The effect of fiber post presence and

restorative technique on the biomechanical

behavior of endodontically treated
maxillary incisors: An in vitro study
Andréa Dolores Correia Miranda Valdivia, DDS, MS,a
Luís Henrique Araújo Raposo, DDS, MS,b Paulo Cézar
Simamoto-Júnior, DDS, MS, PhD,c Veridiana Rezende Novais,
DDS, MS, PhD,d and Carlos José Soares, DDS, MS, PhDe
School of Dentistry, Federal University of Uberlândia, Uberlândia,
MG, Brazil
Statement of problem. Unresolved controversy exists concerning the best technique for restoring endodontically
treated teeth. Prefabricated posts have been recommended with some debate as a method of restoring pulpless teeth
before placement of the final restoration.

Purpose. The purpose of this in vitro study was to evaluate the effect on the strain, fracture resistance, and fracture
mode of endodontically treated human incisors of 4 types of restorations with or without glass fiber posts after ther-
mal cycling.

Material and methods. Ninety human maxillary central incisors were selected and divided into 1 control group and 8
treatment groups (n=10). The teeth were endodontically treated and received 2 large Class III preparations simulating
coronal destruction. The teeth were then restored with or without glass fiber posts and by using 4 restorative tech-
niques: direct composite resin, direct composite veneer, feldspathic ceramic veneer, or a feldspathic ceramic crown.
After artificial aging by thermal cycling, the incisors were loaded at a 135-degree angle, and the strain was measured
by using strain gauges placed on the facial and proximal root surfaces. Specimens were subsequently loaded to the
point of fracture. Strain and fracture resistance results were analyzed by a 1-way and 2-way ANOVA, followed by the
Tukey HSD and Dunnett tests (α=.05).

Results. One-way analyses showed that the control group and the direct composite resin restoration with and without
post groups presented significantly higher fracture resistance values (P<.001). Two-way analyses showed no signifi-
cant differences for the post system. Proximal strain values were higher in the control and composite resin groups and
lower in the feldspathic ceramic groups.

Conclusions. The presence of glass fiber posts did not increase the fracture resistance of endodontically treated
incisors. Conservative composite resin restorations showed higher fracture resistance values. (J Prosthet Dent
2012;108:147-157)

Clinical Implications
Tooth structure removal directly influenced strain values and fracture resistance
within the remaining tooth structure of endodontically treated incisors. The use
of glass fiber posts did not increase the resistance of these teeth to fracture. Teeth
restored with direct composite resin veneers without posts showed more root frac-
tures than the teeth restored with the same technique but with glass fiber posts.

Supported by grant no. APQ-02400/09 from the Research Support Foundation of the State of Minas Gerais (FAPEMIG).

a
Graduate student, Biomechanics Group, Department of Operative Dentistry and Dental Materials.
b
Graduate student, Biomechanics Group, Department of Operative Dentistry and Dental Materials.
c
Professor, Biomechanics Group, Technical Course in Dental Prosthesis, Technical School of Health.
d
Professor, Biomechanics Group, Department of Operative Dentistry and Dental Materials.
e
Professor, Research Coordinator, Biomechanics Group, Department of Operative Dentistry and Dental Materials.
Valdivia et al
148
Endodontically treated teeth com-
monly require post and core restora-
tions for retention purposes because
of extensive structural defects result-
ing from dental caries and access
cavity preparation.1,2 The use of glass
fiber posts together with compos-
ite resin core foundation materials
for the restoration of endodontically
treated teeth is currently a widely ac-
cepted viable alternative to cast posts
and cores.3-5 The major advantage of
glass fiber posts is their similar elas-
tic modulus to dentin,6,7 which may
result in a better distribution of the
occlusal loads along the root.7-9 The
choice of an appropriate restoration
for endodontically treated maxillary
anterior teeth should be guided by the
amount of remaining tooth structure
and by functional and esthetic con-
siderations.10 The use of composite
resins has become routine in restor-
ative dentistry; however, resin-based
composite materials are prone to
degradation during their service life in
the oral environment.11 When a tooth
contains multiple discolored com-
posite resin restorations with several
restoration cycles, the replacement of
a preexisting composite resin resto-
ration first requires the complete re-
moval of composite resin with rotary
instruments, which may inadvertently
result in the removal of sound enamel
and/or dentin.12
Direct composite veneers have
been indicated as an esthetic and
conservative treatment alternative
in a variety of clinical scenarios.13,14
This technique can be used to mask
tooth discoloration and modify tooth
shape.15 However, these veneers have
limited longevity because they are
susceptible to discoloration, wear,
and marginal fractures.11,15 With the
improvement of resin cements, some
ceramics can be successfully bonded
to tooth structure13,16 to improve frac-
ture resistance15,17 and provide good
alternatives for reestablishing esthet-
ics.12,18,19 Ceramic veneers were intro-
duced and have been used over the
last several decades20 offering durable
anterior restorations with superior
The Journal of Prosthetic Dentistry
esthetics.15,21,22 However, feldspathic
ceramics, which remain widely used,
present the potential for brittle cata-
strophic fractures.22 Additionally, the
preparation for complete crown resto-
rations is undoubtedly more invasive,
with substantial removal of sound
tooth substance.15,23
The deformation, fracture resis-
tance, and failure mode of restored
teeth are the result of the interaction
among multiple mechanical proper-
ties.24-26 These parameters can be in-
vestigated both with destructive test-
ing and nondestructive measuring of
deformation.9,27 The deformation of
tooth structure has been measured
by using strain gauges in studies that
analyzed the influence of restorative
materials,27 cuspal deflection,28 end-
odontic therapy,29 load fatigue,30 and
post insertion.9,31 Also, in vitro simu-
lations are often performed because
clinical trials are costly and time-
consuming.32-34 Artificial aging of
specimens by using cyclic mechanical
loading and thermal cycling provides
the desired in vitro design for a study
simulating the physiological func-
tions of the oral environment,35 but
the regimens used vary considerably.32
Generally, the artificial aging of den-
tal materials is indicated because it
accelerates the degradation process,
which causes a significant decrease in
mechanical properties36,37 and bond
strength.38
The purpose of this in vitro study
was to evaluate the effectiveness of
using glass fiber posts and different
procedures for restoring the function
of severely compromised endodon-
tically treated human incisors. The
null hypotheses tested were 1) that
mechanical behavior (expressed as
deformation and fracture resistance)
would not be affected by the different
restorative techniques (direct com-
posite restoration, direct composite
veneer, feldspathic ceramic veneer,
and feldspathic ceramic crown) or by
the presence of glass fiber posts; and
2) the restorative procedures would
not demonstrate mechanical behavior
similar to that of the intact tooth.
Volume 108 Issue 3
MATERIAL AND METHODS
Specimen preparation
Ninety extracted human maxillary
central incisors were selected for this
study (approved by the Federal Uni-
versity of Uberlândia Ethics Commit-
tee). Teeth of similar size and shape
were selected according to root and
coronal dimensions after their height
and faciolingual and mesiodistal
widths in millimeters had been mea-
sured, allowing a maximum deviation
of 10% from the determined mean.39
All external debris was removed with a
hand scaler, and the teeth were stored
in distilled water at 37°C. The teeth
were divided into 9 groups (n=10), 1
control group of sound teeth without
restorations (IT) and 8 experimental
groups defined by the 2 factors in-
vestigated: post system and restor-
ative technique (Fig. 1). The roots
were covered with a 0.3-mm layer of
a polyether impression material (Imp-
regum Soft; 3M ESPE, St Paul, Minn)
to simulate the periodontal ligament
and embedded in a polystyrene resin
(AM 190 resin; AeroJet, Santo Amaro,
SP, Brazil) up to 2.0 mm below the
coronal level to simulate the alveo-
lar bone.40 Conservative endodontic
access was performed and the post
spaces were instrumented with a size
80 master apical file (K-file; Dentsply,
Maillefer, Ballaigues, Switzerland).
Post spaces were rinsed with 1.0% so-
dium hypochlorite (Cloro-Rio Hipo-
clorito de Sódio; Rioquimica, São
José do Rio Preto, SP, Brazil).) and
physiological saline and filled with
gutta percha (Dentsply, Petrópolis,
RJ, Brazil) and calcium hydroxide-
based endodontic sealer (Sealer 26;
Dentsply).41 Post spaces were enlarged
with an apical 1.0 mm to coronal
1.6-mm-diameter conical drill (Exacto
drill no. 3; Angelus Science and Tech-
nology, Londrina, Paraná, Brazil) to a
depth of 10.0 mm for the glass fiber
posts (Fig. 2). Standardized Class III
cavity preparations were made on the
mesial and distal surfaces of all speci-
mens (Fig. 2) except for those in the
Valdivia et al
September 2012 149

B C D

E F G

1 Representative specimens of experimental groups. A, IT, intact tooth. B, CIIICR, tooth restored with
composite resin. C, CRV, tooth restored with direct composite veneer. D, tooth restored without post. E, CV,
feldspathic ceramic veneers. F, CC, feldspathic ceramic crowns. G, P, tooth restored with glass fiber post.

A B
2 A. Endodontic preparation. B, Parameters for definition of Class III preparations: a, mesiodistal tooth dimension;
b, cervicoincisal tooth dimension; c, mesiodistal dimension of Class III preparation, corresponding to 1/4 of mesio-
distal tooth dimension; d, cervicoincisal class III preparation dimension corresponding to 1/3 of cervicoincisal tooth
dimension.

control group. The mesiodistal (a) and
cervicoincisal (b) dimensions of each
tooth were measured with digital cali-
pers (Mitutoyo Sul Americana Ltda,
São Paulo, Brazil) and then divided
into 4 and 3 equal parts, respectively.
The final horizontal size (c) of the Class
III preparations corresponded to 1/4
of the mesiodistal dimension, and the
vertical size (d) corresponded to 1/3
of the cervicoincisal dimension. ceramic veneer without glass fiber post;
Valdivia et al
Restoration fabrication
The groups were classified as fol-
lows: 1) IT, intact teeth unprepared and
untreated; 2) CIIICR, Class III prepara-
tion and endodontic treatment restored
with composite resin without glass fiber
post; 3) CRV, restored with direct com-
posite resin veneer without glass fiber
post; 4) CV, restored with feldspathic
5) CC, restored with feldspathic ce-
ramic crown without glass fiber post;
6) CIIICR-P, Class III preparation and
endodontic treatment restored with
composite resin associated with glass fi-
ber post; 7) CRV-P, restored with direct
composite resin veneer and glass fiber
post; 8) CV-P, restored with feldspathic
ceramic veneer and glass fiber post; and
9) CC-P, restored with feldspathic ce-
ramic crown and glass fiber post.
150
Prefabricated glass fiber posts (Ex-
actoTranslucido no. 3; Angelus Sci-
ence and Technology) were cleaned
and immersed in 24% hydrogen per-
oxide (H2O2, Dinâmica, SP, Brazil) for
1 minute.4,5,42 Afterwards, the 1-bottle
silane coupling agent (Silano, Angelus
Science and Technology) was actively
applied for 1 minute over the post
surface. All posts were luted with self-
adhesive resin cement (RelyX Unicem
2; 3M ESPE) by hand mixing the pre-
dispensed cement portions in a mix-
ing pad for 20 seconds until the ma-
terial reached homogeneity. After 5
minutes, the resin cement was light
polymerized on each surface for 40
seconds with an LED unit (Radii-Cal;
SDI Ltd, Bayswater, Australia).9,29,39
The enamel and dentin were etched
with 37% phosphoric acid (Cond
AC 37; FGM Produtos Odontologi-
cos Ltd, Joinville, SC, Brazil); then
an etch-and-rinse 3 step adhesive
system (Scotchbond Multi-Purpose;
3M ESPE) was applied, following the
manufacturer’s instructions. The ad-
hesive primer was applied over the
etched structures and gently dried
with an air stream for 5 seconds, fol-
lowed by application of the bond
component and polymerization for
10 seconds with an LED unit. Class III
restorations were incrementally built
with nanofilled composite resin (Filtek
Supreme, A3 shade; 3M ESPE). Before
preparation for the direct composite
veneers, each tooth had an acetate
matrix (Bio-art, São Carlos, SP, Brazil)
developed in a vacuum forming ma-
chine (Plastivac P7; Bio-art) to obtain
plastic molds. Then, the facial reduc-
tion for veneers was accomplished by
using diamond rotary cutting instru-
ments (no. 1014 and no. 3145; KG
Sorensen, Cotia, SP, Brazil), resulting
in 0.60 mm depth preparation with
preservation of the incisal edge. Ve-
neer preparations were restored with
nanofilled composite (Filtek Supreme;
3M ESPE). The last facial increment of
the composite resin was then applied
by using the acetate matrix to ensure
adequate crown anatomy.
The Journal of Prosthetic Dentistry
Preparations for feldspathic ce-
ramic veneers were made with similar
instruments, resulting in 0.70-mm
depth preparation at the cervical re-
gion and 1.0-mm at the middle and
incisal thirds.43 For the incisal over-
lap, the incisal edge was reduced by
1.0 mm, and a 1.0-mm-deep and
0.50-mm-wide chamfer was prepared
on the incisal/lingual surface. Com-
plete crown coverage preparations
featuring 1.5 mm of axial reduction
and 6 degrees of axial convergence
of the walls were also performed
with diamond rotary cutting instru-
ments (no. 1014, no. 4138, and no.
3168; KG Sorensen). Impressions of
the specimen crowns were made by
using a polyether impression mate-
rial (Impregum Soft; 3M ESPE). Af-
ter 1 hour, impressions were poured
with a refractory investment mate-
rial (BegoForm; Bego Bremer Gold-
schiagerel Wilh. Herbst GmbH &
Co. KG, Bremen, Germany). The
feldspathic ceramic (EX-3; Noritake
Dental Intl, Nagoya, Japan) was ap-
plied in layers over the dies and fired
at 920°C. Upon cooling, the ceramic
restorations were divested with a
diamond rotary instrument (no. 703
PM; KG Sorensen) after airborne-
particle abrasion with aluminum ox-
ide particles. The internal surfaces
of the restorations were etched with
10% hydrofluoric acid for 2 minutes
(Cond AC Porcelana; FGM Produtos
Odontologicos Ltd).44 Both ceramic
veneers and crowns were luted with a
self-adhesive resin cement, following
the manufacturer’s instructions for
manipulation as previously described
(RelyX Unicem 2; 3M ESPE). Thermal
variations were induced in a thermal
cycling machine (Thermal Cycler ER-
26000; ERIOS equipamentos, São
Paulo, SP, Brazil). All specimens were
subjected to 6000 cycles, simulat-
ing 5 years of clinical service.33,36-38,45
Thermal cycling was performed in
bath temperatures of 5°C, 37°C, and
55°C. The dwell time was 30 seconds
and the transfer time was 10 seconds,
resulting in 2 minutes for each cycle.
Volume 108 Issue 3
Strain measurement tests
To measure tooth deformation,
2 strain gauges (PA-06-038AA-120-
LEN; Excel Sensores, Embú, SP, Bra-
zil) were attached to 5 specimens per
group. One gauge was placed on the
facial surface parallel to the long axis,
and the other was placed on the prox-
imal surface in a transverse direction;
both were 1.0 mm below the cervical
limit of the root.9 The strain gauges
were bonded with a cyanoacrylate ad-
hesive (Super Bonder; Loctite, Itape-
ví, SP, Brazil) and connected to a data
acquisition device (ADS0500IP; Lynx
Tecnologia Eletrônica, SP, Brazil). A
control specimen was mounted adja-
cent to the tested tooth to compen-
sate for temperature fluctuations due
to electrical gauge resistance or local
environment.27,29,39 The specimens
with strain gauges were subjected to
a nondestructive ramp load from 0
to 100 N in a mechanical testing ma-
chine (EMIC DL2000; EMIC Ltd, São
José dos Pinhais, Paraná, Brazil).9 The
load was applied along the long axis
of the tooth on the incisal edge at a
135-degree angle by using a cylindri-
cal tip.11-13 The crosshead speed was
0.5 mm/min. The data were recorded
on a computer that performed the
signal transformation and data analy-
sis (AqDados 7.02 and AqAnalysis;
Lynx, SP, Brazil).
Fracture tests
All specimens were loaded to frac-
ture by using the same compressive
loading design as was used during the
strain gauge tests. The force required
(N) to cause fracture was recorded by
a 500-KN load cell hardwired to soft-
ware (TESC; EMIC Ltd) that detected
any sudden load drop in load cell dur-
ing the compression tests. Fractured
specimens were visually evaluated to
determine the fracture modes by us-
ing a classification system modified
from Chun et al13 and Potiket et al17:
(I) cohesive fracture of tooth or res-
toration less than half of crown; (II)
cohesive fracture of tooth or resto-
Valdivia et al
 September 2012
ration more than half of crown; (III)
cohesive fracture of restoration; (IV)
root fracture in the cervical third.
In the initial analysis, the goal was
to determine the influence of the 2
factors involved in this study: post
system and restorative technique.
Therefore, the data were analyzed with
a 2-way ANOVA (4×2) and the Tukey
Honestly Significant Difference (HSD)
test. In the second analysis, the data
were analyzed with 1-way ANOVA, and
the Dunnett test was used to evaluate
the differences in strain and fracture
resistance of the intact teeth group
(control) and the 8 experimental
groups. For all tests, groups were con-
sidered statistically different at α=.05.
RESULTS
Facial strains
The mean and standard deviation
values for the strains (µS) at the facial
root surface for all groups compared
without the intact teeth group are
shown in Figure 3 and with the intact
teeth group in Figure 4. The factors
presence of post (P=.554) and re-
storative technique (P=.376) and the
interaction between them (P<.939)
were not significant for the facial
strain values as shown by the 2-way
ANOVA (Table I). No significant dif-
ference (df=8, F=1.83, P=.104) was
found among all treatment groups
151
and the control group for the facial
root surface strain values as deter-
mined by the 1-way ANOVA.
Proximal strains
The mean and standard devia-
tion values for the strains (µS) at the
proximal root surface for all groups
compared without the intact teeth
group are shown in Figure 5 and with
the intact teeth group in Figure 6. The
factors presence of post (P=.006) and
restorative technique (P<.001) and
the interaction between them (P=.02)
were significant for the proximal
strain values, as shown by the 2-way
ANOVA (Table II). The Tukey HSD test

2500

2000
Facial Strain- µS

Aa
Aa Aa
1500 Aa
Aa
Aa Aa
Aa
1000

500

0
With Post Without Post With Post Without Post With Post Without Post With Post Without Post
Composite Resin Direct Composite Veneer Ceramic Veneer Ceramic Crown

3 Mean facial strain values and distribution by statistical categories. Same uppercase letter represents no significant
difference identified by Tukey HSD test for restorative technique; same lowercase letter represents no significant differ-
ence identified by Tukey HSD test for post presence (P>.05).
2500

2000
a
Facial Strain- µS

a a
a
1500
a
a a
a
1000 a

500

0
Intact Composite Composite Direct Direct Ceramic Ceramic Ceramic Ceramic
Resin Resin Composite Composite Veneer Veneer Crown Crown
without Post with Post Veneer Veneer without Post with Post with Post without Post
without Post with Post

4 Mean facial strain values and distribution by statistical categories. Same letters represent no significant difference
identified by Dunnett test among intact tooth and treatment groups (P>.05).

Valdivia et al
152 Volume 108 Issue 3

Table I. Two-way ANOVA (4×2) for facial strain values of restored groups
Source Sum of Mean
of Variation df Squares Square F P

Restorative technique 3 1905853 635284 3.168 .376

Post presence 1 71789 71789 .358 .554
Restorative technique × post presence 3 80848 26949 .134 .939
Error 32 6417589 200550
Total 40 75157134
Corrected total 39 8476079

600

500
Proximal Strain- µS

Cb
400

300 Ca

200 Ba
Ba
200 Aa Aa Aa Aa

0
With Post Without Post With Post Without Post With Post Without Post With Post Without Post
Composite Resin Direct Composite Veneer Ceramic Veneer Ceramic Crown

5 Mean proximal strain values and distribution by statistical categories. Different uppercase letters represent signifi-
cant difference identified by Tukey HSD test for restorative technique; different lowercase letters represent significant
difference identified by Tukey HSD test for post presence (P<.05).
500
c
400
Proximal Strain- µS

300
b
b
200 b
ab
100
a a a a

0
Intact Composite Composite Direct Direct Ceramic Ceramic Ceramic Ceramic
Resin Resin Composite Composite Veneer Veneer Crown Crown
without Post with Post Veneer Veneer without Post with Post with Post without Post
without Post with Post

6 Mean proximal strain values and distribution by statistical categories. Different letters represent significant differ-
ence identified by Dunnett test among intact tooth and treatment groups (P<.05).

showed that the groups restored with
feldspathic ceramics (CV, CC, CV-P,
and CC-P) had lower strain values for
the proximal root surface than did the
direct composite resin veneer groups
(CRV and CRV-P). Nevertheless,
teeth restored with direct composite
resin veneers (CRV and CRV-P) had
The Journal of Prosthetic Dentistry
lower strain values than the Class III
composite resin groups (CIIICR and
CIIICR-P). The presence of glass fiber
posts did not significantly modify the
strain values for the groups restored
with feldspathic ceramic (CV, CC, CV-
P, and CC-P) or with direct compos-
ite veneers (CRV and CRV-P) on the
proximal root surface. However, the
use of glass fiber posts significantly re-
duced the proximal strain in teeth re-
stored with composite resin (Fig. 5).
The 1-way ANOVA showed significant
differences (df=8, F=20.2, P<.001)
among all groups with respect to
proximal strain values. The Dunnett
Valdivia et al
September 2012 153

Table II. Two-way ANOVA (4 × 2) for proximal strain values of restored groups
Source Sum of Mean
of Variation df Squares Square F P

Restorative technique 3 28609 28609 8.64 .006

Post presence 1 459722 153241 46.28 <.001
Restorative technique × post presence 3 37610 12537 3.79 .020
Error 32 105956 3311
Total 40 1632353
Corrected total 39 631896

Table III. Mean fracture resistance (SD) values; statistical categories defined by Tukey HSD test
Post System
Restorative Technique Without Glass Fiber Post With Glass Fiber Post

CIIICR (class III composite resin) 838.8 (197.1)Aa 894.1 (397.4)Aa

CRV (direct composite veneers) 559.4 (129.0)Ba 612.1 (154.6)Ba
CV (ceramic veneers) 562.3 (138.6) Ba
484.3 (97.2)Ba
CC (ceramic crowns) 470.9 (121.9)Ba 432.1 (89.1)Ba

The same lower case letter in horizontal lines indicate no significant difference. Different uppercase letters in
vertical columns indicate significant differences (P<.05).

Table IV. Mean fracture resistance values (N) and standard deviations (SD) of
control and experimental groups (n=10), and P values calculated by Dunnett test
Groups Mean SD P

IT (Intact teeth – Control group) 844.8 186.5 -

CIIICR-P (Class III composite resin with glass fiber post) 894.1 397.4 .604
CIIICR (Class III composite resin) 838.8 197.1 .354
CRV-P (Direct composite veneer with glass fiber post) 612.1 154.6 .103
CV (Ceramic veneer) 562.3 138.6 .027
CRV (Direct composite veneer) 559.4 129.0 .010
CV-P (Ceramic veneer with glass fiber post) 484.3 97.2 <.001
CC (Ceramic crown) 470.9 121.7 <.001
CC-P (Ceramic crown with glass fiber post) 432.1 89.1 <.001

*Data were analyzed by 1-way ANOVA and Dunnett test. P value is significant at P<.05

test showed that the groups restored
with feldspathic ceramic (CV, CC, CV-
P, and CC-P), irrespective of glass fi-
ber post presence, presented lower
strain values than the control group
(IT). The teeth restored with direct
composite resin veneers (CRV and
CRV-P), irrespective of post presence,
and the teeth restored with class III
composite resin and glass fiber posts
(CIIICR-P) presented proximal strain
values statistically similar to those of
Valdivia et al
the control group. Teeth restored with
class III composite resin without glass
fiber posts (CIIICR) showed higher
strain values than the control group
(Fig. 6).
Fracture resistance and mode
Mean and standard deviation fail-
ure forces in each tested group with-
out and with the intact teeth group
are shown in Table III and IV, respec-
tively. The factor of restorative tech-
nique (P<.001) was significant for
fracture resistance, as shown by the
2-way ANOVA (Table V). However,
the factor of post system (P=.959)
and the interaction between these 2
factors (P=.605) were not significant.
The Tukey HSD test indicated that
teeth restored with class III compos-
ite resin (CIIICR) presented a higher
resistance to fracture than teeth with
other restorative techniques (Table
154 Volume 108 Issue 3

Table V. Two-way ANOVA (4×2) for fracture resistance of restored groups

Source Sum of Mean
of Variation df Squares Square F P

Restorative technique 3 1979031 659677 18.3 <.001

Post presence 1 97 97 0.003 .959
Restorative technique × post presence 3 67027 22342 0.619 .605
Error 72 2598760 36094
Total 80 34096560
Corrected total 79 4644915

Table VI. Fracture mode distribution among groups

Fracture Mode
Groups I II III IV

ST (Intact teeth) I - - -
CIIICR (Class III composite resin) 10 2 - 2
CIIICR-P (Class III composite resin with glass fiber post) 6 1 - 1
CRV (Direct composite veneer) 8 1 2 6
CRV-P (Direct composite veneer with glass fiber post) 1 - 2 3
CV (Ceramic veneer) 5 1 8 1
CV-P (Ceramic veneer with glass fiber post) - - 9 -
CC (Ceramic crown) 1 - 9 1
CC-P (Ceramic crown with glass fiber post) - - 10 -
-

IV). The 1-way ANOVA indicated
that there were significant differenc-
es (df=8, F=8.9, P<.001) among all
groups with respect to fracture resis-
tance. The Dunnett test showed the
teeth restored with direct composite
veneer (CRV, P=.010), the teeth re-
stored with ceramic veneer with glass
fiber post (CV-P, P<.001), and finally
the teeth restored with ceramic crown
irrespective of post presence (CC and
CC-P, P<.001) presented significantly
lower resistance to fracture than the
intact teeth group. Fracture mode dis-
tribution is described in Table VI. Co-
hesive fracture of the tooth involving
less than half of the crown (mode I)
was dominant for intact teeth (100%)
and for teeth restored with compos-
ite resin and glass fiber posts (80%).
Root fracture in the cervical third
(mode IV) was dominant (60%) for
teeth restored with direct compos-
ite veneers without posts. Cohesive
The Journal of Prosthetic Dentistry
fracture of the restoration (mode III)
was dominant (90%) for all groups re-
stored with feldspathic ceramics.
DISCUSSION
The first null hypothesis that the
mechanical behavior of anterior end-
odontically treated teeth would not
be affected by the different restorative
techniques or by the presence of glass
fiber posts was rejected. The second
null hypothesis that the restorative
procedures would not recover me-
chanical behavior similar to that of
the intact tooth was also rejected. The
biomechanical behavior of endodonti-
cally treated maxillary central incisors
was affected by restorative techniques
in varying degrees compared to the in-
tact tooth. The interactions between
various combinations of these factors
were assessed by measuring deforma-
tions (strains) and fracture strengths
and by observing failure modes.9,24,29,39
The oral cavity experiences ther-
mal cycling during food intake and
functional load, which may result in
slow incremental structural degrada-
tion, often referred to as fatigue.30
Thermal stresses increase marginal
leakage at the composite resin/tooth
interface and decrease the fracture
resistance of restored teeth.13 In this
study, the specimens were subjected
to 6000 thermal cycles, simulating
5 years of clinical service, before the
tests.33,36-38,45 Standardization of the
specimens is essential for obtaining
results that can be compared because
deformation and fracture resistance
depend on tooth geometry.29 The
similarity observed in this study was
in agreement with other studies that
allowed a maximal deviation of 10%
from the determined mean.24,26,39
However, even if acceptable, this vari-
ability in shape and morphology may
Valdivia et al
September 2012
explain the greater variability of the
results, especially those related to
strain. Clinical evidence indicates that
failures related to adhesively bonded
veneers are more frequent with ex-
isting composite restorations.12,14,15
In this study, 2 standardized large
proximal Class III preparations that
removed approximately 25% of the
coronal dental structure volume were
made to simulate the clinical situation
of severely compromised endodonti-
cally treated maxillary anterior teeth.
In this clinical situation, clinicians are
often in doubt whether to use glass
fiber posts or indirect restorations to
improve biomechanical behavior. This
dilemma raises an important ques-
tion as to which is the best restorative
procedure for endodontically treated
incisors with severely compromised
structure.
The current results show that the
strain gauges attached to the facial
and proximal surfaces could detect
differences among the various restor-
ative procedures.29 It has been dem-
onstrated that when a single-rooted
tooth is subjected to a load applied to
the long axis of the tooth, the greatest
compressive and tensile stresses oc-
cur at the lingual or facial root surface
of the coronal third of the root.11-13,31
The results obtained in this study for
facial strain values showed no differ-
ences among the groups, probably
because of the point of loading ap-
plication, which resulted in compres-
sive stresses to the facial root surface.
Measuring the deformation preceding
fracture may contribute to a better
understanding of the entire fracture
process, from initiation to ultimate
rupture.24 The strain gauges were at-
tached near sites where fractures were
expected to start.29 The proximal
strain gauges were oriented perpen-
dicular to the root’s long axis because
this strain component can initiate
cracks that lead to catastrophic frac-
tures of the structure.9 Teeth restored
with ceramic crown and indirect ve-
neer had lower proximal strain values
than the sound teeth. Ceramics have
high elastic moduli and low deforma-
Valdivia et al
tion capability; therefore, they tend to
concentrate stress inside the body of
the restoration.26 Because stress can-
not be relieved by deformation, mate-
rial fracture occurs before stresses can
be transferred to the tooth.
The direct composite resin veneer
groups (CRV and CRV-P) presented
similar strain values as the control
group in the proximal region, prob-
ably because of the similarity of the
mechanical properties between com-
posite resin and the dentin.27 Teeth re-
stored with Class III composite resin
without posts (CIIICR) had higher
proximal strain values than the con-
trol group. This may be because the
amount of dentin loss was not com-
pletely replaced by direct composite
resin restorations. In the biomechani-
cal analysis of tooth structures and
restorative materials, destructive me-
chanical tests used to determine frac-
ture resistance are important means
of analyzing tooth behavior in situa-
tions of concentrated and high inten-
sity load application. These tests usu-
ally produce failure loads that exceed
the average masticatory forces (the
anterior region varies between 22 N
and 222 N).11-13 According to the re-
sults of this study, the control group
and Class III composite resin groups
(CIIICR and CIIICR-P) presented a
similar fracture resistance. These find-
ings confirm the clinical results of a
study by Signore et al10 which demon-
strated that the survival of endodonti-
cally treated teeth was higher for teeth
retaining 4 and 3 coronal walls after
5.3 years.
The results of the present study
also indicated that the presence of
posts was not significant to the frac-
ture strength,14 confirming that the
primary purpose of a post is to re-
tain a core or restoration and not to
reinforce a root canal.1-3,10 Although
the incisors were structurally compro-
mised by the endodontic treatment
and the 2 large Class III preparations,
they still retained dentin sufficient to
maintain a similar stress-strain com-
plex to that of intact teeth. Class III
composite resin restorations pre-
155
sented a higher resistance than other
restorative techniques (Table III). This
behavior can be explained by the fact
that composite resin restorations
bonded to dentin and enamel showed
a mechanical behavior (fracture resis-
tance and stiffness) much closer to
that of an intact sound tooth6 and
that maximal preservation of healthy
tooth structure is important for lon-
gevity of the tooth-restoration com-
plex.2,6,27 The direct and indirect ve-
neer restoration groups presented no
significant differences in fracture re-
sistance.11 It is important to consider
the clinical significance of this study.
Conservation of tooth structure 21
and the selection of direct restorative
materials with biomechanical behav-
ior similar to sound teeth would ap-
pear to be a fast, relatively low cost
alternative25 when compared with
indirect restorations. Nevertheless,
ceramic restorations are considered
the most esthetic and biocompatible
option in dentistry with the ability to
match sound enamel.15 Traditional
recommendations for extending ve-
neer preparations indicate veneers to
finishing lines in enamel.13 Clinical evi-
dence indicates that failures related to
adhesively bonded ceramic laminates
are more frequent on existing com-
posite resin restorations,15 and these
failures are mainly due to the differ-
ence in thermal expansion coefficient
between the ceramic and composite
resin.12 Additionally, dental restora-
tions do not fail only by fracture; they
also fail by marginal leakage, wear,
and marginal degradation. These fac-
tors were not analyzed in the present
study. However, ceramic clearly be-
haves better than composite resin in
these regards.15,21,22 The analysis of
fracture mode revealed that the con-
trol group had predominant (100%)
cohesive fracture of the tooth involv-
ing less than half the crown (mode I).
Teeth restored with composite resin
and glass fiber posts presented simi-
lar fracture distribution (80%). This
result can be explained by the use of
a nanofilled composite resin (Filtek
Supreme; 3M ESPE) associated with
156
the glass fiber posts, resulting in a
unique bonded complex providing fa-
vorable stress distribution.2,7,39 Teeth
restored with direct composite resin
veneers without posts (CRV) showed
more root fractures in the cervical
third (60%) than the same restorative
technique associated with glass fiber
posts (CRV-P). This information is
important for clinicians as it clarifies
that a glass fiber post may be indicat-
ed when a large amount of the tooth
structure is removed. D’Arcangelo
et al25 also showed that a fiber post
restoration can be suggested when
endodontic treatment is associated
with veneer restorations. All teeth in
this study restored with feldspathic
ceramics had a 90% prevalence of
cohesive fractures of the restoration.
Important factors in this behavior
were the high friability 18,22,26,44 and
concentration of stresses within the
material, especially when compared
with composite resin restorations.
This study presents some intrin-
sic limitations, including the lack
of mechanical cycling and the use
of only feldspathic ceramics, which
have lower fracture resistance than
the reinforced ceramics available
to produce veneers and ceramic
crowns.13,17,18,22,26,29,30 Clinical trials
are encouraged to validate the results
of these investigations and to evalu-
ate stress distribution with strain
gauge tests and fracture resistance by
using finite element analysis. How-
ever, it can be concluded that conser-
vation of healthy tooth structure and
the selection of a restorative strategy
compatible with patient expectations
and finances are crucial to ensuring
adequate strain and fracture resis-
tance and an improved prognosis for
endodontically treated, severely com-
promised anterior teeth.
CONCLUSIONS
Within the limitations of this in
vitro study, the following conclusions
were drawn:
1. The presence of glass fiber posts
did not increase the fracture resis-
The Journal of Prosthetic Dentistry
tance of endodontically treated max-
illary anterior teeth (P=.959). The use
of glass fiber posts for restoring end-
odontically treated incisors reduced
root fractures and the proximal de-
formation of teeth restored only with
Class III composite resin (P<.001).
2. Endodontically treated incisors
restored with composite resin, with
or without glass fiber posts, showed
higher fracture resistance than those
restored with direct composite resin
veneers, ceramic veneers, and ce-
ramic complete crown restorations
(P<.001).
3. Ceramic restorations (both ve-
neer and crown) presented fractures
involving only restorative material
(90%).
4. Tooth structure removal had a
direct influence on strain values and
fracture resistance within the remain-
ing tooth structure of endodontically
treated incisors.
REFERENCES
1. Heydecke G, Peters MC. The restoration
of endodontically treated, single-rooted
teeth with cast or direct posts and cores:
a systematic review. J Prosthet Dent
2002;87:380-6
2. Bateman G, Ricketts DN, Saunders WP.
Fibre-based post systems: a review. Br Dent
J 200312;195:43-8.
3. Schwartz RS, Robbins JW. Post place-
ment and restoration of endodontically
treated teeth: a literature review. J Endod
2004;30:289-301.
4. Monticelli F, Toledano M, Tay FR, Sadek
F T, Goracci C, Ferrari M. A simple etching
technique for improving the retention of
fiber posts to resin composites. J Endod
2006;32:44-7.
5. Vano M, Goracci C, Monticelli F, Tognini
F, Gabriele M, Tay FR, et al. The adhe-
sion between fibre posts and composite
resin cores: the evaluation of microtensile
bond strength following various surface
chemical treatments to posts. Int Endod J
2006;39:31-9.
6. Dietschi D, Duc O, Krejci I, Sadan A. Biome-
chanical considerations for the restoration
of endodontically treated teeth: a systemat-
ic review of the literature--Part 1. Composi-
tion and micro- and macrostructure altera-
tions. Quintessence Int 2007;38:733-43.
7. Pegoretti A, Fambri L, Zappini G, Bianchetti
M. Finite element analysis of a glass fibre
reinforced composite endodontic post.
Biomaterials 2002;23:2667-82.
8. Newman MP, Yaman P, Dennison J, Rafter
M, Billy E. Fracture resistance of endodon-
tically treated teeth restored with compos-
ite posts. J Prosthet Dent 2003;89:360-7.
Volume 108 Issue 3
9. Santos-Filho PC, Castro CG, Silva GR,
Campos RE, Soares CJ. Effects of post sys-
tem and length on the strain and fracture
resistance of root filled bovine teeth. Int
Endod J 2008;41:493-501.
10.Signore A, Benedicenti S, Kaitsas V,
Barone M, Angiero F, Ravera G. Long-
term survival of endodontically treated,
maxillary anterior teeth restored with either
tapered or parallel-sided glass-fiber posts
and full-ceramic crown coverage. J Dent
2009;37:115-21.
11.Gresnigt MM, Ozcan M. Fracture strength
of direct versus indirect laminates with and
without fiber application at the cementation
interface. Dent Mater 2007;23:927-33.
12.Ozcan M, Mese A. Fracture strength of
indirect resin composite laminates to teeth
with existing restorations: an evaluation
of conditioning protocols. J Adhes Dent
2009;11:391-7.
13.Chun YH, Raffelt C, Pfeiffer H, Bizhang M,
Saul G, Blunck U, et al. Restoring strength
of incisors with veneers and full ceramic
crowns. J Adhes Dent 2010;12:45-54.
14.Baratieri LN, De Andrada MA, Arcari GM,
Ritter AV. Influence of post placement in
the fracture resistance of endodontically
treated incisors veneered with direct com-
posite. J Prosthet Dent 2000;84:180-4.
15.Peumans M, Van Meerbeek B, Lambrechts
P, Vanherle G. Porcelain veneers: a review of
the literature. J Dent 2000;28:163-77.
16.Borges GA, Sophr AM, de Goes MF,
Sobrinho LC, Chan DC. Effect of etch-
ing and airborne particle abrasion on the
microstructure of different dental ceramics.
J Prosthet Dent 2003;89:479-88.
17.Potiket N, Chiche G, Finger IM. In vitro
fracture strength of teeth restored with dif-
ferent all-ceramic crown systems. J Prosthet
Dent 2004;92:491-5.
18.Strub JR, Beschnidt SM. Fracture strength
of 5 different all-ceramic crown systems. Int
J Prosthodont 1998;11:602-9.
19.Barnes D, Gingell JC, George D, Adachi
E, Jefferies S, Sundar VV. Clinical evalua-
tion of an all-ceramic restorative system:
a 36-month clinical evaluation. Am J Dent
2010;23:87-92.
20.Horn HR. Porcelain laminate veneers
bonded to etched enamel. Dent Clin North
Am 1983;27:671-84.
21.Mahn DH, Polack MA. Esthetic rehabilita-
tion of maxillary incisors in conjunction with
flapless surgical techniques, an implant zir-
conia crown, and porcelain veneers. J Esthet
Restor Dent 2009;21:294-302.
22.Castelnuovo J, Tjan AH, Phillips K, Nicholls
JI, Kois JC. Fracture load and mode of fail-
ure of ceramic veneers with different prepa-
rations. J Prosthet Dent 2000;83:171-80.
23.Edelhoff D, Sorensen JA. Tooth structure
removal associated with various prepara-
tion designs for anterior teeth. J Prosthet
Dent 2002;87:503-9.
24.Soares PV, Santos-Filho PC, Martins LR,
Soares CJ. Influence of restorative tech-
nique on the biomechanical behavior of
endodontically treated maxillary premolars.
Part I: fracture resistance and fracture
mode. J Prosthet Dent 2008;99:30-7.
Valdivia et al
September 2012
25.D’Arcangelo C, De Angelis F, Vadini M,
D’Amario M, Caputi S. Fracture resistance
and deflection of pulpless anterior teeth re-
stored with composite or porcelain veneers.
J Endod 2010;36:153-6.
26.Soares CJ, Martins LR, Fonseca RB, Correr-
Sobrinho L, FernandesNeto AJ. Influence
of cavity preparation design on fracture
resistance of posterior Leucite-reinforced
ceramic restorations. J Prosthet Dent
2006;95:421-9.
27.Soares PV, Santos-Filho PC, Gomide
HA, Araujo CA, Martins LR, Soares CJ.
Influence of restorative technique on the
biomechanical behavior of endodontically
treated maxillary premolars. Part II: strain
measurement and stress distribution. J
Prosthet Dent 2008;99:114-22.
28.Taha NA, Palamara JE, Messer HH. Cuspal
deflection, strain and microleakage of
endodontically treated premolar teeth re-
stored with direct resin composites. J Dent
2009;37:724-30.
29.da Silva NR, Raposo LH, Versluis A,
Fernandes-Neto AJ, Soares CJ. The effect
of post, core, crown type, and ferrule pres-
ence on the biomechanical behavior of end-
odontically treated bovine anterior teeth. J
Prosthet Dent 2010;104:306-17.
30.Ma PS, Nicholls JI, Junge T, Phillips KM.
Load fatigue of teeth with different ferrule
lengths, restored with fiber posts, compos-
ite resin cores, and all-ceramic crowns. J
Prosthet Dent 2009;102:229-34.
31.Yoldas O, Akova T, Uysal H. An experimental
analysis of stresses in simulated flared root
canals subjected to various post-core appli-
cations. J Oral Rehabil 2005;32:427-32.
32.Gale MS, Darvell BW. Thermal cycling
procedures for laboratory testing of dental
restorations. J Dent 1999;27:89-99.
33.Stappert CF, Ozden U, Gerds T, Strub
JR. Longevity and failure load of ceramic
veneers with different preparation designs
after exposure to masticatory simulation. J
Prosthet Dent 2005;94:132-9.
34.Heydecke G, Butz F, Hussein A, Strub JR.
Fracture strength after dynamic loading of
endodontically treated teeth restored with
different post-and-core systems. J Prosthet
Dent 2002;87:438-45.
35.Geminiani A, Lee H, Feng C, Ercoli C. The
influence of incisal veneering porcelain
thickness of two metal ceramic crown
systems on failure resistance after cyclic
loading. J Prosthet Dent 2010;103:275-82.
36.Hahnel S, Henrich A, Bürgers R, Handel G,
Rosentritt M. Investigation of mechanical
properties of modern dental composites
after artificial aging for one year. Oper Dent
2010;35:412-9.
37.Naumann M, Sterzenbach G, Rosentritt
M, Beuer F, Frankenberger R. In vitro
performance of self-adhesive resin cements
for post-and-core build-ups: influence of
chewing simulation or 1-year storage in
0.5% chloramine solution. Acta Biomater
2010;6:4389-95.
38.Fischer J, Zbären C, Stawarczyk B, Häm-
merle CH. The effect of thermal cycling
on metal-ceramic bond strength. J Dent
2009;37:549-53.
39.Santana FR, Castro CG, Simamoto-Júnior
PC, Soares PV, Quagliatto PS, Estrela C,
Soares CJ. Influence of post system and
remaining coronal tooth tissue on bio-
mechanical behaviour of root filled molar
teeth. Int Endod J 2011;44:386-94.
40.Soares CJ, Pizi EC, Fonseca RB, Martins
LR. Influence of root embedment mate-
rial and periodontal ligament simulation
on fracture resistance tests. Braz Oral Res
2005;19:11-6.
157
41.Soares CJ, Santana FR, Silva NR, Preira JC,
Pereira CA. Influence of the endodontic
treatment on mechanical properties of root
dentin. J Endod 2007;33:603-6.
42.de Sousa Menezes M, Queiroz EC, Soares
PV, Faria-e-Silva AL, Soares CJ, Martins LR.
Fiber post etching with hydrogen peroxide:
effect of concentration and application
time. J Endod 2011;37:398-402.
43.Turkaslan S, Tezvergil-Mutluay A, Bagis
B, Shinya A, Vallittu PK, Lassila LV. Effect
of intermediate fiber layer on the fracture
load and failure mode of maxillary incisors
restored with laminate veneers. Dent Mater
J 2008;27:61-8.
44.Soares CJ, Soares PV, Pereira JC, Fon-
seca RB. Surface treatment protocols in
the cementation process of ceramic and
laboratory-processed composite restora-
tions: a literature review. J Esthet Restor
Dent 2005;17:224-35.
45.Naumann M, Preuss A, Rosentritt M. Ef-
fect of incomplete crown ferrules on load
capacity of endodontically treated maxillary
incisors restored with fiber posts, compos-
ite build-ups, and all-ceramic crowns: an in
vitro evaluation after chewing simulation.
Acta Odontol Scand 2006;64:31-6.
Corresponding author:
Dr Carlos José Soares
Faculdade de Odontologia, Área de Dentística
e Materiais Odontológicos
Universidade Federal de Uberlândia
Av. Pará, n° 1720
CEP 38400-902, Bloco B, 2B24
Uberlândia, Minas Gerais
BRAZIL
Fax: +55-34-3218-2279
E-mail: carlosjsoares@umuarama.ufu.br
Copyright © 2012 by the Editorial Council for
The Journal of Prosthetic Dentistry.

Receive Tables of Contents by E-mail

To receive tables of contents by e-mail, sign up through our Web site at http://www.journals.elsevierhealth.com/peri-
odicals/ympr.
Instructions
Log on and click “Register” in the upper right-hand corner. After completing the registration process, click on “My
Alerts,” then “Add Table of Contents Alert.” Select the category “Mosby” or type The Journal of Prosthetic Dentistry in
the search field and click on the Journal title. The title will then appear, and having already completed the Registra-
tion process, you may add tables of contents alerts by accessing an issue of the Journal and clicking on the “Add TOC
Alert” link.
You will receive an e-mail message confirming that you have been added to the mailing list.

Note that tables of contents e-mails will be sent when a new issue is posted to the Web site.

Valdivia et al