Clin Oral Invest

DOI 10.1007/s00784-015-1576-3

ORIGINAL ARTICLE

Short fibre-reinforced composite for extensive direct restorations:

a laboratory and computational assessment
Bruno Castro Ferreira Barreto 1,2 & Annelies Van Ende 1 & Diogo Pedrollo Lise 1 &
Pedro Yoshito Noritomi 3 & Siegfried Jaecques 2 & Jos Vander Sloten 2 & Jan De Munck 1 &
Bart Van Meerbeek 1

Received: 7 July 2015 / Accepted: 18 August 2015

# Springer-Verlag Berlin Heidelberg 2015

Abstract (p = 0.000). Failure mode analysis revealed more repairable

Objectives The objective of the study was to evaluate the
effectiveness of a short fibre-reinforced composite (FRC) ap-
plied in combination with a conventional filler composite
(CFC) on the fatigue resistance, fracture strength, failure mode
and stress distribution, for restorations of premolars under two
loading angles.
Material and methods Thirty-two inferior premolars received
extensive cavities with removal of the lingual cusp. Teeth
were restored directly using ‘FRC (EverX Posterior, GC) +
CFC (G-aenial, GC)’ or ‘CFC only’ and received two fatigue/
fracture loadings at two different angles (0°/45°) (n = 8). Data
were submitted to two-way ANOVA (α = 5 %) and Tukey
test. Failure mode was analysed using SEM. Four 3D finite
element (FE) models were constructed and static, linear and
elastic analyses were performed. Maximum principal and von
Mises stresses were evaluated.
Results All specimens survived the mechanical fatigue simu-
lation. No statistical difference in fracture resistance was re-
corded between FRC + CFC and CFC only, considering both
loading angles (p = 0.115). However, the 0° loading showed a
statistical significant higher strength than the 45° loading
* Bart Van Meerbeek
bart.vanmeerbeek@med.kuleuven.be
1
BIOMAT, Department of Oral Health Sciences, KU Leuven
(University of Leuven) & Dentistry, University Hospitals Leuven,
Leuven, Belgium
2
BMGO, Biomechanics Section (BMe), Department of Mechanical
Engineering, KU Leuven (University of Leuven), Leuven, Belgium
3
Division of Tridimensional Technologies, Center for Information
Technology Renato Archer (DT3D/CTI), Campinas, São Paulo,
Brazil
fractures upon 0° loading, versus more root fractures
(unrepairable) upon 45° loading. FE revealed a higher amount
of stress upon 45° loading, with tensile stress being imposed
to the lingual cervical area.
Conclusion The fracture strength was not increased using the
FRC. Loading at a 45° decreased significantly the fracture
resistance.
Clinical relevance The restoration of extensive cavities in
posterior tooth is a challenge for the clinicians and the choice
of the material that increases the fracture strength of tooth-
restoration complex is required.
Keywords Fatigue resistance . Finite element analysis .
Fracture strength . Loading angle . Short fibre-reinforced
composites
Introduction
Loss of tooth structure as result of caries, trauma or restorative
and endodontic procedures increases the risk of cusp fracture
of posterior teeth. In order to restore severely damaged teeth,
especially when requiring cusp replacement, indirect (or semi-
direct) restorative techniques are employed, thereby signifi-
cantly increasing the treatment cost [1]. Moreover, the least
invasive and most tooth tissue-preserving approach should be
followed at all times, this to postpone more invasive restor-
ative techniques, such as ‘post & core’, as long as possible in
light of the long-term tooth survival [1–3]. Intending to pro-
vide dentists restorative means that are more affordable for
patients, a tendency today exists to also apply direct compos-
ites to restore large cavities, including build-up of one or more
missing cusps [1]. Despite today’s composites presenting ap-
propriate physico-mechanical properties (along with aesthetic

potential) [4], not much laboratory data and hardly any clinical
data are available on the performance and longevity of cusp-
replacing composite restorations [5].
Furthermore, a major clinical problem remains fracture,
varying from small chipping at proximal margins to large bulk
fractures of posterior composite restorations [6]. This inferior
fracture resistance should be ascribed to its composition of a
resin matrix highly loaded with filler [7]. As a result, a rela-
tively high brittleness and low fracture toughness of current
conventional filler composites (CFCs) still limit their use in
large stress-bearing restorations [8]. In addition, the polymer-
ization shrinkage that is inherent of dental composites can
generate stress within the restored tooth and in particular at
their mutual interface [9]. This stress has been associated with
several clinical symptoms, such as cusp deflection, micro-
fractures in enamel, marginal leakage and post-operative sen-
sitivity [10].
In order to overcome this shortcoming, a short fibre-
reinforced composite (FRC) (EverX Posterior, GC, Tokyo,
Japan) has been introduced as a dental restorative composite
with the goal to enlarge the indication of direct composites to
extensive posterior cavities, as well as to mimic the stress-
absorbing properties of dentin and the dentin-enamel junction
(DEJ) [11]. The FRC is intended to be used in high stress-
bearing areas [2], thanks to an improved load-bearing capac-
ity, a higher flexural strength and fracture toughness as com-
pared to those of a CFC [12–14]. Furthermore, the FRC has
also been claimed to control polymerization-shrinkage stress
more favourably within the restoration, this based on aniso-
tropic fibre orientation [6, 15].
The aim of this study was to assess the effect of using FRC
as dentin base replacement in combination with CFC as outer
enamel replacement (‘FRC + CFC’) versus ‘CFC only’ on the
fatigue resistance, fracture strength and failure mode of direct-
ly restored premolars, including a cusp build-up, and when
tested under two loading angles (0° and 45°). The stress dis-
tribution of FRC + CFC and CFC only restored premolars,
when subjected to the two angle loadings (0° and 45°), was
assessed by mathematical models in a computational simula-
tion using finite element analysis (FEA). The first null hypoth-
eses tested were that both FRC + CFC and CFC only restored
premolars have similar mechanical behaviour, and the second
one is that there is no difference in mechanical characteristics
upon loading following the 0° and 45° loading angle.
Materials and methods
Sample preparation
Thirty-two non-carious and non-restored human mandibular
premolars (gathered following informed consent approved by
the Commission for Medical Ethics of KU Leuven under the
Clin Oral Invest
file number S57622) were stored in 0.5 % chloramine in water
at 4 °C and used within 3 months after extraction. The teeth
were selected to have an inter-cuspal width within a maximum
deviation of not more than 10 % of the determined mean. The
measured inter-cuspal width varied between 6.3 and 7.7 mm.
The periodontal ligaments were imitated using an elastic self-
hardening material (Erkoskin, Erkodent, Wembley, Australia).
After that, the teeth were embedded in acrylic resin (Clarocit,
Struers, Ballerup, Denmark) up to 2 mm below the cementum-
enamel junction to simulate the alveolar bone, and impres-
sions of all tooth crowns were taken using a transparent poly-
vinyl siloxane impression material (Memosil, Heraeus Kulzer,
Hanau, Germany) to facilitate the reconstruction of the origi-
nal proximal anatomy of each tooth.
Mesial-occlusal-distal (MOD) class-II cavities were pre-
pared using a custom-adapted MicroSpecimen Former (The
University of Iowa, Iowa city, IA, USA), equipped with a
cylindrical diamond bur (842.314.014, Komet, Lemgo, Ger-
many). The width of the cavity was determined as the inter-
cuspal distance, the occlusal preparation was 4 mm deep and
there were no proximal boxes. Next, the lingual cusp was
removed up to the same depth as the cavity bottom. The buc-
cal cusp was approximately 2 mm thick at the cervical area. In
addition, the pulp chamber was accessed using a cylindrical
bur up to a 2-mm depth, leaving a shoulder with a 1-mm
thickness (Fig. 1). The root canals were irrigated with 1 %
of NaOCl and a zinc-phosphate cement (Dentsply,
Addlestone, Surrey, UK) was placed in the canal orifice up
to the level of the cavity floor.
Fig. 1 The dimensions of the cavity prepared in each premolar
Clin Oral Invest
The prepared teeth were randomly divided into four groups
(n = 8), following combinations of the two study factors. Di-
rect composite cusp build-ups were fabricated using a short
fibre-reinforced composite (FRC, EverX Posterior, GC) to
replace dentin; this base was then covered with a 2-mm layer
of conventional filler composite (CFC, G-aenial, GC). This
experimental group is further being referred to as FRC +
CFC. As control, only CFC is used and being referred to as
CFC only. All specimens were submitted to fatigue and frac-
ture testing following two loading directions (0° and 45°).
The enamel margins were selectively etched using
phosphoric acid (Scotchbond Etchant Gel 35 %, 3M
ESPE, Seefeld, Germany) for 15 s, rinsed with water for
15 s and gently dried with air. The transparent impressions
were cut at the top to remove the occlusal part and were
subsequently positioned on the corresponding tooth. The
one-step self-etch adhesive G-aenial Bond (GC) was ap-
plied following the IFU of the manufacturer and then
light-cured for 5 s using the light-curing unit G-light
(GC), which has a light-curing output of 1200 mW/cm2.
The proximal walls were reconstructed with three incre-
ments of CFC with a 1-mm thickness each, this consecu-
tively to restore the mesial, distal and lingual wall; each
increment was light-cured for 20 s (G-light, GC).
For the FRC + CFC group, a bulk increment with a 4-mm
base of FRC was inserted and condensed inside the cavity,
before being light-cured for 10 s. A 2-mm cover layer of
CFC was put on the FRC, after which a ceramic ball with a
6-mm diameter was pressed on the last increment to create a
niche for loading in the fatigue resistance and fracture strength
test. This niche is expected to distribute the loading forces in
all directions, independent of the height of the cusps. The CFC
increment was light-cured twice for 20 s (G-light, GC), before
and after the removal of the ceramic ball.
For the CFC only group, the remaining class-I cavity was
restored using two 2-mm increments of CFC that were light-
cured for 20 s each (G-light, GC). A final 2-mm cover layer of
CFC was applied, in which the abovementioned ceramic ball
was pressed and light-cured twice for 20 s (G-light, GC).
After 24-h water storage at 4 °C, the restorations were
finished with ultra-fine diamond burs (8882.314.014, Komet,
Lemgo, Germany) and finally polished with soft-lex discs
(3M ESPE).
Fatigue resistance and fracture strength test
Specimens of the four experimental groups were submitted to
1,200,000 cycles with a frequency of 2 Hz and 50-N loading
using a chewing simulation machine (SD Mechatronik
GmbH, Chewing Simulator, Willytec, Feldkirchen-
Westerham, Germany). Two sets of specimens were loaded
following a 0° angle (parallel to the long axis of the tooth)
or a 45° angle (oblique to the long axis of the tooth), this using
an antagonist ceramic ball with a 6-mm diameter (Steatite,
Hoechst CeramTec, Wunsiedel, Germany) directed towards
the floor of the cavity for the 0°-angle loading and towards
the remaining cusp for the 45°-angle loading (Fig. 2). The
specimens were kept humid throughout the dynamic loading
in the chewing simulator.
The specimens that survived the chewing simulation were
mounted in a fracture-test setup loaded in compression until
fracture (Instron 5848 microtester, Norwood, MA, USA) and
this following the same loading angles as applied in the
chewing fatigue test. As antagonist structure, a steel ball with
a 6-mm diameter was employed. The loading was applied on
the niche foreseen at the occlusal surface using a 0.5-
mm min−1 cross-head speed.
The force needed to fracture each tooth was recorded in
Newton (N). The mean and the corresponding 95 % confi-
dence interval of the fracture loading for each group were
calculated and statistically compared using a Tukey test.
Failure mode analysis using SEM
After fracture, all specimens were fixed in 2.5 % glutaralde-
hyde in cacodylate buffer solution, dehydrated in ascending
concentrations of ethanol and chemically dried using
hexamethyldisilazane (HMDS). After they were mounted on
aluminium stubs and gold sputter-coated using a gold-
sputtering device (Sputtering Device 07 120, Balzers Union,
Balzers, Liechtenstein), the fractured specimens were exam-
ined with a field-emission-gun scanning electron microscope
(Feg-SEM XL30, Philips, Eindhoven, Netherlands). Fractures
were identified as either ‘repairable’, ‘root fracture’ or ‘cusp
de-bonding’.
Stress distribution—FEA
A sound human premolar with the average size was selected.
The tooth was subjected to contact scanning readings
Fig. 2 The loading angles of 0° in a and of 45° in b

(MDX-40, Roland Models, Cotia, SP, Brazil) in order to
register the outer enamel contour in STL-format file. The
root surface was next protected using wax and the crown
immersed in 10 % hydrochloric acid solution for 10 min
to completely dissolve enamel. After enamel dissolution,
the coronal portion was scanned again to determine the
outer dentin contour. The files were exported to a model-
ling software (Rhinoceros 4.0, McNeel, USA), and a Bio-
CAD protocol was used to generate the 3D model [16].
The preparation cavity and the composite cusp build-up
restoration were derived from the standard model, gener-
ating four models, considering FRC + CFC and CFC only
groups upon two different loading angles (0° and 45°).
The adhesive interface was not modelled separately but
represented by glued contacts between composite and
tooth structure. The periodontal ligament was simulated
using polyether properties and a cylinder support with
polystyrene resin properties was also modelled. These
models were mounted and exported to FEA software de-
veloped for stress analysis (Ansys Workbench 15, PA,
USA). The meshes were generated for each structure with
manual control to ensure proper refinement at the areas of
interest, focusing on homogeneity, congruency and ade-
quate connectivity between the structures. Solid quadratic
tetrahedral elements with ten nodes were used. The me-
chanical properties used are described in Table 1; all struc-
tures were considered isotropic, linear and homogeneous.
The boundary conditions, restrictions of the model, load-
ing application and stresses were also defined at this stage.
To simulate the 0° loading, a constant force of 50 N was
applied to the total area of the niche, as a pressure parallel
to the long axis of the tooth model (Fig. 2). To simulate
the 45° loading, a constant force of 50 N was applied as a
pressure perpendicular to the facial area of the niche at the
buccal cusp and directed to the buccal cusp (Fig. 2). The
restrictions of the model were performed at both the base
and lateral surface of the cylinder in 6 degrees of freedom
(X-, Y- and Z-axis). Maximum principal and von Mises
stresses were used for data analysis and comparison
among the different models.
Table 1 Substrate and material properties extracted out of literature
Substrate/Material Elastic modulus (E) (GPa)
Enamel 46.8
Dentin 18.0
G-aenial Posterior 8.2a
a
EverX Posterior 12.3
Polyether 0.05
Polystyrene resin 13.5
a
Data provided by the manufacturer (GC, Tokyo, Japan)
Clin Oral Invest
Results
All specimens survived the mechanical fatigue simulation.
The fracture resistance results are shown in Table 2 for the
four experimental groups. No statistical difference in fracture
resistance was recorded between the FRC + CFC and CFC
only experimental groups, considering both loading angles
(p = 0.115). The 45° loading angle significantly decreased
the fracture resistance of the restored teeth, but this is irrespec-
tive of whether the fibre-reinforced composite was used or not
(p = 0.000).
The failure mode analysis revealed a predominance of re-
pairable fractures for the 0° angle loading; for the 45° angle
loading, unrepairable root fracture and cusp de-bonding oc-
curred more often (Fig. 3a–e).
For the FRC + CFC models, FEA disclosed a higher
amount of von Mises stress concentrated at the cervical
area of the buccal surface, with a stress peak of 900 MPa
at the dentin-enamel junction (DEJ) for the 0° angle load-
ing model (Fig. 4a), and a stress peak of 1104 MPa at the
DEJ for the 45° angle loading model (Fig. 4b). For the
CFC only models, FEA disclosed the same pattern of
stress distribution as for the FRC + CFC models, with a
stress peak of 959 MPa at the DEJ for the 0° angle
loading model (Fig. 4c) and a stress peak of 1185 MPa
at the DEJ for the 45° angle loading model (Fig. 4d).
When maximum principal stress was analysed, the 0° an-
gle loading models showed a high concentration of stress
at the cavo-surface angle of the niche on the occlusal area
of the composite restoration, which absorbed the stress
and did not transmit it to the tooth structure. For those
models that received 45° loading, the peak of tensile stress
was concentrated at DEJ of the lingual surface area. In the
FRC + CFC model, the peak of tensile stress was
298 MPa for 0° angle loading model (Fig. 5a) and
485 MPa for the 45° angle loading model (Fig. 5b). In
the CFC only model, the peak of tensile stress was
318 MPa in the model for 0° angle loading model
(Fig. 5c) and 539 MPa for 45° angle loading model
(Fig. 5d).
Poisson’s ratio (ϑ) References
0.3 Wright and Yettram, [17]
0.31 Rees et al., 1994 [18]
0.24 Versluis et al., 2004 [10]
0.24 Versluis et al., 2004 [10]
0.45 Farah et al., 1981 [19]
0.31 Stafford et al., 2004 [20]
Clin Oral Invest
Table 2 Fracture resistance in Newton (N)
Loading angle ‘CFC only’
(G-aenial Posterior, GC)
0° 1222.0 (±251.9) Aa
45° 628.3 (±163.9) Ab
Different capital letters and lower case letters represent statistical differences in the rows and in the columns, respectively (two-way ANOVA with 5 % as
level of significance and Tukey test)
Discussion
In accordance with the results of the present study, the con-
ventional composite restoration showed a similar mechanical
resistance as that of the short fibre-reinforced composite res-
toration, by which the first null hypothesis was accepted. The
values of the fracture strength test were influenced by the
different loading angles, with lower data being recorded for
the 45° loading, this irrespective of the composite used. There-
fore, the second null hypothesis was rejected.
The fibre-reinforced composite has been developed with
the intention to enhance the longevity of medium-to-large-
sized composite restorations in posterior teeth [11]. Its resin
Fig. 3 Failure mode for the 0° loading in a and the 45° loading in b; the failures were categorized as ‘repairable’ (black arrows) in c or root fracture
(black arrows) in d. Cusp de-bonding (black arrows) is shown in e
‘FRC + CFC’
(EverX Posterior + G-aenial Posterior, GC)
1329.2 (±185.4) Aa
765.8 (±236.4) Ab
matrix contains conventional methacrylate monomers like
Bis-GMA and TEGDMA, to which PMMA is added to form
a matrix that has been referred to as a ‘semi-interpenetrating
polymer network’ (semi-IPN net-poly(methylmethacrylate)-
inter-net-poly(bis-glycidyl-A-dimethacrylate) [21]. This
semi-IPN has been claimed to provide good bonding and wet-
ting properties to the tooth substrate as well as to improve the
fracture toughness of the polymer matrix [21]. Previous stud-
ies showed an improvement in mechanical and
polymerization-stress properties as compared to conventional
composite [15, 22, 23]. The presence of fibres results in an
anisotropic characteristic [15] that has been suggested to re-
lieve stress and prevent crack propagation, thereby increasing
 Clin Oral Invest

Fig. 4 Distribution of von Mises

stress (MPa) for the ‘FRC + CFC’
models upon 0° loading in a and
upon 45° loading in b. The ‘CFC
only’ model upon 0° loading is
shown in c and upon 45° loading
in d

the load-bearing capacity of the restoration itself [5]. The re-
inforcing effect of the fibres is based on stress transfer from
the polymer matrix to the fibres but also on behaviour of
individual fibres as crack stoppers [24].
Nevertheless, our results showed that no restored tooth has
broken during the fatigue test in the chewing simulator. This
finding suggests a good performance of both the FRC and
CFC composite restoration. This study tested the fatigue re-
sistance as a dynamic laboratory assessment of restored teeth
[25], which is thought to comply more with clinical findings
than a test involving a single episode of high occlusal stress
[26]. It is known that teeth are subjected to repetitive occlusal
loading during normal function; the chewing loading is esti-
mated to reach peaks of 50–60 N and approximately 1,200,
000 chewing cycles were suggested to correspond to a time
period of 5 years [27, 28].
Since the restorations did not fail in this study upon
chewing, the fatigue simulation served as artificial ageing of

Fig. 5 Distribution of the

maximum principal stress (MPa)
for the ‘FRC + CFC’ models
upon 0° loading in a and upon 45°
loading in b. The ‘CFC only’
model upon 0° loading is shown
in c and upon 45° loading in d
Clin Oral Invest
the restored teeth for the subsequent fracture strength test.
Although not intended and thus not executed in this study,
additional thermo-cycling of the restored teeth might have
accelerated the fatigue process and possibly led to fatigue
failures. Another experiment is needed to test this hypothesis.
Measuring the fracture strength is a static test used to pre-
dict the failure of restored teeth under compression [25]. As
for the chewing fatigue, no difference in fracture strength was
recorded between the FRC + CFC and CFC only restored
premolars. Some studies have shown that FRC failed to im-
prove the fracture resistance of restored teeth but decreased
the number of fractures below the cementum-enamel junction
that should be regarded as unfavourable and thus unrepairable
fractures [5, 29]. In the present study, failure was initiated at
the adhesive interface where the stress concentrated. After
cusp de-bonding, the stress pathway followed the adhesive
interface until some irregularity inside the cavity was reached,
where stress concentrated; further on, a pathway preferentially
in enamel or dentin was followed, by which stress was spread
until the restored tooth completely failed. The bond failure
remains a major concern with resin composite restoration
[25]. The number of root fractures and cusp de-bonding in-
creased substantially when the restored teeth were loaded fol-
lowing a 45° angle (Fig. 3a, b). For the 45° loading angle
models, a high tensile stress peak of 485 MPa was detected
at the DEJ of the lingual area of the FRC + CFC model
(Fig. 5b). A corresponding high tensile stress peak of
539 MPa was revealed at the DEJ of the lingual area of the
CFC only model (Fig. 5d). Dissipation of stress towards the
gingival wall initiated the failure at the adhesive interface. At
this failure origin, stress was concentrated and triggered the
fracture to propagate along the adhesive interface. The frac-
ture then continued in an oblique direction, due the anisotropy
of dentin, and eventually resulted in an unrepairable fracture.
This failure pattern is suggestive for a relationship between
cusp de-bonding and root fracture (Fig. 3a, b) and is supported
by the maximum principal tensile stress concentration at the
lingual DEJ area found in the 45° loading angle FE models
(Fig. 5b, d).
There is no consensus in the literature regarding the angu-
lation of cyclic fatigue and static fracture strength loading
being most relevant for clinical chewing; the loading angles
of 0°, 30° and 45° have been applied in different studies [25,
27, 30]. In our study, the data significantly differed following
the loading angle, with a significantly higher fracture resis-
tance for the 0° loading. The latter is understandable, consid-
ering that the 0° loading-induced stress is distributed along the
long axis of the tooth and so absorbed by the root dentin,
periodontal ligament and alveolar bone. The 45° loading angle
however caused the teeth to bend with stress being concen-
trated more at the cervical area. Tooth structure is more resis-
tant to axial compression than bending; this thus explains the
higher fracture resistance for the 0° loading angle. Teeth are
most vulnerable to fracture when eccentric forces are applied,
while the failure point is also reached at a lower load for a
larger loading angle [31, 32]. These findings were supported
by FEA that revealed significantly higher von Mises stress
(1104 and 1185 MPa, respectively) at the buccal DEJ area
for both the FRC + CFC and CFC only models upon 45°
loading (Fig. 4b, d), this as opposed to a lower stress concen-
tration (900 and 959 MPa, respectively) at the buccal DEJ area
for the FRC + CFC and CFC only models upon 0° loading
angle (Fig. 4a, c).
In most cases, the bond failed before the composite frac-
tured, once again proving that the adhesive interface remains
the weakest part of the restoration. In the case of this study, the
adhesive selected was a one-step self-etch adhesive G-aenial
Bond (GC) for the same manufacture of composite. The kind
of adhesive used and its adhesive potential might have influ-
enced the study outcome. To test this hypothesis, further re-
search is warranted, for instance using a ‘gold standard’ three-
step etch-and-rinse adhesive.
Conclusion
The fatigue resistance and fracture strength of extensively
restored premolars with a missing cusp were not increased
using the fibre-reinforced composite. The loading angle de-
creased significantly the fracture resistance of the restored
teeth. FEA revealed that the zones with the highest von Mises
stress were located at the buccal DEJ area. Analysis of the
maximum principal stress disclosed tensile stress at the DEJ
lingual area for the 45° loading angle models. These stress
concentrations correspond well with the failure mode ob-
served in the experiments.
Acknowledgments The CAPES ‘Science without borders’ Program
from the Brazilian Government awarded a Post-Doc scholarship to Bruno
de Castro Ferreira Barreto. GC Europe is gratefully acknowledged for
providing the materials needed for this research.
Compliance with ethical standards
Funding This study was funded in part by the CAPES ‘Science without
borders’ Program from the Brazilian Government, who awarded a Post-
Doc scholarship to Bruno de Castro Ferreira Barreto.
Conflict of interest The authors declare that they have no competing
interests.
Ethical approval This article does not contain any studies with human
participants or animals performed by any of the authors. The human
premolars employed in the study were gathered with informed consent
as approved by the Commission for Medical Ethics of KU Leuven under
the file number S57622.

References
1. van Dijken JW, Sunnegardh-Gronberg K (2006) Fiber-reinforced
packable resin composites in class II cavities. J Dent 34:763–769
2. Garoushi S, Mangoush E, Vallittu M, Lassila L (2013) Short fiber
reinforced composite: a new alternative for direct onlay restora-
tions. Open Dent J 7:181–185
3. Krejci I, Duc O, Dietschi D, de Campos E (2003) Marginal adap-
tation, retention and fracture resistance of adhesive composite res-
torations on devital teeth with and without posts. Oper Dent 28:
127–135
4. Ferracane JL (2011) Resin composite—state of the art. Dent Mater
27:29–38
5. Fennis WM, Tezvergil A, Kuijs RH, Lassila LV, Kreulen CM,
Creugers NH, Vallittu PK (2005) In vitro fracture resistance of fiber
reinforced cusp-replacing composite restorations. Dent Mater 21:
565–572
6. Garoushi S, Sailynoja E, Vallittu PK, Lassila LV (2013) Physical
properties and depth of cure of a new short fiber reinforced com-
posite. Dent Mater 29:835–841
7. Lutz F, Phillips RW (1983) A classification and evaluation of com-
posite resin systems. J Prosthet Dent 50:480–488
8. Bayne SC, Thompson JY (1996) Mechanical property analysis of
two admixed PRIMM-modified commercial dental composites.
Trans Acad Dent Mater 9:238
9. Bicalho AA, Valdivia AD, Barreto BC, Tantbirojn D, Versluis A,
Soares CJ (2014) Incremental filling technique and composite
material-part II: shrinkage and shrinkage stresses. Oper Dent 39:
E83–E92
10. Versluis A, Tantbirojn D, Pintado MR, DeLong R, Douglas WH
(2004) Residual shrinkage stress distributions in molars after com-
posite restoration. Dent Mater 20:554–564
11. Frater M, Forster A, Kereszturi M, Braunitzer G, Nagy K (2014)
In vitro fracture resistance of molar teeth restored with a short fibre-
reinforced composite material. J Dent 42:1143–1150
12. Garoushi S, Vallittu PK, Lassila LV (2007) Short glass fiber rein-
forced restorative composite resin with semi-inter penetrating poly-
mer network matrix. Dent Mater 23:1356–1362
13. Garoushi S, Vallittu PK, Lassila LV (2007) Direct restoration of
severely damaged incisors using short fiber-reinforced composite
resin. J Dent 35:731–736
14. Garoushi S, Vallittu PK, Watts DC, Lassila LV (2008)
Polymerization shrinkage of experimental short glass fiber-
reinforced composite with semi-inter penetrating polymer network
matrix. Dent Mater 24:211–215
15. Vallittu PK (2014) High-aspect ratio fillers: fiber-reinforced com-
posites and their anisotropic properties. Dent Mater 31:1–7
16. Soares PV, de Almeida MG, Pereira FA, Reis BR, Soares CJ, de
Sousa MM, de Freitas Santos-Filho PC (2013) Rapid prototyping
and 3D-virtual models for operative dentistry education in Brazil. J
Dent Educ 77:358–363
Clin Oral Invest
17. Wright KW, Yettram AL (1979) Reactive force distributions for
teeth when loaded singly and when used as fixed partial denture
abutments. J Prosthet Dent 42:411–416
18. Rees JS, Jacobsen PH, Hickman J (1994) The elastic modulus of
dentine determined by static and dynamic methods. Clin Mater 17:
11–15
19. Farah JW, Clark AE, Ainpour PR (1981) Elastomeric impression
materials. Oper Dent 6:15–19
20. Stafford CM, Harrison C, Beers KL, Karim A, Amis EJ,
VanLandingham MR, Kim HC, Volksen W, Miller RD, Simonyi
EE (2004) A buckling-based metrology for measuring the elastic
moduli of polymeric thin films. Nat Mater 3:545–550
21. Lastumaki TM, Lassila LV, Vallittu PK (2003) The semi-
interpenetrating polymer network matrix of fiber-reinforced com-
posite and its effect on the surface adhesive properties. J Mater Sci
Mater Med 14:803–809
22. Garoushi S, Vallittu PK, Watts DC, Lassila LV (2008) Effect of
nanofiller fractions and temperature on polymerization shrinkage
on glass fiber reinforced filling material. Dent Mater 24:606–610
23. Garoushi S, Kaleem M, Shinya A, Vallittu PK, Satterthwaite JD,
Watts DC, Lassila LV (2012) Creep of experimental short fiber-
reinforced composite resin. Dent Mater J 31:737–741
24. Garoushi S, Tanner J, Vallittu P, Lassila L (2012) Preliminary clin-
ical evaluation of short fiber-reinforced composite resin in posterior
teeth: 12-months report. Open Dent J 6:41–45
25. Taha NA, Palamara JE, Messer HH (2014) Fracture strength and
fracture patterns of root-filled teeth restored with direct resin com-
posite restorations under static and fatigue loading. Oper Dent 39:
181–188
26. Torbjorner A, Fransson B (2004) A literature review on the pros-
thetic treatment of structurally compromised teeth. Int J
Prosthodont 17:369–376
27. ElAyouti A, Serry MI, Geis-Gerstorfer J, Lost C (2011) Influence of
cusp coverage on the fracture resistance of premolars with end-
odontic access cavities. Int Endod J 44:543–549
28. Stappert CF, Att W, Gerds T, Strub JR (2006) Fracture resistance of
different partial-coverage ceramic molar restorations: an in vitro
investigation. J Am Dent Assoc 137:514–522
29. Sengun A, Cobankara FK, Orucoglu H (2008) Effect of a new
restoration technique on fracture resistance of endodontically treat-
ed teeth. Dent Traumatol 24:214–219
30. Soares PV, Santos-Filho PC, Martins LR, Soares CJ (2008)
Influence of restorative technique on the biomechanical behavior
of endodontically treated maxillary premolars. Part I: fracture resis-
tance and fracture mode. J Prosthet Dent 99:30–37
31. Christian GW, Button GL, Moon PC, England MC, Douglas HB
(1981) Post core restoration in endodontically treated posterior
teeth. J Endod 7:182–185
32. Plasmans PJ, Visseren LG, Vrijhoef MM, Kayser AF (1986)
In vitro comparison of dowel and core techniques for endodonti-
cally treated molars. J Endod 12:382–387