Removal Torque Values of Titanium Implants

in the Maxilla of Miniature Pigs

Daniel Buser, DDS, Prof Dr med dent*/Thomas Nydegger, Dipl Ing HTL**/
Hans Peter Hirt, DDS, Dr med dent***/David L. Cochran, DDS, PhD****/
Lutz-Peter Nolte, PhD*****

The purpose of this study was to compare side-by-side two different titanium screw-type implants in the maxillae
of miniature pigs. The test implants had a machined and acid-etched surface (Osseotite) whereas the control
implants were sandblasted and acid-etched (SLA). After 4, 8, and 12 weeks of healing, removal torque testing was
performed to evaluate the shear strength of the bone-implant interface for both implant types. The results
demonstrated significant differences between both implant types (P < .01). Osseotite implants revealed mean
removal torque values (RTV) of 62.5 Ncm at 4 weeks, 87.6 Ncm at 8 weeks, and 95.7 Ncm at 12 weeks of healing.
In contrast, the SLA implants demonstrated mean RTV of 109.6 Ncm, 196.7 Ncm, and 186.8 Ncm at correspond-
ing healing periods. The mean RTV for SLA implants was 75% to 125% higher than for Osseotite implants up to 3
months of healing.
(INT J ORAL MAXILLOFAC IMPLANTS 1998;13:611–619)

Key words: interface shear strength, removal torque values, Osseotite surface, sandblasted and acid-etched surface,
SLA surface, titanium implants

O ver the past 30 years, the use of osseointegrated

implants has become a scientifically accepted
and well-documented treatment modality for the
osseointegrated implants was reported in clinical
studies using CP titanium implants.6–14 Other clinical
studies, however, reported increased failure rates in
rehabilitation of completely and partially edentulous areas with low bone density or reduced bone height,
patients. This development is based on fundamental such as in the posterior maxilla, especially for screw-
studies by the research teams of Brånemark et al1,2 type implants with a machined surface.15–18 There-
and Schroeder et al3–5 of commercially pure (CP) fore, attempts have been made over the past 12 years
titanium implants. Successful long-term stability of to improve bone anchorage of dental implants.
Thomas and Cook 19 examined the variables that
could potentially influence implant anchorage in
bone. Of 12 parameters examined, only the implant
*****Associate Professor, Department of Oral Surgery, School of
Dental Medicine, University of Berne, Berne, Switzerland.
surface had a significant effect on bone integration.
*****Research Assistant, Division of Orthopedic Biomechanics, This observation has been confirmed over the past 10
Maurice E. Müller Institute for Biomechanics, University of years in a series of in vivo studies evaluating different
Berne, Berne, Switzerland. surface configurations of titanium implants both in
*****Assistant Professor, Department of Oral Surgery, School of the long bones and in the jaws.20–23 Among the tested
Dental Medicine, University of Berne, Berne, Switzerland.
*****Professor and Chair, Department of Periodontics, University
surfaces, a new sandblasted and acid-etched surface
of Texas Health Science Center, San Antonio, Texas. (SLA) consistently showed the best results both in
*****Division Head of Orthopedic Biomechanics, Maurice E. histometric and biomechanical testing. A recent
Müller Institute for Biomechanics, University of Berne, removal torque study by Buser et al24 in the maxillae
Berne, Switzerland. of miniature pigs compared this SLA surface with the
Reprint requests: Prof Dr Daniel Buser, Department of Oral
two best-documented titanium surfaces in implant
Surgery, University of Berne, Freiburgstrasse 7, CH-3010 Bern, dentistry, the machined and the titanium plasma-
Switzerland. E-mail: [email protected] sprayed (TPS) surface. The study confirmed that the

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF The International Journal of Oral & Maxillofacial Implants 611
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
Buser et al

Berne guidelines for the care and use of laboratory

animals were followed.
Implant Shapes and Surfaces. Two different CP
titanium screw-type implants with different shapes
and surfaces were placed in edentulous areas of the
anterior maxillae of miniature pigs. Both implant types
are used clinically in patients (Fig 1). For the test
implant, a standard, 3.75-mm, self-cutting, screw-type
implant with a length of 10 mm and four grooves in
the apical portion was used. This implant has a stan-
dard hex on top and is characterized by a hybrid
design, a short, approximately 2- to 3-mm-long
machined surface in the crestal area, and an acid-
Fig 1 The two tested titanium implants: the 8-mm-long SLA etched (sulfuric acid–hydrochloric acid) surface over-
implant without apical grooves, and the 10-mm-long Osseotite all (Osseotite, Implant Innovations, Palm Beach Gar-
implant with four apical grooves.
dens, FL). For the control group, an 8-mm-long
solid-screw implant (4.05 mm diameter), character-
ized by a sandblasted (large grit of 250 to 500 µm) and
machined surface had 8 to 10 times lower removal acid-etched (sulfuric acid–hydrochloric acid) (SLA)
torque values (RTV) when compared with the two surface (Institut Straumann, Waldenburg, Switzer-
rough surfaces, and that the SLA surface showed land). This implant shape, which has been commer-
higher RTV than the TPS surface (without reaching cially available for more than 10 years with a TPS sur-
statistical significance) during initial healing of 4 face, has no macroscopic retentive elements such as
weeks. At 8 and 12 weeks, both surfaces showed sim- vents or grooves in the apical portion, but has a square
ilar values. top to allow for proper removal torque testing.
In 1996, another acid-etched titanium surface was The surface characteristics of both implants were
introduced to the implant market under the brand examined qualitatively by scanning electron micro-
name Osseotite. When introduced, the dearth of sci- scopy. To determine the profile quantitatively, an addi-
entific documentation on this surface provoked critical tional profilometric analysis was performed using a
remarks by Taylor25 in an editorial in The Interna- Form Talysurf Series-2 laser interferometric system
tional Journal of Oral & Maxillofacial Implants. (Rank Taylor Hobson, Leicester, UK) equipped with a
Recently, an animal study published by Klokkevold et custom-made 0.6-µm-diameter diamond stylus. For
al26 compared an acid-etched surface with a machined both implant types, two samples were scanned along
titanium surface in a removal torque test. The study the circumference in three or four different areas over
demonstrated significantly better results for the acid- a length of 2 mm. Thirty-one different amplitude,
etched surface. As of today, however, no data are avail- spacing, and hybrid parameters were calculated from
able on how this acid-etched surface compares with the profile data. Average roughness Ra and the mean
other rough titanium surfaces that have been used in spacing of adjacent local peaks S were selected as the
implant patients for many years. variables to best describe the surface characteristics.
The purpose of this study was to compare the sur- R a is a universally recognized and widely used
face of this new implant (Osseotite) with the sand- parameter to describe roughness. It is the arithmetic
blasted and acid-etched surface of the scientifically mean of the departures of the roughness profile from
well-documented SLA implant in a side-by-side the mean line, and is calculated as:
analysis in the maxillae of miniature pigs by measur-
ing removal torque values after 4, 8, and 12 weeks of L
1
healing. Ra =
L ∫ y(x) dx
0
Materials and Methods
where L is the assessment length and y(x) is the pro-
Approval for Animal Research. The protocol for file amplitude. S is the mean spacing of adjacent local
the animal study was approved by the standing com- peaks, and is calculated as:
mittee on Animal Research at the University Hospi-
tal, Medical Faculty, University of Berne, Switzer- n
S1 + S 2 + K + S n
∑s
1
land, and by the Committee for Animal Research, S= i =
State of Berne (approval no. 97/04). The State of N i=1
n

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

612 Volume 13, Number 5, 1998 OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Buser et al

Fig 2a Self-cutting Osseotite implants served as the test Fig 2b SLA implants were pretapped into place in the con-
implants in this split-mouth study design. tralateral side and served as controls.

where n is the number of spacings over the assess- for Osseotite implants, primary wound closure was
ment length and Si is the spacing between the local achieved with resorbable sutures.
peaks. A local peak is the highest part of the profile Removal Torque Testing of the Bone-Implant
measured between two adjacent minima. Interface. The miniature pigs were sacrificed after
Surgical Procedures. A total of nine adult minia- healing periods of 4, 8, or 12 weeks. Each subgroup
ture pigs with a minimum age of 2 years were used in consisted of three miniature pigs, each with six
the study. In each animal, two surgical interventions implants for biomechanical testing. Immediately after
were performed. First, the anterior teeth in the max- sacrifice, the soft tissues in the edentulous areas of the
illa were removed under general anesthesia using maxilla were removed to expose the integrated
extended mucoperiosteal flaps to provide sufficient implants. Subsequently, the maxilla was excised and
access to the alveolar crest containing the teeth to be the left and right halves were isolated with a diamond-
removed (Surgical Research Unit ESI and Clinic for plated saw (Makro Trennsystem, Exakt Apparatebau
Large Animals, University of Berne). Prior to surgery, AG, Norderstedt, Germany). To improve further han-
the animals were given 1 g of prophylactic amoxicillin dling and for temperature isolation, each of the sam-
intramuscularly. Following tooth removal, the ele- ples was molded into dental cement (Kerr Suprastone
vated flaps were repositioned and closed with inter- Green, Kerr Europe AG, Basel, Switzerland).
rupted sutures. Second, after a healing period of at The removal torque testing was performed on a
least 6 months, 6 to 8 implants were placed in each biaxial servohydraulic materials-testing machine
animal. Six implants were scheduled for biomechani- (MTS Minibionix 358.02, MTS Systems, Minneapo-
cal evaluation with removal torque testing, and one lis, MN). To apply pure axial moments in the test, the
implant per side was scheduled for histologic analysis. axis of the implant to be tested had to correspond
If the anatomic situation allowed, this implant was exactly with the axis of the testing machine. For this
always located closest distally to the canine. In all, 70 reason, the implant was first attached to the actuator,
implants were placed, 54 of which were intended for thereby guaranteeing the implant-actuator align-
biomechanical testing. The recipient sites in the cre- ment. The implant-bone–dental cement complex was
ated edentulous areas of the maxilla were exposed by lowered into a tub on the rigid part of the machine,
elevation of buccal mucoperiosteal flaps. When nec- which was then filled with low melting temperature
essary, the alveolar crest was flattened to allow pre- metal alloy (Legierung 47 Grad, Billiton Witmetaal
cise preparation of the implant recipient sites. The BV, Naarden, Netherlands). The cooling of the alloy
sites were prepared under copious irrigation with effectively fixed the implant-bone–dental cement
sterile physiologic saline using standard commercially complex to the machine (Figs 3a and 3b). To allow
available drills for both implant types. For the SLA for proper removal torque testing, a commercially
implants, the thread was pretapped into the bone cav- available insertion device for screw-type implants
ity, while the Osseotite implants were self-cutting, as with a standard hex (Nobel Biocare, Lucerne, Swit-
recommended by the manufacturer. A split-mouth zerland) was used for Osseotite implants, which fit
design, using one side for the test and the contralat- precisely onto the hex. For SLA implants, a specially
eral side for the control implants (Figs 2a and 2b), manufactured adapter with a square shape that fit
was employed. Following application of healing caps precisely onto the head of SLA implants was used.

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF The International Journal of Oral & Maxillofacial Implants 613
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
Buser et al

Fig 3a One specimen with three Fig 3b The entire setup for removal torque testing, using the
implants is embedded in dental cement MTS servohydraulic materials testing machine, the affixed
and attached to the actuator, guarantee- implant-bone–dental cement complex attached to the actuator,
ing a perfect implant-actuator alignment. and the PC.
The implant-bone–dental cement com-
plex is now affixed to the tub by filling in
low temperature melting metal alloy.

The removal torque test was performed by applying a used separately for each implant position. To investi-
counterclockwise rotation to the implant axis at a rate gate the influence of implant type, a sign test was
of 0.1 degree/second. The resulting torques were used to calculate the differences between the paired
measured by an axial-torsional load transducer (MTS variables at each implant position (position 1 being
662.20D-04, MTS Systems). mesial, position 3 distal).
After testing of one implant was completed, the
alloy was melted to remove the implant-bone–dental Results
cement complex from the fitting tub. The next implant
was secured to the actuator, and the entire process was Characterization of Both Surfaces. The scanning
repeated until all implants were tested. To avoid dry- electron microscope (SEM) analysis (Figs 4a and 4b)
ing of the bone, the specimens were sprayed with revealed similar features for the two implant sur-
saline solution every 15 minutes. For each implant, the faces: both showed small micropits with a diameter
torque-rotation curve was recorded. To characterize of 1 to 2 µm produced by the acid-etching procedure.
the bone-implant interface, the removal torque was The Osseotite surface, however, seemed to have a
defined as the maximum torque on the curve. flatter profile when compared with the SLA surface.
Failure Torque Testing of Hex Connection of This impression was confirmed by profilometric
Osseotite Implants. Since the hex of Osseotite analysis (Figs 5a and 5b) since the SLA surface
implants is rather short, and thus possibly not able to yielded higher values for average roughness (Ra = 2.0
withstand the occurring shear stresses during µm) than the Osseotite surface (Ra = 1.3 µm). Con-
removal torque testing of the bone-implant interface, cerning mean spacing of peaks, both implants
the failure torque of the hex connection of four demonstrated similar values (S = 12.0 µm for SLA
Osseotite implants inserted in dental cement was also versus 15.0 µm for Osseotite).
tested in vitro. The same insertion device as was used At sacrifice and following soft tissue removal, four
for reverse torque testing was mounted onto the hex implants, two of each implant type, demonstrated a
of the four individually embedded implants. Each penetration of their apical portion into the nasal cav-
single implant was precisely aligned to the long axis ity as a result of a reduced vertical bone height at this
of the actuator as outlined above. specific site and/or a misangled implant, thus reduc-
Statistical Analysis. Because of the small num- ing the extent of bone-anchoring surface for these
ber of implants being tested, the assumption of nor- four implants. Consequently, it was decided to
mally distributed groups could not be made. A multi- exclude these implants from further analysis. The
factor analysis of variance (ANOVA) could not be remaining 50 implants, which demonstrated firm
applied because of missing values in the test proto- anchorage in the maxilla, were used for removal
col. Therefore, simple nonparametric methods were torque measurements.

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

614 Volume 13, Number 5, 1998 OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Buser et al

Osseotite
SLA

Fig 4a SEM analysis showing the Osseotite surface with micro- Fig 4b SEM analysis showing the SLA surface with a rough-
pits of 1 to 2 µm in diameter produced by the acid-etching pro- blasted surface recognized as “valleys” and micropits of 1 to 2
cedure. Note that the Osseotite surface has a rather flat profile µm in diameter produced by the acid-etching procedure (origi-
when compared to the SLA surface (original magnification  nal magnification  620).
620).

Osseotite SLA
10 10
8 8
6 6
4 4
Y (µm)

2
Y (µm)

2
0 0
–2 –2
–4 –4
–6 –6
–8 –8
–10 –10
200 300 400 500 600 700 800 900 1000 1100 1200 200 300 400 500 600 700 800 900 1000 1100 1200

X (µm) X (µm)

Fig 5a Profilometric analysis of the Osseotite surface. Values Fig 5b Profilometric analysis of the SLA surface. Values for
for average roughness were 1.3 µm, lower than those for the average roughness were 2.0 µm.
SLA surface.

The removal torque testing resulted in a typical between 62.6 and 95.7 Ncm for Osseotite implants
curve for both implants, where its peak was assumed (Fig 7). The differences between the two implant
to be the failure (removal) torque of the bone- types were highly significant (implant position 1: P <
implant interface. The Osseotite implants showed a .008; implant positions 2 and 3: P < .004). The meas-
flatter curve without a clear reduction following fail- urements of the failure torque for the hex connection
ure torque after 12 to 18 degrees of counterclockwise of four Osseotite implants demonstrated values
rotation (Fig 6). In contrast, the curve for SLA between 148 and 163 Ncm, with a mean failure
implants demonstrated a higher slope to the maxi- torque of 156.0 ± 6.8 Ncm (Table 2). The curve
mum level at approximately 10 to 12 degrees of demonstrated a constant increase and reached the
counterclockwise rotation, and a subsequent clear failure torque at approximately 30 degrees of coun-
reduction following fracture at the bone-implant terclockwise rotation (Fig 6).
interface (Fig 6). A summary of all removal torque
values (RTVs) for both implant types at the three dif- Discussion
ferent healing periods is shown in Table 1. At all time
periods, SLA implants demonstrated 75 to 125% In the present study, two titanium screw-type
higher mean RTV than Osseotite implants. The mean implants of different shapes and surface characteris-
RTV ranged between 109.6 and 196.7 Ncm for SLA tics were tested in the maxillae of miniature pigs.
implants, whereas the corresponding values ranged This animal model was chosen to measure removal

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF The International Journal of Oral & Maxillofacial Implants 615
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
Buser et al

200

180 SLA

160

140
Torque (Ncm)

120 Hex

100 Fig 6 Removal torque testing of SLA and

Osseotite implants after 3 months of healing.
80
Osseotite implants showed a rather flat curve
60 and the interface failure occurred around 95
Osseotite Ncm at about 15 degrees of counterclockwise
40 rotation. In contrast, SLA implants yielded a
constant increase to the maximum level,
20
around 180 Ncm at approximately a 12-
0 degree rotation, followed by a steep reduction
0 5 10 15 20 25 30 35 40 after failure. The failure torque testing of the
hex connection demonstrated an increase to
Angle (degrees) the maximum level of around 150 Ncm at
approximately 30 degrees of counterclock-
wise rotation.

Table 1 Removal Torque Values (Ncm) for Both Implant Types at Three Different
Healing Periods*
Healing periods

1 (4 weeks) 2 (8 weeks) 3 (12 weeks)

Animal Osseotite SLA Osseotite SLA Osseotite SLA

74 124 96 218 70 228

1 78 119 94 214 82 200
70 98 88 181 80 180
57 108 94 194 121 —
2 46 110 83 215 93 203
46 92 79 139 92 129
73 116 — 227 123 161
3 57 121 85 — 113 209
— 98 81 185 88 186
Mean RTV† 62.5 109.6 87.6 196.7 95.7 186.8
12.9 11.6 6.4 30.6 18.9 30.8

*Three animals were tested at each healing period.

†RTV = removal torque value.

torques of implants in maxillary bone because its ent roughness values.27 This analysis demonstrated
bone structure is comparable to that of humans, in that SLA implants have a rougher surface (Ra = 2.0
whom dental implants are placed in daily practice. µm) when compared with Osseotite implants (Ra =
The two tested implants were chosen for their 1.3 µm). This difference was also apparent in SEM
rough titanium surface in the bone-anchoring por- analysis. The SLA surface is characterized by a pri-
tion, which can be attributed in part to a sulfuric mary roughness produced by the sandblasting proce-
acid–hydrochloric acid etching procedure. The two dure that creates “valleys” whereas the acid-etching
implants differ, however, in that the SLA surface is procedure removes remnants of grits and attacks the
sandblasted (large grits of 250 to 500 µm) and acid- titanium surface, producing 1- to 2-µm-diameter
etched, whereas the Osseotite surface is acid-etched micropits superimposed on the rough-blasted sur-
superimposed on a machined surface. The character- face. The Osseotite surface has similar micropits pro-
ization of the surface topography was performed with duced by the etching procedure, but its profile is flat-
stylus profilometry. This method has been used fre- ter, since no sandblasting procedure is used prior to
quently to characterize titanium surfaces with differ- etching. It is important to note, however, that stylus

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

616 Volume 13, Number 5, 1998 OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Buser et al

240
220
200
180
160

Torque (Ncm)
140
120
100
80
60
40
Fig 7 Comparison of mean removal torque 20
values for Osseotite and SLA implants at 4, 8, 0
and 12 weeks (± SD). The observed differ- 4 weeks 8 weeks 12 weeks
ences between the two implant types are sig-
nificant (P < .01).

profilometry cannot resolve the finer features of both Table 2 Failure Torque Testing of the Hex of Four
surfaces with their micropits. Development of a more Osseotite Implants (Ncm)
precise method to capture the small-range (1 to 10 Implant Failure torque of hex
µm) as well as the long-range features (10 to 30 µm)
1 148
is required. However, the measured values represent 2 163
the major topographic features of each surface. 3 160
The significant difference observed in removal 4 153
Mean 156.0
torque testing can most likely be attributed to the SD 6.8
different surface characteristics, although the tested
implants also differed concerning their macroscopic
implant shape. On the one hand, the SLA implants
profited from a slightly larger diameter (4.05 mm argued that the relatively short hex connection would
versus 3.75 mm). On the other hand, Osseotite not resist the shear stresses during removal torque
implants were favored as a result of four grooves in testing. The examination of four Osseotite implants
the apical portion for the self-cutting procedure dur- embedded in dental cement demonstrated failure
ing implant placement, and this feature provided torques of between 148 and 163 Ncm (mean RTV of
additional bone anchorage against removal torque 156 Ncm). These values are clearly higher than all
testing. The SLA implants were solid screws and measured RTVs of the 25 osseointegrated Osseotite
lacked any macroscopic retention elements. This dif- implants. In addition, the hex failure torque was
ference in implant shape most likely explains the always observed at approximately 30 degrees of coun-
observed difference in the characteristics of the terclockwise rotation, whereas Osseotite implants
removal torque curves. The SLA curve showed an demonstrated failure torques at 12 to 18 degrees of
immediate decrease following the peak in the curve, counterclockwise rotation (Fig 6). Therefore, it can be
whereas Osseotite implants were characterized by a concluded that the mean RTVs, ranging between 62
slow, creeping failure of the interface, since the four and 96 Ncm, do in fact represent the failure torques
grooves in the apical portion continued to provide at the bone-implant interface at various time points.
resistance to shear by their macroscopic anchorage in The results of this study confirm the findings of
the bone. Summarizing all these aspects of micro- previous biomechanical studies in various animal
scopic and macroscopic properties of the two tested models. These studies all demonstrated higher RTV
implants, it can be assumed that the surface charac- for roughened titanium surfaces when compared
teristics had the most significant impact on the inter- with fine-structured or machined titanium sur-
face shear strength, since SLA implants without api- faces.20,22,24,26,28–35 The two most recent studies eval-
cal grooves had 75 to 125% higher mean RTV than uated the two implant surfaces that were tested in
Osseotite implants with four apical grooves. the present study. The study by Klokkevold et al26
The in vitro testing of the hex connection of examined screw-type implants with an Osseotite sur-
Osseotite implants was necessary since it could be face and a length of 4 mm, but no macroscopic

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF The International Journal of Oral & Maxillofacial Implants 617
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
Buser et al

grooves in the tibiae of rabbits. Titanium implants of Acknowledgments

the same shape but with a machined surface served
as controls. The removal torque testing after 8 weeks The authors are grateful to Dr D. Mettler, VD, and the entire
team of the Surgical Research Unit ESI and Clinic for Large Ani-
of healing revealed a mean RTV four times higher
mals, University of Berne, for their excellent support during the
than that for machined implants, although the values study. The help of Ms C. Grau during surgeries and of Dr D.
were rather low (mean RTV of 20.50 Ncm for Osseo- Snétivy for the manufacturing of SLA implants is greatly appreci-
tite and 4.95 Ncm for machined implants). The ated. The authors are also grateful to Prof Hüsler, Institute for
recent study by Buser et al 24 compared titanium Mathematical Statistics, University of Berne, for his statistical
advice. The study was supported by a grant from the ITI Founda-
screw implants of identical shape but with three dif-
tion for the Promotion of Oral Implantology (no. 94-080).
ferent surfaces. The SLA and TPS surfaces demon-
strated 8 to 10 times higher mean RTV than the
References
machined surface. These two studies, though carried
out in different animal models, compare well with 01. Brånemark P-I, Breine U, Adell R, Hansson BO, Lindström J,
the results of the present study. Olsson A. Intraosseous anchorage of dental prostheses. I.
The SLA surface, produced by a sandblasting and Experimental studies. Scand J Plast Reconstr Surg
acid-etching procedure, has the potential to become 1969;3:81–100.
02. Brånemark P-I, Hansson BO, Adell R, Breine U, Lindström J,
important in implant dentistry in the future. Based
Hallen O, Ohman A. Osseointegrated implants in the treat-
on more than 10 years of scientific evaluation, this ment of the edentulous jaw. Experience from a 10-year
surface is scientifically well-documented in both in period. Scand J Plast Reconstr Surg 1977;11(suppl 16):1–132.
vitro36–41 and in vivo20–24 studies. All of these studies 03. Schroeder A, Pohler O, Sutter F. Gewebereaktion auf ein
demonstrated that the SLA surface offers equal or Titan-Hohlzylinderimplantat mit Titan-Spritzschichtober-
flüche. Schweiz Monatsschr Zahnheilk 1976;86:713–727.
even better results when compared with the TPS sur-
04. Schroeder A, Stich H, Straumann F, Sutter F. Über die
face, in particular during the initial healing period. Anlagerung von Osteozement an einen belasteten Implan-
The trend for faster bone integration of SLA im- tatkörper. Schweiz Monatsschr Zahnheilk 1978;88:1051–1058.
plants might create the opportunity to reduce the 05. Schroeder A, van der Zypen E, Stich H, Sutter F. The reac-
current healing period of 3 months without func- tion of bone, connective tissue and epithelium to endosteal
implants with sprayed titanium surfaces. J Maxillofac Surg
tional loading, a standard routinely used for implants
1981;9:15–25.
with a TPS surface in clinical practice.9,10,13 A multi- 06. Adell R, Lekholm U, Rockler B, Brånemark PI. A 15-year
center study is currently testing this hypothesis using study of osseointegrated implants in the treatment of the
6 weeks of healing for implants in sites with normal edentulous jaw. J Oral Surg 1981;10:387–416.
bone density (Classes I to III). 07. Babbush CA, Kent JN, Misiek DJ. Titanium plasma-sprayed
(TPS) screw implants for the reconstruction of the edentulous
mandible. J Oral Maxillofac Surg 1986;44:274–282.
Conclusions 08. Adell R, Eriksson B, Lekholm U, Brånemark PI, Jemt T. A
long-term follow-up study of osseointegrated implants in the
The present biomechanical study in the maxillae of treatment of totally edentulous jaws. Int J Oral Maxillofac
miniature pigs compared two different titanium screw Implants 1990;5:347–359.
implants with a rough surface in the bone-anchoring 09. Buser D, Weber HP, Lang NP. Tissue integration of non-
submerged implants. 1-year results of a longitudinal study
portion. The study demonstrated significant differ- with ITI hollow-screw and hollow-cylinder implants. Clin Oral
ences (P < .01) in mean removal torque values Implants Res 1990;1:78–85.
between two different titanium screw implants with a 10. Buser D, Weber HP, Brügger U, Balsiger C. Tissue integra-
sandblasted and acid-etched surface (SLA) and a tion of one-stage ITI implants. 3-year results of a longitudinal
machined and acid-etched surface (Osseotite) at 4, 8, study with hollow-cylinder and hollow-screw implants. Int J
Oral Maxillofac Implants 1991;6:405–412.
and 12 weeks of healing. Mean RTVs ranged between 11. Lekholm U, van Steenberghe D, Herrmann I, Bolender C,
62 and 96 Ncm for Osseotite implants, while SLA Folmer T, Gunne J, et al. Osseointegrated implants in the
implants revealed mean RTVs between 109 and 196 treatment of partially edentulous jaws: A prospective 5-year
Ncm. Differences exist between roughened titanium multicenter study. Int J Oral Maxillofac Implants
implants with different surface characteristics; there- 1994;9:627–635.
12. Mericske-Stern R, Steinlin-Schaffner T, Marti P, Geering AH.
fore, proper scientific evaluation of new titanium sur-
Peri-implant mucosal aspects of ITI implants supporting over-
faces, including both in vitro and in vivo studies, is dentures. A 5-year longitudinal study. Clin Oral Implants Res
mandatory prior to clinical application in patients. 1994;5:9–18.
13. Buser D, Mericske-Stern R, Bernard JP, Behneke A, Behneke
N, Hirt HP, et al. Long-term evaluation of non-submerged
ITI Implants. Part I: 8-year life table analysis of a prospective
multi-center study with 2359 implants. Clin Oral Implants Res
1997;8:161–172.

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

618 Volume 13, Number 5, 1998 OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Buser et al

14. Behneke A, Behneke N, d’Hoedt B, Wagner W. Hard and soft 28. Claes L, Hutzschenreuter P, Pohler O. Lösemomente von
tissue reactions to ITI screw implants: 3-year longitudinal Corticalis-schrauben in Abhängigkeit von Implantationszeit
results of a prospective study. Int J Oral Maxillofac Implants und Oberflüchenbeschaffenheit. Arch Orthop Unfall-Chir
1987;12:749–757. 1976;85:155–159.
15. Jaffin RA, Berman CL. The excessive loss of Brånemark fix- 29. Thomas KA, Kay JF, Cook SD, Jarcho M. The effect of sur-
tures in type IV bone: A 5-year analysis. J Periodontol face macrotexture and hydroxylapatite coating on the mechan-
1991;62:2–4. ical strengths and histologic profiles of titanium implant mate-
16. Quirynen M, Naert I, van Steenberghe D, Schepers E, Cal- rials. J Biomed Mater Res 1987;21:1395–1414.
berson L, Theuniers G, et al. The cumulative failure rate of 30. Carlsson L, Röstlund T, Albrektsson B, Albrektsson T.
the Brånemark system in the overdenture, the fixed partial, Removal torques for polished and rough titanium implants.
and the fixed full prostheses design: A prospective study on Int J Oral Maxillofac Implants 1988;3:21–24.
1273 fixtures. J Head Neck Pathol 1991;10:43–53. 31. Gotfredsen K, Nimb L, Hjørting-Hansen E, Jensen JS, Hol-
17. Jemt T. Implant treatment in resorbed edentulous upper jaws. men A. Histomorphometric and removal torque analysis for
A three-year follow-up study in 70 patients. Clin Oral smooth and TiO2 blasted titanium implants in dogs. Clin Oral
Implants Res 1993;4:187–194. Implants Res 1992;3:77–84.
18. Jemt T, Chai J, Harnett J, Heath MR, Hutton JE, John RB, et 32. Gotfredsen K, Wennerberg A, Johansson C, Skovgaard LT,
al. A 5-year prospective multicenter follow-up report on over- Hjørting-Hansen E. Anchorage of TiO2 blasted, HA-coated,
dentures supported by osseointegrated implants. Int J Oral and machined implants: An experimental study with rabbits. J
Maxillofac Implants 1996;11:291–298. Biomed Mater Res 1995;29:1223–1231.
19. Thomas KA, Cook S. An evaluation of variables influencing 33. Wennerberg A, Albrektsson T, Andersson B, Krol JJ. A histo-
implant fixation by direct bone apposition. J Biomed Mater metric and removal torque study on screw-shaped titanium
Res 1985;19:875–901. implants with three different surface topographies. Clin Oral
20. Wilke HJ, Claes L, Steinemann S. The influence of various Implants Res 1995;6:24–30.
titanium surfaces on the interface shear strength between 34. Wennerberg A, Albrektsson T, Lausmaa J. A torque and histo-
implants and bone. In: Heimke G, Soltész U, Lee AJC (eds). morphometric evaluation of c.p. titanium screws, blasted with
Advances in Biomaterials, Vol. 9: Clinical Implant Materials. 25 and 75 µm sized particles of AI2O3. J Biomed Mater Res
Amsterdam: Elsevier Science, 1990:309–314. 1996;30:251–260.
21. Buser D, Schenk RK, Steinemann S, Fiorellini JP, Fox CH, 35. Wennerberg A, Ektessabi A, Albrektsson T, Johansson C,
Stich H. Influence of surface characteristics on bone integra- Andersson B. A 1-year follow-up of implants of differing sur-
tion of titanium implants. A histometric study in miniature face roughness placed in rabbit bone. Int J Oral Maxillofac
pigs. J Biomed Mater Res 1991;25:889–902. Implants 1997;12:486–494.
22. Wong M, Eulenberger J, Schenk RK, Hunziker E. Effect of 36. Cochran DL, Simpson J, Weber HP, Buser D. Attachment
surface topography on the osseointegration of implant materi- and growth of periodontal cells on smooth and rough tita-
als in trabecular bone. J Biomed Mater Res nium. Int J Oral Maxillofac Implants 1994;9:289–297.
1995;29:1567–1575. 37. Martin JY, Schwartz Z, Hummert TW, Schraub DM, Simpson
23. Cochran DL, Schenk RK, Lussi A, Higginbottom FL, Buser J, Lankford J Jr, et al. Effect of titanium surface roughness on
D. Bone response to unloaded and loaded titanium implants proliferation, differentiation, and protein synthesis of human
with a sandblasted and acid-etched surface. A histometric osteoblast-like cells (MG63). J Biomed Mater Res
study in the canine mandible. J Biomed Mater Res 1995;29:389–401.
1988;40:1–11. 38. Martin JY, Dean DD, Cochran DL, Simpson J, Boyan BD,
24. Buser D, Nydegger T, Oxland T, Cochran DL, Schenk RK, Schwartz Z. Proliferation, differentiation, and protein synthe-
Hirt HP, et al. The interface shear strength of titanium sis of human osteoblast-like cells (MG63) cultured on previ-
implants with a sandblasted and acid-etched surface. A biome- ously exposed titanium surfaces. Clin Oral Implants Res
chanical study in the maxilla of miniature pigs. J Biomed 1996;7:27–37.
Mater Res 1998 (in press). 39. Schwartz Z, Martin JY, Dean DD, Simpson J, Cochran DL,
25. Taylor T. Again and again [editorial]. Int J Oral Maxillofac Boyan BD. Effect of titanium surface roughness on chrondo-
Implants 1996;11:709–710. cyte proliferation, matrix production, and differentiation
26. Klokkevold PR, Nishimura RD, Adachi M, Caputo A. depends on the state of cell maturation. J Biomed Mater Res
Osseointegration enhanced by chemical etching of the tita- 1996;30:145–155.
nium surface. A torque removal study in the rabbit. Clin Oral 40. Kieswetter K, Schwartz Z, Hummert TW, Cochran DL, Simp-
Implants Res 1997;8:442–447. son J, Dean DD, et al. Surface roughness modulates the local
27. Wennerberg A, Albrektsson T, Ulrich H, Krol J. An optical 3D production of growth factors and cytokines by osteoblast-like
method for topographical description of surgical implants. J MG-63 cells. J Biomed Mater Res 1996;32:55–63.
Biomed Eng 1992;14:412–418. 41. Boyan BD, Batzer R, Kieswetter K, Liu Y, Cochran DL,
Szmuckler-Moncler S, et al. Titanium surface roughness alters
responsiveness of MG63 osteoblast-like cells to 1alpha,25-
(OH)2D3. J Biomed Mater Res 1998;39:77–85.

COPYRIGHT © 2000 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING

OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF The International Journal of Oral & Maxillofacial Implants 619
THIS ARTICLE MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITH-
OUT WRITTEN PERMISSION FROM THE PUBLISHER.