Influence of cavity preparation design on fracture resistance of ...

<strong>Influence</strong> <strong>of</strong> <strong>cavity</strong> <strong>preparation</strong> <strong>design</strong> on fracture resistance <strong>of</strong> posterior
Leucite-reinforced ceramic restorations
Carlos Jose Soares, DDS, MS, PhD, a Luis Roberto Marcondes Martins, DDS, MS, PhD, b
Rodrigo Borges Fonseca, DDS, MS, PhD, c Lourenco Correr-Sobrinho, DDS, MS, PhD, d
and Alfredo Julio Fernandes Neto, DDS, MS, PhD e
School <strong>of</strong> Dentistry, Federal University <strong>of</strong> Uberlandia, Minas Gerais, Brazil; Piracicaba School <strong>of</strong>
Dentistry, State Univerty <strong>of</strong> Campinas, Sao Paulo, Brazil
Statement <strong>of</strong> problem. Controversy exists concerning the preferred <strong>cavity</strong> <strong>design</strong> for posterior ceramic restorations
to improve their resistance to fracture under occlusal load.
Purpose. The aim <strong>of</strong> this study was to assess the resistance to fracture <strong>of</strong> leucite-reinforced ceramic restorations
placed on molars with different <strong>cavity</strong> <strong>preparation</strong> <strong>design</strong>s.
Material and methods. Ninety noncarious molars were selected, stored in 0.2% thymol solution, and divided
into 9 groups (n=10): IT, intact teeth; CsI, conservative inlay; ExI, extensive inlay; CsO/mb, conservative onlay
with mesio-buccal cusp coverage; ExO/mb, entensive onlay with mesio-buccal cusp coverage; CsO/b, conservative
onlay with buccal cusp coverage; ExO/b, entensive onlay with buccal cusp coverage; CsO/t, conservative
onlay with total cusp coverage; ExO/t, extensive onlay with total cusp coverage. Teeth were restored with a
Leucite-reinforced ceramic (Cergogold). The fracture resistance (N) was assessed under compressive load in a
universal testing machine. The data were analyzed with 1-way and 2-way analyses <strong>of</strong> variance, followed by
the Tukey HSD test (a=.05). Fracture modes were recorded, based on the degree <strong>of</strong> tooth structure and
restoration damage.
Results. One-way analysis showed that intact teeth had the highest fracture resistance values. Two-way analyses
showed no significant differences for the isthmus extention factor, but showed a significant difference for the
<strong>preparation</strong> <strong>design</strong> type <strong>of</strong> fracture (P=.03), and also for the interaction between both factors (P=.013). The
fracture mode observed in all groups tended to involve only restorations.
Conclusion. Within the limitations <strong>of</strong> this study, it was observed that cuspal coverage does not increase fracture
resistance <strong>of</strong> the posterior tooth-restoration complex restored with leucite-reinforced ceramics. (J Prosthet
Dent 2006;95:421-9.)
CLINICAL IMPLICATIONS
In this in vitro study, the <strong>cavity</strong> <strong>preparation</strong> involving cuspal reduction for restoration with
Leucite-reinforced ceramic did not improve the resistance to fracture. The fracture modes
observed were predominantly restricted to the restorative material.
The presence <strong>of</strong> extensive carious lesions, unsatisfactory
restorations, and tooth fracture results in controversy
regarding the optimal restorative procedure.
Choosing between use <strong>of</strong> a direct or indirect technique
Presented at the IADR Meeting, Hawaii, March 10-13, 2004.
Supported by Fundacao de Amparo a Pesquisa do Estado de Minas
Gerais (FAPEMIG), Grant 1987-03.
a
Pr<strong>of</strong>essor, Operative Dentistry and Dental Materials, Dental School
<strong>of</strong> Federal University <strong>of</strong> Uberlandia.
b
Pr<strong>of</strong>essor, Operative Dentistry, Dental School <strong>of</strong> Piracicaba, State
University <strong>of</strong> Campinas.
c
Pr<strong>of</strong>essor, Operative Dentistry and Dental Materials, Dental School
<strong>of</strong> Federal University <strong>of</strong> Uberlandia.
d
Pr<strong>of</strong>essor, Dental Materials, Dental School <strong>of</strong> Piracicaba, State University
<strong>of</strong> Campinas.
e
Pr<strong>of</strong>essor, Prosthodontic and Dental Materials, Dental School <strong>of</strong>
Federal University <strong>of</strong> Uberlandia.
when placing a posterior restoration is difficult and involves
esthetic, biomechanical, anatomical, and financial
considerations. When an indirect restoration is determined
to be the best treatment option, the clinician must then
determine the geometric configuration <strong>of</strong> the <strong>cavity</strong>
<strong>preparation</strong>. 1-3 Several <strong>design</strong>s have been proposed for
preparing posterior resin-bonded all-ceramic restorations,
4-7 due to the particular mechanical and structural
characteristics presented by ceramic restorative
materials. 8-10
For cast metal restorations, nonfunctional and functional
cusps tend to be reduced during <strong>cavity</strong> <strong>preparation</strong>,
producing better stress distribution. 11 Since
metal restorations are generally cemented with a nonadhesive
cement, such as zinc-phosphate cement, complete
cusp coverage is indicated to increase dental
structure resistance. 12 Tooth <strong>preparation</strong> <strong>design</strong>s
JUNE 2006 THE JOURNAL OF PROSTHETIC DENTISTRY 421

THE JOURNAL OF PROSTHETIC DENTISTRY SOARES ET AL
advocated for posterior ceramic restorations have been
based on traditional cast metal restoration <strong>design</strong>s, but
with more occlusal tooth reduction and a slightly increased
taper. 3 These <strong>preparation</strong>s can involve the removal
<strong>of</strong> considerable tooth structure, 13 and as more
structure is removed, a tooth will have less resistance
to fracture. 14 However, in <strong>preparation</strong>s for posterior ceramic
restorations, some authors have demonstrated
that occlusal reduction results in a reduced chance <strong>of</strong> restoration
failure, likely increasing longevity <strong>of</strong> the restoration.
4,6,15 Fracture resistance tests have been used to
determine the forces that may induce fracture <strong>of</strong> such
restorations, and thus enable a <strong>preparation</strong> <strong>design</strong> to
be suggested for providing greatest resistance to fracture.
10,16-26
Dental ceramics are considered to be esthetic restorative
materials with desirable characteristics, such as
translucence, fluorescence, and chemical stability. 17,27
They are also biocompatible, have high compressive
strength, and their thermal expansion coefficient is similar
to that <strong>of</strong> the tooth structure. 27 In spite <strong>of</strong> their
many advantages, ceramics are fragile under tensile
strain, making them susceptible to fracture during the
luting procedure and under occlusal force. 28-31 This
dichotomy raises an important question as to which is
the best <strong>cavity</strong> <strong>preparation</strong> <strong>design</strong> for posterior teeth
restored with ceramic restorations.
Therefore, the aim <strong>of</strong> this study was to assess the
in vitro resistance to fracture <strong>of</strong> Leucite-reinforced
ceramic-restored posterior teeth, and to analyze the
modes <strong>of</strong> fracture with various <strong>preparation</strong> <strong>design</strong>s.
The null hypothesis was that different <strong>preparation</strong>
<strong>design</strong>s have no effect on the fracture resistance <strong>of</strong> teeth
restored with Leucite-reinforced ceramics.
MATERIAL AND METHODS
Fig. 1. Cavity <strong>preparation</strong> <strong>design</strong>s <strong>of</strong> different experimental groups.
Ninety freshly extracted, sound, caries-free human
mandibular molars <strong>of</strong> similar size and shape were
selected by measuring the buccolingual and mesiodistal
widths in millimeters, allowing a maximum deviation <strong>of</strong>
10% from the determined mean. Teeth were stored in
0.2% thymol solution. Calculus and s<strong>of</strong>t-tissue deposits
were removed with a hand scaler. The teeth were cleaned
using a rubber cup and fine pumice water slurry and then
stored in 0.9% saline solution at 4°C until completion <strong>of</strong>
the experiment. The roots were covered with a 0.3-mm
layer <strong>of</strong> a polyether impression material (Impregum; 3M
ESPE, St Paul, Minn) to simulate the periodontal ligament,
and embedded in a polystyrene resin (Cristal,
Piracicaba, Sao Paulo, Brazil) up to 2 mm below
the cementoenamel junction to simulate the alveolar
bone. 17,24 The teeth were divided into 9 groups
(n=10) as follows: IT, intact teeth (control group);
CsI, conservative inlay; ExI, extensive inlay; CsO/mb,
onlay with conservative isthmus covering the mesiobuccal
cusp; ExO/mb, onlay with extensive isthmus
covering the mesio-buccal cusp; CsO/b, onlay with
conservative isthmus covering all buccal cusps; ExO/
b, onlay with extensive isthmus covering all buccal
cusps; CsO/t, onlay with conservative isthmus covering
all cusps; ExO/t, onlay with extensive isthmus covering
all cusps (Fig. 1).
Using a 6-degree taper diamond rotary cutting
instrument (3131; KG Sorensen, Barueri, Sao Paulo,
Brazil), 8 different <strong>preparation</strong>s, with internal rounded
angles, were defined. A <strong>preparation</strong> machine (Federal
University <strong>of</strong> Uberlandia, Uberlandia, Minas Gerais,
Brazil) was used to standardize <strong>preparation</strong> dimensions
(Fig. 2). 17 This device consists <strong>of</strong> a high-speed handpiece
(KaVodo Brasil Ltd, Joinville, SC, Brazil) coupled to a
mobile base. The mobile base moves vertically and horizontally
with the aid <strong>of</strong> 3 micrometers (Mitutoyo,
Tokyo, Japan) with a 0.1-mm level <strong>of</strong> accuracy. The isthmus
floor <strong>of</strong> the mesio-occluso-distal (MOD) cavities
was prepared following principles for ceramic and indirect
composite resin <strong>preparation</strong>s. 32 The pulpal floor
was prepared to a depth <strong>of</strong> 2.5 mm from the occlusal
422 VOLUME 95 NUMBER 6

SOARES ET AL
Fig. 2. Cavity <strong>preparation</strong> instrument. (A) Micrometer controls
quantity <strong>of</strong> vertical movement, which is moved by connecting
rod, B; (C) metal device allows 180-degree
movement around its longitudinal axis and 360-degree movement
around its transverse axis; (D) high-speed handpiece
with diamond rotary cutting instrument; (E) specimen being
prepared; (F) micrometer.
cavosurface margin <strong>of</strong> the <strong>preparation</strong>; the occlusal
isthmus was 5 mm wide for extensive isthmus groups
and 2.5 mm wide for conservative isthmus groups.
The buccolingual widths on mesial and distal boxes
were similar to the occlusal isthmus width. Each box
had a gingival floor depth <strong>of</strong> 1.5 mm mesiodistally and
an axial wall height <strong>of</strong> 2 mm. Margins were prepared
with 90-degree cavosurface angles.
A single-stage impression was made <strong>of</strong> each prepared
tooth using a double-viscosity vinyl polysiloxane
(Panasil; Kettenbach GmbH, Eschenburg, Germany)
in a stock plastic tray (Tigre, Sao Paulo, Brazil).
After 2 hours, the impressions were poured with
Type IV stone (Velmix; Kerr Italia SpA, Scafati,
Italy). One technician fabricated all restorations using
a standardized technique and following the manufacturer’s
instructions. 27 Restorations were made with a
leucite-reinforced ceramic (Cergogold; Degussa Dental,
Hanau, Germany). A spacer (Isolit; Degussa Dental)
was applied over the high-density stone dies, and a
0.7-mm-thick wax coping (Plastodent; Degussa
Dental) was fabricated. The wax coping was invested
(Cerg<strong>of</strong>it Investment; Degussa Dental) and placed in
a burnout furnace (F1800 1P; EDG, Sao Paulo,
Brazil) to eliminate the wax. The burnout furnace
was preheated to 270°C, and the temperature was
gradually increased to approximately 850°C for 40
minutes with the alumina plunger kept inside the
furnace. Ingots (Cergogold, shade A3; Degussa Dental)
were pressed in an automatic press furnace (Cerampress
Qex; Dentsply Ceramco, York, Pa). After cooling,
the specimens were divested using 50-mm glass
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 3. Compressive loading using 6-mm-diameter steel
sphere placed in center <strong>of</strong> occlusal leucite-reinforced
ceramic molar restoration.
beads at 4-bar pressure, followed by airborne-particle
abrasion with 100-mm aluminum oxide at 2-bar pressure
to remove the refractory material. Finally, the
specimens were treated with 100-mm aluminum oxide
airborne-particle abrasion at 1-bar pressure.
The ceramic restorations were then luted (RelyX
ARC; 3M ESPE), following the manufacturer’s instructions.
Ceramic inlays were etched with 10% hydr<strong>of</strong>luoric
acid (Condicionador de Porcelanas; Dentsply, Sao
Paulo, Brazil) for 60 seconds, and then a silane agent
(Rely-X ceramic primer; 3M ESPE) was applied for
60 seconds and dried. 27 The <strong>cavity</strong> <strong>preparation</strong>s (enamel
and dentin) were etched using 37% phosphoric acid for
15 seconds, rinsed, and blotted dry with absorbent paper.
With a fully saturated brush tip, 2 consecutive coats
<strong>of</strong> an adhesive system (Adper Single Bond; 3M ESPE)
were applied to the tooth, gently dried for 5 seconds
with compressed air, and polymerized with a halogen
light-polymerization unit (XL 3000; 3M ESPE) for 20
seconds at an intensity <strong>of</strong> 800 mW/cm 2 and a sourceto-specimen
distance <strong>of</strong> 1 cm. The resin luting agent was
dispensed onto a mixing pad and mixed for 10 seconds.
A thin layer <strong>of</strong> the material was applied to the ceramic
restoration, which was seated in place. Excess luting
agent was removed with a brush. The luting agent was
polymerized (800 mw/cm 2 , XL 3000; 3M ESPE)
from the facial, lingual, and occlusal directions for 40
seconds in each direction. Finishing rotary cutting instruments
(#2135 F and #2135 FF; KG Sorensen)
were used to remove excess luting agent.
The teeth were subjected to axial compressive loading
with a metal sphere 6 mm in diameter (Fig. 3) at a crosshead
speed <strong>of</strong> 0.5 mm/min in a universal testing
machine (Model DL2000; EMIC, Sao Jose dos Pinhais,
Brazil). The force required (N) to cause fracture was
JUNE 2006 423

THE JOURNAL OF PROSTHETIC DENTISTRY SOARES ET AL
Fig. 4. Fracture modes: (I) isolated fracture <strong>of</strong> restoration; (II) restoration fractures involving small portion <strong>of</strong> tooth; (III) fracture
involving more than half <strong>of</strong> tooth, without root involvement; (IV) fracture with root involvement.
Table I. One-way ANOVA <strong>of</strong> mechanical compression test
values
Source
<strong>of</strong> variation df
Sum <strong>of</strong>
squares
Mean
square
Calculated
F P
Critical
F
Types <strong>of</strong>
<strong>preparation</strong>s
8 184925.4 23115.7 11.444 .001 2.055
Error 81 163606.9 2019.8
Total 89 348532.2
Variation coefficient = 199.76.
Table III. Two-way ANOVA (4 3 2) for mechanical
compression test <strong>of</strong> ceramic restored groups
Source <strong>of</strong> variation df
Sum <strong>of</strong>
squares
Mean
square
Calculated
F P
Type <strong>of</strong> <strong>preparation</strong> 3 49541.5 16513.8 9.454 .003
Buccolingual extent 1 899.8 899.8 0.515 .095
Interaction 3 22482.3 7494.1 4.290 .013
Treatments 7 72923.6 10417.7
Error 72 125770.7 1746.8
Total 79
Variation coefficient 196.8; significant difference P,.05.
recorded by a 5-kN load cell hardwired to s<strong>of</strong>tware
(TESC; EMIC), which was able to detect any sudden
load drop during compression.
The fractured specimens were evaluated to determine
fracture patterns using a modified classification system
based on the classification system proposed by Burke
et al 8 : (I) isolated fracture <strong>of</strong> the restoration; (II) restoration
fracture involving a small tooth portion; (III)
fracture involving more than half <strong>of</strong> the tooth, without
periodontal involvement; and (IV) fracture with periodontal
involvement (Fig. 4).
Table II. Mean fracture resistance values (SDs) and statistical
categories <strong>of</strong> all experimental groups (n=10)
Groups Failure load mean (N)
Statistical Tukey
category
IT 3143.1 (635.5) A
CsI 2465.4 (318.7) B
ExI 2278.1 (586.4) B
ExO/b 2204.5 (353.0) BC
CsO/b 2158.4 (321.7) BCD
CsO/t 2062.3 (488.4) BCD
ExO/mb 2001.5 (337.3) BCD
CsO/mb 1612.2 (349.1) CD
ExO/t 1551.4 (443.3) D
IT, Intact teeth.
Minimum significant difference = 628.4; different letters indicate significant
differences (P,.05).
Table IV. Mean fracture resistance (SD) values and
statistical categories defined by Tukey HSD test for
interaction between isthmus extension and <strong>cavity</strong>
<strong>preparation</strong> <strong>design</strong> factors (n=10)
Groups Failure load mean values General mean values
Conservative
CsI 2465.4 (318.7) A
CsO/b 2158.4 (321.7) A
CsO/t 2062.3 (488.4) AB
CsO/mb 1612.2 (349.1) B
Extensive
ExI 2278.1 (586.4) a
ExO/b 2204.5 (353.0) a
Ex/Omb 2001.5 (337.3) ab
ExO/t 1551.4 (443.3) b
2074.6 (369.5) A
2008.9 (430.0) A
Minimum significant difference = 482.59.
Different uppercase (conservative groups) or lowercase (extensive groups)
letters indicate significant differences (P,.05). General mean values are
compared by uppercase letters (P..05).
424 VOLUME 95 NUMBER 6

SOARES ET AL
Fig. 5. Mean fracture resistance values and distribution by statistical categories. Different letters represent significant differences
identified by Tukey test for <strong>preparation</strong>s characterized by conservative occlusal isthmus (P,.05).
Fig. 6. Mean fracture resistance values and distribution by statistical categories. Different letters represent significant differences
identified by Tukey test for <strong>preparation</strong>s characterized by extensive occlusal isthmus (P,.05).
In the initial analysis, the fracture resistance data <strong>of</strong>
the 9 groups were submitted to statistical analysis by
1-way analysis <strong>of</strong> variance (ANOVA) and the Tukey
Honestly Significant Difference (HSD) test. In the
second analysis, the aim was to determine the influence
<strong>of</strong> 2 factors involved in this study, the dimension <strong>of</strong> the
isthmus floor and cuspal coverage. Therefore, the data
were analyzed with a 2-way ANOVA (4 3 2) and the
Tukey HSD test. For all tests, groups were considered
statistically different at a=.05.
RESULTS
The 1-way ANOVA showed that there were significant
differences (P=.001) among all groups with respect
to resistance to fracture (Table I). The Tukey HSD test
showed that intact teeth presented a higher resistance to
fracture under occlusal load, which was significant when
compared to the other groups (P,.05). The mean and
SD <strong>of</strong> the forces applied to cause failure in each tested
group are shown in Table II. Two-way ANOVA showed
that there were significant differences (P=.013) for the
interaction between occlusal isthmus floor dimension
and cuspal coverage procedure (Table III). The Tukey
test was applied to determine the significance <strong>of</strong> the interaction
<strong>of</strong> the 2 factors (Table IV) and indicated that
the cuspal coverage did not result in higher fracture resistance
(Figs. 5 and 6). The fracture mode analysis indicated
that all groups tended to demonstrate only
fractures <strong>of</strong> restorations, rather than tooth structure
(Fig. 7).
DISCUSSION
THE JOURNAL OF PROSTHETIC DENTISTRY
Performing in vitro experiments that aim to analyze
indirect restoration failures, characterized by the
JUNE 2006 425

THE JOURNAL OF PROSTHETIC DENTISTRY SOARES ET AL
Fig. 7. Representative fracture modes observed after compressive loading test. A, Inlay. B, Onlay covering 1 cusp. C, Onlay
covering all buccal cusps. D, Onlay covering all cusps.
fracture <strong>of</strong> either the restorative material or dental structure,
is an important method for improving restorative
procedures. 1,2,10,12,16,17,28 There are a number <strong>of</strong> factors
that may interfere with resistance to fracture, such
as the tooth embedment method, type <strong>of</strong> load application
device, and crosshead speed. 8 Thus, the experimental
methods used for in vitro analyses do not faithfully
represent real clinical conditions, in which failures occur
primarily due to fatigue. 18 To minimize the discrepancy
between experimental assessments and clinical failures,
different methods have been used, such as the joint
use <strong>of</strong> mechanical tests and fracture mode analysis according
to predefined scales. 8,17,19
Mechanical fracture tests are performed to numerically
quantify the influence <strong>of</strong> restorative material
types, 2,10,16,17,19,20 luting procedures, 12,21 and <strong>preparation</strong>
characteristics 14,22 for resistance to fracture
when submitted to a concentrated and increasing load.
These tests usually produce failure loads that exceed
the load limit exerted by normal stomatognathic system
movements. 33 In spite <strong>of</strong> this fact, higher loading situations
can be compared to the situation in which the
individual grinds a solid body <strong>of</strong> small dimensions and
the force that would be distributed over the occlusal
surfaces <strong>of</strong> posterior teeth is concentrated over a single
tooth. If this tooth is structurally debilitated or prepared
with an inadequate <strong>cavity</strong> <strong>design</strong>, the result may be
fracture <strong>of</strong> the tooth, the restoration, or both.
When performing mechanical tests, some factors are
important to more closely approximate the clinical situation,
such as the root embedment method to simulate
the periodontal membrane, the loading apparatus, and
the mode <strong>of</strong> load transmission. 8 The simulation <strong>of</strong> the
periodontal ligament should be done with an elastomeric
material that is able to undergo elastic deformation
and reproduce the accommodation <strong>of</strong> the tooth
in the alveolus, providing the nonconcentration <strong>of</strong>
stresses in the cervical region <strong>of</strong> the tooth. Moreover, a
simulated periodontal ligament is highly influential on
the fracture pattern. 23 In this experiment, a polyether
impression material was used in association with a polystyrene
resin as an adequate method for fracture resistance
tests. 17,24
Occlusal loading is another important factor. Burke
et al 8 and Burke and Watts 9 concluded that the best
method for measuring the resistance <strong>of</strong> premolars to
fracture is the use <strong>of</strong> a cylinder <strong>of</strong> a defined diameter. 19
The use <strong>of</strong> a 6-mm steel sphere for resistance to fracture
testing by Dietschi et al 25 and Soares et al 17 was shown
to be ideal for molars because it contacts the functional
and nonfunctional cusps in positions close to those
found clinically.
The results <strong>of</strong> the present study showed that the mean
resistance to load <strong>of</strong> healthy teeth (IT:3143.1 6 635.5
N) was significantly higher than the resistance to load
<strong>of</strong> teeth prepared with different types <strong>of</strong> <strong>preparation</strong>s
and restored with a Leucite-reinforced ceramic; thus,
the null hypothesis was rejected. This demonstrates
that the restorative process, even when adhesive techniques
are associated with cuspal coverage, is not able
to restore the total resistance to load <strong>of</strong> healthy molars.
This result is in agreement with the studies <strong>of</strong> Morin
426 VOLUME 95 NUMBER 6

SOARES ET AL
et al, 1 St-Georges et al, 26 and particularly with that <strong>of</strong>
Mondelli et al, 14 who showed a reduction in the resistance
to fracture for teeth that had been prepared with
greater removal <strong>of</strong> dental structure. Since ceramics
have high elastic moduli and tend to concentrate stress
inside the body <strong>of</strong> the restoration, 4,18,31 they have lower
resistance to fracture than healthy teeth, even though
the ceramic is reinforced by the inclusion <strong>of</strong> oxides. 27
This is because ceramics are not capable <strong>of</strong> undergoing
elastic deformation at the same rate as tooth structure
and resinous materials. Thus, stress concentrations are
dependent on the geometry <strong>of</strong> the specimen material,
loading conditions, and presence <strong>of</strong> intrinsic or extrinsic
flaws. In addition, the resin luting agent under a ceramic
restoration may act as a s<strong>of</strong>t layer and could reduce the
effects <strong>of</strong> stress concentration. In spite <strong>of</strong> the adhesive
process being fundamental for luting a leucite-reinforced
ceramic restoration, 12,27 the cushioning effect
<strong>of</strong> the resinous cement did not seem to be sufficient
to absorb the stresses, which remain inside the ceramic
restoration and demand deformation; if deformation
does not exist, fracture may occur.
Analyses <strong>of</strong> the different types <strong>of</strong> <strong>preparation</strong>s in this
study showed that <strong>preparation</strong>s resulting in a greater loss
<strong>of</strong> tooth structure appear to decrease the resistance to
fracture <strong>of</strong> the tooth-restoration complex. Assif et al 22
analyzed the extent <strong>of</strong> <strong>preparation</strong> for amalgam restorations
and found that endodontically treated teeth with a
small amount <strong>of</strong> structure removed (conservative occlusal
isthmus) and total cuspal coverage produced better
resistance values. The discrepancies between the results
<strong>of</strong> Assif et al 22 and the findings <strong>of</strong> the present study are
likely due to the differences in the mechanical properties
and adhesive characteristics <strong>of</strong> the respective restorative
materials used. Amalgam, as a metal, can undergo elastic
deformation; however, ceramics cannot due to their
ionic and covalent bonds. Moreover, amalgam adhesion
to the tooth structure is minimal; thus, extensive amalgam
restorations demand cuspal coverage to protect
teeth from fracture when the restorative material is undergoing
deformation. Ceramics, in general, present a
high elastic modulus and low strain capacity, so the
stress tends to concentrate inside the material since
stress can not be relieved by deformation the material
fractures before stresses are transferred to the tooth.
Leucite-reinforced ceramics have good adhesion to
tooth structures, and according to St-Georges et al, 26
this enables an internal rather than an external splinting
with cuspal coverage 10 to be created in some instances.
Thus, the adhesive strength should be enough to support
cusp deflection. With respect to the fracture resistance
<strong>of</strong> all tested groups, the conservative inlay
<strong>preparation</strong> showed the highest numerical value
(2465.4 6 318.7 N), but was statistically similar to
the extensive inlay group, in which a greater amount
<strong>of</strong> tooth structure had been removed during <strong>preparation</strong>
THE JOURNAL OF PROSTHETIC DENTISTRY
(2278.1 6 586.4 N). The ceramic restoration presented
less resistance to fracture when more tooth structure was
removed during <strong>preparation</strong>, and the effect <strong>of</strong> cuspal
coverage was not a benefit.
In addition to discussing fracture resistance values, it
may be important to analyze the fracture modes in each
experimental group. For all groups, fracture was observed
in the restoration itself. Similar findings were
reported by Soares et al 17 when the authors tested feldspathic
ceramic in extensive inlays, and also by Burke, 21
who affirmed that the ceramic fractures before the
tooth. For the CsI group, it was found that the majority
<strong>of</strong> the fractures occurred exclusively in the restoration,
probably because <strong>of</strong> the larger volume <strong>of</strong> dental structure
in the cusps, which resulted in a greater capacity
for undergoing deformation rather than fracture.
Cuspal coverage with an onlay apparently does not
produce a clear benefit to the fracture resistance <strong>of</strong> the
restored tooth, as has been shown for metal restorations.
11 This is related to the fact that the failure, as
seen by the fracture modes, occurs almost exclusively
in ceramics. If the restoration is expected to fail irrespective
<strong>of</strong> the <strong>preparation</strong> <strong>design</strong>, then extending <strong>preparation</strong>s
will not change this behavior, as seen in this
study.
The clinician may encounter a situation in which the
tooth presents loss <strong>of</strong> a cusp or fracture <strong>of</strong> a portion <strong>of</strong> it.
It is not advisable to establish an occlusal contact at the
tooth-restoration interface, due to the difference in the
mechanical behavior <strong>of</strong> the 2 structures. In this situation
the need for cuspal coverage must be determined.
Another concern is the need for covering nonfunctional
cusps, which may be observed by comparing the groups
<strong>of</strong> teeth with onlay <strong>preparation</strong>s covering just the functional
cusps (CsO/b, 2158.4 6 321.7 N and ExO/b,
2204.5 6 353.0 N), and the onlay <strong>preparation</strong>s with
total cusp coverage (CsO/t, 2062.3 6 488.4 N and
ExO/t, 1551.4 6 443.3 N). According to the results
<strong>of</strong> this study, it appears that there is no advantage in
covering nonfunctional cusps.
The comparison <strong>of</strong> the behavior <strong>of</strong> groups with great
volumes <strong>of</strong> ceramic in occlusal isthmus (CsO/t and
ExO/t) requires discussion. In spite <strong>of</strong> being statistically
similar, ExO/t was almost 500 N less resistant than
CsO/t, and it may be that the volume <strong>of</strong> ceramic material
in the occlusal box and the thickness <strong>of</strong> remaining tooth
structure in prepared cusps are responsible for this finding.
Ceramic thickness, when either very thin or very
thick, seems to be detrimental with regard to fracture. 34
In addition, sharp angles and knife-edge–prepared
cusps tend to concentrate stress, resulting in greater susceptibility
to ceramic restoration fracture 32 ; therefore,
rounded internal angles are preferred for ceramic restorations.
When tooth structure loss results in an extensive
occlusal isthmus, this may necessitate coverage <strong>of</strong> all the
cusps, a feasible alternative for restoration in the occlusal
JUNE 2006 427

THE JOURNAL OF PROSTHETIC DENTISTRY SOARES ET AL
isthmus and also to eliminate the knife-edge form in
the prepared cusps. However, further investigation is
needed to determine the effect <strong>of</strong> this procedure.
This study has several limitations. The compressive
load applied to the restored tooth was increased until
failure; however, dental ceramics typically fail as a result
<strong>of</strong> many loading cycles or from an accumulation <strong>of</strong> damage
from stress and water. 35 In terms <strong>of</strong> in vivo loading,
the masticatory cycle consists <strong>of</strong> a combination <strong>of</strong> vertical
and lateral forces, subjecting the ceramic to a variety <strong>of</strong>
<strong>of</strong>f-axis loading forces. 36 Cyclic loading may be more
adequate to reproduce fatigue failures verified clinically.
Future analyses should involve both methods <strong>of</strong> load applications.
Destructive tests are important for predicting
and comparing the behavior <strong>of</strong> a restored tooth under
specific situations; however, future studies should perhaps
use nondestructive methodologies such as finite element
analysis or structural deformation by employment
<strong>of</strong> strain gauges, enabling not only analysis at the moment
<strong>of</strong> failure but also the possible causes for the failure.
In addition, because only one material type was used in
the current study, it is not possible to apply these results
to other esthetic materials. 17 The results <strong>of</strong> this study
suggest guidelines when preparing teeth for ceramic
restorations, but more analyses are required before extrapolating
these results to other ceramic restorative
systems.
CONCLUSION
Within the limitations <strong>of</strong> this in vitro study, the following
conclusions were drawn:
1. Intact teeth are more resistant to fracture than teeth
prepared and restored with Leucite-reinforced ceramics,
irrespective <strong>of</strong> the <strong>cavity</strong> <strong>preparation</strong> <strong>design</strong>.
2. Specific <strong>design</strong>s <strong>of</strong> different <strong>preparation</strong>s facilitate the
occurrence <strong>of</strong> fractures <strong>of</strong> ceramic restorations, with
fracture modes generally restricted to the restoration.
3. The cuspal coverage did not increase fracture resistance
<strong>of</strong> the posterior tooth-restoration complex
restored with Leucite-reinforced ceramics.
REFERENCES
1. Morin D, Delong R, Douglas WH. Cusp reinforcement by acid-etch technique.
J Dent Res 1984;63:1075-8.
2. Steele A, Johnson BR. In vitro fracture strength <strong>of</strong> endodontically treated
premolars. J Endod 1999;25:6-8.
3. Etemadi S, Smales RJ, Drummond PW, Goodhart JR. Assessment <strong>of</strong> tooth
<strong>preparation</strong> <strong>design</strong>s for posterior resin-bonded porcelain restorations.
J Oral Rehabil 1999;26:691-7.
4. Banks RG. Conservative posterior ceramic restorations: a literature
review. J Prosthet Dent 1990;63:619-26.
5. Broderson SP. Complete-crown and partial-coverage tooth <strong>preparation</strong>
<strong>design</strong>s for bonded cast ceramic restorations. Quintessence Int 1994;25:
535-9.
6. Garber DA, Goldstein RE. Porcelain and composite inlays and onlays.
Esthetic posterior restorations. 1st ed. Chicago: Quintessence; 1994. p. 38.
7. Milleding P, Ortengren M, Karlsson S. Ceramic inlay systems: some clinical
aspects. J Oral Rehabil 1995;22:571-80.
8. Burke FJ, Wilson NH, Watts DC. The effect <strong>of</strong> <strong>cavity</strong> wall taper on fracture
resistance <strong>of</strong> teeth restored with resin composite inlays. Oper Dent 1993;
18:230-6.
9. Burke FJ, Watts DC. Fracture resistance <strong>of</strong> teeth restored with dentinbonded
crowns. Quintessence Int 1994;25:335-40.
10. Bremer BD, Geurtsen W. Molar fracture resistance after adhesive restoration
with ceramic inlays or resin-based composites. Am J Dent 2001;14:
216-20.
11. Reeh ES, Douglas WH, Messer HH. Stiffness <strong>of</strong> endodontically-treated
teeth related to restoration technique. J Dent Res 1989;68:1540-4.
12. Eakle WS, Staninec M. Effect <strong>of</strong> bonded gold inlays on fracture resistance
<strong>of</strong> teeth. Quintessence Int 1992;23:421-5.
13. Moscovich H, Creugers NH, Jansen JA, Wolke JG. Loss <strong>of</strong> sound tooth
structure when replacing amalgam restorations by adhesive inlays. Oper
Dent 1998;23:327-31.
14. Mondelli J, Steagall L, Ishikiriama A, de Lima Navarro MF, Soares FB.
Fracture strength <strong>of</strong> human teeth with <strong>cavity</strong> <strong>preparation</strong>s. J Prosthet
Dent 1980;43:419-22.
15. Fuzzi M, Bonfiglioli R, Difebo G, Marin C, Caldari R, Tonelli MP. Posterior
porcelain inlay: clinical procedures and laboratory technique. Int J Periodontics
Restorative Dent 1989;9:274-87.
16. Neiva G, Yaman P, Dennison JB, Razzoog ME, Lang BR. Resistance to
fracture <strong>of</strong> three all-ceramic systems. J Esthet Dent 1998;10:60-6.
17. Soares CJ, Martins LRM, Pfeifer JM, Giannini M. Fracture resistance <strong>of</strong>
teeth restored with indirect-composite and ceramic inlay systems. Quintessence
Int 2004;35:281-6.
18. Hondrum SO. A review <strong>of</strong> the strength properties <strong>of</strong> dental ceramics.
J Prosthet Dent 1992;67:859-65.
19. Mak M, Qualtrough AJE, Burke FJ. The effect <strong>of</strong> different ceramic materials
on the fracture resistance <strong>of</strong> dentin-bonded crowns. Quintessence
Int 1997;28:197-203.
20. Cotert HS, Sen BH, Balkan M. In vitro comparison <strong>of</strong> cuspal fracture
resistances <strong>of</strong> posterior teeth restored with various adhesive restorations.
Int J Prosthodont 2001;14:374-8.
21. Burke FJ. The effects <strong>of</strong> variations in bonding procedure on fracture resistance
<strong>of</strong> dentin-bonded all-ceramic crowns. Quintessence Int 1995;26:293-300.
22. Assif D, Nissan J, Gafni Y, Gordon M. Assessment <strong>of</strong> the resistance to fracture
<strong>of</strong> endodontically treated molars restored with amalgam. J Prosthet
Dent 2003;89:462-5.
23. Salis SG, Hood JÁ, Stokes NA, Kirk EE. Patterns <strong>of</strong> indirect fracture in
intact and restored human premolar teeth. Endod Dent Traumatol 1987;3:
10-4.
24. Behr M, Rosentritt M, Leibrock A, Schneider-Feyrer S, Handel G. In-vitro
study <strong>of</strong> fracture strength and marginal adaptation <strong>of</strong> fiber-reinforced
adhesive fixed partial inlay dentures. J Dent 1999;27:163-8.
25. Dietschi D, Maeser M, Meyer JM, Holz J. In vitro resistance to fracture <strong>of</strong>
porcelain inlays bonded to tooth. Quintessence Int 1990;21:823-31.
26. St-Georges AJ, Sturdevant JR, Swift EJ Jr, Thompson JY. Fracture resistance
<strong>of</strong> prepared teeth restored with bonded inlay restorations. J Prosthet Dent
2003;89:551-7.
27. Borges GA, Sophr AM, de Goes MF, Sobrinho LC, Chan DC. Effect <strong>of</strong>
etching and airborne particle abrasion on the microstructure <strong>of</strong> different
dental ceramics. J Prosthet Dent 2003;89:479-88.
28. McLean JW, Hughes TH. The reinforcement <strong>of</strong> dental porcelain with
ceramic oxides. Br Dent J 1965;119:251-67.
29. Jones DW. Development <strong>of</strong> dental ceramics. An historical perspective.
Dent Clin North Am 1985;29:621-44.
30. van Noort R. Introduction to dental materials. 1st ed. St. Louis: CV Mosby;
1994. p. 201-14.
31. Kelly JR, Nishimura I, Campbell SD. Ceramics in dentistry: historical roots
and current perspectives. J Prosthet Dent 1986;75:18-32.
32. McDonald A. Preparation guidelines for full and partial coverage ceramic
restorations. Dent Update 2001;28:84-90.
33. Gibbs CH, Mahan PE, Mauderli A, Lundeen HC, Walsh EK. Limits <strong>of</strong> human
bite strength. J Prosthet Dent 1986;56:226-9.
34. Scherrer SS, de Rijk WG, Belser UC, Meyer JM. Effect <strong>of</strong> cement film
thickness on the fracture resistance <strong>of</strong> a machinable glass-ceramic. Dent
Mater 1994;3:172-7.
35. Jung YG, Peterson IM, Kim DK, Lawn BR. Lifetime-limiting strength
degradation from contact fatigue in dental ceramics. J Dent Res 2000;79:
722-31.
36. Pallis K, Griggs JA, Woody RD, Guillen GE, Miller AW. Fracture resistance
<strong>of</strong> three all-ceramic restorative systems for posterior applications. J Prosthet
Dent 2004;91:561-9.
428 VOLUME 95 NUMBER 6

SOARES ET AL
Reprint requests to:
DR CARLOS JOSE SOARES
FACULDADE DE ODONTOLOGIA
UNIVERSIDADE FEDERAL DE UBERLÂNDIA
AV. PARÁ, N.1720, BLOCO 2B, SALA 24
CAMPUS UMUARAMA
CEP: 38400-902
UBERLÂNDIA, MINAS GERAIS, BRAZIL
FAX: 55 34 32182279
E-MAIL: carlosjsoares@umuarama.ufu.br
Noteworthy Abstracts
<strong>of</strong> the
Current Literature
0022-3913/$32.00
Copyright Ó 2006 by The Editorial Council <strong>of</strong> The Journal <strong>of</strong> Prosthetic
Dentistry.
doi:10.1016/j.prosdent.2006.03.022
THE JOURNAL OF PROSTHETIC DENTISTRY
A confocal microscopic evaluation <strong>of</strong> resin-dentin interface using
adhesive systems with three different solvents bonded to dry and
moist dentin—An in vitro study
Mohan B, Kandaswamy D. Quintessence Int 2005;36:511-21.
Objective: Total dehydration <strong>of</strong> acid-etched dentin is known to cause the collapse <strong>of</strong> collagen fiber, which leads
to poor hybridization. Dentin-bonding systems with water as a solvent are found to rehydrate the collapsed
collagen. Acetone-based adhesives are found to compete with moisture, and the acetone carries the resin
deep into the dentin. The question arises whether to dry the dentin and use a water-based adhesive, or to
keep the dentin moist and use an acetone- or alcohol-based adhesive. The aim <strong>of</strong> this study was to compare
different bonding systems and techniques to assess which is most successful. A confocal microscope was
used to evaluate the amount <strong>of</strong> hybrid layer formation and the depth <strong>of</strong> resin tag formation.
Method and Materials: Superficial occlusal dentin specimens from 120 noncarious, freshly extracted human
premolars were used for the study. The dentin was etched using 36% phosphoric acid for 15 seconds and then
rinsed. The specimens were then randomly divided into 4 groups for different drying procedures; group I:
air-dried for 30 seconds; group II: air-dried for 3 seconds; group III: blotted dry; group IV: overwet.
The specimens were further subdivided into 3 groups to be tested with different bonding systems: subgroup
A: acetone-based adhesive (Prime & Bond NT); subgroup B: water-based adhesive (Syntac Single
Component); subgroup C: water- and ethanol-based adhesive (Single Bond). The resulting resin-dentin
interfaces were then examined and categorized via confocal microscopy, and relative values were assigned to
each specimen.
Results: Group IV (overwet) showed the lowest values, and the highest values were obtained in group III. The
highest values were seen in group III, subgroup A (blotted dry, acetone-based bonding agent, Prime & Bond
NT).
Conclusion: Under these conditions, using these three bonding systems, maximum hybridization and resin
tag formation were achieved using acetone-based adhesive on etched dentin kept moist by blot drying.—
Reprinted with permission <strong>of</strong> Quintessence Publishing.
JUNE 2006 429