Accepted Manuscript

Title: Amalgam and resin composite longevity of posterior

restorations: A systematic review and meta-analysis

Authors: Vittorio Moraschini DDS, MS, PhD Cheung Ka Fai

DDS, MS Raphael Monte Alto DDS, MS, PhD Gustavo
Oliveira dos Santos DDS, MS, PhD

PII: S0300-5712(15)00144-X
DOI: http://dx.doi.org/doi:10.1016/j.jdent.2015.06.005
Reference: JJOD 2479

To appear in: Journal of Dentistry

Received date: 1-4-2015

Revised date: 16-6-2015
Accepted date: 19-6-2015

Please cite this article as: Moraschini V, Fai CK, Alto RM, Santos GO, Amalgam and
resin composite longevity of posterior restorations: A systematic review and meta-
analysis, Journal of Dentistry (2015), http://dx.doi.org/10.1016/j.jdent.2015.06.005

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
*Title Page with Author Details

Amalgam and resin composite longevity of posterior restorations:

a systematic review and meta-analysis.

Vittorio Moraschini, DDS, MS, PhD

Substitute professor, Fluminense Federal University, School of Dentistry,
Department of Periodontology.

t
Rua Mario dos Santos Braga, 30, Centro, Niterói, Rio de Janeiro, Brazil. Cep.:

ip
24020-140.

cr
Cheung Ka Fai, DDS, MS
Graduate student, Fluminense Federal University, School of Dentistry, Department

us
of Integrated Clinic.
Rua Mario dos Santos Braga, 30, Centro, Niterói, Rio de Janeiro, Brazil. Cep.:
24020-140.

Raphael Monte Alto, DDS, MS, PhD

an
Adjunt Professor, Fluminense Federal University, School of Dentistry, Department of
Integrated Clinic.
M
Rua Mario dos Santos Braga, 30, Centro, Niterói, Rio de Janeiro, Brazil. Cep.:
24020-140.
d

Gustavo Oliveira dos Santos, DDS, MS, PhD

Adjunt Professor, Fluminense Federal University, School of Dentistry, Department of
e

Integrated Clinic.
Rua Mario dos Santos Braga, 30, Centro, Niterói, Rio de Janeiro, Brazil. Cep.:
pt

24020-140.
ce

Corresponding Author:
Vittorio Moraschini Filho
Ac

Fluminense Federal University - Department of Periodontology - School of Dentistry.

Rua Mario dos Santos Braga, 30, Centro, Niterói, Rio de Janeiro, Brazil. Cep.:
24020-140.
E-mail: vittoriomf@terra.com.br

Key Words: Restorative Dentistry, Longevity, Amalgam, Comopiste Resin, Meta-

analysis.

Page 1 of 24
*Manuscript

Amalgam and resin composite longevity of posterior restorations:

a systematic review and meta-analysis.

Abstract

Objectives: The aim of the present review was to evaluate by means of a

t
ip
systematic review and meta-analysis the hypothesis of no difference in failure rates
between amalgam and composite resin posterior restorations.

cr
Data: Randomized controlled trials, controlled clinical trials and prospective and
retrospective cohort studies were included in this review. The eligibility criteria

us
included clinical trials in humans with at least 12 months of follow-up comparing the
failures rates between occlusal and occlusoproximal amalgam and composite resin

an
restorations. Clinical questions were formulated and organized according to the
PICOS strategy.
Source: An electronic search without restriction on the dates or languages was
M
performed in PubMed/MEDLINE, Cochrane Central Register of Controlled
Trials, and Web of Science up until March 2015.
d

Study selection: The initial search resulted in 938 articles from PubMed/MEDLINE,
89 titles from the Cochrane Central Register of Controlled Trials, and 172 from the
e

Web of Science. After an initial assessment and careful reading, 8 studies published
pt

between 1992 and 2013 were included in this review. According to the risk of bias
evaluation, all studies were classified as high quality.
ce

Conclusions: The results of this review suggest that composite resin restorations in
posterior teeth still have less longevity and a higher number of secondary caries
Ac

when compared to amalgam restorations. In relation to fractures, there was no

statistically significant difference between the two restorative materials regarding the
time of follow-up.
Clinical significance: There is currently a worldwide trend toward replacing
amalgam restorations with mercury-free materials, which are adhesive and promote
aesthetics. It is important to perform an updated periodic review to synthesize the
clinical performance of restorations in the long-term.

1
Page 2 of 24
1. Introduction
For decades, various materials have been used in direct restorations of
posterior teeth, such as amalgam and composite resin. In recent years, on account
of an increasing demand for aesthetic restorations, composites have gained a
prominent role in restorative dentistry. However, despite aesthetic requirements
being fundamental, the mechanical properties, longevity and mainly the functional
rehabilitation should be the most important criteria when choosing the restorative

t
ip
material.1
Although amalgam restorations still have the highest functional durability, 2 its

cr
use has been questioned in recent decades due to the incorporation of mercury to
the metal alloy.3 In addition, the need for more dental preparation, necessary to

us
promote greater restoration retention, make amalgam questionable for conservative
dentistry. For these reasons, the use of composite resins has been increasing
an
throughout the world for direct posterior teeth restorations.4,5
The higher sensitivity in the manufacturing technique, in addition to limitations
such as the contraction during polymerization and possibility of forming marginal
M
gaps, can be critical factors for the durability of composites. 6 However,
studies7,8 have shown a low annual failure average for composite resins in occlusal
d

and occlusoproximal restorations, varying from 1 to 3 %. The most frequent reason

e

for failure are recurrent or secondary marginal restoration caries,9 thus indicating
pt

possible failures in the adhesion process. On the other hand, amalgam restorations
reduce the possibility of secondary caries over time by forming oxides in the margin
ce

of the cavities as a result of the natural corrosion of the material, mainly in alloys with
high copper content.
Data from longitudinal clinical studies comparing the longevity of restorations,
Ac

especially in posterior teeth, should be interpreted with caution, because numerous

confounding factors may be involved. The experience and skill of the professional,
the size of the cavities, the quality and correct indication of material and type of
occlusion are factors that can influence the restorations performance. Due to these
variables, randomized clinical trials (RCTs) are necessary for this type of research.
However, to date, few RCTs10,11 have compared the longevity of
amalgam versus composite resin restorations.

2
Page 3 of 24
 The aim of this study was to evaluate by means of a systematic review and
meta-analysis the hypothesis of no difference in failure rates between amalgam and
composite resin posterior restorations.

2. Material and Methods

The methodology of this study followed the recommendations of the Cochrane
Handbook for Systematic Reviews of Interventions12 and PRISMA (Preferred

t
ip
Reporting Items for Systematic Reviews and Meta-Analyses).13 The clinical
reasoning was broken down and organized according to the PICOS strategy.

cr
2.1 Objective

us
The aim of the present review was to evaluate by means of a systematic
review and meta-analysis the hypothesis of no difference in failure rates between
an
amalgam and composite resin posterior restorations.

2.2. Focused question

M
What is the longevity of occlusal and occlusoproximal amalgam and
composite resin posterior restorations?
e d

2.3. Search strategy

pt

An unrestricted electronic search of dates or languages was performed in

PubMed/MEDLINE, Cochrane Central Register of Controlled Trials, and Web of
ce

Science until March 2015. The search strategy and the PICOS tool can be seen in
Table 1. In addition, the list of references of included studies was accessed in search
of new studies.
Ac

2.4. Selection criteria

This review searched for randomized controlled trials (RCTs), controlled
clinical trials and prospective and retrospective cohort studies. The eligibility criteria
included clinical trials in humans with at least 12 months of follow-up comparing the
rate of failures between occlusal and occlusoproximal amalgam and composite resin
restorations. The exclusion criteria were animal studies, in vitro studies, involving
complex restorations, case studies, case reports and reviews.

3
Page 4 of 24
2.5. Screening process
The search and screening process was conducted by two of the authors (V.
M. F and C. K. F) with the preliminary review of titles and abstracts. In a second
step, full articles were selected for careful reading and analyzed according to the
eligibility criteria (inclusion/exclusion) for future data extraction. Disagreement
between reviewers was resolved through careful discussion. The search
concordance between the two reviewers was evaluated by Cohen's kappa coefficient

t
ip
(k). The authors of the studies, when necessary, were contacted by e-mail to answer
any questions.

cr
2.6. Quality assessment

us
The quality analysis of the studies included was conducted in accordance with
the Newcastle-Ottawa scale (NOS), designed to be used in systematic reviews that
an
include non-randomized, specifically cohort, studies.14 For the analysis, three main
categories are addressed: selection, comparison and results. For the selection and
results categories, the studies may obtain one star/point for each item. For the
M
comparison category, two stars/points may be assigned. According to NOS, the
maximum score assigned to a study is nine stars/points (highest scientific evidence).
d

Studies scoring 6 stars and above are regarded as high quality.

e
pt

2.7. Data extraction

The following data were extracted from the studies included (when available):
ce

authors, study design, follow-up period, number of volunteers, number of dropouts,

mean age and range, type and size of the cavities, teeth, number of restorations,
Ac

adhesive system, restoration technique, amalgam or composite resin, survival of

amalgam, survival of composite resin, total number of failures, secondary caries and
fractures.

2.8. Statistical analysis

The binary variables (failure of restorations, secondary caries and

fractures) of the included studies were analyzed by means of meta-analysis when at
least two studies analyzed the same data types. The estimate of the effects of
intervention was expressed as risk ratio (RR) with a confidence interval (CI) of 95%.

4
Page 5 of 24
The inverse-variance method was used as random effects model or fixed effects
model. The I2 statistic was used to express the percentage of heterogeneity of the
studies. Values up to 25% were classified as low heterogeneity and values of 50 and
70% were classified as medium and high heterogeneity, respectively. When a
significant heterogeneity was found (P < 0.10), the results of the random effects
model were validated. When a low heterogeneity was observed, the fixed effects
model was considered. The level of statistical significance was set at P < 0.05.

t
ip
Publication bias was graphically explored through a funnel plot. The
asymmetry in the funnel plot can indicate possible publication bias.

cr
The data were analyzed using statistical software Review Manager (version
5.2.8; The Nordic Cochrane Center, The Cochrane Collaboration, Copenhagen,

us
Denmark, 2014).

3. Results an
3.1. Literature search
M
The initial search identified 938 titles in PubMed/MEDLINE, 89 titles in the
Cochrane Central Register of Controlled Trials, and 172 in the Web of Science. After
d

the initial assessment, 21 full articles were selected. After careful reading, 13 studies
e

were excluded because they did not fit the eligibility criteria of this review. Thus,
8 studies10,11,15-20 published between 1992 and 2013 were included in this review.
pt

The reasons for excluding the studies and the selection process can be viewed in
ce

Figure 1.
The k concordance value between the reviewers of the potential articles to be
included (titles and abstracts) was 0.97 and for the articles selected it was 0.85,
Ac

showing an "almost perfect" agreement according to the criteria proposed by Landis

& Koch.21

3.2. Study Characteristics

The characteristics of the studies included are presented in Table 2. Only one
included study authors11 returned the contact via e-mail to clarify any doubts. Two
randomized clinical trials,10,11 five prospective cohort studies15-19, and one
20
retrospective cohort study were included. The number of participants in the studies
ranged from 2715 to 472,11 with a mean age of 21.6 years. The follow-up period

5
Page 6 of 24
ranged from 1210,19 to 12018 months, with a mean of 55 months. Three thousand,
four hundred and eighty-six occlusal and occlusoproximal cavities were restored with
amalgam (1844) and composite resin (1642). All studies used amalgam alloys with
high copper content and dispersed phase and hybrid or microhybrid composite
resins. The adhesive systems employed as technique were applied in two or three
steps for enamel and dentin.
Only four studies11,15,17,18 described the technique used to insert the

t
ip
composite resin and control of the polymerization contraction. No study reported the
mechanisms used for the isolation of the operative field.

cr
3.3. Quality assessment

us
All studies had a score ≥ 6 stars and were classified as high quality. The
scores of each study are summarized in Table 3.

3.4 Meta-analysis
an
The mean survival of amalgam and composite resin varied from 76.320 to 100
M
%15 and 5619 to 100%10,15 with a mean annual failure of 1.71 and 3.17%,
respectively. The random effects model was used for the analysis of failures
d

between the two types of restorations analyzed due to the considerable

heterogeneity found (I2 = 78%; P < 0.00001). The meta-analysis presented a RR of
e

0.46 (95% CI: 0.28 - 0.78), demonstrating a statistically significant difference

pt

(P =0.003) in favour of amalgam restorations (Figure 2). For the assessment of the
ce

risk of secondary caries, the fixed effects model was used due to low evidence of
heterogeneity (I2 = 1%; P = 0.39), which presented a RR of 0.23 (95% CI: 018 -
0.30), with a statistically significant difference (P < 0.00001) in favour of amalgams
Ac

(Figure 3). Considering the risk of fracture, the fixed effects model was used due to
the absence of heterogeneity (I2 = 0%; P = 0.77), with no statistically significant
difference between the two types of restoration (P = 0.46), with a RR of 1.24 (95%
CI: 0.71 - 2.16) (Figure 4).

Publication Bias
The funnel plot showed no asymmetry when the failure of the restorations was
analyzed, indicating the possibility of no publication bias (Figure 5).

6
Page 7 of 24
4. Discussion
The aim of this study was to evaluate by means of a systematic review and
meta-analysis the hypothesis of no difference in failure rates between amalgam and
composite resin posterior restorations.
This systematic review identified studies comparing the longevity of amalgam
and composite resin restorations. After the search, only two RCTs were within the
inclusion criteria of this review. The other studies were prospective and retrospective

t
ip
cohort trials. Despite the inclusion of cohort studies in systematic reviews increasing
the amount of data and enabling the consolidation of the clinical reasoning, 22 the

cr
absence or a small number of RCTs can increase the risk of bias. 12,23 Thus, the data
presented in this systematic review should be analyzed and interpreted with caution.

us
The quality analysis performed, based on NOS, and regarded the studies
included as of high quality. However, some studies failed to report important
information in their methodologies, an
such as information about study
participants,11,15,16,18-20 full description of the operative technique,16,19,20 and the
materials used.16,19,20 The absence of these data hinders the interpretation and
M
analysis of these methodological studies.
Despite the use of the more than 150 years in dentistry for tooth restoration,
d

the amalgam is suffering questions in recent years mainly by incorporation of

e

mercury to the metal alloy,3 added to factors such as an increase in demand of

pt

aesthetic procedures and the conception of a conservative dentistry. Therefore, the

composite resin restorations is gaining more space in restorative dentistry. 4,5
ce

However, by requiring a more complex technique6 and a greater chance of

microleakage,9 the replacement of the amalgam must be careful and proportional to
the development and improvement of the physical characteristics of the composites.
Ac

This systematic review revealed that occlusal and occlusoproximal amalgam

posterior restorations have greater clinical longevity when compared to composite
resin restorations. All amalgam alloys used in the studies had a high copper content,
which provides a better clinical performance of the restorations by inhibiting the
gamma-2 phase.24 However, the most recent study included in this review20 was
published in 2013, and this may have influenced the quality of the composites used,
in view of the constant improvement in the physical characteristics and mechanical
performance of the composite resins. Other factors may also influence the
performance and longevity of the restorations, such as: the skill of the operator,

7
Page 8 of 24
materials used, operative technique, field isolation, patient cooperation, and oral
conditions. None of the RCTs included opted for a split-mouth design, which would
be ideal to treat the patients’ oral conditions, such as occlusion, diet and
parafunctional habits.
The results of this meta-analysis were expressed as relative risk (RR), a
statistical analysis often used in binary results, which is defined as the probability of
an event to occur. Regarding restoration failures, this meta-analysis indicated a RR

t
ip
of 0.46 (95% CI: 0.28 - 0.78), i.e. the composite resin restorations have a 46% higher
probability of failure when compared to amalgam restorations.

cr
The study20 that presented one of the highest number of restoration failures
did not report the brands or characteristics of the materials used, which makes the

us
interpretation of the data difficult. However, this study used the USPHS 25 index as a
success criterion, which characterizes restorations that could be repaired or adjusted
an
as absolute failures, as is the case of small fractures or marginal imbalances, which
could explain the high number of failures reported by the article.
This systematic review calculated a 92.8% mean rate of survival for
M
amalgams and 86.2% for the composite resins, with a mean of 55 months of follow-
up. These data are similar to the one reported by a recent systematic review 26 that
d

also compared the longevity of amalgam vs. composite resin, with a mean survival
e

rate of 92.5 and 85.8%, respectively, with a mean of 72 months of follow-up.

However, the review cited included only RCTs, performing a meta-analysis with only
pt

two studies.
ce

The main causes of failures reported in the studies assessed were the
occurrence of secondary caries and fractures of the restorations or teeth, which had
already been previously reported by other studies.26 -28 The presence of secondary
Ac

caries was significantly higher (P < 0.00001) in composite resin restorations. The
oxides formed in the tooth vs. amalgam interface help seal the margins, which may
explain the lower incidence of caries.29 In contrast, factors such as adhesive
technique, adhesive system, polymerization shrinkage, type of dental substrate
(enamel / dentin) and the quality of the hybrid layer can act critically towards
adhesive failure in the composites, thus increasing the risk of recurrent caries. With
regard to fractures, there was no statistically significant difference between the two
materials (P = 0.46), which was also observed in another study,26 indicating a lower
sensitivity of the posterior restorations to fracture when compared to recurrent caries.

8
Page 9 of 24
 In summary, the meta-analysis based on the results of the eight studies
included showed a higher longevity of the amalgam restorations compared to
composite resin restorations. With the exception of one paper,10 all the studies
selected for this review indicated a higher longevity of the amalgam restorations in
posterior teeth.

5. Conclusions

t
ip
The results of this review suggest that posterior composite resin restorations
still have less longevity and a greater number of secondary caries when compared to

cr
amalgam restorations. With regard to fractures, there was no statistically significant
difference between the two restorative materials in relation to the time of follow-

us
up. The data from this review should be interpreted with caution due to the inclusion
of only two RCTs. The publication of a greater number of RCTs based on the
an
CONSORT-statement30 and preferably with a Split-mouth design is crucial for a
better understanding and monitoring of the long-term performance of restorations.
M
References
d

1. Qvist V, Qvist J, Mjör I. Placement and longevity of tooth-colored restorations

e

in Denmark. Acta Odontologica Scandinavica 1990;48(5):305-311.

pt

2. Sjögren P, Halling A. Survival time of Class II molar restorations in relation to

patient and dental health insurance costs for treatment. Swedish Dental
ce

Journal 2002;26(2):59-66.

3. Fuks AB. The use of amalgam in pediatric dentistry. Pediatric Dentistry

2002;24(5):448-455.
Ac

4. Lynch CD, Opdam NJ, Hickel R, Brunton PA, Gurgan S, Kakaboura A, et al.
Guidance on posterior resin composites: Academy of Operative Dentistry -
European Section. Journal of Dentistry 2014;42(4):377-383.

5. Zöchbauer H. Number of dental restorations worldwide. Market Research

Ivoclar Vivadent 2011.

6. Davidson CL, de Gee AJ, Feilzer A. The competition between the composite-
dentin bond strength and the polymerization contraction stress. Journal of
Dental Research 1984;63(12):1396–1399.

9
Page 10 of 24
7. Manhart J, Chen H, Hamm G, Hickel R. Buonocore Memorial Lecture. Review
of the clinical survival of direct and indirect restorations in posterior teeth of
the permanent dentition. Operative Dentistry 2004;29(5):481-508.

8. Heintze SD, Rousson V. Clinical effectiveness of direct class II restorations - a

meta-analysis. Journal of Adhesive Dentistry 2012;14(5):407-431.

9. Mjör IA. The reasons for replacement and the age of failed restorations in
general dental practice. Acta Odontologica Scandinavica 1997;55(1):58-63.

t
ip
10. Wilson MA, Cowan AJ, Randall RC, Crisp RJ, Wilson NH. A practice-based,
randomized, controlled clinical trial of a new resin composite restorative: one-

cr
year results. Operative Dentistry 2002;27(5):423-429.

us
11. Bernardo M, Luis H, Martin MD, Leroux BG, Rue T, Leitão J, DeRouen TA.
Survival and reasons for failure of amalgam versus composite posterior
restorations placed in a randomized clinical trial. Journal of American Dental
Association 2007;138(6):775-783. an
12. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews
M
of Interventions 4.2.6 [updated September 2006]. In: The Cochrane Library
2006;4.

13. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred

d

reporting items for systematic reviews and meta-analyses: the PRISMA

e

statement. Annals of Internal Medicine 2009;151(4):264-269.

pt

14. Wells GA, Shea B, O'Connel D, Peterson J, Welch V, Losos M, et al. The
Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized
ce

studies in meta-analysis. 2000. Available from:

http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp [accessed
22.03.15].
Ac

15. Johnson GH, Bales DJ, Gordon GE, Powell LV. Clinical performance of
posterior composite resin restorations. Quintessence International
1992;23(10):705-711.

16. Mjör IA, Kokstad A. Five-year study of Class II restorations in permanent

teeth using amalgam, glass polyalkenoate (ionomer) cerment and resin-based
composite materials. Journal of Dentistry 1993;21(6):338-343.

17. Collins CJ, Bryant RW, Hodge KL. A clinical evaluation of posterior composite
resin restorations: 8-year findings. Journal of Dentistry 1998;26(4):311-317.

18. Mair LH. Ten-year clinical assessment of three posterior resin composites
10
Page 11 of 24
 and two amalgams. Quintessence International 1998;29(8):483-490.

19. Levin L, Coval M, Geiger SB. Cross-sectional radiographic survey of

amalgam and resin-based composite posterior restorations. Quintessence
International 2007;38(6):511-514.

20. Kim KL, Namgung C, Cho BH. The effect of clinical performance on the
survival estimates of direct restorations. Restorative Dentistry & Endodontics
2013;38(1):11-20.

t
ip
21. Landis JR, Koch GG. The measurement of observer agreement for
categorical data. Biometrics 1977;33(1):159-174.

cr
22. Shrier I, Boivin JF, Steele RJ, Platt RW, Furlan A, Kakuma R, Brophy J,

us
Rossignol M. Should meta-analyses of interventions include observational
studies in addition to randomized controlled trials? A critical examination of
underlying principles. Am J Epidemiol 2007;166(10):1203-1209.
an
23. Pihlstrom BL, Curran AE, Voelker HT, Kingman A. Randomized controlled
trials: what are they and who needs them? Periodontology 2000
M
2000;59(1):14-31.

24. Letzel H, van't Hof MA, Marshall GW, Marshall SJ. The influence of the
amalgam alloy on the survival of amalgam restorations: a secondary analysis
d

of multiple controlled clinical trials. Journal of Dental Research

e

1997:76(11):1787-1798.
pt

25. Roulet JF. Longevity of glass ceramic inlays and amalgam--results up to 6

years. Clinical Oral Investigations 1997;1(1):40-46.
ce

26. Hurst D. Amalgam or composite fillings--which material lasts longer?

Evidence-Based Dentistry 2014;15(2):50-51.
Ac

27. Burke FJ, Wilson NH, Cheung SW, Mjör IA. Influence of patient factors on
age of restorations at failure and reasons for their placement and
replacement. Journal of Dentistry 2001;29(5):317-324.

28. Forss H, Widström E. The post-amalgam era: a selection of materials and

their longevity in the primary and young permanent dentitions. International
Journal of Paediatric Dentistry 2003;13(3):158-164.

29. Grossman ES, Matejka JM. Effect of restorative materials and in vitro carious
challenge on amalgam margin quality. Journal of Prosthetic Dentistry
1996;76(3):239-245.

11
Page 12 of 24
30. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated
guidelines for reporting parallel group randomised trials. BMC Medicine
2010;8:1

t
ip
cr
us
an
M
e d
pt
ce
Ac

12
Page 13 of 24
Table 1. Systematic search strategy (PICOS strategy)

Search strategy

Population #1 Dental caries[MeSH] OR dental restoration failures[MeSH] OR dental

restorations (permanent)[MeSH] OR posterior teeth OR molar[MeSH] OR
premolar[MeSH] OR class I OR class II OR class I cavities OR class II

t
cavities OR occluso cavities OR occlusoproximal cavities.

ip
Intervention #2 Dental restoration[MeSH] OR amalgam restoration OR composite

cr
restoration OR dental amalgam[MeSH] OR dental composite OR dental
composite restoration OR restoration posterior teeth OR composite

us
posterior teeth OR direct class I OR direct class II OR class I restoration
OR class II restoration OR occlusal restoration OR occlusoproximal

Comparisons
restoration.

Amalgam vs. composite resin

an
M
Outcomes #3 Survival OR success OR failure OR longevity OR amalgam longevity
OR resin longevity OR composite resin longevity OR long-term OR follow-
up OR prospective study[MeSH] OR retrospective study OR randomized
d

controlled trial[MeSH] OR controlled trial.

e
pt

Study design Randomized controlled trials, controlled clinical trials, prospective and
retrospective cohort studies
ce

Search combination #1 AND #2 AND #3

Database search
Ac

Language No restriction

Eletronic databases PubMed/MEDLINE, Cochrane Central Register of Controlled Trials and

Web of Science

13
Page 14 of 24
 t
ip
Figure 1 - Flow diagram (PRISMA format) of the screening and selection process.

cr
us
Medline/PubMed Cochrane (CENTRAL) Web of Science
Ide Records identified through Records identified through Records identified through

an
ntifi database searching database searching database searching
cati (n = 938) (n = 89) (n = 172)
on

M
Records excluded
(n = 1178)
Scr
een

d
ing Full-text articles assessed for

Elig
te eligibility (n = 21)
ep
ibilit Full-text articles excluded
y (n = 13)
c

Incl
Ac

ude 13 full-text articles excluded:

Studies included in the
d present meta-analysis 2 review paper
(n = 8) 1 in vitro study
5 resin-modified
1 complex amalgam restoration
4 not reported survival rates

14
Page 15 of 24
 t
ip
Table 2. Main characteristics of selected studies.

cr
us
Study design No. of subjects Age Range Cavity type No. of Adhesive
Author (year) Observation Dropouts (%) Mean Age Cavity size Restorations system
period (years) Tooth type

an
Johnson et al.15 (1992) Prospective 27 NR Occlusal and 40 (AM) Scotchbond L/C
3 15 NR occlusoproximal 88 (CR)
Small, medium,
large

M
Premolar and
molar

Mjor and Mokstad16 (1993) Prospective 142 NR Occlusoproximal 88 (AM) NR

d
3 37 13 Small 91 (CR)
Premolar and

Collins et al.17 (1998)

te
Prospective 72 13-32.4
molar

Occlusal and 52 (AM) Ketac Bond

ep
8 36 16.8 occlusoproximal 161 (CR)
Medium
Premolar and
molar
c

Mair et al.18 (1998) Prospective NR NR Occlusoproximal 60 (AM) Clearfil Bonding

Ac

10 NR NR NR 90 (CR) Agent, Occlusion

NR Bond,
Scotchbond

Wilson et al.10 (2002) RCT 49 18-75 Occlusal and 52 (AM) Singlebond

1 2 35 occlusoproximal 52 (CR)
Medium
Premolar and
molar

15
Page 16 of 24
 t
ip
Bernardo et al.11 (2007) RCT 472 8-12 Occlusal and 856 (AM) Scotchbond Multi-

cr
7 35 NR occlusoproximal 892 (CR) Purpose
Small, medium,
large

us
Premolar and
molar

Levin et al.19 (2007) Prospective 459 18-19 Occlusoproximal 557 (AM) NR

an
1 NR NR NR 93 (CR)
NR

Kim et al.20 (2013) Retrospective 232 NR Occlusal and 139 (AM) NR

M
5 0 NR occlusoproximal 175 (CR)
NR
NR

d
Continuation
te
ep
Author (year) Resin Amalgam / Amalgam Composite Failed / total Secundary caries (%)
restorative Composite resin survival Resin restorations Fracture restoration
technique brand (%) survival (%) (%)
c

Johnson et al.15 (1992) Incremental Dispersalloy / 100 100 0 / 40 (AM) AM = 0 / CR = 0

Ac

Bisfil-P, 0 / 88 (CR) AM = 0 / CR = 0
P-30

Mjor and Mokstad16 (1993) NR Dispersalloy / P-10 95.4 90.1 4 / 88 (AM) AM = 0 / CR = 4.39
9 / 91 (CR) AM = 3.41 / CR = 2.19

Collins et al.17 (1998) Incremental Dispersalloy / 94.2 86.4 3 / 52 (AM) AM = 1.92 / CR = 4.34
Heliomolar, 22 / 161 (CR) AM = 3.84 / CR = 3.72
Herculite XR, P-30

16
Page 17 of 24
 t
ip
Mair et al.18 (1998) Incremental New True 96.6 95.5 2 / 60 (AM) NR

cr
Dentalloy, Solola 4 / 90 (CR) AM = 0 / CR = 0
Nova / Clearfil
Posterior, Occlusin,

us
P-30

Wilson et al.10 (2002) NR Dispersalloy / Z250 98 100 1 / 52 (AM) AM = 0 / CR = 0

0 / 52 (CR) AM = 1.92 / CR = 0

an
Bernardo et al.11 (2007) Incremental Dispersalloy / Z100 94.4 85.6 48 / 856 (AM) AM = 3.7 / CR = 12.7
129 / 892 (CR) AM = 1.9 / CR = 1.8

M
Levin et al.19 (2007) NR NR / NR 88 56 67 / 557 (AM) AM = 8 / CR = 43
47 / 93 (CR) AM = 4 / CR = 1

Kim et al.20 (2013) NR NR 76.3 71.5 33 / 139 (AM) NR

d
50 / 175 (CR) NR

te
No.= number, NR= not reported, CL= class, AM= amalgam, CR= composite resin.
c ep
Ac

17
Page 18 of 24
 t
ip
Table 3. Quality assessment of the studies by the Newcastle-Ottawa scale.

cr
Authors (year) Selection Comparability Outcome

us
Representative Selection Ascertaiment Outcome Comparability Assessment Was Adequacy Total
ness of the of of exposure of of cohorts on of outcome follow-up of follow- 9/9

an
exposed external interest the basis of long up of
cohort control not the design or enough cohorts
present at analysisa for
start outcomes

M
occurb

Johnson et al.15 0 ★ ★ ★ ★0 ★ ★ ★ 7/9

(1992)

d
Mjor and 0 ★ ★ ★ ★0 ★ ★ ★ 7/9
Mokstad16
(1993)
te
ep
Collins et al.17 0 ★ ★ ★ ★0 ★ ★ ★ 7/9
(1998)
c

Mair et al.18 0 ★ ★ ★ ★0 ★ 0 ★ 6/9

(1998)
Ac

Wilson et al.10 0 ★ ★ ★ ★0 ★ ★ 0 6/9

(2002)

Bernardo et al.11 0 ★ ★ ★ ★0 ★ ★ ★ 7/9

(2007)

Levin et al.19 0 ★ ★ ★ ★0 ★ ★ 0 6/9

(2007)

18
Page 19 of 24
 t
ip
Kim et al.20 0 ★ ★ ★ ★★ ★ ★ ★ 7/9

cr
(2013)

a
A study can be awarded a maximum of one star for each item within the selection and outcome categories. A maximum of two star can be given for

us
comparability.
b
Two years of follow-up was chosen to be enough for the outcome survival to occur.

an
M
d
te
c ep
Ac

19
Page 20 of 24
 t
ip
Figure 2. Forest plot for the event restoration failure rate.

cr
us
an
M
d
te
c ep
Ac

20
Page 21 of 24
 t
ip
Figure 3. Forest plot for the event secundary caries.

cr
us
an
M
d
te
c ep
Ac

21
Page 22 of 24
 t
ip
Figure 4. Forest plot for the event fracture.

cr
us
an
M
d
te
c ep
Ac

22
Page 23 of 24
 t
ip
Figure 5. Funnel plot for the studies reporting the outcome event restoration failure rate.

cr
us
an
M
d
te
c ep
Ac

23
Page 24 of 24