Fracture resistance of thermal cycled

and endodontically treated premolars

with adhesive restorations
Luciano de V. Habekost, DDS, MS,a Guilherme B. Camacho,
DDS, PhD,b Eduardo C. Azevedo, DDS,c and Flavio F.
Demarco, DDS, PhDd
Department of Operative Dentistry, School of Dentistry, Federal
University of Pelotas, RS, Brazil

Statement of problem. Cavity preparations have routinely been associated with decreased fracture strength of re-
stored teeth.

Purpose. The purpose of this study was to evaluate the influence of endodontic treatment and thermal cycling on the
fracture resistance of teeth restored with ceramic or composite resin inlay restorations.

Material and methods. One hundred sound, maxillary premolars were selected. Twenty intact teeth served as a con-
trol group. Eighty teeth were prepared with MOD cavity preparations; half of them were also endodontically treated.
Inlay restorations were prepared with composite resin (Filtek Z250) or feldspathic ceramic (Vitadur Alpha). Half of the
specimens for each group were submitted to thermal cycling (500 cycles, between 5 and 55°C, dwell time of 30 sec-
onds). The specimens were subjected to compressive axial loading using a steel ball. Data were analyzed using 3-way
ANOVA and post hoc Tukey’s test (α=.05).

Results. Except for those teeth restored with ceramic without endodontic treatment and thermal cycling, no other
group achieved fracture strength similar to sound teeth. Ceramic and composite resin restorations provided similar
resistance to fracture. There was a significant interaction between endodontic therapy and thermal cycling (P<.001).
In the nonthermal cycled groups, endodontically treated teeth showed significantly lower fracture resistance (P<.001).
All experimental groups had similar fracture strength when submitted to thermal cycling.

Conclusions. Both restorative techniques provided similar fracture resistance. Endodontic treatment decreased the
fracture resistance of nonthermal cycled specimens, while thermal cycling decreased the fracture resistance of non-
endodontically treated specimens. (J Prosthet Dent 2007;98:186-192)

Clinical Implications
Premolars with adhesive intracoronal restorations do not have
the same fracture resistance following endodontic treatment
as intact teeth.

Cavity preparations have been rou-
tinely associated with decreased frac-
ture strength of restored teeth.1,2,3,4 In
general, for both molar and premo-
lar teeth, the higher the number of
restored surfaces and the wider the
isthmus of the restoration, or both,
the greater the chance of cuspal frac-
ture with time.5
Mesio-occlusodistal (MOD) in-
tracoronal preparations commonly
result in the creation of elongated
cusps that are further subjected to
stress and potential cuspal fracture.6
Measurements of cuspal flexure with

a
Postgraduate student, Operative Dentistry.
b
Associate Professor.
c
Postgraduate student, Operative Dentistry.
d
Associate Professor.
The Journal of Prosthetic Dentistry Habekost et al
September 2007
strain gauges showed that a bonded
restoration has the potential for re-
covering original nonrestored tooth
stiffness.7 Therefore, a restorative ma-
terial should not only replace the lost
tooth structure but also increase the
fracture resistance and promote ef-
fective marginal sealing.7,8
Adhesive restorations have been in-
dicated to obtain an “internal splint”,
preserve tooth structure, and provide
an esthetic and functional restora-
tion.9 Composite resin gained popu-
larity among clinicians due to its ease
of handling, excellent esthetic and
mechanical properties, and reported
ability to reinforce weakened dental
structure.9 Although hybrid compos-
ite resins are the first choice for res-
toration of small- and medium-sized
occlusal cavities, direct composite
resin restorations are highly technique
sensitive, presenting disadvantages
primarily related to polymerization
shrinkage, postoperative sensitivity,
and wear resistance.10
When a cavity preparation exceeds
the recommended limits for the direct
application of composite resins, indi-
rect total- or partial-coverage restora-
tions have been indicated.10 Ceramic
materials are a viable alternative be-
cause of superior esthetics, biocom-
patibility, resistance to wear, and a
similar coefficient of thermal expan-
sion as that observed for dental struc-
ture.10,11
Endodontic therapy is important
to preserve teeth that would other-
wise be lost. Nevertheless, endodonti-
cally treated teeth are generally weaker
because of the loss of tooth structure
caused by caries and endodontic ac-
cess and instrumentation of the root
canal.4 If a normal occlusion, without
parafunctional habits, is present, ad-
hesive inlay restorations may be an
alternative restorative treatment for
nonvital teeth not requiring posts or
cuspal coverage.12,13
Conflicting outcomes regarding
the strengthening effect of bonded
restorations on weakened teeth have
been reported. Some studies reported
no significant differences in fracture
Habekost et al
strength between intact and directly
or indirectly restored teeth,2,9,14,15,16
with contrasting findings reported in
other studies.3,11,17,18
The clinical survival of restored
teeth is dependent on the restorative
material, technique, remaining tooth
structure, and the interactions be-
tween material, teeth, and the oral
environment.19 Thermal cycling is an
in vitro aging process that consists of
subjecting a tooth-restoration com-
plex to temperature extremes, in an at-
tempt to simulate the thermal stresses
that normally occur intraorally.20,21
Although several studies have in-
vestigated the influence of thermal
cycling on restorations, little is known
about the combined effect of end-
odontic treatment and thermal cy-
cling on fracture strength of indirectly
restored teeth. The purpose of this
study was to evaluate the influence of
endodontic treatment and thermal cy-
cling on the fracture strength of teeth
restored with ceramic inlays or direct
composite resin restorations. The
null hypothesis was that there was no
significant difference among experi-
mental groups restored with ceramics
or composite resins, endodontically
treated or not treated, and subjected
to thermal cycling or not.
MATERIAL AND METHODS
The research protocol had the ap-
proval of the Ethical Research Com-
mittee of the School of Dentistry,
Federal University of Pelotas, Brazil.
One hundred sound human premo-
lars were selected. The teeth were
extracted to meet orthodontic treat-
ment requirements. After soft tissue
removal, the teeth were stored in 1%
chloramine solution (Laboratory of
Biochemistry, Federal University of
Pelotas, Pelotas, RS, Brazil) for 72
hours.22
To be included in the study, the
premolars were required to have the
following crown dimensions: 9.0- to
9.6-mm bucco-lingual distance, 7.0-
to 7.4-mm mesio-distal distance,
and 7.7- to 8.8-mm cervico-occlusal
187
distance.18 Also, the teeth were re-
quired to be free of cracks when ob-
served under magnification (x10) in
a stereomicroscope (Model SQF-F;
Biosystems, Curitiba, PR, Brazil) and
fiber-optic transillumination (FOTI)
(Eurotek International Ltd, Warsaw,
Poland). The teeth were washed in
tap water and stored in distilled water
at 37°C, which was changed every 5
days during the study.
The teeth had their roots embed-
ded in a cylindrical polyvinyl chloride
(PVC) matrix (1.5 x 2.0 cm; Tigre SA,
Sao Paulo, SP, Brazil), using acrylic
resin (Artigos Odontologicos Classi-
co Ltda, Sao Paulo, SP, Brazil), up to
1 mm apical to the cemento-enamel
junction (CEJ). An impression of each
tooth was made with a heavy-bodied
vinyl polysiloxane material (Express;
3M ESPE, St. Paul, Minn). The im-
pression was sectioned for use as an
anatomical guide during tooth reduc-
tion. Eighty teeth were randomly se-
lected using the Research Randomizer
(www.randomizer.org/form.htm) and
submitted to MOD cavity prepara-
tion. Forty teeth were endodontically
treated. Twenty premolars remained
intact as the control group.
For the endodontically treated
teeth, the pulp chamber was accessed
with a diamond rotary cutting instru-
ment (#1014; KG Sorensen, Barueri,
SP, Brazil). Root canals were instru-
mented to an ISO size 40 (Dentsply,
Petropolis, RJ, Brazil). After intermit-
tent rinsing with 1% sodium hypo-
chlorite solution (Biodinamica, Ibi-
pora, PR, Brazil), the root canals were
obturated with laterally condensed
gutta-percha points (Dentsply) and
sealer (Endofill; Dentsply). The supe-
rior aspect of the gutta-percha ma-
terial extended 1 mm apical to the
CEJ, forming an apical location for
the composite resin (Filtek Z250; 3M
ESPE) used to fill the access prepara-
tion.
For cavity preparations, diamond
rotary cutting instruments (#3131;
KG Sorensen) were used in a high-
speed handpiece (KaVo, Joinville, SC,
Brazil), under copious air-water cool-
188 Volume 98 Issue 3
ing. Burs were replaced after 4 prepa-
rations, in order to ensure high cutting
efficacy. The occlusal preparation was
2 mm deep, with a width of one half
the intercuspal distance. The proxi-
mal boxes were prepared at a width
one half the bucco-lingual distance,
1.5 mm deep, axially, and the cervical
wall was 1 mm coronal to the CEJ (Fig.
1).2 In order to finish the preparations
and to standardize the convergence
angle of the cavity walls, the same
diamond rotary cutting instrument
was mounted in a milling machine (S.
Colombano, Milano, Italy) and used
at low speed. Cavity dimensions were
measured with the vinyl polysiloxane
index and with a digital caliper (Mitu-
toyo, Suzano, SP, Brazil). Half of the 1 MOD ceramic inlay preparation sizes (IC, intercuspal
specimens were restored with feld- distance; BL, buccolingual distance).
spathic ceramic (Vitadur Alpha; Vita
Zahnfabrik, Bad Sackingen, Germany)
and half with hybrid composite resin Table I. Experimental groups (n=10)
(Filtek Z250; 3M ESPE). All materials
Restorative Endodontic Thermal
were used following manufacturers’
directions. Code Material Treatment Cycling
The ceramic restorations were
VT Vitadur Alpha No Yes
submitted to 4 burnout processes
(600 - 930°C, model EDGCON 5P; V Vitadur Alpha No No
EDG, Sao Carlos, SP, Brazil), finished
VET Vitadur Alpha Yes Yes
with silicone points (Dentsply), and
glazed. The internal surfaces were VE Vitadur Alpha Yes No
airborne-particle abraded with 50-
µm glass particles (Danville Materi- FT Filtek Z250 No Yes
als, San Ramon, Calif ), etched with F Filtek Z250 No No
10% hydrofluoric acid (Dentsply) for
4 minutes, washed for 20 seconds, FET Filtek Z250 Yes Yes
and coated with silane coupling agent FE Filtek Z250 Yes No
(Dentsply).
Tooth surfaces received 37% phos- CT No restoration (control) ––– Yes
phoric acid etching (Dentsply) for 15
C No restoration (control) ––– No
seconds. After washing and lightly
drying, the dentin was kept moist,
then the dentin-bonding agent (Sin-
gle Bond; 3M ESPE) was applied in 2 cement (RelyX ARC; 3M ESPE), ac- Excess cement was removed with a
coats to the etched surfaces and light cording to manufacturer’s instruc- scalpel. Mesial and distal interfaces
polymerized for 10 seconds, using a tions. Equal amounts of base and were light-activated for 40 seconds
light-polymerization unit with energy catalyst were dispensed and mixed. (XL3000; 3M ESPE).
higher than 450 mW/cm2 (XL3000; The ceramic restoration with the resin The direct composite resin resto-
3M ESPE). When light polymerization cement was placed into the cavity and rations (Filtek Z250; 3M ESPE) were
was required, the tip of the light unit held in place under pressure, a 1-kg placed in 2-mm increments. After the
was placed as close as possible to the load, using a Vicat needle (Viatest, application of Single Bond as previ-
tooth surface without contacting it. Belo Horizonte, MG, Brazil), which ously described, metal matrix strips
The ceramic restorations were ce- had the point rounded to avoid dam- (Tofflemire; Jon, Sao Paulo, SP, Bra-
mented using dual-polymerized resin age to the restoration, for 2 minutes. zil) were placed around the cavity to
The Journal of Prosthetic Dentistry Habekost et al
September 2007 189

2 Representation of sphere application used for axial compression test.

Table II. Three-way ANOVA evaluating effect of different variables investi-

gated and their interactions (control groups were excluded from analysis)

Source of Variation df MS F P

Endodontic treatment 1 10023.23 14.92 <.001

Material 1 6798.02 10.12 .003

Thermal cycling 1 11178.53 16.64 <.001

Endodontic x material 1 36.16 0.059 .818

Endodontic x thermal 1 6394.33 9.52 .003

Material x thermal 1 2100.88 3.13 .083

Endodontic x material x thermal 1 105.88 0.16 .693

Residual 48 671.81

Total 55 1252.43

reestablish the proximal surfaces of
the restorations. Light activation of
each increment was performed from
the occlusal aspect, for 20 seconds
(XL3000; 3M ESPE). For both restor-
ative materials, finishing and polish-
ing procedures were accomplished
with aluminum oxide discs (Sof-Lex;
3M ESPE), and the specimens were
stored in distilled water at 37°C for 7
days. Half of the specimens were sub-
mitted to 500 thermal cycles between
Habekost et al
5°C and 55°C using a thermal cycling
machine (Etica Odontologica, Sao
Paulo, SP, Brazil), with a dwell time
of 30 seconds. The 10 experimental
groups (n=10) are listed in Table I.
The specimens were subjected to
an axial compression test in a uni-
versal testing machine (MEM 2000;
EMIC Ltda, Sao Jose dos Pinhais,
PR, Brazil), using a 9-mm-diameter
steel sphere (Fig. 2). The sphere was
applied in the center of the occlusal
surfaces, with a 500-kgf load, at a
crosshead speed of 0.5 mm/min until
specimen fracture.
The fracture pattern was evalu-
ated based on a standard ranking
developed by Habekost et al18: (1)
fracture restricted to the restoration;
(2) fracture of the tooth structure,
but not through the long axis of the
tooth; (3) fracture of the tooth and
the restoration, but not through the
long axis of the tooth; and (4) fracture
190 Volume 98 Issue 3
Table III. Mean values (standard deviations) (kgf ) of fracture resistance for different experimental groups

Vitadur Alpha Filtek Z250

Without With Without With
Control Endodontics Endodontics Endodontics Endodontics

Not thermal 161 159 109 123 76

cycled (26) (17) (25) (19) (26)

A, a A, ab A, cd A, bc A, d

Thermal cycled 153 94 93 89 79

(33) (40) (28) (28) (12)

A, a B, b A, b B, b A, b

Different uppercase letters indicate significant differences in columns (P<.05)

Different lowercase letters indicate significant differences in row (P<.05)

100%

80%

60%
Percent

40%

20%

VT V VET VE FT F FET FE CT C

Type 1 Type 2 Type 3 Type 4

3 Frequency rate of fracture type observed for each group.

through the long axis of the tooth, in
the tooth/restoration or only at the
tooth. Data from the fracture resis-
tance test were submitted to a 3-way
analysis of variance (ANOVA) and
post hoc Tukey Honestly Different
test. Since the values from the control
group were inflating the results and
impairing the comparison between
experimental groups, they were com-
pared with the other groups using a
1-way ANOVA. The evaluation of the
fracture patterns was performed us-
The Journal of Prosthetic Dentistry
ing a nonparametric Kruskal-Wallis
test, with an alpha value equal to .05
for all tests.
RESULTS
For the 3-way ANOVA, the control
group was excluded from the analy-
sis (Table II). Three variables and
their interactions were investigated:
endodontic treatment, restorative
material, and thermal cycling. The 3
variables demonstrated significant
differences, as well as the interac-
tion between the factors, endodontic
treatment and thermal cycling.
Table III presents the mean val-
ues and standard deviations for each
group. Except for those teeth restored
with ceramic, without endodontic
treatment and not submitted to ther-
mal cycling, no other groups demon-
strated the fracture resistance shown
for the control group (P<.01). When
the variable, material, was investigat-
ed, the only significant difference ob-
Habekost et al
September 2007
served was for those teeth not submit-
ted to thermal cycling. Teeth without
endodontic treatment and restored
with ceramic restorations exhibited
higher fracture resistance than those
teeth with endodontic treatment and
restored with direct composite resin
(P<.05).
A significant interaction between
endodontic treatment and ther-
mal cycling conditions was detected
(P<.01). For the specimens submit-
ted to thermal cycling, no difference
between teeth with and without end-
odontic treatment was observed. Oth-
erwise, thermal cycling significantly
decreased the fracture resistance of
specimens without endodontic treat-
ment (P<.001). Endodontically treat-
ed teeth were not affected by thermal
cycling. For the specimens not ther-
mal cycled, the teeth with endodontic
therapy showed lower fracture resis-
tance (P<.001). Figure 3 exhibits the
percentage of failure patterns for all
groups. Only the data from restored
teeth were submitted for statistical
analysis, and there were no significant
differences among the fracture pat-
terns for any group.
DISCUSSION
The results from this study do not
support the null hypotheses, since the
3 experimental factors exhibited sig-
nificant differences. The primary rea-
sons for failure of restorations are the
presence of secondary caries, tooth
or restoration fracture, marginal dis-
crepancies, and wear. While sound
teeth rarely fracture due to stresses
from mastication, fracture may occur
in teeth weakened by caries and cavity
preparations.1
The present study was conducted
with maxillary premolars because
these teeth have a similar fracture
potential compared to molar teeth.5
Several investigations regarding tooth
fracture resistance have used premo-
lars, which facilitates the comparison
of results.8,11,15,16,18 Since tooth size is
variable, and standard preparation
could weaken smaller specimens, cav-
Habekost et al
ities with approximately 50% of buc-
colingual intercuspal distance width
were prepared, as recommended by
Bremer and Geurtsen.9
In general, the findings of this
study showed that adhesively bonded
restorations were not capable of fully
restoring the fracture strength of un-
prepared teeth. The ceramic group
without endodontic treatment and
without thermal cycling was the only
group that exhibited fracture strength
comparable to healthy teeth. How-
ever, when these ceramic restorations
were submitted to thermal cycling, a
significant decrease in fracture resis-
tance was observed. Other studies
have detected higher fracture strength
for intact teeth when compared to
those with adhesively bonded MOD
restorations.3,11,17,18
Generally, the restorative mate-
rial did not influence the resistance to
fracture of the restored teeth. Other
authors have not observed differ-
ences between composite resin and
ceramic MOD restorations, which
is in agreement with the findings of
this study.2,3,17 Evaluating the fatigue
life of Class II MOD restorations in
premolars restored with composite
resin or ceramic restorations, no sig-
nificant difference in fatigue cycles
to failure between the 2 materials
was observed.16 Similar resistance
to fracture was observed for premo-
lars with endodontic treatment when
restored with composite resin or ce-
ramic systems.3 Since conservative
preparations (inlays) were used in
this study, the amount of remaining
tooth structure could have a signifi-
cant role in the resistance to fracture.
In fact, higher resistance to fracture
was observed for teeth with ceramic
inlay restorations than for those with
partial or total ceramic onlays.18 Per-
haps if a larger cavity preparation was
made, the restorative material may
have been more influential. In rela-
tion to the ceramic type, it has been
reported that ceramic type could be
important to the fracture strength
of MOD restorations,9 while other
findings demonstrated similar results
191
among different ceramic systems.18
The thermal cycling process signif-
icantly affected the strength of teeth
without endodontic treatment. The
artificial aging induced by thermal cy-
cling can occur in 2 ways. Hot water
may accelerate hydrolysis of interfacial
composite resin components, and the
higher thermal contraction/expan-
sion coefficient of the restorative ma-
terial could generate stresses at the
tooth-material interface,21 thereby
weakening the adhesive bonding and
decreasing the fracture resistance.20
Nevertheless, thermal cycling only
significantly influenced the groups
without endodontic therapy. For end-
odontically treated teeth, there is a
need for additional tooth structure
removal, including important ana-
tomical aspects such as the roof of
the pulpal chamber.3 This causes an
additionally significant decrease in
the resistance to fracture,3,4 and end-
odontically treated teeth are more
likely to fracture than posterior teeth
with vital pulps.15 Therefore, end-
odontically treated teeth could have
a lower fracture resistance when com-
pared to those not treated, and, as
a consequence, thermal cycling did
not produce an additional reduction
in these already lowered values. In 1
study, Gonzalez-Lopez et al6 showed
that teeth with MO preparations and
endodontic access resulted in a dou-
bling of cuspal flexure with occlusal
loading, when compared to teeth
with only MO preparations.
Another possible reason for the
thermal cycling influence only in those
groups without endodontic therapy is
the larger volume of restorative mate-
rial in teeth with endodontic access.
Therefore, if a large volume of material
were present, it could reduce the ther-
mal stress at the adhesive interface.
The specimens should be immersed
in different temperatures (thermal
cycling) for longer times to effectively
cause thermal stress at such an inter-
face.20
Although teeth that crack or frac-
ture intraorally usually do so as a re-
sult of repeated episodes of stress that
192
fatigue the crystalline structure over a
longer period of time,7 in this study,
a continually increasing load was ap-
plied until tooth fracture. These tests
are largely used in other studies re-
ported in the literature, and they are
an important source of information,
indicating which component is most
fragile: restoration, tooth structure,
or adhesive interface.2 Clinical ex-
trapolation of the experimental find-
ing should be carefully made, since
no experimental method truly mim-
ics the conditions under which teeth
function in an oral environment. The
high standard deviations observed
are typical of mechanical testing on
irregular anatomic shapes, and this
fact is accounted for in the statistical
analysis.5
Dietschi et al11 reported that the
extent of fracture increased propor-
tionally with the fracture resistance of
the specimens. According to the out-
comes from the present study, the type
1 mode of failure was predominantly
observed for all tested groups. None-
theless, when an alteration of this
predominant pattern was observed,
teeth restored with ceramic presented
a more harmful mode of failure (pat-
tern 3) than did the composite resin-
filled teeth (pattern 2), as shown in
Figure 3. This is probably related to
the higher elastic modulus of the ce-
ramic when compared to composite
resin, which results in greater stress
concentrations within the bulk of the
restoration, which is liberated to sur-
rounding structures when fracture
occurs. The fracture of restorations
may be influenced by other factors,
and, despite the importance of short
laboratory studies to answer some
questions, the actual performance of
restorations can only be determined
by long-term clinical trials.
The Journal of Prosthetic Dentistry
CONCLUSIONS
Within the limitations of this in
vitro study, the following conclusions
were drawn:
1. In general, adhesively bonded
restorations did not provide resistance
to fracture similar to that achieved for
nonrestored teeth.
2. Generally, ceramic and compos-
ite resin restorations exhibited similar
resistance to fracture.
3. Thermal cycling significantly re-
duced the fracture resistance of teeth
not endodontically treated.
REFERENCES
1. Mondelli J, Steagall L, Ishikiriama A, de
Lima Navarro MF, Soares FB. Fracture
strength of human teeth with cavity prepa-
rations. J Prosthet Dent 1980;43:419-22.
2. Dalpino PH, Francischone CE, Ishikiriama
A, Franco EB. Fracture resistance of teeth
directly and indirectly restored with com-
posite resin and indirectly restored with
ceramic materials. Am J Dent 2002;15:389-
94.
3. Santos MJ, Bezerra RB. Fracture resistance
of maxillary premolars restored with direct
and indirect adhesive techniques. J Can
Dent Assoc 2005;71:585.
4. Sagsen B, Aslan B. Effect of bonded resto-
rations on the fracture resistance of root
filled teeth. Int Endod J 2006;39:900-4.
5. Cavel WT, Kelsey WP, Blankenau RJ. An
in vivo study of cuspal fracture. J Prosthet
Dent 1985;53:38-42.
6. Gonzalez-Lopez S, De Haro-Gasquet F,
Vilchez-Diaz MA, Ceballos L, Bravo M. Ef-
fect of restorative procedures and occlusal
loading on cuspal deflection. Oper Dent
2006;31:33-8.
7. Lopes LM, Leitao JG, Douglas WH. Effect
of a new resin inlay/onlay restorative mate-
rial on cuspal reinforcement. Quintessence
Int 1991;22:641-5.
8. de Freitas CR, Miranda MI, de Andrade MF,
Flores VH, Vaz LG, Guimaraes C. Resis-
tance to maxillary premolar fractures after
restoration of class II preparations with
resin composite or ceromer. Quintessence
Int 2002;33:589-94.
9. Bremer BD, Geurtsen W. Molar fracture
resistance after adhesive restoration with
ceramic inlays or resin-based composites.
Am J Dent 2001;14:216-20.
10.Ritter AV, Baratieri LN. Ceramic restora-
tions for posterior teeth: guidelines for the
clinician. J Esthet Dent 1999;11:72-86.
11.Dietschi D, Maeder M, Meyer JM, Holz J.
In vitro resistance to fracture of porcelain
inlays bonded to tooth. Quintessence Int
Volume 98 Issue 3
1990;21:823-31.
12.Krejci I, Duc O, Dietschi D, de Campos
E. Marginal adaptation, retention and
fracture resistance of adhesive composite
restorations on devital teeth with and with-
out posts. Oper Dent 2003;28:127-35.
13.Steele A, Johnson BR. In vitro fracture
strength of endodontically treated premo-
lars. J Endod 1999;25:6-8.
14.Brunton PA, Cattell P, Burke FJ, Wilson
NH. Fracture resistance of teeth restored
with onlays of three contemporary tooth-
colored resin-bonded restorative materials.
J Prosthet Dent 1999;82:167-71.
15.Hannig C, Westphal C, Becker K, Attin
T. Fracture resistance of endodontically
treated maxillary premolars restored with
CAD/CAM ceramic inlays. J Prosthet Dent
2005;94:342-9.
16.Shor A, Nicholls JI, Phillips KM, Libman
WJ. Fatigue load of teeth restored with
bonded direct composite and indirect
ceramic inlays in MOD class II cavity prepa-
rations. Int J Prosthodont 2003;16:64-9.
17.St-Georges AJ, Sturdevant JR, Swift EJ Jr,
Thompson JY. Fracture resistance of pre-
pared teeth restored with bonded inlay res-
torations. J Prosthet Dent 2003;89:551-7.
18.Habekost Lde V, Camacho GB, Pinto MB,
Demarco FF. Fracture resistance of premo-
lars restored with partial ceramic restora-
tions and submitted to two different load-
ing stresses. Oper Dent 2006;31:204-11.
19.da Rosa Rodolpho PA, Cenci MS, Donas-
sollo TA, Loguercio AD, Demarco FF. A
clinical evaluation of posterior composite
restorations: 17-year findings. J Dent
2006;34:427-35.
20.Rossomando KJ, Wendt SL. Thermocycling
and dwell times in microleakage evalua-
tion for bonded restorations. Dent Mater
1995;11:47-51.
21.De Munck J, Van Landuyt K, Peumans M,
Poitevin A, Lambrechts P, Braem M, et al. A
critical review of the durability of adhesion
to tooth tissue: methods and results. J Dent
Res 2005;84:118-32.
22.Haller B, Hofmann N, Klaiber B, Bloching
U. Effect of storage media on microleakage
of five dentin bonding agents. Dent Mater
1993;9:191-7.
Corresponding author:
Dr Luciano de Vargas Habekost
Rua Almirante Tamandaré, 251/408
Pelotas, RS
BRAZIL
CEP: 96015-750
Fax: 55-53255581
E-mail: luvh@bol.com.br or flavio.demarco@
pesquisador.cnpq.br
Acknowledgments
The authors thank Dr John M. Powers for revi-
sion and suggestions during the preparation of
the manuscript.
Copyright © 2007 by the Editorial Council of
The Journal of Prosthetic Dentistry.
Habekost et al