Influence of Microgap Location and Configuration

on Peri-implant Bone Morphology in

Nonsubmerged Implants:
An Experimental Study in Dogs
Dietmar Weng, DDS, Dr Med Dent1/Maria José Hitomi Nagata, DDS, MSc, PhD2/
Matthias Bell, DDS, Dr Med Dent3/Luiz Gustavo Nascimento de Melo, DDS, MSc, PhD4/
Alvaro Francisco Bosco, DDS, MSc, PhD2

Purpose: It is unknown whether different microgap configurations can cause different peri-implant bone
reactions. Therefore, this study sought to compare the peri-implant bone morphologies of two implant
systems with different implant-abutment connections. Materials and Methods: Three months after
mandibular tooth extractions in six mongrel dogs, two oxidized screw implants with an external-hex con-
nection were inserted (hexed group) on one side, whereas on the contralateral side two grit-blasted
screw implants with an internal Morse-taper connection (Morse group) were placed. On each side, one
implant was inserted level with the bone (equicrestal) and the second implant was inserted 1.5 mm
below the bony crest (subcrestal). Healing abutments were inserted immediately after implant place-
ment. Three months later, the peri-implant bone levels, the first bone-to-implant contact points, and the
width and steepness of the peri-implant bone defects were evaluated histometrically. Results: All 24
implants osseointegrated clinically and histologically. No statistically significant differences between the
hexed group and Morse group were detected for either the vertical position for peri-implant bone levels
(Morse equicrestal –0.16 mm, hexed equicrestal –0.22 mm, Morse subcrestal 1.50 mm, hexed subcre-
stal 0.94 mm) or for the first bone-to-implant contact points (Morse equicrestal –2.08 mm, hexed
equicrestal –0.98 mm, Morse subcrestal –1.26 mm, hexed subcrestal –0.76 mm). For the parameters
width (Morse equicrestal –0.15 mm, hexed equicrestal –0.59 mm, Morse subcrestal 0.28 mm, hexed
subcrestal –0.70 mm) and steepness (Morse equicrestal 25.27 degree, hexed equicrestal 57.21 degree,
Morse subcrestal 15.35 degree, hexed subcrestal 37.97 degree) of the peri-implant defect, highly sig-
nificant differences were noted between the Morse group and the hexed group. Conclusion: Within the
limits of this experiment, it can be concluded that different microgap configurations influence the size
and shape of the peri-implant bone defect in nonsubmerged implants placed both at the crest and sub-
crestally. INT J ORAL MAXILLOFAC IMPLANTS 2010;25:540–547

Key words: bone morphology, equicrestal placement, histometric study, microgap, subcrestal placement

mplant therapy has become a very predictable and
I highly successful treatment option to restore eden-
tulous sites.1 The more successful it has become, the
1Visiting Scholar, Department of Prosthodontics, Propaedeutics,
and Dental Materials, School of Dentistry, Christian-Albrechts-Uni-
versity at Kiel, Germany; Private Practice, Starnberg, Germany.
2Associate Professor, Division of Periodontics, Department of
Surgery and Integrated Clinic, Dental School of Araçatuba,
UNESP–Universidade Estadual Paulista, Araçatuba, Brazil.
3Private practice, Gießen, Germany.
4Private practice, Goiania, Brazil.
Correspondence to: Dr Dietmar Weng, Private Practice Böhm &
Weng, Maximilianstraße 17, 82319 Starnberg, Germany.
Fax: +49-8151-652511. Email: dw@max-17.de
more attention has been paid to factors that may
alter or improve the implant-host interaction.2–5 Cur-
rently the connection between implant and abut-
ment is a focus of research.6 There seems to be a shift
from external to internal connection types in most
implant systems, based on the knowledge that differ-
ent connection types influence microbial leakage7
and mechanical stability,8 factors that, if left uncon-
trolled, can cause a series of peri-implant hard and
soft tissue reactions.
The microgap is the connection line between an
implant and an abutment. This gap has been a matter
of intense investigation during the last 10 years. Sev-
eral radiographic and histologic animal studies of sub-
merged and nonsubmerged implants with butt-joint

540 Volume 25, Number 3, 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Fig 1 (Left) Equicrestal insertion of an
implant of the hexed group. The implant
shoulder is level with the surrounding bone.
Fig 2 (Right) A Morse group implant
inserted 1.5 mm subcrestally.
connections came to the conclusions that (1) the first
bone-to-implant contact settles at a vertical radiologic
distance of 2 mm from the microgap, irrespective of
the location of the microgap relative to the surround-
ing bone level at implant insertion9; (2) the bone-to-
implant contact histologically keeps a distance of 1.3
to 2.6 mm from the microgap depending on the loca-
tion of the microgap relative to the surrounding bone
level at implant insertion10,11; and (3) the actual size of
the microgap itself does not influence the amount of
peri-implant bone resorption, as long as micromove-
ment does not become an additional factor.11,12 These
radiographic and histologic findings have been
observed on radiographs for more than 20 years as
“dish-shaped” defects around the implant shoulder
and termed “peri-implant bone remodeling (down to
the first thread).” They were taken into account for the
definition of the implant success criteria from 198613
and reinforced in 199814 by eliminating from consid-
eration the amount of peri-implant bone loss in the
vertical dimension prior to the first year of loading as
a relevant factor to an implant’s success.
If seen from the perspective of implant function, the
amount of peri-implant bone loss during the healing
and early loading phases might not be a key factor in
the overall success or survival rate of a given implant.
Seen from the esthetic viewpoint, however, any loss of
peri-implant hard tissue might signal the beginning of
peri-implant soft tissue collapse, since bony support of
soft tissues is still considered the most important
determinant in peri-implant soft tissue stability.
It was the objective of this histometric experiment
in dogs to study the influence of different vertical
microgap locations on the peri-implant bone reac-
tions in two different implant-abutment connection
types after nonsubmerged healing.
© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Weng et al
MATERIALS AND METHODS
The protocol of this animal study was approved by the
Ethical Committee for Animal Investigations of the
Dental School of Araçatuba, UNESP–Universidade
Estadual Paulista, Araçatuba, Brazil. All surgical proce-
dures were performed under general anesthesia using
atropine sulfate, xylazine, tiletamine/zolazepam, and
lidocaine with epinephrine. Perioperatively, antibiotics
(espiramizine/metronidazol) and non-steroidal anti-
inflammatory drugs (flunixine/meglumine) were applied.
During early wound healing, chlorhexidine diglu-
conate rinses were administered three times per week.
Sample size calculations were based on the pre-
sumption that a mean difference of 0.5 mm with a
standard deviation of 0.5 should be detected at a sig-
nificance level of .05, warranting a statistical power of
80%.
Surgery
Six mongrel dogs were utilized in this investigation. At
the beginning of the study, all mandibular premolars
(P1 to P4) and the first mandibular molar (M1) were
extracted bilaterally. After extraction, the sockets were
left untreated for 3 months. Full mucoperiosteal flaps
were then raised, and the ridges were flattened with
surgical burs to obtain a buccolingual width of 6 to
7 mm. Thereafter, osteotomies for two implants were
drilled on each side of the mandible according to the
implant manufacturers’ protocols. On one side, two
screw-type implants with an oxidized surface and an
external-hex connection were placed (TiUnite Bråne-
mark Mk III, Nobel Biocare; diameter 3.75 mm, length
8.5 mm, smooth collar height 0.75 mm; hexed group)
in such a manner that the shoulder of one implant was
located at the same level as the surrounding bone
(equicrestal position) (Fig 1), whereas the shoulder of
The International Journal of Oral & Maxillofacial Implants 541
 Weng et al
HBD
PBL
MG
SLO BICP
the second implant was placed 1.5 mm below the level
of the surrounding bone (subcrestal position). The
same procedure was carried out on the contralateral
side of the mandible, but two implants with a smooth
collar, a grit-blasted body, and an internal Morse-taper
connection were inserted (Ankylos A8, Dentsply Fri-
adent; diameter 3.5 mm, length 8 mm, smooth collar
height 1.5 mm; Morse group) (Fig 2). In the six animals,
left and right sides were alternated between implant
systems as well as the anterior and posterior positions
of the subcrestal and equicrestal groups. On each side,
however, only one implant system was used, and the
vertical positions of the implant shoulders were the
same on each side of the mandible. Finally, healing
abutments were attached manually to the implants
according to the manufacturers’ guidelines. The heal-
ing abutments of the subcrestally inserted implants
were chosen to be 1.5 to 2 mm higher than those for
the equicrestal implants so as to keep the tops of the
healing abutments at a similar height above the
mucosa. The flaps were then adapted around the heal-
ing abutments and sutured. The implants were left in
this nonsubmerged healing mode for 3 months.
Sutures were removed after 1 week. Swabbing and
rinsing with chlorhexidine digluconate was carried out
three times per week.
542 Volume 25, Number 3, 2010
© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Fig 3 (Left) An implant of the hexed group
presenting with healthy peri-implant soft
tissue conditions after 3 months of non-
submerged healing.
Fig 4 (Right) An implant of the Morse group
showing uninflamed peri-implant soft tissue
structures after 3 months of nonsubmerged
healing.
Fig 5 Example of parameter assessment. MG (white line) =
microgap level; PBL = peri-implant bone levels; BICP = first bone-
to-implant contact point; HBD = width of the peri-implant bone
defect; SLO = steepness of the peri-implant bone defect.
Histology
After 3 months of healing (Figs 3 and 4) the animals
were sacrificed by exsanguination. Two to three
mesiodistally cut ground sections were produced
from each implant and subsequently stained with
toluidine blue solution.15 The following parameters
were evaluated histometrically (Fig 5):
1. Peri-implant bone level (PBL): vertical distance
between the most coronal point of the peri-
implant bone and the implant shoulder; expressed
as a positive value if the peri-implant bone was
located coronal to the implant shoulder
2. Bone-to-implant contact point (BICP): vertical dis-
tance between the most coronal bone-to-implant
contact point and the implant shoulder; expressed
as a negative value if the bone-to-implant contact
point was located apical to the implant shoulder
3. Horizontal bone distance (HBD): horizontal dis-
tance between the most coronal point of the peri-
implant bone and a vertical line along the outer
implant surface; expressed as a negative value
unless bone had grown onto the implant shoulder
4. Peri-implant bone slope (SLO): angle between a
line extending along the peri-implant bone defect
and a vertical line along the outer implant surface

Fig 6 (Left) Implant of the equicrestal
Morse group. Note the overall maintenance
of PBL and the tight cuff (small SLO value) of
the peri-implant bone around the microgap
area. Red dots indicate microgap. Yellow line
shows bone level at implant placement.
Fig 7 (Right) Implant of the equicrestal
hexed group. Note the typical “dish-shaped”
character of the bone defect extending to the
first implant thread. Red dots = microgap;
yellow line = bone level at implant placement.
1 mm
Statistics
Each parameter was measured on both the mesial
and distal sides of a histologic section and summa-
rized to result in a mean value per section. These
means of the two to three sections available per
implant were then summarized to obtain a mean per
implant. Therefore the statistical unit was the implant
(equal to the animal). Comparisons of the two
implant systems within the same vertical groups and
of the same vertical positions within the same
implant system were made by using paired t tests.
RESULTS
Clinical Results
The 3-month healing period was uneventful. All
implants osseointegrated clinically and did not show
any signs of mobility at the time of sacrifice. Loosen-
ing of the healing abutments was not noted. The clin-
ical peri-implant tissue condition was characterized
by healthy or occasional slightly inflamed mucosal
margins.
Histologic Findings
Regardless of the implant system and the vertical
© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Weng et al
1 mm
insertion depth of the implants, the histologic sec-
tions presented well-osseointegrated implants in
mature bone structures (Figs 6 to 9). When the
implants were inserted subcrestally, the PBLs were
located coronal to the implant shoulder (Figs 8 and 9).
With equicrestal placement, PBL usually was found at
or slightly below the implant shoulder (Figs 6 and 7).
In most of the sections, BICP was located apical to the
implant shoulder. Only around implants from the sub-
crestally placed Morse group was bone tissue found
overgrowing the implant shoulder and establishing
the first bone-to-implant contact on the crestal face of
the implant shoulder (Fig 8). The horizontal extent of
the peri-implant bone defect and its “dish-shaped”
character were more pronounced in the hexed group,
especially when placed subcrestally (Fig 9). The “fun-
nel” of bone around the implants was narrower in the
Morse groups, with regular bone growth over the
implant shoulder when the Morse implants were
inserted subcrestally (Fig 8).
Histometric Measurements
Means, standard deviations, and P values of the four
histometric parameters are summarized in Table 1. A
graphic depiction of those results can be found in
Figs 10 and 11. Between the Morse group and the
The International Journal of Oral & Maxillofacial Implants 543
 Weng et al

Fig 8 (Left) Implant of the subcrestal

Morse group. Note the narrow “funnel”
establishing close proximity between the
peri-implant bone, the healing abutment,
and the microgap area. Bone tissue has
grown onto the implant shoulder. Red dots
= microgap; yellow line = bone level at
implant placement.

Fig 9 (Right) Implant of the subcrestal

hexed group. Note the establishment of a
classical “dish-shaped” defect around the
microgap-abutment complex. The peri-
implant bone tissue seems to avoid direct
contact with the microgap. Red dots =
microgap; yellow line = bone level at
implant placement.

1 mm

1 mm

Table 1 Histometric Measurements (Means ± Standard Deviations)

Group (n = 6 each) PBL (mm) BICP (mm) HBD (mm) SLO (deg)

Equicrestal Morse group

Mean –0.16 ± 0.83 –2.08 ± 1.20 –0.15 ± 0.16* 25.27 ± 25.20**
Range –0.61 to 1.38 –0.86 to 4.31 –0.42 to 0.05 8.33 to 76.01
Equicrestal hexed group
Mean –0.22 ± 0.80 –0.98 ± 0.41 –0.59 ± 0.16 57.21 ± 29.84
Range –1.44 to 0.76 –1.52 to –0.52 –0.39 to –0.75 33.76 to 113.94
Subcrestal Morse group
Mean 1.50 ± 0.66 –1.26 ± 1.48 0.25 ± 0.22** 15.35 ± 11.70*
Range 0.61 to 2.38 –4.09 to 0.25 –0.07 to 0.47 9.07 to 39.06
Subcrestal hexed group
Mean 0.94 ± 0.44 –0.76 ± 0.41 –0.70 ± 0.28 37.91 ± 18.57
Range 0.35 to 1.44 –1.53 to –0.32 –1.13 to –0.32 22.76 to 72.35
PBL = peri-implant bone level; BICP = bone-to-implant contact point; HBD = horizontal bone distance of the peri-implant defect;
SLO = peri-implant bone slope of the peri-implant defect.
*P < .005 versus hexed group placed at same level; **P < .001 versus hexed group placed at same level; all other values P > .05
between groups placed at same level.

hexed group, statistically significant differences were
detected for the parameters HBD and SLO in the
equicrestal and subcrestal groups. Comparisons of
the two vertical positions within the same implant
system group (with regard to BICP, HBD, and SLO)
showed a significant difference for HBD (P < .05) in
the Morse implants and no differences for the hexed
implants. A comparison of PBL was not carried out
because differences in the height of the peri-implant
bone can be expected whenever the implants are
placed in different vertical positions.

544 Volume 25, Number 3, 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Fig 10 (Left) Illustration of the average
bone morphology for the Morse group (left
side) and the hexed group (right side) in
equicrestally placed implants. Drawing is
based on the mean values of the parame-
ters PBL, BICP, and HBD. The horizontal line
separating the orange from the red area
delineates the bone level at the time of
implant placement.
Fig 11 (Right) Illustration of the average
bone morphology for the Morse group (left
side) and the hexed group (right side) in
subcrestally placed implants. Drawing is
based on the mean values of the parame-
ters PBL, BICP, and HBD. The horizontal line
separating the orange from the red area
delineates the bone level at the time of
implant placement.
1 mm
DISCUSSION
Regardless of the implant system used or the vertical
position of the microgap, it was generally observed
that the level of the peri-implant bone was main-
tained quite well. The PBL was kept close to 0 mm in
the equicrestal group and close to 1.5 mm in the sub-
crestal group. Although the subcrestal hexed group
presented a difference of more than 0.5 mm in PBL
compared to the Morse group, the difference was not
significant. Despite these stable results for PBL, it may
be noteworthy that the BICP established itself consid-
erably below the microgap in both the equicrestal
group (hexed group –0.98 mm, Morse group –2.08
mm) as well as in the subcrestal group (hexed group
–0.76 mm, Morse group –1.26 mm). The equicrestal
BICP difference of 1.10 mm was not significant
because of the higher standard deviation in the Morse
group, which was caused by an outlier. Without this
outlier, the BICP value for the equicrestal Morse group
would have amounted to a mean of 1.63 mm with a
standard deviation of 0.55 mm. Nevertheless, it seems
that BICP in equicrestally inserted, nonsubmerged
implants with a smooth collar can be found around 1
to 1.5 mm below the microgap. This finding is con-
firmed by Hermann et al,10 who demonstrated a BICP
of –1.68 mm in equicrestal nonsubmerged implants.
© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Weng et al
1 mm
Also, Piattelli et al16 observed the BICP at –2.11 mm in
equicrestal, nonsubmerged, titanium plasma–sprayed
implants. However, when they inserted the implants 1
to 1.5 mm subcrestally, Piattelli et al16 found a BICP of
–3.65 mm. A possible explanation might be that their
implants underwent early and immediate loading.
Considerably better BICP values were found by Abra-
hamsson et al 17; their equicrestal, nonsubmerged
implants exhibited a BICP of –0.68 mm after a healing
time of 6 months. A reason for their finding might be
the fact that they used implants with a rough surface
that extended to the implant shoulder, without a
smooth collar. This may be a hint that microroughen-
ing of implants to the level of the microgap might
help to decrease the distance between the implant
shoulder and the BICP in equicrestally placed
implants. The fact that in the present study BICP val-
ues were better (although insignificantly so) for both
equicrestally and subcrestally placed implants in the
hexed groups (0.75-mm smooth collar height) than in
the Morse groups (1.5 mm smooth collar height)
lends support to this idea.
Pronounced differences were noticed when the
horizontal parameters HBD and SLO were compared.
These parameters measure how close the peri-
implant bone is to the microgap area in a horizontal
direction. The most coronal point of the peri-implant
The International Journal of Oral & Maxillofacial Implants 545
 Weng et al
bone was chosen as the reference point for HBD. In
histologic sections with an asymptotic shape (eg, Fig
5, right side), this might overestimate HBD. As an alter-
native, the bone tangents as described by Gomez-
Roman et al18 could have been used. However, the
use of the most coronal point accounts for the fact
that the original bone level was a flattened ridge. In
both the equicrestal group and the subcrestal group,
the bone was four times closer to the implant-abut-
ment connection in Morse implants versus hexed
implants. This difference in HBD was statistically
highly significant. Earlier studies did not evaluate the
horizontal effects of the implant-abutment connec-
tion. Only Tarnow et al19 assessed HBD radiographi-
cally in 36 patients 1 to 3 years after stage-two
surgery in implants with a microgap configuration
similar to that used in the present external-hex
implants. Tarnow et al observed considerably higher
HBD values (–1.34 mm and –1.40 mm) than in the
present study, but the earlier study described a clini-
cal situation several years after loading. As a conse-
quence of the low HBD values in the Morse groups,
SLO, ie, the angulation of the peri-implant bone
defect, was only half as big in the Morse group as in
the hexed group, again at a high significance level.
Obviously, the implant connection type of the hexed
group (non–Morse taper butt joint with no horizontal
offset) seemed to exert a more horizontally pro-
nounced influence on the peri-implant bone tissue
than the connection type of the Morse group (Morse-
taper connection with horizontal offset). For both the
Morse and hexed groups, SLO was 10 to 20 degrees
smaller when a subcrestal insertion mode was cho-
sen compared to an equicrestal placement. However,
this difference was not statistically significant.
Another significant difference was found when
subcrestal insertion was compared to equicrestal
insertion in the Morse group with regard to HBD (0.25
mm versus –0.15 mm). The low standard deviation of
0.22 mm in the subcrestal group showed that, in
almost all the sections of the Morse group, bone had
grown onto the crestal face of the implant shoulder
when they were inserted apical to the bone level. It
seemed that the peri-implant bone accepted the
space offered by the horizontal offset and
approached the implant-abutment interface by
actively growing into this void. Interestingly enough,
the remaining horizontal distance between the bone
and the healing abutment was always smaller in the
subcrestally placed Morse group than in the subcre-
stally inserted hexed group; with a manufactured hor-
izontal offset of 0.5 mm and a mean HBD of 0.25 mm,
the mean distance between bone and healing abut-
ment amounted to only 0.25 mm, compared to the
mean HBD of –0.70 mm seen in the subcrestal hexed
546 Volume 25, Number 3, 2010
© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
group. Of course the question arises whether the hori-
zontal offset between implant and healing abutment
is the only factor accounting for this difference. As
shown by Zipprich et al,8,20 mechanical micromove-
ment is extremely low in Morse taper–connected
implant systems, which may allow for reduced micro-
bial contamination of the microgap and subsequently
lower concentrations of bacterially produced bone-
resorbing agents. Furthermore, Hansson21,22 was able
to show that crestal bone receives less stress upon
axial loading if the implant-abutment connection has
a Morse-taper design. The overall impression in the
subcrestal as well as in the equicrestal Morse group
was that of a considerably narrower “dish-shaped”
defect compared to the hexed groups.
Because in the present study the healing abut-
ments were not repeatedly loosened and retightened
to simulate prosthetic procedures (impression, try-in,
crown insertion), one may ask whether such proce-
dures would have a worsening effect on the appear-
ance and shape of the “bone funnel” around the
implant. In the studies of Hermann et al,9,10 the non-
submerged, equicrestally placed implant type C
(butt-joint connection without horizontal offset) was
subjected to a triple loosening and retightening pro-
cedure of the healing abutment, and histologic exam-
ination after 6 months showed a typical dish-shaped
defect. This defect and the corresponding loss of peri-
implant bone contact (ie, BICP value), however, were
already detectable in the very first radiograph 4
weeks after implant placement, ie, 3 months before
the first loosening and retightening procedure was
performed. Furthermore, the triple unscrewing and
screwing procedures, done 4, 5, and 5.5 months after
implant placement, did not seem to produce addi-
tional detrimental effects on the peri-implant bone.
Therefore, it may be concluded that the bone-resorb-
ing effect of a micromechanically and microbiologi-
cally challenged implant-abutment connection
balances out the possibly detrimental effects of
repeated abutment changes.
CONCLUSIONS
Within the limits of this animal experiment it can be
concluded that, 3 months after insertion of nonsub-
merged implants: (1) the resorption of the original
peri-implant bone height can be kept within 0.5 mm,
(2) the first bone-to-implant contact establishes itself
1 to 1.5 mm apical to the implant shoulder regardless
of the vertical position of the microgap, and (3) the
“dish-shaped” defect configuration is more pro-
nounced in a non–Morse taper, butt-joint connection
without horizontal offset. The clinical consequences of

these observations might be that the amount and
shape of a peri-implant “dish-shaped” bone defect
depend on the implant-abutment connection (espe-
cially if the implant is placed subcrestally), and that
narrowing the “bone funnel” of such defects might be
beneficial for the peri-implant soft tissue support.
ACKNOWLEDGMENTS
The authors wish to express their gratitude to Mrs Waltraut Schnei-
der for the preparation of the histologic sections in this study. This
study was supported by a grant from Dentsply Friadent.
REFERENCES
1. Pjetursson BE, Tan K, Lang NP, Brägger U, Egger M, Zwahlen M.
A systematic review of the survival and complication rates of
fixed partial dentures (FPDs) after an observation period of at
least 5 years. I. Implant-supported FPDs. Clin Oral Implants Res
2004;15:625–642.
2. Wallace SS, Froum SJ. Effect of maxillary sinus augmentation
on the survival of endosseous dental implants. A systematic
review. Ann Periodontol 2003;8:328–343.
3. Graziani F, Donos N, Needleman I, Gabriele M, Tonetti M. Com-
parison of implant survival following sinus floor augmenta-
tion procedures with implants placed in pristine posterior
maxillary bone: A systematic review. Clin Oral Implants Res
2004;15:677–682.
4. Hinode D, Tanabe S, Yokoyama M, Fujisawa K, Yamauchi E,
Miyamoto Y. Influence of smoking on osseointegrated implant
failure: A meta-analysis. Clin Oral Implants Res 2006;17:473–478.
5. Shalabi MM, Gortemaker A, van't Hof MA, Jansen JA, Creugers
NH. Implant surface roughness and bone healing: A system-
atic review. J Dent Res 2006;85:496–500.
6. Piermatti J, Yousef H, Luke A, Mahevich R, Weiner S. An in vitro
analysis of implant screw torque loss with external hex and
internal connection implant systems. Implant Dent 2006;15:
427–435.
7. Dibart S, Warbington M, Su MF, Skobe Z. In vitro evaluation of
the implant-abutment bacterial seal: The locking taper system.
Int J Oral Maxillofac Implants 2005;20:732–737.
8. Zipprich H, Weigl P, Lange B, Lauer HC. Erfassung, ursachen
und folgen von mikrobewegungen am implantat-abutment-
interface. Implantol 2007;15:31–46.
9. Hermann JS, Cochran DL, Nummikoski PV, Buser D. Crestal
bone changes around titanium implants. A radiographic eval-
uation of unloaded nonsubmerged and submerged implants
in the canine mandible. J Periodontol 1997;68:1117–1130.
© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Weng et al
10. Hermann JS, Buser D, Schenk RK, Cochran DL. Crestal bone
changes around titanium implants. A histometric evaluation
of unloaded non-submerged and submerged implants in the
canine mandible. J Periodontol 2000;71:1412–1424.
11. Hermann JS, Schoolfield JD, Schenk RK, Buser D, Cochran DL.
Influence of the size of the microgap on crestal bone changes
around titanium implants. A histometric evaluation of
unloaded non-submerged implants in the canine mandible.
J Periodontol 2001;72:1372–1383.
12. King GN, Hermann JS, Schoolfield JD, Buser D, Cochran DL.
Influence of the size of the microgap on crestal bone levels in
non-submerged dental implants: A radiographic study in the
canine mandible. J Periodontol 2002;73:1111–1117.
13. Albrektsson T, Zarb G, Worthington P, Eriksson AR. The long-
term efficacy of currently used dental implants: A review and
proposed criteria of success. Int J Oral Maxillofac Implants
1986;1:11–25.
14. Zarb GA, Albrektsson T. Consensus report: Towards optimized
treatment outcomes for dental implants. Int J Prosthodont
1998;11:389.
15. Donath K, Breuner G. A method for the study of undecalcified
bones and teeth with attached soft tissues. J Oral Pathol 1982;
11:318–326.
16. Piattelli A, Vrespa G, Petrone G, Iezzi G, Annibali S, Scarano A.
Role of the microgap between implant and abutment: A retro-
spective histologic evaluation in monkeys. J Periodontol 2003;
74:346–352.
17. Abrahamsson I, Berglundh T, Moon IS, Lindhe J. Peri-implant
tissues at submerged and non-submerged titanium implants.
J Clin Periodontol 1999;26:600–607.
18. Gomez-Roman G, d’Hoedt B, Axmann D, Schulte W. Visual-met-
ric measurement of peri-implant bone defects on radiographs.
Z Zahnärztl Implantol 1996;12:104–109.
19. Tarnow DP, Cho SC, Wallace SS. The effect of inter-implant dis-
tance on the height of inter-implant bone crest. J Periodontol
2000;71:546–549.
20. Zipprich H, Weigl P, Fischbach-Sedlatschek S, Lauer HC. Failure
mode of implant-abutment connections after horizontal
cyclic loading. Int Poster J Dent Oral Med 2004;6:238.
21. Hansson S. Implant-abutment interface: Biomechanical study
of flat top versus conical. Clin Implant Dent Relat Res 2000;2:
33–41.
22. Hansson S. A conical implant-abutment interface at the level
of the marginal bone improves the distribution of stresses in
the supporting bone. An axisymmetric finite element analysis.
Clin Oral Implants Res 2003;14:286–293.
The International Journal of Oral & Maxillofacial Implants 547