See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/319959539

Clinical studies in restorative dentistry: New directions and new demands

Article  in  Dental Materials · September 2017

DOI: 10.1016/j.dental.2017.08.187

CITATIONS READS

9 841

11 authors, including:

Niek J M Opdam Kauê Collares

Radboud University Universidade de Passo Fundo
157 PUBLICATIONS   5,096 CITATIONS    21 PUBLICATIONS   393 CITATIONS   

SEE PROFILE SEE PROFILE

Reinhard Hickel Stephen C Bayne

Ludwig-Maximilians-University of Munich University of Michigan
521 PUBLICATIONS   14,618 CITATIONS    139 PUBLICATIONS   5,967 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Secondary caries lesion & Local factors: evidences from complex in vitro and in situ studies View project

trial in the field, clinical study design View project

All content following this page was uploaded by Kauê Collares on 22 September 2017.

The user has requested enhancement of the downloaded file.

DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

Available online at www.sciencedirect.com

ScienceDirect

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

Review

Clinical studies in restorative dentistry: New

directions and new demands

N.J.M. Opdam a,∗ , K. Collares b , R. Hickel c , S.C. Bayne d , B.A. Loomans a ,

M.S. Cenci b , C.D. Lynch e , M.B. Correa b , F. Demarco b , F. Schwendicke f ,
N.H.F. Wilson g
a Radboud University Medical Centre, College of Dental Sciences, P.O. Box 9101, NL 6500 HB Nijmegen, The
Netherlands
b Graduate Program in Dentistry, Federal University of Pelotas, Gonçalves Chaves, 457, 96015560 Pelotas, Brazil
c Klinikum, Ludwig-Maximilians Universität München, Goethestr. 70, 80336 München, Germany
d Department of Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, 1011

North University, Ann Arbor, MI 48109, USA

e Cork University Dental School & Hospital, University College Cork, College Road, Cork T12 YN60, Ireland
f Abteilung für Zahnerhaltung und Präventivzahnmedizin, Charité — Universitätsmedizin Berlin, Aßmannshauser

Str. 4-6, 14197 Berlin, Germany

g King’s College London Dental Institute, King’s College London, SE19RT London, UK

a r t i c l e i n f o a b s t r a c t

Article history: Clinical research of restorative materials is confounded by problems of study designs, length
Received 24 May 2017 of trials, type of information collected, and costs for trials, despite increasing numbers and
Accepted 24 August 2017 considerable development of trials during the past 50 years. This opinion paper aims to
Available online xxx discuss advantages and disadvantages of different study designs and outcomes for evaluat-
ing survival of dental restorations and to make recommendations for future study designs.
Keywords: Advantages and disadvantages of randomized trials, prospective and retrospective longitu-
Clinical trials dinal studies, practice-based, pragmatic and cohort studies are addressed and discussed.
Restoration The recommendations of the paper are that clinical trials should have rational control
Dental materials groups, include confounders such as patient risk factors in the data and analysis and should
Longevity use outcome parameters relevant for profession and patients.
Survival © 2017 The Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Retrospective
Prospective
Restorative dentistry

Contents

1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .00
2. Clinical studies of different designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

∗
Corresponding author.
E-mail address: Niek.opdam@radboudumc.nl (N.J.M. Opdam).
http://dx.doi.org/10.1016/j.dental.2017.08.187
0109-5641/© 2017 The Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
2 d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

3. Relevant research questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

4. Outcomes and outcome measures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .00
5. What is the ideal study design? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
5.1. RCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
6. Retrospective longitudinal studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
7. Prospective longitudinal studies and pragmatically designed trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
8. Population-based (birth) cohort studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
9. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
10. Recommendations for future clinical studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
Appendix A. Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

required number of restorations, and for the recall of patients

1. Introduction at specific time points for the detailed assessment of the
qualities of individual restorations, typically including clin-
Clinical research of restorative materials is confounded by
ical photographs and the recording of impressions. Costs
problems of study designs, length of trials, type of informa-
may limit a study’s follow-up period, frequency of recall
tion collected, and costs for trials, despite increasing numbers
visits, let alone the number of patients included in the
and considerable development of trials during the past 50
trial — factors critical to the viability of the investiga-
years. In the 1970s the first clinical studies on the perfor-
tions. Moreover, the relevance of such outcome measures
mance of direct composites were published at a time when
to patients, providers and payers needs to be questioned
amalgam remained the gold standard material for direct
[10].
restorations in posterior teeth [1–3]. At that time performance
To avoid selection bias and assessments of the efficacy of
investigations were more about the potential of test mate-
restorative materials and different treatment modalities, ran-
rials rather than effectiveness in everyday clinical practice,
domized controlled clinical trials (RCT) have been favoured
resulting in highly controlled studies, typically undertaken
in the past, being widely considered to be the best design to
in academic institutions. Such studies revealed a great deal
answer specific questions in clinical research. However, RCTs
about the clinical performance of the materials under inves-
in dentistry are extremely resource intensive and costly and
tigation, but had dubious applicability in the primary care
oftentimes performed under artificial conditions with limited
settings. Today, there is an upcoming of so-called practice
external validity. Furthermore, many RCTs in dentistry suf-
based studies that attempt to collect data from general prac-
fer from limited sample sizes and relatively high participant
tices, but at the same time include the risk to be flawed
attrition rates over, in particular, extended observation times,
by untrained practitioners in placement, management, and
while materials under investigation are already replaced on
assessment.
the market.
To reduce selection bias and allow restorations of dif-
The often proclaimed demand for more randomized clin-
ferent materials to be evaluated according to standardised
ical trials with longer observation times could be considered
and controlled protocols, well defined criteria for evaluat-
to be unhelpful, indeed unrealistic. Consequently, additional
ing restorations were developed and introduced. For direct
alternative designs and outcomes should be considered
restorations, USPHS or Ryge criteria were published, with
and accepted by funders and reviewers of research on the
modified versions still being used today to assess various
clinical performance of dental restorations. For example,
features of restorations [4,5]. To modernise this method
retrospective clinical studies have the capacity to provide
for the evaluation of dental restorations, new criteria were
data over extended observation times on large numbers of
published by the FDI in 2007 [6–8]. These criteria were
restorations placed by general practitioners [10–12]. These
updated in 2010 [6–9]. Basically, this method for the assess-
trials seek different outcomes (oftentimes less standard-
ment of the clinical performance of restorations requires
ised and granular than RCTs), which merit more detailed
that:
consideration for their validity, sensitivity and reliability in
guiding practitioners in primary care settings [11]. More-
- Patients should be recalled for restoration evaluation visit over, other designs like cohort studies should be considered.
(best within specific timeframes) Cross sectional studies, that are often used to reflect the
- All restorations should be clinically evaluated by calibrated situation in general practice are unreliable for providing
evaluators. longevity data and should not be used for that purpose
- The evaluators should use prescribed list of criteria to assess [13].
qualities of the restorations, based on specific grades vary- This opinion paper aims to discuss advantages and dis-
ing from excellent to poor in need of replacement [4,5]. The advantages of different study designs and outcomes for
number of used grades depends on the aim of the evalua- evaluating survival of dental restorations and to make rec-
tion. ommendations for future study designs. It should serve as
a basis for further discussions on the most efficient and
However, the detailed type of evaluation requires huge, effective use of resources available for clinical research in
sustained effort and incurs high costs for the recruitment restorative dentistry to better serve both the advancement of
of patients, the placement and baseline evaluation of the

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
 DENTAL-3007; No. of Pages 12
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater

Table 1 – Descriptive analysis on studies evaluating longevity of posterior composite restoration (n = 87) according study characteristic and statistical analysis
performed. (Studies published between 2005 and 2015).
Variables Retrospective studies References Prospective studies References
n (%) n (%)
Study characteristic
Funding of the study
By dental 0 (0.0) – 28 (37.3) [31,32,51–76]
manufac-
tures
By private 3 (25.0) [33,42,77] 23 (30.7) [17,19,27,28,78–96]
or public
institutions
Not 9 (75.0) [43,44,97–103] 24 (32.0) [30,104–126]

ARTICLE IN PRESS
reported

d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx
Observation time
Up to 5 2 (16.7) [98,100] 57 (76.0) [30,32,33,42–44,51–62,64–73,75–80,82,84–87,92,94,95,97–108,110–114,117–126]
years
5 years or 10 (83.3) [33,42–44,77,97,109,101–103] 18 (24.0) [17,19,27,28,31,63,74,81,83,88–91,93,96,109,115,116]
more
Restorations included
Up to 100 2 (16.7) [97,100] 49 (65.3) [30–33,42–44,52–54,56–58,61,63,64,66–73,75–80,82–84,86–88,90,96–115,117,122–126]
More than 10 (83.3) [33,42–44,77,98,109,101–103] 26 (34.7) [17,19,27,28,51,55,59,60,62,65,74,81,85,89,91–95,116,118–121]
100
Participants included
Up to 100 5 (41.7) [42–44,97,109] 67 (89.3) [30–33,42–44,51–54,56–64,66–115,117,119–126]
More than 7 (58.3) [33,77,98,100–103] 8 (10.7) [17,19,27,28,55,65,116,118]
100
Risk Patients excluded
Yes 4 (33.3) [33,42,101,102] 45 (60.0) [19,30–32,51–53,56–64,66–69,72,75,76,82–85,87,104,105,107,108,110–113,117–126]
No 3 (25.0) [43,98,109] 20 (26.7) [27,54,55,62,65,70,71,74,81,86,88–96,116]
No 5 (41.7) [44,77,97,100,103] 10 (13.3) [17,28,73,78–80,106,109,114,115]
reported

Statistical analysis performed

Survival rate
No 0 (0.0) – 36 (48.0) [17,27,30,51–53,59,61,65,67–69,72,73,75,76,79,80,82,85,87,88,104,107–110,112,114,115,117,120,123,124,126]
Yes 12 (100.0) [33,42–44,77,97–103] 39 (52.0) [28,31,32,54–58,60,62–64,66,70,71,74,78,81,83,84,86,89–96,105,106,111,113,116,118,120,122,125]
Multivariable regression model
No 3 (25.0) [33,109,100] 71 (94.7) [17,19,30–33,42–44,51–54,56–105,107–126]
Yes 9 (75.0) [42–44,77,97,98,101–103] 4 (5.3) [27,28,55,106]
Reasons for failure reported
No 1 (8.3) [103] 3 (4.0) [59,114,123]
Yes 11 (91.7) [33,42–44,77,97–102] 57 (76.0) [17,19,27,28,32,33,42–44,51–59,62–65,68–72,74,77,79–86,88–109,111–113,115,116,118–122,124]
No failures 0 (0.0) – 15 (20.0) [30,31,60,61,66,67,75,76,78,87,110,117,125,126]

Total 12 (100.0) 75 (100.0)

3
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
4 d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

the art and science of dentistry and the consumers of oral resin techniques [16,17], repair versus replacement [18], inter-
healthcare—our patients [12]. vention versus no intervention. The number of prospective
clinical studies addressing such issues is limited, possibly
given the lack of funding and sponsoring opportunities —
2. Clinical studies of different designs manufacturers have limited interest in these types of stud-
ies, and only RCT’s comparing amalgam and composite have
In an attempt to get an overview of different types of stud- been supported by the American NIH in recent times [17,19].
ies, research questions, clinical trial methodologies, variables Another problem is the number of variables involved, espe-
and outcomes, a literature search was performed. An overview cially when fundamentally different techniques and materials
of longevity studies on direct posterior composite restora- are compared, e.g. amalgam vs. composite. However, these
tions published over the last 10 years (from January 1st, types of comparisons are important for dentists when it comes
2005 to September 30th, 2015) is set out in Table 1. The to effecting a change in behaviour, such as a shift to the
selection of studies was based on a search of the National use of composite rather than amalgam in the management
Library of Medicine international electronic database — MED- of primary lesions of caries requiring operative intervention.
LINE/PubMed (Table 3). The search was restricted to papers Behavioural change with many new technologies is highly rel-
written in English. Table 1 divides the 87 selected studies evant to the further development of the profession.
into “prospective” and “retrospective” and includes details on
study characteristics and the statistical analyses performed.
2 Materials testing
Detailed information on the assessment of each individual
study is presented in Appendix A (Supplementary material).
From Table 1, it can be seen that more than 60% of the Research questions testing a new material or technique
prospective clinical studies had less than 100 restorations as an alternative to an existing (gold standard) material or
included, observation times of <5 years, and 60% of the stud- technique that might offer major improvements in longevity,
ies did not provide information on important patients’ risk applicability, safety or efficiency [20,21]. Studies comparing a
factors such as caries experience and parafunctional habits. new material with an existing material (e.g. a new, suppos-
Moreover, only a limited number of the prospective stud- edly improved formulation vs. the current formulation) might
ies assessed the influence of risk factors on the longevity not have high clinical relevance if the existing standard mate-
of restorations by means of multivariable statistical anal- rial already performs sufficiently [22,23]. Moreover, one should
yses. More importantly, given relatively short observation note that any performance against the gold standard materials
times, 20% of the studies included no data on restoration fail- may well require high sample sizes to yield statistical sig-
ures. nificant differences, given the excellent performance of ‘gold
From the 10 year look-back review, prospective controlled standard’ materials. Considerations of the relevance of dif-
studies, despite having better quality of design (e.g. risk of ferences which are hard to demonstrate should be a focus in
bias definition — something we admittedly did not assess future trial reports [24].
here), it may be concluded that the considerable effort (and
cost) of running the typical prospective trial results in impor- 3 Assessing patient risk factors
tant limitations linked to observation time, low number of
retained patients and statistical underpowering. Furthermore,
Studies investigating the influence of risk factors on
there is evidence that an increasing number of RCTs are being
longevity that are related to the human factors in dentistry:
performed to investigate materials that “do not outperform
patients and dentists. While most clinical studies focus on
in vitro, lack fundamental properties, or do not fulfil the pre-
materials and their properties, it is increasingly recognised
requisites required for clinical trials” [14].
that other factors, including caries risk, bruxism, socio-
economic status and operator variables [25–28] play a major,
3. Relevant research questions and even dominant role in restoration longevity.

When a clinical study is being planned, one of the main 4 High risk testing
questions should be: will the design and expected outcome
contribute to better care in dentistry? Clinical relevance of
Studies aiming to test the performance of restorative work
studies will be improved if problems and questions general
in challenging clinical situations [29] may be considered to
practitioners face every day will be addressed by the research.
be ‘high risk testing’. Many clinical studies exclude high risk
In our opinion, the following questions could have such rele-
patients, notably high caries risk patients and patients with
vance:
bruxism from their study population [30,31]. Moreover, the size
of restorations in clinical trials is oftentimes limited, and cases
1 Technique comparisons involving deep subgingival margins or cusp replacement are
excluded. This is understandable as RCTs aim to remove con-
Studies comparing two fundamentally different restora- founding variables, but might also be linked to sponsors and
tive techniques for a specific situation, such as indirect funders of clinical research wishing to see positive outcomes
versus direct restoration [15], mechanistic, interventive amal- from expensive studies. Dentists, however, face such chal-
gam techniques versus minimum intervention composite lenges on a daily basis, and need meaningful research data

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx 5

Table 2 – Summary of characteristics of different trial types.

Characteristics RCTs Pro- or retrospective
non-randomized trials
Risk of selection bias Low High
Risk of performance bias Probably high Probably high
Risk of detection bias Possibly high Possibly high
Risk of selective reporting Low (if registered) Higher, registration possible prior
conduct (in prospective studies) or
analysis
Follow-up Usually short Longer
Risk of attrition High Low
Sample size and power Usually low Possibly high
External validity Usually low (could be high in Usually high
practice-based RCTs)
Outcome measures Should be relevant to all Should be relevant to all
stakeholders stakeholders
Costs and administrative efforts High Low (especially when using
routine data)

to guide them in their adoption, or rejection of new materials Alternative to Ryge or FDI criteria, the functional presence
and techniques. of a restoration may be viewed as critical, indicating that a
restoration is still functioning in the mouth (success) or has
been replaced or repaired (failure). This method is used in
5 Long-term longevity studies
practice based studies, prospective as well as retrospective
[27–29,33] but lacks detailed quality assessment of a restora-
Studies planned with less than 2–3 years observation time tion. It is highly dependent on the treating dentist and the
have limited clinical relevance as most, modern day materials criteria he or she uses to decide whether a restoration needs
will be able to perform satisfactorily for that length of time to be replaced or not. Therefore, the reasons for failure or
in clinical service [21,32]. Moreover, one of the most impor- replacement, and the criteria used for such decision need to
tant reasons for restoration failure, secondary caries, tends to be recorded. When it comes to defining what is a success and
occur after more than 2–3 years in clinical service [28]. Future a failure, Anusavice [34] suggest that a restoration is “success-
studies should compare materials or treatments over a rele- ful” if no intervention is indicated, and a “failure”, when the
vant timeframe. entire restoration must be replaced or the tooth extracted.
A third category is restorations that are repaired or received
endodontic treatment with the restoration remaining in place;
4. Outcomes and outcome measures these are classified as “survived” [32]. For direct restorations,
previously all interventions were considered indicative of fail-
As described, most clinical trials on restorative materials use ure, including repairs where the original restoration is still (at
standardised criteria, applied by calibrated observers, which least in large part) in place. For standardisation reasons, and
yield minute differences between materials or treatments on a to be able to compare longevity of direct and indirect restora-
range of aspects (colour, margin, surface behaviour etc.) [20,26]. tions, it would be good to use the same definitions of success,
Moreover, they often deem specific situations as requiring survival and failure for all types of restorations:
replacement (the Ryge criteria, for example, include Charlie
and Delta ratings, indicating that restorations need immedi-
ate replacement or replacement in the near future). In clinical
dental practice however, the replacement of a restoration 1. Success: at evaluation, the restoration is still functioning
depends not only on the status of the restored tooth but also and no intervention (repair or replacement) is indicated. In
on the oral and general health of the individual patient, risk this regard, refurbishment, recontouring and polishing is
assessment outcomes, and, last but not least, informed con- not considered to be an intervention.
sent of the patient. Given that restorations nowadays are often 2. Survival: a restoration requires repair. This category would
monitored, refurbished or repaired, even those rated Charlie or also include teeth that require endodontic intervention, but
Delta might continue to function (i.e. remain in clinical service) with the restoration remaining in place, with the access
for many years. As well defined standardised criteria are the opening restored following the endodontic therapy.
key to compare the clinical status of restorations over time, 3. Failure: cases where a restoration must be replaced or the
the FDI criteria might be an alternative, and could even be tooth removed for reason related to the restorations, such
simplified by merging grade 1–3 to one grade — clinically sat- as tooth fracture, but unrelated to periodontal health or
isfactory. Similarly, not all FDI criteria need to be used in all trauma.
studies; in fact, single criteria can be selected for the specific 4. If a restoration is removed for reasons unrelated to material
goal of the study and the remaining criteria could be omitted performance; for example, an abutment tooth for a bridge
(e.g. aesthetics). or a tooth removed for periodontal reasons, the event may

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
6 d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

be noted but not recorded as a failure, resulting in censoring However, in clinical research in restorative dentistry, e.g. in
in the analysis. trials comparing crowns and fillings, blinding is impossible
and the likelihood of these types of bias is inevitable, albeit
Besides longevity, there is increasing acceptance that other that it may be reduced by using more operators and evalua-
outcomes, including, subjective judgment and material con- tors. The fact that patients are aware of the applied therapy
siderations by patients or practitioners are also relevant. As includes the risk for sampling bias, as many patients will not
occurs in other disciplines, core outcome sets should be estab- be enrolled because they do not want to have, for example,
lished for restorative trials; these are defined in consensus amalgam restorations, and, as a result, designing a clinical
between patients, practitioners and other stakeholders [22,35]. trial to address this still important research question [38] will
The resulting sets are minimum standards of outcomes to be include the risk of a non-representative sample of the popu-
assessed; they help to report on the most relevant aspects of lation. It is acknowledged that many of these disadvantages
trials, but also prevent selective reporting [10,36]. apply also to non-randomized clinical studies.
More recently, patient-based outcomes measures (PBOs)
have been emerged as an important aspect to complement
2 Time of follow-up
conventional clinical measures that have been the main focus
of oral health research [37]. The assessment of the opinion
of patients about their own health is a key point to evaluate Differences in effectiveness of therapies may only be mea-
the effectiveness of an intervention. In this way, the assess- sured after several years, as failure behaviour may vary -and
ment of the impact that restorations can have on patients’ oral one type of material may be more susceptible to caries and
health-related quality of life (OHRQoL) should be stimulated. the other to tooth fracture on the long term [33]. As a result,
The present digital revolution provides opportunities for long observation times are required, sometimes exceeding 10
designing new assessment methods that combine the need years to record all relevant effects and differences. To main-
for standardisation using a well-defined list of criteria and tain a population of trial participants over an extended period
data acquisition from electronic patient files and digital imag- is very challenging. Attrition bias is a common phenomenon in
ing. Intra-oral electronic photography and inspection of digital clinical research in dentistry, making it very difficult to achieve
bitewing radiographs may provide standardised information long observation times in clinical trials (Table 1). Recently,
on the quality of restorative work and provide insight on results from two clinical trials with a long observation time
the application of replacement criteria by practitioners. New were published by a Danish research group, but this is excep-
intra-oral scanning methods, including colour reproduction tional [39,40].
capacities have been developed, which might become impor-
tant tools in future clinical studies. However, patients still
have to attend to obtain images, so data acquisition remains 3 Inclusion bias
dependent on patients being available for recall and sufficient
resources to fund recall procedures. Also, images have to be
If inclusion criteria are not well defined, it may lead to
stored securely and evaluated using sophisticated method-
inclusion bias, as the restorations are placed for a variety of
ologies which are both time consuming and costly. Anyway,
different reasons including active caries, fracture due to brux-
dentistry should start to put effort on the establishment of
ism and aesthetic considerations. Moreover, it is likely that
guidelines advising on how dentists should record clinical data
in some clinical trials highly motivated patients with a good
and images to allow proper follow up of restorative treatments.
oral hygiene tend to be included, a feature which is seldom
mentioned in trial reports. A typical example of this is two
5. What is the ideal study design? randomized clinical trials comparing the longevity of amal-
gam and composite restorations in young children [17,19]. The
5.1. RCT inclusion criteria were that children should require one or
more restorations, which is likely oftentimes a result of pri-
The randomized clinical trial is most often considered to mary caries, indicating high caries experience. Obviously, the
be the ideal design for comparing different treatments in inclusion bias is that a high risk population is investigated, but
medicine and as a consequence, procedures and the use of dif- this is not mentioned in any of the relevant reports [11,17,41].
ferent materials in dentistry. However, as discussed, the RCT
approach has several disadvantages when testing restorative 4 Independent evaluation
materials:

1 Blinding To avoid reporting bias, dentists that place restorations in

a clinical study should not evaluate their own restorations at
In RCTs, blinding of patients and researchers for the applied recall. Therefore, independent evaluation is mandatory, mak-
therapy is considered to be most important in the prevention ing the process of clinical testing very expensive and, out of
of performance bias (operators perform different therapies necessity often limited to university settings.
differently if they know the group allocation) and detection In Table 2, characteristics of RCTs in restorative dentistry
bias (patients and evaluators being aware of the allocation are summarised. Given the above mentioned issues, alterna-
and being more or less prone to detect or report differently). tive trial designs should be considered.

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx 7

Table 3 – Structured search strategy carried out in MEDLINE/PubMed database.

Search Topic and terms
#4 Search #1 AND #2 AND #3

#3 Composite resin/dental restoration:

((“composite resins” [MeSH Terms] OR “resins, composite” [Title/Abstract] OR (“composite” AND “resins”)
[Title/Abstract] OR “composite resins” [Title/Abstract] OR (“composite” AND “resin”) [Title/Abstract] OR
“composite resin” [Title/Abstract] OR “Dental Restoration, Permanent” [Mesh] OR “Dental Restoration,
Permanent” [Title/Abstract] OR “Permanent Dental Restoration” [Title/Abstract] OR “Restoration, Permanent
Dental” [Title/Abstract] OR “Restorations, Permanent Dental” [Title/Abstract] OR “Dental Restorations,
Permanent” [Title/Abstract] OR “Permanent Dental Restorations” [Title/Abstract] OR “Dental Permanent Fillings”
[Title/Abstract] OR “Fillings, Permanent Dental” [Title/Abstract] OR “Permanent Dental Fillings” [Title/Abstract]
OR “Permanent Fillings, Dental” [Title/Abstract] OR “Permanent Filling, Dental” [Title/Abstract] OR “Dental
Filling, Permanent” [Title/Abstract] OR “Dental Permanent Filling” [Title/Abstract] OR “Filling, Dental Permanent”
[Title/Abstract] OR “Filling, Permanent Dental” [Title/Abstract] OR “Permanent Dental Filling” [Title/Abstract] OR
“Fillings, Dental Permanent” [Title/Abstract] OR “Dental Fillings, Permanent” [Title/Abstract]

#2 Survival analysis:
(“Survival” [All Fields] OR “Success” [All Fields] OR “Longevity” [All Fields] OR “Annual failure rate” [All Fields] OR
“Clinical evaluation” [All Fields] OR “Survival Analysis”[Mesh] OR “Survival Analysis” [All Fields] OR “Analysis,
Survival” [All Fields] OR “Analyses, Survival” [All Fields] OR ”Survival Analyses” [All Fields] OR “Failure” [All
Fields] OR “Dental Restoration Failure”[Mesh] OR “Dental Restoration Failure”[All Fields] OR “Failure, Dental
Restoration” [All Fields] OR “Restoration Failures, Dental” [All Fields] OR“ Failures, Dental Restoration” [All Fields]
OR“ Restoration Failure, Dental” [All Fields] OR“ Dental Restoration Failures” [All Fields])

#1 Clinical trial/Longitudinal study/Retrospective study/Cohort study:

((“clinical” [Title/Abstract] AND “trial” [Title/Abstract]) OR “clinical trials” [MeSH Terms] OR “clinical trial”
[Publication Type] OR random*[Title/Abstract] OR “random allocation” [MeSH Terms] OR “therapeutic use” [MeSH
Subheading]) OR “Longitudinal Studies”[MeSH Terms] OR “Longitudinal Studies” [All Fields] OR “Longitudinal
Study” [All Fields] OR “Studies, Longitudinal” [All Fields] OR “Study, Longitudinal” [All Fields] OR “Longitudinal
Survey” [All Fields] OR “Longitudinal Surveys”[All Fields] OR “Survey, Longitudinal” [All Fields] OR “Surveys,
Longitudinal” [All Fields] OR “Prospective Studies” [Mesh] OR “Prospective Studies” [All Fields] OR “Retrospective
Studies” [MeSH Terms] OR “Studies, Retrospective” [All Fields] OR “Study, Retrospective” [All Fields] OR
“Retrospective Study” [All Fields] OR “Clinical Evaluation” [All Fields] OR “Follow-up” [All Fields] OR “Cohort
Studies” [Mesh] OR “Cohort Studies” [All Fields] OR “Cohort Study” [All Fields] OR “Studies, Cohort” [All Fields]
OR “Study, Cohort” [All Fields])

to specialists, patient files have limited value in retrospective

6. Retrospective longitudinal studies studies.
Moreover, in retrospective studies, quality of patient files
For robust longevity analyses of restoration longevity, it is is dependent on the practice investigated. If the dentist who
important that for all restorations placed in a certain time placed the restoration also decided when a restoration should
period, not only date and reason of failure, but also date of be replaced, there is significant risk of detection — and proba-
placement, and even more importantly, the assessment date bly also reporting bias. In certain studies, this risk and the risk
for restorations remaining in clinical service are recorded. that certain restorations may be replaced in another practice
Comparing only the number of failed and remaining restora- is overcome by independent evaluators visiting the practice
tions in the comparator groups leads to erroneous conclusions [42–44], but this limits the number of patient records that can
[13]. be analysed and greatly increases research logistics and costs.
When it comes to comparing restorative techniques and Typical characteristics of non-randomized and retrospective
dental materials, retrospective studies run considerable risk studies are summarised in Table 2.
of selection bias, as dentists invariably choose the restora- Given the opportunities afforded by intraoral, digital cam-
tive technique that is most appropriate for a given indication. eras and scanning devices, it is possible in the future that
Using data on cohorts of restorations placed at a different time photographs and scans of restored teeth can be obtained and
for the same indication (i.e. composites in situations where assessed for qualitative analysis by independent investiga-
amalgams had been placed in the past) might help to over- tors, reducing the risk for reporting bias. This would also offer
come this problem [33]. Accounting for recorded confounders the opportunity to evaluate the influence of operators on the
as indicators of selection bias and considering them during treatment result.
statistical evaluation is another way of reducing the impact of Furthermore, contractual, commissioning and reimburse-
selection bias. The reliability of patient records used for ret- ment systems may highly influence the decisions of dentists
rospective studies is highly dependent upon the number of and patients, significantly influencing the selection of mate-
individuals who remain patients and receive all their dental rials and techniques as well as replacement decisions and, in
treatment in the practice(s) over extended periods of time. In turn, longevity.
situations where patients often change dentists, or where it
is common practice to refer patients for certain treatments

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
8 d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

Oftentimes, in clinical studies in dentistry, more than one

7. Prospective longitudinal studies and tooth is treated in participating patients. Multilevel statistical
pragmatically designed trials models are required to properly account for this clustering of
statistical units [25,29]. Similarly, statistical evaluation should
Recently, a number of large prospective clinical studies, based make best use of specific designs such as split-mouth studies,
on survival data obtained by the treating dentists from which are common in restorative trials [47,48].
patients records in Scandinavian public health clinics and
practices participating in the practice based research network
in the USA have been published [27,28,45]. As with retrospec- 10. Recommendations for future clinical
tive studies based on patient clinical records, reporting bias studies
is a limitation of these prospective study designs. However,
when it is possible to access large datasets, pragmatic study To further development restorative dentistry, several specific
designs make it possible to compare the performance of differ- research questions are of central importance. Clinical trial
ent materials and restorative techniques in the primary care designs should address these questions:
setting, assuming data, including information on all relevant
potential risk factors could be obtained from electronic files. 1 Choose a rational control

Studies comparing new materials to their predecessors are

8. Population-based (birth) cohort studies necessary prior to the launch of new products. Such stud-
ies are, however, of limited priority in the quest to further
In cohort studies, a group of individuals is observed for a develop restorative dentistry. Clinical studies comparing dif-
period of time allowing the investigation of the effects of living ferent treatment options are of much greater importance.
life on health outcomes. The use of population-based sam- Direct versus indirect restorations, tooth replacement versus
ples allows extrapolation of data to the target population. tooth retention, restoration repair versus replacement, bridge
Recently, data from the Pelotas 1982 Birth Cohort were used versus implant, are all examples of research comparisons
for analysing factors affecting the quality of posterior restora- which need to be addressed. If trials compare materials, such
tions [25] and the placement of amalgam or composite fillings comparisons should be made against a gold standard, not
in posterior teeth [46]. These studies showed that individual “straw men” comparators which yield favourable results for
characteristics, such as socioeconomic status, type of service the material to be tested [49]. For such material testing trials,
provided and individual caries risk, influence the longevity clear reporting of all sponsorship should be mandatory [50].
of posterior restorations. The inclusion of restoration assess-
ment in such large multidisciplinary studies creates access to 2 Choose relevant outcome parameters
data on a wide range of variables pertaining aspects of den-
tal care and general health and wellbeing. This is especially Regardless of what is investigated, the chosen outcomes of
interesting when investigating patients’ risk factors. With this trials should reflect what is most relevant to patients, dentists
approach, information on events occurring in life is collected. and other stakeholders. The use of an agreed outcome set for
However, the observational nature of cohort studies does not future trials would be desirable also to reduce risk of selective
allow researchers to obtain precise data on, for example, the reporting.
date of restoration placement.
3 Include data on patient risk and interventions

9. Statistical analysis Major confounders such as caries risk and bruxism should
be recorded as part of trial methodologies and appropriately
In the ideal RCT, all variables are standardised, with only considered during analyses. The time and reasons for any
the experimental variable differing between the test and interventions should also be recorded to allow discrimination
control groups. In such cases, the Kaplan Meier statistical between success, survival and failure, and to perform statisti-
model helps generate survival graphs and enables compar- cally valid survival analyses.
isons between experimental groups, for example a log-rank Different trial designs should be considered when planning
test which shows if survival curves are statistically different. to compare restorative treatments and materials: while ran-
However, in RCTs not all confounding variables are perfectly domised clinical trials reduce selection bias, their observation
balanced (accounting for all possible confounders during ran- times are usually short and sample sizes limited. Retrospec-
domisation is theoretically possible, but practically hard to tive and prospective practice based studies circumvent these
achieve). Parameters such as caries risk [33,42], bruxism [42], limitations, but have considerable risk of indication bias and
socioeconomic status [25], age [33,42], number of teeth in confounding which need to be dealt with appropriately. Per-
the dentition [19], as well as tooth type and details of the haps the answer lies in coordinated, multicentre studies of
practitioner(s) who treated the patient [19,45] should thus be different design, no one design being capable of yielding all the
recorded during the trial and included in the evaluation using required outcomes. Whatever the way forward funders and
multi-variate analyses. In planning prospective trials, the esti- sponsors of clinical research in restorative dentistry should
mation of the required sample size should take into account encourage investigators to come forward with new innovative
the need to undertake such analyses. approaches to answers priority questions.

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx 9

Children’s Amalgam Trial. J Am Dent Assoc

Appendix A. Supplementary data 2007;138:763–72.
[18] Gordan VV, Riley 3rd JL, Rindal DB, Qvist V, Fellows JL,
Supplementary data associated with this arti- Dilbone DA, et al. Repair or replacement of restorations: a
prospective cohort study by dentists in The National
cle can be found, in the online version, at
Dental Practice-Based Research Network. J Am Dent Assoc
http://dx.doi.org/10.1016/j.dental.2017.08.187.
2015;146:895–903.
[19] Bernardo M, Luis H, Martin MD, Leroux BG, Rue T, Leitao J,
references et al. Survival and reasons for failure of amalgam versus
composite posterior restorations placed in a randomized
clinical trial. J Am Dent Assoc 2007;138:775–83.
[20] Heintze SD, Rousson V. Clinical effectiveness of direct class
[1] Chandler HH, Bowen RL, Paffenbarger GC, Mullineaux AL. II restorations — a meta-analysis. J Adhes Dent
Clinical evaluation of a radiopaque composite restorative 2012;14:407–31.
material after three and a half years. J Dent Res [21] van Dijken JW. Three-year performance of a calcium-,
1973;52:1128–37. fluoride-, and hydroxyl-ions-releasing resin composite.
[2] Osborne JW, Gale EN, Ferguson GW. One-yer and two-year Acta Odontol Scand 2002;60:155–9.
clinical evaluation of a composite resin vs. amalgam. J [22] Smail-Faugeron V, Fron Chabouis H, Durieux P, Attal JP,
Prosthet Dent 1973;30:795–800. Muller-Bolla M, Courson F. Development of a core set of
[3] Liatukas EL. A clinical investigation of composite resin outcomes for randomized controlled trials with multiple
restorations in anterior teeth. J Prosth Dent 1972;27:616–21. outcomes—example of pulp treatments of primary teeth
[4] Ryge G. Clinical criteria. Int Dent J 1980;30:347–58. for extensive decay in children. PLoS One 2013;8:e51908.
[5] Ryge G, Snyder M. Evaluating the clinical quality of [23] Williamson PR, Altman DG, Blazeby JM, Clarke M, Devane
restorations. J Am Dent Assoc 1973;87:369–77. D, Gargon E, et al. Developing core outcome sets for clinical
[6] Hickel R, Peschke A, Tyas M, Mjor I, Bayne S, Peters M, et al. trials: issues to consider. Trials 2012;13:132.
FDI World Dental Federation: clinical criteria for the [24] Schwendicke F, Goestemeyer G, Gluud C. Cavity lining after
evaluation of direct and indirect restorations-update and excavating caries lesions: meta-analysis and trial
clinical examples. Clin Oral Investig 2010;14:349–66. sequential analysis of randomized clinical trials. J Dent
[7] Hickel R, Roulet JF, Bayne S, Heintze SD, Mjor IA, Peters M, 2015;43(11):1291–7.
et al. Recommendations for conducting controlled clinical [25] Correa MB, Peres MA, Peres KG, Horta BL, Barros AJ,
studies of dental restorative materials. Science Committee Demarco FF. Do socioeconomic determinants affect the
Project 2/98—FDI World Dental Federation study design quality of posterior dental restorations: a multilevel
(Part I) and criteria for evaluation (Part II) of direct and approach. J Dent 2013;41:960–7.
indirect restorations including onlays and partial crowns. J [26] Heintze SD, Ruffieux C, Rousson V. Clinical performance of
Adhes Dent 2007;9(Suppl. 1):121–47. cervical restorations—a meta-analysis. Dent Mater
[8] Hickel R, Roulet JF, Bayne S, Heintze SD, Mjor IA, Peters M, 2010;26:993–1000.
et al. Recommendations for conducting controlled clinical [27] Kopperud SE, Tveit AB, Gaarden T, Sandvik L, Espelid I.
studies of dental restorative materials. Int Dent J Longevity of posterior dental restorations and reasons for
2007;57:300–2. failure. Eur J Oral Sci 2012;120:539–48.
[9] Hickel R, Roulet JF, Bayne S, Heintze SD, Mjor IA, Peters M, [28] Pallesen U, van Dijken JW, Halken J, Hallonsten AL,
et al. Recommendations for conducting controlled clinical Hoigaard R. Longevity of posterior resin composite
studies of dental restorative materials. Clin Oral Investig restorations in permanent teeth in Public Dental Health
2007;11:5–33. Service: a prospective 8 years follow up. J Dent
[10] Schwendicke F, Lamont T, Innes N. Outcomes in Trials for 2013;41:297–306.
Management of Caries Lesions (OuTMaC): protocol. Trials [29] Collares K, Correa MB, Laske M, Kramer E, Reiss B, Moraes
2015;16:397. RR, et al. A practice-based research network on the survival
[11] Göstemeyer G, Blunck U, Paris S, Schwendicke F. Design of ceramic inlay/onlay restorations. Dent Mater
and validity of randomized controlled dental restorative 2016;32:687–94.
trials. Materials (Basel) 2016;9:316. [30] Deliperi S, Bardwell DN, Alleman D. Clinical evaluation of
[12] Ioannidis JP, Greenland S, Hlatky MA, Khoury MJ, Macleod stress-reducing direct composite restorations in
MR, Moher D, et al. Increasing value and reducing waste in structurally compromised molars: a 2-year report. Oper
research design, conduct, and analysis. Lancet Dent 2012;37:109–16.
2014;383:166–75. [31] Kramer N, Garcia-Godoy F, Reinelt C, Feilzer AJ,
[13] Opdam NJ, Bronkhorst EM, Cenci MS, Huysmans MC, Frankenberger R. Nanohybrid vs: fine hybrid composite in
Wilson NH. Age of failed restorations: a deceptive longevity extended class II cavities after six years. Dent Mater
parameter. J. Dent 2011;39:225–30. 2011;27:455–64.
[14] Ozcan M. Lost in translation? J Adhes Dent 2014;16:403. [32] Kramer N, Garcia-Godoy F, Frankenberger R. Evaluation of
[15] Manhart J, Chen H, Hamm G, Hickel R. Buonocore Memorial resin composite materials: part II: in vivo investigations.
Lecture: review of the clinical survival of direct and indirect Am J Dent 2005;18:75–81.
restorations in posterior teeth of the permanent dentition. [33] Opdam NJ, Bronkhorst EM, Loomans BA, Huysmans MC.
Oper Dent 2004;29:481–508. 12-year survival of composite vs: amalgam restorations. J
[16] Schwendicke F, Gostemeyer G, Gluud C. Cavity lining after Dent Res 2010;89:1063–7.
excavating caries lesions: meta-analysis and trial [34] Anusavice KJ. Standardizing failure, success, and survival
sequential analysis of randomized clinical trials. J Dent decisions in clinical studies of ceramic and metal-ceramic
2015;43:1291–7. fixed dental prostheses. Dent Mater 2012;28:102–11.
[17] Soncini JA, Maserejian NN, Trachtenberg F, Tavares M, [35] Williamson P, Altman D, Blazeby J, Clarke M, Devane D,
Hayes C. The longevity of amalgam versus Gargon E, et al. Developing core outcome sets for clinical
compomer/composite restorations in posterior primary trials: issues to consider. Trials 2012;13:132.
and permanent teeth: findings from the New England

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
10 d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

[36] Lamont T, Schwendicke F, Innes N. Why we need a core composites used to restore worn posterior teeth. Int J
outcome set for trials of interventions for prevention and Prosthodont 2006;19:613–7.
management of caries. Evid Based Dent 2015;16:66–8. [55] Beck F, Dumitrescu N, Konig F, Graf A, Bauer P, Sperr W,
[37] Tsakos G, Allen PF, Steele JG, Locker D. Interpreting oral et al. One-year evaluation of two hybrid composites placed
health-related quality of life data. Community Dent Oral in a randomized-controlled clinical trial. Dent Mater
Epidemiol 2012;40:193–200. 2014;30:824–38.
[38] Rasines Alcaraz MG, Veitz-Keenan A, Sahrmann P, [56] Bekes K, Boeckler L, Gernhardt CR, Schaller HG. Clinical
Schmidlin PR, Davis D, Iheozor-Ejiofor Z. Direct composite performance of a self-etching and a total-etch adhesive
resin fillings versus amalgam fillings for permanent or system — 2-year results. J Oral Rehabil 2007;34:855–61.
adult posterior teeth. Cochrane Database Syst Rev [57] Boeckler A, Boeckler L, Eppendorf K, Schaller HG,
2014;3:CD005620. Gernhardt CR. A prospective, randomized clinical trial of a
[39] Pallesen U, van Dijken JW. A randomized controlled 27 two-step self-etching vs two-step etch-and-rinse adhesive
years follow up of three resin composites in class II and SEM margin analysis: four-year results. J Adhes Dent
restorations. J Dent 2015;43:1547–58. 2012;14:585–92.
[40] Pallesen U, van Dijken JW. A randomized controlled 30 [58] Boeckler A, Schaller HG, Gernhardt CR. A prospective,
years follow up of three conventional resin composites in double-blind, randomized clinical trial of a one-step,
class II restorations. Dent Mater 2015;31:1232–44. self-etch adhesive with and without an intermediary layer
[41] Schwendicke F, Blunck U, Paris S, Gostemeyer G. Choice of of a flowable composite: a 2-year evaluation. Quintessence
comparator in restorative trials: a network analysis. Dent Int 2012;43:279–86.
Mater 2015;31:1502–9. [59] Bottenberg P, Alaerts M, Keulemans F. A prospective
[42] van de Sande FH, Opdam NJ, Rodolpho PA, Correa MB, randomised clinical trial of one bis-GMA-based and two
Demarco FF, Cenci MS. Patient risk factors’ influence on ormocer-based composite restorative systems in class II
survival of posterior composites. J Dent Res cavities: three-year results. J Dent 2007;35:163–71.
2013;92:78s–83s. [60] Dresch W, Volpato S, Gomes JC, Ribeiro NR, Reis A,
[43] Da Rosa Rodolpho PA, Donassollo TA, Cenci MS, Loguercio Loguercio AD. Clinical evaluation of a nanofilled composite
AD, Moraes RR, Bronkhorst EM, et al. 22-Year clinical in posterior teeth: 12-month results. Oper Dent
evaluation of the performance of two posterior composites 2006;31:409–17.
with different filler characteristics. Dent Mater [61] Ergucu Z, Turkun LS. Clinical performance of novel resin
2011;27:955–63. composites in posterior teeth: 18-month results. J Adhes
[44] Baldissera RA, Correa MB, Schuch HS, Collares K, Dent 2007;9:209–16.
Nascimento GG, Jardim PS, et al. Are there universal [62] Ernst CP, Brandenbusch M, Meyer G, Canbek K, Gottschalk
restorative composites for anterior and posterior teeth. J F, Willershausen B. Two-year clinical performance of a
Dent 2013;41:1027–35. nanofiller vs a fine-particle hybrid resin composite. Clin
[45] McCracken MS, Gordan VV, Litaker MS, Funkhouser E, Oral Investig 2006;10:119–25.
Fellows JL, Shamp DG, et al. A 24-month evaluation of [63] Frankenberger R, Reinelt C, Kramer N. Nanohybrid vs. fine
amalgam and resin-based composite restorations: findings hybrid composite in extended class II cavities: 8-year
from The National Dental Practice-Based Research results. Clin Oral Investig 2014;18:125–37.
Network. J Am Dent Assoc 2013;144:583–93. [64] Goncalves FS, Leal CD, Bueno AC, Freitas AB, Moreira AN,
[46] Correa MB, Peres MA, Peres KG, Horta BL, Barros AD, Magalhaes CS. A double-blind randomized clinical trial of a
Demarco FF. Amalgam or composite resin: factors silorane-based resin composite in class 2 restorations:
influencing the choice of restorative material. J Dent 18-month follow-up. Am J Dent 2013;26:93–8.
2012;40:703–10. [65] Koubi G, Colon P, Franquin JC, Hartmann A, Richard G,
[47] Göstemeyer G, Blunck U, Paris S, Schwendicke F. Design and Faure MO, et al. Clinical evaluation of the performance and
validity of randomized controlled dental restorative trials. safety of a new dentine substitute, Biodentine, in the
Materials 2016;9:316, http://dx.doi.org/10.3390/ma9050372. restoration of posterior teeth — a prospective study. Clin
[48] Lesaffre E, Philstrom B, Needleman I, Worthington H. The Oral Investig 2013;17:243–9.
design and analysis of split-mouth studies: what [66] Kramer N, Reinelt C, Garcia-Godoy F, Taschner M, Petschelt
statisticians and clinicians should know. Stat Med A, Frankenberger R. Nanohybrid composite vs: fine hybrid
2009;28:3470–82. composite in extended class II cavities: clinical and
[49] Schwendicke F, Blunck U, Paris S, Gostemeyer G. Choice of microscopic results after 2 years. Am J Dent 2009;22:228–34.
comparator in restorative trials: a network analysis. Dent [67] Kramer N, Reinelt C, Richter G, Petschelt A, Frankenberger
Mater 2015;31(12):1502–9. R. Nanohybrid vs: fine hybrid composite in class II cavities:
[50] Schwendicke F, Tu YK, Blunck U, Paris P, Göstemeyer G. clinical results and margin analysis after four years. Dent
Effect of industry sponsorship on dental restorative trials. J Mater 2009;25:750–9.
Dent Res 2016;95(1):9–16. [68] Manhart J, Chen HY, Hickel R. Three-year results of a
[51] Andrade AK, Duarte RM, Silva FD, Batista AU, Lima KC, randomized controlled clinical trial of the posterior
Pontual ML, et al. Clinical evaluation of nanofill and composite QuiXfil in class I and II cavities. Clin Oral
nanohybrid composite in class I restorations: a 12-month Investig 2009;13:301–7.
randomized trial. Gen Dent 2012;60:e255–62. [69] Manhart J, Chen HY, Hickel R. Clinical evaluation of the
[52] Baracco B, Perdigao J, Cabrera E, Ceballos L. Two-year posterior composite Quixfil in class I and II cavities: 4-year
clinical performance of a low-shrinkage composite in follow-up of a randomized controlled trial. J Adhes Dent
posterior restorations. Oper Dent 2013;38:591–600. 2010;12:237–43.
[53] Baracco B, Perdigao J, Cabrera E, Giraldez I, Ceballos L. [70] Schirrmeister JF, Huber K, Hellwig E, Hahn P. Two-year
Clinical evaluation of a low-shrinkage composite in evaluation of a new nano-ceramic restorative material. Clin
posterior restorations: one-year results. Oper Dent Oral Investig 2006;10:181–6.
2012;37:117–29. [71] Schirrmeister JF, Huber K, Hellwig E, Hahn P. Four-year
[54] Bartlett D, Sundaram G. An up to 3-year randomized evaluation of a resin composite including nanofillers in
clinical study comparing indirect and direct resin posterior cavities. J Adhes Dent 2009;11:399–404.

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx 11

[72] Shi L, Wang X, Zhao Q, Zhang Y, Zhang L, Ren Y, et al. [90] van Dijken JW, Lindberg A. Clinical effectiveness of a
Evaluation of packable and conventional hybrid resin low-shrinkage resin composite: a five-year evaluation. J
composites in class I restorations: three-year results of a Adhes Dent 2009;11:143–8.
randomized, double-blind and controlled clinical trial. Oper [91] van Dijken JW, Pallesen U. Clinical performance of a hybrid
Dent 2010;35:11–9. resin composite with and without an intermediate layer of
[73] Swift Jr EJ, Ritter AV, Heymann HO, Sturdevant JR, Wilder Jr flowable resin composite: a 7-year evaluation. Dent Mater
AD. 36-month clinical evaluation of two adhesives and 2011;27:150–6.
microhybrid resin composites in class I restorations. Am J [92] van Dijken JW, Pallesen U. Four-year clinical evaluation of
Dent 2008;21:148–52. class II nano-hybrid resin composite restorations bonded
[74] van Dijken JW, Pallesen U. A randomized 10-year with a one-step self-etch and a two-step etch-and-rinse
prospective follow-up of class II nanohybrid and adhesive. J Dent 2011;39:16–25.
conventional hybrid resin composite restorations. J Adhes [93] van Dijken JW, Pallesen U. A six-year prospective
Dent 2014;16:585–92. randomized study of a nano-hybrid and a conventional
[75] Yazici AR, Ustunkol I, Ozgunaltay G, Dayangac B. hybrid resin composite in class II restorations. Dent Mater
Three-year clinical evaluation of different restorative 2013;29:191–8.
resins in class I restorations. Oper Dent 2014;39:248–55. [94] van Dijken JW, Pallesen U. Randomized 3-year clinical
[76] Manhart J, Chen HY, Neuerer P, Thiele L, Jaensch B, Hickel evaluation of class I and II posterior resin restorations
R. Clinical performance of the posterior composite QuiXfil placed with a bulk-fill resin composite and a one-step
after 3, 6, and 18 months in class 1 and 2 cavities. self-etching adhesive. J Adhes Dent 2015;17:81–8.
Quintessence Int 2008;39:757–65. [95] Van Dijken JW, Sunnegardh-Gronberg K. A four-year clinical
[77] Rho YJ, Namgung C, Jin BH, Lim BS, Cho BH. Longevity of evaluation of a highly filled hybrid resin composite in
direct restorations in stress-bearing posterior cavities: a posterior cavities. J Adhes Dent 2005;7:343–9.
retrospective study. Oper Dent 2013;38:572–82. [96] van Dijken JW, Sunnegardh-Gronberg K. Fiber-reinforced
[78] Cetin AR, Unlu N. One-year clinical evaluation of direct packable resin composites in class II cavities. J Dent
nanofilled and indirect composite restorations in posterior 2006;34:763–9.
teeth. Dent Mater J 2009;28:620–6. [97] Adolphi G, Zehnder M, Bachmann LM, Gohring TN. Direct
[79] Demarco FF, Cenci MS, Lima FG, Donassollo TA, Andre Dde resin composite restorations in vital versus root-filled
A, Leida FL. Class II composite restorations with metallic posterior teeth: a controlled comparative long-term
and translucent matrices: 2-year follow-up findings. J Dent follow-up. Oper Dent 2007;32:437–42.
2007;35:231–7. [98] Al-Samhan A, Al-Enezi H, Alomari Q. Clinical evaluation of
[80] Demarco FF, Pereira-Cenci T, de Almeida Andre D, de Sousa posterior resin composite restorations placed by dental
Barbosa RP, Piva E, Cenci MS. Effects of metallic or students of Kuwait University. Med Princ Pract
translucent matrices for class II composite restorations: 2010;19:299–304.
4-year clinical follow-up findings. Clin Oral Investig [99] da Rosa Rodolpho PA, Cenci MS, Donassollo TA, Loguercio
2011;15:39–47. AD, Demarco FF. A clinical evaluation of posterior
[81] Lindberg A, van Dijken JW, Lindberg M. Nine-year composite restorations: 17-year findings. J Dent
evaluation of a polyacid-modified resin composite/resin 2006;34:427–35.
composite open sandwich technique in class II cavities. J [100] Moura FR, Romano AR, Lund RG, Piva E, Rodrigues Jr SA,
Dent 2007;35:124–9. Demarco FF. Three-year clinical performance of composite
[82] Loguercio AD, Reis A, Hernandez PA, Macedo RP, Busato AL. restorations placed by undergraduate dental students. Braz
3-Year clinical evaluation of posterior packable composite Dent J 2011;22:111–6.
resin restorations. J Oral Rehabil 2006;33:144–51. [101] Opdam NJ, Bronkhorst EM, Roeters JM, Loomans BA.
[83] Pazinatto FB, Gionordoli Neto R, Wang L, Mondelli J, Longevity and reasons for failure of sandwich and
Mondelli RF, Navarro MF. 56-month clinical performance of total-etch posterior composite resin restorations. J Adhes
class I and II resin composite restorations. J Appl Oral Sci Dent 2007;9:469–75.
2012;20:323–8. [102] Opdam NJ, Bronkhorst EM, Roeters JM, Loomans BA. A
[84] Poon EC, Smales RJ, Yip KH. Clinical evaluation of packable retrospective clinical study on longevity of posterior
and conventional hybrid posterior resin-based composites: composite and amalgam restorations. Dent Mater
results at 3.5 years. J Am Dent Assoc 2005;136:1533–40. 2007;23:2–8.
[85] Sadeghi M, Lynch CD, Shahamat N. Eighteen-month [103] Vahanikkila H, Kakilehto T, Pihlaja J, Pakkila J, Tjaderhane
clinical evaluation of microhybrid, packable and nanofilled L, Suni J, et al. A data-based study on survival of permanent
resin composites in class I restorations. J Oral Rehabil molar restorations in adolescents. Acta Odontol Scand
2010;37:532–7. 2014;72:380–5.
[86] Stefanski S, van Dijken JW. Clinical performance of a [104] Arhun N, Celik C, Yamanel K. Clinical evaluation of
nanofilled resin composite with and without an resin-based composites in posterior restorations: two-year
intermediary layer of flowable composite: a 2-year results. Oper Dent 2010;35:397–404.
evaluation. Clin Oral Investig 2012;16:147–53. [105] Banomyong D, Harnirattisai C, Burrow MF. Posterior resin
[87] Sundfeld RH, Scatolin RS, Oliveira FG, Machado LS, composite restorations with or without resin-modified,
Alexandre RS, Sundefeld ML. One-year clinical evaluation glass-ionomer cement lining: a 1-year randomized, clinical
of composite restorations in posterior teeth: effect of trial. J Invest Clin Dent 2011;2:63–9.
adhesive systems. Oper Dent 2012;37:E1–8. [106] Brackett WW, Browning WD, Brackett MG, Callan RS,
[88] van Dijken JW. Durability of resin composite restorations in Blalock JS. Effect of restoration size on the clinical
high C-factor cavities: a 12-year follow-up. J Dent performance of posterior packable resin composites over
2010;38:469–74. 18 months. Oper Dent 2007;32:212–6.
[89] van Dijken JW. A 6-year prospective evaluation of a [107] Browning WD, Myers ML, Chan DC, Downey MC, Pohjola
one-step HEMA-free self-etching adhesive in class II RM, Frazier KB. Performance of 2 packable composites at 12
restorations. Dent Mater 2013;29:1116–22. months. Quintessence Int 2006;37:361–8.

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187
 DENTAL-3007; No. of Pages 12
ARTICLE IN PRESS
12 d e n t a l m a t e r i a l s x x x ( 2 0 1 7 ) xxx–xxx

[108] Celik C, Arhun N, Yamanel K. Clinical evaluation of [119] Mahmoud SH, Ali AK, Hegazi HA. A three-year prospective
resin-based composites in posterior restorations: a 3-year randomized study of silorane- and methacrylate-based
study. Med Princ Pract 2014;23:453–9. composite restorative systems in class II restorations. J
[109] Cetin AR, Unlu N, Cobanoglu N. A five-year clinical Adhes Dent 2014;16:285–92.
evaluation of direct nanofilled and indirect composite resin [120] Mahmoud SH, El-Embaby AE, AbdAllah AM. Clinical
restorations in posterior teeth. Oper Dent 2013;38:E1–11. performance of ormocer, nanofilled, and nanoceramic
[110] de Souza FB, Guimaraes RP, Silva CH. A clinical evaluation resin composites in class I and class II restorations: a
of packable and microhybrid resin composite restorations: three-year evaluation. Oper Dent 2014;39:32–42.
one-year report. Quintessence Int 2005;36:41–8. [121] Mahmoud SH, El-Embaby AE, AbdAllah AM, Hamama HH.
[111] Efes BG, Dorter C, Gomec Y. Clinical evaluation of an Two-year clinical evaluation of ormocer, nanohybrid and
ormocer, a nanofill composite and a hybrid composite at 2 nanofill composite restorative systems in posterior teeth. J
years. Am J Dent 2006;19:236–40. Adhes Dent 2008;10:315–22.
[112] Efes BG, Dorter C, Gomec Y, Koray F. Two-year clinical [122] Monteiro PM, Manso MC, Gavinha S, Melo P. Two-year
evaluation of ormocer and nanofill composite with and clinical evaluation of packable and nanostructured
without a flowable liner. J Adhes Dent 2006;8:119–26. resin-based composites placed with two techniques. J Am
[113] Ermis RB, Kam O, Celik EU, Temel UB. Clinical evaluation of Dent Assoc 2010;141:319–29.
a two-step etch&rinse and a two-step self-etch adhesive [123] Ozakar-Ilday N, Zorba YO, Yildiz M, Erdem V, Seven N,
system in class II restorations: two-year results. Oper Dent Demirbuga S. Three-year clinical performance of two
2009;34:656–63. indirect composite inlays compared to direct composite
[114] Fagundes TC, Barata TJ, Bresciani E, Cefaly DF, Jorge MF, restorations. Med Oral Patol Oral Cir Bucal 2013;18:e521–8.
Navarro MF. Clinical evaluation of two packable posterior [124] Perdigao J, Dutra-Correa M, Anauate-Netto C, Castilhos N,
composites: 2-year follow-up. Clin Oral Investig Carmo AR, Lewgoy HR, et al. Two-year clinical evaluation of
2006;10:197–203. self-etching adhesives in posterior restorations. J Adhes
[115] Fagundes TC, Barata TJ, Carvalho CA, Franco EB, van Dijken Dent 2009;11:149–59.
JW, Navarro MF. Clinical evaluation of two packable [125] Spreafico RC, Krejci I, Dietschi D. Clinical performance and
posterior composites: a five-year follow-up. J Am Dent marginal adaptation of class II direct and semidirect
Assoc 2009;140:447–54. composite restorations over 3.5 years in vivo. J Dent
[116] Fennis WM, Kuijs RH, Roeters FJ, Creugers NH, Kreulen CM. 2005;33:499–507.
Randomized control trial of composite cuspal restorations: [126] Walter R, Boushell LW, Heymann HO, Ritter AV, Sturdevant
five-year results. J Dent Res 2014;93:36–41. JR, Wilder Jr AD, et al. Three-year clinical evaluation of a
[117] Gianordoli Neto R, Santiago SL, Mendonca JS, Passos VF, silorane composite resin. J Esthet Restor Dent
Lauris JR, Navarro MF. One year clinical evaluation of two 2014;26:179–90.
different types of composite resins in posterior teeth. J
Contemp Dent Pract 2008;9:26–33.
[118] Lange RT, Pfeiffer P. Clinical evaluation of ceramic inlays
compared to composite restorations. Oper Dent
2009;34:263–72.

Please cite this article in press as: Opdam NJM, et al. Clinical studies in restorative dentistry: New directions and new demands. Dent Mater
(2017), http://dx.doi.org/10.1016/j.dental.2017.08.187

View publication stats