Received: 19 March 2018 | Revised: 31 May 2018 | Accepted: 4 June 2018

DOI: 10.1111/clr.13339

ORIGINAL RESEARCH

Peri‐implant tissue healing at implants with different designs

and placement protocols: An experimental study in dogs

Monika Oskarsson | Motohiro Otsuki | Maria Welander | Ingemar Abrahamsson

Department of Periodontology, Institute

of Odontology, the Sahlgrenska Academy Abstract
at University of Gothenburg, Gothenburg, Objectives: To compare the peri‐implant tissue response to subcrestal or crestal
Sweden
placement of implants with or without “platform‐switched” implant/abutment
Correspondence connections.
Ingemar Abrahammson, Department of
Periodontology, Institute of Odontology, Material and methods: On each side of the mandible in six Labrador dogs, two “plat‐
The Sahlgrenska Academy at University of form‐switched” (PS) implants and two implants with matching implant/abutment di‐
Gothenburg, Medicinaregatan 12F, PO Box
450, SE 405 30 Gothenburg, Sweden. ameters (PM) were placed in a crestal or a 1.5 mm subcrestal position. Sulcus formers/
Email: ingemar.abrahamsson@odontologi. abutments were connected, baseline radiographs obtained and the distance between
gu.se
the abutment‐fixture junction (A/F) and the marginal bone level (B) was determined.
Funding information Five months later, radiographic and clinical examinations were repeated and biopsies
Dentsply Implants Manufacturing GmbH,
Grant/Award Number: DF1810-1-299-2- were obtained and processed for histological analysis.
1; TUA Research Funding, Grant/Award Results: The radiographic analysis showed that marginal bone level changes occurred
Number: TUAGBG-363841
during the first observation period (13 weeks). PS‐implants showed minor marginal
bone level alterations during the full observation period (20 weeks), regardless of
placement depth. For PM‐implants, placement had a significant impact on bone level
changes; crestal placement resulted in a mean bone loss of 0.4 mm and subcrestal
placement in a 1.5 mm bone loss. The histometric measurements resulted in a dis‐
tance A/F‐B of 0.6 and 0.0 mm for PS implants, placed in a crestal or subcrestal posi‐
tion, respectively. The corresponding distance at PM‐implants was 1.5 mm
irrespective of placement protocol. The peri‐implant mucosa was thicker at subcre‐
stally placed implants as compared to a crestal placement.
Conclusions: Bone remodeling occurred within the early healing phase. During the
full 20 weeks of observation, PS‐implants demonstrated significantly less crestal
bone loss compared to PM‐implants.

KEYWORDS
animals, bone level, dental implants, histology, platform switching, radiographs, subcrestal
placement

1 | I NTRO D U C TI O N is essential for the preservation of the marginal bone support

and osseointegration. A healthy mucosa with proper dimensions
The peri‐implant soft tissue attachment is a well‐recognized con‐ is also a prerequisite for an optimal esthetic outcome. The early
cept and is established during healing after dental implant surgery healing of peri‐implant soft and hard tissues as well as the main‐
(Abrahamsson, Berglundh, & Lindhe, 1997; Berglundh & Lindhe, tenance of a healthy soft tissue barrier may, however, influence at
1996; Goldberg, Higginbottom, & Wilson, 2001; Stanford, 2002). which position the soft tissue attachment will become established
It serves as a protective barrier against the oral environment and (Yeung, 2008).

Clin Oral Impl Res. 2018;1–8. wileyonlinelibrary.com/journal/clr © 2018 John Wiley & Sons A/S. | 1
Published by John Wiley & Sons Ltd
2 | OSKARSSON et al.

The early healing of the peri‐implant tissues is influenced by host of Gothenburg. The animals were fed a soft diet two times per day
factors, surgical technique, and implant‐design. The choice of implant with free access to water. They also had free access to an outdoor
system determines, in the majority of cases, the surgical technique exercise yard during daytime. ARRIVE guidelines (Kilkenny, Browne,
to be applied. With regard to implant systems, a number of charac‐ Cuthill, Emerson, & Altman, 2011) were followed. During all the sur‐
teristics and designs have been claimed to affect marginal bone level gical procedures, general anesthesia was administered intravenously
preservation and the maintenance of a healthy soft tissue barrier. In with propofol (Diprivan, 10 mg/ml, 0.6 ml/kg) and sustained with
addition, the design of the trans‐mucosal part of the implant may in‐ N2O:O2 (1:1.5–2) and isoflurane employing endotracheal intubation.
fluence the preservation of soft tissue health during function.
The idea behind “platform‐switching” (PS; nonmatching implant/
2.2 | Implant placement and follow‐up
abutment diameters) is that the diameter of the trans‐mucosal abut‐
ment is smaller than that of the implant shoulder. The length of the To provide recipient sites for implants, all mandibular premolars and
interface between the soft tissue and the abutment/implant surface the 1st, 2nd, and 3rd maxillary premolars were extracted bilaterally.
will thus increase without influencing the vertical dimension of the Three months later, mucoperiosteal flaps were elevated in both sides
soft tissues. This feature is proposed to promote favorable conditions of the mandible. After osteotomy preparations, four implants were
for the establishment of a soft tissue attachment with minimal re‐ placed on each side using a nonsubmerged technique. A total of 48
sorption of marginal bone during healing (Baffone et al., 2012; Guerra implants were placed. In each side of the mandible, two implants
et al., 2014; Lazzara & Porter, 2006; Telleman, Meijer, Vissink, & of the Ankylos® C/X system (implant A; Ankylos® C/X A11, 3.5,
Raghoebar, 2012) and hence, minimal soft tissue recessions after the 11 mm long; DENTSPLY Implants Manufacturing GmbH, Mannheim,
healing phase. In addition, a 1–2‐mm subcrestal placement of implants Germany) and two Nobel Replace Tapered Groovy implants (implant
with a PS design has been proposed to promote bone formation on N; 3.5 mm, length 10 mm; Nobel Biocare, Kloten, Switzerland) were
the implant shoulder and to favor soft tissue healing and stability placed. While implants of type A had an internal conical Morse taper
(Huang et al., 2012). and a PS design of the implant/abutment connection, implants of
The overall aim of this experiment was to study soft and hard type N had an internal tri‐channel and a PM design of the connec‐
tissues around implants with “platform‐switching” (PS) and “plat‐ tion between the implant and abutment (Figure 2). The drilling pro‐
form‐matched” (PM) implant/abutment connections, placed in either tocols used followed the guidelines for each implant type. In each
a crestal or subcrestal position and maintained with or without oral side of the mandible, one pair of implants (one of each type) was
hygiene for 5 months. placed with the abutment/implant connection in a crestal position,
while the other pair of implants (one of each type) was placed with
the connection in a 1.5 mm subcrestal position. The placement
2 | M ATE R I A L A N D M E TH O DS sequence was randomized. Sulcus formers/abutments (Ankylos®
Gingiva Former/C [4.2, GH4.5] or Healing Abutment NobRpl NP
[3.5 × 5 mm]) were connected to the implants. Flaps were adapted
2.1 | Animals
and sutured.
Six female 12‐month‐old destination bred Labrador dogs (mean The sutures were removed 2 weeks later and a plaque control
weight 30 kg) were used. The outline of the study is shown in Figure 1. program consisting of implant cleaning using a toothbrush 5 days/
The regional Ethics Committee for Animal Research, Gothenburg, week was started. Baseline radiographs were obtained using a cus‐
Sweden, approved the study protocol (Dnr 138–2011). The ex‐ tom‐made film‐holder device (Hawe Super Bite; Hawe Neos Dental,
periment was conducted in 2011 and 2012 at the Laboratory for Bioggio, Switzerland). The radiographs were analyzed using an
Experimental Bio‐Medicine at the Sahlgrenska Academy, University Olympus SZH10 stereo macroscope (Olympus optical Co., BmbH,

FIGURE 1 Outline of the study

OSKARSSON et al. | 3

F I G U R E 2 The implants used in the study: (a) Ankylos®

system (implant A; ∅ 3.5 mm, length 11 mm; DENTSPLY Implants
Manufacturing GmbH, Mannheim, Germany); (b) Nobel Replace®
Tapered Groovy implants (implant N; ∅ 3.5 mm, length 10 mm;
Nobel Biocare®, Kloten, Switzerland)

Hamburg, Germany) and digital images were obtained using a Leica

DFC280 camera (Leica, GmbH, Wetzlar, Germany). In the radio‐
graphs, the vertical distance between the abutment‐fixture junction
(A/F) and the marginal bone level (B) (i.e., the marginal position at F I G U R E 3 Radiographs from implant A and N obtained
at Baseline (a) and 20 weeks (b). The yellow lines indicate the
which bone was judged to be in contact with the implant surface)
abutment‐fixture junction (A/F) and the red arrows indicate
was determined at the mesial and distal aspect of each implant using
marginal level of bone to implant contact (B)
the QWin software (Leica QWin Standard V3.2.0; Leica Imaging
Systems Ltd., Cambridge, UK) (Figure 3). After 6 weeks of nonsub‐
merged healing, healthy noninflamed soft tissue conditions were es‐ Berglundh, Lindhe, Jonsson, and Ericsson (1994). In brief, the tis‐
tablished and oral hygiene measures were abandoned on one side sue samples were placed in EDTA. Before the hard tissue was fully
of the mandible. Thirteen weeks after implant placement, new ra‐ decalcified, incisions parallel with the long axis of the implant were
diographs were obtained from all implant sites. Five months after made through the peri‐implant tissues and four different blocks
implant placement surgery, new radiographs were obtained and the (mesio‐buccal, disto‐buccal, mesio‐lingual, disto‐lingual) were
experiment was terminated. thereby obtained and separated from the implant. Decalcification
was completed in EDTA and dehydration was performed in a se‐
ries of ethanol concentrations. The specimens were then de‐fat‐
2.3 | Biopsy and histological preparation
ted in xylene, rinsed in absolute ethanol (99.9%), and subsequently
The dogs were euthanized by an overdose of sodium pentothal embedded in LR White Resin (Hard grade; London Resin Company
(Hospira Enterprises B.V., Hoofddorp, the Netherlands) and perfused Ltd, Reading, Berkshire, England). Semithin sections were pro‐
through the carotid arteries by a fixative (4% formaldehyde). The duced with the microtome set at 3 μm and staining of sections was
mandibles were removed and placed in the fixative. Each implant performed with PAS and toluidine blue (Schroeder, 1969).
site, including the implant and the soft and hard peri‐implant tissues,
was dissected using a diamond saw (Exakt; Kulzer, Norderstedt,
2.4 | Histological analysis
Germany).
The specimens were processed using a modification of the The histological analysis was performed using a Leica DM‐RBE
“fracture technique” (Thomsen & Ericson, 1985) described by microscope (Leica, Heidelberg, Germany) equipped with an image
4 | OSKARSSON et al.

F I G U R E 4 Histological landmarks
depicted in sections from an A‐implant
(a) and N‐implant (b). A/F, the abutment/
fixture borderline; aJE, the level of the
apical termination of the junctional
epithelium; B, the marginal level of bone
to implant contact; PM, the marginal
portion of the peri‐implant mucosa

system (Q‐500 MC; Leica, Wetzlar, Germany). In each histological

3.1 | Clinical findings
section the following landmarks were identified and used for linear
measurements (Figure 4): the marginal portion of the peri‐implant On the oral hygiene side, abutments were free from visible plaque
mucosa (PM), the apical termination of the barrier epithelium (aJE), and the peri‐implant mucosa showed no clinical signs of inflamma‐
the marginal level of bone‐to‐implant contact (B) and the position tion at the end of the plaque control period. On the nonoral hygiene
of the abutment/fixture border (A/F). The linear distances between side, the abutments were covered by plaque and the peri‐implant
the landmarks were measured on six sections from each implant site. mucosa showed evident signs of inflammation.

2.5 | Data analysis 3.2 | Radiological findings

The Stata Statistical Software: Release 13 (StataCorp LP, College Results from the radiographic assessments are presented in Table 1
Station, TX, USA) was used for the statistical analysis. and Figure 5.
A two‐sample t test with the animal as the unit of analysis was Changes in marginal bone levels were more pronounced
used to explore differences in soft tissue dimensions between oral during the initial 13‐week period. Thereafter, only minor changes
hygiene vs. no oral hygiene sites at different implant types and dif‐ occurred.
ferent placement protocols. A‐implants showed small changes in marginal bone levels during
Mean values and SD were calculated for all parameters with the the 20 weeks of observation, regardless of crestal or subcrestal
implant as the unit of analysis (n = 45). A regression analysis was per‐ placement. For N‐implants, however, placement had a significant im‐
formed. Implant type and implant position were entered as predictors, pact on bone level changes. N‐implants installed in a crestal position
considering interaction between the two terms. To test the robust‐ demonstrated a mean bone loss of about 0.4 mm, while N‐implants
ness of the results from the regression analysis, sensitivity analysis on placed in a subcrestal position showed a bone loss of 1.5 mm. The
the regression model was performed using a two‐sample t test. regression analysis revealed an interaction between implant type N
A p‐value < 0.05 was considered to be significant. and placement depth.

3.3 | Histological findings

3 | R E S U LT S
As data for the histological parameters were found to be similar
One subcrestally placed A‐implant did not integrate and was re‐ within each subgroup (A‐crestal, A‐subcrestal, N‐crestal, and N‐sub‐
moved 6 weeks after placement. Two crestally placed N‐implants crestal) in the comparison between sites with oral hygiene and no
showed extensive bone loss due to external trauma during the first oral hygiene (final 14 weeks; Table 2), data were collapsed.
weeks of healing. These three implants were excluded from all analy‐ The results of the histological assessments are presented in
ses. Healing for the remaining 45 implants was uneventful. Tables 2 and 3.
OSKARSSON et al. | 5

TA B L E 1 Results from the radiographic assessments TA B L E 3 Results from the histologic measurements

Implant type Crestal Subcrestal Implant type Crestal Subcrestal

A 0.09 (0.51) 0.04 (0.21) NS Distance between A/F and B

N −0.36 (0.73) −1.50 (0.63) S A 0.63 (0.98) −0.02 (0.19) S
NS S N 1.51 (0.49) 1.51 (0.84) NS

Notes. Bone level alterations between baseline and final evaluation S S

(20 weeks). Mean values in mm and (SD). Implant‐level analysis. Distance between aJE and B
Significances as tested by regression analysis and t test for sensitivity. S
A 1.36 (0.62) 1.54 (0.35) NS
= significant (p < 0.05). NS = not significant (p ≥ 0.05).
N 1.24 (0.45) 2.54 (0.92) S
NS S
Distance between PM and aJE
A 1.94 (0.40) 2.37 (0.50) S
N 2.17 (0.66) 2.04 (0.44) NS
NS NS
Distance between PM and B
A 3.30 (0.71) 3.91 (0.49) S
N 3.41 (0.72) 4.58 (0.72) S
NS NS

Notes. Mean values in mm and (SD). Implant‐level analysis. Significances

as tested by regression analysis and t test for sensitivity. S = significant
FIGURE 5 Marginal bone level changes (mm) assessed in (p < 0.05). NS = not significant (p ≥ 0.05).
radiographs

For A‐implants, the distance between A/F and B was signifi‐ N‐implants, this dimension was significantly larger at implants
cantly shorter for subcrestally placed implants than for implants placed in the subcrestal than the crestal position. In addition,
placed in a crestal position. For N‐implants, however, no difference the regression analysis indicated an interaction between implant
regarding the distance A/F‐B was observed between crestal‐ and type and placement protocol. Thus, for the subcrestal group, N
subcrestal placement. The distance A/F‐B was significantly greater implants showed a significantly larger vertical aJE‐B distance
at N‐implants than A‐implants, irrespective of placement depth. In than A implants.
fact, at nine of the A‐implants (two with crestal and seven with sub‐ The length of the junctional epithelium (PM‐aJE) varied between
crestal placement), bone to implant contact was observed on the top 1.94 and 2.37 mm. A‐implants presented with statistically higher
of the implant shoulder (Figure 4a). values in the subcrestal than in the crestal group, while no such dif‐
The vertical dimensions of the supracrestal connective ference was observed for N‐implants.
tissue zone (aJE‐B) at A‐implants were 1.36 mm (crestal) and The overall soft tissue dimension (PM‐B) at both A and N im‐
1.54 mm (subcrestal). The corresponding connective tissue di‐ plants was significantly longer at subcrestally placed implants than
mensions at N‐implants were 1.24 and 2.54 mm, respectively. At at crestally placed implants.

TA B L E 2 Results from the histologic measurements comparing oral hygiene to no oral hygiene sites within each subgroup (A‐crestal,
A‐subcrestal, N‐crestal, and N‐subcrestal)

Crestal/no oral Subcrestal/no

Implant type Crestal/oral hygiene hygiene Subcrestal/oral hygiene oral hygiene

PM‐B A 3.13 (0.49) 3.48 (0.89) 4.12 (0.36) 3.66 (0.54)

PM‐B N 3.58 (0.76) 3.30 (0.75) 4.22 (0.43) 4.95 (0.79)
PM‐aJE A 1.98 (0.51) 1.90 (0.30) 2.71 (0.31) 1.96 (0.35)*
PM‐aJE N 2.64 (0.77) 1.86 (0.38) 1.95 (0.42) 2.13 (0.47)
A/F‐B A 0.24 (0.30) 1.02 (1.30) 0.01 (0.19) −0.05 (0.21)
A/F‐B N 1.80 (0.36) 1.32 (0.50) 1.11 (0.48) 1.91 (0.98)

Notes. Vertical distances between the landmarks: A/F, the abutment/fixture borderline; aJE, the level of the apical termination of the junctional epithe‐
lium; B, the marginal level of bone to implant contact; PM, the marginal portion of the peri‐implant mucosa.
Mean values in mm and (SD). Two‐sample t test.
*
p < 0.05.
6 | OSKARSSON et al.

4 | D I S CU S S I O N Recent reviews (Annibali et al., 2012; Atieh, Ibrahim, & Atieh,

2010; Santiago et al., 2016; Strietzel, Neumann, & Hertel, 2015) in‐
In the present study, the influence of implant geometry and crestal cluding studies with follow‐up periods ranging from 12 to 60 months
and subcrestal placement on the soft and hard tissue conditions have also described significantly smaller mean crestal bone level
were studied. It was demonstrated that platform‐switched im‐ changes at PS implants, as compared to PM implants.
plants showed minimal changes in marginal bone levels during the A limitation in the present study is that the different implant
20 weeks of observation, regardless of placement depth. For N‐im‐ types differed not only in terms of PS and PM design but also with
plants, however, placement had a significant impact on bone level respect to different types of implant‐abutment connections; that
changes. N‐implants installed in a crestal position demonstrated a is, internal conical or external connection. In a recent systematic
mean bone loss of about 0.4 mm, while N‐implants placed in a subcr‐ review and meta‐analysis (Caricasulo, Malchiodi, Ghensi, Fantozzi,
estal position showed a bone loss of 1.5 mm. & Cucchi, 2018), based on 14 RCT’s, studying the influence of
It was also found that the major part of bone level alterations fol‐ implant‐abutment connection to peri‐implant bone loss, it was
lowing implant placement occurred within 13 weeks, and only small indicated that crestal bone levels were better maintained when
bone level alterations occurred during the following 7 weeks of the internal kinds of interfaces were adopted. However, PS seemed to
experiment. positively affect bone levels, nonregarding the implant connection
The position of the bone‐to‐implant contact was found signifi‐ it was applied to.
cantly closer to A/F at A‐implants placed in a subcrestal position The finding that the placement of implants in a crestal or subcr‐
than at crestally placed A‐implants. The distance between A/F and estal position had a significant impact on bone level changes during
B was significantly longer at N‐implants than at A‐implants irrespec‐ healing at PM implants but not at PS implants is in agreement with
tive of crestal or subcrestal placement. the results of previous animal studies (Jung et al., 2008; Weng et al.,
In addition, while no significant differences regarding the total 2011). Weng et al. (2011), in a study on six mongrel dogs, radiolog‐
soft tissue dimensions (PM‐B) were found between different implant ically assessed the peri‐implant bone levels at implants with either
designs, crestal placement resulted in a significantly thinner soft tis‐ a PM (TiUnite Brånemark; Nobel Biocare) or PS design (Ankylos,
sue barrier at both implant types. Dentsply Friadent), placed in a crestal or 1.5 mm subcrestal posi‐
The finding that the marginal bone level is better preserved tion. Three months after implant placement, similar amounts of
during healing after implant placement at implants with a PS im‐ marginal bone loss was reported at crestally placed PM and PS im‐
plant/abutment connection than at implants with a PM connection plants (0.69 mm vs. 0.46 mm, respectively), while subcrestal place‐
is in agreement with previous animal and human studies (Baffone ment resulted in more bone loss at PM implants than at PS implants
et al., 2012; Guerra et al., 2014; Rocha et al., 2016; Telleman et al., (1.56 mm vs. 0.79 mm, respectively). In another study using five
2012; Weng, Nagata, Bosco, & Melo, 2011). Baffone et al. (2012) foxhounds, Jung et al. (2008) made radiological assessments of ex‐
evaluated the effect of a mismatch between abutment and implant perimental two‐piece PS‐implants (Straumann), placed in different
platform diameter in an experiment on six Labrador dogs. Four positions in relation to the marginal bone crest. Three months after
months after implant placement with the implant shoulder in level implant placement, implants placed 1 mm subcrestal demonstrated
with the buccal alveolar crest, the crestal bone level was identified significantly more crestal bone loss as compared to implants placed
at a more coronal position at PS implants (mismatch 0.85 mm) than in a supracrestal position.
at PM implants. In another animal study (Weng et al., 2011) peri‐im‐ Another important finding of the present study was that most
plant bone levels at implants with either a PM (TiUnite Brånemark; of the bone level alterations occurred within 13 weeks after implant
Nobel Biocare) or a PS (Ankylos, Dentsply Friadent) design were placement, while bone levels thereafter remained stable. This is in
assessed radiologically. The crestal bone level was identified in a agreement with findings from a study on five foxhounds in which
significantly more coronal position at PS‐implants than at PM‐im‐ bone level alterations at implants placed in a submerged or nonsub‐
plants 3 months after implant placement. Guerra et al. (2014), in a merged position were studied radiographically (Hermann, Cochran,
RCT used radiographs to assess crestal bone level changes during Nummikoski, & Buser, 1997). It was found that most bone remodel‐
12 months postloading at 74 single implants with a PS‐design and ing occurred within 1 month after implant placement in the nonsub‐
72 implants with a PM‐design using implants from the CAMLOG merged implant group and within 1 month after abutment connection
system. They reported a small bone gain or no bone loss for 67.1% in the submerged implant group. Berglundh, Abrahamsson, and Lindhe
of the sites with PS implants as compared to 49.2% for the implants (2005), in another animal study including six beagle dogs, reported
with PM design. Rocha and co‐workers made a 3‐year follow‐up that the major part of the marginal bone level alterations occurred fol‐
study of the same subjects included in the study by Guerra and co‐ lowing implant surgery and abutment connection but remained stable
workers and results remained stable during this period. In another after prosthetic delivery. In a series of experiments in 20 Labrador
RCT (Telleman et al., 2012) 149 implants with either PS (n = 76) or dogs (Abrahamsson, Berglundh, Linder, Lang, & Lindhe, 2004;
PM (n = 73) implant‐abutment connections were followed for 1 year Berglundh, Abrahamsson, Welander, Lang, & Lindhe, 2007), different
after delivery of the implant crown. Significantly less peri‐implant phases of wound healing in the peri‐implant tissues following implant
bone loss was observed in the PS group. placement were studied. It was found that after about 6–8 weeks of
OSKARSSON et al. | 7

nonsubmerged healing, the marginal bone level and soft tissue barrier ORCID
dimensions were established.
Monika Oskarsson http://orcid.org/0000-0003-0700-3944
Moreover, in the study by Berglundh et al. (2005), the initial bone
loss at implants with a PM implant/abutment connection (Brånemark
System®) was more pronounced, compared to implants with a PS de‐ REFERENCES
sign (Astra Tech Implants® Dental System).
Abrahamsson, I., Berglundh, T., Linder, E., Lang, N. P., & Lindhe, J.
In the present study, no significant differences regarding the
(2004). Early bone formation adjacent to rough and turned en‐
soft tissue dimensions were found between different implant de‐ dosseous implant surfaces. An experimental study in the dog.
signs. This finding is partly in agreement with another experiment on Clinical Oral Implants Research, 15(4), 381–392. https://doi.
Labrador dogs (Baffone et al, 2012). In that study, a reduced height org/10.1111/j.1600-0501.2004.01082.x.
Abrahamsson, I., Berglundh, T., & Lindhe, J. (1997). The mucosal barrier
on the buccal aspect of the peri‐implant soft tissues of PM implants
following abutment dis/reconnection. An experimental study in
was reported, while no significant differences were reported on lin‐ dogs. Journal of Clinical Periodontology, 24(8), 568–572. https://doi.
gual or proximal aspects. org/10.1111/j.1600-051X.1997.tb00230.x.
In addition, when N‐implants in our study were placed subcre‐ Annibali, S., Bignozzi, I., Cristalli, M. P., Graziani, F., La Monaca, G.,
& Polimeni, A. (2012). Peri‐implant marginal bone level: A sys‐
stally, the vertical dimension of the supracrestal connective tissue
tematic review and meta‐analysis of studies comparing plat‐
zone was significantly increased. form switching versus conventionally restored implants. Journal
In the present study, new bone formation was observed on of Clinical Periodontology, 39(11), 1097–1113. https://doi.
top of the implant shoulder at seven out of 11 PS implants placed org/10.1111/j.1600-051X.2012.01930.x.
Atieh, M. A., Ibrahim, H. M., & Atieh, A. H. (2010). Platform switching for
subcrestally and at two of 12 PS implants placed crestally. New
marginal bone preservation around dental implants: A systematic
bone formation coronal of A/F did not occur at any of the PM im‐ review and meta‐analysis. Journal of Periodontology, 81(10), 1350–
plants. This finding is corroborated by previous findings in animal 1366. https://doi.org/10.1902/jop.2010.100232.
experiments (Welander, Abrahamsson, & Berglundh, 2009; Weng, Baffone, G. M., Botticelli, D., Canullo, L., Scala, A., Beolchini, M., &
Lang, N. P. (2012). Effect of mismatching abutments on im‐
Nagata, Bell, Melo, & Bosco, 2010). Weng et al. (2010), in a study
plants with wider platforms – An experimental study in dogs.
on six mongrel dogs, reported an overgrowth of bone tissue on the Clinical Oral Implants Research, 23(3), 334–339. https://doi.
implant shoulder at PS/Morse taper implants placed subcrestally, org/10.1111/j.1600-0501.2011.02320.x.
while no such overgrowth was found at PM/external hex implants. Berglundh, T., Abrahamsson, I., & Lindhe, J. (2005). Bone reactions to
In addition, Welander et al. (2009), in a study made on five mon‐ longstanding functional load at implants: An experimental study in
dogs. Journal of Clinical Periodontology, 32(9), 925–932. https://doi.
grel dogs, found bone in contact with the abutment at PS implants
org/10.1111/j.1600-051X.2005.00747.x.
placed 2‐mm subcrestally. Berglundh, T., Abrahamsson, I., Welander, M., Lang, N. P., & Lindhe, J.
(2007). Morphogenesis of the peri‐implant mucosa: An experimen‐
tal study in dogs. Clinical Oral Implants Research, 18(1), 1–8. https://
doi.org/10.1111/j.1600-0501.2006.01309.x.
5 | M A I N FI N D I N G S Berglundh, T., & Lindhe, J. (1996). Dimension of the periimplant mucosa.
Biological width revisited. Journal of Clinical Periodontology, 23(10),
Bone remodeling occurred within the early healing phase (13 weeks) 971–973. https://doi.org/10.1111/j.1600-051X.1996.tb00520.x.
after implant placement. Berglundh, T., Lindhe, J., Jonsson, K., & Ericsson, I. (1994). The topog‐
raphy of the vascular systems in the periodontal and peri‐implant
The preservation of the bone level was similar irrespective of
tissues in the dog. Journal of Clinical Periodontology, 21, 189–193.
crestal or subcrestal placement of the PS implants up to 5 months https://doi.org/10.1111/j.1600-051X.1994.tb00302.x.
after placement. Caricasulo, R., Malchiodi, L., Ghensi, P., Fantozzi, G., & Cucchi, A.
Subcrestal placement of PM implants was associated with more (2018). The influence of implant‐abutment connection to peri‐
implant bone loss: A systematic review and meta‐analysis. Clin
bone loss 5 months after placement compared to crestal placement.
Implant Dent Relat Res, 1–12. [Epub ahead of print]. https://doi.
PS implants demonstrated significantly less crestal bone loss org/10.1111/cid.12620.
compared to PM implants. Goldberg, P. V., Higginbottom, F. L., & Wilson, T. G. (2001).
Periodontal considerations in restorative and implant ther‐
apy. Periodontology 2000, 25(1), 100–109. https://doi.
AC K N OW L E D G E M E N T S org/10.1034/j.1600-0757.2001.22250108.x.
Guerra, F., Wagner, W., Wiltfang, J., Rocha, S., Moergel, M., Behrens,
The study was supported by grants from DENTSPLY Implants E., & Nicolau, P. (2014). Platform switch versus platform match in
Manufacturing GmbH (grant no. DF1810‐1‐299‐2‐1) and TUA the posterior mandible ‐ 1‐year results of a multicentre random‐
Research Funding (The Sahlgrenska Academy at the University of ized clinical trial. Journal of Clinical Periodontology, 41(5), 521–529.
https://doi.org/10.1111/jcpe.12244.
Gothenburg /Region Västra Götaland, Sweden).
Hermann, J. S., Cochran, D. L., Nummikoski, P. V., & Buser, D. (1997).
Crestal bone changes around titanium implants. A radiographic
evaluation of unloaded nonsubmerged and submerged implants in
C O N FL I C T O F I N T E R E S T the canine mandible. Journal of Periodontology, 68(11), 1117–1130.
https://doi.org/10.1902/jop.1997.68.11.1117.
The authors report no conflicts of interest related to this study.
8 | OSKARSSON et al.

Huang, B., Meng, H., Piao, M., Xu, L., Zhang, L., & Zhu, W. (2012). Telleman, G., Meijer, H. J. A., Vissink, A., & Raghoebar, G. M. (2012).
Influence of placement depth on bone remodeling around tapered Short implants with a nanometer‐sized CaP surface provided with
internal connection implant: A clinical and radiographic study either a platform‐switched or platform‐matched abutment connec‐
in dogs. Journal of Periodontology, 83(9), 1164–1171. https://doi. tion in the posterior region: A randomized clinical trial. Clinical Oral
org/10.1902/jop.2012.110617. Implants Research, 24(12), 1316–1324. https://doi.org/10.1111/
Jung, R. E., Jones, A. A., Higginbottom, F. L., Wilson, T. G., Schoolfield, J., clr.12000.
Buser, D., … Cochran, D. L. (2008). The influence of non‐matching Thomsen, P., & Ericson, L. (1985). Light and transmission electron microscopy
implant and abutment diameters on radiographic crestal bone lev‐ used to study the tissue morphology close to implants. Biomaterials, 6,
els in dogs. Journal of Periodontology, 79(2), 260–270. https://doi. 421–424. https://doi.org/10.1016/0142-9612(85)90104-8.
org/10.1902/jop.2008.070132. Welander, M., Abrahamsson, I., & Berglundh, T. (2009). Subcrestal place‐
Kilkenny, C., Browne, W., Cuthill, I. C., Emerson, M., Altman, D. G., & ment of two‐part implants. Clinical Oral Implants Research, 20(3),
National Centre for the Replacement, Refinement and Reduction 226–231. https://doi.org/10.1111/j.1600-0501.2008.01637.x.
of Animals in Research (2011). Animal research: Reporting in Weng, D., Nagata, M. J. H., Bell, M., de Melo, L. G. N., & Bosco, A. F.
vivo experiments‐the ARRIVE guidelines. Journal of Cerebral (2010). Influence of microgap location and configuration on peri‐
Blood Flow & Metabolism, 31, 991–993. https://doi.org/10.1038/ implant bone morphology in nonsubmerged implants: An experi‐
jcbfm.2010.220. mental study in dogs. The International Journal of Oral & Maxillofacial
Lazzara, R. J., & Porter, S. S. (2006). Platform switching: A new concept in Implants, 25(3), 540–547.
implant dentistry for controlling postrestorative crestal bone levels. Weng, D., Nagata, M. J. H., Bosco, A. F. H., & de Melo, L. G. N. (2011).
Int J Periodontics Restorative Dent, 26, 9–17. Influence of microgap location and configuration on radiographic
Rocha, S., Wagner, W., Wiltfang, J., Nicolau, P., Moergel, M., Messias, bone loss in nonsubmerged implants: An experimental study in
A., … Guerra, F. (2016). Effect of platform switching on crestal dogs. The International Journal of Prosthodontics, 24(5), 445–452.
bone levels around implants in the posterior mandible: 3 years Yeung, S. C. H. (2008). Biological basis for soft tissue management in im‐
results from a multicentre randomized clinical trial. Journal of plant dentistry. Australian Dental Journal, 53 Suppl, 1(s1), S39–S42.
Clinical Periodontology, 43(4), 374–382. https://doi.org/10.1111/ https://doi.org/10.1111/j.1834-7819.2008.00040.x.
jcpe.12522.
Santiago Jr, J. F., de Souza Batista, V. E., Verri, F. R., & Hono ŕ io, H. M., …
… E. P. (2016). Platform‐switching implants and bone preservation: S U P P O R T I N G I N FO R M AT I O N
A systematic review and meta‐analysis. International Journal of Oral
and Maxillofacial Surgery, 45, 332–345. https://doi.org/10.1016/j. Additional supporting information may be found online in the
ijom.2015.11.009. Supporting Information section at the end of the article.
Schroeder, H. (1969). Ultrastructure of the junctional epithelium of the
human gingiva. Helvetica Odontologica Acta, 13, 65.
Stanford, C. M. (2002). Achieving and maintaining predictable implant
How to cite this article: Oskarsson M, Otsuki M, Welander
esthetics through the maintenance of bone around dental implants.
Compendium of Continuing Education in Dentistry (Jamesburg, N.J. : M, Abrahamsson I. Peri‐implant tissue healing at implants
1995), 23(9 Suppl 2), 13–20. with different designs and placement protocols: An
Strietzel, F. P., Neumann, K., & Hertel, M. (2015). Impact of platform experimental study in dogs. Clin Oral Impl Res. 2018;00:1–8.
switching on marginal peri‐implant bone‐level changes. A system‐
https://doi.org/10.1111/clr.13339
atic review and meta‐analysis. Clinical Oral Implants Research, 26(3),
342–358. https://doi.org/10.1111/clr.12339.