A Retrospective Radiographic Study on the Effect of Natural Tooth-Implant Proximity and an Introduction to the Concept of a Bone-Loading Platform Switch

Abstract

Purpose:
The aim of this study was to evaluate the effect of tooth-implant proximity using an implant system with a double platform shift that was designed to load bone coronal to the implant-abutment cohort study was conducted between January 2008 and December 2009. The sample was composed of patients who had received at least one 5-mm-wide hydroxyapatite-coated single-tooth Bicon implant that had been placed adjacent to at least one natural tooth. Descriptive statistics and univariate and multivariate linear mixed-effects regression models, adjusted for multiple implants in the same patient, were utilized. The primary predictor variable was the horizontal distance between implant and adjacent tooth, and the primary outcome variable was the change in peri-implant bone levels over time.

Results:
Two hundred six subjects who received 235 plateau root-form implants were followed for an average of 42 months. Tooth-implant distance ranged between 0 and 14.6 mm. Out of 235 implants, 43 implants were placed < 1 mm to an adjacent natural tooth on mesial and/or distal sides. The proximity of a plateau root-form implant was not associated with complications on the adjacent tooth such as bone loss, root resorption, endodontic treatment, pain, or extraction. The proximity of an adjacent tooth was not a risk factor for the failure of a plateau root-form implant. After adjusting for other covariates in a multivariate model, the proximity of a natural tooth did not have a statistically significant effect on peri-implant bone levels (P = .13). The extraction of an adjacent tooth was associated with a significant increase in peri-implant bone loss (P = .008).

Conclusion:
The placement of a plateau root-form implant with a sloping shoulder in close proximity to an adjacent tooth did not cause damage to that tooth or lead to bone loss or the failure of the implant.

Full text

1412 Volume 29, Number 6, 2014
©2014 by Quintessence Publishing Co Inc.
A Retrospective Radiographic Study on the Eect
of Natural Tooth–Implant Proximity and an Introduction
to the Concept of a Bone-Loading Platform Switch
Rainier A. Urdaneta, DMD1/Rudolf Seemann, DDS2/Irina-Florentina Dragan, DDS3/
William Lubelski, BS4/Joseph Leary, DMD5/Sung-Kiang Chuang, DMD, MD6
Purpose:
The aim of this study was to evaluate the effect of tooth-implant proximity using an implant
system with a double platform shift that was designed to load bone coronal to the implant-abutment
interface.
Materials and Methods:
A retrospective cohort study was conducted between January 2008
and December 2009. The sample was composed of patients who had received at least one 5 -mm–wide
hydroxyapatite-coated single-tooth Bicon implant that had been placed adjacent to at least one natural
tooth. Descriptive statistics and univariate and multivariate linear mixed-effects regression models, adjusted
for multiple implants in the same patient, were utilized. The primary predictor variable was the horizontal
distance between implant and adjacent tooth, and the primary outcome variable was the change in peri-
implant bone levels over time.
Results:
Two hundred six subjects who received 235 plateau root-form
implants were followed for an average of 42 months. Tooth-implant distance ranged between 0 and 14.6
mm. Out of 235 implants, 43 implants were placed < 1 mm to an adjacent natural tooth on mesial and/
or distal sides. The proximity of a plateau root-form implant was not associated with complications on the
adjacent tooth such as bone loss, root resorption, endodontic treatment, pain, or extraction. The proximity
of an adjacent tooth was not a risk factor for the failure of a plateau root-form implant. After adjusting for
other covariates in a multivariate model, the proximity of a natural tooth did not have a statistically signicant
effect on peri-implant bone levels (P = .13). The extraction of an adjacent tooth was associated with a
signicant increase in peri-implant bone loss (P = .008).
Conclusion:
The placement of a plateau root-form
implant with a sloping shoulder in close proximity to an adjacent tooth did not cause damage to that tooth
or lead to bone loss or the failure of the implant. Int J Ora l Ma xIllOfac IMplan ts 2014;29:1412–1424. doi:
10.11607/jomi.3699
Key words: adjacent structures, platform switching, retrospective cohort study, single-tooth implants, tooth-
implant distance
Endosseous implants are a viable treatment alterna-
tive for the replacement of single teeth.1 Minimal
or no crestal bone resorption is considered to be an
indicator of the long-term success of implant restora-
tions.2 Marginal bone loss has been associated with
high occlusal stress in several clinical studies using
splinted smooth-surface implants.3–6 Nonetheless,
studies on the eect of forces in peri-implant bone
have shown conicting results, and a cause-and-
eect relationship between overloading and peri-
implant bone loss has not been demonstrated.7 On
the contrary, crestal bone growth has also been asso-
ciated with increased loads in long-term studies.8 –10
Furthermore, it is well known that mechanical stimuli
are critical for bone maintenance and repair and that
maintaining bone mass requires a continuous, load-
related osteoregulatory stimulus.11–16 It is generally
accepted that the implant and the abutment work as
a unit in transferring occlusal loads to bone. Crestal
bone loss to the rst thread has been commonly ob-
served in conventional implant-abutment platform
designs. In conventional platform designs, the rst
1Prosthodontist, Private Practice, Implant Dentistry Centre,
Boston, Massachusetts, USA.
2Oral and Maxillofacial Surgeon, University of Vienna, Austria.
3Postgraduate Resident, Dep artment of Periodontolog y,
Tufts University School of Dental Medicine, Boston,
Massachusetts, USA.
4Biomedical Engineer, Bicon, Boston, Massachusetts, USA.
5Periodontist, Private Practice, Norwood, Massachusetts, USA.
6Associate Professor, Department of Oral and Maxillofacial
Surger y, Massachusetts General Hospital and Harvard S chool
of Dental Medicine, Boston, Massachusetts, USA.
Correspondence to: Dr Rainier A. Urdaneta, 25 Prairie Ave,
Auburndale, MA 02466, USA . Fax: +1-617-390- 004 3.
Email: rainieru@yahoo.com
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
The International Journal of Or al & Maxillofacial Implants 1413
on its adjacent teeth, to investigate the eect of tooth
proximity on implant survival and peri-implant bone
levels, and to introduce the concept of a bone-loading
platform switch.
MATERIALS AND METHODS
Study Design and Sample
The present study was designed as a retrospective co-
hort study. The cohort was derived from a population
of patients who had at least one 5-mm–wide Bicon
Short implant (Bicon) placed between January 1, 2008
and December 31, 2009 at the Implant Dentistry Cen-
tre in Boston, Massachusetts.
Patients of record treated at Implant Dentistry
Centre were selected if they satised the following
inclusion criteria: (1) hydroxyapatite-coated implants
placed using a two-stage surgical protocol; (2) at least
one 5-mm–wide plateau root-form implant (5, 6, or 8
mm in length) restored with a single implant restora-
tion28 in which the denitive abutments were inserted
the same day as crown insertion; (3) an implant adja-
cent to at least one natural tooth.
Following manufacturer recommendations, the im-
plants were seated into similar-sized osteotomies with-
out torqueing, so that only the tip of the n or plateau
was in contact with the adjacent bone, leaving space
between the plateaus and bone. The implants were
placed 2 to 3 mm apical to the crest of the bone. The
number of plateaus or ns per implant was 6, 7, and
10, corresponding to the implant lengths of 5, 6, and
8 mm, respectively. On the day of insertion of the de-
nitive abutment, the bone and/or soft tissues coronal
to the implant were prepared to match the spherical
shape of the abutment’s base with the use of a sulcus
reamer matching the abutment’s coronal prole (Fig 1).
thread is also the most coronal area where bone is
loaded in compression.17
Platform switching, the creation of a horizontal dis-
crepancy between the diameters of both the implant
platform and the abutment, has been shown to reduce
postrestorative crestal bone loss.18,19
The biomechanical design of plateau root-form im-
plants has a double platform switch that was designed
to load bone coronal to the implant-abutment inter-
face through the base of the abutment.20 A nite el-
ement study reported high stress concentration (von
Mises strains of 1,750 to 3,000) in the bone surround-
ing the spherical base of abutments when intracrest-
ally placed plateau root-form implants and abutments
were subjected to nonaxial loads of 100 N.21 Vertical
bone growth coronal to the implant shoulder and
toward the spherical base of the abutment has been
reported in several clinical studies using plateau root-
form implants.22–25
The literature has reported contrasting results with
regard to how close an implant may be placed to an
adjacent tooth without causing damage to it. Esposi-
to et al26 evaluated implants with matching implant-
abutment platforms and reported increased bone
loss to adjacent teeth as the horizontal tooth-implant
distance between the two structures decreased. In
contrast, Vela et al27 reported no signicant correla-
tion between implant proximity and resorption on the
adjacent tooth’s bony crest and suggested that the dif-
fering results may be explained by platform switching.
With regard to changes in peri-implant bone levels,
Vela et al27 reported that, for implants with a horizon-
tal platform switch placed ≤ 1 mm away, the proximity
of an adjacent tooth was associated with vertical and
horizontal peri-implant bone loss.
The purposes of the present report were to evaluate
the eect of proximity of a plateau root-form implant
Fig 1 Relationship between the base of
the abutment and bone. Bone coronal to
a plateau root-form implant is prepared to
match the shape of the abutment’s base
prior to the insertion of the abutment: (a)
Guide pin is inserted into the implant well.
(b) Bone is prepared using a sulcus reamer
inserted into the guide pin. (c) An abutment
matching the reamer’s size is inserted.
(d) Contact bet ween abutment and bone
is achieved (Courtesy of Bicon Implants,
Boston, Massachusetts.). Periapical ra-
diographs of an implant restoration with
a bone-loading platform switch are shown
(e) at crown insertion (1999) and at rec all
appointments (f) in 2008 and (g) in 2011.
The bone mineralization observed under
the spherical base af ter 13 years of load-
ing suggests that occlusal loads may be
transferred to crestal bone through the
abutment’s base.
abcd
e f g
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
1414 Volume 29, Number 6, 2014
Study Variables
Study variables were grouped into the following cat-
egories using previously published criteria:
Demographics. The patient’s sex and age at im-
plant placement were recorded.
Health Status. General health status was classied
according to the American Society of Anesthesiology
system.29 Current tobacco use was recorded. The daily
intake of the following medications or supplements
was recorded: nonsteroidal anti-inammatory drugs
(NSAIDs), omega-3 fatty acids from sh oil, and mul-
tivitamins. Whether the patient had been diagnosed
with diabetes mellitus was also recorded.
Ana tomy. This category included implant position
(maxilla, mandible, anterior, posterior), bone quality
(types 1 to 4), and proximity of the implant relative to
teeth or other implants. To indicate what structures
were immediately adjacent to the implants on the
day of implant placement, the following categories
were used: one natural tooth, two natural teeth, or one
natural tooth–one implant.30 The extraction of a tooth
immediately adjacent to a study implant was recorded.
Other Variables. These variables included implant
length (5, 6, and 8 mm), whether or not there was an
immediate extraction and placement on the same day,
and opposing structures (natural teeth vs implant-
supported restorations). The size of the abutment was
documented as length of the abutment shaft (short
[2 mm] or long [ ≥ 3 mm]) and abutment width (5
mm or 6.5 mm). Possible presence of sleep bruxism
or awake bruxism as self-reported by study subjects31
was recorded.
Reconstructive Procedures
Bone graft augmentation procedures prior to, at the
time of, or after implant placement were recorded.
These reconstructive procedures included internal or
lateral sinus elevations and/or synthetic bone-sub-
stitute grafting using β-tricalcium phosphate (β-TCP)
(SynthoGraft, Bicon).
Complications
Implant failure was dened as removal of the implant.
For each implant, the date the implant was placed,
the date the denitive restoration was placed, and the
date of the patient’s last visit was recorded. If appli-
cable, the date of implant removal was recorded. The
time between implant placement and the patient’s last
visit or implant removal was dened as the duration
of implant survival. Augmentation procedures needed
after implant placement were documented. Changes
occurring on the mesial and/or distal of adjacent teeth
following the placement of the implant were record-
ed. These changes included extraction of an adjacent
tooth, presence of lesions of endodontic origin or
This created, whenever possible, intimate contact be-
tween the abutment’s spherical base and bone.20
All the implants in the present study were the same
width, 5 mm, and the abutment post width was also
constant, 3 mm, representing a platform switch of ap-
proximately 1 mm on each side. Coronally, the implant
abutment reversed the platform switch by increasing
its width from the 3-mm post to a 5-mm or 6.5-mm
coronal prole, depending upon the size of the tooth
restored, as shown in Fig 2.
Three clinicians, two oral maxillofacial surgeons and
a prosthodontist, placed the implants; three dierent
clinicians, two general dentists and a prosthodontist,
restored the implants. Each subject was given a de-
tailed description of the procedures and required to
sign informed consent forms. An independent review
board approved the study (New England Institutional
Review Board, Newton, Massachusetts, USA).
Fig 2 Periapical radiographs of a plateau root-form implant re-
storing a maxillar y right rst premolar with an integrated abut-
ment crown. Measurements: measured implant length (MIL),
distance to mesial adjacent tooth (DATM) and distal adjacent
tooth (DATD), crest of bone on mesial (CBM) and distal (CBD),
and bony crest of adjacent tooth (CAT). IAI, implant-abutment
interf ace. Notice the stability of the bony crest on the adjacent
tooth and bone mineralization toward the abutment’s base on
the mesial side 3 years af ter crown insertion despite the proxim -
ity to the mesial adjacent root surface. Radiographs were taken
at crown insertion on (left) February 3, 2009 and at a recall ap-
pointment on (right) Januar y 23, 2012.
20122009
MIL
DAT D DAT M
IAI
CBD
CBM
CAT
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
The International Journal of Or al & Maxillofacial Implants 1415
published.22–24 Periapical radiographs were obtained
with the use of the long-cone technique and an XCP/
Rinn lm-holding system (Dentsply-Rinn). Digital peri-
apical radiographs were stored using a digital extraoral
imaging system (Digora PCT, Soredex). Stored images
were displayed on a monitor (UltraSharp E173FPB,
Dell), and direct measurements were performed in mil-
limeters by two dierent examiners (RS and IFD). The
actual implant length, available from the manufactur-
er, was used to obtain a margin of error on each radio-
graph evaluated, and measurements were adjusted for
calibration error.
Sample Size Evaluation
There is an 80% probability that a sample size of 235
implants will detect a relationship between the tooth-
implant distance and the changes in bone levels over
time given (1) a two-sided 0.05 signicance level and
(2) if the true change in bone levels over time is 0.2
mm per 1 mm change in tooth-implant distance. Con-
sequently, a larger sample size would have improved
the likelihood of detecting signicant correlations be-
tween the study variables.
Data Management and Statistics
A database was created using Excel 2007 (Microsoft)
with appropriate checks to identify errors. Descriptive
statistics were computed for all study variables. A rst
univariate regression model (Tables 1a and 1b) was
developed to identify those variables that were associ-
ated with changes in bone levels over time (outcome
variable). The second univariate model (Table 2) was
developed to assess the possible correlation between
complications (outcome variable) and tooth-implant
proximity on the mesial and distal sides. Risk factors
with P values ≤ .15 on the rst univariate analysis to-
gether with biologically relevant factors (age and sex)
and the primary predictor variable (average tooth-im-
plant distance) were entered into a multivariate linear
mixed-eects regression model (Table 3). This analytic
method adjusted for clustering implant observations
within the same patient32 using the procedure com-
mand PROC MIXED with the option of exchangeable
correlation matrix in SAS Statistical Software Comput-
ing Environment Version 9.3 (SAS). The follow-up time
was estimated by the dierence in months between
the day of implant placement and the patient’s last
visit or implant removal. Kappa statistics were used to
evaluate examiner reliability.
RESULTS
Between January 1, 2008 and December 31,
2009, 206 subjects had at least one 5-mm–wide
endodontic treatment, radiographic evidence of exter-
nal or internal root resorption, damage to the adjacent
root surfaces, and pain (self-reported cases of pain as
documented in the patient’s chart).
Radiographic Measurements
Linear measurements (in millimeters) from the im-
plant-abutment interface (IAI) to the highest crestal
bone level were obtained at the mesial and distal22
(see Fig 2). An average mesiodistal crestal bone level
was obtained for each implant. The changes in peri-
implant bone levels were measured mesially and dis-
tally by comparing the average mesiodistal crestal
bone levels on a periapical radiograph obtained on
the day of the insertion of the denitive restoration to
the average bone level observed in the most recent
radiograph available. The average bone level was also
documented on the day of implant placement and re-
corded as the depth of implant placement.
The average change in mesiodistal peri-implant
bone levels was obtained for each implant restoration.
A negative average change in bone levels implied peri-
implant bone loss. A positive average change in bone
levels suggested an increase in crestal bone levels over
time. The primary predictor variable was the horizontal
distance between the implant and the adjacent natu-
ral tooth on the mesial and/or distal sides. The distance
(in millimeters) between the most coronal implant
plateau and the root surface of the adjacent natural
tooth was recorded for each implant on the periapical
radiograph obtained on the day of implant placement.
This distance was recorded for the mesial (distance to
mesial adjacent tooth [DATM], Fig 2) and distal sides
(distance to distal adjacent tooth [DATD], Fig 2). The
implants were categorized according to tooth-implant
distance on mesial and distal sides in the following
groups: ≤ 1 mm, 1.01 to 2 mm, 2.01 to 3 mm, and
≥ 3.01 mm. An average mesiodistal distance to adja-
cent teeth was obtained for each implant restoration.
Additionally, linear vertical measurements (in milli-
meters) from the IAI on the mesial side of the implant
to the highest crestal bone level of the adjacent tooth
were obtained (crest of adjacent tooth [CAT], Fig 2). The
changes in the distal crestal bone level of the adjacent
tooth were measured by comparing the average bone
levels on periapical radiographs obtained on the day
of implant placement and the most recent radiograph
available. The bony crest of the adjacent tooth was not
recorded in cases of immediate implant placement af-
ter tooth extraction and/or where there was an extrac-
tion of an adjacent tooth after implant placement.
Calibration of Radiographic Measurements
The methodology used to calibrate the radiographic
measurements and reviewers has been previously
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
1416 Volume 29, Number 6, 2014
posterior areas, and 140 implants (59.6%) were placed
in the maxilla. Twenty-three implants were placed in
close proximity to a mesial adjacent tooth ( ≤ 1 mm);
the average tooth-implant distance in that particular
group was 0.67 mm, and the average change in bone
level was 0.001 mm. Twenty implants were placed in
close proximity to a distal adjacent tooth ( ≤ 1 mm);
the average tooth-implant distance in that group was
0.76 mm, and the average change in bone level was
−0.002 mm. Ninety-nine implants were adjacent to a
tooth on both the mesial and distal sides, and 12 of
those implants were ≤ 2 mm from the adjacent tooth
hydroxyapatite-coated plateau root-form implant
(lengths of 5, 6, or 8 mm) placed adjacent to a natural
tooth on the mesial and/or distal sides, using a two-
stage surgical protocol. The implants were restored
with a single-tooth implant restoration and were eligi-
ble for study inclusion. Of these, 51.5% or 106 implants
were placed in female subjects. The average age was
62 years with a range of 22 to 89 years, and the average
follow-up for all implants was 42 months with a range
of 31 to 54 months. The average change in mesiodistal
bone levels was 0.11 mm. A total of 235 Bicon implants
were placed; 212 implants or 90.2% were placed in
Table 1a Descriptive Statistics for the Study Variables and the Univariate Model of the Correlation
Between Study Variables and Changes in Peri-implant Bone Levels (Outcome Variable):
Demographic, Health Status, and Anatomical Variables
Variable
Number or
mean Percentage
Parameter
estimate
95% Condence
interval
P
value
Demographic variables
(n = 206 patients, k = 235 implants)
Sex n/k n/k −0.01 (−0.13, 0.1 2) .91
Women 10 6/115 5 1. 5 /4 8 .9
Men 100/12 0 48 .5/51.1
Age at implant placement (years) 62 Range (22 –8 9) −0.002 (−0.007, 0.00 4) .54
Health status variables
ASA status
n = 200 patients, k = 229 implants
0.01 (−0.13, 0.15) .86
ASA I 23/26 11.5/11.4
ASA II 144/163 72 .0/ 71. 2
ASA III 33/4 0 16.5/17.5
Systemic disease/medications
NSAIDs 55/6 3 26.7/26.8 0.06 (−0.08, 0.2) .39
Diabetes 13/15 6.3/6.4 −0.05 (−0.31, 0. 21) .73
Fish oil 27/31 13.1/13.2 0.06 (−0.1 2, 0.24) .49
Multivitamins 6 0/70 29.1/29.8 − 0.02 (−0.17, 0.12) .68
Both NSAIDs and sh oil 13/16 6.3/6.8 0.13 (−0.11, 0.38) .27
Current tobacco use 18/23 8 . 7/ 9. 8 0.06 (−0.15, 0.27) .59
Smoking (ef fect on mandibular implants only) 8/10 3 . 9/4 . 3 − 0.28 (−0.6, 0.05) .11
Anatomical variables
k = 235 implants
Arch −0.03 (−0.16, 0.10) .69
Maxilla 140 59.6
Mandible 95 40.4
Location 0.23 (−0.00 03, 0.4 5) .06
Anterior 23 9.8
Posterior 212 90. 2
Adjacent structures
One tooth 27 11.5 0.11 (−0.09, 0.30) .28
Two t eeth 99 42.1 0.12 (−0.00 3, 0.25) .07
One tooth and one implant 109 46.4 − 0.17 (−0. 29, −0.0 4) .01
Extraction of an adjacent tooth 29 12.3 −0. 3 (−0.48, −0.12) .004
Mean mesial + distal dept h of implant placement (mm) 2.7 7 0.01 (−0.04, 0.06) .72
Bone quality (k = 225) −0.01 (−0.11, 0.0 8) .76
Typ e I 41.8
Typ e II 11 4.9
Type III 78 34 .7
Typ e I V 132 58 .7
Total on distal side 137
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
The International Journal of Or al & Maxillofacial Implants 1417
cases of internal or external root resorption were doc-
umented. The distance of an implant to an adjacent
tooth was not correlated to postoperative pain, extrac-
tion of adjacent teeth, or endodontic treatment of the
adjacent tooth when evaluated statistically (Table 2).
Effect of Implant Proximity on the Bony Crest
of the Adjacent Tooth
Changes in the bony crest of 100 teeth mesially adja-
cent to an implant were evaluated. The eect of tooth-
implant distance on the distal bony crest of those
teeth was evaluated from the day of implant place-
ment to the last recall appointment. Average changes
in the distal bony crest of the mesial adjacent teeth of
−0.42 mm and −0.27 mm were recorded when the
tooth-implant distances were ≤ 1 mm and > 3 mm,
respectively. There was no signicant statistical rela-
tionship between the distance of an adjacent plateau
on both sides. Kappa statistics showed substantial
interexaminer agreement and almost perfect intraex-
aminer agreement. The average intraexaminer and
interexaminer kappa coecients were 0.87 and 0.74,
respectively. The descriptive statistics for the study
variables are presented in Tables 1a and 1b.
Effect of Implant Proximity on Adjacent Teeth
Because the implants were not commonly placed equi-
distant from each adjacent tooth, the complications
observed on each side (mesial and distal) were evalu-
ated independently and correlated to the distance of
the implant from the natural tooth on that side. Table 2
presents the descriptive statistics of the complications
stratied by the distance from mesial and distal adja-
cent teeth as well as the univariate statistical correla-
tion between each specic complication (outcome)
and the distance to adjacent teeth (predictor). No
Table 1b Descriptive Statistics for the Study Variables and the Univariate Model of the Correlation
Between Study Variables and Changes in Peri-implant Bone Levels (Outcome Variable):
Other Variables and Primary Predictor
Variable
Number or
mean Percentage
Parameter
estimate
95% Condence
interval
P
value
Other variables
k = 235 implants
Implant length (mm) 0.03 (−0.02, 0.0 8) .31
531 13.2
691 38 .7
8113 48.1
Immediate extrac tion 111 47. 2 0.02 (−0.1, 0.15) . 74
Opposing structures (k = 213) −0.13 (−0.31, 0.06) .20
Natural teeth 179 84.0
Implant-supported restorations 34 16.0
Bone augmentation at implant (k = 228)
Internal sinus elevation 27 11.8 0.12 (−0.07, 0.31) .24
Use of β-TCP 63 2 7. 6 0.09 (−0.0 5, 0.23) .21
Abutment length (k = 225) 0.05 (− 0.08, 0.18) .49
Short (2 mm) 126 56.0
Long ( > 3 mm) 99 44.0
Abutment width (k = 2 17) 0.1 (−0.0 4, 0.23) .17
5 mm 133 61.3
6.5 mm 84 38 .7
Bruxism (k = 235) 46 19.6 −0.0 3 (−0.16, 0.16) .97
Primary predictor
Mean mesial + distal distance to adjacent teeth (mm) 3.20 Range (0.6 –9) 0.02 (−0.02, 0.0 6) .31
Distance to mesial adjacent tooth (mm) Mean 3.0 R ange (0−10.6) 0.0 03 (− 0.03, 0.0 4) .87
≤ 1 23 11.8
1.01–2 44 22.6
2.01– 3 44 22.6
> 3.01 84 4 3.1
Total on mesial side 195
Distance to distal adjacent tooth Mean 3.6 Range (0.3−14.6) 0.0 2 (− 0.01, 0.05) .18
≤ 1 20 14.6
1.01–2 18 13.1
2.01– 3 32 23.3
> 3 67 48.9
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
1418 Volume 29, Number 6, 2014
restorations where an adjacent tooth had been ex-
tracted and implant restorations where adjacent teeth
remained in place. Univariate and multivariate statis-
tical models were developed to evaluate the eect
of proximity of adjacent teeth on changes in peri-im-
plant bone levels. According to the univariate analysis
shown in Table 1, there were signicant statistical cor-
relations between changes in peri-implant bone levels
over time and the following variables (P ≤ .15): eect
of smoking on mandibular implants (P = .11); implant
location, anterior vs posterior (P = .06); adjacent struc-
tures, two teeth (P = .07); adjacent structures, one
tooth/one implant (P = .01); and extraction of an ad-
jacent tooth (P = .004). The results of the multivariate
analysis are summarized in Table 3. The primary predic-
tor variable was the average distance to adjacent teeth,
and the primary outcome variable was the change
in peri-implant bone levels over time. The other vari-
ables included in the model were considered biologi-
cally important (age at implant placement and sex) or
potential confounders (variables with P values ≤ .15
in the univariate analysis [Table 1]). Distance to adja-
cent teeth was not statistically signicantly correlated
to changes in peri-implant bone levels (P = .13). The
extraction of an adjacent tooth was correlated with
an increase in peri-implant bone loss (P = .008). The
association between smoking and peri-implant bone
loss in the mandible (P = .08) was close to statistical
signicance.
root-form implant and changes in the bony crest on
the adjacent tooth (P = .32, Table 2).
Tooth Proximity and Implant Failures
There were eight documented implant failures. There
were no failures of implants placed ≤ 1 mm to an ad-
jacent tooth.
Tooth-implant proximity on the distal side (P = .21,
Table 2) was not correlated to the failure of a plateau
root-form implant when evaluated statistically. A ten-
dency for increased implant failures was documented
with an increase in the distance to mesial adjacent
teeth; this tendency almost reached statistical signi-
cance (P = .06, Table 2).
Tooth Proximity and Peri-implant Bone Levels
Average peri-implant bone levels of 0.12 mm and
0.10 mm were recorded in the mesial and distal sides,
respectively. Average changes in peri-implant bone
levels for implants distributed by distance to adjacent
teeth as well as the univariate correlation between
tooth-implant proximity on the mesial and distal sides
and changes in peri-implant bone levels over time are
presented in Table 2. The proximity of a natural tooth
on the mesial (P = .87) or distal sides (P = .18) was not
statistically signicantly associated with peri-implant
bone loss (Table 2).
Average changes in bone levels of −0.18 mm and
+0.15 mm, respectively, were recorded around implant
Table 2 Descriptive Statistics and Univariate Correlation on the Effect of Complications
Documented on the Mesial and Distal Adjacent Tooth After Implant Placement
for Implants Stratied by Tooth-Implant Distance
Mesial adjacent tooth
Complications ≤ 1 1.01–2 2 .01–3 > 3 P value
Number of implants 23 44 44 84
Implant failures 0123.06
Extractions of mesial adjacent teet h 0 5 2 7 .96
Pain 42114 .09
Lesion of endodontic origin present or RCT performed 0 0 0 2 .24
Average (mm) bone loss on distal crest of mesial adjacent tooth − 0.42 − 0.3 −0.3 3 −0. 27 .32
Average (mm) change in mesial peri-implant bone levels 0.001 0.20 0.05 0.16 .87
Distal adjacent tooth
Complications ≤ 1 1.01–2 2 .01–3 > 3 P value
Number of implants 20 18 32 67
Implant failures 0013.21
Extractions of distal adjacent teeth 1 2 5 4 .39
Pain 0239.18
Lesion of endodontic origin present or RCT performed 0 0 0 1 *
Average (mm) change in distal peri-implant bone levels −0.002 −0.15 0.13 0.18 .18
RCT = root c anal ther apy. *Not calcu lated due to small numb er of obser vations.
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
The International Journal of Or al & Maxillofacial Implants 1419
tooth-implant proximity using two-dimensional ra-
diographs that gave an approximate measurement,
whereas the aforementioned animal studies used
histologic methods that provided an exact distance.
Therefore, it is possible that the implants in the pres-
ent study only appear to be touching the roots but
were actually superimposed in the radiograph. It is
also possible that changes that may have occurred in
the adjacent root surfaces could not be identied with
periapical radiographs.
The dierences between mini-implants and plateau
root-form implants may also play a role. Mini-implants,
like conventional screw-type implants, are screwed
into an osteotomy (sometimes an undersized oste-
otomy) with a certain amount of torqueing, and con-
sequently they exert pressure on the adjacent bone
and root surface. On the other hand, plateau root-form
implants are seated into a similar-sized osteotomy
without torqueing and without pressure to adjacent
structures. This study’s ndings suggest that a plateau
root-form implant may be placed in close proximity
( ≤ 1 mm) to an adjacent natural tooth without a del-
eterious eect on that structure.
Load-Bearing Platform Switching, Tooth-
Implant Proximity, and Bone Preservation
In the clinical technique used when restoring plateau
root-form implants, contact is created between the
base of the abutment and crestal bone with the pur-
pose of loading bone coronal to the implant-abutment
interface (see Fig 1), a concept called load-bearing
platform switching (Fig 3). The objective of load-bear-
ing platform switching is to convert the transcrestal
portion of the implant-abutment complex into a load-
transferring structure, with the objective of loading
bone coronal to the rst implant thread or plateau,
in the same way that bone is loaded apically to that
thread. In the successful use of load-bearing platform
DISCUSSION
This study suggests that close proximity to adjacent
teeth is not a risk factor for the survival or peri-implant
bone levels of single-tooth plateau root-form im-
plants. The results also demonstrated that proximity
of a plateau root-form implant to an adjacent tooth
or teeth does not have an adverse eect (measured as
increased pain, presence of root resorption, develop-
ment of endodontic complications, number of extrac-
tions of adjacent teeth, and crestal bone loss) on these
structures. However, the extraction of an adjacent
tooth had a signicant eect on peri-implant bone
levels. These results are consistent with previous stud-
ies that have reported low complication rates on teeth
adjacent to single-tooth implants.33–35
In this study, postoperative pain was used to gauge
the possibility of damage to an adjacent tooth. It was
hypothesized that the closer the implant was to an
adjacent tooth, the more likely it would be that the
adjacent tooth could have become symptomatic. This
hypothesis was not supported by the data of the study.
There was no correlation between the number of pa-
tients reporting pain postoperatively and the distance
of the implant to the adjacent teeth.
Even though there were 43 implants placed
< 1 mm to an adjacent tooth and in some instances
apparently touching the adjacent root surface, there
were no cases of root resorption documented. There
is limited research on the subject of root resorption
as a result of dental implant placement. Some animal
studies have reported root surface resorption after the
placement of mini-implants less than 1 mm from the
root surface36 and that the incidence of root resorption
increased when mini-implants were less than 0.6 mm
from the root.37 The dierences between those ani-
mal studies and the present investigation may explain
the dissimilar ndings. The present study measured
Table 3 Multivariate Analysis of the Correlation Between the Study Variables and the Outcome
Variable (Change in Peri-implant Bone Levels)
Variable PE 95% CI P value
Primary predictor variable
Average distance to adjacent teeth 0.03 −0.01, 0.07 .13
Biologically relevant factors
Age at implant placement 0.0001 −0.006, 0.006 .98
Sex −0.0 9 −0.22, 0.0 4 .19
Potential confounders
Location, anterior vs posterior 0.17 − 0.06, 0.4 .18
Adjacent structures, one tooth and one implant −0.19 −0.4, 0.02 .10
Adjacent structures, two teeth 0.01 −0. 2, 0.2 .93
Smoking, effect on mandibular implants −0.3 −0.64, 0.02 .08
Extraction of an adjacent tooth −0.3 −0.5, −0.1 .00 8*
CI = cond ence inte rval; PE = p aramete r estimat e.
*Statistically signicant at P ≤ .05.
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
1420 Volume 29, Number 6, 2014
restoration that will follow. Thus designed, the abut-
ment base creates a horizontal surface opposing the
space created with the implant’s sloping shoulder,
which in turn will transmit compressive loads to ex-
isting or potential crestal bone (see Fig 1 and label 2,
Fig 3b).
This study reported that proximity of a plateau root-
form implant was not associated with the loss of the
crestal bone around the adjacent tooth. These ndings
are consistent with those of Vela et al27 but dier from
the results reported by Esposito et al26 and Krennmair
et al.34 Vela et al27 reported that placing platform-
switched implants 1 mm from teeth does not have
an adverse eect on the bone level adjacent to them,
whereas Krennmair et al34 and Esposito et al26 report-
ed increased bone loss on teeth adjacent to implants
with matching implant-abutment platforms.
Furthermore, proximity of an adjacent tooth was
not correlated to changes in peri-implant bone levels
around plateau root-form implants restored with the
load-bearing platform-switch concept. These ndings
are consistent with previous studies using this type of
switching, two conditions will necessarily result. First,
space must exist for the creation or maintenance of
crestal bone; second, an opposing horizontal loading
surface must have been created. The biomechani-
cal design used in the implants in this study includes
a double platform shift that meets the above criteria,
and is explained as follows.
The rst element, the implant’s sloping shoulder, is
a space-creating feature. The creation of space is ac-
complished through the use of a gradual vertical and
horizontal platform shift that occurs as the implant
shoulder slopes inwardly and coronally toward the
implant-abutment interface, adding a vertical com-
ponent to the concept of platform switching. This rst
shift provides space coronoapically to the implant-
abutment interface for the adjacent tooth’s bony crest
in cases where there is close tooth-implant proximity
(see label 1, Fig 3b).
The loading element, which represents a reversing
platform switch, is the spherical base of the abutment.
This shift occurs as the abutment post transitions out-
ward to achieve the proper emergence prole for the
Fig 3 Implant abutment platforms and tooth-implant proximit y: (a) platform switching (PS): a 4 mm–wide implant placed at the level
of the bone crest and restored with a 3-mm abutment (Reprinted from Vela et al27 with permission.); (b) load-bearing platform switch-
ing (LBPS): intracrest ally placed sloping shouldered implant restored with a nonshouldered abutment with a titanium spherical base;
(c) matching platforms (MP): both implant and abutment are of similar size (note that image was rotated when enlarged to provide
for comparison purposes). (b) With LBPS, there is a double platform shift that has a ver tical and a horizontal component. In the rst
shift (labeled 1), the implant sloping shoulder allows for space for hard and sof t tissues and permits a slow transition bet ween the
wider implant and narrower abutment. In the second shif t (labeled 2), the abutment base is the loading component. In LBPS, occlusal
loads (green arrows) may be transferred to bone (white arrows) coronal to the implant-abutment interface (IAI, dotted green line). In
PS and MP, bone coronal of the IAI (dotted green line) remains unloaded.
a
a
b
b
c
c
1
2
Conventional
platform
Load-bearing
platform
switching
Platform
switching
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
The International Journal of Or al & Maxillofacial Implants 1421
restorative platforms of similar size restored with the
platform-switching concept (a design also used in the
Vela study27 and shown in Fig 3a), when placed in close
proximity to an adjacent tooth the implant shoulder
and the rst thread would contact bone at the same
time, and the resulting coronal space created by the
horizontal reduction of the abutment platform would
be insucient for the development of adequate crest-
al bone between the implant and tooth. Consequently
such implants cannot be placed as close to the adja-
cent tooth as a plateau root-form implant without an
adverse outcome.
For implants with matching implant abutment
platforms that have been placed supracrestally, as in
the case of the implants used in the Krennmair et al34
study, if the implant platform is in contact with the ad-
jacent tooth, there is no space left for bone or papilla.
Consequently, the dierences between the implant-
abutment platform designs used may explain the cor-
relation between tooth-implant proximity and bone
loss reported in these previous studies.26,27,34 More-
over, a certain degree of loss of the adjacent tooth’s
bony crest would be expected immediately after the
placement of a screw-design implant that exerts pres-
sure on that bone. The implants used in the previously
cited studies26,27,34 were screw-type or press-t. On the
other hand, a plateau root-form implant that is placed
several millimeters subcrestally, such as the implants
in the present study, leaves space between the sloping
shoulder and the adjacent tooth’s bony crest (Figs 3b
and 4), and it is seated, not torqued, into the osteoto-
my (see Fig 1).
Finally, after creating space for bone, any result-
ing bone must be loaded to be maintained.8–16 It is
widely accepted that bone mass and architecture are
regulated in response to the local strains produced by
functional loading, and an increase in mechanical us-
age tends to increase bone deposits.13 Unusual strain
distributions, high strains, and high strain rates seem
to be particularly osteogenic.12,14 Stress in the magni-
tude of 2.48 × 10 N/mm2 has been shown to cause an
increase in bone growth.15 Bone increases volume and
thickens trabeculae in response to the strains experi-
enced during loading.16
Studies on the eect of forces on crestal bone sur-
rounding dental implants have shown conicting
results. Some clinical studies using splinted smooth-
surfaced implants have associated high occlusal stress
with bone loss.3–6 However, there is not sucient evi-
dence to support a cause-and-eect relationship be-
tween overload and bone or implant loss.7 Moreover,
bone growth on dental implants has also been associ-
ated with increased loads in long-term studies. A ret-
rospective study reported an increase in vertical bone
height beneath the cantilever area of mandibular xed
implant but dier from the results of other studies with
dierent implant platform designs.26,27,34 Urdaneta et
al23 reported on factors associated with changes in
crestal bone levels on 326 single-tooth plateau root-
form implants that were followed for an average of 5.9
years. No signicant correlation between the distance
of an implant to its mesial or distal adjacent structure
and vertical changes in peri-implant bone levels were
reported. In contrast, other studies26,27,34 have report-
ed a signicant increase in peri-implant bone loss with
decreasing tooth-implant distance. The present inves-
tigation reports on the outcome of 235 implants, some
of which were placed in very close proximity to adja-
cent roots, with 43 implants being placed ≤ 1 mm (av-
erage, 0.71 mm) from adjacent roots; this is closer than
the average distance of 0.99 mm reported in the Vela
et al27 study. The fact that the implants in the present
study could be placed closer to adjacent teeth without
apparent sequelae may be explained by dierences
between the implant abutment platform designs used
in both studies as well as their surgical placement tech-
niques. The present study evaluated plateau root-form
implants with a sloping shoulder that were placed on
average 2.77 mm apical to the crest of the bone (Fig 2).
The implants in the Vela et al27 study (Fig 3a) were
placed at the level of the bone crest and restored with
platform-switching implant-abutment platforms, and
the implants in the Krennmair et al34 study, restored
with matching platforms, were placed supracrestally.
The implants in the present study were restored
with a consistent implant platform reduction, the neck
of the implant being narrower than its body by 2 mm
(1 mm on each side). This creates space between the
rst implant plateau to the base of the abutment (vol-
ume of approximately 23.4 mm3). This space, initially
designed for bone to be loaded by the spherical base,
also allows for the development and maintenance of
crestal bone associated with the adjacent tooth, if the
implant plateau itself is in close proximity or contact
with the adjacent root surface (Figs 2, 3b, and 4). Be-
cause this implant platform reduction is signicantly
larger than the platform switch documented in the
Vela study27 (where the abutment was narrower than
the neck of the implant shoulder by 1 mm, or 0.5 mm
on each side), it is possible that the larger platform
switch could have played a role in the reduced bone
loss observed. Clinical studies have reported that there
is an inverse relationship between the extent of mis-
matching and the amount of bone loss.38
For implants with a wider neck placed at the crest of
the bone in close proximity to an adjacent tooth, and
then subsequently restored with platform switching as
done in the Vela study,27 the implant platform would
necessarily contact the adjacent tooth before the
rst implant thread. For implants with diameters and
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
1422 Volume 29, Number 6, 2014
loads of 100 N. These researchers reported nding von
Mises strains between 1750 and 3000 in the bone sur-
rounding the spherical base of the abutments when
the implants had been placed 1 and 2 mm subcrest-
ally. Even though contact between the abutment’s
base and bone is usually created on the day of inser-
tion of the abutment, the authors of the study have
observed that when removing this type of abutment
after several years of loading, a thin layer of soft tissue
usually exists between the abutment and bone. On the
other hand, the authors have also observed in some
instances, such as the case shown in Fig 1, that the soft
tissue immediately adjacent to the abutment’s base
(shown as a radiolucency in the periapical radiograph
in Fig 1e) appears to be replaced by bone over time
(Figs 1f and 1g), which suggests load transmission to
bone through those tissues. However, the exact com-
position of the tissue and its capacity to transfer loads
will need to be evaluated in future histologic studies.
Peri-implant bone growth coronal to the implant-
abutment interface toward the abutment base, and
in some instances in apparent contact with the ti-
tanium spherical base, has been reported in several
peer-reviewed clinical studies using plateau root-form
implants.22–25 In fact, a crown cemented on a prefabri-
cated abutment with a spherical base has been shown
to be one of ve predictors of peri-implant bone
growth surrounding plateau root-form implants.23
implant-supported prostheses followed from 5 to 19
years, and the authors hypothesized that the bone
gain may be associated with the increased stress dis-
tributed from the prosthesis and transmitted directly
to bone through the osseointegrated implants.8 A 24-
year retrospective study reported signicant increases
in bone height in edentulous mandibles (from 4 to 13
mm) when the loads of complete dentures were sup-
ported by ramus-frame implants.9 Clinical studies us-
ing newer implant designs with rough-surface texture
treatments, some of them supporting unsplinted resto-
rations, have shown that increased stress, as measured
by increased crown-to-implant ratios, is not associated
with peri-implant bone loss,10,22,39 but rather, it is cor-
related with prosthetic complications22 and with bone
growth,10 and the bone loss observed could have been
caused by stress shielding–induced bone atrophy.39
Bone coronal to the implant-abutment interface
(IAI) cannot be physically loaded with matching im-
plant-abutment platforms or with conventional plat-
form switching. As shown in Figs 3a and 3c, traditional
platform designs lack a loading surface for bone cor-
onal to the IAI. In contrast, the base of the abutment
in plateau root-form implants was designed to load
bone coronal to the IAI.20 This rationale was tested in
vitro by Chou et al.21 In a nite element stress analysis
of a simulated mandibular model, plateau root-form
implants and abutments were subjected to nonaxial
Fig 4 Periapical radiographs and clinical photograph of a plateau root-form implant restoring a mandibular rig ht central incisor with
an integrated abutment crown. Notice the crestal bone mineralization, healthy distal dental papilla, and the stabilit y of the bony
crest of the adjacent tooth 1 year after crown insertion, despite the fact that the implant plateaus seem to be contacting the root
surface of the distal adjacent tooth. The increased space provided by the LBPS is occupied by (c and e) bone and (f) sof t tissues.
Radiographs: (a) implant placement (April 11, 2011), (b and d) crown/abutment insertion (October 3, 2011), and (c and e) recall ap -
pointment (October 8, 2012). (f) Photograph, February 5, 2013.
a b c
d e f
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
The International Journal of Or al & Maxillofacial Implants 1423
3. The placement of a plateau implant in close prox-
imity to an adjacent tooth did not cause detect-
able damage to the root surface or to the crestal
bone on the adjacent tooth.
4. This study introduced the concept of a bone-
loading platform switch. In this platform design,
an implant shoulder gradually slopes inward and
coronally, toward the implant-abutment interface,
creating space for crestal bone, while the base
of the implant abutment oers a loading surface
through which compressive loads may be exerted
on existing or potential crestal bone. The combi-
nation of such design elements oered signicant
advantages when placing implants in close prox-
imity to adjacent teeth.
ACKNOWLEDGMENTS
The authors would like to recognize Megan McKenna and Kim -
berly Panetta for their contribution to this study. Thomas Driskell
designed the concept of a bone -loading implant abutment plat-
form and gave it the name “load-bearing platform switch.” The
invaluable contribution of Fred Weekley is also acknowled ged.
Drs Seemann, Dragan, Leary, and Chuang have no nancial in -
terests related to any products involved in this study. Dr Urda -
neta is a faculty at the Bicon Institute and lectures on Bicon
dental implants. Mr Lubelski is a par t-time employee of Bicon
(Boston, Massachuset ts, USA). This study was funded by the
authors and their institutions.
REFERENCES
1. Mayer T, Hawley C , Gunsolley J, Feldman S. The single-tooth
implant: A viab le alternative for single-tooth replacement. J Peri-
odontol 2002;73:687–693.
2. Smith DE, Zarb GA. Criteria for success of osseointegrated endosse-
ous implants. J Prosthet Dent 20 02;69:567–572.
3. Quirynen M, Naert I, van Steenberghe D. Fixture design and
overload inuence marginal bo ne loss and xture success in the
Brånemark System. Clin Oral Imp lants Res 1992;3:104–111.
4. Naert I, Quirynen M , van Steenberghe D, Darius P. A study of 589
consecutive implants supporting complete xed prosth eses. Part II:
Prosthetic aspects. J Prosthet Dent 1992;68:949–956.
5. Jemt T, Lekholm U, Adell R . Osseointegrated implants in the
treatment of partially edentulous patients: A preliminary study on
876 consecutively placed xtures. Int J Oral Maxillofac Implants
1989;4: 211–217.
6. Linquist LW, Rockler B, Carlsson GE. B one resorption around
xtures in e dentulous patients treated with mandibular xed tissue
integrated prosthesis. J Prosthet Dent 1988;59:59– 63.
7. Naert I, Duyck J, Vandamme K. Occlusal overload and b one/implant
loss. Clin Oral Implants Res 2012;23(suppl 6):95–107.
8. Dhima M, Balshi T, Wolnger G, Petropoulus VC, Bashi S. A retro-
spective analysis of mandibular bone height changes asso ciated
with 81 screw-retained implant-supported prostheses with distal
cantilevers: A long-term follow-up analysis. Int J Oral Maxillofac
Implants 2013;28:854–859.
9. Rober ts RA. A 24-year retrospective study of bone growth
after implant placement. J O ral Implantol 2005;31:98–103.
10. Blanes R J, Bernard JP, Blanes ZM, Belser UC. A 10 -year prospective
study of ITI dental implants placed in the posterior region. I: Clinical
and radiographic results. Clin Oral Implants Res 2007;18:699–706.
Additionally, the absence of a titanium spherical
base has been signicantly correlated to peri-implant
bone loss on this type of implants.25 Consequently, it
is hypothesized that the bone growth coronal to the
implant-abutment interface surrounding plateau root-
form implants may be the bone’s response to occlusal
loads directed through the abutments’ bases, and the
absence of bone coronal to the rst implant thread,
commonly observed around implants with traditional
platforms, could be the result of disuse atrophy due to
under-loading. These theories will need to be investi-
gated in future studies.
Factors Affecting Peri-implant Bone Levels on
Plateau Root-Form Implants
The multivariate analysis demonstrated that the ex-
traction of an adjacent tooth was signicantly correlat-
ed with peri-implant bone loss after crown insertion, as
well as a tendency for mandibular peri-implant bone
loss in patients who smoked tobacco. These ndings
were expected and consistent with previous studies.
Retrospective studies on plateau root-form implants
have reported that a change in adjacent structures
(extraction of a tooth, placement of an implant) was
one of 13 predictors for peri-implant bone loss23 and
that the correlation between smoking and mandibular
bone loss was close to statistical signicance.24
This study’s limitations included the retrospective
design, the sample size, and the use of a single center
for implant placement. Retrospective cohort studies
rely on complete data entered into the patient’s his-
tory and have less validity than randomized prospec-
tive clinical trials due to possible selection bias and
confounding factors. A larger sample would have en-
hanced the power of the study. However, it is impor-
tant to point out that this study’s sample is signicantly
larger (235 implants, 206 subjects) than the sample
used in other similar investigations (less than 80 im-
plants).26,27,34 Prospective studies with larger sample
sizes are necessary to validate the current ndings.
CONCLUSIONS
Within the limitations of this study, the following con-
clusions may be drawn:
1. The distance between the rst implant plateau and
root surface of an adjacent tooth was not a risk fac-
tor for the failure of plateau root-form implants.
2. There was no signicant correlation between
tooth-implant proximity and changes in peri-im-
plant bone levels surrounding plateau root-form
implants.
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Urdaneta et al
1424 Volume 29, Number 6, 2014
26. Esposito M, Ekestubbe A, Grondahl K. Radiological evaluation
of marginal bone loss at tooth sur faces facing single Brånemark
implants. Clin Oral Implants Res 1993;4:151–157.
27. Vela X, Mendez V, Rodriguez X, Segala M, Tarnow DP. Crestal bone
changes on platform-s witched implants and adjacent teeth when
the tooth-implant distance is less than 1.5 mm. Int J Periodontics
Restorative D ent 2012;32:149–155.
28. Urdaneta RA, Marincola M. The Integrated Abutment Crown, a
screwless and cementless restoration for single-tooth implants: A
report on a new technique. J Prosthodont 2007;16:311–318.
29. American Academy of Anesthesiologists (ASA). Physical status
classes. In: Clinical Anesthesia Procedures of the Massachuset ts
General Hospital, ed 5, vol 3. Philadelphia: Lip pincott-Raven, 1998.
30. Truhlar RS, Orenstein IH, Morris HF, Ochi S. Distribution of bone
quality in patients receiving endosseous dental implants. J Oral
Maxillofac Surg 1997;55(suppl 5):38–45.
31. Lobbezoo F, Ahlberg J, Glaros AG, et al. Bruxism dened and
graded: An international consensus. J Oral Rehabil 2013;40:2–4.
32. Chuang SK, Wei L J, Douglas CW. Risk factors for dental implant
failure: A strategy for the analysis of clustered failure-time observa-
tions. J Dent Res 2002;1:572–577.
33. Priest G. Single-tooth implants and their role in preserving remain -
ing teeth: A 10 year survival study. Int J Oral Maxillofac Implants
199 9;14 :181–18 8.
34. Krennmair G, Piehslinger E, Wagner H. Status of teeth adjacent to
single tooth implants. Int J Prosthodont 2003;16:524–528.
35. Misch CE, Misch- Dietsh F, Silc J, et al. Posterior implant single-tooth
replacement and status of adjacent teeth during a 10-year period: A
retrospec tive report. J Periodo ntol 2008;79:2378–2382.
36. Kim H, Kim TW. Histolo gic evaluation of root-surf ace healing after
root contact or approximation during placement of mini-implants.
Am J Orthod Dentofacial Orthop 2011;139:752–760.
37. Lee YK, Kim JW, Baek SH, Kim TW. Chang YI. Root and bone
response to the proximity of a mini-implant under orthodontic
loading. Angle Orthod 2010;80:452–458.
38. Canullo L, Fedele GR , Ianello G, Jepsen S. Plat form switching and
marginal bone alterations: The results of a randomized- controlled
trial. Clin Oral Implants Res 2010;21:115–121.
39. Rokni S, Todescan R , Watson P, Pharoah M, Adegb embo AO,
Deporter D. An assessment of crown-to-root ratios with sintered
porous-sur faced implants suppor ting prostheses in par tially eden-
tulous patients. Int J Oral Ma xillofac Implants 2005;20:69–76.
11. Wol J. The Law of Transformation of Bone. Berlin: August
Hirschwald, 1892.
12. Lanyon LE. Using functional loading to inuence bone mass and ar-
chitecture: Objectives, mechanisms, and relationship with estrogen
of the mechanically adaptive process in bone. B one 1996;18(suppl
1):37S–43S.
13. Frost HM. The mechanostat: A proposed pathogenic mechanism
of osteoporosis and the bone mass eects of mechanical and non-
mechanical agents. Bone Miner 1987;2:73–85.
14. Turner CH, Warden SJ, Bellido T, et al. Mechanobiology of the skel-
eton. Science Signaling 2009;2:3.
15. Hassler CR, Rylicky EF, Cummings KD, Clark LC. Quantication of
bone stress during remode ling. J Biomech 1980;13:185–190.
16. Van der Meulen MC, Yang X, Morgan TG, Bostrom MP. The eects
of loading on cancellous bon e in the rabbit. Clin Ortho p Relat Res
2009;467:2000–2006.
17. Misch CE, Bidez MW. A scientic rationale for dental implant design.
In: Misch CE (ed). Contemporary Implant Dentistr y, ed 2. St Louis:
Mosby, 1999:329–343.
18. Laz zara RJ, Porter SS. Plat form switching: A new concept in implant
dentistry for controlling p ost restorative crestal bone levels. Int J
Periodontic s Restorative Dent 2006;26:9–17.
19. Al -Nsour MM, Cjhan HL, Wang HL. Eect of the platform-switching
technique on preservation of peri-implant marginal bone: A sys-
tematic review. Int J Oral Maxillof ac Implants 2012;27:138–145.
20. Bicon Dental Implants. Surgical Manual: Step By Step Techniques.
http://www.bicon.com/tech/t_manuals.html. Accessed February
20, 2014.
21. Chou H-Y, Muftu S, Bozkaya D. Combine d eects of implant inser-
tion depth and alveolar bone quality on peri-implant bone strain
induced by a wide-diame ter, short implant and a narrow- diameter,
long implant. J Prosthet Dent 2010;104:293–300.
22. Urdaneta RA, Rodriguez S, McN eill DC, Weed M, Chuang SK. The
eect of increased crown-to-implant ratio on single-tooth locking-
taper implants. Int J Oral Maxillofac Implants 2010;25:729–743.
23. Urdaneta RA, Daher S, Leary J, Emanuel K, Chuang SK. Factors
associated with crestal bone gain on single -tooth locking-taper
implants: The eect of nonsteroidal anti -inammatory drugs. Int J
Oral Maxillofac Implant s 2011;26:1063–1078.
24. Urdaneta RA, Leary J, Lubelski W, Emanuel KM, Chuang SK. The
eect of implant size 5 x 8 mm on cres tal bone levels around single
tooth implants. J Periodontol 2012;83:1235–1244.
25. Urdaneta RA, Leary J, Panetta KM, Chuang SK. The ee ct of
opposing structures, natural teeth vs. implants on crestal bone
levels surrounding single-tooth implant s. Clin Oral Implants Res
2014 ;2 5: e179 – e18 8.
a
ERR ATA
In the July/August 2014 issue of JOMI, in the ar ticle
“Microcomputed Tomographic and Histomorphometric
Analyses of Novel Titanium Mesh Membranes for Guided Bone
Regeneration: A Study in Rat Calvarial Defects” by Rakhmatia
et al (Int J Oral Maxillofac Implants 2014;29:826–835), on
page 831, the wrong image was placed for the “T100 -8w”
image in Fig 7. The correct image is as follows:
T100–8w
The publisher regrets these errors.
In the September/October 2014 issue of JOMI, in the article
“Biologic Width Around Different Implant-Abutment Interface
Congurations. A Radiographic Evaluation of the Effect of
Horizontal Offset and Concave Abutment Prole in the Canine
Mandible” by Caram et al (Int J Oral Maxillofac Implants
2014;29:1114–1122), on page 1117, the wrong image was
placed for Fig 3a. The correct image is as follows:
a
© 2014 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.