REVIEW ARTICLE

Efficacy of Pneumococcal Vaccination in Children

Younger Than 24 Months: A Meta-Analysis
Maria Pavia, MD, MPHa, Aida Bianco, MDa, Carmelo G. A. Nobile, MDa, Paolo Marinelli, MDb, Italo F. Angelillo, DDS, MPHb
a

Department of Hygiene, Medical School, University of Catanzaro Magna Grcia, Catanzaro, Italy; bDepartment of Public, Clinical, and Preventive Medicine, Second
University of Naples, Naples, Italy

Financial Disclosure: Prof Angelillo attended a consensus conference (Rome, Italy; November 26 27, 2007) on the pneumococcal vaccination in newborns as a speaker; the other authors have no nancial
relationships relevant to this article to disclose.

ABSTRACT
CONTEXT. Pneumococcal conjugate bacterial vaccines that are able to prevent invasive

disease and mucosal infections have been developed.

OBJECTIVE. A meta-analysis of published data from trials on pneumococcal conjugate

vaccine was performed to determine the efficacy in reducing the incidence of

invasive disease caused by Streptococcus pneumoniae, pneumonia, and acute otitis
media in healthy infants younger than 24 months.
METHODS. A systematic search of the literature was conducted. Controlled clinical trials
had to compare the protective efficacy of the pneumococcal conjugate vaccine in
reducing the incidence of invasive disease caused by S pneumoniae, pneumonia, and
acute otitis media in healthy infants with placebo or control vaccines. Information
was extracted by using a standardized protocol.
RESULTS. The efficacy of pneumococcal conjugate vaccine in the reduction of invasive

pneumococcal disease was 89% involving vaccine serotypes in both the intentionto-treat and per-protocol analyses and ranged from 63% to 74% for all serotypes.
The efficacy to prevent acute otitis media sustained by vaccine serotypes was 55% in
the intention-to-treat and 57% in the per-protocol analyses, whereas it was 29% to
prevent otitis involving all serotypes in the per-protocol analysis. Finally, in the
intention-to-treat and per-protocol analyses, the efficacy to prevent clinical pneumonia was 6% and 7%, respectively, whereas for the prevention of radiographconfirmed pneumonia it was 29% and 32%, respectively.
CONCLUSIONS. The pneumococcal conjugate vaccine produces a significant effect regarding prevention of invasive pneumococcal disease. Results on prevention of otitis or
pneumonia have been less striking, but considering the high burden of these diseases
in infants, even a low efficacy has potential for tremendous impact on the health of
infants in developing and industrialized countries. Pediatrics 2009;123:e1103e1110

www.pediatrics.org/cgi/doi/10.1542/
peds.2008-3422
doi:10.1542/peds.2008-3422
This work was presented in part as poster
343 at the 2nd Global Congress on
Vaccine; December 79, 2008; Boston, MA.
Key Words
acute otitis media, efcacy, infants, invasive
pneumococcal disease, pneumococcal
conjugate vaccine, pneumonia
Abbreviations
PCVpneumococcal conjugate vaccine
PCV-nn-valent pneumococcal conjugate
vaccine
IPDinvasive pneumococcal disease
ITTintention to treat
PPper protocol
RRrelative risk
CI condence interval
Accepted for publication Feb 10, 2009
Address correspondence to Italo F. Angelillo,
DDS, MPH, Second University of Naples,
Department of Public, Clinical, and Preventive
Medicine, Via Luciano Armanni, 5, 80138
Naples, Italy. E-mail: italof.angelillo@unina2.it
PEDIATRICS (ISSN Numbers: Print, 0031-4005;
Online, 1098-4275). Copyright 2009 by the
American Academy of Pediatrics

TREPTOCOCCUS PNEUMONIAE IS one of the major bacterial pathogens responsible for a wide spectrum of invasive

disease and acute respiratory tract infections such as community-acquired pneumonia, acute otitis media,
bacteremia, and meningitis, especially in children younger than 2 years in both developing and industrialized
countries.1 The high morbidity and mortality rates resulting from these diseases2 and the increase of multidrugresistant pneumococcal strains3,4 have emphasized the urgent need for introduction of effective vaccines against
diseases caused by S pneumoniae.57
Pneumococcal conjugate bacterial vaccines that are able to prevent invasive disease and mucosal infections have
been developed. In the United States, the Advisory Committee on Immunization Practices of the Centers for Disease
Control and Prevention and the American Academy of Pediatrics have recommended routine administration of a
heptavalent pneumococcal conjugate vaccine (PCV-7), which contains the 7 serotypes (4, 6B, 9V, 14, 18C, 19F, and
23F) that are the most prevalent in causing invasive diseases among young children, concurrently with other
childhood immunizations for all infants younger than 2 years as well for older children (aged 25 years) at greater
risk of developing invasive pneumococcal disease (IPD).810
Since the current PCV-7 was introduced in many industrialized countries, several studies have been performed on
populations of infants and toddlers more representative of those for whom the vaccine is recommended in the
Western world. These studies have evaluated the immunogenicity, reactogenicity, safety, and efficacy of this vaccine
in preventing IPD, pneumonia, and acute otitis media and its complications in healthy children. However, there is
still discussion about the efficacy of PCV-7, especially for the prevention of acute otitis media. Moreover, vaccines
PEDIATRICS Volume 123, Number 6, June 2009

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

e1103

prepared with a higher number of serotypes (ie, 9 or 11)

have been produced and evaluated for their efficacy
toward the same outcomes.
Therefore, we performed a meta-analysis of all available published data from controlled clinical trials of anyvalency PCV to determine the protective clinical efficacy
in reducing the incidence of IPD, pneumonia, and acute
otitis media in healthy infants younger than 24 months.
METHODS
This study was performed with a prospectively developed protocol that prespecified the research objective,
literature-search strategy, study-inclusion criteria, methods of quality assessment, data extraction, and statistical
analysis. All subgroup variables were defined before
analysis. Reporting of the studys findings was in accordance with the Quality of Reporting of Meta-analyses
(QUORUM) conference statement.11
A detailed literature-search strategy was developed to
identify published reports of controlled trials that evaluated the protective efficacy of the PCV (any valency) in
reducing the incidence of IPD, pneumonia, and acute
otitis media in healthy infants younger than 24 months.
A systematic bibliographic search of the medical literature between January 2000 and June 2008 was conducted through electronic databases, including the National Library of Medicine (Medline) and Embase. To
achieve the maximum sensitivity of the search strategy
and identify all trials, the search was performed by exploring and combining the following medical subject
headings (MeSH) terms and text words: respiratory
tract infections, respiratory infections, pneumonia,
otitis media, acute otitis media, recurrent otitis media, consolidation, invasive pneumococcal disease,
pneumococcal conjugate vaccine, pneumococcal vaccine, efficacy, effectiveness, effect, effects,
child, children, infant, infants, toddler, and
toddlers. Titles and abstracts of trials generated by
electronic search were reviewed, and hard copies of
potentially suitable reports were retrieved for detailed
assessment. In an effort to identify other possibly undetected published articles, a systematic manual search of
bibliographies of every relevant trial retrieved from the
electronic search and from review articles was performed.
To be eligible for inclusion in the meta-analysis, controlled clinical trials had to compare the protective efficacy of the PCV (any valency) in reducing the incidence
of IPD, pneumonia, and acute otitis media in healthy
infants younger than 24 months with placebo or control
vaccines, irrespective of number of events and duration
of follow-up. The contents of all potentially relevant
abstracts or full-text articles identified through the literature search were reviewed independently by 2 investigators in duplicate to determine if they met eligibility
criteria for inclusion. When discrepancies between investigators occurred for inclusion or exclusion, additional evaluation of the study was conducted, and disagreements were resolved by consensus. Studies were
included in the meta-analysis if they met all of the
following prespecified criteria: (1) the study was a rane1104

PAVIA et al

domized, controlled trial; (2) healthy infants younger

than 24 months were enrolled; (3) the intervention
consisted of a primary series of immunizations (3 4
doses) of PCV (any valency) given before 12 months of
age and a control group with placebo or with other
vaccines; (4) the study measured at least 1 of the following outcomes, through intention-to-treat (ITT) or perprotocol (PP) analysis, that occurred during the study in
the intervention and control groups: (a) number of
events of IPD, to include all or specific serotypes; (b)
number of events of pneumonia (of unspecified etiology) with or without radiograph confirmation; and (c)
number of events of acute otitis media; (5) the study was
published in the English language; and (6) the study was
published through June 2008. Clinical pneumonia was
defined by the treating physician on the basis of an
individual appraisal of a physical examination including
or not ancillary laboratory data, or as history of cough or
breathing difficulty of 14 days duration, with a raised
respiratory rate for age or lower chest-wall indrawing.
Acute otitis media was defined as a visually abnormal
tympanic membrane (color, position, and/or mobility)
suggesting middle-ear effusion, with at least 1 or 2 of the
following symptoms or signs of acute infection: fever,
earache, irritability, diarrhea, vomiting, acute otorrhea
not caused by otitis externa, and other symptoms of
respiratory infection. If the same population appeared in
more than 1 report and there was a possibility of overlapping data, the most recent report and/or the one with
the largest number of observations was used.
Two investigators independently assessed the quality
of each individual trial included in the meta-analysis.
Each trial was read and scored for quality, and all studies
had blinded investigators, institutions, country, and
journal, thereby reducing the possibility of detection
bias. The quality assessment of the trials was performed
according to the widely used methods of Chalmers et al12
and Jadad et al.13 A Jadad et al score ranging from 0 to 5
points was assigned to the included trials according to
whether the investigators described the study as randomized and double-blind, reported the methods used to
randomly assign patients and blind the intervention, and
reported the number of withdrawals and dropouts and
the reason. Higher scores indicate higher quality in the
conduct or reporting of trials. The readers discussed their
evaluation, and discrepancies in quality ratings were
resolved by consensus between the 2 reviewers. To avoid
selection bias, no study was rejected because of these
quality criteria.
Two investigators independently assessed articles according to the predetermined eligibility criteria, and information was extracted by using a standardized protocol and a data-collection form that was created for this
study. Differences between reviewers data were resolved through discussion until a consensus was
reached. Because agreement was by consensus, no formal methods to assess interobserver agreement were
used. We did not contact authors to request additional
information. The following information was recorded
from each trial: (1) first authors name, year of publication, and geographic setting of the population studied;

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

370

potentially relevant citations

identified through electronic
database and hand searches

14 studies excluded because they were irrelevant

356

potentially relevant studies

identified and screened on the
basis of title or abstract for
retrieval
307 studies excluded
(editorials, nonrandomized trials, dealt with
pneumococcal polysaccaride vaccine, observational or
laboratory studies, cost-effectiveness studies, narrative
review)

49

potentially appropriate studies

to be included/reviewed in detail
40 studies excluded due to
23 data on immunogenicity or nasopharyngeal carriage
9 older age groups, low birth weight or preterm infants
5 duplicate studies
3 incomplete data

studies with usable information

included in meta-analysis
1 had IPD data only
2 had AOM data only
1 had pneumonia data only
3 had IPD and AOM data
2 had IPD and pneumonia data

FIGURE 1
Flowchart of the published trials evaluated for inclusion in the meta-analysis. AOM indicates acute otitis media.

(2) study design; (3) description of intervention; (4) type

of controls; (5) total number and ages of study participants in each treatment group; (6) outcomes data (including the number of IPD cases, to include all or specific
serotypes; the number of clinical pneumonia cases [of
unspecified etiology] with or without radiograph confirmation; the number of clinical episodes of acute otitis
media, of first episode of acute otitis media, of acute
otitis media cases caused by S pneumoniae [to include
serotypes present in the vaccine], of acute otitis media
cases caused by all pneumococcal serotypes, and of recurrent episodes of otitis media and of tympanoplasty);
and (7) type of analysis (ITT or PP).
The efficacy of PCV (any valency) was evaluated in
several meta-analyses by combining the results regarding the reported number of events caused by vaccine
serotypes of IPD, the number of pneumonia cases, and
the number of clinical episodes of acute otitis media in
infants who received PCV compared with those who
received placebo or another vaccine. In every study, the
efficacy of PCV on each end point of interest was expressed as 1 minus relative risk (RR) along with the 95%
confidence interval (CI). When data on efficacy were not
available, the number of events and number of participants or participants/years in the vaccine as well as in
the control group was extracted, and RRs and 95% CIs
were calculated. The outcome measures were pooled by
use of the random-effects model by the method of DerSimonian and Laird,14 with which it is assumed that
there is variation between studies and the calculated risk
ratio, thus, has a more conservative value. A statistically

significant heterogeneity was considered when the P

value was .1 among the results of the included studies.
In cases with heterogeneity, a random-effects model was
applied, as opposed to a fixed-effects model,15 because it
includes both within-study sampling error (variance)
and between-study variation in the assessment of the
uncertainty (CI) of the results of a meta-analysis. In the
absence of significant heterogeneity, overall results led
by fixed-effects models were preferred only when they
were substantially similar to those emerging from random-effects model meta-analyses.
To explore the reasons for observed heterogeneity,
sensitivity analyses were performed by grouping studies
that used only the PCV-7. Finally, the presence of publication bias was assessed with a funnel plot for asymmetry, a scatter plot of individual studies that related the
magnitude of the treatment effect against a measure of
its precision,16 using for formal statistical testing an adjusted rank correlation test.17,18
All P values are 2-sided. Results were considered to be
statistically significant at a P value of .05. All statistical
analyses were performed with the use of Stata 10 software.19
RESULTS
Figure 1 summarizes the flowchart of studies from identification through final inclusion in the meta-analysis. A
total of 370 citations were identified through electronic
database searching and scanning reference lists. Of
these, 356 abstracts were screened and identified and
retrieved potentially relevant studies for further scruPEDIATRICS Volume 123, Number 6, June 2009

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

e1105

TABLE 1 Selected Characteristics of Studies Included in the Meta-Analysis

Authors

20

Black et al
Eskola et al21
Black et al22
OBrien et al23
Kilpi et al24
Klugman et al25
Cutts et al26
Prymula et al27
OBrien et al28

Country

United States
Finland
United States
United States
Finland
South Africa
Gambia
Czech and Slovak Rep
United States

Year

2000
2001
2002
2003
2003
2003
2005
2006
2008

Age at
Enrollment,
mo

No. in Each Group

Experimental

Control

215
212
215
224
212
27
212
25
224

18 927
831
18 927
2974
835
19 922
8718
2455
424

18 942
831
18 942
2818
831
19 914
8719
2452
432

tiny. Forty-nine retrieved articles were reviewed for inclusion criteria and data extraction. A total of 9 articles
fulfilled all inclusion criteria and were included in the
meta-analysis.2028
Table 1 shows the principal characteristics of the
included clinical trials. The trials were undertaken between 2000 and 2008, were all randomized on an individual basis except for 1 study23 that was group-randomized, varied in participant size from 856 to 39 836, all but
1 study provided data on the ITT and PP populations,
and the end-point of IPD was assessed in 6 trials, radiograph-confirmed pneumonia of unspecified etiology in 3
trials, clinical pneumonia with or without radiograph
confirmation in 2 trials, and acute otitis media in 5 trials.
Some of the trials reported 2 outcomes (IPD and acute
otitis media; IPD and clinical/radiograph-confirmed
pneumonia). One study included HIV-1negative and
positive infants, and the results regarding those who
tested negative were included and pooled,25 and 1 study
was performed with a high-risk population (Navajo
American Indian children).23,28 All but 3 of the studies
used PCV-7: 2 used the 9-valent vaccine25,26 and 1 used
the 11-valent vaccine.27
Table 2 lists selected items of the quality-assessment
systems used, along with the number of articles that
scored adequate (ie, completely addressed the issue)
on each item. The mean quality scores of the individual
studies using the Chalmers et al12 scale ranged from 0.24
to 0.74 (mean: 0.53), for the protocol from 0.19 to 0.79
(mean: 0.54), and for the data analysis and presentation
from 0.3 to 0.74 (mean: 0.49). All trials received full
credit for description of therapeutic regimens, start and
stop dates, CIs of trials end points, and laboratory tests to
evaluate the adsorption or pharmacologic effect of the
treatment. A few trials reported adequate methods of
randomization (33.3%) or correct techniques of allocation concealment, such as assignment via telephone
communication by an individual not involved in the
actual treatment or treatments randomly precoded by
the pharmacy (55.6%) and could be considered doubleblinded (55.6%). No trials reported both test statistics
and P values. With regard to the Jadad et al13 criteria, the
mean score was 3.89 (median: 4), and most trials were
classified as double-blinded (88.9%); a few trials addressed adequately the problems of withdrawals or dropouts after randomization (44.4%).
e1106

PAVIA et al

Schedule,
mo

Vaccine
Valency

Outcomes

Type of
Analysis

2, 4, 6, 1215
2, 4, 6, 12
2, 4, 6, 1215
2, 4, 6, 1215
2, 4, 6, 12
2, 4, 7
2, 56
3, 4, 5, 1215
2, 4, 6, 1215

7
7
7
7
7
9
9
11
7

IPD, otitis
IPD, otitis
Pneumonia
IPD
IPD, otitis
IPD, pneumonia
IPD, pneumonia
Otitis
Otitis

ITT, PP
ITT, PP
ITT, PP
ITT, PP
ITT
ITT, PP
ITT, PP
ITT, PP
ITT, PP

TABLE 2 Distribution of Studies by Quality-Scoring Values

According to the Chalmers et al12 and Jadad
et al13 Methods
Quality Items

Chalmers et al scale
Selection description
No. and reasons for eligible patients not included in the
study
Regimen denition
Blinding of randomization
Patients blinded to therapy
Physicians/observers blinded to therapy
Physicians/observers blinded to ongoing results
Statistical estimate of sample size
Testing randomization
Testing blinding
Biological equivalent
Dates of study
Results of prerandomization
Both test statistics and P values given
CIs given
Regression/correlation
Statistical analysis
No. and reasons for patients withdrawn after randomization
Withdrawals handled in several ways
Adverse-effects discussion
Subgroups retrospective analysis
Jadad et al scale
Randomization
Double-blinding
Withdrawals and dropouts

Adequatea
No.

4
1

44.4
11.1

9
3
6
5
0
7
2
1
9
9
1
0
9
4
1
4
3
5
0

100.0
33.3
66.7
55.6
0.0
77.8
22.2
11.1
100.0
100.0
11.1
0.0
100.0
44.4
11.1
44.4
50.0
62.5
0.0

5
8
4

55.6
88.9
44.4

The items withdrawals handled in several ways and adverse-effects discussion were
evaluated in 6 and 8 studies, respectively.
a Completely addressed the issue.

Table 3 shows the overall preventive effect led by the

fixed-effects models comparing the reduction in the risk
of IPD, otitis, and pneumonia by PCV (any valency) with
that by placebo or other vaccines recommended for
healthy infants. The estimate of treatment effect on IPD
favoring PCV-7 was 89% involving vaccine serotypes in
both the ITT (95% CI: 73%96%) and PP (95% CI:
65%96%) analyses, whereas the reduction of IPD
caused by all serotypes was still significant, although
with a lower efficacy, with values of 74% (95% CI:

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

TABLE 3 Summary of Meta-Analysis Results of Effectiveness of PCV in the Prevention of IPD, Otitis, and Pneumonia
Outcome

IPD
Vaccine serotypes
Vaccine serotypes
All serotypes
All serotypes
Vaccine serotypes
Vaccine serotypes
All serotypese
All serotypes
Otitis
Vaccine serotypes
Vaccine serotypes
All episodes
All episodes
Recurrent otitis
Recurrent otitis
Ventilatory tube placement
Ventilatory tube placement
All serotypes
First episode
Vaccine serotypes
All serotypese
All episodese
All episodese
First episode
Recurrent otitis
Ventilatory tube placement
Pneumonia
Clinical pneumonia
Clinical pneumonia
Radiograph-conrmed pneumonia
Radiograph-conrmed pneumonia

No. of
Studiesa

No. of
Events
V/Cb

No. Enrolled V/Cb

RR
(Efcacy, %)

ITT
PP
ITT
PP
ITT
PP
ITT
PP

420,21,23,24
320,23,24
420,21,23,24
223,24
620,21,2326
520,2326
620,21,2326
323,24,26

5/62
3/49
17/76
6/15
8/79
14/100
28/95
36/74

23 567/22 590
18 842/18 613
23 567/22 590
3277/3077
23 567/22 590c,d
27 031/26 764d
23 567/22 590d
11 466/11 228

0.11 (89)
0.11 (89)
0.26 (74)
0.37 (63)
0.2 (80)
0.18 (82)
0.38 (62)
0.47 (53)

0.04; 0.27 (73 to 96)

0.04; 0.35 (65 to 96)
0.15; 0.46 (54 to 85)
0.15; 0.96 (4 to 85)
0.12; 0.34 (66 to 88)
0.1; 0.33 (67 to 90)
0.22; 0.67 (33 to 78)
0.31; 0.72 (28 to 69)

.001
.001
.004
.042
.001
.001
.001
.001

7
7
7
7
7
7
7
7
7
7
7, 11
7, 11
7, 11
7, 11
7, 11
7, 11
7, 11

ITT
PP
ITT
PP
ITT
PP
ITT
PP
PP
PP
PP
PP
ITT
PP
PP
PP
PP

220,21
320,21,24
220,28
420,21,24,28
220,21
220,21
220,28
220,28
221,24
221,24
420,21,24,27
321,24,27
320,27,28
520,21,24,27,28
321,24,27
320,21,27
320,27,28

6/17d
221/262
1092/1059d
3321/2081d
158/174d
123/149d
13/17d
11/16d
585/414
/d
281/403
677/603
1458/1612d
3654/2580d
57/118d
131/167d
15/26d

19 758/19 772
17 156/16 330
19 692.8/19 720.3c
17 679.9/16 876.7c
831/831d
811/821d
765.8/779.3c,d
16 088.9/16 082.7c
1591/794
/d
19 611/18 782
4046/3246
22 181.8/22 199.3c
20 134.9/19 328.7c
2455/2452d
3266/3273d
18 543.9/18 534.7c

0.45 (55)
0.43 (57)
0.94 (6)
0.94 (6)
0.91 (9)
0.9 (10)
0.8 (20)
0.8 (20)
0.71 (29)
0.49 (51)
0.43 (57)
0.63 (37)
0.86 (14)
0.91 (9)
0.49 (51)
0.9 (10)
0.78 (22)

0.36; 0.57 (43 to 64)

0.36; 0.52 (48 to 64)
0.91; 0.96 (4 to 9)
0.91; 0.96 (4 to 9)
0.86; 0.96 (4 to 14)
0.85; 0.96 (4 to 15)
0.66; 0.96 (4 to 34)
0.65; 0.98 (2 to 35)
0.62; 0.8 (20 to 38)
0.41; 0.58 (42 to 59)
0.37; 0.5 (50 to 63)
0.51; 0.79 (21 to 49)
0.68; 1.1 (10 to 32)
0.83; 1.00 (0 to 17)
0.42; 0.57 (43 to 58)
0.84; 0.96 (4 to 16)
0.64; 0.96 (4 to 36)

.001
.001
.001
.001
.001
.001
.018
.032
.001
.001
.001
.001
.23
.06
.001
.001
.02

7, 9
7, 9
7, 9
7, 9

ITT
PP
ITT
PP

222,26
222,26
322,25,26
322,25,26

1327/1428d
3118/3289
806/1003
600/866

39 019/39 046c
25 834/25 846c
39 019/27 661c,d
25 834/25 846c,d

0.94 (6)
0.93 (7)
0.71 (29)
0.68 (32)

0.9; 0.98 (2 to 10)

0.89; 0.98 (2 to 11)
0.65; 0.78 (22 to 35)
0.61; 0.76 (24 to 39)

.003
.007
.001
.001

Vaccine
Valency

Type of
Analysis

7
7
7
7
7, 9
7, 9
7, 9
7, 9

95% CI of RR
(Efcacy %)

Shown are the number of studies and the reference numbers of the studies included.
Vaccine/control groups.
c Patients and person-years.
d Data not available for all studies.
e Random-effects model.
b

54% 85%) in the ITT and 63% (95% CI: 4% 85%) in

the PP analyses. Additional meta-analyses were performed by pooling data from studies testing the efficacy
against IPD of any valency of the PCV compared with
that in the control group. The pooled RR estimate was
significant for the vaccine versus control either for vaccine serotypes or all serotypes in both the ITT (efficacy:
80% [95% CI: 66% 88%] vs 62% [95% CI: 33%
78%], respectively) and PP (efficacy: 82% [95% CI:
67%90%] vs 53% [95% CI: 28% 69%], respectively)
analyses, although the efficacy was always lower than
that observed with the PCV-7.
The efficacy of PCV-7 against acute and recurrent
otitis was less striking, and several specific outcomes
could be investigated. The efficacy to prevent acute otitis
sustained by vaccine serotypes was 55% (95% CI: 43%
64%) in the ITT and 57% (95% CI: 48% 64%) in the PP
analyses, whereas it was substantially lower, although
still significant, with a value of 29% (95% CI: 20%
38%) to prevent otitis involving all serotypes in the PP
analysis. Even lower efficacy (6%) resulted from the

pooled analysis of both the ITT (95% CI: 4%9%) and

PP (95% CI: 4%9%) analyses regarding prevention of
all acute otitis episodes, whereas the prevention of first
episode of otitis was more effective in the PP analysis,
with a value of 51% (95% CI: 42%59%). Moreover, in
the meta-analyses performed to investigate the efficacy
of PCV-7 in the prevention of recurrent otitis, a significant efficacy of 9% for the ITT (95% CI: 4%14%) and
10% (95% CI: 4%15%) for the PP analyses was observed, whereas for the prevention of tube placement,
the efficacy was 20% in both the ITT and PP analyses.
The results of the meta-analyses conducted by pooling
data from studies with PCV-7 and PCV-11 with regards
to acute otitis media and recurrent otitis were not substantially changed, but for all episodes, no significant
efficacy was observed.
The protective efficacy of the PCV in reducing the
incidence of pneumonia was established by pooling data
from trials testing the 7-valency and 9-valency vaccines.
The results showed that in the ITT and PP analyses, the
efficacy to prevent clinical pneumonia was 6% (95% CI:
PEDIATRICS Volume 123, Number 6, June 2009

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

e1107

2%10%) and 7% (95% CI: 2%11%), whereas for the

prevention of radiograph-confirmed pneumonia it was
29% (95% CI: 22%35%) and 32% (95% CI: 24%
39%), respectively.
Funnel plots did not show any significant asymmetry
in studies that explored the effectiveness of PCV in the
prevention of IPD and acute otitis media caused by vaccine serotypes, thus reducing the potential for publication bias.
DISCUSSION
This meta-analysis provides an up-to-date summary of
average effect of all the relevant randomized, controlled
trials that have assessed the efficacy of the PCV (any
valency) in reducing the incidence of IPD caused by
S pneumoniae, pneumonia, and clinical episodes of acute
otitis media in healthy infants and reliable guidance for
clinical practice and future research. It demonstrated
that the PCV-7 was highly effective in the prevention of
IPD serotypes included in the vaccine (89%) and by all
serotypes (63%74%). This may be partly because of the
high incidence of circulating S pneumoniae that belong to
serotypes in the vaccine and/or to cross reactions of
antibodies produced against the vaccine with other serotypes. In the analysis including all trials regardless of
the type of vaccine used (any valency), the results
showed a lower efficacy ranging from 80% to 82%
against vaccine serotypes and from 53% to 62% for all
serotypes. Therefore, although the efficacy of different
vaccine preparation could not formally be compared by
pooling data from head-to-head trials, we may indirectly suggest a lower efficacy of PCV-9 compared with
PCV-7. However, the 9-valent studies occurred in Africa,
where the incidence of disease is greater and the age at
which children contract the disease is different, the
range of infecting serotypes is broader, the population is
potentially sicker (with poor nutritional status and more
comorbidities, with higher infection pressure, including
HIV, diarrhea, and measles), and serotype-specific protection even for the PCV-7 is lower than that in the
developed country studies. A previous meta-analysis
was performed without restrictions to type of vaccines
involved, and data were pooled on efficacy toward IPD
caused by serotypes of S pneumoniae not included in the
vaccine or cross-reacting with those included.29 Therefore, our meta-analysis provided data specifically concerning efficacy of PCV-7 that were not available in the
previous one, and the meta-analysis on all vaccines comprised also data from a PCV-9 trial,26 which was not
available at that time. In another meta-analysis that
involved 3 of the trials included in ours,20,23,26 the minimum protective antibody concentration that correlates
with clinical protection from IPD after immunization
with PCV was derived.30
In this meta-analysis, the results regarding the several
outcomes involving prevention of acute otitis media
showed a modest, although significant, efficacy toward
prevention of all episodes of acute otitis media (10%).
This was consistent in all studies involving the PCV-7,
whereas a recent trial on the efficacy of PCV-11 yielded
a considerably higher estimate (34%); indeed, in the
e1108

PAVIA et al

meta-analysis including all types of vaccines, a significant heterogeneity was detected. There were a number
of methodologic differences among these studies, including case definitions, case detection, time periods, and
sites of investigation. A recent study verified whether
the differences in efficacy could depend on the case
definition used and reanalyzed the data of a trial21 by
using a definition by Prymula et al.27 However, the efficacy did not vary significantly, and it was concluded that
differences may be related to varying epidemiology of
S pneumoniae and its serotypes and/or different casedetection methods.31 It should be noted that the inclusion in the meta-analyses of the study by Prymula et al27
always strikingly influenced the overall results. This is of
concern, because the vaccine used was formulated by
using the H influenzae derived protein D as a carrier
protein for pneumococcal polysaccharides; therefore, it
protected also against acute otitis media caused by nontypeable H influenzae.
Another meta-analysis addressed the effect of PCV on
acute otitis media32 by pooling studies that evaluated
conjugate and polysaccharide vaccines on infants, toddlers, and children in day care centers, and only 2 of the
studies included20,21 were also pooled in this metaanalysis. Therefore, the comparison is difficult, although
the results on the prevention of all episodes of otitis were
very similar to ours, regardless of the differences in the
inclusion criteria used.
The findings on the effectiveness toward prevention
of pneumonia must be interpreted within the context of
certain limitations. Only 3 published randomized, controlled trials have investigated this outcome, and it never
was the primary outcome. To offset this limitation, however, our analysis was performed by pooling the data,
whenever possible, from all 3 trials. Moreover, despite
results of single trials being very similar, 1 of them did
not follow the World Health Organizations standardized
criteria for the radiologic diagnosis of pneumonia in
children.22 However, the data were recently reanalyzed
according to World Health Organization criteria, and the
results were not very dissimilar from those of the original one.33 To assess whether these changes would have
an impact on the meta-analysis, these data were pooled,
but results were almost identical, demonstrating the robustness of analysis and that the diagnostic criteria seem
to have had a small impact on overall results. A previous
meta-analysis,29 on the prevention of radiographconfirmed pneumonia pooling from 2 studies,22,25 demonstrated a 22% efficacy in the ITT and 24% in the PP
analyses. In our meta-analysis, which includes also data
from Cutts et al,26 the efficacy was higher in both the ITT
(29%) and PP (32%) analyses.
The results indicate that there is a reasonable amount
of evidence to suggest that PCV produces a significant
greater effect regarding prevention of IPD than did placebo or another vaccine and confirmed the current
guidelines. Results on prevention of otitis or pneumonia
are less striking, but considering the high burden of
these diseases in infants, even a low efficacy has potential for tremendous impact on the health of infants in
developing and industrialized countries. Indeed, a re-

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

duction of all-cause pneumonia hospitalizations among

US children aged 24 months has been documented,
and it has been associated to the introduction of routine
PCV-7 use.34
A key finding from this meta-analysis is that the
studies included differed in patient population and
methodologies, rendering comparison difficult even
with current meta-analysis techniques. Many different
comparisons were used, and many different outcomes
were reported, particularly on acute and recurrent otitis
media, with considerable variability in the definitions
used. One study included pediatric patients whose immune function may have differed from that of adults
with HIV for a number of years,27 although removing
this study from the analysis did not significantly affect
the results. It is sometimes possible, when sources of
variability are clearly identified, to conduct secondary
analyses to determine the extent to which that covariate
is contributing to the overall heterogeneity. In this article, however, the sources of variability were not reported with enough consistency to conduct such analysis, so subanalyses were not conducted to determine if 1
of these factors was influencing the results. It should be
pointed out, however, that in all meta-analyses involving PCV-7, no significant heterogeneity was revealed
regardless of the outcome investigated, whereas significant heterogeneity was observed in many of those involving any-valency vaccine, and this was the only
source of heterogeneity.
Our meta-analysis has highlighted that the methodologic quality of the studies was variable, with several
scoring 5 on the Jadad et al13 scale, particularly the first
studies, whereas more recent trials gained full scores.
However, there is still debate on the opportunity of using
quality scores as an inclusion criterion in meta-analysis,
and we pooled all studies regardless of their score. Sometimes subgroup analysis is performed according to quality score, but because studies were few, we preferred to
avoid such analysis. This choice was supported by the
fact that statistical heterogeneity can also be caused by
defects of methodologic quality, but that was not our
case, because in the meta-analysis restricted to PCV-7,
no statistical heterogeneity was found and, as already
stated, the only source of heterogeneity was related to
vaccine valency.
This meta-analysis has several strengths and potential
limitations. The strengths include the comprehensive
literature search that improved the likelihood of identifying all the relevant published studies. Duplicate extraction of data reduced the potential for bias in this component of the synthesis process. By limiting this review
only to randomized prospective trials, we ensured that
the included studies would have reduced likelihood of
systematic bias and, therefore, have high internal validity. The large number of patients provided an adequate
statistical power to address the study questions, as reflected by the narrow CIs for RRs, whereas most individual trials were unable to do so because of their small
sample sizes. The results are robust and consistent, as
shown by the extensive search for potential bias by use
of publication-bias analysis.

Meta-analysis also has inherent potential limitations.

Only published trials have been included, and they have
been judged according to what is recorded in the publication. The results might have been affected by publication bias, because positive studies are more likely than
negative ones to be published, and so it is possible that
other trials have been conducted and their results never
published. However, because PCV is a relatively new
vaccine, it is improbable that other negative studies exist. Moreover, results from the tests do not suggest publication bias as a main issue in this meta-analysis. We
purposefully did not contact authors of the articles included in this meta-analysis, because we wished to assess the evidence as it stands in the public domain. The
search was limited to articles published in the English
language, so the possibility of quality studies in other
languages does exist. Finally, the findings are affected by
the limitations of the individual trials included.
Lessons about the reporting of studies also emerged
from this meta-analysis. First, accurate, detailed descriptions of the source of patients are required in all reports
so that readers can judge which studies are duplicate
from a single sample and which are true replication from
a new sample. When multiple reports are justified, they
should be clearly cross-referenced. Second, many researchers describe their samples as consecutive series but
provide no information on how many patients were
excluded or refused to participate. Therefore, readers
cannot judge whether samples are truly representative
or whether an apparently consecutive series is merely a
convenience sample. These problems could be limited if
journals refused to publish articles that do not specify the dates of data collection or give the absolute number of patients who were potentially available for
recruitment.
CONCLUSIONS
The findings of this meta-analysis offer a global estimate
of overall response to PCV in healthy infants; given the
incidence of S pneumoniae infection in developing and
industrialized countries, accurate assessment of the
probability of response to PCV is critical.
ACKNOWLEDGMENTS
This study was supported by a research grant from the
Second University of Naples (Naples, Italy). The funding
source had no role in study design, data collection, data
analysis, data interpretation, or writing of the report. All
authors had access to all the data. Prof Angelillo had
final responsibility for the decision to submit the manuscript for publication.
We express our appreciation to Prof. Pietro Crovari
(Department of Health Sciences, University of Genoa,
Genoa, Italy) for the thoughtful comments provided in
the preparation of the manuscript.
REFERENCES
1. Williams BG, Gouws E, Boschi-Pinto C, Bryce J, Dye C. Estimates of world-wide distribution of child deaths from acute
respiratory infections. Lancet Infect Dis. 2002;2(1):2532

PEDIATRICS Volume 123, Number 6, June 2009

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

e1109

2. Bridy-Pappas AE, Margolis MB, Center KJ, Isaacman DJ. Streptococcus pneumoniae: description of the pathogen, disease epidemiology, treatment, and prevention. Pharmacotherapy. 2005;
25(9):11931212
3. Johnson DM, Stilwell MG, Fritsche TR, Jones RN. Emergence
of multidrug-resistant Streptococcus pneumoniae: report from the
SENTRY Antimicrobial Surveillance Program (1999 2003). Diagn Microbiol Infect Dis. 2006;56(1):69 74
4. Van Bambeke F, Reinert RR, Appelbaum PC, Tulkens PM,
Peetermans WE. Multidrug-resistant Streptococcus pneumoniae
infections: current and future therapeutic options. Drugs. 2007;
67(16):23552382
5. OBrien KL, Santosham M. Potential impact of conjugate pneumococcal vaccines on paediatric pneumococcal diseases. Am J
Epidemiol. 2004;159(7):634 644
6. Straetemans M, Sanders E, Veenhoven R, Schilder AG, Damoiseaux RA, Zielhuis GA. Review of randomized controlled
trials on pneumococcal vaccination for prevention of otitis
media. Pediatr Infect Dis J. 2003;22(6):515524
7. Oosterhuis-Kafeja F, Beutels P, Van Damme P. Immunogenicity, efficacy, safety and effectiveness of pneumococcal conjugate vaccines (1998 2006). Vaccine. 2007;25(12):2194 2212
8. Centers for Disease Control and Prevention. Preventing pneumococcal disease among infants and young children: recommendations of the Advisory Committee on Immunization Practices. MMWR Recomm Rep. 2000;49(RR-9):135
9. American Academy of Pediatrics, Committee on Infectious Diseases. Policy statement: recommendations for the prevention
of pneumococcal infections, including the use of pneumococcal
conjugate vaccine (Prevnar), pneumococcal polysaccharide
vaccine, and antibiotic prophylaxis. Pediatrics. 2000;106(2 pt
1):362366
10. Centers for Disease Control and Prevention. Updated recommendation from the Advisory Committee on Immunization
Practices (ACIP) for use of 7-valent pneumococcal conjugate
vaccine (PCV7) in children aged 24 59 months who are not
completely vaccinated. MMWR Morb Mortal Wkly Rep. 2008;
57(13):343344
11. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF.
Improving the quality of reports of meta-analyses of randomised controlled trials: the QUORUM statement. Quality of Reporting of Meta-analyses. Lancet. 1999;354(9193):1896 1900
12. Chalmers TC, Smith H Jr, Blackburn B, et al. A method for
assessing the quality of a randomized control trial. Control Clin
Trials. 1981;2(1):31 49
13. Jadad AR, Moore A, Carroll D, et al. Assessing the quality of
reports of randomized clinical trials: is blinding necessary?
Control Clin Trials. 1996;17(1):112
14. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control
Clin Trials. 1986;7(3):177188
15. Mantel N, Haenszel W. Statistical aspects of the analysis of data
from retrospective studies of disease. J Natl Cancer Inst. 1959;
22(4):719 748
16. Sterne JA, Egger M, Smith GD. Systematic reviews in health
care: investigating and dealing with publication bias and other
biases in meta-analysis. BMJ. 2001;323(7304):101105
17. Begg CB, Mazumdar M. Operating characteristics of a rank
correlation test for publication bias. Biometrics. 1994;50(4):
1088 1099
18. Egger M, Davey Smith G, Schneider M, Minder C. Bias in
meta-analysis detected by a simple, graphical test. BMJ. 1997;
315(7109):629 634
19. Stata Corp. 2007. Stata Statistical Software [computer program].
Release 10. College Station, TX: Stata Corp LP

e1110

PAVIA et al

20. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and

immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J. 2000;19(3):187195
21. Eskola J, Kilpi T, Palmu A, et al. Efficacy of a pneumococcal
conjugate vaccine against acute otitis media. N Engl J Med.
2001;344(6):403 409
22. Black SB, Shinefield HR, Ling S, et al. Effectiveness of heptavalent pneumococcal conjugate vaccine in children younger than
five years of age for prevention of pneumonia. Pediatr Infect Dis
J. 2002;21(9):810 815
23. OBrien KL, Moulton LH, Reid R, et al. Efficacy and safety of
seven-valent conjugate pneumococcal vaccine in American
Indian children: group randomised trial. Lancet. 2003;
362(9381):355361
24. Kilpi T, hman H, Jokinen J, et al. Protective efficacy of a
second pneumococcal conjugate vaccine against pneumococcal
acute otitis media in infants and children: randomized, controlled trial of a 7-valent pneumococcal polysaccharidemeningococcal outer membrane protein complex conjugate
vaccine in 1666 children. Clin Infect Dis. 2003;37(9):11551164
25. Klugman KP, Madhi SA, Huebner RE, Kohberger R, Mbelle N,
Pierce N. A trial of a 9-valent pneumococcal conjugate vaccine
in children with and those without HIV infection. N Engl J Med.
2003;349(14):13411348
26. Cutts FT, Zaman SMA, Enwere G, et al. Efficacy of nine-valent
pneumococcal conjugate vaccine against pneumonia and invasive pneumococcal disease in the Gambia: randomised, doubleblind, placebo-controlled trial [published correction appears in
Lancet. 2005;366(9479):28]. Lancet. 2005;365(9465):
1139 1146
27. Prymula R, Peeters P, Chrobok V, 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(9512):740 748
28. OBrien KL, David AB, Chandran A, et al. Randomized, controlled trial efficacy of pneumococcal conjugate vaccine against
otitis media among Navajo and White Mountain Apache infants. Pediatr Infect Dis J. 2008;27(1):7173
29. Lucero MG, Dulalia VE, Parreno RN, et al. Pneumococcal conjugate vaccines for preventing vaccine-type invasive pneumococcal disease and pneumonia with consolidation on x-ray in
children under two years of age. Cochrane Database Syst Rev.
2004;(4):CD004977
30. Siber GR, Chang I, Baker S, et al. Estimating the protective
concentration of anti-pneumococcal capsular polysaccharide
antibodies. Vaccine. 2007;25(19):3816 3826
31. Palmu A, Jokinen J, Kilpi T; Finnish Otitis Media Study Group.
Impact of different case definitions for acute otitis media on the
efficacy estimates of a pneumococcal conjugate vaccine. Vaccine. 2008;26(20):2466 2470
32. Straetemans M, Sanders EAM, Veenhoven RH, Schilder AGM,
Damoiseaux RAMJ, Zielhuis GA. Pneumococcal vaccines for
preventing otitis media. Cochrane Database Syst Rev. 2004;(1):
CD001480
33. Hansen J, Black S, Shinefield H, et al. Effectiveness of heptavalent pneumococcal conjugate vaccine in children younger than
5 years of age for prevention of pneumonia: updated analysis
using World Health Organization standardized interpretation
of chest radiographs. Pediatr Infect Dis J. 2006;25(9):779 781
34. Centers for Disease Control and Prevention. Pneumonia hospitalizations among young children before and after introduction of pneumococcal conjugate vaccine: United States,
19972006. MMWR Morb Mortal Wkly Rep. 2009;58(1):1 4

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

Efficacy of Pneumococcal Vaccination in Children Younger Than 24 Months: A

Meta-Analysis
Maria Pavia, Aida Bianco, Carmelo G. A. Nobile, Paolo Marinelli and Italo F.
Angelillo
Pediatrics 2009;123;e1103
DOI: 10.1542/peds.2008-3422
Updated Information &
Services

including high resolution figures, can be found at:

http://pediatrics.aappublications.org/content/123/6/e1103.full.
html

References

This article cites 33 articles, 5 of which can be accessed free

at:
http://pediatrics.aappublications.org/content/123/6/e1103.full.
html#ref-list-1

Citations

This article has been cited by 6 HighWire-hosted articles:

http://pediatrics.aappublications.org/content/123/6/e1103.full.
html#related-urls

Subspecialty Collections

This article, along with others on similar topics, appears in

the following collection(s):
Ear, Nose & Throat Disorders
http://pediatrics.aappublications.org/cgi/collection/ear_nose__throat_disorders_sub
Otitis Media
http://pediatrics.aappublications.org/cgi/collection/otitis_medi
a_sub
Infectious Diseases
http://pediatrics.aappublications.org/cgi/collection/infectious_
diseases_sub

Permissions & Licensing

Information about reproducing this article in parts (figures,

tables) or in its entirety can be found online at:
http://pediatrics.aappublications.org/site/misc/Permissions.xht
ml

Reprints

Information about ordering reprints can be found online:

http://pediatrics.aappublications.org/site/misc/reprints.xhtml

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright 2009 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015

Efficacy of Pneumococcal Vaccination in Children Younger Than 24 Months: A

Meta-Analysis
Maria Pavia, Aida Bianco, Carmelo G. A. Nobile, Paolo Marinelli and Italo F.
Angelillo
Pediatrics 2009;123;e1103
DOI: 10.1542/peds.2008-3422

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/123/6/e1103.full.html

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2009 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on July 27, 2015