to investigate potential

sources of heterogeneity and identify

any differences by vaccine or study

design. All analyses were conducted

by using the metafor package in

R.24,25

For the analysis of combined

moderate and severe varicella or

severe only we used the definitions

and classification from each study

and excluded studies that did not

provide these classifications. In

most studies, severity of disease was

defined by the number of skin lesions

and incidence of complications or

hospitalization as mild, <50 lesions;

moderate, 50 to 500 lesions; and

severe, ≥500 lesions or a serious

complication or hospitalization;

the combined moderate and severe

category includes cases in the last

2 categories. Several studies used

a different number of lesions to

define severe disease, or assessed

severity based on a disease-severity

score modified from the clinical

trials, a combination of criteria that

included number of days with fever,

number of lesions, number of days

the patient needed rest and presence

of complications, hospitalization

only, or parental assessment of

severity (Supplemental Information).

VE estimates against moderate

varicella only26,27 were included in

the analysis of VE against combined

moderate and severe varicella. One

study that also reported VE against

moderate disease alone provided

data in the publication that allowed

us to calculate VE against combined

moderate and severe varicella; we

used VE against combined moderate

and severe varicella for this study.20

We used descriptive statistics to

present the findings on 1-dose VE

against severe disease alone because

none of the publications reported

CIs for VE against severe disease,

mainly due to the absence of severe

cases among vaccinated patients.

Additionally, publications that

reported no cases of severe disease

were assigned a VE of 100% against

severe disease for our analysis.

One study reported VE against

varicella-related hospitalizations23;

this estimate was included in the

summary of findings for severe

disease.

RESULTS

We screened 872 nonduplicate

articles for eligibility and identified

105 potentially relevant studies

for further review. After excluding

63 studies, 42 original studies met

our inclusion criteria for meta-

analysis (Fig 1; references listed in

Supplemental Table 2 of the Online

Supplemental Information). These

studies originated from the United

States (23), China (4), Germany (3),

Israel (3), Italy (2), Spain (2), Taiwan

(2), Australia (1), Turkey (1), and

Uruguay (1).

One-Dose VE

Almost two-thirds of studies of

1-dose varicella VE (64%, 27

of 42) were conducted during

outbreak investigations and used

a retrospective cohort study

design; 2 additional studies used a

retrospective cohort study design

based on data from electronic

databases (Supplemental Table

2). Other methods used included

matched case-control study (24%,

10 of 42), prospective cohort

study (1), household contact study

(1), and time-series modeling

(1) (Supplemental Table 2). The

populations studied included

children in different settings such

as child-care centers, schools,

community clinical practices,

hospitals, outpatient setting, and

households. Ages of participants

varied, overall ranging from 12

months to 18 years, but VE was

predominantly calculated among

preschool and elementary school–

age children (Supplemental Table

2). Most studies used as an outcome

clinically diagnosed varicella with

details on illness obtained from

parents; 5 studies used laboratory

confirmed varicella as their outcome

(Supplemental Table 2).

The 42 publications reported for

1-dose monovalent vaccines 58

VE estimates for prevention of all

varicella (Fig 2), 34 estimates for

prevention of combined moderate

and severe varicella (including 2

studies that provided estimates for

moderate varicella alone; Fig 3), and

25 estimates for prevention of severe

varicella. Only 1 study reported

MMRV VE (1 and 2 doses).27

The pooled 1-dose VE for prevention

of all varicella was 81% (95% CI:

78%–84%, I2 = 88%; Fig 2). Most

postlicensure studies were reported

from the United States and as a result,

most VE estimates were for Varivax

(n = 26; Fig 2 and Supplemental

Table 2). Studies conducted in other

countries assessed Varilrix, Okavax,

and various other varicella vaccines.

When stratified by vaccine, pooled

1-dose estimates were as follow:

Varivax 82% (95% CI: 79%–85%,

I2 = 62%), Varilrix 77% (95% CI:

62%–85%, I2 = 92%), other vaccines

86% (95% CI: 78%–91%, I2 = 39%),

or mixed/multiple vaccines 81%

(95% CI: 76%–85%, I2 = 85%; Fig

2). The only VE reported for 1 dose

of MMRV/Priorix-Tetra was 55%

(95% CI: 8%–78%). Pooled VE

estimates were also similar by study

design: 81% (95% CI: 76%–84%, I2

= 92%) for cohort and 83% (95% CI:

3 by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

MARIN et al

79%–86%, I2 = 42%) for case-control

studies (Supplemental Fig 5).

The pooled 1-dose VE for prevention

of combined moderate and severe

varicella was 98% (95% CI: 97%–

99%, I2 = 85%; Fig 3). The estimates

by vaccine were similar to the overall

pooled VE: Varivax 98% (95% CI:

95%–99%, I2 = 86%), Varilrix 98%

(95% CI: 89%–100%, I2 = 79%),

mixed/multiple vaccines 99% (95%

CI: 95%–100%, I2 = 86%; Fig 3).

When stratified by study design the

pooled estimates were 98% (95%

CI: 97%–99%, I2 = 87%) for cohort

and 97% (95% CI: 93%–99%, I2

= 69%) for case-control studies

(Supplemental Fig 6).

In the meta-regression analyses,

no significant association was

found between VE and vaccine

type or study design (all Ps > .1;

Supplemental Table 3).

Of the 25 estimates for VE for

prevention of severe disease, for 24

VE was 100%; 1 study reported a VE

of 85% (for prevention of varicella-

related hospitalizations).23 The only

VE for 1 dose MMRV/Priorix-Tetra

for prevention of severe disease was

100%.

Two-Dose VE

There were 8 publications that

reported 9 estimates for 2-dose VE,

8 for monovalent varicella vaccines

and 1 for MMRV/Priorix-Tetra. The

pooled 2-dose VE for monovalent

vaccines was 92% (95% CI:

88%–95%, I2 = 57%) and by study

design: 91% (95% CI: 84%–95%, I2

= 74%) for cohort and 95% (95% CI:

90%–97%, I2 = 0%) for case-control

studies (Fig 4). The only VE reported

for 2 doses of MMRV/Priorix-Tetra

was 91% (95% CI: 65%–98%).

There was evidence of publication

bias as indicated by Egger’s

regression test in some subgroups

analyzed but not in others

(Supplemental Table 4).

DISCUSSION

This is the first study to

systematically assess the

effectiveness of varicella vaccines

currently available worldwide. In

our analyses that included children

in 42 studies, within the first decade

after vaccination, 1 dose of varicella

vaccine was moderately effective

at preventing all varicella (81%)

and highly effective at preventing

combined moderate and severe

varicella (98%). Noteworthy is the

consistency of the findings, with

similar pooled estimates whether

the analysis was stratified by

individual vaccines or study design.

Additionally, 2 doses of varicella

vaccine were highly effective at

preventing all varicella.

Most of the reported 1-dose VE

estimates for prevention of all

varicella were close to the pooled

estimate; however, 1 estimate

for MMRV was 55%, and 2

estimates were <50%, 1 each for

Varivax and Varilrix, 44% and

20%, respectively.27–29 These

estimates were all calculated

during outbreak investigations

4

FIGURE 1Study selection. *Articles excluded from the analysis originated from the United States (15), Australia (5), China (5), Germany (5), Japan (5), France (4), Spain (4), Brazil (3), Canada (2), Chile (2), Italy (2), Turkey (2), United Kingdom (2), and 1 each from Denmark, Finland, Hungary, Israel, Korea, Poland, and Thailand. **Articles included in the analysis originated from the United States (23), China (4), Germany (3), Israel (3), Italy (2), Spain (2), Taiwan (2), Australia (1), Turkey (1), and Uruguay (1).

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

PEDIATRICS Volume 137 , number 3 , March 2016

that tend to underestimate the

performance of vaccines; outbreaks

that come to public health attention

may be exceptions in which the

vaccine failed and therefore may

represent the lower range of vaccine

effectiveness versus situations in

which the vaccine worked well and

prevented outbreaks and where no

investigations are done.30 Although

there are no definitive explanations

for the low effectiveness, in outbreak

investigations, there are confounders

not able to be controlled for, the

force of infection may be high in

some outbreaks or the degree of

exposure may be variable across

study participants; additionally, such

values could be identified by chance,

especially during investigations

conducted in settings where there

is epidemiologic evidence of vaccine

failure (outbreaks). This highlights

that more than a few estimates are

needed to accurately assess VE. A

previous meta-analysis that included

VE calculated during varicella

outbreaks from 14 publications

through 2004 reported a pooled

1-dose VE of 72.5%.31

Fewer studies assessed 2-dose

VE in children. Overall, 2 doses of

vaccine provided 10% or better

protection than 1 dose. However, 2

of the estimates, both from outbreak

investigations, were <90% and

that may have lowered the pooled

estimate to the lower 90s.32,33

The pooled VE estimates we report

for 1 dose are a compilation of VE

at different points of time since

vaccination, primarily within the

first decade. Considering the age

of participants in the studies and

vaccine recommendations in each

country, the median time since

vaccination is likely lower than 10

years. Within this time frame, some

studies described a higher risk

for vaccine failure with time since

vaccination (using a cutoff of 3, 4, or

5 years),22,26,28,34–36 but other studies

did not find this association27,37–41;

in all these studies, conducted during

outbreak investigations, the sample

sizes were usually insufficient to

assess the independent effect of time

since vaccination as a risk factor.

Four studies reported decline in VE

with time since vaccination; however,

the differences did not reach

statistical significance.42–45 Vazquez

et al used laboratory confirmed

cases and described a decline in VE

for Varivax between years 1 and 2

(from 97% to 86%) after vaccination

but not subsequently (up to 7 years

of follow-up, 84%)46; another study

in the United States reported a

significant decline in 1-dose VE from

94% within 5 years after vaccination

to 88% for 5 to 9 years and 82%

for ≥10 years after vaccination.47

Additionally, Bayer et al in their

meta-analysis of outbreak data,

concluded waning immunity based

on data from four studies which all

showed a decrease in VE by time

since vaccination (data available

for an average of 4 to 6 years since

5

FIGURE 2Random effects model of 1-dose varicella VE for prevention of all varicella, by vaccine.

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

MARIN et al

vaccination).31 When interpreting the

results of these studies, consideration

should be given to the fact that

they did not adjust for likelihood

of exposure or force of infection

which declined over time due to

changing varicella epidemiology with

the implementation of vaccination

programs. This results in differential

exposure by age group and can

confound interpretation of waning

effectiveness. One large study that

controlled for likelihood of exposure

and age examined 10 years of active

surveillance data (1995–2004) from

a sentinel population and reported

an increase in incidence of varicella

among vaccinated persons with time

since vaccination after the first dose

although the rate of breakthrough

varicella was still very low.48 Only

1 study examined 2-dose VE by

time since vaccination and found

no difference in VE through 5 years

since the second dose; there were too

few subjects to confidently examine

VE >5 years after the second dose.47

The VE findings are supported by

immunologic and epidemiologic

data. Immunogenicity studies from

the United States reported that

primary vaccine failure occurs in 9%

to 14% of children after 1 dose of

vaccine.49,50 A small study found that

24% of infants lacked VZV antibody

measured by fluorescent antibody

to membrane antigen a median of

4 months after vaccination.51 The

second dose of varicella vaccine

in children produced an improved

immunologic response as measured

by the proportion of subjects with

titers of ≥5 glycoprotein-based

enzyme-linked immunosorbent

assay, an approximate correlate of

protection, (99.6% vs 86% 6 weeks

after the second and first dose,

respectively) and higher geometric

mean titers.49 Countries that

introduced varicella vaccination have

experienced substantial reductions in

varicella incidence, severe morbidity

and mortality, and evidence of herd

protection beyond the age groups

targeted by vaccination. By the end

of the decade of the 1-dose program

in the United States, when vaccine

coverage had reached ∼90%,

varicella incidence had declined 90%

or more, hospitalization in children

decreased ≥95%, and deaths had

been nearly eliminated3,52,53; effects

of vaccination were documented

within 5 years of the program

in communities where vaccine

coverage among young children

had reached ∼80%.54 In Germany,

where a 1-dose national vaccination

program was implemented in 2004,

a decline of 55% in cases and 82%

in varicella complications was

observed from 2005 to 2009 using

data from physician-based sentinel

surveillance.7 In Canada, where the

1-dose varicella vaccination program

was recommended in 1999, declines

of 81% to 88% in the number of

hospitalized varicella cases were

reported between 2000 and 2008,

with impact of the vaccination

program being noted beginning 1 to

2 years after the start of the program;

6

FIGURE 3Random effects model of 1-dose varicella VE for prevention of combined moderate and severe varicella, by vaccine.

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

PEDIATRICS Volume 137 , number 3 , March 2016

vaccination coverage ranged from

74% to 91% in 2007–2008.6 The

impact of a 1-dose vaccination

program on varicella and its severe

morbidity has also been described

from Taiwan, Uruguay, Australia, and

Italy.8,10–12 Protection against severe

disease among those vaccinated

appears to be maintained after 1

dose of vaccine. No increases in rate

of complications, hospitalizations

or deaths have been reported with

time since vaccination in countries

with a 1-dose program. The 1 study

mentioned earlier that used active

surveillance data reported that

among vaccinated persons, varicella

was twice as likely to be moderate/

severe in those who developed

disease >5 years after vaccination

compared with those who became ill

<5 years after vaccination; moderate/

severe varicella was defined as >50

skin lesions in this study.48

Despite considerable success

in controlling varicella and its

severe complications with the

1-dose varicella vaccine program

in the United States, cases and

varicella outbreaks (although less

in number, smaller in size, and of

shorter duration) continued to

occur in highly vaccinated school

students. This, coupled with the

evidence that 2 doses induce a

higher immune response and

higher effectiveness, resulted in

adoption of a routine 2-dose policy

for children in 2006.2,55 During the

first 5 years after introduction of the

2-dose program, reported varicella

incidence has declined further to

the lowest level since the start of

the vaccine program, with fewer

outbreaks and additional declines in

varicella-related hospitalizations.4,56

In Navarre region, Spain, where a

2-dose routine childhood program

was introduced from the beginning

and high coverage was achieved early

in the program, a decline of 98.5%

in incidence in children and 88%

in hospitalizations were achieved

within a period of only 5 years.9

Our study has several limitations. The

majority of studies assessed VE using

clinically diagnosed varicella cases

and the evaluations occurred during

outbreak investigations that tend

to underestimate the performance

of vaccines. We did not perform a

formal quality assessment of the

studies included; all were published

in peer-review publications however,

we excluded 2 due to methodologic

issues identified during the review.

When stratified by vaccine, there was

some evidence of heterogeneity in

the 1-dose estimates, suggested by

I2 >75%. This finding may be due to

variations in study design and size.

There was also some evidence of

publication bias in the estimates from

the publications that reported on

moderate-severe varicella, suggesting

that higher ORs could have come

from smaller studies, which could

lead to some overestimation of our

pooled estimates. Most studies are

from high-income countries with few

from middle-income and none from

low-income countries.

Available data support similar

performance of the various 1-dose

monovalent varicella vaccines in

preventing varicella. One-dose

varicella vaccine is moderately

effective (81%) for preventing all

varicella and highly effective (98%)

for preventing combined moderate

and severe varicella. The second

dose adds improved protection

against all varicella (92%). The

impact of varicella vaccination in

decreasing morbidity and mortality

due to varicella is well documented

during the first 2 decades of program

implementation. The duration of

protection after 1 dose is not fully

understood or studied, especially in

settings of low varicella incidence

however, live viral vaccines because

they actively replicate in the body

commonly provide durable immunity.

It is unknown what role waning

immunity after an initial response

and primary immunologic failure

rates play in the inability of 1 dose

of vaccine to provide complete

protection. The fact that varicella in

vaccinated persons is usually highly

modified suggests partial protection

or inadequate induction of a totally

protective immune response, not

primary vaccine failure. To date, there

is no evidence of increased severe

outcomes (death or hospitalization

rates) at population level with time

since 1 dose of vaccine. Assessment

of vaccine effectiveness in recipients

who are >10 to 20 years after

vaccinations with both 1 and 2 doses

is needed. One dose is sufficient

to reduce mortality and severe

morbidity from varicella but not to

prevent limited virus circulation and

outbreaks. To further reduce the

number of cases and outbreaks and

transmission 2 doses are needed.

Additionally, the widespread use of

7

FIGURE 4Random effects model of 2-dose varicella VE for prevention of all varicella, by study design.

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

MARIN et al

varicella vaccine could reduce herpes

zoster incidence among the vaccinated

populations; several studies reported

a lower risk for herpes zoster among

varicella-vaccinated children57–61 and

a decline in herpes zoster incidence

among cohorts targeted for varicella

vaccination.59,62

ACKNOWLEDGMENTS

We thank Centers for Disease Control

and Prevention colleagues Chengbin

Wang, MD, PhD, for reviewing the

studies in Chinese; Joanna Taliano

for performing the literature search;

Ben Lopman, PhD, for meta-analysis

technical expertise; Daniel Feikin,

MD, for the suggestion to perform

meta-analysis; and Jessica Allen

for editorial assistance. We also

acknowledge the members of the

World Health Organization Strategic

Advisory Group of Experts Working

Group on varicella and herpes

zoster vaccines and Jon Abramson,

MD, Chair of the Working Group,

Wake Forest University School of

Medicine, Winston-Salem, NC, and

Philippe Duclos, DVM, PhD, World

Health Organization, Switzerland, for

their advice for analysis during the

leadership of the work group.

8

ABBREVIATIONS

CI:  confidence interval

MMRV:  combination measles,

mumps, rubella, varicella

vaccine

OR:  odds ratio

RR:  relative risk

VE:  vaccine effectiveness

VZV:  varicella zoster virus

Dr Jeram’s current affi liation is Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada

DOI: 10.1542/peds.2015-3741

Accepted for publication Nov 25, 2015

Address correspondence to Mona Marin, MD, Centers for Disease Control and Prevention, 1600 Clifton Rd, NE; MS A-47, Atlanta, GA 30333. E-mail: mmarin@cdc.gov

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2016 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they have no fi nancial relationships relevant to this article to disclose.

FUNDING: Anita Kambhampati’s work was supported by an appointment to the Research Participation Program at the Centers for Disease Control and Prevention

administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and CDC.

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential confl icts of interest to disclose.

REFERENCES

1. Takahashi M, Otsuka T, Okuno Y,

Asano Y, Yazaki T. Live vaccine used

to prevent the spread of varicella

in children in hospital. Lancet.

1974;2(7892):1288–1290

2. Marin M, Güris D, Chaves SS,

Schmid S, Seward JF. Prevention of

varicella: recommendations of the

Advisory Committee on Immunization

Practices (ACIP). MMWR Recomm Rep.

2007;56(RR-4):1–40

3. Guris D, Jumaan AO, Mascola L, et

al. Changing varicella epidemiology

in active surveillance sites—United

States, 1995–2005. J Infect Dis.

2008;197(suppl 2):S71–S75

4. Bialek SR, Perella D, Zhang J, et al.

Impact of a routine two-dose varicella

vaccination program on varicella

epidemiology. Pediatrics. 2013;132(5).

Available at: www. pediatrics. org/ cgi/

content/ full/ 132/ 5/ e1134

5. Leung J, Bialek SR, Marin M. Trends

in varicella mortality in the United

States: data from vital statistics

and the national surveillance

system. Hum Vaccin Immunother.

2015;11(3):662–668

6. Tan B, Bettinger J, McConnell A,

et al; Members of the Canadian

Immunization Monitoring Program,

Active (IMPACT). The effect of funded

varicella immunization programs on

varicella-related hospitalizations in

IMPACT centers, Canada, 2000–2008.

Pediatr Infect Dis J. 2012;31(9):956–963

7. Siedler A, Arndt U. Impact of the

routine varicella vaccination

programme on varicella epidemiology

in Germany. Euro Surveill. 2010;15(13)

8. Bechini A, Boccalini S, Baldo V, et

al. Impact of universal vaccination

against varicella in Italy. Hum Vaccin

Immunother. 2015;11(1):63–71

9. Garcia Cenoz M, Castilla J, Chamorro

J, Martinez-Baz I, Martinez-Artola V,

Irisarri F, et al Impact of universal

two-dose vaccination on varicella

epidemiology in Navarre, Spain, 2006 to

2012. Euro Surveill. 2013;18(32):20552

10. Heywood AE, Wang H, Macartney KK,

McIntyre P. Varicella and herpes zoster

hospitalizations before and after

implementation of one-dose varicella

vaccination in Australia: an ecological

study. Bull World Health Organ.

2014;92(8):593–604

11. Quian J, Rüttimann R, Romero C,

et al. Impact of universal varicella

vaccination on 1-year-olds in

Uruguay: 1997–2005. Arch Dis Child.

2008;93(10):845–850

12. Chao DY, Chien YZ, Yeh YP, Hsu PS,

Lian IB. The incidence of varicella

and herpes zoster in Taiwan during a

period of increasing varicella vaccine

coverage, 2000–2008. Epidemiol Infect.

2012;140(6):1131–1140

13. Weibel RE, Neff BJ, Kuter BJ, et al. Live

attenuated varicella virus vaccine.

Effi cacy trial in healthy children. N Engl

J Med. 1984;310(22):1409–1415

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

PEDIATRICS Volume 137 , number 3 , March 2016

14. Varis T, Vesikari T. Effi cacy of high-titer

live attenuated varicella vaccine in

healthy young children. J Infect Dis.

1996;174(suppl 3):S330–S334

15. Ma FB, Luo LY, Zhang LH. Study on

epidemiological effect of the freeze-

dried attenuated live varicella vaccine

[in Chinese]. Zhongguo Yi Miao He

Mian Yi. 2009;15(3):193–195

16. Merck & Co. Inc. PROQUAD (measles,

mumps, rubella, and varicella [Oka/

Merck] virus vaccine live) [package

insert]. Whitehouse Station, NJ: Merck

& Co, Inc; 2013. Available at: http://

www. merck. com/ product/ usa/ pi_

circulars/ p/ proquad/ proquad_ pi. pdf

17. GlaxoSmithKline Inc. PRIORIX-TETRA

(measles, mumps, rubella, and

varicella [Oka/Merck] virus vaccine

live) [product monograph]. Ontario, CA:

GlaxoSmithKline Inc; 2013. Available at:

http:// www. gsk. ca/ english/ docs- pdf/

product- monographs/ Priorix- tetra. pdf.

Accessed January 12, 2016

18. Prymula R, Bergsaker MR, Esposito

S, et al. Protection against varicella

with two doses of combined measles-

mumps-rubella-varicella vaccine

versus one dose of monovalent

varicella vaccine: a multicentre,

observer-blind, randomised, controlled

trial. Lancet. 2014;383(9925):1313–1324

19. Nguyen MD, Perella D, Watson B, Marin

M, Renwick M, Spain CV. Incremental

effectiveness of second dose varicella

vaccination for outbreak control at

an elementary school in Philadelphia,

Pennsylvania, 2006. Pediatr Infect Dis

J. 2010;29(8):685–689

20. Seward JF, Zhang JX, Maupin TJ,

Mascola L, Jumaan AO. Contagiousness

of varicella in vaccinated cases:

a household contact study. JAMA.

2004;292(6):704–708

21. Tafuri S, Martinelli D, Prato R,

Germinario C. Vaccine effectiveness

evaluation during a varicella outbreak

among children of primary schools

and day-care centers in a region which

adopted UMV. Hum Vaccin Immunother.

2013;9(1):184–188

22. Wang Z, Yang H, Li K, et al. Single-

dose varicella vaccine effectiveness

in school settings in China. Vaccine.

2013;31(37):3834–3838

23. Huang WC, Huang LM, Chang

IS, Tsai FY, Chang LY. Varicella

breakthrough infection and vaccine

effectiveness in Taiwan. Vaccine.

2011;29(15):2756–2760

24. Viechtbauer W. Conducting meta-

analyses in R with the metafor

package. J Stat Softw. 2010;36(3):1–48

25. R Development Core Team2008). R:

A language and environment for

statistical computing. Vienna, Austria:

R Foundation for Statistical Computing.

Available at: http:// www. R- project. org.

Accessed January 12, 2016

26. Lee BR, Feaver SL, Miller CA, Hedberg

CW, Ehresmann KR. An elementary

school outbreak of varicella attributed

to vaccine failure: policy implications.

J Infect Dis. 2004;190(3):477–483

27. Spackova M, Wiese-Posselt M, Dehnert

M, Matysiak-Klose D, Heininger U,

Siedler A. Comparative varicella

vaccine effectiveness during outbreaks

in day-care centres. Vaccine.

2010;28(3):686–691

28. Galil K, Lee B, Strine T, et al. Outbreak

of varicella at a day-care center

despite vaccination. N Engl J Med.

2002;347(24):1909–1915

29. Miron D, Lavi I, Kitov R, Hendler A.

Vaccine effectiveness and severity of

varicella among previously vaccinated

children during outbreaks in day-care

centers with low vaccination coverage.

Pediatr Infect Dis J. 2005;24(3):233–236

30. Fine PE, Zell ER. Outbreaks in highly

vaccinated populations: implications

for studies of vaccine performance.

Am J Epidemiol. 1994;139(1):77–90

31. Bayer O, Heininger U, Heiligensetzer C,

von Kries R. Metaanalysis of vaccine

effectiveness in varicella outbreaks.

Vaccine. 2007;25(37–38):6655–6660

32. Gould PL, Leung J, Scott C, et al. An

outbreak of varicella in elementary

school children with two-dose varicella

vaccine recipients—Arkansas, 2006.

Pediatr Infect Dis J. 2009;28(8):678–681

33. Mahamud A, Wiseman R, Grytdal S, et

al. Challenges in confi rming a varicella

outbreak in the two-dose vaccine era.

Vaccine. 2012;30(48):6935–6939

34. Centers for Disease Control and

Prevention (CDC). Outbreak of varicella

among vaccinated children—Michigan,

2003. MMWR Morb Mortal Wkly Rep.

2004;53(18):389–392

35. Haddad MB, Hill MB, Pavia AT, et

al. Vaccine effectiveness during

a varicella outbreak among

schoolchildren: Utah, 2002–2003.

Pediatrics. 2005;115(6):1488–1493

36. Tugwell BD, Lee LE, Gillette H, Lorber

EM, Hedberg K, Cieslak PR. Chickenpox

outbreak in a highly vaccinated school

population. Pediatrics. 2004;113(3 pt

1):455–459

37. Lee LE, Ho H, Lorber E, Fratto J,

Perkins S, Cieslak PR. Vaccine-era

varicella epidemiology and vaccine

effectiveness in a public elementary

school population, 2002–2007.

Pediatrics. 2008;121(6). Available at:

www. pediatrics. org/ cgi/ content/ full/

121/ 6/ e1548

38. Lopez AS, Guris D, Zimmerman L, et al.

One dose of varicella vaccine does not

prevent school outbreaks: is it time for

a second dose? Pediatrics. 2006;117(6).

Available at: www. pediatrics. org/ cgi/

content/ full/ 117/ 6/ e1070

39. Lu L, Suo L, Li J, et al. A varicella

outbreak in a school with high one-

dose vaccination coverage, Beijing,

China. Vaccine. 2012;30(34):5094–5098

40. Marin M, Nguyen HQ, Keen J, et al.

Importance of catch-up vaccination:

experience from a varicella outbreak,

Maine, 2002–2003. Pediatrics.

2005;115(4):900–905

41. Galil K, Fair E, Mountcastle N,

Britz P, Seward J. Younger age at

vaccination may increase risk of

varicella vaccine failure. J Infect Dis.

2002;186(1):102–105

42. Cenoz MG, Martínez-Artola V, Guevara

M, Ezpeleta C, Barricarte A, Castilla

J. Effectiveness of one and two doses

of varicella vaccine in preventing

laboratory-confi rmed cases in

children in Navarre, Spain. Hum Vaccin

Immunother. 2013;9(5):1172–1176

43. Fu C, Wang M, Liang J, Xu J, Wang C,

Bialek S. The effectiveness of varicella

vaccine in China. Pediatr Infect Dis J.

2010;29(8):690–693

44. Liese JG, Cohen C, Rack A, et al. The

effectiveness of varicella vaccination

in children in Germany: a case-

control study. Pediatr Infect Dis J.

2013;32(9):998–1004

9 by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

MARIN et al

45. Wu LP, Liu G, Zhang H, et al. Effect of

time after vaccination and varicella

attack rate on effectiveness of a single

dose of varicella vaccine [in Chinese].

Chinese J Biol. 2013;26(9):1304–1308

46. Vázquez M, LaRussa PS, Gershon AA, et

al. Effectiveness over time of varicella

vaccine. JAMA. 2004;291(7):851–855

47. Thomas CA, Shwe T, Bixler D, et

al. Two-dose varicella vaccine

effectiveness and rash severity in

outbreaks of varicella among public

school students. Pediatr Infect Dis J.

2014;33(11):1164–1168

48. Chaves SS, Gargiullo P, Zhang JX, et

al. Loss of vaccine-induced immunity

to varicella over time. N Engl J Med.

2007;356(11):1121–1129

49. Kuter B, Matthews H, Shinefi eld H, et

al; Study Group for Varivax. Ten year

follow-up of healthy children who

received one or two injections of

varicella vaccine. Pediatr Infect Dis J.

2004;23(2):132–137

50. Shinefi eld H, Black S, Digilio L, et al.

Evaluation of a quadrivalent measles,

mumps, rubella and varicella vaccine

in healthy children. Pediatr Infect Dis J.

2005;24(8):665–669

51. Michalik DE, Steinberg SP, Larussa

PS, et al. Primary vaccine failure

after 1 dose of varicella vaccine

in healthy children. J Infect Dis.

2008;197(7):944–949

52. Lopez AS, Zhang J, Brown C, Bialek S.

Varicella-related hospitalizations in the

United States, 2000–2006: the 1-dose

varicella vaccination era. Pediatrics.

2011;127(2):238–245

53. Marin M, Zhang JX, Seward JF. Near

elimination of varicella deaths in

the US after implementation of the

vaccination program. Pediatrics.

2011;128(2):214–220

54. Seward JF, Watson BM, Peterson

CL, et al. Varicella disease after

introduction of varicella vaccine in

the United States, 1995–2000. JAMA.

2002;287(5):606–611

55. American Academy of Pediatrics

Committee on Infectious

Diseases. Prevention of varicella:

recommendations for use of varicella

vaccines in children, including a

recommendation for a routine 2-dose

varicella immunization schedule.

Pediatrics. 2007;120(1):221–231

56. Leung J, Harpaz R. Impact of the

maturing varicella vaccination

program on varicella and related

outcomes in the United States:

1994–2012. J Pediatric Infect Dis Soc.

2015;piv044

57. Hardy I, Gershon AA, Steinberg

SP, LaRussa P; Varicella Vaccine

Collaborative Study Group. The

incidence of zoster after immunization

with live attenuated varicella vaccine. A

study in children with leukemia. N Engl

J Med. 1991;325(22):1545–1550

58. Weinmann S, Chun C, Schmid DS, et al.

Incidence and clinical characteristics

of herpes zoster among children in

the varicella vaccine era, 2005–2009. J

Infect Dis. 2013;208(11):1859–1868

59. Civen R, Chaves SS, Jumaan A,

et al. The incidence and clinical

characteristics of herpes zoster

among children and adolescents

after implementation of varicella

vaccination. Pediatr Infect Dis J.

2009;28(11):954–959

60. Son M, Shapiro ED, LaRussa P, et al.

Effectiveness of varicella vaccine in

children infected with HIV. J Infect Dis.

2010;201(12):1806–1810

61. Stein M, Cohen R, Bromberg M, Tasher

D, Shohat T, Somekh E. Herpes zoster

in a partially vaccinated pediatric

population in central Israel. Pediatr

Infect Dis J. 2012;31(9):906–909

62. Marin M, Civen R, Zhang J, et al.

Update on incidence of herpes zoster

among children and adolescents

following implementation of varicella

vaccination, Antelope Valley, CA,

2000–2010. Presented at IDweek 2015;

October 7–11 2015; San Diego, CA

10 by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

DOI: 10.1542/peds.2015-3741 originally published online February 16, 2016; 2016;137;Pediatrics 

SewardMona Marin, Melanie Marti, Anita Kambhampati, Stanley M. Jeram and Jane F.

Global Varicella Vaccine Effectiveness: A Meta-analysis

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/137/3/e20153741including high resolution figures, can be found at:

Supplementary Material

015-3741.DCSupplementalhttp://pediatrics.aappublications.org/content/suppl/2016/02/14/peds.2Supplementary material can be found at:

References

f-list-1http://pediatrics.aappublications.org/content/137/3/e20153741.full#reThis article cites 52 articles, 7 of which you can access for free at:

Subspecialty Collections

lth_subhttp://classic.pediatrics.aappublications.org/cgi/collection/public_heaPublic Healthmunization_subhttp://classic.pediatrics.aappublications.org/cgi/collection/vaccine:imVaccine/Immunizationdiseases_subhttp://classic.pediatrics.aappublications.org/cgi/collection/infectious_Infectious Diseasefollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

https://shop.aap.org/licensing-permissions/in its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://classic.pediatrics.aappublications.org/content/reprintsInformation about ordering reprints can be found online:

. ISSN:60007. Copyright © 2016 by the American Academy of Pediatrics. All rights reserved. Print

American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since . Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from

DOI: 10.1542/peds.2015-3741 originally published online February 16, 2016; 2016;137;Pediatrics 

SewardMona Marin, Melanie Marti, Anita Kambhampati, Stanley M. Jeram and Jane F.

Global Varicella Vaccine Effectiveness: A Meta-analysis

http://pediatrics.aappublications.org/content/137/3/e20153741located on the World Wide Web at:

The online version of this article, along with updated information and services, is

. ISSN:60007. Copyright © 2016 by the American Academy of Pediatrics. All rights reserved. Print

American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since . Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on May 12, 2018http://pediatrics.aappublications.org/Downloaded from