Caries-preventive effect of glass ionomer and resin-based fissure sealants on permanent teeth: a meta analysis.

373
Abstract: The purpose of this quantitative
systematic review was to appraise the evidence on the
caries-preventive effect of glass ionomer cement (GIC)
in relation to resin-based fissure sealants. Nine English
and two Portuguese databases were searched (15
January 2008). Randomized clinical trials and
systematic reviews were considered for inclusion. Trial
exclusion criteria were: drop-out rates > 33%; no
randomization; baseline differences in groups not
statistically adjusted; and no clinically important
outcomes were presented. Two authors reviewed the
articles independently. The outcome measure for the
caries preventive effect was caries absence on sealed
teeth. Of the 112 identified articles, 25 were selected
for review. Of these, 14 were excluded and 11 accepted
(8 trials; 3 systematic reviews). The accepted reviews
provided no evidence of superiority of either sealant
material. Six trials were included for meta-analysis. The
pooled odds ratio was 0.96, 95% CI 0.62-1.49, indicating
no difference in the caries-preventive effect of GIC
and resin-based fissure sealant material. This systematic
review with meta-analysis found no evidence that either
material was superior to the other in the prevention of
dental caries. Thus, both materials appear equally
suitable for clinical application as a fissure sealant
material. (J Oral Sci 51, 373-382, 2009)
Keywords: glass ionomer cement; resin composite;
fissure sealing; meta-analysis.
Introduction
Pits and fissures of posterior molar teeth are considered
to be highly susceptible to the adhesion of microorganisms
and, consequently, caries. Therefore, a significant amount
of tooth decay occurs at these sites. Fissure sealants are
used to prevent occlusal caries with 71% percent of occlusal
decay preventable after a once-off fissure sealant application
(1). The evidence for the efficacy and cost-effectiveness
of sealants in reducing occlusal caries in molars has been
highlighted in a number of articles in highly rated journals
(1-5). The most commonly used sealant material is resin
composite (6-8). Its caries-preventive effect relies on the
sealing of pits and fissures through micro-retention, created
through tags after enamel acid etching. However, these are
easily destroyed by saliva contamination, which reduces
micro-retention and consequently, the caries-preventive
effect (9). Under the generally wet conditions in the oral
cavity, Glass Ionomer Cement (GIC) offers an alternative.
Owing to its hydrophilic properties, GIC is not as moisture-
sensitive as hydrophobic resin (10). It has been suggested
that the ‘gold standard’ in caries prevention through sealant
administration should not be based on physical (material
retention on the tooth surface) but rather, on biological
outcomes (11). Such biological outcomes are measured in
relation to the absence of caries in pits and fissures after
sealant application. So far, three systematic reviews
(2,11,12) including appraisals on the effectiveness of GIC
fissure sealant have been published. One of these, by
Mejare et al. (12), did not include a direct comparison
between GIC and resin-based sealants. Two other
systematic reviews (2,11) have compared the effect of
GIC with that of resin based fissure sealants. One of these
was a Cochrane Systematic Review (2) that used strict
inclusion criteria, which resulted in a large number of
Journal of Oral Science, Vol. 51, No. 3, 373-382, 2009
Correspondence to Dr. Steffen Mickenautsch, Division of Public
Oral Health, University of the Witwatersrand, P.O. Box 2779,
Houghton 2041, South Africa
Tel: +27-11-717-2594
Fax: +27-11-717-2625
E-mail: neem@global.co.za
Caries-preventive effect of glass ionomer and resin-based
fissure sealants on permanent teeth: a meta analysis
Veerasamy Yengopal1), Steffen Mickenautsch1), Ana C. Bezerra2) and Soraya C. Leal2)
1)Division of Public Oral Health, University of the Witwatersrand Johannesburg, Houghton, South Africa
2)School of Dentistry, University of Bras ı´lia, DF, Brazil
(Received 30 October 2008 and accepted 16 April 2009)
Original

374
trials being excluded from the final analysis. The systematic
review by Beiruti et al. (11) excluded studies lacking
sufficient reported statistics for calculation of relative and
attributable risk. In all these three systematic reviews,
only English databases were searched and English articles
reviewed. Additionally, the inconclusive findings reported
in each of these reviews were based on the authors’
assessment of each included trial using a PICOS (patient;
intervention; controls; outcome; study authors’ conclusions)
format and a narrative synthesis of the included articles.
However, the disadvantage of a narrative synthesis in
systematic reviews is that bias may be introduced if the
outcomes of some studies are inappropriately stressed
over others (13). The advantages of a meta-analysis over
narrative synthesis are that it provides the chance to detect
a treatment effect as statistically significant (P < 0.05) and
to improve the estimation of a treatment effect by
quantifying its outcome, thus making its estimation more
precise (13). Therefore, whilst methodological weaknesses
limit what can be inferred in terms of efficacy, the
cumulative weight of evidence (as highlighted, where
possible, in a meta-analysis) provides a more objective
assessment of a systematic analysis of the literature. The
inconclusive findings reported in the three published
systematic reviews may reflect the opposite should a meta-
analysis of trials that report on the same outcome be added.
Indeed, this has been shown to be the case in a number of
systematic reviews where the individual studies had varied
outcomes but the cumulative weight of the evidence (done
by pooling together the results of trials with similar
outcomes) were found to be conclusive for that particular
outcome (14-16).
Due to the lack of a conclusive quantitative analysis in
past reviews, the aim of this systematic review is not only
to extend the evidence search and review to non-English
clinical trials, but also to conduct a meta-analysis in order
to quantitatively appraise the current evidence regarding
the caries-preventing effect of GIC in comparison to that
of resin-based fissure sealants for the first time.
Materials and Methods
Search strategy
The literature search covered nine Anglophone databases:
Biomed Central, Cochrane Oral Health Reviews, Cochrane
Library, Directory Of Open Access Journals, Expanded
Academic ASAP PLUS, Meta Register Of Controlled
Trials – mRCT, PubMed, Science-Direct, Research
Findings Electronic Register – ReFeR and two Lusophone
databases: Bibliografia Brasileira Em Odontologia – BBO,
Literatura Latino-Americana E Caribenha Em Ciências Da
Saúde – LILACS. In order to search databases, strings of
search terms were constructed, consisting of relevant text
words and Boolean links. The string of English search
terms: “(GIC sealant* OR Glass ionomer cement sealant)
AND (caries OR tooth decay)” was used to search the
Anglophone databases and the string of Portuguese search
terms: “SELANTE” [Palavras] and “CIMENTOS DE
IONOMEROS DE VIDRO” [Palavras] and “CARIE”
[Palavras]” was used to search the Lusophone databases.
All publications listed in the databases until 15 January
2008 were included in the search.
Inclusion and exclusion criteria
Both clinical trials and systematic reviews by other
authors were eligible for inclusion. Publications were
included from the search results on the basis that their titles
Table 1 Exclusion criteria for trials and literature reviews

375
and abstracts were in accordance with broad inclusion
criteria: (i) titles/abstracts were relevant to the review
objective; and (ii) the article was published in English,
German, Portuguese or Spanish. Where only a relevant title
without a listed abstract was available, a full copy of the
publication was assessed for inclusion. In accordance with
published recommendations (17), included articles were
reviewed independently by two reviewers. Disagreements
were resolved through discussion and consensus. After
review, articles were accepted only if they complied with
all the exclusion criteria described in Table 1. In cases of
multiple reports regarding the same trial, the report covering
the longest period and lacking the exclusion criteria was
accepted. For the systematic reviews, only a descriptive
analysis was attempted.
Data extraction from accepted trials
The outcome measure of the caries preventive effect was
the caries absence on sealed teeth. Two reviewers (VY and
SM) independently extracted data from the accepted
articles, using a pilot-tested data-extraction form that
included information contained in Table 2. Wherever
Table 2 Details of accepted trials

376
possible, missing data was calculated from information
given in tables and text of trials in order to complete the
2 × 2 table for meta analysis. Disagreements between
reviewers during data extraction were resolved through
discussion and consensus.
It was anticipated that the majority of studies eligible
for inclusion would be split-mouth in design. The split-
mouth study design is commonly used in dentistry to test
interventions and has the advantage of having an individual
serve as both the experiment and control. In this study
design, one or more pairs of teeth (e.g. primary molars)
form the unit of randomization. Strictly, these pairs are not
independent and should be analyzed as “paired data” on
a patient basis. However, similar to other reviews where
split-mouth trials are included (2), it was decided to analyze
the pairs independently as it would have meant that most
trials considered for inclusion here would have been
excluded.
Quality of trials
The quality assessment of the included trials was
undertaken independently by two reviewers (VY and SM).
The quality assessment process was piloted using trials not
included in this review and subsequently; quality assessment
rating scored by both the reviewers was derived by
consensus within the review group. Four main quality
criteria were examined:
1) Generation of randomization sequence (Allocation),
recorded as:
(A) Adequate – e.g. computer generated random numbers,
table of random numbers.
(B) Unclear.
(C) Inadequate – e.g. case record number, date of birth,
date of administration, alternation.
2) Allocation concealment, recorded as:
(A) Adequate – e.g. central randomization, sequentially
numbered sealed opaque envelopes.
(B) Unclear.
(C) Inadequate – e.g. open allocation schedule, unsealed
or non-opaque envelopes.
3) Blind outcome assessment, recorded as:
(A) Yes.
(B) Unclear.
(C) No.
(D) Not used/possible.
4) Completeness of follow up (clear explanation for
withdrawals and loss-to-follow-up in each treatment group)
assessed as:
(A) Yes, drop outs less than 30%.
(B) Yes, drop outs more than 30%.
(C) No explanation.
Meta-analysis
The caries absence and caries presence in sealed teeth
were treated as dichotomous data. Trials were assessed for
their clinical and methodological heterogeneity following
Cochrane guidelines (13). Trials were considered
homogenous, if they did not differ substantially in the
following clinical and methodological aspects: age of
patients; follow-up period; type of sealant material used;
frequency of sealant material application; as well as
measured outcome. Only trials considered to be clinically
and methodologically homogenous were included for
meta-analysis, for which the fixed effects model of the meta-
analysis software, RevMan 4.2 was used. The differences
in the caries preventive effect were computed on the basis
of odds ratios (OR) from each trial and the respective
95% confidence interval (CI). Studies were assigned a
Mantel-Haenszel weight in direct proportion to their
sample size.
Results
From the initial search results, 112 articles were
identified, 25 of which were selected for review.
Independent review of these 25 articles resulted in further
exclusion of 2 reviews (8,18) and 12 trials (19-30). Table
3 provides information on the reasons for exclusion. Four
trials (19,20,23,29) were excluded because the drop-out
rates of participants were greater than 33%. The trial by
Boksman et al. (21) was abandoned 6 months into the 3-
year trial period, because only 1.7% of the GIC fissure
sealants placed were available for evaluation.
Eight trials (31-38) and three literature review articles
(2,11,12), were accepted and thus formed the basis for the
evaluation of evidence regarding the caries-preventive
effect of GIC versus that of resin-based fissure sealants.
Description of accepted reviews
Three literature reviews (2,11,12) were accepted. The
Cochrane systematic review (2) sought to evaluate the
caries preventive effect of resin and GIC cements in trials
comparing these two interventions with each other or with
a placebo (or no treatment). The strict inclusion and
exclusion criteria meant that 40 of the 56 studies included
for review were excluded, e.g. split-mouth trials, in which
the authors did not present data in a paired way were
excluded in this review without the attempt to calculate
the missing data from available information. These criteria
added to the strength of methodological rigor of this review

377
but resulted in similar findings in the review presented by
Mejare et al. (12): although there was evidence regarding
the effectiveness of resin sealants, the evidence related to
GIC based sealants was perceived to be less convincing
or incomplete. Moreover, the results from the comparison
of resin sealants and GIC sealants were conflicting, as two
of the assessed trials (23,31) were in favor of resin, while
one trial (35) reported that GIC fissure sealants performed
significantly better at 44 months after placement. As the
results of these trials differed substantially, the authors did
not attempt a meta-analysis.
The second review by Mejáre et al. (12) did not include
trials comparing one type of fissure sealant material with
another. Therefore, trials that pitted GIC fissure sealants
against resin-based sealants for a variety of outcome
measures were excluded. All of the 13 studies assessed in
the review by Mejáre et al. (12) contained control groups
that did not receive any intervention (i.e., fissure sealant
caries preventive effect per tooth/child was compared to
‘no treatment’). Of these studies, none was graded as
providing “high value” evidence; only 2 were graded as
offering “moderate” evidence and most were rated as
having “limited value”. The main outcome measures were
relative risk reduction (the number of decayed occlusal
surfaces in the controls minus the number of decayed
surfaces in the sealed teeth, divided by the number of
decayed surfaces in the controls) or prevented fraction
(caries increment in the control minus caries increment in
the sealed group, divided by the caries increment in the
controls). The relative risk reductions reported were
variable; ranging between 4% and 93% for all of the
studies assessed. A meta-analysis, reporting on the caries-
preventive effect of a single application of resin-based
fissure sealants on the occlusal surfaces of 1st molars,
showed that the relative risk of developing caries in a
fissure-sealed tooth in relation to an untreated control was
0.67 (Confidence interval 95% CI: 0.55-0.83), which
corresponded to a relative risk reduction of 33%. Only 2
of the 13 studies in the Mejáre et al. (12) review dealt
specifically with GIC-type fissure sealants (31,39). Both
trials reported significant caries preventive effects for GIC
sealants but the strength of the evidence was rated as
being of limited value. Consequently the authors’ concluded
that the evidence regarding use of GIC fissure sealants was
incomplete.
The systematic review by Beiruti et al. (11) was critical
of the Cochrane (2) and Mejáre et al. (12) reviews, as the
former excluded many trials and the latter only considered
trials in which the control groups did not receive an
intervention. Beiruti et al. (11) also limited their search to
Medline and PubMed database entries to December 2004
and analyzed articles published in English only (94
publications identified and 12 analyzed). Of these, only
randomized-control trials (RCT) were analyzed, from
which a relative risk (RR) or an attributable risk (AR) could
be calculated as an outcome measure for a caries-preventive
effect. The GIC materials were categorized as medium
viscosity, low-viscosity, and low-viscosity resin-modified
(cavity liner). The resin-based materials were grouped
into ‘auto-cured’ and ‘light-cured’. Although such
methodology was conceived as being more appropriate for
reviewing trials comparing GIC and resin based sealants,
the conclusions reached were similar to that regarding the
Cochrane Review: that no evidence is provided regarding
the relative superiority of resin-based or GIC sealants
materials in preventing caries development in pits and
Table 3 Excluded articles and main reasons for exclusion