d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617

available at www.sciencedirect.com

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

A study of the bending resistance of implant-supported

reinforced alumina and machined zirconia
abutments and copies

Anders Sundh a,b , Göran Sjögren a,∗

aDental Materials Science, Faculty of Medicine, Umeå University, SE-901 87 Umeå, Sweden
bNational Dental Health Service, County Council of Västerbotten, Norsjö, Sweden and
Manager of Research and Development, Cad.esthetics AB, Skellefteå, Sweden

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

Article history: Objectives. The purpose of the present study was to evaluate the bending resistance of
Received 30 August 2006 implant-supported CAD/CAM-processed restorations made out of zirconia or manually
Accepted 31 May 2007 shaped made out of reinforced alumina.
Methods. Units of abutments and copies made of (i) a prefabricated hot isostatic pressed
(HIPed) yttrium oxide partially-stabilized zirconia (Y-TZP) (Denzir), (ii) a prefabricated
Keywords: densely-sintered magnesia partially stabilized zirconia (Mg-PSZ) (Denzir-M) or, copies made
CAD/CAM of (iii) a prefabricated partially-sintered, porous reinforced alumina ceramic (RN synOcta-
Dental ceramics In-Ceram) were subjected to static loading perpendicularly at the long axis. The abutments
Dental implants were attached to either stainless steel analogs or titanium implant fixtures. The Y-TZP and
Mg-PSZ Mg-PSZ copies were bonded onto the ceramic abutments with a dual-cured resin composite
Titanium (Rely-X Unicem). Units of titanium abutment attached to a titanium implant fixtures were
Y-TZP used as reference.
Results. The units comprising Denzir abutments as delivered (p < 0.05) and heat-treated Den-
zir copies bonded to the Denzir abutments as delivered (p < 0.01) and attached to stainless
steel analogs exhibited significantly higher bending resistance than the control. The heat-
treated Denzir copies bonded to the heat-treated Denzir M abutments attached to titanium
implant fixtures and the In-Ceram specimens attached to stainless steel analogs showed
significantly (p < 0.05) lower bending resistance than the heat-treated Denzir copies bonded
to the Denzir abutments as delivered and attached to stainless steel analogs. No statistically
significant (p > 0.05) differences were seen among the other groups studied.
Significance. All the ceramic abutments and copies exhibited values that were equal or supe-
rior to that of the control and exceeded the reported value, up to 300 N, for maximum incisal
bite forces. To assess the clinical behavior long-term clinical studies should be conducted.
© 2007 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

1. Introduction restorations [1]. Metal restorations can, however, cause gin-

gival discoloration and the surrounding soft tissue to have
During the past few decades, unalloyed titanium has been an unnatural appearance [2–4]. Therefore, demands for den-
extensively used as a material of dental implant-supported tal restorations with a better esthetic appearance and scares

∗
Corresponding author. Tel.: +46 90 785 61 32; fax: +46 90 77 56 17.
E-mail address: Goran.Sjogren@odont.umu.se (G. Sjögren).
0109-5641/$ – see front matter © 2007 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dental.2007.05.021
612 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617

among the public concerning allegedly adverse effects of den-
tal metals and alloys have resulted in demands for esthetically
improved and biocompatible materials.
One ceramic system used for implant-supported abut-
ments is In-Ceram Zirconia (Vita Zahnfabrik, Bad Säckingen,
Germany) [4]. The In-Ceram Zirconia core system is based
on a partially-sintered, porous alumina structure supple-
mented with 33 wt% ZrO2 partially stabilized zirconia, which
is infiltrated with molten lanthanum glass [5]. When dental
restorations are made using this system either the slip-cast or
dry-pressed technique can be used [4,5].
Another ceramic of special interest because of its superior
mechanical properties compared to other dental ceramics is
yttrium oxide partially-stabilized zirconia (Y-TZP) [6–8]. Dental
restorations using Y-TZP are made using recently developed
CAD/CAM-systems and can be made either out of homo-
geneous ceramic green-body blanks of zirconia by milling
enlarged restorations, which are then sintered and shrunk
to the desired final dimensions [9], or by milling the restora-
tions directly to the final dimensions out of highly dense
sintered, hot isostatic pressed (HIPed) prefabricated Y-TZP
blanks [7,8,10]. Yttrium oxide is added to zirconia as a phase
stabilizer to maintain the high temperature tetragonal phase
(t) and in that way reduces the spontaneous transformation
into the monoclinc phase (m) at room temperature [6]. The
spontaneous transformation of t into m is known as ‘aging’
and it has been stated that the t → m transformation causes
mechanical property degradation of the material [6].
Other oxide additives than yttria can be used as phase
stabilizers and another zirconia ceramic for dental restora-
tions has recently been presented, Denzir-M, a prefabricated
densely-sintered magnesia partially stabilized zirconia (Mg-
PSZ) [11]. In Denzir-M, MgO is added as the ‘stabilising’ oxide.
Mg-PSZ is known to be one of the most commonly used engi-
neering ceramics based on zirconia [12] and to be one of
the toughest zirconia-based ceramics in use [13] and more
resistant to low temperature aging than Y-TZP [14]. The high
fracture toughness of the Mg-PSZ prefabricated blanks seems
to make it easier to mill dental ceramic restorations of reduced
thickness [11]. In an earlier paper it was reported that the hard-
ness of Mg-PSZ prefabricated blanks is less than that of the
HIPed Y-TZP and that the time needed to produce restorations
from prefabricated Mg-PSZ blanks and the wear on the cutting
tools were markedly reduced compared to when HIPed Y-TZP
blanks were milled [11].
In previous papers, it has been suggested that machin-
ing and other mechanical surface treatments can affect the
generation of phase transformation of Y-TZP and Mg-PSZ and
introduce residual compressive stresses on the surface, which
influence the mechanical properties of zirconium ceramics
[6–8,11]. In addition, it has been stated that subsequent heat-
treatment seems to induce relaxation of the tensile stresses
generated during machining, which also influences the prop-
erties of the ceramic [7,8,11]. This information has to be taken
into consideration when those materials are subjected to sur-
face treatments or veneering.
Since there is an increasing demand for ceramic implants
the fracture resistance, among other things, of units compris-
ing implant-supported ceramic abutments and copies are of
interest. Ceramics are brittle and have a tendency to crack
but should withstand compressive stresses better than ten-
sile stresses [15]. Compressive loads to cantilever designed
implant-retained prostheses are, however, said to result in
bending and tensile forces within certain regions of the pros-
thesis [16,17]. The bending resistance of implant-supported
ceramic abutments and copies is, thus, of particular interest.
According to a survey of the literature, no data were avail-
able on the strength of implant-supported abutments and
copies made of Mg-PSZ, HIPed Y-TZP or zirconia reinforced alu-
mina ceramics. The aim of the present work, therefore, was to
determine the bending resistance of implant-supported abut-
ments and copies made from prefabricated Mg-PSZ or HIPed

Table 1 – Summary of the components studied

Type of material Manufacturer

RelyX Unicem Resin composite cement 3M ESPE, St. Paul, MN, USA
Straumann solid screw implant, SLA. Ø Titanium Institut Straumann AG,
3.3 mm, 8.0 mm PLUS Waldenburg, Switzerland
Straumann SCS occlusal screw, long, L 7.6 Titanium Institut Straumann AG
mm
Straumann analog for RN synOcta, L 12.0 Stainless steel Institut Straumann AG
mm
Straumann RN synOcta abutment for Titanium Institut Straumann AG
cement-retained crowns/bridges,
height 5.5 mm
Straumann RN synOcta abutment, height Titanium Institut Straumann AG
2.5 mm
Denzir M Prefabricated densely-sintered magnesia partially Cad.esthetics AB,
stabilized zirconia (Mg-PSZ) Skellefteå, Sweden
Denzir Prefabricated hot isostatic pressed (HIPed) yttrium Cad.esthetics AB
oxide partially-stabilized zirconia (Y-TZP)
RN synOcta-In-Ceram Partially-sintered, porous alumina structure Vita Zahnfabrik, Bad
reinforced with 33 wt% ZrO2 partially stabilized Säckingen, Germany
zirconia and infiltrated with molten lanthanum
glass
 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617 613

Y-TZP zirconia blanks and from a zirconia reinforced alumina

ceramic.

2. Materials and methods

Units of threaded titanium screw implant fixtures or stain-

less steel analogs with ceramic abutments and ceramic
copies were studied in various combinations. As a con-
trol, units of titanium abutments attached to threaded
titanium screw implant fixtures were used. The materials
used and the combinations evaluated are summarized in
Tables 1 and 2, respectively. Using the Cad.esthetics CAD/CAM
system (Cad.esthetics AB) Mg-PSZ (Denzir-M) abutments and
HIPed Y-TZP (Denzir) abutments and copies were manu-
factured from prefabricated ceramic blocks. The zirconia
reinforced alumina ceramic was delivered as prefabricated
blanks, RN synOcta-In-Ceram (In-Ceram Zirconia), which
was screwed onto their corresponding titanium abutments
(RN synOcta). By means of rotating diamonds, the copies
were then manually shaped by one dental technician at
Cad.esthetics AB. Another dental technician at the Depart-
ment of Dental Materials Science, Umeå University, Umeå,
Sweden then carried out the glass infiltration of the In-Ceram
Zirconia copies. The Denzir-M and the Denzir specimens were
studied either as delivered directly from the manufacturer
or after the copies and/or abutments had been subjected to
heat-treatment at 950 ◦ C for 1 min. The shapes and dimen-
sions of the Denzir-M and Denzir abutments and copies
are shown in Fig. 1A and B, the In-Ceram Zirconia copies
in Fig. 1 C, and the titanium abutments used for the In-
Ceram Zirconia copies in Fig. 1 D. The wall thickness of
the CAD/CAM-processed Denzir and Denzir abutments was
occlusally 0.5 mm and cervically 0.3 mm, respectively, and
the overall wall thickness of the CAD/CAM-processed Den-
zir and Denzir copies was 0.5 mm. The wall thickness of the
manually shaped In-Ceram Zirconia copies, determined for
each specimen by means of a measuring instrument with
an accuracy of 0.1 mm, at four preselected measuring points

Table 2 – Summary of the various combinations used

Unit

(A) Titanium screw implant fixture/Titanium abutment

(B) Stainless steel analog/Denzir abutment as delivered
(C) Stainless steel analog/Denzir heat-treated abutment
(D) Stainless steel analog/Denzir M abutment as
delivered
(E) Stainless steel analog/Denzir abutment as delivered
and Denzir heat-treated copy
(F) Stainless steel analog/Denzir M heat-treated Fig. 1 – Shape and dimensions of: (A) Denzir and Denzir-M
abutment and Denzir heat-treated copy ceramic copies; (B) Denzir and Denzir-M ceramic
(G) Titanium screw implant fixture/Denzir M abutments; (C) In-Ceram Zirconia ceramic copies; (D) RN
heat-treated abutment and Denzir heat-treated copy synOcta titanium abutments used for the In-Ceram
(H) Stainless steel analog/Denzir heat-treated abutment
Zirconia copies. The letters a–d in Fig. 1C denote
and Denzir heat-treated copy
preselected measuring points for measurement of the wall
(I) Titanium screw implant fixture/Denzir heat-treated
abutment and Denzir heat-treated copy thickness of the In-Ceram Zirconia copies.
(J) Stainless steel analog/Titanium abutment and
In-Ceram Zirconia copy
614 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617

3. Results

Medians, first and third quartiles, minimum and maximum

values, outliers and extreme outliers of the load (Newton)
determined for the 10 series of specimens are presented in
Fig. 3. The failure mode of each unit is presented in Table 3.
Fig. 4A–D shows representative images of the specimens after
the load-to-fracture test.
The load necessary to fracture the units comprising the
Denzir abutments as delivered (p < 0.05) and the heat-treated
Denzir copies bonded to the Denzir abutments as delivered
Fig. 2 – (A) Ceramic abutment; (B) ceramic copy bonded to a
(p < 0.01) and attached to stainless steel analogs was signifi-
ceramic abutment fixed in the specimen holder and
cantly higher than that of the control. The load necessary to
subjected to a vertical load by means of a hardened
fracture the heat-treated Denzir copies bonded to the heat-
chisel-shaped steel blade.
treated Denzir-M abutments assembled on titanium implant
fixtures (p < 0.05) and the units comprising the In-Ceram Zirco-
nia copies assembled on stainless steel analogs (p < 0.05) were
significantly lower than that of the heat-treated Denzir copies
(Fig. 1C), was 1.2 ± 0.1 mm. Each abutment was screwed onto bonded to the Denzir abutments as delivered and attached
a titanium implant fixture or a stainless steel analog with to stainless steel analogs. No statistically significant (p > 0.05)
a torque of 35 N cm. The same torque was used for each differences were seen among the other groups studied.
In-Ceram Zirconia copy when they were screwed onto the After the test all the ceramic copies were still bonded to
corresponding titanium abutments. The Denzir and Denzir- the abutments. Twenty-nine of the titanium abutment screws
M abutments were screwed onto the stainless steel analogs were visibly deformed and six were fractured. All the tita-
or the titanium implant fixtures with abutment titanium nium implant fixtures were markedly bent (Fig. 4), whereas
screws 7.6 mm long, whereas the In-Ceram Zirconia copies only five of the stainless steel analogs were slightly deformed
were screwed onto their corresponding titanium abutments (Table 3). All the In-Ceram Zirconia copies were fractured in
using titanium screws 5 mm long. Using a dual-cured resin the cervical part of the copies (Fig. 4 D). No visible fracture
composite (RelyX Unicem) the Denzir-M and Denzir ceramic was seen with the naked eye of the copies made out of the
copies, i.e. the specimens E-I in Table 2, were cemented HIPed Y-TZP (Denzir) prefabricated ceramic blanks, with the
onto the abutments at room temperature. A constant load
of 115 N was applied on the top of the central part of the
copy. The dual-cured resin composite was then polymerized
by means of a halogen lamp (Norlite, Germany) for 20 s in
four directions. Five specimens from each group were stud-
ied.
The titanium implant fixture or the stainless steel analog
of the assembled units was fixed with 2 horizontal screws in a
stainless steel specimen holder (Fig. 2A and B). Using a univer-
sal testing machine (Alwetron TCT 5/10, Lorentzon and Wettre,
Spånga, Sweden) the assembled units were then subjected to
compressive loading by means of a hardened chisel-shaped
steel blade (thickness 0.7 mm and width 6.5 mm) with the force
applied perpendicularly at the long axis of the implant fixture
or of the stainless steel analog (Fig. 2A and B). The load was
applied perpendicularly at the abutment or at the copy 3 mm
from the border of the titanium implant fixture or the stain-
less steel analog and 6 mm from the level of the specimen
holder (Fig. 2A and B) with a cross-head speed of 0.5 mm/min.
The loads were run with increasing force until the testing
Fig. 3 – Box-plot diagram comprising the load at fracture (N)
machine automatically registered that the force was 1% below
of the 10 series of the implant-supported units studied. The
the highest level recorded during the test and the testing was
capitals denote the units in accordance with Table 2. Five
interrupted.
specimens in each test group. Data are presented as
medians and 1st and 3rd quartiles. A horizontal line within
2.1. Statistical analysis the box presents the median. The maximum and minimum
values are illustrated via the upper and lower strokes. ()
The values were statistically analyzed using the Kruskal– marks outliers. ( ) marks extreme outliers. Statistically
Wallis test supplemented with Bonferroni at a significance significant difference: *p < 0.05, **p < 0.01; Kruskal–Wallis
level of p < 0.05. test supplemented with Bonferroni.
 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617 615

Table 3 – Location and mode of failure after load-to-fracture test. The letters in the head-line refer to the groups of units
listed in Table 2. (n = 5 in each group)
Fractured or deformed component(s) A B C D E F G H I J

Titanium abutment screw fractured 1 2 1 2

Titanium abutment screw deformed 5 2 5 4 5 3 5
Abutment fractured 5 5 3 4 5 5 5 5
Copy fractured 1 2 5
Titanium implant fixture or stainless steel analog deformed 5 2 3 5 5
No visible fracture of abutment and copy 2 1

The figures in the columns denote number of specimens. For some of the units more than 1 failure mode is given.

exception of three copies (Table 3). Fractures were observed in

all the Denzir-M and the Denzir ceramic abutments in close
4. Discussion
proximity to the implant fixture-abutment interface or to the
In the current study, the bending resistance of implant-
stainless steel analog-abutment interface (Fig. 5), except for
supported units of various types of abutments and copies
three of the Denzir abutments tested as delivered (Table 3). In
was determined. The load was applied perpendicularly at the
11 cases the fractures in the implant fixture-abutment inter-
long axis of the implant fixture or of the stainless steel ana-
face or the stainless steel analog-abutment interface were
log, 6 mm from the border with the surrounding material and
undetected until the abutments were unscrewed.
3 mm from the border with the implant titanium fixture or
the stainless steel analog. In that way, the specimens should
have been subjected to relatively high stress. Usually, in in vitro
studies of implant-supported specimens the fracture loads are
determined by force application at 30◦ to the long axis of the
implants [2,18–20]. However, since the aim of the current study
was to evaluate the bending resistance of all-ceramic abut-
ments and copies, this experimental set up was chosen in
order to create a worst-case situation and subject the spec-
imens to severe stress and because the load on anterior teeth
can sometimes be almost horizontal. In addition, the load
was applied to the specimens by means of a chisel-shaped

Fig. 4 – Representative images of specimens after the

load-to-fracture test: (A) Titanium abutment assembled on
a titanium implant screw fixture; (B) Denzir-M copy bonded
onto a Denzir-M abutment and assembled on a titanium
implant screw fixture; (C) Denzir copy bonded onto a Denzir Fig. 5 – Representative image after the load-to-fracture test
abutment and assembled on a stainless steel analog; (D) of ceramic abutments with fracture in close proximity to
In-Ceram Zirconia copy assembled on a titanium implant the abutment/stainless steel analog or abutment/implant
screw fixture. fixture interface.
616 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617
steel blade, which should have aggravated the stress on the
specimens.
The bending resistance of the abutments and copies was
determined for the units assembled either on stainless steel
analogs or titanium implant fixtures. This was because the
modulus of elasticity of stainless steel is reported to be higher
than that of unalloyed titanium, around 200 GPa and 100 GPa,
respectively [21]. The rigidity of the stainless steel analogs was,
thus, considered to be higher than that of the implant fix-
tures and should therefore resist bending forces better. That
is, using stainless steel analogs was meant to provide fur-
ther information about the bending resistance of the ceramic
abutments and the copies studied. However, no statistically
significant differences were found between the bending resis-
tance determined with similar combinations of abutments
and/or copies assembled on the titanium implant fixtures and
those assembled on the stainless steel analogs. After fracture
was registered all the titanium implant fixtures were, how-
ever, significantly more bent than the stainless steel analogs.
In addition, the range of variation in the values obtained for
the bending resistance was considerably wider for the units
assembled on the stainless steel analogs.
After the load-to-fracture test no visible fracture of the
copies was seen with the naked eye, with the exception
of three copies. When the abutments were unscrewed and
inspected after the test, fractures were however observed in all
the ceramic abutments in the proximity of the implant screw-
abutment interface or of the stainless steel analog-abutment
interface, except for three of the abutments. This implies that
this part of implant-supported restorations is an area of exces-
sive stress concentration. The finding is comparable to the
results reported in previous studies of the fracture resistance
of alumina and zirconia implant-supported abutments [18,19].
In those in vitro studies, it was stated that the crack started
from the cervical portion of the abutments close to the implant
interface and the abutment screw [18,19].
Among the 10 unit combinations tested, the only com-
ponent that was included in all the combinations was the
titanium abutment screw. Since statistically significant differ-
ences were only found between a small number of groups one
can speculate that the titanium abutment screw was the com-
ponent that largely determined the bending resistance of the
assembled units. Thirty-five of the titanium abutment screws
were markedly bent, deformed, and/or fractured. This is con-
sistent with the findings in a study by Tripodakis et al. [2] of
the fracture strength and the failure mode of all-ceramic In-
Ceram Alumina abutments under static load, in which it was
concluded that the weakest link was the abutment screw.
Even though only two groups showed statistically signifi-
cantly higher values compared to the control, the mean values
of the bending resistance of the units comprising ceramic
abutments, irrespective of type of ceramic, pretreatment or
combination, were superior to those of the units comprising
titanium abutments. One possible explanation for this may
be that the ceramic abutments, with a higher modulus of
elasticity and flexural strength than the titanium abutments,
stabilized the implant-supported units and, thus, reduced the
stress on the titanium abutment screws. Values for the mod-
ulus of elasticity provided by the manufacturers were 200 GPa
for Denzir-M and Denzir, 258 GPa for In-Ceram Zirconia, and
110 GPa for the titanium abutments, respectively. The flex-
ural strength, according to the manufacturers’ information,
was 500 MPa for Denzir-M, 1000 MPa for Denzir, 600 MPa for
In-Ceram Zirconia, and 400 MPa for the titanium abutments,
respectively. In this context it is worth noting that the mean
values obtained for the units comprising Mg-PSZ (Denzir-M) or
HIPed Y-TZP (Denzir) abutments as delivered were superior,
but not to a statistically significant level, to those obtained
for the units comprising heat-treated abutments. This is in
agreement with the findings in previous studies, where it has
been shown that heat treatment of machined zirconia reduces
its fracture resistance [8,11]. Another interesting finding was
that in a preparatory study of four implant-supported units
comprising titanium abutments assembled on stainless steel
analogs, minor fractures could be seen in the upper part of
the stainless steel analogs in close proximity to the titanium
abutment interface after the bending test. This implies that
the stainless steel analogs and titanium abutments were sub-
jected to considerable stress concentration in this area. The
mean value of those specimens, 380 ± 189 N, was the lowest of
all the unit combinations tested, and almost exactly the same
as the mean value of the titanium abutments assembled to the
titanium implant fixtures (380 ± 7 N). The theory that ceramic
abutments can strengthen the implant-supported units is to
some extent similar to the theory presented in a previous
study of the protective effect from bending forces that the
coping-abutment interface has on the retaining screw [17]. In
this earlier study, it was suggested that the configuration of the
coping-abutment interface would prevent rotation around the
axis of the bending moment and that long, vertical mating sur-
faces would protect the retaining screw [17]. Thus, the design
of the head of the implant fixture and of the fixture/abutment
interface seems to play an important role in reducing bending
forces on the abutments, copies and retaining screws. How-
ever, more studies are needed to confirm such hypotheses.
The mean values determined for the bending resistance in
the present study ranged from 380 N to 649 N. The titanium
abutment assembled on titanium implant fixtures showed
the lowest mean value (380 ± 7 N), whereas the Denzir heat-
treated copies bonded to the Denzir abutments as delivered,
and assembled on the stainless steel analogs showed the high-
est (649 ± 130 N). Reported mean value for the physiological
maximal bite force for men in the anterior region in a study
by Waltimo and Könönen [22] was up to around 300 N and
up to around 850 N in the molar region. However, since the
stresses and strains of dental restorations in vivo are complex,
it is difficult to extrapolate in vitro values to clinical situa-
tions. Values reported in previous in vitro studies addressing
the bending resistance of implant-supported all-ceramic abut-
ments and/or crowns range from 170 N to 738 N [2,18–20,23]. In
all those studies, the load was applied at 30 or 45◦ to the long
axis of the implant and in some of the studies the specimens
were subjected to dynamic loading and thermal cycling, and,
in addition, the size and type of implant fixtures and the sur-
rounding materials varied, making direct comparison between
the results difficult [2,18–20]. For example, in the study by
Gehrke et al. [20] the specimens that were subjected to static
loading exhibited a maximum fracture load of 672 N, whereas
only 269 N was recorded for those subjected to 800,000
cycles. Thus, the experimental set up significantly affected the
 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 611–617
outcome. In the study by Att et al. [18], the mean value of the
fracture resistance of titanium abutment/Al2 O3 crown units
subjected to vertical loading was 1344 N, which was around
three times that determined for the units in the same study
that comprised Al2 O3 crowns bonded to ZrO2 or to Al2 O3 abut-
ments. This contrasts with the outcome in the present study,
in which the titanium abutment exhibited inferior or almost
equal mean values compared to the ceramic units. However,
in another study it was shown that the loading direction could
influence the results [23]. In that study, under vertical loading
the fracture resistance of restorations that comprised tita-
nium abutments was almost twice that of those comprising
ceramic abutments, whereas under oblique loading (45◦ ) no
statistically significant differences in fracture resistance were
seen between the restorations with ceramic abutments and
those with titanium abutments [23]. All the mean values were
also significantly reduced when the specimens were subjected
to oblique loading [23]. For example, the fracture resistance of
In-Ceram Alumina crowns bonded to alumina abutments was
reduced from 858 ± 91 N at vertical loading to 182 ± 55 N at
oblique loading [23].
To summarize, all the combinations of ceramic abutments
and copies in the present study exceeded the reported value of
up to 300 N for the physiological maximal incisal bite force. In
addition, the bending resistance of the ceramic specimens was
equal or superior to the titanium control. The combinations of
ceramic abutments and copies studied thus showed promis-
ing bending resistance under the tested conditions. However,
long-term clinical studies should be conducted to assess the
clinical behavior of the ceramic abutments and copies under
clinical conditions.
references
[1] Adell R, Lekholm U, Rockler B, Brånemark PI. A 15-year study
of osseointegrated implants in the treatment of the
edentulous jaw. Int J Oral Surg 1981;10:387–416.
[2] Tripodakis AP, Strub JR, Kappert HF, Witkowski S. Strength
and mode of failure of single implant all-ceramic abutment
restorations under static load. Int J Prosthodont
1995;8:265–72.
[3] Yildirim M, Edelhoff D, Hanisch O, Spiekermann H. Ceramic
abutments—a new era in achieving optimal esthetics in
implant dentistry. Int J Periodontics Restor Dent
2000;20:81–91.
[4] Jackson MC. Posterior implants using a new ceramic
material. J Dent Technol 1999;16:19–22.
[5] Apholt W, Bindl A, Lüthy H, Mörmann WH. Flexural strength
of Cerec 2 machined and jointed InCeram-Alumina and
InCeram-Zirconia bars. Dent Mater 2001;17:260–7.
617
[6] Piconi C, Maccauro G. Zirconia as a ceramic biomaterial. Rev
Biomater 1999;20:1–25.
[7] Sundh A, Sjögren G. A comparison of fracture strength of
yttrium-oxide-partially-stabilized zirconia ceramic crowns
with varying core thickness, shapes and veneer ceramics. J
Oral Rehabil 2004;31:682–8.
[8] Sundh A, Molin M, Sjögren G. Fracture resistance of yttrium
oxide partially-stabilized zirconia all-ceramic bridges after
veneering and mechanical fatigue testing. Dent Mater
2005;21:476–82.
[9] Filser F, Kocher P, Weibel F, Luthy H, Scharer P, Gauckler LJ.
Reliability and strength of all-ceramic dental restorations
fabricated by direct ceramic machining (DCM). Int J Comput
Dent 2001;4:89–106.
[10] Sjölin R, Sundh A, Bergman M. The Decim system for the
production of dental restorations. Int J Comput Dent
1999;2:197–207.
[11] Sundh A, Sjögren G. Fracture resistance of all-ceramic
zirconia bridges with differing phase stabilizers and quality
of sintering. Dent Mater 2006;22:778–84.
[12] Liu Z, Spargo AE. Identification of various phases in HRTEM
images of MgO-PSZ. J Electron Microsc 2001;50:443–6.
[13] Hannink RHJ, Howard CJ, Kisi EH, Swain MV. Relationship
between fracture toughness and phase assemblage in
Mg-PSZ. J Am Ceram Soc 1994;77:571–9.
[14] Masaki T. Mechanical properties of Y-TZP after aging at low
temperatures. Int J High Tech Ceram 1986;2:85–98.
[15] Lüthy H, Filser F, Loeffel O, Schumacher M, Gauckler LJ,
Hammerle CH. Strength and reliability of four-unit
all-ceramic posterior bridges. Dent Mater 2005;21:
930–7.
[16] Rangert B, Jemt T, Jörneus L. Forces and moments on
Brånemark implants. Int J Oral Maxillofac Implants
1989;4:241–7.
[17] Jaarda MJ, Razzoog ME, Gratton DG. Effect of preload torque
on the ultimate tensile strength of implant prosthetic
retaining screws. Implant Dent 1994;3:17–21.
[18] Att W, Kurun S, Gerds T, Strub JR. Fracture resistance of
single-tooth implant-supported all-ceramic restorations: an
in vitro study. J Prosthet Dent 2006;95:111–6.
[19] Yildirim M, Fischer H, Marx R, Edelhoff D. In vivo fracture
resistance of implant-supported all-ceramic restorations. J
Prosthet Dent 2003;90:325–31.
[20] Gehrke P, Dhom G, Brunner J, Wolf D, Degidi M, Piattelli A.
Zirconium implant abutments: fracture strength and
influence of cyclic loading on retaining-screw loosening.
Quintessence Int 2006;37:19–26.
[21] Anusavice KJ. Phillip’s science of dental materials. 11th ed.
St. Louis: Saunders; 2003. p. 765.
[22] Waltimo A, Könönen M. A novel bite force recorder and
maximal isometric bite force values for healthy young
adults. Scand J Dent Res 1993;101:171–5.
[23] Cho HW, Dong JK, Jin TH, Oh SC, Lee HH, Lee JW. A study on
the fracture strength of implant-supported restorations
using milled ceramic abutments and all-ceramic crowns. Int
J Prosthodont 2002;15:9–13.