journal of dentistry 40 (2012) 703–710

Available online at www.sciencedirect.com

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

Amalgam or composite resin? Factors influencing the choice

of restorative material

M.B. Correa a, M.A. Peres b, K.G. Peres b, B.L. Horta c, A.D. Barros c, F.F. Demarco a,c,*
a
Department of Operative Dentistry, Faculty of Dentistry, Federal University of Pelotas, RS, Brazil
b
Oral Epidemiology and Dental Public Health Research Group, Post-Graduate Program in Public Health, Health Sciences Center,
Federal University of Santa Catarina, SC, Brazil
c
Post-Graduate Program in Epidemiology, Federal University of Pelotas, RS, Brazil

article info abstract

Article history: Objectives: This study aimed to investigate the patient and tooth factors associated with
Received 14 December 2011 selection of restorative material in direct posterior restorations in young adults from a
Received in revised form population-based birth cohort.
20 April 2012 Methods: A representative sample (n = 720) of all 5914 births occurring in Pelotas in 1982
Accepted 23 April 2012 were prospectively investigated, and posterior restorations were assessed in 2006, when the
patients were 24 years old. Tooth-related variables (individual level) included restorative
material (amalgam or composite), type of tooth, size of cavity, and estimated time in mouth.
Keywords: Data regarding demographic and socio-economic characteristics, oral health, and service
Epidemiology utilization patterns during the life course were also assessed (contextual level).
Cohort studies Results: Logistic Regression Multilevel models showed that individuals who have accessed
Dental restorations dental services by private insurance by age 15 [odds ratio (OR) = 1.66 (0.93–2.95)] and who had
Posterior teeth a higher dental caries index at age 15 (high DMFT tertile) [OR 2.89 (1.59–5.27)] presented more
Dental materials amalgam restorations in the posterior teeth. From tooth-level variables, the frequency of
amalgams decreases with increasing number of surfaces enrolled in the cavity preparation
( p < 0.001) and was almost 5 times greater in molars than in premolars.
Conclusions: The present findings suggest that variables related to type of dental service,
dental caries (higher DMFT index), and cavity characteristics (tooth type, size) determine the
choice of dentists for restorative materials. Other individual characteristics such as demo-
graphic and socioeconomic status have not influenced this choice.
Clinical significance: This is the first population-based study that assesses the determinant
factors for the choice of dentists for composite or amalgam in posterior direct restorations,
showing that, independently of socioeconomic and demographic characteristics, type of
payment of dental services and clinical factors are associated with this choice.
# 2012 Published by Elsevier Ltd.

1. Introduction Although indirect restorations may have more longevity,

direct restorations continue to be the first choice of dentists to
Despite the decrease in caries prevalence in many countries, treat posterior decayed teeth because of the lower need for
there is still a high need for posterior restorative treatment. sound tooth removal, advances in direct restorative materials

* Corresponding author at: Universidade Federal de Pelotas, Faculdade de Odontologia, Rua Gonçalves Chaves, 457, 5o andar – Centro, CEP:
96015568 Pelotas, RS, Brazil. Tel.: +55 53 3222 66 90x135; fax: +55 53 3222 66 90x135.
E-mail address: flavio.demarco@pq.cnpq.br (F.F. Demarco).
0300-5712/$ – see front matter # 2012 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.jdent.2012.04.020
704 journal of dentistry 40 (2012) 703–710

technology that has increased materials survival, and the low interviewed and examined for several oral health conditions,
cost of direct restorations compared to that of indirect ones.1 such as dental caries, periodontal status, soft tissue oral
Dental amalgam has been the predominant preferred restor- lesions, and quality of posterior restorations. Methodological
ative material for posterior teeth during the twentieth century; details of this study are available in a recent publication.15
it presents some advantages, including relatively low cost,
longevity, and less sensitivity to clinical techniques, than the 2.1. Outcome and tooth-level variables
other direct restorative materials used for this purpose.
Despite these material strengths, amalgam also presents The outcome of the present study was the restorative material
some important limitations, such as the absence of adhesion used in posterior restorations—amalgam or composite.
to dental tissues, which is manifested in the need to remove Posterior restorations were also assessed according to: (1)
sound tooth structure, and aesthetics.2 tooth location: molars or premolars; (2) type of cavity: class
Since the composite resins became a viable alternative to and number of surfaces affected; (3) estimated time in mouth,
posterior restorations, a dramatic change in the election of reported by the individual: up to 1 year, 1–5 years, 6–10 years,
restorative material occurred, not only due to patients’ priority or more than 10 years. The time reported by the individual was
on aesthetics, but also because, in combination with the total- collected in the interview, when the categories above
etch acid technique, composites present adhesive properties mentioned were presented as answer options.
to the dental structure, restricting the cavity preparation to the
removal of carious tissue, which is not possible with amalgam 2.2. Personal-level variables
restorations.3 Regarding longevity, clinical trials have shown
that amalgam restorations are comparable to or slightly better The independent variables that were used in this study were
than composite resin restorations.4,5 In a recent study a good obtained from the different assessments made in this cohort.
clinical performance of posterior composite restorations was Skin colour was referred by the individual as white, brown, or
observed, with an annual failure rate varying between 1.5 and black and then the variable was dichotomized in whites and
2.2% after 22 years of follow-up for two different composites.6,7 blacks/browns. Maternal schooling was collected in years of
Considering the possibilities of restorative materials avail- successful studies and then categorized into four groups: 12;
able for posterior teeth, some studies have investigated which 9–11; 5–8; and 4 years. Family income data were collected in
factors would be associated with a dentist’s election, demon- 1982, using five categories of Brazilian minimum wages (<1, 1–
strating that variables related to the tooth, the patient and the 3, 3.1–6, 6.1–10, and >10). Unfortunately, information on the
dentist could influence the material choice.8–12 However, most continuous level of income was not available because the
of these studies were conducted before or around 2000, when variable was already collected in the five categories above. To
amalgam use was highly prevalent; furthermore, none of the classify families into tertiles for the data analyses, it was
previous studies was population-based.8,9 necessary to regroup the five categories. A principal compo-
Therefore, the aim of the present study was to investigate nent analysis was carried out using four variables strongly
the patient- and tooth-related factors associated with dentists’ related to wealth in our sample—delivery care payment mode
selection for restorative material in direct posterior restora- (out-of-pocket, public free or private health insurance),
tions in young adults from a population-based birth cohort. schooling, height, and mother’s skin colour. The groups did
not have exactly the same number of individuals due to ties in
the derived score. After this, the second and third tertiles were
2. Methods consolidated in one category (non-poor), while the first tertile
was referred to as the ‘‘poor’’ category.16
Pelotas is a medium-sized city located in a relatively affluent Family incomes were also collected at ages 15 and 23 in a
area in southern Brazil. All infants born at three maternity continuous level, and subjects were divided into tertiles. The
hospitals in the city were identified in 1982, and the 5914 live- middle and higher tertiles were merged into a group referred
born infants and their mothers were weighed and measured; to as ‘‘non-poor,’’ while the lower tertile was referred to as
then the mothers were interviewed. This population has been ‘‘poor.’’ We performed a group-based trajectory analysis to
followed-up several times since then. A detailed explanation estimate family income trajectory groups with the PROCTRAJ
about methodological procedures is available elsewhere.13 In macro in SAS version 9.1 (SAS Institute Inc., Cary, NC). The
1997 (when individuals were 15 years of age), a systematic combination of this classification resulted in four different
sample of 27% of the total of census tracts (70) was selected, family income trajectories from birth to 23 years of age: (1)
and every household in these tracts was visited. A total of 1076 those who were always poor; (2) those who were never poor; (3)
cohort members were found, and a sample of 900 individuals those who were poor at birth and non-poor later on (upwardly
was randomly selected for the Oral Health Study (OHS-97), mobile); and (4) those who were non-poor at birth and then
which consisted of (1) dental examinations to evaluate the became poor (downwardly mobile).17
presence of dental caries, malocclusion, and soft tissue lesions Dental service payment mode (out-of-pocket, public free or
in the adolescents, and (2) an interview, containing questions private health insurance) was verified in Oral Health Studies,
about oral hygiene habits and dental service utilization. The at 15 and 24 years of age. Dental caries at age 24 was
diagnosis criteria for dental caries followed the World Health determined by the DMFT14 index, which was divided into
Organization’s (WHO) guidelines.14 tertiles for analysis.
In 2006, at 24 years of age, the 888 adolescents (98.7%) A team of six dentists and four advanced dental students
evaluated in the OHS-97 were assessed again. Individuals were from the Federal University of Pelotas (UFPel) participated in
 journal of dentistry 40 (2012) 703–710 705

the fieldwork. Examiners and interviewers were trained and

calibrated following the methodology previously described.18
Examiner reliabilities were calculated using weighted and
simple kappa tests and the intraclass correlation coefficient,
when appropriate. The lowest kappa value for restorations
variables (restorative material, size of cavity) was 0.70,
showing a good reliability. For quality data control, 10% of
the interviews were repeated by telephone with a short
version of the questionnaire, allowing concordance level
calculation.

2.3. Data analysis

Fig. 1 – Relative frequencies (%) of posterior restorations by
The software STATA version 11.0 (2009, Stata Corporation,
restorative material and estimated time in mouth in a
College Station, TX, USA) was used for the analysis. A
sample of young adults of 24 years. of age in Pelotas, RS,
descriptive analysis was made to assess the distribution of
Brazil. n = 1955 restorations.
restorative materials in posterior restorations. Associations
between variables were tested using the chi-squared test and
the chi-squared test for linear trends, when appropriate. To
determine the factors associated with the material used for fewer years of education at the time of childbirth ( p = 0.019). In
posterior restorations, a multilevel analysis model was addition, amalgam restorations were more commonly ob-
utilized to adjust the results considering the effects of served in individuals who accessed dental services by private
variables from both individual and tooth levels. A logistic insurance or public free at ages 15 ( p = 0.025) and 24 ( p = 0.034)
multilevel regression analysis considered two levels of data and in those presenting a greater DMFT index at age 24
organization: tooth-related variables and patient-related ( p < 0.001). From tooth-level variables, the frequency of
variables. The odds ratio and 95% confidence interval were amalgams decreased with increasing number of surfaces
estimated. Variability at levels 1 and 2 were represented in the enrolled in the cavity preparation ( p < 0.001) and was greater
model as random coefficients. Model selection was carried out in molars than in premolars ( p < 0.001). Fig. 1 shows the
using deviance-based significance testing. distribution of posterior restorations by estimated time in
To determine their entrance in the multivariable model, mouth, as reported by the participants. It was observed that
variables were also grouped in four blocks as follows: Block 1: the number of amalgam restorations substantially decreases
socioeconomic and demographic factors at birth and socio- in relation to composites from older to newer restorations
economic trajectory; Block 2: variables collected at age 15 ( p < 0.001).
(dental service payment mode); Block 3: variables collected at Table 2 displays the results of the crude and adjusted
age 24 (dental caries and dental service payment model); and Multilevel Logistic Regression analyses for the presence of
Block 4: tooth-level variables. The only variables that present a amalgam restorations. In the final model, from personal-level
p-value < 0.20 in the crude analysis were kept in the final variables, dental service payment mode at age 15 and DMFT
model. index at age 24 maintained the association with posterior
restorative material found in crude analysis. Individuals who
2.4. Ethical issues accessed dental services by private insurance at age 15
presented an odds ratio that was 136% greater for amalgam
This project was approved by the UFPel Ethics Committee. All restorations compared with those accessed dental services
the examinations and interviews were performed with out-of-pocket. In addition, the odds of the presence of
individual authorization after participants signed informed amalgam restorations was almost three times greater in
consent forms. Individuals with treatment needs were individuals who had a higher dental caries index compared
referred to the Dental Clinic of the Graduate Program in with those located in the lower tertile of dental caries. From
Dentistry, Federal University of Pelotas. tooth-level variables, molars presented an odds ratio almost
five times greater than premolars, while restorations with four
or more surfaces presented a reduction of 75% in the odds
3. Results compared with restorations with one surface.

A total of 720 individuals were dentally examined (response

rate of 80% of OHS-97), and 503 (69.9%) of them presented at 4. Discussion
least one posterior restoration in the mouth. From the total of
2135 restorations observed, 943 (43.9%) were composites, and To the best of our knowledge, this is the first population-based
1207 (56.1%) were amalgam. Table 1 displays the distribution study that assessed factors associated with the choice of
of restorations by personal and tooth-level variables with restorative material in posterior teeth. Although the assess-
bivariate analysis. Amalgam restorations were found with ment of posterior restorations was cross-sectional, the
more frequency in males ( p = 0.100), blacks/browns present study was nested in a representative sample of a
( p = 0.001), and in individuals whose mothers had four or birth cohort monitored regularly since 1982,13 providing
706 journal of dentistry 40 (2012) 703–710

Table 1 – Association between posterior restorative material and socio-economic, dental service utilization, oral health,
and tooth in a sample of young adults of 24 years of age in Pelotas, RS, Brazil. Bivariate analysis (n = 503 individuals; 2135
restorations).
Variable/category Restorative material, n (%)

Tooth coloured Amalgam p-Value

Total 943 (43.9) 1207 (56.1)

Level 2 – individual
Sex 0.010
Male 401 (41.3) 570 (58.7)
Female 541 (46.9) 613 (53.1)
Skin colour 0.001
White 815 (45.0) 995 (55.0)
Black/Brown 89 (34.5) 169 (65.5)
Family’s socio-economic trajectory from birth to age 23 0.134
Never poor 590 (45.8) 699 (54.2)
Downwardly mobile 170 (40.9) 246 (59.1)
Upwardly mobile 98 (43.2) 129 (56.8)
Always poor 85 (39.0) 133 (61.0)
Mother’s educational level at birth (years) 0.019*
12 196 (48.4) 209 (51.6)
9–11 120 (44.8) 148 (55.2)
5–8 434 (44.7) 538 (55.3)
0–4 188 (40.0) 282 (60.0)
Dental service payment mode at age 15 0.025
Out-of-pocket 409 (52.1) 376 (47.9)
Private health insurance 270 (37.0) 460 (63.0)
Public free 252 (43.8) 324 (56.2)
Dental caries (DMFT) at age 24 (tertiles) <0.001
1 229 (54.8) 189 (45.2)
2–4 327 (42.2) 448 (57.8)
>5 386 (41.4) 546 (58.6)
Dental service payment mode at age 24 0.034
Out-of-pocket 341 (51.4) 323 (48.6)
Private health insurance 215 (46.7) 245 (53.3)
Public free 108 (42.2) 148 (57.8)

Level 1 – restoration
Dental group <0.001
Pre-molars 248 (65.1) 133 (34.9)
Molars 695 (39.3) 1074 (60.7)
Type of cavity (class)
I (1 surface) 652 (40.6) 955 (59.4) <0.001
I (>1 surface) 52 (38.0) 85 (62.0)
II (1 proximal surface) 159 (56.6) 122 (43.4)
II (3 surfaces) 41 (59.4) 28 (40.6)
II (4 surfaces) 39 (70.9) 16 (29.1)
Estimated time in mouth
Up to 1 year 182 (66.4) 92 (33.6) <0.001
1–5 years 345 (59.0) 240 (41.0)
6–10 years 257 (35.0) 477 (65.0)
More than 10 years 35 (9.7) 327 (90.3)

reliable data about each individual’s life course and thus restorations are cheaper.25 Thus, health insurance companies
reinforcing our findings. In addition to teeth-related variables, are expected to promote the use of dental amalgam in posterior
the individual level of dental caries and dental service teeth to improve their profits. Although free public dental care
payment mode influenced the choice of restorative material has not shown statistical association with restorative material,
for posterior teeth. it was possible to observe the same tendency in terms of
Individuals who accessed dental services at age 15 by private amalgam selection [OR = 1.66 CI95% (0.93–2.95)], as opposed to
health insurance were more prone to having their teeth filled private services assessed by out-of-pocket. Since posterior
with dental amalgam than whose accessed dental services out- restorations do not affect aesthetic factors, from a financial
of-pocket. Despite the good performance of composites showed perspective, the use of amalgam should still be recommended
in recent studies, 19–22 dental amalgam has been considered as a good alternative, mainly in public services, which are
more cost effective in comparison with composites,23,24 once funded exclusively by the government.23 The fact that private
the longevity of both materials is comparable,5 but amalgam practitioners use relatively more composite materials may
 journal of dentistry 40 (2012) 703–710 707

Table 2 – Crude (c) and adjusted (a) odds ratios (OR) for independent variables from tooth and personal levels and presence
of amalgam in posterior restorations in a sample of young adults of 24 years of age in Pelotas, RS, Brazil. Multilevel
analysis (n = 466 individuals; 2008 restorations).
Variable/category RO (c) (CI95%) p-Value RO (a) (CI95%) p-Value
Level 2 – individual
Sex 0.328 – –
Male 1 1
Female 0.79 (0.50–1.26)
Skin colour 0.034 0.073
White 1 1
Black/Brown 2.08 (1.06–4.08) 1.88 (0.94–3.75)
Family’s socio-economic trajectory from birth to age 15 0.391 – –
Never poor 1
Downwardly mobile 1.59 (0.86–2.96)
Upwardly mobile 1.05 (0.50–2.19)
Always poor 1.52 (0.75–3.12)
Mother’s educational level at birth (years) 0.406 – –
12 1
9–11 1.65 (0.68–3.98)
5–8 1.57 (0.80–3.09)
0–4 1.88 (0.90–3.95)
2 log likelihood (Block 1) 2414.6

Dental service payment mode at age 15 0.002 0.011

Out-of-pocket 1 1
Private health insurance 2.70 (1.54–4.72) 2.36 (1.34–4.17)
Public free 1.82 (1.04–3.19) 1.66 (0.93–2.95)
2 log likelihood (Block 1 + Block 2) 2353.4

Dental caries (DMFT) at age 24 (tertiles) 0.002 0.001

0–1 1 1
2–4 2.13 (1.19–3.82) 2.45 (1.35–4.44)
>5 2.11 (1.17–3.81) 2.89 (1.59–5.27)
Dental service payment mode at age 24 0.226 – –
Out-of-pocket 1
Private health insurance 1.73 (0.87–3.46)
Public free 1.65 (0.74–3.65)
2 log likelihood (individual level) 2347.4

Level 1 – restoration
Dental group <0.001 <0.001
Pre-molars 1 1
Molars 4.72 (3.32–6.69) 4.79 (3.31–6.93)
Type of cavity (class) <0.001 <0.001
I (1 surface) 1 1
I (>1 surface) 0.98 (0.59–1.63) 0.73 (0.43–1.24)
II (1 proximal surface) 0.50 (0.34–0.72) 0.65 (0.44–0.96)
II (3 surfaces) 0.49 (0.25–0.97) 0.47 (0.22–0.97)
II (4 surfaces) 0.25 (0.11–0.57) 0.24 (0.11–0.54)
2 log likelihood (restoration level + individual level) 2239.6
Variables that presented a p-value > 0.200 were not included in the final model.

reflect a difference in the patient populations in private practice, restorative material. This fact can be explained by the
where the demand for aesthetics is greater for attending increased use of composites for posterior restorations world-
individuals with a high income level, while in public dental wide.26,27 In fact, the use of dental amalgam is decreasing
health practice, despite having universal access in Brazil, dramatically,8 and the same tendency could be observed in our
mainly low income population is attended.10 Furthermore, the findings. Despite the lack of consistent evidence about the
prices for composite restoration placement are higher than damage caused by mercury present in amalgam,28,29 some
those for amalgam and; despite the placement of composites governments have recommended reducing the use of this
takes longer than amalgam, the gains for the professional are material, and others have decided to ban it from dental
increased. Actually, companies are investing strongly in the practice entirely.9 Moreover, in Netherlands, the teaching of
development of new composites in detriment of amalgam, and amalgam restoration for undergraduate students has been
the pressure exercised by the market probably influences the completely halted at some dental schools for almost one
choice of dentists. decade.26 In other countries, teaching of composite has
In contrast, the mode of payment of dental services substantially increases in last yeas, and students often gain
reported at the age of 24 did not show any association with more experience in the placement of posterior composite than
708 journal of dentistry 40 (2012) 703–710

amalgam.30–32 In addition to the potential risk of mercury, the the increase in cavity size, especially when marginal ridges are
main arguments in favour of universal composite use are the removed.41 Our findings also reinforce the belief that adhesive
adoption of a more conservative approach, preserving direct composite restoration will help to reinforce a highly
more dental structure due to the adhesive dentistry develop- fragile tooth, where a considerable amount of tooth structure
ment, and the aesthetic benefit of these materials, given that was removed.27,28 However, dentists should be aware that
patients seeking dental treatment are now more concerned despite the improvement in resistance to fracture, there is a
with aesthetics.26,27 highly increased chance of composite restoration failure when
Even though the use of composites is increasing consis- place in larger cavities; this increased failure rate is directly
tently, in our sample, amalgam restorations are still more proportional to the number of surfaces enrolled in the cavity
common when all restorations were considered. Considering preparation.6,42 Nevertheless, our data differ from a recent
the different periods, amalgam represents more than 90% of study’s findings that did not find significant differences for the
restorations with more than 10 years in the mouth, and an selection of restorative material in relation to cavity size.8
inversion could be observed for restorations placed more The variable of estimated time in mouth reported by the
recently (Fig. 1), being the composite restorations more individual was excluded from the multilevel analysis because
prevalent (approximately two thirds of restorations) in periods considerable information loss was detected (186 restorations),
of up to one year of placement. It is possible that this variable, resulting in many clusters with fewer than 2 units, which is
which was reported by the individuals for each restoration, not recommended for this type of analysis.43 The multilevel
have presented some lack of precision due to the difficulty to approach allows the achievement of reliable estimations
remember the date when the restorations were placed. when two levels of variables were analysed simultaneously,44
However, the adoption of categorical option in data collection avoiding incorrect and potentially misleading results that
and the nature of this information, which not lead to could be generated by bivariate and single-level analysis.
estimative errors in the same direction for all individuals, Other strengths of this study include high repeatability, few
reduce the possibility of bias. missing data, and blinding of interviewers and examiners, all
Previous caries experience was also associated with the of which contributed to the study’s internal validity.
presence of amalgam restorations. The fact that in patients with
high DMFT the use of amalgam was more frequent, could
possibly indicate a common belief that tooth-coloured materi- 5. Conclusions
als should not be placed in individuals with higher caries risk.9
In fact, some studies showed that in patients with caries risk, Once both composite and amalgam demonstrate strong
the incidence of secondary caries is greater when composite clinical performance for posterior teeth, dentists should focus
restorations were placed, in comparison with amalgam.33,34 on oral health promotion, especially considering that the main
However dentists should take into consideration that no reasons for direct posterior restorations are more related to
material is able to prevent caries recurrence, and the prognosis the patients and the operator, rather than to the material used
for patients with uncontrolled disease is not satisfactory for for restoration.22 Such preventive approach would help to
both materials.35 Thus, more than the choice of restorative improve the longevity of restorative treatments. The present
option, professionals should prioritize educational approaches findings suggest that the choice of dentists for restorative
to change the behaviours that are putting patients at risk for materials are related to the type of payment of dental service,
these issues, so they avoid the restorative cycle36 that could dental caries (higher DMFT index), and cavity characteristics,
culminate in tooth loss and the need for dental prosthesis.37 such as the tooth type and the size of the cavity, in detriment
Molar teeth were strongly associated with the presence of of other individual characteristics such as demographic and
amalgam. In opposition, tooth-coloured restorations were socioeconomic status that showed no association with this
preferred in premolars where teeth are more apparent, outcome.
justifying the choice of an aesthetic alternative.9 Yet pre-
molars were found to be the most frequently fractured teeth,38
mainly due to their anatomy, which—added to the cavity Acknowledgements
design required for amalgam restorations—can result in
longitudinal fractures in these teeth. The use of adhesive The authors are grateful to CNPq for the research grant to the
materials in this case, besides reducing the removal of sound researchers (MAP, KGP, BLH, ADB, FFD) and to this study and to
tissue, can also reinforce the remaining dental structure.39 CAPES for the postgraduate fellowship (MBC).
In the same way, composites were chosen more frequently
with the increased number of surfaces affected by caries. In
the past, the indication of composite resins was restricted to references
small cavities. A recent study, however, revealed the better
performance of composite restorations in comparison to
amalgam for large cavities.40 Our findings suggest that 1. Hickel R, Heidemann D, Staehle HJ, Minnig P, Wilson NH.
Direct composite restorations: extended use in anterior and
dentists tend to rely on adhesive materials in cases where
posterior situations. Clinical Oral Investigations 2004;8:43–4.
large amounts of dental structure were removed in detriment
2. Cenci MS, Piva E, Potrich F, Formolo E, Demarco FF, Powers
of amalgam restorations, which depend on mechanical JM. Microleakage in bonded amalgam restorations using
retention to present a good prognosis. It is well known that different adhesive materials. Brazilian Dental Journal
the reduction of tooth resistance to fracture is proportional to 2004;15:13–8.
 journal of dentistry 40 (2012) 703–710 709

3. Demarco FF, Cenci MS, Lima FG, Donassollo TA, André D, de 22. Wirsching E, Loomans BAC, Klaiber B, Dörfer CE. Influence
A, et al. Class II composite restorations with metallic and of matrix systems on proximal contact tightness of 2- and
translucent matrices: 2-year follow-up findings. Journal of 3-surface posterior composite restorations in vivo. Journal of
Dentistry 2007;35:231–7. Dentistry 2011;39:386–90.
4. Brunthaler A, Konig F, Lucas T, Sperr W, Schedle A. Longevity 23. Beazoglou T, Eklund S, Heffley D, Meiers J, Brown LJ, Bailit H.
of direct resin composite restorations in posterior teeth. Economic impact of regulating the use of amalgam
Clinical Oral Investigations 2003;7:63–70. restorations. Public Health Reports (Washington DC 1974)
5. Manhart J, Chen H, Hamm G, Hickel R. Buonocore Memorial 2007;122:657–63.
Lecture. Review of the clinical survival of direct and indirect 24. Tobi H, Kreulen CM, Vondeling H, van Amerongen WE. Cost-
restorations in posterior teeth of the permanent dentition. effectiveness of composite resins and amalgam in the
Operative Dentistry 2004;29:481–508. replacement of amalgam Class II restorations. Community
6. Da Rosa Rodolpho PA, Donassollo TA, Cenci MS, Loguercio Dentistry and Oral Epidemiology 1999;27:137–43.
AD, Moraes RR, Bronkhorst EM, et al. 22-Year clinical 25. Tan SS, Ken Redekop W, Rutten FF. Costs and prices of
evaluation of the performance of two posterior composites single dental fillings in Europe: a micro-costing study. Health
with different filler characteristics. Dental Materials Economics 2008;17:S83–93.
2011;27:955–63. 26. Roeters FJ, Opdam NJ, Loomans BA. The amalgam-free
7. Niek JM, Opdam. Ewald M, Bronkhorst. Max S, Cenci. Marie- dental school. Journal of Dentistry 2004;32:371–7.
Charlotte DNJM, Huysmans. Nairn HF, Wilson. Age of failed 27. Demarco FF, Correa MB, Cenci MS, Moraes RR, Opdam NJM.
restorations: A deceptive longevity parameter. Journal of Longevity of posterior composite restorations: Not only a
Dentistry 2011;39:225–30. matter of materials. Dental Materials 2011;28:87–101.
8. Lubisich EB, Hilton TJ, Ferracane JL, Pashova HI, Burton B. 28. Bellinger DC, Trachtenberg F, Zhang A, Tavares M, Daniel D,
Association between caries location and restorative material McKinlay S. Dental amalgam and psychosocial status: the
treatment provided. Journal of Dentistry 2011;39: 302–8. New England Children’s Amalgam Trial. Journal of Dental
9. Vidnes-Kopperud S, Tveit AB, Gaarden T, Sandvik L, Espelid I. Research 2008;87:470–4.
Factors influencing dentists’ choice of amalgam and tooth- 29. Shenker BJ, Maserejian NN, Zhang A, McKinlay S. Immune
colored restorative materials for Class II preparations in function effects of dental amalgam in children: a
younger patients. Acta Odontologica Scandinavica 2009;67:74–9. randomized clinical trial. Journal of American Dental
10. Mjor IA, Moorhead JE, Dahl JE. Selection of restorative Association 2008;139:1496–505.
materials in permanent teeth in general dental practice. 30. Lynch CD, Frazier KB, McConnell RJ, Blum IR, Wilson NHF.
Acta Odontologica Scandinavica 1999;57:257–62. State-of-the-art techniques in Operative Dentistry:
11. Espelid I, Tveit AB, Mejàre I, Sundberg H, Hallonsten A-L. contemporary teaching of posterior composites in UK and
Restorative treatment decisions on occlusal caries. Acta Irish dental schools. British Dental Journal 2010;209:129–36.
Odontologica Scandinavica 2001;59:21–7. 31. Lynch CD, Frazier KB, McConnell RJ, Blum IR, Wilson NHF.
12. Tran LA, Messer LB. Clinicians’ choices of restorative Minimally invasive management of dental caries:
materials for children. Australian Dental Journal 2003;48: contemporary teaching of posterior resin composites in U.S.
221–32. and Canadian dental schools. Journal of the American Dental
13. Victora CG, Barros FC. Cohort profile: the 1982 Pelotas Association 2011;142:612–20.
(Brazil) birth cohort study. International Journal of Epidemiology 32. Lynch CD, Stewardson DA, Shortall ACC, Tomson PL, Burke
2006;35:237–42. FJT. Teaching posterior composite resin restorations in the
14. World Health Organization/FDI. Oral health surveys: basic United Kingdom and Ireland: Consensus views of teachers.
methods. Genebra; 1997. p. 47. British Dental Journal 2007;205:183–7.
15. Peres KGA, Peres MA, Demarco FF, Tarquinio SB, Horta BL, 33. Bernardo M, Luis H, Martin MD, Leroux BG, Rue T, Leitao J,
Gigante DP. Oral health studies in the 1982 Pelotas (Brazil) et al. Survival and reasons for failure of amalgam versus
birth cohort: methodology and principal results at 15 and 24 composite posterior restorations placed in a randomized
years of age. Cadernos de Saude Publica 2011;27:1569–80. clinical trial. Journal of American Dental Association
16. Barros AJD, Victora CG, Horta BL, Gonçalves HD, Lima RC, 2007;138:775–83.
Lynch J. Effects of socioeconomic change from birth to early 34. Opdam NJ, Bronkhorst EM, Roeters JM, Loomans BA. A
adulthood on height and overweight. International Journal of retrospective clinical study on longevity of posterior
Epidemiology 2006;35:1233–8. composite and amalgam restorations. Dental Materials
17. Peres MA, Peres KG, Thomson WM, Broadbent JM, Gigante 2007;23:2–8.
DP, Horta BL. The influence of family income trajectories 35. Opdam NJ, Bronkhorst EM, Roeters JM, Loomans BA.
from birth to adulthood on adult oral health: findings from Longevity and reasons for failure of sandwich and total-etch
the 1982 Pelotas Birth Cohort. American Journal of Public posterior composite resin restorations. Journal of Adhesive
Health 2011;101:730–6. Dentistry 2007;9:469–75.
18. Peres MA, Traebert JL, Marcenes W. Calibration of 36. Elderton RJ. Preventive (evidence-based) approach to quality
examiners for dental caries epidemiologic studies. Cadernos general dental care. Medical Principles and Practice International
de Saude Publica 2001;17:153–9. Journal of the Kuwait University Health Science Centre 2003
19. van Dijken JWV, Pallesen U. Four-year clinical evaluation of 2003;12:12–21.
Class II nano-hybrid resin composite restorations bonded 37. Correa MB, Peres MA, Peres KG, Horta BL, Gigante DP,
with a one-step self-etch and a two-step etch-and-rinse Demarco FF. Life-course determinants of need for dental
adhesive. Journal of Dentistry 2011;39:16–25. prostheses at age 24. Journal of Dental Research 2010;89:
20. van Dijken JWV. Durability of resin composite restorations 733–8.
in high C-factor cavities: A 12-year follow-up. Journal of 38. Nothdurft FP, Seidel E, Gebhart F, Naumann M, Motter PJ,
Dentistry 2010;38:469–74. Pospiech PR. The fracture behavior of premolar teeth with
21. Andrade AKM, Duarte RM, Silva FDSCM. Batista AUD, Lima class II cavities restored by both direct composite
KC, Pontual MLA. Montes MAJR. 30-Month randomised restorations and endodontic post systems. Journal of
clinical trial to evaluate the clinical performance of a Dentistry 2008;36:444–9.
nanofill and a nanohybrid composite. Journal of Dentistry 39. Coelho-De-Souza FH, Camacho GB, Demarco FF, Powers JM.
2011;39:8–15. Fracture resistance and gap formation of MOD restorations:
710 journal of dentistry 40 (2012) 703–710

influence of restorative technique, bevel preparation and 42. da Rosa Rodolpho PA, Cenci MS, Donassollo TA, Loguercio
water storage. Operative Dentistry 2008;33:37–43. AD, Demarco FF. A clinical evaluation of posterior
40. Opdam NJ, Bronkhorst EM, Loomans BA, Huysmans MC. 12- composite restorations: 17-year findings. Journal of Dentistry
Year survival of composite vs. amalgam restorations. Journal 2006;34:427–35.
of Dental Research 2010;89:1063–7. 43. Rabe-Hesketh S, Skrondal A. Multilevel and longitudinal
41. Coelho-de-Souza FH, Rocha Ada C, Rubini A, Klein-Júnior modeling using stata. College Station: Stata Press; 2008.
CA. Demarco FF. Influence of adhesive system and bevel 44. Diez-Roux AV. Multilevel analysis in public health research.
preparation on fracture strength of teeth restored with Annual Reviews in Public Health 2000;21:171–92.
composite resin. Brazilian Dental Journal 2010;21:
327–31.