Wolf-Dieter Mueller

Ulrich Gross
Thomas Fritz
Christian Voigt
Peter Fischer
Georg Berger
Sigfried Rogaschewski
Klaus-Peter Lange

Authors affiliations:
Wolf-Dieter Mu eller, Peter Fischer, Klaus-Peter
Lange, Dental School, Medical Faculty Charite,
Humboldt-University of Berlin, Berlin, Germany
Ulrich Gross, Thomas Fritz, Institute of Pathology,
Christian Voigt, Department of Traumatology and
Recontructive Surgery,
Klinikum B. Franklin, Free University of Berlin,
Berlin, Germany
Georg Berger, Federal Institute for Material
Research and Testing, Berlin, Germany
Sigfried Rogaschewski, Institute of Physics,
Humboldt-University of Berlin, Berlin, Germany
Correspondence to:
Dr. Wolf-Dieter Mueller
Zentrum fur Zahnmedizin
Medizinische Fakultat Charite
Humboldt-Universitat zu Berlin
Fohrer Str. 15, D 13353 Berlin
Germany
Tel.: +49 30 4505 62074
Fax: +49 30 4505 62920
e-mail: wolf-dieter.mueller@charite.de

Date:
Accepted 18 January 2002
To cite this article:
Mueller W-D, Gross U, Fritz T, Voigt C, Fisher P, Berger
G, Rogaschewski S, Lange K-P. Evaluation of the
interface between bone and titanium surfaces being
blasted by aluminium oxide or bioceramic particles
Clin. Oral Impl. Res, 14, 2003; 349356

Copyright r Blackwell Munksgaard 2003

ISSN 0905-7161

Evaluation of the interface between bone

and titanium surfaces being blasted by
aluminium oxide or bioceramic particles

Key words: dental implants, surface modification, animal experiment, bioceramics

Abstract: The surface structure, in particular the surface roughness, and the surface
chemistry of titanium implants influence their anchoring in bone. The aim of this study
was to analyse metalbone contact (MBC) after modification of the implant surface, using
different materials for blasting. The surface modification of titanium was produced by
blasting it with particles made of Al2O3 or bioceramics. The biological effects were then
investigated experimentally using 27 rabbits, analysed after 7, 28 and 84 days after the
implantation of titanium cylinders treated accordingly. The MBC showed a tendency for
more bone after bioceramics were used as a blasting material, compared to Al2O3.

The quality of an interface is determined by

the chemical, physical and mechanical
properties involved, including the topographic characteristics o f the contacting
faces. The complete characterization of the
surface properties of metallic implants is
necessary (Kasemo & Lausmaa, 1988) to
compare the function of the implant with
its surface characteristics and to control the
effects of the manufacturing process. The
chemical composition, contamination at
the surface, and thickness and structure of
the surface oxide layer all contribute to the
biocompatibility of an implant. The release
of ions from the implant is closely correlated to the chemical composition of the
surface, its roughness and thickness of the
passivation oxide layer. The chemical
composition of implant surfaces is often
different from the bulk of the material and
depends on the processing and surface
treatment of the material. Surface layers
can contain reactive structures, they can

react with water molecules or different ions

in solution in the vicinity of the surface,
and influence the binding of proteins to the
surface (Williams et al. 1990). These
conditions influence subsequent cell reactions (Baier & Meyer 1988; Smith et al.
1991, Smith 1993). Comparative analyses
of different implant surfaces by Wennerberg
(1996) resulted in a description of parameters that could influence the quantity of
the metalbone contact (MBC). Different
diameters of particles used as blasting
material did not influence the MBC dramatically. The highest amount of bone at
the interface was found with hydroxyapatite (HAP)-coated Ti implants in comparison to blasted or untreated, as machined,
surfaces (Cook et al. 1987; Wennerberg
1996). The disadvantage of HAP-covered
implants is the low bonding strength between the HAP layer and the Ti bulk
material. Tiller (1985), Lotz & Fleischfresser
(1986) and Wennerberg (1996) demonstrated

349

Mueller et al . Evaluation of the interface between bone and titanium

that particles used in the blasting of surfaces

can persist on these blasted surfaces.
The objectives of this investigation were
to assess the usage of bioceramics particles
versus Al2O3 particles as a blasting material. To achieve this, cylinders made of c.p.
titanium were treated accordingly, implanted in a rabbit model, analysed morphometrically for MBC and then evaluated
statistically.

Material and methods

For the animal experiments, cylinders of
c.p. titanium (8 mm in length and 4 mm in
diameter) were used. Three different surfaces were prepared:
1. Control specimens were left untreated.
2. The surface was blasted with Al2O3
particles (carborundum) with a particle
size of 110 mm, called Al2O3 surfaces.
3. The surface was blasted with particles of
a mixture of bioceramics with a particle
size of 50 mm, referred to as GB surface.
The blasting procedures were realised in
a chamber with a pressure of 4 bar and a
distance of 1 cm between tube and surface.
For equal treatment, the cylindrical specimens were rotated and moved along the
length. A schematic representation is
shown in Fig. 1.
The bioceramic mixture contained 50%
of a bio-vitro-ceramic (AP40) and of a
resorbable glass ceramic (GB14). The glass
ceramic GB14 with a composition of (in
weight percentages) CaO 30.67, P2O5
43.14, NaO 9.42, K2O 9.42, MgO 2.45
contained as its main crystal phase Ca2KNa(PO4)6 (Berger et al. 1995). AP40
contained (in weight percentages) CaO

Fig. 1. Schematic of surface treatment.

350 |

Clin. Oral Impl. Res. 14, 2003 / 349356

31.89, P2O5 11.21, K2O 0.19, Na2O 4.6,

MgO 2.82, CaF2 4.99 and SiO2 44.3 and
displayed apatitic phases with a mixture of
fluorapatite and hydroxyapatite as well as
wollastonite (calcium silicate) (Berger et al.
1989). The different components provided
the basis for different structures and differences in the solubility of the components.
Therefore, GB14 can be called a resorbable
glass ceramic and AP40 a more stable ceramic.
A total of 63 implants were used, divided
into three groups of 21 specimens, according to their three different surface modifications as detailed above. For each time
interval six specimens were implanted,
two per animal. Three specimens of each
group were used for surface analysis of the
initial conditions by measuring the roughness, scanning electron microscopy (SEM)
and energy-dispersive X-ray analysis
(EDXA). The surface roughness was measured with the help of a mechanical
contacting instrument, Perthometer S6P
(Corp. Feinpruf Perthen GmbH, Gottingen, Germany), three times on three specimens and to a length of 5 mm. The fivepoint height of the profile (Rz), Rz 1/
5S(Rmax), the arithmetic mean deviation of
the profile (Ra), Ra 1/nS(y(xi)), and the
maximal roughness depth (Rmax) were detected.
Twenty-seven female adult Chinchilla
rabbits, strain CH bb-CH, body weight 3
4 kg, were anaesthetized by an intramuscular injection of a mixture of 6 ml Ketanests (ketamine hydrochloride) and 3 ml
Rompuns (xylazine hydrochloride) and
operated under sterile conditions bilaterally
using the distal femur epiphysis as the
implantation site. The animals had closed
epiphysial lines. The knee joint was opened
by a medial approach, the patella luxated
and a hole with a diameter of 3.95 mm

drilled sagittally into the epiphysis below

the patella sliding plane by using a diamonded hollow cylinder burr with inner
cooling by physiological NaCl solution
being pumped through the hollow cylinder.
The implant cylinder was inserted and the
patella repositioned, the fasciae and skin
carefully sutured. For infection prophylaxis
the animals received an antibiotic (Refobacin 80, 0.5 and Aubikal 0.5 ml) postoperatively.
After 7, 28 and 84 days, the animals were
anaesthetized by an overdose as mentioned
above, and sacrificed. The samples were
excised with the surrounding bone and
cartilage, and immersed in 5% (v/v) formaldehyde solution buffered according to
Lillie (pH 7.3) for more than 3 days. The
specimens were dehydrated in a series of
ethanols 50, 70, 80, 96, and twice 100% for
3 days each step, defatted in etherchloroform, soaked in polymethylmethacrylate
monomer and polymerized into blocks.
Sections were made from these blocks
using a saw with a rotating diamonded disk
(Leitz 1600, Leitz, Wetzlar, Germany). The
orientation of the blocks enabled three
sections to be gained from the middle of
the cylinder in a sagittal plane, and other
sections in a frontal plane, the thickness of
the sections being in the order of 70 mm.
The sections were polished and stained
using Giemsa staining solution, and von
Kossa/Paragon staining solution, i.e. with
fuchsin and toluidine blue. Only the superficial 7 mm, i.e. the surface layer, of the
sections was stained. The stained sections
were analysed using a conventional light
microscope for transmitted or reflected
light. The various tissues at the interface
between the metal and the surroundings
(bone, osteoid, chondroid and soft tissue)
were determined morphometrically by
using equipment described elsewhere
(Gross 1999). In brief, a computer-assisted
workstation with a measuring tablet connected to a microscope and adapted PC
software (Systec, Berlin, Germany) were
used. Images were linked to the PC via a
generic video camera and a Rausch (Munich, Germany) graphic board. The measurements were carried out with constant
magnification,
100, of the stained sections.
To quantify the tissues at the interface,
the length of the interface was determined
(UI) and the other tissues, i.e. bone, osteoid

Mueller et al . Evaluation of the interface between bone and titanium

and chondroid, were recorded in percentages of the interface. The values for MBC
were given in percentages:


lMBC

100
MBC%
Ul
where lMBC is the length of bone contacts to
the implant surface and UI whole length
of the interface.

Results
The modified surfaces of the implant
cylinders prior to implantation are depicted
in Figs. 24. In the case of modified
surfaces, differential EDXA spectra were
recorded in relation to untreated surfaces.
Fig. 2a depicts the typical morphology of an

untreated surface in SEM. The parallel

grooves visible are due to the machining
of the cylinders. The surface contained a
higher than usual amount of nitrogen (N2),
which was probably chemically titanium
nitrite (Fig. 2b). The differences in the
roughness of the treated surfaces are
obvious, i.e. in between the treated surfaces
and compared to the untreated cylinders
(Fig. 5). The EDX analyses, performed
over an area of 120
90 mm, indicate
the existence of blasting particles. The
signals in the EDXA differential spectra
(different spectra between a treated and an
untreated Ti surface) are characteristic for
the particles. The signal for aluminium in
Fig. 4b characterizes Al2O3 (carborundum).
The calcium (Ca) and phosphorous (P)

Fig. 2. (a) SEM picture of untreated, as-machined Ti surface. (b) EDX spectra of the
same surface.

signals in Fig. 3b characterize the mixture

of bioceramics. The main values and the
standard deviation of Rz, Ra, Rmax are
shown in Fig. 5. The Al2O3-treated
samples display a roughness greater by
a factor of 2 than the GB14-treated
samples.
The implantbone contact was analysed
after 7, 28 and 84 days postoperatively.
After 7 days the wound healing and
bone regeneration was not advanced enough to perform morphometric measurements. Therefore, measurements were
only carried out after 28 and 84 days
postoperatively. Five sections from each
sample were used to measure the
mean values. After 28 days (Fig. 6)
the Al2O3-treated specimens displayed

Fig. 3. (a) SEM picture of bioceramic (GB14/AP40) blasted Ti surface. (b) EDX
spectra of bioceramic (GB14/AP40) blasted Ti surface compared to untreated Ti
surface (background, light grey).

351 |

Clin. Oral Impl. Res. 14, 2003 / 349356

Mueller et al . Evaluation of the interface between bone and titanium

Fig. 5. Maximum roughness (Rmax), mean roughness (Rz), and average roughness
(Ra) of modified Ti surfaces, box plots (KruskalWallis) showing 2575 percentiles,
extremes, median and arithmetic mean.

Fig. 4. (a) SEM picture of Al2O3 blasted Ti surface. (b) EDX spectra of Al2O3 blasted
Ti surface compared to untreated surface (background, light grey).

Fig. 6. Metalbone contact after 28 days as a function of the surface modification;

differences (Pr0.10) between Al2O3-treated and bioceramic-treated as well as
untreated specimens. Mean standard deviation.

352 |

Clin. Oral Impl. Res. 14, 2003 / 349356

Fig. 7. Metalbone contact after 84 days as a function of the surface modification;

differences between untreated and Al2O3-treated specimens (Pr0.15) as well as
bioceramic-treated specimens (Pr0.10). Mean standard deviation.

Mueller et al . Evaluation of the interface between bone and titanium

Fig. 8. As-machined Ti implant, sections after 84 days (v. Kossa/Paragon stain): (a) overview, position of the
implant; (b) thin frame of bone; (c) different density of soft tissue, only some macrophages at the interface; no
attachment between bone and implant surface; border of a cyst-like structure in the upper part of this figure; (d)
sagittal cut, thin frame of bone.

significantly more MBC than the two

other surfaces. After 84 days (Fig. 7) all
samples showed an increase in the MBC.
The increase of MBC from 8% to 63 %
with the GB-treated samples is the
most important finding. Statistically, the
increased MBC of the GB-modified surfaces is significant on a level of 85%
(Pr0.15) in relation to the Al2O3-treated
surfaces.

The histological findings at the three

different surface modifications are also
characteristic after 84 days (Figs. 810). In
the surroundings of the untreated cylinder,
a dense ring of bone was developed similar
to cortical bone with regard to density and
development of Haversian canals (Fig. 8a,b
and d). The thickness of this bone ring is up
to 500 mm. The implant surface was partially covered by mineralized bone (Fig. 8b).

Some areas also displayed a small seam of

not fully mineralized bone. Occasionally,
there was a tissue regenerate with a
metachromatic extracellular matrix between the regenerating cells indicating
chondroid, bone or cartilage. Otherwise,
some loose or medium dense fibrous tissue
developed, in some places separating from
the metal surface like a cyst, i.e. a fluidfilled space with some fibres at the border
to loose soft tissue (Fig. 8c). In some places
there were also fat cells and haematopoietic
cells visible.
In the surroundings of the cylinders
treated with GB14/AP40 (Fig. 9a), there is
a rather dense bone ring with a thickness of
approximately 300 mm. From this ring
some smaller foot-like processes of bone
extend to the surface at the cylinder (Fig.
9bd) where a smaller bone ring covering
the implant surface was present. These
trabeculae show surface morphologies comparable to Bauds curves (Mattheck, 1990),
indicating a homogeneous surface stress
state, i.e. avoiding clefts which are sites of
increasing risk for fractures. The mineralization of the regenerated bone was
up to the metal surface. Therefore, load
transfer from the implant to the bone is
enabled.
A partial bone ring with a thickness of
approximately 200 mm at the cylinders
treated with Al2O3 could be found.
Furthermore, some foot-like attachments
of the bone to the metal surface (Fig. 10ac)
were visible, and in between the trabeculae
loose connective tissue with fat cells,
haematopoietic cells and a few macrophages were also in evidence. In contrast
to the bone structure around the bioceramic-modified implant surface, in this case
the mineralization of the bone was homogeneous between the pre-existing lamellar
structures and the metal surface. An
apposition of new to old bone could be
observed. Load transmission could be assumed.

Discussion
In this study the different surface roughnesses and surface compositions influenced
the MBC. During the experiment two
phases could be observed: one in the
development of the MBC at the Al2O3,

353 |

Clin. Oral Impl. Res. 14, 2003 / 349356

Mueller et al . Evaluation of the interface between bone and titanium

Fig. 9. GB14/AP40 blasted implant after 84 days (v. Kossa/Paragon stain): (a) overview, position of the implant;
(b) bone frame with protuberances to the implant surface; (c) bone and soft tissues in the interface, arrow shows
Bauds curve; (d) new trabecular bone (n) in connection with old trabeculae (o) in the bone frame around the
implant.

and the second at the bioceramic-treated

surfaces.
The initial phase was observed in the
first 28 days. Thereafter, during 8 weeks up
until the 84th day postoperatively, a phase
of consolidation was seen. During the first
4 weeks the surface roughness influenced
the MBC positively, as shown by the
Al2O3-treated surfaces. This observation
confirms the findings of Wennerberg. The

354 |

Clin. Oral Impl. Res. 14, 2003 / 349356

increase of the MBC as given in Fig. 11

corresponded with a rate of 1 for bone
development as given in the equation
 
DMBC %
vMBC
Dt
day
rate of bone development

This means that after 30 days, bone had

covered 30% of the implant surface.

It was surprising that the MBC at the

bioceramic-treated surfaces and the untreated control surfaces was small (Fig. 6)
in this initial period of up to 28 days. At the
surface of the bioceramic-treated surfaces,
with little difference regarding roughness in
respect to Al2O3-treated surfaces, more
bone was expected since the release of Ca
ions from impacted bioceramic particles
was assumed.
During the second phase, effects that
were apparently related to the different
composition of the surface could be seen,
otherwise the increase of the MBC of
bioceramic-treated surfaces could not be
explained. The ion release from the implant
surface (Pischzik et al. 2000) is assumed to
be a stimulus for bone cell multiplication in
the surroundings of the implant. A positive
influence of Bioglasss 45S5 eluate on
growth and differentiation of undifferentiated stem cells or osteoblast precursors
was described recently (Xynos et al. 2001).
Since the elements in the bioceramic
particles (GB14AP40) are close to those in
the mentioned Bioglasss45S5, the mentioned interpretation is supported. The
many points of bone regeneration on the
surface of old trabecular bone (Fig. 9) can be
explained accordingly. Kohler & Retemeyer (1981) characterized AP40 in vitro
and in animal experiments. The biocompatibility of GB14 was investigated in cell
culture by using fibroblasts (Picker et al.
1993) and osteoblasts (Knabe et al. 1998),
and in vivo using rabbits (Muller-Mai et al.
1997).
In this context, an increase of the surface
roughness due to the dissolution of impacted bioceramic particles could be
taken into consideration but not proven
by the mentioned measurements of the
surface roughness. The rates of bone development were compiled and are shown in
Table 1.
The blasting of titanium surfaces using
bioceramic particles provides an interesting
variant for the development of an increased
implant bone interface as shown. Furthermore, the study could be repeated with
implants of the same surface roughness and
morphology after bioceramic particles
treatment as after Al2O3 particles treatment. It can be speculated that under
conditions of a rougher surface, treated
with bioceramics, the rate of MBC could
be 1%/day during the first phase too. This

Mueller et al . Evaluation of the interface between bone and titanium

study could not determine how much MBC

is necessary for an optimal long-term
function of a dental implant.

Conclusions
This investigation demonstrates that bone
growth around implants depends on the
topography and composition of an implant
surface customized by the manufacturing
technique. Initially the effect of the roughness is more dominant than a change of
composition attained by impacting or blasting particles for the MBC. Possible changes
to the surface roughness and composition
participate in the development of the
implantbone interface. The application of
bioceramic particles as a blasting material
for the treatment of implant surfaces seems
to be one way of creating a bioreactive
surface. The elucidation of the molecular
mechanisms behind the positive influence
of bioceramic particles on bone development remains a topic for additional investigations.

Resume
La structure de surface, en particulier la rugosite de
surface et la chimie de surface des implants en titane
influencent leur ancrage dans los. Le but de cette
etude a ete danalyser le contact os-metal (MBC)
apre`s modification de la surface implantaire par
lutilisation de differents materiaux pour le soufflage.
La modification de surface du titane a ete produite en
le souflant avec des particules faites dAl2O3 ou de
bioceramique. Les effets biologiques ont ensuite ete
etudies experimentalement a` laide de 27 lapins
analyses apre`s 7, 28 et 84 jours apre`s limplantation
de cylindres en titane traites comme indique cidessus. Le MBC saccompagne dun contact osseux
plus important apre`s lutilisation de la bioceramique
comme materiel de soufflage compare a` Al2O3.
Fig. 10. Al2O3 blasted implant after 84 days (v. Kossa/Paragon stain): (a) overview, position of the implant; (b)
trabecular bone with protuberances to the interface; (c) bone and soft tissue with some macrophages at the
interface; (d) bone with foot-like attachment, and soft tissue.

Table 1. Rate of bone development in phases 1 and 2n

Surface treatment

vMBC, phase 1
[%/day]

vMBC, phase 2
[%/day]

Untreated as machined
Bioceramic particles blasting
Al2O3 particles blasting

0.25
0.25
1

0.25
1
0.35

The rate of bone growth was determined in Chinchilla rabbits.

Zusammenfassung
Die Untersuchung der Verbindungzwischen Knochenund der Titanoberflache, dieentwe der mit
Aluminium oxido der biokeramischen Partikelns
andge strahltworde nwaren . Die Oberflachen
struktur, insbeson der eaber die Oberflachen rauhigkeitund die chemische. Zusammen setzung der
Oberflache von Titanim plantaten beeinflusse nihre
V e r a n k e r u n g i m Knochen. Das Zieldieser
Studiewareine Analyse des Metall-Knochen-Kontaktes (MBC) nach Veran der ung der Implantat
oberflache durch Sandstrahl ung mit verschiedenen
Materialien. Die Oberflachen ver an der ungdes
Titanserfolgte dur ch Sandstrahlungmit Al2O3-Partikelnodermit Biokeramikpartikeln. Der biologische

355 |

Clin. Oral Impl. Res. 14, 2003 / 349356

Mueller et al . Evaluation of the interface between bone and titanium

hueso-implante (MBC) tras la modificacion de la

superficie del implante, usando diferentes materiales
para pulverizar. La modificacion de la superficie de
titanio se produjo pulverizando partculas de Al2O3 o
bioceramica. Los efectos biologicos se investigaron
experimentalmente usando 27 conejos, analizados
tras 7, 28 y 84 das tras la implantacion de cilindros
de titanio tratados previamente. EL MBC mostro una
tendencia para mas hueso tras el uso de bioceramica
como material de pulverizacion, comparado con
Al2O3.

Fig. 11. Development of the MBC as a function of time for three different implant surface modifications.
Mean standard deviation.

Einflusswurdean 27 Kaninchenuntersucht, dievor7,

28 und 84 Tageneinenentsprechendbehandelten Titanzylin der implantierterhaltenhatten. Der MBC
zeigteeineten den ziellverstarkte Knochena nlage
rungen, wenndie Biokeramikanstelle des Al2O3 zur
Sandstrahlung ver wendet wordenwar.

Resumen
La estructura de la superficie, en particular la
rugosidad superficial y la qumica superficial de los
implantes de titanio influyen su anclaje en el hueso.
La intencion de este estudio fue analizar el contacto

References
Baier, R.E. & Meyer, A.E. (1988) Implant surface
preparation. International Journal of Oral and
Maxillofacial Implants 3: 920.
Berger, G., Gildenhaar, R. & Ploska, U. (1995) Rapid
resorbable glassycrystalline materials on the basis
of calcium alkali orthophosphate. Biomaterials
16: 12411248.
Berger, G., Sauer, R., Steinborn, G., Wihsmann,
F.G., Thieme, V., Kohler, ST. & Dressel, H.
(1989) Clinical application of surface reactive
apatite/wollastonite containing glass-ceramics.
In: XV. International Congress on Glass, Leningrad, USSR, 37 July 1989, Vol 3a, 120126.
Cook, S.D., Kay, J.F., Thomas, K.A. & Jarcho, M.
(1987) Interface mechanics and histology of
titanium and hydroxylapatite-coated titanium for
dental implant applications. International
Journal of Oral and Maxillofacial Implants 2:
1522.
Gross, U.M. (1999) Quantitative morphology of the
implantbone interface. In: Recum, A.V., ed.
Handbook of biomaterials evaluation, 739742.
Philadelphia: Taylor and Francis.
Kasemeo, B. & Lausmaa, J. (1988) Biomaterial and
implant surfaces: On the role of cleaning, contamination, and preparation procedures. Journal of
Biomedical Materials Research 22: 145158.
Knabe, C., Ostapowicz, W., Radlanski, R., Gildenhaar, R., Berger, G., Fitzner, R. & Gross, U. (1998)
In vitro investigation of novel calcium phosphates

356 |

Clin. Oral Impl. Res. 14, 2003 / 349356

using osteogenic cultures. Journal of Materials

Science: Materials in Medicine 9: 337345.
Kohler, St. & Retemeyer, K. (1981) Untersuchungen zur Biokompatibilitat und Bioaktivitat
zahnarztlicher Implantatmaterialien unter besonderer Berucksichtigung der Bio-Vitrokeramik
Ap40. Habilitationsschrift, Humboldt-Universitat zu Berlin.
LeGeros, J.P., Daculsi, G. & LeGeros, R.Z.
(1998) Tissue response to grit-blasted Ti-alloy,
Proceedings of the 24th Annual Meeting of the
Society for Biomaterials, San Diego, 229.
Lotz, J. & Fleischfresser, C. (1986) Einflu der
Metalloberflachenvorbehandlung auf die Festigkeit des NEMPlastVerbundes, Diploma, Humboldt-University of Berlin.
Mattheck, C. (1990) Engineering components grow
like trees. Mat.-Wiss. U. Werkstofftechnik 21:
143168.
Muller-Mai, C., Berger, G., Voigt, C., Bakki, B. &
Gross, U. (1997) The bony reaction to rapidly
degradable glass-ceramics based on the new phase
Ca2KNa(PO4)2. In: Sedel, L. & Rey, C., eds.
Bioceramics, Vol. 10, 5356. Amsterdam: Elsevier
Science Ltd.
Picker, H.-U., Arps, H., Kohler, St. & Berger, G.
(1993) Zytotoxizitatsprufung von resorbierbaren
glaskeramischen
Knochenersatzmaterialien.
Zeitschrift fur Zahnarztliche Implantologie 9:
111116.

Pischzik, L., Muller, W.-D., Rogaschewski, S.,

Lange, K.-P. & Berger, G. (2001) Journal of Dental
Research 80: 1245, Abstract 356.
Smith, D.C. (1993) Dental implants: Materials and
design consideration. International Journal of
Prosthodontics 6: 106117.
Smith, D.C., Pillar, R.M. & Chernecky, R. (1991)
Dental implant materials. I. Some effects of
preparative procedures on surface topography.
Journal of Biomedical Materials Research 25:
10451068.
Tiller, H.-J. (1985) Der Sandstrahlproze und seine
Einwirkung auf den Oberflachenzustand von
Dentallegierungen. Quintessenz 36: 19271934
(part 1): 21512158 (part 2)..
Wennerberg, A. (1996) On surface roughness and
implant incorporation. Thesis, Goteborg.
Williams, R.L., Higgins, S.J., Hamnett, A. &
Williams, D.F. (1990) The Characteristics of
protein adsorption onto metallic biomaterials. In:
Heimke, G., Soltesz, U. & Lee, A.J.C., eds.
Clinical implant materials. advances in biomaterials, Vol. 9, 327332. Amsterdam: Elsevier
Science Publishers B.V.
Xynos, I.D., Edgar, A.J., Buttery, L.D.K., Hench,
L.L. & Polak, J.M. (2001) Gene-expression profiling of human osteoblasts following treatment with
the ionic products of Bioglasss 45S5 dissolution.
Journal of Biomedical Materials Research 55:
151157.