Journal of the Mechanical Behavior of Biomedical Materials 91 (2019) 32–37

Contents lists available at ScienceDirect

Journal of the Mechanical Behavior of

Biomedical Materials
journal homepage: www.elsevier.com/locate/jmbbm

Bacterial leakage and bending moments of screw-retained, composite- T

veneered PEEK implant crowns
Andreas Wachtela, Tycho Zimmermanna, Mona Sütela, Ufuk Adalia, Mohamad Abou-Emarab,
⁎
Wolf-Dieter Müllera, Sven Mühlemannc, Andreas Dominik Schwitallaa,
a
Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Dental Materials and
Biomaterial Research, Department of Prosthodontics, Geriatric Dentistry and Craniomandibular Disorders, Aßmannshauser Str. 4-6, 14197 Berlin, Germany
b
Beuth University of Applied Sciences, Luxemburger Str. 10, 13353 Berlin, Germany
c
University of Zurich, Clinic of Fixed and Removable Prosthodontics and Dental Material Science, Plattenstrasse 11, 8032 Zürich, Switzerland

A R T I C LE I N FO A B S T R A C T

Keywords: Due to its elastic modulus close to bone, the high-performance material PEEK (polyetheretherketone) represents
PEEK an interesting material for implant-supported dental prostheses. Besides a damping effect of masticatory forces,
Dental implant it might have a sealing effect against bacterial leakage of the implant-abutment interface (IAI). So far, PEEK has
Implant crown only been used for provisional implant crowns. Therefore, the aim of the study was the evaluation of the bac-
Conical implant-abutment connection
terial tightness of screw-retained PEEK crowns on titanium implants with conical IAI during masticatory si-
Morse taper
mulation and subsequent bending moment testing.
Bacterial leakage
Ten screw-retained implant crowns in the shape of an upper central incisor consisting of a PEEK crown
framework veneered with composite were connected to NobelActive RP titanium implants (4.3 × 11.5 mm,
Nobel Biocare AB, Göteborg, Sweden) with a tightening torque of 15 Ncm.
Prior to tightening, the interior of the implant was inoculated with a bacterial suspension of Enterococcus
faecium. The specimens were overmolded with indicating agar (Kanamycin-Aesculin-Azid-Agar (KAAA), Oxoid
Limited, Basingstoke United Kingdom), that turns black in contact with E. faecium. The specimens were sub-
jected to a cyclic masticatory simulation whereby a force of 50 N cm was applied at an angle of 30° to the implant
axis for 1.2 million cycles. Afterwards, the specimens were subjected to a static loading test according to ISO
14801:2007 to determine the bending moment.
During masticatory simulation neither a loosening of the implant screws nor any damage to the veneer or the
PEEK framework occurred. Furthermore, no bacterial leakage could be observed in any of the specimens. The
average maximum bending moment was measured at 352.13 ± 48.96 N cm.
Regarding masticatory forces, PEEK implant crowns seem to be applicable as definitive implant-supported
restorations. Furthermore, the bacterial tightness of the IAI of screw-retained one-piece PEEK implant crowns is
advantageous compared to superstructures of conventional materials. Further studies are necessary to sub-
stantiate the clinical significance of these findings.

Introduction likely attributable to a closer fit between implant and abutment surfaces

Regarding long-term success, implant systems with an implant-
abutment interface (IAI) exhibiting a conical implant-abutment con-
nection demonstrate advantages compared to other systems (Pessoa
et al., 2017; Goiato et al., 2015; Mangano et al., 2015). One of their
main assets is their lower bacterial leakage (Mishra et al., 2017). It has
been suggested that bacterial colonization of the implant interior may
be a co-factor in the development of peri-implantitis (Jervøe-Storm
et al., 2015). The higher bacterial tightness of a conical IAI is most
within the IAI (Larrucea Verdugo et al., 2014). This taper- and torque-
dependent decrease in microscopic gap width leads to a reduction in
micro-scale movements along the IAI (Liu and Wang, 2017). Excessive
micromovements within the IAI can increase abrasive wear and po-
tentially cause damage to the IAI. Furthermore, micromovements have
been shown to result in a pumping effect, which increases bacterial
leakage as well as leakage of the toxic byproducts of bacterial meta-
bolism (Liu and Wang, 2017).
Conventional implant abutments may be fabricated out of metals

⁎
Corresponding author.
E-mail address: andreas.schwitalla@charite.de (A.D. Schwitalla).

https://doi.org/10.1016/j.jmbbm.2018.11.027
Received 17 October 2018; Received in revised form 26 November 2018; Accepted 27 November 2018
Available online 28 November 2018
1751-6161/ © 2018 Elsevier Ltd. All rights reserved.
A. Wachtel et al. Journal of the Mechanical Behavior of Biomedical Materials 91 (2019) 32–37

such as titanium or of ceramics such as zirconium oxide ceramic. D = 12 mm

Zirconium oxide is preferentially utilized in the anterior region of the
mouth due to its white color (Ioannidis et al., 2017), whereas titanium
is favored in the posterior region due to its higher mechanical strength
(Truninger et al., 2012). Metals and ceramics, however, are very rigid
restorative materials with an elastic modulus of 110 GPa for titanium

H = 14 mm
and 210 GPa for zirconium oxide (Kaleli et al., 2018). This high rigidity
is believed to negatively impact the peri-implant bone, due to the un-
mitigated transfer of masticatory forces through the ankylotic implant-
bone junction (Kaleli et al., 2018). Stress peaks imparted on the bone-
implant interface by excessive masticatory loads may cause damage to
the peri-implant bone and thus lead to marginal bone resorption and
implant loss (Esposito et al., 1998). Therefore, a more elastic abutment
material may be of great benefit, as it would be able to absorb the
masticatory forces through elastic self-deformation (Magne et al.,
2013), thus reducing the stress transferred to the peri-implant bone. * h = 2.55 mm
Additionally, an elastic abutment material would be able to prevent the
deleterious micromovements along the IAI (Magne et al., 2013).
Fig. 1. Cylinder shaped PEEK blank with apically protruding IAI (*).
Recently, the high performance polymer polyether ether ketone
(PEEK) was introduced as a restorative material in dental medicine. It
has a comparatively low elastic modulus of about 3.5 GPa (Kaleli et al.,
2018) and the advantages of an unobtrusive light grey, opaque coloring
(Bodden et al., 2017) and biological inertness (Johansson et al., 2018).
Dental restorations manufactured from PEEK have a high wearing
comfort for the patient (Zoidis, 2018a) due to their low density (Bodden
et al., 2017). Its good polishability and accompanying low plaque af-
finity make it a promising candidate for the fabrication of implant
abutments (Hahnel et al., 2015). For these reasons, PEEK has become a
serious contender among the list of materials in consideration for dental
implant abutments (Al-Rabab'ah et al., 2017).
So far, the long-term performance of a single implant crown com-
plex made out of PEEK has not been thoroughly investigated. Therefore,
the aim of the present study was to test the bacterial leakage and me- a b c
chanical stability of screw-retained single implant crowns in an implant
system with a conical IAI during and after repeated flexural loading. Fig. 2. Digital STL model of the optically scanned implant-abutment-crown-
The hypotheses were, that when using PEEK for the implant-crown combination (a); Modified digital model with smoothed surface (b); Digital
complex (abutment and crown consisting of a single piece) i) bacterial model reduced to framework for subsequent veneering. Arrows indicate
leakage is reduced and ii) simulated long-term stability is sufficiently boundary of veneering area.
high for being employed as definitive restoration.
the crown contour to reproducibly define the marginal extension of the
Materials and methods veneering material. Coronally of this margin, the crown form was re-
duced in order to provide sufficient space for the subsequent veneering.
A cylinder shaped PEEK blank served as raw material for the pro- Hereby the incisal edge was cut back by 1.5 mm and the circumferential
duction of the screw retained one-piece PEEK implant crowns with contour was cut back by 0.5 mm, resulting in a crown framework as
internal conical connection. The blank was manufactured by Nobel seen in Fig. 2c.
Biocare AB, Göteborg, Sweden, for the purpose of this study. The blank According to this digital blueprint, the crown frameworks were
featured a centrally located screw channel as well as a conical implant- manufactured using CAD/CAM milling (Klingel medical metal GmbH,
abutment-interface (IAI) compatible with the implant-side IAI of a Pforzheim, Germany). Thereby the PEEK blanks were fixated on la-
4.3 mm diameter NobelActive RP implant of the same manufacturer boratory analogues (Implant Replica CC RP, Nobel Biocare AB) using
(Fig. 1). the corresponding implant screws (Lab Screw RP/WP, Nobel Biocare
To be able to compare the results of the present study with results AB), which were manually tightened. The laboratory analogues served
from the literature, the design of the abutment-crown-complex of an as mounting support for the chucks of a dual axis (Mazak, Oguchi,
upper central incisor evaluated in a previously published study served Japan) and subsequently a five axis milling machine (Fanuc, Oshino,
as a template for the shape of the crown (Mühlemann et al., 2014). A Japan).
template specimen was provided, which consisted of a glass ceramic Ten crown frameworks were produced (Fig. 3a). To initiate the
crown (IPS Empress CAD, Ivoclar Vivadent, Schaan, Liechtenstein) on a veneering process, the final crown shape was modelled onto the PEEK
titanium abutment with a 1-piece internal implant-abutment connec- framework using modeling wax. A silicone mold of the wax contour was
tion (CARES abutment on Bone Level RC implant with implant diameter produced by investing the crown up to the margin of the IAI into
of 4.1 mm, Straumann, Basel, Switzerland). The crown shape was di- transparent silicone (Regofix Dreve, Unna, Deutschland).
gitally scanned (D700, 3shape, Copenhagen, Denmark) and converted After the complete curing of this transparent silicone mold, the wax
to a surface tessellation language (STL) file, which in turn was loaded was removed from the crown framework with warm water to produce a
into a 3-D modeling software (Creo parametric 3.9, Parametric Tech- lost form negative between the crown framework and the mold. The
nology Corporation, Needham, Massachusetts, USA) (Fig. 2a and 2b). surface of the crown framework area was micro-abraded using cor-
A commercially available zirconium abutment (Procera Esthetic undum (particle size of 110 µm, pressure of 2.5 bar). Then, a primer
Abutment #8, Nobel Biocare AB) was scanned and digitized as well. (ML Primer, Shofu, Ratingen, Germany) was applied and subsequently
The location of its circumferential chamfer was digitally projected onto

33
A. Wachtel et al.
a b
Fig. 3. Crown framework (a) and fully veneered screw retained implant crown
connected to titanium implant (b).
air dried according to the manufacturer's recommendations. This was
followed by applying two layers of opaquing agent (Ceramage Opaker,
Shofu); hereby the first layer was light cured for 1 min and the second
layer was light cured for 2 min (HiLite power, Heraeus Kulzer, Hanau,
Germany). The crown framework was placed back into the silicone
mold and the lost form between the silicone mold and the crown fra-
mework was filled with flowable composite (Ceramage Flowable
Composite, Shofu) and light cured for three minutes. Excess composite
tags were removed with a fine bur resulting in the finished test speci-
mens (Fig. 3b).
Ten implant mounting supports were produced by filling a stainless
steel tube (length:50 mm, diameter: 28 mm, wall thickness: 1 mm) with
poly methyl methacrylate (PMMA; PalaXpress-transluzent, Heraeus
Kulzer, Hanau, Germany) and curing it. Beforehand, the interior of the
stainless steel tube was preconditioned using the Rocatec procedure
(3 M ESPE, St. Paul, Minnesota, USA) in order to permit a strong ad-
hesive bond between polymer and stainless steel. The upper surface of
the cured polymer was smoothed out and furnished with a centrally
located vertical bore of 4.5 mm diameter. To simulate a peri-implant
bone loss of 3 mm as required by ISO 14801:2007 the bores had a depth
of 8.5 mm. The bore was equipped with implants with a diameter of
4.3 mm and a length of 11.5 mm (NobelActive RP, Nobel Biocare AB).
The implants were glued into the bore using PMMA (PalaXpress-
transluzent, Heraeus Kulzer).
For the bacterial leakage test, a bacterial suspension of harmless
Enterococcus faecium was prepared according to a previously described
technique (Wachtel et al., 2016). 6.5 µl of the bacterial suspension were
injected into the implant hollow, a volume corresponding to the im-
plants interior volume. Subsequently, the screw retained and veneered
implant crowns were fixated on the implants (Fig. 3b). The implant
screws were tightened with a torque of 15 N·cm as recommended by
BEGO Implant Systems GmbH & Co. KG for PEEK abutments. The
screws were protected with cotton pellets before the screw channels
were sandblasted with corundum (particle size: 50 µm), conditioned
with primer (ML Primer, Shofu) and sealed with veneering composite
(Ceramage Flowable Composite, Shofu), in order to prevent bacterial
leakage through the screw channel. Impressions of the palatal faces of
the specimens were produced using kneadable silicone (Provil novo,
Putty soft regular set, Heraeus Kulzer, Hanau, Germany). The im-
pressions were cut along the sagittal plane. These impressions had the
purpose of serving as index for the identification of possible plastic
deformations of the implant crowns resulting from the cyclic loading
tests. The bacterial tightness of the IAI under static, unloaded condi-
tions was verified by overmolding the test specimens with 50 °C warm,
liquid Kanamycin-Aesculin-Azid-Agar (KAAA, Oxoid Limited,
34
Journal of the Mechanical Behavior of Biomedical Materials 91 (2019) 32–37
Basingstoke United Kingdom). After solidification of the agar, the test
devices were wrapped in Parafilm (Pechiney Plastic Packaging, Chi-
cago, Illinois, USA) and incubated for 24 h at 37 °C. After incubation,
the agar was inspected as to the presence of black discolorations, which
would indicate bacterial leakage.
In brief, the indicator effect of the KAAA is based on the ability of E.
faecium to hydrolyse Aesculin. The products of this primary reaction
form black iron-phenole complexes with the contained iron, which are
clearly visible in the otherwise transparent agar. Furthermore, the agar
contains Kanamycin and sodium azide in order to inhibit any undesired
bacterial flora (Mossel et al., 1978). As a result, the agar is highly se-
lective for E. faecium.
If no discoloration of the agar could be observed in this static phase,
the agar was removed and the test specimens were further evaluated in
the subsequent dynamic masticatory simulation phase by cyclic
loading. In preparation, a short tube made of Plexiglas was glued to the
top of the implant mounting support in extension, in order to keep the
agar in shape and prevent it from drying out while at the same time
permitting observation of the IAI area. Once again, the specimens were
overmolded with warm KAAA. After solidification, 1 ML of NaCl 0.9%
was added to the agar, to prevent the agar from drying out. The
Plexiglas tube was sealed off at the top with Parafilm to prevent de-
hydration and contamination of the agar.
The test devices were clamped into a tabletop testing machine
(DYNA1dent, DYNA-MESS Prüfsysteme GmbH, Stolberg, Germany)
with an angle of 30° to the load axis as required by ISO 14801:2007.
The contact point to the test objects was set at the intercept between
implant axis and load axis near the cusp of the crown and was designed
in the form of a fin in order achieve reliable force transfer and prevent
horizontal slippage (Fig. 4).
The test objects were loaded with a frequency of 2 Hz and a force of
50 N, for a number of 1,200,000 cycles. The necessary incubation
temperature of 37 °C was maintained throughout the experiment by
aligning two fan heaters towards the test devices, which were in turn
temperature controlled through a water recirculation and heating
pump. A reliable optical assessment of the area of the IAI was made
possible by orienting a light source at the test devices from below, using
a glass fiber light conductor. Hereby, the translucency of the acrylate
core of the mounting support permitted a uniform illumination of the
agar in the area of the IAI.
During the experiment, a camera (D50, Nikon, Tokio, Japan)
equipped with a macro lens (AT-X PRO D, Tokina, Machida, Japan) was
oriented towards the test objects in order to record a possible color
change of the indicating agar as a response to bacterial contamination.
load axis implant axis
F
Indicator-agar
(KAAA) to detect
leakage of
E. faecium from
the IAI
Fig. 4. Test device clamped into test machine for the evaluation of bacterial
tightness of the IAI under cyclic masticatory loading.
A. Wachtel et al. Journal of the Mechanical Behavior of Biomedical Materials 91 (2019) 32–37

Controlled by a computer program (DCamCapture, Bernd Peretzke,

400
Ascheffel, Germany), the camera was set to record a photo once every
hour, corresponding to 7200 load cycles. The photos were exported to a 350

Bending moments [Ncm]

computer and matched to the numbers of load cycles.
300
The following technical complications were defined:
250
a) Loosening or fracture of the abutment screw
b) Repairable fracture limited to the composite veneer 200
c) Plastic deformation or fracture of the PEEK crown framework
150

The samples were recorded for 24 h after the end of the cyclic 100
loading phase to detect a possible delayed color change of the agar.
50
After removal of the indicating agar, the silicone indexes were ap-
plied to the palatal faces of the crowns and the combination of the two 0
was examined under a microscope (VHX-2000, Keyence, Osaka, Japan)
to determine the possible occurrence of plastic deformations. Fig. 5. Plot of the determined bending moments.
Subsequently, the test objects were clamped into a universal testing
machine (Z050, Zwick, Ulm, Germany) in order to measure the bending
moment of the specimens. The palatal surfaces of the crowns were
loaded at a point 8 mm above the implant shoulder with a fin-shaped
punch at an angle of 30° to the implant axis. A progression speed of
1 mm/min was set and the samples were loaded until fracture had oc-
curred and a subsequent force drop of 10% of the maximum value had
taken place.
The bending moment M was calculated by the equation M = 0.5 × l
× F outlined in ISO 14801:2007. Hereby l is defined as the distance of
the intercept between the implant axis and force axis on the one hand
and the bone crest on the other hand, which was 11 mm for the present
setup. F stands for the force determined from the experiment.
Microscopic pictures of the fractured specimens were recorded
using a digital optical microscope (Keyence VHX-2000, Osaka, Japan).
If no bacterial leakage could be detected during the cyclic loading
phase, the test specimens were disassembled after the static load test
and the implant interiors where wiped out with a little brush. The
brushes were put into KAAA and incubated for 24 h at 37 °C in order to Fig. 6. Damaged crown after bending moment testing.
verify presence of bacterial activity of E. faecium. After positive ver-
ification, samples of the KAAA surrounding the specimens during the
cyclic loading phase to detect bacterial leakage were added to these
bacterial cultures and incubated for another 24 h at 37 °C to verify the
proper functioning of the indicating agar. During the two incubation
periods, KAAA color changes were recorded every hour by the camera.
The statistical interpretation of the results was carried out with the
aid of the commercial software Origin-Pro 7.5G SR2 (OriginLab
Corporation, Northampton, MA, USA).

Results

During the masticatory loading simulation, none of the test speci-

mens exhibited any of the technical complications as defined above.
Similarly, none of the 10 test specimens exhibited bacterial leakage
at the IAI. At the same time, the functionality of the KAAA and the
bacterial activity of E. faecium extracted from the IAI after the final
mechanical test could be verified for each specimen.
A maximum bending moment of 352.13 ± 48.96 N·cm was calcu-
lated from the results of the static load test. The individually achieved
bending moments are represented in Fig. 5.
For the static load testing a typical fracture pattern could be ob-
served, which was characterized by a spalling of the palatal composite
veneer and a corresponding sagittal crack along the palatal side of the
PEEK framework ( Fig. 6). In one specimen the PEEK framework frac-
Fig. 7. Longitudinal section through specimen no. 7 after bending moment
tured horizontally in the area of the implant shoulder (Fig. 7). The same
testing.
specimen demonstrated the highest maximum load bearing capacity in
the bending moment test at 750.2 N. None of the specimens displayed
deformation or fracture of the implant itself or the abutment screw
(Fig. 7).

35
A. Wachtel et al.
Discussion
The results of the experiments confirmed the hypothesis that a
screw-retained single implant crown made of PEEK does not show
bacterial leakage when combined with a titanium implant by a conical
IAI.
When judging whether or not the implant crown in consideration is
suitable as final restoration with respect to its mechanical properties,
the masticatory forces naturally occurring in the oral cavity have to be
taken into account. Various findings regarding the extent of the mas-
ticatory forces have been published. One study, which aimed to de-
termine the masticatory forces acting on single teeth of young, fully
toothed individuals, can serve as valid reference point for comparison
(Ferrario et al., 2004). In that study, the highest values for upper in-
cisors were achieved by male individuals and amounted to
146 ± 44.44 N. This corresponds to a bending moment of about
80 N cm. For first molars, the maximum masticatory force was reported
for young male individuals at 306.07 ± 41.99 N, corresponding to a
bending moment of about 168 N cm. Both values of bending moment
are below the ones attained by the PEEK based crowns in the present
study (352.13 ± 48.96 N cm). Consequently, the tested implant crown
complex would be able to withstand the forces encountered in the oral
cavity when serving as a superstructure to an implant inserted as re-
placement up to the first molar.
Regarding the results of the previous study by Mühlemann et al., the
bending moments of the present study were generally lower
(Mühlemann et al., 2014). Compared to the present results, the next
higher bending moment of 464.9 ± 106.6 N cm was evaluated for a
group consisting of 1-piece zirconia abutments (“CARES abutments on
Bonelevel RC implants with implant diameter of 4.1 mm, Straumann,
Basel, Switzerland”) (Mühlemann et al., 2014). However, for another
group of 1-piece zirconia abutments (“Atlantis abutments on OsseoS-
peed implants with implant diameter of 4.5 mm, Astra Tech AB,
Mölndal, Sweden”), a bending moment of 216.4 ± 90 N cm was re-
ported (Mühlemann et al., 2014). This value was below the value of the
1-piece PEEK superstructure of the present study.
When evaluating the fracture modes ( Figs. 6 and 7), it became
apparent that due to the elastic properties of the PEEK framework the
titanium implant and abutment screw were protected from damage,
whereas the bending point was displaced more towards the coronal
portions. In practice, in the event of excessively high forces that exceed
the fracture resistance of the PEEK superstructure, the superstructure
can easily be remanufactured in short time via CAD/CAM manu-
facturing using the existing 3D data, while the implant itself can con-
tinue to serve its purpose. This is a considerable advantage compared to
much more rigid superstructures made of metal or zirconium oxide. The
behavior of metal and zirconium oxide abutments has been studied
under flexural loading, which has, by contrast, resulted in plastic de-
formation of the implant shoulder (Mühlemann et al., 2014; Sailer
et al., 2018).
Another possible advantage of prosthetic reconstructions made of
PEEK is their transparency to x-rays (Zoidis, 2018b). That way, artefacts
can be avoided. This facilitates the interpretation of radiographs, which
especially benefits three-dimensional x-ray methods.
PEEK offers patients a metal-free alternative to conventional pros-
theses made from dental alloys, which in certain cases can adversely
affect patient health (Elshahawy et al., 2009).
In theory, replacement of the rigid conventional superstructure
materials with elastic PEEK should lead to a reduction in the level of
masticatory forces imparted on the implant-bone interface. This as-
sumption could not be confirmed by a Finite Element Analysis (FEA)
study by Kaleli et al. (2018). It was not mentioned in the study how the
contacts between the individual components were defined. This how-
ever is a substantial factor when calculating FEA results and might be
the reason for this particular outcome.
Conical IAIs display the highest bacterial tightness compared to
36
Journal of the Mechanical Behavior of Biomedical Materials 91 (2019) 32–37
other IAI designs. Nevertheless, even conical IAIs tested with combi-
nations of conventional materials did not achieve absolute bacterial
tightness ((Mishra et al., 2017)). In contrast, an absolute bacterial
tightness under cyclic masticatory loading could be demonstrated in the
present study using a PEEK superstructure on conventional titanium
implants. This is most likely due to the high elasticity of the PEEK su-
perstructure, which brings about two benefits: (1) Primary bacterial
tightness through the close alignment of the contacting surfaces within
the IAI and (2) Secondary bacterial tightness during masticatory si-
mulation through the prevention of micromovements in the IAI due to
self-deformation of the superstructure.
The employed test bacterium E. faecium has the dimensions of
0.6–2.0 × 0.6–2.5 µm (Devriese and Pot, 1995). It is comparable in size
to that of conventional periodontal pathogens such as Prevotella inter-
media, Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis
(Callan et al., 2005) and Streptococcus mutans (Whittaker and Drucker,
1970). As no bacterial leakage could be detected in the present study, it
can be assumed that the IAI showed no gap or at least a gap smaller in
width than the test germ.
Ideally, the mechanical properties as well as the bacterial tightness
of the presented specimens would need to be tested during immersed
thermocycling at temperatures between 5 and 55 °C, as has previously
been performed in a study by Mühlemann et al. (2014). This was not
possible in the present study because that would have flushed away the
indicating agar. To achieve this, the test setup will have to be modified
accordingly.
Another limitation of the present study is the low number of n = 10
tested samples and the absence of control groups. Therefore, a higher
number of samples would serve to strengthen the statistical significance
of the results, especially in comparison with control groups.
Additionally, in vivo studies are necessary to prove the clinical long-
term success of such implant crowns.
Conclusion
Within the limitations of the present study, screw retained dental
implant superstructures made of PEEK display favorable properties with
respect to bacterial tightness of the IAI and fracture mode compared to
conventional implant superstructures. Further in vitro and in vivo stu-
dies will be necessary in order to substantiate the clinical significance of
these benefits.
Acknowledgements
The authors would like to thank Nobel Biocare AB, Göteborg,
Sweden for providing the implant components through Grant no 2012-
1143, and Klingel medical metal GmbH for the manufacturing of the
crown frameworks.
References
Al-Rabab'ah, M., Hamadneh, W., Alsalem, I., Khraisat, A., Abu Karaky, A., 2017. Use of
high performance polymers as dental implant abutments and frameworks: a case
series report. J. Prosthodont. https://doi.org/10.1111/jopr.12639.
Bodden, L., Lümkemann, N., Köhler, V., Eichberger, M., Stawarczyk, B., 2017. Impact of
the heating/quenching process on the mechanical, optical and thermodynamic
properties of polyetheretherketone (PEEK) films. Dent. Mater. 33, 1436–1444.
Callan, D.P., Cobb, C.M., Williams, K.B., 2005. DNA probe identification of bacteria co-
lonizing internal surfaces of the implant-abutment interface: a preliminary study. J.
Periodontol. 76 (1), 115–120.
Devriese, L.A., Pot, B., 1995. The genus Enterococcus. In: Wood, B.J.B., Holzapfel, W.H.
(Eds.), The Genera of Lactic Acid Bacteria. Blackie Academic & Professional, London,
pp. 327–367.
Elshahawy, W.M., Watanabe, I., Kramer, P., 2009. In vitro cytotoxicity evaluation of
elemental ions released from different prosthodontic materials. Dent. Mater. 25,
1551–1555.
Esposito, M., Hirsch, J.M., Lekholm, U., Thomsen, P., 1998. Biological factors con-
tributing to failures of osseointegrated oral implants. (II). Etiopathogenesis. Eur. J.
Oral Sci. 106, 721–764.
Ferrario, V.F., Sforza, C., Serrao, G., Dellavia, C., Tartaglia, G.M., 2004. Single tooth bite
A. Wachtel et al.
forces in healthy young adults. J. Oral Rehabil. 31, 18–22.
Goiato, M.C., Pellizzer, E.P., da Silva, E.V., Bonatto Lda, R., dos Santos, D.M., 2015. Is the
internal connection more efficient than external connection in mechanical, biolo-
gical, and esthetical point of views? A systematic review. Oral Maxillofac. Surg. 19,
229–242.
Hahnel, S., Wieser, A., Lang, R., Rosentritt, M., 2015. Biofilm formation on the surface of
modern implant abutment materials. Clin. Oral Implants Res. 26, 1297–1301.
Ioannidis, A., Cathomen, E., Jung, R.E., Fehmer, V., Hüsler, J., Thoma, D.S., 2017.
Discoloration of the mucosa caused by different restorative materials - a spectro-
photometric in vitro study. Clin. Oral Implants Res. 28, 1133–1138.
Jervøe-Storm, P.M., Jepsen, S., Jöhren, P., Mericske-Stern, R., Enkling, N., 2015. Internal
bacterial colonization of implants: association with peri-implant bone loss. Clin. Oral
Implants Res. 26, 957–963.
Johansson, P., Barkarmo, S., Hawthan, M., Peruzzi, N., Kjellin, P., Wennerberg, A., 2018.
Biomechanical, histological, and computed X-ray tomographic analyses of hydro-
xyapatite coated PEEK implants in an extended healing model in rabbit. J. Biomed.
Mater. Res A 106, 1440–1447.
Kaleli, N., Sarac, D., Külünk, S., Öztürk, Ö., 2018. Effect of different restorative crown and
customized abutment materials on stress distribution in single implants and periph-
eral bone: a three-dimensional finite element analysis study. J. Prosthet. Dent. 119,
437–445.
Larrucea Verdugo, C., Jaramillo Núñez, G., Acevedo Avila, A., Larrucea San Martín, C.,
2014. Microleakage of the prosthetic abutment/implant interface with internal and
external connection: in vitro study. Clin. Oral Implants Res. 25, 1078–1083.
Liu, Y., Wang, J., 2017. Influences of microgap and micromotion of implant-abutment
interface on marginal bone loss around implant neck. Arch. Oral Biol. 83, 153–160.
Magne, P., Silva, M., Oderich, E., Boff, L.L., Enciso, R., 2013. Damping behavior of im-
plant-supported restorations. Clin. Oral Implants Res. 24, 143–148.
Mangano, C., Iaculli, F., Piattelli, A., Mangano, F., 2015. Fixed restorations supported by
Morse-taper connection implants: a retrospective clinical study with 10-20 years of
follow-up. Clin. Oral Implants Res. 26, 1229–1236.
37
Journal of the Mechanical Behavior of Biomedical Materials 91 (2019) 32–37
Mishra, S.K., Chowdhary, R., Kumari, S., 2017. Microleakage at the different implant
abutment interface: a systematic review. J. Clin. Diagn. Res. 11, ZE10–ZE15.
Mossel, A.A., Bijker, P.G., Eelderink, I., 1978. Streptococci of lancefield groups A, B and D
and those of buccal origin in foods: their public health significance, monitoring and
control. Soc. Appl. Bacteriol. Symp. Ser. 7, 315–334.
Mühlemann, S., Truninger, T.C., Stawarczyk, B., Hämmerle, C.H., Sailer, I., 2014.
Bending moments of zirconia and titanium implant abutments supporting all-ceramic
crowns after aging. Clin. Oral Implants Res. 25, 74–81.
Pessoa, R.S., Sousa, R.M., Pereira, L.M., Neves, F.D., Bezerra, F.J., Jaecques, S.V., Sloten,
J.V., Quirynen, M., Teughels, W., Spin-Neto, R., 2017. Bone remodeling around im-
plants with external hexagon and morse-taper connections: a randomized, controlled,
split-mouth, clinical trial. Clin. Implant Dent. Relat. Res. 19, 97–110.
Sailer, I., Asgeirsson, A.G., Thoma, D.S., Fehmer, V., Aspelund, T., Özcan, M., Pjetursson,
B.E., 2018. Fracture strength of zirconia implant abutments on narrow diameter
implants with internal and external implant abutment connections: a study on the
titanium resin base concept. Clin. Oral Implants Res. 29, 411–423.
Truninger, T.C., Stawarczyk, B., Leutert, C.R., Sailer, T.R., Hämmerle, C.H., Sailer, I.,
2012. Bending moments of zirconia and titanium abutments with internal and ex-
ternal implant-abutment connections after aging and chewing simulation. Clin. Oral
Implants Res. 23, 12–18.
Wachtel, A., Zimmermann, T., Spintig, T., Beuer, F., Müller, W.D., Schwitalla, A.D., 2016.
A novel approach to prove bacterial leakage of implant-abutment connections in
vitro. J. Oral Implantol. 42, 452–457.
Whittaker, D.K., Drucker, D.B., 1970 NovN. Scanning electron microscopy of intact co-
lonies of microorganisms. J. Bacteriol. 104, 902–909.
Zoidis, P., 2018a. Polyetheretherketone overlay prosthesis over high noble ball attach-
ments to overcome base metal sensitivity: a clinical report. J. Prosthodont. https://
doi.org/10.1111/jopr.12747.
Zoidis, P., 2018b. The all-on-4 modified polyetheretherketone treatment approach: a
clinical report. J. Prosthet. Dent. 119, 516–521.