journal of dentistry 43 (2015) 1043–1050

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.intl.elsevierhealth.com/journals/jden

Review article

Amalgam and resin composite longevity

of posterior restorations: A systematic
review and meta-analysis

Vittorio Moraschini a,*, Cheung Ka Fai b, Raphael Monte Alto b,

Gustavo Oliveira dos Santos b
a
Fluminense Federal University, Department of Periodontology, School of Dentistry, Rua Mario dos Santos Braga, 30,
Centro, Niterói, Rio de Janeiro CEP 24020-140, Brazil
b
Fluminense Federal University, School of Dentistry, Department of Integrated Clinic, Rua Mario dos Santos Braga, 30,
Centro, Niterói, Rio de Janeiro CEP 24020-140, Brazil

article info abstract

Article history: Objectives: The aim of the present review was to evaluate by means of a systematic review
Received 1 April 2015 and meta-analysis the hypothesis of no difference in failure rates between amalgam and
Received in revised form composite resin posterior restorations.
16 June 2015 Data: Randomized controlled trials, controlled clinical trials and prospective and retrospec-
Accepted 19 June 2015 tive cohort studies were included in this review. The eligibility criteria included clinical trials
in humans with at least 12 months of follow-up comparing the failures rates between
Keywords: occlusal and occlusoproximal amalgam and composite resin restorations. Clinical questions
Restorative dentistry were formulated and organized according to the PICOS strategy.
Longevity Source: An electronic search without restriction on the dates or languages was performed in
Amalgam PubMed/MEDLINE, Cochrane Central Register of Controlled Trials, and Web of Science up
Comopiste resin until March 2015.
Meta-analysis 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 Web of Science.
After an initial assessment and careful reading, 8 studies published between 1992 and 2013
were included in this review. According to the risk of bias evaluation, all studies were
classified as high quality.
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 when compared to
amalgam restorations. In relation to fractures, there was no statistically significant differ-
ence between the two restorative materials regarding the time of follow-up.
Clinical significance: There is currently a worldwide trend towards 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.
# 2015 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +55 2130268190.

E-mail address: vittoriomf@terra.com.br (V. Moraschini).
http://dx.doi.org/10.1016/j.jdent.2015.06.005
0300-5712/# 2015 Elsevier Ltd. All rights reserved.
1044 journal of dentistry 43 (2015) 1043–1050

can influence the restorations performance. Due to these

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 material.1
Although amalgam restorations still have the highest
functional durability,2 its 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 promote greater restoration retention, make
amalgam questionable for conservative dentistry. For these
reasons, the use of composite resins has been increasing
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 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 and occlusoproximal restora-
tions, varying from 1 to 3%. The most frequent reason for
failure are recurrent or secondary marginal restoration
caries,9 thus indicating 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 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, 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
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.
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 Reporting Items for Systematic
Reviews and Meta-Analyses).13 The clinical reasoning was
broken down and organized according to the PICOS strategy.
2.1. Objective
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 amalgam and composite
resin posterior restorations.
2.2. Focused question
What is the longevity of occlusal and occlusoproximal
amalgam and composite resin posterior restorations?
2.3. Search strategy
An unrestricted electronic search of dates or languages was
performed in PubMed/MEDLINE, Cochrane Central Register of
Controlled Trials, and Web of 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.

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 cavities OR occluso
cavities OR occlusoproximal cavities.
Intervention #2 Dental restoration[MeSH] OR amalgam restoration OR composite restoration OR dental amalgam[MeSH] OR dental
composite OR dental composite restoration OR restoration posterior teeth OR composite posterior teeth OR direct
class I OR direct class II OR class I restoration OR class II restoration OR occlusal restoration OR occlusoproximal
restoration.
Comparisons Amalgam vs. composite resin
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 controlled
trial[MeSH] OR controlled trial.
Study design Randomized controlled trials, controlled clinical trials, prospective and retrospective cohort studies
Search combination #1 AND #2 AND #3
Database search
Language No restriction
Eletronic databases PubMed/MEDLINE, Cochrane Central Register of Controlled Trials and Web of Science
 journal of dentistry 43 (2015) 1043–1050
2.4. Selection criteria
This review searched for randomized controlled trials
(RCTs), controlled clinical trials and prospective and retro-
spective 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.
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
(k). The authors of the studies, when necessary, were
contacted by e-mail to answer any questions.
2.6. Quality assessment
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 include non-random-
ized, 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 comparison category, two
stars/points may be assigned. According to NOS, the maxi-
mum score assigned to a study is nine stars/points (highest
scientific evidence). Studies scoring 6 stars and above are
regarded as high quality.
2.7. Data extraction
The following data were extracted from the studies included
(when available): 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, adhesive system, restoration technique, amal-
gam 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%. 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
1045
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 signifi-
cance was set at P < 0.05.
Publication bias was graphically explored through a funnel
plot. The asymmetry in the funnel plot can indicate possible
publication bias.
The data were analyzed using statistical software Review
Manager (version 5.2.8; The Nordic Cochrane Center, The
Cochrane Collaboration, Copenhagen, Denmark, 2014).
3. Results
3.1. Literature search
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 the initial assessment, 21 full
articles were selected. After careful reading, 13 studies 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. The reasons for excluding
the studies and the selection process can be viewed in Fig. 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, 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 studies,15–19 and one
retrospective cohort study20 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 ranged from
1210,19 to 12018 months, with a mean of 55 months. Three
thousand, four hundred and eighty-six occlusal and occluso-
proximal 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 dentine.
Only four studies11,15,17,18 described the technique used to
insert the composite resin and control of the polymerization
contraction. No study reported the mechanisms used for the
isolation of the operative field.
3.3. Quality assessment
All studies had a score 6 stars and were classified as
high quality. The scores of each study are summarized in
Table 3.
1046 journal of dentistry 43 (2015) 1043–1050
[(Fig._1)TD$IG]

Fig. 1 – Flow diagram (PRISMA format) of the screening and selection process.

3.4. Meta-analysis
The mean survival of amalgam and composite resin varied
from 76.320 to 100%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
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 0.46 (95% CI: 0.28–0.78),
demonstrating a statistically significant difference
(P = 0.003) in favour of amalgam restorations (Fig. 2). For
the assessment of the 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 (Fig. 3). Considering the risk of
[(Fig._2)TD$IG]
fracture, the fixed effects model was used due to the
absence of heterogeneity (I2 = 0%; P = 0.77), with no statisti-
cally significant difference between the two types of
restoration (P = 0.46), with a RR of 1.24 (95% CI: 0.71–2.16)
(Fig. 4).
3.4.1. Publication bias
The funnel plot showed no asymmetry when the failure of the
restorations was analyzed, indicating the possibility of no
publication bias (Fig. 5).
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

[(Fig._3)TD$IG] Fig. 2 – Forest plot for the event restoration failure rate.

Fig. 3 – Forest plot for the event secondary caries.

 journal of dentistry 43 (2015) 1043–1050 1047

Table 2 – Main characteristics of selected studies.

Author (year) Study design No. of subjects Age range Cavity type cavity size No. of Adhesive
observation dropouts (%) mean age tooth type restorations system
period (years)
Johnson et al.15 (1992) Prospective 27 NR Occlusal and occlusoproximal 40 (AM) Scotchbond L/C
3 15 NR Small, medium, large 88 (CR)
Premolar and molar
Mjor and Mokstad16 Prospective 142 NR Occlusoproximal 88 (AM) NR
(1993)
3 37 13 Small 91 (CR)
Premolar and molar
Collins et al.17 (1998) Prospective 72 13–32.4 Occlusal and occlusoproximal 52 (AM) Ketac Bond
8 36 16.8 Medium 161 (CR)
Premolar and molar
Mair et al.18 (1998) Prospective NR NR Occlusoproximal 60 (AM) Clearfil Bonding
Agent, Occlusion
Bond, Scotchbond
10 NR NR NR 90 (CR)
NR
Wilson et al.10 (2002) RCT 49 18–75 Occlusal and occlusoproximal 52 (AM) Singlebond
1 2 35 Medium 52 (CR)
Premolar and molar
Bernardo et al.11 (2007) RCT 472 8–12 Occlusal and occlusoproximal 856 (AM) Scotchbond
Multi-Purpose
7 35 NR Small, medium, large 892 (CR)
Premolar and molar
Levin et al.19 (2007) Prospective 459 18–19 Occlusoproximal 557 (AM) NR
1 NR NR NR 93 (CR)
NR
Kim et al.20 (2013) Retrospective 232 NR Occlusal and occlusoproximal 139 (AM) NR
5 0 NR NR 175 (CR)
NR

Author (year) Resin Amalgam/composite Amalgam Composite Failed/total Secundary

restorative resin brand survival (%) resin restorations caries (%) fracture
technique survival (%) restoration (%)
Johnson et al.15 (1992) Incremental Dispersalloy/Bisfil-P, P-30 100 100 0/40 (AM) AM = 0/CR = 0
0/88 (CR) AM = 0/CR = 0
Mjor and Mokstad16 NR Dispersalloy/P-10 95.4 90.1 4/88 (AM) AM = 0/CR = 4.39
(1993)
9/91 (CR) AM = 3.41/CR = 2.19
Collins et al.17 (1998) Incremental Dispersalloy/Heliomolar, 94.2 86.4 3/52 (AM) AM = 1.92/CR = 4.34
Herculite XR, P-30
22/161 (CR) AM = 3.84/CR = 3.72
Mair et al.18 (1998) Incremental New True Dentalloy, 96.6 95.5 2/60 (AM) NR
Solola Nova/Clearfil
Posterior,
Occlusin, P-30
4/90 (CR) AM = 0/CR = 0
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
Bernardo et al.11 Incremental Dispersalloy/Z100 94.4 85.6 48/856 (AM) AM = 3.7/CR = 12.7
(2007)
129/892 (CR) AM = 1.9/CR = 1.8
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
50/175 (CR) NR

No. = number, NR = not reported, CL = class, AM = amalgam, CR = composite resin.

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 retro-
spective 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
absence or a small number of RCTs can increase the risk of
1048 journal of dentistry 43 (2015) 1043–1050

bias.12,23 Thus, the data presented in this systematic review

Total 9/9

7/9
7/9
7/9
6/9
6/9
7/9
6/9
7/9
should be analyzed and interpreted with caution.
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
Adequacy of
follow-up of
cohorts
methodologies, such as information about study partici-

$
$
$
$

$
pants,11,15,16,18–20 full description of the operative tech-

0
nique,16,19,20 and the materials used.16,19,20 The absence of
these data hinders the interpretation and analysis of these

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 comparability.
methodological studies.
Was follow-up

for outcomes
long enough

Despite the use of the more than 150 years in dentistry for
Outcome

occurb

tooth restoration, the amalgam is suffering questions in recent

$
$
$

$
$
$
$
years mainly by incorporation of mercury to the metal alloy,3
0

added to factors such as an increase in demand of aesthetic

procedures and the conception of a conservative dentistry.
Therefore, the composite resin restorations is gaining more
Assessment
of outcome

space in restorative dentistry.4,5 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.
basis of the design
Comparability of

This systematic review revealed that occlusal and occlu-

cohorts on the
Comparability

or analysisa

soproximal amalgam posterior restorations have greater

clinical longevity when compared to composite resin restora-
$$
$0
$0
$0
$0
$0
$0
$0

tions. 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
Outcome of
interest not
present at

composites used, in view of the constant improvement in

start
$
$
$
$
$
$
$
$

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
Table 3 – Quality assessment of the studies by the Newcastle-Ottawa scale.

skill of the operator, materials used, operative technique, field

Ascertaiment

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

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.
Selection

The results of this meta-analysis were expressed as relative

of external

risk (RR), a statistical analysis often used in binary results,

Selection

control

which is defined as the probability of an event to occur.

$
$
$
$
$
$
$
$

Regarding restoration failures, this meta-analysis indicated a

RR of 0.46 (95% CI: 0.28–0.78), i.e. the composite resin
restorations have a 46% higher probability of failure when
Representativeness

compared to amalgam restorations.

of the exposed

The study20 that presented one of the highest number of

cohort

restoration failures did not report the brands or characteristics

0
0
0
0
0
0
0
0

of the materials used, which makes the interpretation of the

data difficult. However, this study used the USPHS25 index as a
success criterion, which characterizes restorations that could
be repaired or adjusted as absolute failures, as is the case of
small fractures or marginal imbalances, which could explain
Mjor and Mokstad16 (1993)

the high number of failures reported by the article.

Bernardo et al.11 (2007)
Johnson et al.15 (1992)

Wilson et al.10 (2002)

Collins et al.17 (1998)

This systematic review calculated a 92.8% mean rate of

Levin et al.19 (2007)
Mair et al.18 (1998)

Kim et al.20 (2013)

survival for amalgams and 86.2% for the composite resins,

Authors (year)

with a mean of 55 months of follow-up. These data are similar

to the one reported by a recent systematic review26 that also
compared the longevity of amalgam vs. composite resin, with
a mean survival rate of 92.5 and 85.8%, respectively, with a
b
a

mean of 72 months of follow-up. However, the review cited

 journal of dentistry 43 (2015) 1043–1050
[(Fig._4)TD$IG]
Fig. 4 – Forest plot for the event fracture.
[(Fig._5)TD$IG]
Fig. 5 – Funnel plot for the studies reporting the outcome
event restoration failure rate.
included only RCTs, performing a meta-analysis with only two
studies.
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 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/dentine) 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.
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 restora-
tions. 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
The results of this review suggest that posterior composite
resin restorations still have less longevity and a greater
1049
number of secondary caries when compared to amalgam
restorations. With regard to fractures, there was no statisti-
cally significant difference between the two restorative
materials in relation to the time of follow-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 CONSORT-statement30 and
preferably with a Split-mouth design is crucial for a better
understanding and monitoring of the long-term performance
of restorations.
references
1. Qvist V, Qvist J, Mjör I. Placement and longevity of tooth-
colored restorations in Denmark. Acta Odontologica
Scandinavica 1990;48:305–11.
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 Journal
2002;26:59–66.
3. Fuks AB. The use of amalgam in pediatric dentistry. Pediatric
Dentistry 2002;24:448–55.
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;42:377–83.
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;63:1396–9.
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;29:481–508.
8. Heintze SD, Rousson V. Clinical effectiveness of direct class
II restorations – a meta-analysis. Journal of Adhesive Dentistry
2012;14:407–31.
9. Mjör IA. The reasons for replacement and the age of failed
restorations in general dental practice. Acta Odontologica
Scandinavica 1997;55:58–63.
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-year results. Operative
Dentistry 2002;27:423–9.
11. Bernardo M, Luis H, Martin MD, Leroux BG, Rue T, Leitão J,
et al. Survival and reasons for failure of amalgam versus
composite posterior restorations placed in a randomized
clinical trial. Journal of American Dental Association
2007;138:775–83.
12. Higgins JPT, Green S, editors. Cochrane handbook for
systematic reviews of interventions 4.2.6 [updated September
2006]. The Cochrane Library; 2006. p. 4.
1050 journal of dentistry 43 (2015) 1043–1050
13. Moher D, Liberati A, Tetzlaff J, Altman DG. PRISMA Group.
Preferred reporting items for systematic reviews and meta-
analyses: the PRISMA statement. Annals of Internal Medicine
2009;151:264–9.
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 studies in meta-analysis. 2000.
Available from: http://www.ohri.ca/programs/
clinical_epidemiology/oxford.asp [accessed 22.03.15].
15. Johnson GH, Bales DJ, Gordon GE, Powell LV. Clinical
performance of posterior composite resin restorations.
Quintessence International 1992;23:705–11.
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;21:338–43.
17. Collins CJ, Bryant RW, Hodge KL. A clinical evaluation of
posterior composite resin restorations: 8-year findings.
Journal of Dentistry 1998;26:311–7.
18. Mair LH. Ten-year clinical assessment of three posterior
resin composites and two amalgams. Quintessence
International 1998;29:483–90.
19. Levin L, Coval M, Geiger SB. Cross-sectional radiographic
survey of amalgam and resin-based composite posterior
restorations. Quintessence International 2007;38:511–4.
20. Kim KL, Namgung C, Cho BH. The effect of clinical
performance on the survival estimates of direct restorations.
Restorative Dentistry & Endodontics 2013;38:11–20.
21. Landis JR, Koch GG. The measurement of observer
agreement for categorical data. Biometrics 1977;33:159–74.
22. Shrier I, Boivin JF, Steele RJ, Platt RW, Furlan A, Kakuma R,
et al. Should meta-analyses of interventions include
observational studies in addition to randomized controlled
trials? A critical examination of underlying principles.
American Journal of Epidemiology 2007;166:1203–9.
23. Pihlstrom BL, Curran AE, Voelker HT, Kingman A.
Randomized controlled trials: what are they and who needs
them? Periodontology 2000;59:14–31.
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 of multiple controlled
clinical trials. Journal of Dental Research 1997;76:1787–98.
25. Roulet JF. Longevity of glass ceramic inlays and amalgam –
results up to 6 years. Clinical Oral Investigations 1997;1:40–6.
26. Hurst D. Amalgam or composite fillings – which material
lasts longer? Evidence-Based Dentistry 2014;15:50–1.
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;29:317–24.
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;13:158–64.
29. Grossman ES, Matejka JM. Effect of restorative materials and
in vitro carious challenge on amalgam margin quality.
Journal of Prosthetic Dentistry 1996;76:239–45.
30. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement:
updated guidelines for reporting parallel group randomised
trials. BMC Medicine 2010;8:1.