J Oral Maxillofac Surg

68:1667-1675, 2010

Biomechanical Evaluation of Endosseous

Implants at Early Implantation Times:
A Study in Dogs
Paulo G. Coelho, DDS, PhD,* Rodrigo Granato, DDS, MSc,†
Charles Marin, DDS, MSc,‡
Estevam A. Bonfante, DDS, MSc, PhD,§
Jose N.O. Freire, DDS, PhD,㛳 Malvin N. Janal, PhD,¶
Jose N. Gil, DDS, MSc, PhD,# and Marcelo Suzuki, DDS**

Purpose: This study tested the null hypothesis that differences in surgical instrumentation, macroge-
ometry, and surface treatment imposed by different implant systems do not affect early biomechanical
fixation in a canine mandible model.
Materials and Methods: The lower premolars of 6 beagle dogs were extracted and the ridges allowed
to heal for 8 weeks. Thirty-six (n ⫽ 12 each group) implants were bilaterally placed, remaining for 1 and
3 weeks in vivo. The implant groups were as follows: group 1, Ti-6Al-4V with a dual acid-etched surface
with nanometer scale discrete crystalline deposition (Nanotite; Certain Biomet-3i, West Palm Springs,
FL); group 2, Ti-6Al-4V with a titanium oxide-blasted fluoride-modified surface chemistry (Osseospeed 4.0
S; Astra Tech, Mölndal, Sweden); group 3: Ti-6Al-4V with a bioceramic microblasted surface (Ossean;
Intra-Lock International, Boca Raton, FL). Following euthanasia, implants were torqued to interface
failure and histologically evaluated. General linear modeling (ANOVA) at 95% level of significance was
performed.
Results: Histology showed that interfacial bone remodeling and initial woven bone formation were
observed around all implant groups at 1 and 3 weeks. Torque values were significantly affected by time in
vivo, implant group, and their interaction (P ⫽ .016, P ⬍ .001, and P ⫽ .001, respectively). Regarding torque
values, group 3, group 2, and group 1 ranked highest, intermediate, and lowest, respectively.
Conclusion: Early biomechanical fixation at 1 and 3 weeks was affected by surgical instrumentation,
macrogeometry, and surface treatment present for one of the implant systems tested. The null hypothesis
was rejected.
© 2010 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 68:1667-1675, 2010

Osseointegration is a phenomenon in which intimate
contact between bone and biomaterials occurs at the
optical microscopy level, enabling dental implants to
replace load-bearing tooth organs and restore their
form and intraoral function.1 Specific to dental im-
plantology, implant therapy success ratios often ex-
ceed 90%.2,3 However, despite the high success rates
reported, basic scientists and clinicians have at-
tempted to decrease treatment time frames by reduc-
ing the healing period for establishment of osseointe-

*Assistant Professor, Department of Biomaterials and Biomimet-
ics, New York University, New York, NY.
†Instructor, Department of Dentistry, Universidade Federal de
Santa Catarina, Florianopolis, Brazil.
‡Instructor, Department of Dentistry, Universidade Federal de
Santa Catarina, Florianopolis, Brazil.
§Private Practice, Bauru, SP, Brazil.
㛳Private Practice, Florianópolis, SC, Brazil.
¶Senior Research Scientist, Department of Epidemiology and
Health Promotion, New York University, New York, NY.
#Associate Professor, Department of Dentistry, Universidade Fed-
eral de Santa Catarina, Florianopolis, Brazil.
**Assistant Professor, Department of Prosthodontics, Tufts Uni-
versity School of Dental Medicine, Boston, MA.
Address correspondence and reprint requests to Dr Coelho:
345 East 24th Street, Room 314a, Department of Biomaterials and
Biomimetics, New York University, New York, NY 10010; e-mail:
pgcoelho@nyu.edu
© 2010 American Association of Oral and Maxillofacial Surgeons
0278-2391/10/6807-0033$36.00/0
doi:10.1016/j.joms.2010.02.050

1667
1668
gration.4 For that purpose, the most common
approach is through modification of different implant
design parameters.4,5
Implant design alterations have included changes in
its structural material,6,7 its macrogeometry and/or
surgical instrumentation,5,8,9 and surface modifica-
tions.10-12 Although it is not well defined whether
biocompatible materials other than titanium and its
alloys improve the host-to-implant response, alter-
ations in macrogeometry and/or surgical instrumenta-
tion and surface modifications have demonstrated sig-
nificant effects at the early stages of bone healing
around endosseous implants.5,8-12
Several studies have demonstrated that depending
on the interplay between final bone drilling and im-
plant geometric dimensions, different bone healing
mechanisms and kinetics may be observed.5,13-15 If
there is intimate surgical fit between bone and im-
plant surface, the interfacial bone undergoes remod-
eling and is gradually substituted by woven bone that
is gradually replaced by mature lamellar bone.13 How-
ever, if healing chambers form in regions in which
bone and implant are not in intimate contact because
of the interplay between drilling and implant dimen-
sions, rapid woven bone filling occurs and long-term
implant stability is ensured by bone modeling and
remodeling processes.5,13,14
Among surface modifications, implant texture has
evolved from the as-turned surfaces toward moder-
ately rough surfaces because several studies have
shown improved histologic and biomechanical host-
to-implant response.10,12,16,17 More recently, several
studies have shown that chemistry alterations to mod-
erately rough surfaces, such as the addition of fluoride
or calcium phosphate in small scales (nanometer or
minute scale), have resulted in further improvements
in bone to implant response at early implantation
times in vivo.12,13,15,18-20 Although several experimen-
tal studies have demonstrated the effectiveness of
surface modifications at early implantation times regard-
less of implant macrogeometry and drilling technique,
further experimentation is desirable to determine what
macrogeometry/drilling dimension/surface modification
results in optimal bone-to-implant response at early im-
plantation times.
From a clinical perspective, it is general consensus
that implant stability immediately and early after
placement is desirable, because relative motion be-
tween implant and bone may risk osseointegra-
tion.4,16,21 Histologic studies have shown that implant
primary stability is rendered during placement, and
because of remodeling and subsequent bone apposi-
tion bridging old bone and implant surface, biological
or secondary stability is achieved.8,13,14 This rationale
has led to the presumption that if osteoclastic activity
undermines primary stability before new bone forma-
BIOMECHANICAL EVALUATION OF ENDOSSEOUS IMPLANTS
tion prevents implant micromotion, a stability de-
crease will take place at early times after implanta-
tion.22
Although substantial work has been published regarding
the effects of macrogeometry and implant surface on early
bone-implant interaction,5,8,13,14,17,18,23,24 comparative
studies regarding the effect of implant surgical ins-
trumentation/macrogeometry/surface on stability at
early implantation times are sparse in the literature.
Thus, this investigation tested the null hypothesis that
different implant systems presenting distinct surgical
instrumentation, macrogeometry, and surface would
not present alterations in biomechanical fixation at
early implantation times in a canine mandible model.
Materials and Methods
This study used 36 endosseous implants with vari-
ations in bulk geometry and surface treatment from
three manufacturers (12 of each type).
The implant groups were as follows: group 1, Ti-
6Al-4V threaded implant with a dual acid-etched sur-
face with nanometer scale discrete crystalline depo-
sition (Nanotite; Certain Biomet-3i, West Palm
Springs, FL); group 2, Ti-6Al-4V threaded implant (pre-
senting microthreads in the cervical third) with a
titanium oxide-blasted fluoride-modified surface
chemistry (Osseospeed, 4.0 S; Astra Tech, Mölndal,
Sweden); group 3, Ti-6Al-4V threaded implant (pre-
senting microthreads in the cervical third) with a
bioceramic microblasted surface (Ossean; Intra-Lock
International, Boca Raton, FL). The group 1 and 3
devices presented the same physical dimensions with
respect to diameter (4 mm) and length (10 mm).
Group 2 implants had a 4-mm diameter and 11-mm
implant length.
Following approval of the bioethics committee for
animal experimentation at the Universidade Federal
de Santa Catarina, Brazil, 6 beagle dogs with closed
growth plates (between 18 and 24 months of age) in
good health were acquired for the study and followed
a 2-week in-house period before surgery.
All surgical procedures were performed under gen-
eral anesthesia. The preanesthetic procedure com-
prised intramuscular administration of atropine sul-
fate (0.044 mg/kg) and xylasin chlorate (8 mg/kg).
General anesthesia was then obtained following an
intramuscular injection of ketamine chlorate (15 mg/
kg).
Bilateral extractions of all 4 premolars (P1, P2, P3,
and P4) were performed. The procedure involved a
full thickness mucoperiosteal flap, teeth sectioning in
the buccolingual direction so individual roots could
be extracted by means of root elevators and forceps
without bone wall damage. The soft tissue was closed
COELHO ET AL 1669

with resorbable sutures (3-0 Vicryl, Ethicon, Lang-

home, PA).
Following a healing period of 8 weeks, 1 implant of
each design was placed by 1 experienced implantolo-
gist in each of the 6 animals, 3 per mandible side;
implant configuration placement was interpolated at
the distal, central, and proximal positions. Two weeks
later, 1 implant of each design was placed in the other
hemiarch following the same distribution of the pre-
vious implant placement procedure. The animals
were euthanized 3 weeks after the first implant place-
ment surgery, and each dog provided implants that
remained for 3 and 1 weeks in vivo. Such experimen-
tal distribution allowed the comparison of the same
number of implants per design per mandibular region
and time in vivo.
For implant placement, a mucoperiosteal flap was
used to expose the alveolar bone (Fig 1A), the distal
implant was placed 0.5 cm from the first molar, and the
remaining 2 implants (central and mesial) were sequen-
tially placed 1 cm from each other (Fig 1B). All implants
were placed following the manufacturer’s surgical pro-
tocol (Table 1), and the soft tissue was closed with
resorbable sutures (3-0 Vicryl, Ethicon; Fig 1C). Postsur-
gical medication included antibiotics (penicillin, 20.000
UI/kg) and analgesics (ketoprofen, 1 mL/5 kg) for a
period of 48 hours postoperatively. The euthanasia was
performed by anesthesia overdose.
For biomechanical testing, the bone blocks with
implants were adapted to an electronic torque ma-
chine equipped with a 200 N-cm torque load cell
(Test Resources, Minneapolis, MN). Custom ma-
chined tooling was adapted to the implants’ internal
connections, and the retrieved bone was carefully
positioned to minimize angulation during testing. The
implants were torqued to interfacial fracture at a rate
of approximately 0.19618 rad/s, and the maximum
torque value was recorded for each specimen. The
torque machine was set to stop automatically when
a torque drop of 25% from the highest load was
detected. The rationale for this procedure was to FIGURE 1. A, A mucoperiosteal flap was used to access the
minimize interface damage, allowing its histomorpho- alveolar bone and allow implant placement (B) 8 weeks after
extraction of premolars. One implant of each design was placed in
logic evaluation.15,23-25 the mandible, bilaterally (3 per mandible side with site interpola-
Following biomechanical testing, the bone blocks tion). C, Resorbable sutures were used to close the soft tissue.
were kept in 10% buffered formalin solution for 24 Coelho et al. Biomechanical Evaluation of Endosseous Implants.
hours, washed in running water for 24 hours, and J Oral Maxillofac Surg 2010.
gradually dehydrated in a series of alcohol solutions
ranging from 70% to 100% ethanol. Following dehy-
dration, the samples were embedded in a methacry- setting time was allowed before grinding and polish-
late-based resin (Technovit 9100; Heraeus Kulzer, Wehr- ing. The sections were then reduced to a final thick-
heim, Germany) according to the manufacturer’s in- ness of 30 ␮m by means of a series of SiC abrasive
structions. The blocks were then cut into slices papers (400, 600, 800, 1,200, and 2,400; Buehler) in a
(⬃300-␮m thickness) aiming the center of the im- grinding/polishing machine (Metaserv 3000, Buehler)
plant along its long axis with a precision diamond saw under water irrigation.26 The sections were then to-
(Isomet 2000; Buehler, Lake Bluff, IL), glued to acrylic luidine blue stained and referred to optical micros-
plates with an acrylate-based cement, and a 24-hour copy evaluation. The histologic features were qualita-
1670 BIOMECHANICAL EVALUATION OF ENDOSSEOUS IMPLANTS

Table 1. DETAILED DRILLING SEQUENCE FOR EACH IMPLANT SYSTEM

Group 1 Group 2 Group 3

Round drill Guide drill 1.5-mm initial drill

2.0-mm twist drill 2.0-mm twist drill 2.5-mm twist drill
Pilot drill 3.2-mm twist drill 2.8-mm twist drill
3.25-mm twist drill 3.7-mm twist drill 3.2-mm twist drill
Countersink drill 4.0-mm cortical drill 3.5-mm twist drill
3.85 twist drill Countersink drill
Drilling speed: 1,200 rpm Drilling speed: 1,500 rpm Drilling speed: 1,200 rpm under
under irrigation under irrigation irrigation, 400 rpm for last 2 drills
Observation: Drilling sequence Observation: Drilling sequence Observation: Drilling sequence
recommended by recommended by recommended by manufacturer for
manufacturer for dense bone manufacturer for dense bone dense bone
Coelho et al. Biomechanical Evaluation of Endosseous Implants. J Oral Maxillofac Surg 2010.

tively evaluated at 50⫻ to 200⫻ magnification (Leica
DM2500M; Leica Microsystems, Wetzlar, Germany).
Statistical evaluation of torque measures employed
a mixed-model analysis of variance (ANOVA) with 1
between-subjects factor (3 levels of implants). Statis-
tical significance was indicated by P levels less than
5%, and post hoc testing employed the Fisher least
significant difference test.
levels at 1 and 3 weeks were also observed for group
3. However, groups 2 and 3 presented comparable
values at 3 weeks, and these values were significantly
higher than group 1 (Table 6).
Table 2. IMPLANT DISTRIBUTION PER ANIMAL,
SITE, TIME IN VIVO, AND THEIR RESPECTIVE
TORQUE VALUES

Time Dog Site (Distal Torque to

Results Implant In Vivo No. to Proximal) Failure (N-cm)
The surgical procedures and follow-up demon- Group 1 1 1 1 25.1
strated no complications regarding procedural condi- Group 1 1 2 2 15.1
tions or other immediate clinical concerns. However, Group 1 1 3 3 18.3
a postoperative complication was detected in one Group 1 1 4 1 15
Group 1 1 5 2 20.1
3-week group hemimandible, which was excluded Group 1 1 6 3 23
from the study because of clinical instability of all Group 2 1 1 2 31.3
implants after euthanization. Group 2 1 2 3 27.3
All torque values per implant group, animal, im- Group 2 1 3 1 26.7
plantation site, and animal are presented in Table 2. Group 2 1 4 2 19.8
Group 2 1 5 3 18.3
The general linear modeling (ANOVA) biomechanical Group 2 1 6 1 17.5
results are presented in Table 3 and showed that time Group 3 1 1 3 109.2
in vivo, implant group, and the interaction between Group 3 1 2 1 107.1
implant group and time in vivo significantly affected Group 3 1 3 2 123.4
torque values (P ⫽ .016, P ⬍ .001, and P ⫽ .001, Group 3 1 4 3 126
Group 3 1 5 1 90.5
respectively). Implantation site within arch (distal, Group 3 1 6 2 89.4
central, and proximal) and the interaction between Group 1 3 1 1 24.5
implant group and time in vivo/implantation site, and Group 1 3 2 2 41
the 3-way interaction between implant site, time in Group 1 3 3 3 20.1
vivo, and implant group did not significantly affect Group 1 3 4 1 15.1
Group 1 3 5 2 27.6
torque values (P ⬎ .98, P ⬎ .69, P ⬎ .81, and P ⬎ .52, Group 2 3 1 2 72.3
respectively). Group 2 3 2 3 92.2
The overall increase in torque values from 1 to 3 Group 2 3 3 1 87.3
weeks is depicted in Table 4, and the overall differ- Group 2 3 4 2 53
ences in torque between groups presented in Table 5. Group 2 3 5 3 66.3
Group 3 3 1 3 117.3
At 1 week, group 3 presented a significantly higher Group 3 3 2 1 101.9
torque than groups 2 and 1. For implant group 1, Group 3 3 3 2 102.3
comparable levels of torque were observed at both Group 3 3 4 3 57.8
implantation times, whereas a significant increase in Group 3 3 5 1 86.8
torque value was observed for implant group 2 as Coelho et al. Biomechanical Evaluation of Endosseous Implants.
time in vivo elapsed from 1 to 3 weeks. Comparable J Oral Maxillofac Surg 2010.
COELHO ET AL 1671

Table 3. GENERAL LINEAR MODELING ANALYSIS OF VARIANCE STATISTICS SUMMARY

Source Numerator df Denominator df F P

Intercept 1 15 246.12 .000

Time in vivo 1 15 0.989 .016
Implant group 2 15 68.57 ⬍.001
Site 2 15 0.014 .986
Time in vivo*implant group 2 15 12.29 .001
Time in vivo*site 2 15 0.372 .695
Implant group*site 4 15 0.393 .811
Time in vivo*implant group*site 4 15 0.829 .527
Asterisk represents interaction between variables.
Coelho et al. Biomechanical Evaluation of Endosseous Implants. J Oral Maxillofac Surg 2010.

The nondecalcified sample processing after con-
trolled torque to interface failure testing showed sim-
ilar bone-to-implant response. Higher magnification of
the bone-implant interface region showed that the
nondecalcified sections obtained following biome-
chanical testing presented minimal morphologic dis-
tortion due to mechanical testing bone disruption
and/or thin section processing (Figs 2-4). Intimate
bone-to-implant interaction was observed for all im-
plant groups (Figs 2-4).
At 1 week, initial woven bone formation was ob-
served around all implant groups. Irrespective of the
presence of microthreads in implant geometry,
throughout the whole bone-implant interface perim-
eter, all groups presented a combination of regions of
initial woven bone formation and gap regions be-
tween old bone and implant surface characteristic of
interfacial bone modelling-remodeling stages (Figs 2A,
3A,B, and 4A,B). Bone cracking due to implant place-
ment was also observed at regions of microthreads
(Figs 3A and 4A), and large pitch threaded regions
(Figs 2A, 3B, and 4B).
At 3 weeks, irrespective of experimental group,
newly formed woven bone bridging the old bone and
implant surface filled the interfacial modelling-remod-
eling regions observed at 1 week, resulting in contin-
uous bone-implant structures (Figs 2B, 3C,D, and 4C,
D). Bone cracking at the thread tips was seldom
observed at 3 weeks as the bone in intimate contact
with the implant thread tips underwent remodeling
(Figs 2B, 3C,D, and 4C,D).
Discussion
Over the past 40 years, surgical and prosthetic
protocols substantially deviating from the classical
2-stage protocol27 have been suggested, typically un-
der the rationale of implant design modifications that
would enable improved healing or biomechanical be-
havior.10,11,17,18,28 Although a substantial amount of
research has been devoted to increasing the host-to-
implant response at early implantation times,10,11,17,18,28,29
little information has been published to date concern-
ing the host-to-implant response considering the in-
terplay between surgical protocols and implant bulk
design.5 This investigation tested the null hypothesis
that different implant systems presenting distinct sur-
gical instrumentation, macrogeometry, and surface
would not present alterations in biomechanical fixa-

Table 5. STATISTICAL SUMMARY FOR IMPLANT

Table 4. IMPLANTATION TIME AND TORQUE (N-cm) GROUP TORQUE (N-cm) IN WHICH SIGNIFICANT
STATISTICAL SUMMARY IN WHICH SIGNIFICANT DIFFERENCES WERE OBSERVED BETWEEN GROUPS
DIFFERENCES WERE OBSERVED BETWEEN 1 AND 3 (P < .001)
WEEKS (P ⴝ .016)
95% CI 95% CI
95% CI 95% CI Implant Lower Upper
Time In Lower Upper Group Mean SE df Bound Bound
Vivo Mean SE df Bound Bound
Group 1 22.08 4.84 15c 11.75 32.41
b
1 Week 50.17 3.662 15 42.36 57.97 Group 2 49.94 4.84 15b 39.61 60.26
3 Weeks 65.28a 4.228 15 56.27 74.30 Group 3 101.16 4.84 15a 90.84 111.50
Abbreviations: CI, confidence interval; SE, standard error. Abbreviations: CI, confidence interval; SE, standard error.
The same superscript letters denote statistically homoge- The same superscript letters denote statistically homoge-
neous groups. neous groups.
Coelho et al. Biomechanical Evaluation of Endosseous Implants. Coelho et al. Biomechanical Evaluation of Endosseous Implants.
J Oral Maxillofac Surg 2010. J Oral Maxillofac Surg 2010.
1672 BIOMECHANICAL EVALUATION OF ENDOSSEOUS IMPLANTS

Table 6. TORQUE (N-cm) STATISTICAL SUMMARY FOR THE IMPLANT GROUP—TIME IN VIVO INTERACTION
TORQUE IN WHICH SIGNIFICANT DIFFERENCES WERE OBSERVED (P ⴝ .001)

95% CI 95% CI
Implant Group Time In Vivo Mean SE df Lower Bound Upper Bound

Group 1 1 Week 19.43b 6.34 15 5.91 32.95

3 Weeks 24.73b 7.32 15 9.124 40.34
Group 2 1 Week 23.48b 6.34 15 9.96 37.00
3 Weeks 76.40a 7.32 15 60.79 92.01
Group 3 1 Week 107.6a 6.34 15 94.08 121.12
3 Weeks 94.73a 7.32 15 79.124 110.34
Abbreviations: CI, confidence interval; SE, standard error.
The same superscript letters denote statistically homogeneous groups.
Coelho et al. Biomechanical Evaluation of Endosseous Implants. J Oral Maxillofac Surg 2010.

tion at early implantation times in a canine mandible

FIGURE 2. A, Optical micrograph of the bone-implant interface in
a group 1 thin section shows woven bone formation in proximity
with implant surface at 1 week. Bone cracking induced by implant
placement occurred in some large pitch threaded areas (arrow).
Note regions of initial woven bone formation and regions between
old bone and implant surface representative of interfacial bone
modeling/remodeling stages. B, At 3 weeks, newly formed woven
bone replacing the modeled/remodeled interfacial bone region,
resulting in a more continuous bone–implant structure.
Coelho et al. Biomechanical Evaluation of Endosseous Implants.
J Oral Maxillofac Surg 2010.
model.
The implant systems used in this study were com-
mercially available and presented distinct implant ge-
ometry and surfaces, as well as surgical instrumenta-
tion. Group 1 implant presented a moderately rough
surface with discrete hydroxyapatite crystalline dep-
osition at the nanometer scale,20 group 2 presented a
moderately rough surface presenting minute content
of fluoride obtained by a sequence of grit-blasting and
acid etching procedures,13 and group 3 presented a
moderately rough structure along with a nanometer
scale roughness pattern with calcium and phosphate
minute quantities, obtained by a grit-blasting me-
thod.19 All surfaces evaluated in the study have shown
superior performance compared with their moder-
ately rough predecessors at early times in different
animal models.13,19,20 From a geometric perspective,
the implants used in this study present commonly
used thread designs; microthreads were included in
groups 2 and 3 but not group 1.
Previous investigations using the same animal
model have described the early events of bone heal-
ing around implants of varied shapes and surfa-
ces.5,8,9,14 It has been demonstrated that a blood clot
in close proximity to the implant surface at the im-
plant surface region is established immediately after
implant placement, and over the next hours to several
days, a sequence of cellular and vascular events orig-
inates a granulation tissue with varied degrees of
inflammatory cell content.5,8 Then, the formation of
woven bone is detected at 1 week5,8 and substantial
interfacial remodeling is observed over the next 3 to
4 weeks5,8,9 until lamellar bone is observed in prox-
imity with the implant surface. Thus, the rationale for
choosing in vivo evaluation times of 1 and 3 weeks
was to evaluate the various implant groups’ interac-
tion with bone at times that would comprise the
COELHO ET AL 1673

FIGURE 3. A, Optical micrograph of a group 2 thin section bone–implant interface shows bone modeling/remodeling represented by areas
of woven bone formation and gaps in the space separating old bone and implant surface at 1 week. Bone cracking induced by implant
placement is observed at the microthreaded region (arrow) as well as (B) large pitch thread regions (arrow). C, At the microthreads and also
(D) at the large pitch thread regions, newly formed woven bone is shown at 3 weeks, filling the gaps present at 1 week, where intimate contact
between bone and implant surface is observed.
Coelho et al. Biomechanical Evaluation of Endosseous Implants. J Oral Maxillofac Surg 2010.

onset and an intermediate period of woven bone
deposition and interfacial remodeling in dog mandi-
bles.5,8,9
Our histologic observations are in direct agreement
with previous studies5,8,9 in which, irrespective of
experimental group, initial woven bone formation
along with bone remodeling at the interfacial region
was detected at 1 week and extended to 3 weeks.5,8,9
All groups evaluated presented higher amounts of
woven bone at the interfacial region at 3 weeks,
which may be explained by subsequent interfacial
bone remodeling and woven bone formation be-
tween 1 and 3 weeks. The observation of bone crack-
ing at the implant thread tip regions at 1 week for all
specimens and its near inexistence at 3 weeks sug-
gests the extent of the region of high bone activity
(formation and remodeling) temporally changed over
time. Such observation reflects the dynamic nature of
the bone–implant interface at early implantation
times despite the 3 systems’ differing surgical drilling
techniques, geometries, and surfaces.
The intimate contact between newly formed bone
and the 3 groups’ different surfaces suggest that they
are all highly biocompatible and osseoconductive.
Such observation is in direct agreement with previ-
ously published experimental animal studies.13,19,20
Although our previous investigations15,23-25 have mea-
sured bone-to-implant contact values from histologic
samples that were biomechanically tested with the
same methods used in the current investigation, those
studies15,23-25 used experimental implant designs that
allowed biomechanical testing on a plane (cylindrical
shaped implants without threads, unlike threaded im-
plants that change in position along their long axis
during testing, potentially causing higher interfacial
disruption).
In contrast with our histologic findings in which all
implant groups presented similar interaction with
1674 BIOMECHANICAL EVALUATION OF ENDOSSEOUS IMPLANTS

FIGURE 4. A, Optical micrograph of group 3 bone-implant interface shows woven bone formation in close proximity with implant surface
at 1 week at the microthread and (B) also at the large pitch threaded regions, where bone cracking is induced by implant placement (arrow).
C and D, At 3 weeks, both regions of microthreads (C) and also large pitch threads (D) present a continuous implant– bone structure bridged
by newly formed woven bone.
Coelho et al. Biomechanical Evaluation of Endosseous Implants. J Oral Maxillofac Surg 2010.

bone at 1 and 3 weeks in vivo, significant differences
were observed in biomechanical testing. The general
linear modeling (ANOVA) test detected a significant
increase of torque values over implantation time
when all implant groups were included. Significantly
higher torque values were also observed between
groups, with group 3, group 2, and group 1 ranked
highest, intermediate, and lowest, respectively. Al-
though these observations are informative because
the different implant systems were placed according
to their manufacturer’s instructions, it should be
noted that mainly because of the different geometric
configuration between systems, any attempt to com-
pare the torque values obtained in this study to stan-
dard insertion torque values commonly observed for
humans or to compare the torque values obtained
between experimental implant groups would be of
limited value.
Directly related to the null hypothesis, which stated
that no differences in biomechanical behavior would
be detected over time for all groups, our results
showed that 2 groups maintained their torque values
at the period of onset of woven bone formation and
interfacial bone remodeling and a period during
which substantial bone activity takes place in the dog
mandible model, rejecting the null hypothesis. Our
results, however, strongly suggest that the interplay
between surgical instrumentation, implant geometry,
and surface directly influences implant systems’ sta-
bility at early implantation times. However, because
this investigation was limited to only 2 evaluation
points in vivo, temporal stability decreases and in-
creases among systems may have been hindered.
Thus, experimental multivariable studies concerning
specific effects of implant design aspects over a larger
number of evaluation times are warranted.
Acknowledgments
This study was supported in part by the Department of Dentistry
at Universidade Federal de Santa Catarina, Brazil. The implants used
COELHO ET AL
were purchased from all manufacturers except Intra-Lock Interna-
tional, which donated the implants. The authors acknowledge
valuable discussions with Dr Robert J. Miller before and during the
preparation of the article.
References
1. Albrektsson T, Branemark PI, Hansson HA, et al: Osseointe-
grated titanium implants. Requirements for ensuring a long-
lasting, direct bone-to-implant anchorage in man. Acta Orthop
Scand 52:155, 1981
2. Chuang SK, Tian L, Wei LJ, et al: Kaplan-Meier analysis of dental
implant survival: A strategy for estimating survival with clus-
tered observations. J Dent Res 80:2016, 2001
3. Chuang SK, Wei LJ, Douglass CW, et al: Risk factors for dental
implant failure: A strategy for the analysis of clustered failure-
time observations. J Dent Res 81:572, 2002
4. Lemons JE: Biomaterials, biomechanics, tissue healing, and
immediate-function dental implants. J Oral Implantol 30:318,
2004
5. Berglundh T, Abrahamsson I, Lang NP, et al: De novo alveolar
bone formation adjacent to endosseous implants. Clin Oral
Implants Res 14:251, 2003
6. Lemons J, Dietch-Misch F: Biomaterials for dental implants. In:
Misch CE, editor. Contemporary Implant Dentistry. Saint Louis,
MO, Mosby, 1999, pp 271-302
7. Bottino MC, Coelho PG, Henriques VA, et al: Processing, char-
acterization, and in vitro/in vivo evaluations of powder metal-
lurgy processed Ti-13Nb-13Zr alloys. J Biomed Mater Res A
88:689, 2009
8. Abrahamsson I, Berglundh T, Linder E, et al: Early bone forma-
tion adjacent to rough and turned endosseous implant surfaces.
An experimental study in the dog. Clin Oral Implants Res
15:381, 2004
9. Leonard G, Coelho P, Polyzois I, et al: A study of the bone
healing kinetics of plateau versus screw root design titanium
dental implants. Clin Oral Implants Res 20:232, 2009
10. Albrektsson T, Wennerberg A: Oral implant surfaces: Part
1—Review focusing on topographic and chemical properties
of different surfaces and in vivo responses to them. Int J
Prosthodont 17:536, 2004
11. Albrektsson T, Wennerberg A: Oral implant surfaces: Part
2—Review focusing on clinical knowledge of different sur-
faces. Int J Prosthodont 17:544, 2004
12. Coelho PG, Granjeiro JM, Romanos GE, et al: Basic research
methods and current trends of dental implant surfaces.
J Biomed Mater Res B Appl Biomater 88:579, 2009
13. Berglundh T, Abrahamsson I, Albouy JP, et al: Bone healing at
implants with a fluoride-modified surface: An experimental
study in dogs. Clin Oral Implants Res 18:147, 2007
1675
14. Coelho PG, Suzuki M, Guimaraes MV, et al: Early bone healing
around different implant bulk designs and surgical techniques:
A study in dogs. Clin Implant Dent Relat Res, 2009
15. Granato R, Marin C, Suzuki M, et al: Biomechanical and histo-
morphometric evaluation of a thin ion beam bioceramic de-
position on plateau root form implants: An experimental study
in dogs. J Biomed Mater Res B Appl Biomater 90:396, 2009
16. Albrektsson T, Sennerby L, Wennerberg A: State of the art of
oral implants. Periodontol 2000 47:15, 2008
17. Butz F, Aita H, Wang CJ, et al: Harder and stiffer bone osseointe-
grated to roughened titanium. J Dent Res 85:560, 2006
18. Buser D, Broggini N, Wieland M, et al: Enhanced bone apposi-
tion to a chemically modified SLA titanium surface. J Dent Res
83:529, 2004
19. Marin C, Granato R, Suzuki M, et al: Removal torque and
histomorphometric evaluation of bioceramic grit-blasted/acid-
etched and dual acid-etched implant surfaces: An experimental
study in dogs. J Periodontol 79:1942, 2008
20. Mendes VC, Moineddin R, Davies JE: The effect of discrete
calcium phosphate nanocrystals on bone-bonding to titanium
surfaces. Biomaterials 28:4748, 2007
21. Branemark PI, Hansson BO, Adell R, et al: Osseointegrated
implants in the treatment of the edentulous jaw. Experience
from a 10-year period. Scand J Plast Reconstr Surg Suppl 16:1,
1977
22. Raghavendra S, Wood MC, Taylor TD: Early wound healing
around endosseous implants: A review of the literature. Int
J Oral Maxillofac Implants 20:425, 2005
23. Coelho PG, Cardaropoli G, Suzuki M, et al: Early healing of
nanothickness bioceramic coatings on dental implants. An ex-
perimental study in dogs. J Biomed Mater Res B Appl Biomater
88:387, 2009
24. Coelho PG, Lemons JE: Physico/chemical characterization and
in vivo evaluation of nanothickness bioceramic depositions on
alumina-blasted/acid-etched Ti-6Al-4V implant surfaces. J Bio-
med Mater Res A 90:351, 2009
25. Granato R, Marin C, Gil JN, et al: Thin bioactive ceramic-coated
alumina-blasted/acid-etched implant surface enhances biome-
chanical fixation of implants: An experimental study in dogs.
Clin Implant Dent Relat Res Aug 3, 2009 [E-pub ahead of print]
26. Donath K, Breuner G: A method for the study of undecalcified
bones and teeth with attached soft tissues. The Sage-Schliff
(sawing and grinding) technique. J Oral Pathol 11:318, 1982
27. Branemark PI: Osseointegration and its experimental back-
ground. J Prosthet Dent 50:399, 1983
28. Coelho PG, Cardaropoli G, Suzuki M, et al: Early healing of
nanothickness bioceramic coatings on dental implants. An ex-
perimental study in dogs. J Biomed Mater Res B Appl Biomater
88:387, 2009
29. Yang Y, Kim KH, Ong JL: A review on calcium phosphate
coatings produced using a sputtering process—An alternative
to plasma spraying. Biomaterials 26:327, 2005