ArticlePDF Available

Peri-implant marginal bone loss rate pre- and post-loading: An exploratory analysis of associated factors

Authors:
Article

Peri-implant marginal bone loss rate pre- and post-loading: An exploratory analysis of associated factors

Abstract and Figures

Objectives: To perform an exploratory analysis of factors influencing annual rates of peri-implant marginal bone loss (RBL) calculated over different time-frames, at implants unaffected by peri-implantitis. Material and Methods: 154 implants from 86 patients were reviewed at 1.6-6.8 years after placement. Marginal bone levels (MBL) were assessed on intraoral-radiographs at 3 time-points: immediately post-placement, time of loading, and least 1-year post-loading. RBLs (mm/year) were computed using these 3 time-frames and corresponding MBL changes as: RBL-placement-loading, RBL-loading-review, RBL-placement-review. Exploratory ordination of 3 RBLs, corresponding time-durations, and 17 background factors was used for visualization. Hierarchical linear mixed effects models (MEM) with predictor selection were applied to RBL outcomes. The correlation of actual MBL with MBLs predicted by RBL placement-loading and RBL loading-review was tested. Results: Median RBL placement-loading was 0.9 mm/year (IQR=2.02), loading-review was 0.06 mm/year (IQR=0.16) and overall RBL placement-review was 0.21 mm/year (IQR=0.33). Among patient variance was highest for RBL placement-loading (sd=0.66). Longer time predicted lower RBL in all time-frames. Shorter time of loading significantly predicted lower RBL placement-review. Augmentation predicted lower RBL placement-loading, while anterior location and older age predicted lower RBLs placement-loading placement-review. Only MBL projected using RBL-placement-loading significantly correlated with actual MBL. Conclusions: Exploratory analysis indicated RBL varied with the time-duration used for calculation in pre-, post-loading, and overall periods. In each period, RBL declined with increasing time. Earlier loading predicted lower overall RBL. Higher pre-loading RBL predicted worse actual bone level.
Content may be subject to copyright.
1
Clinical Oral Implants Research
Peri-implant marginal bone loss rate pre- and post-loading: an exploratory analysis of associated
factors
Acharya A.1,2*, Leung M.C. 1*, Ng K.T.1, Fan M.H.M. 1, Fokas G. 1, Mattheos N.1
*These authors have contributed equally to the manuscript
Authors’ Designation/Affiliations:
1. Implant Dentistry, Prosthodontics, Faculty of Dentistry, University of Hong Kong
2. Department of Periodontology, Dr. D.Y.Patil Dental College and Hospital, Dr. D. Y. Patil
Vidyapeeth, Pimpri, Pune.
Corresponding author:
Nikos Mattheos
Clinical Associate Professor
Prosthodontics, Prince Phillip Dental Hospital
34 Hospital Road, 3F, Bloc B, Sai Ying Pun, Hong Kong SAR PR China
nikos@mattheos.net, Tel. +852 5188 5893
Running title: Rate of peri-implant marginal bone loss
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
2
2
Abstract:
Objectives: To perform an exploratory analysis of factors influencing annual rates of peri-implant
marginal bone loss (RBL) calculated over different time-frames, at implants unaffected by peri-
implantitis. Material and Methods: 154 implants from 86 patients were reviewed at 1.6-6.8 years after
placement. Marginal bone levels (MBL) were assessed on intraoral-radiographs at 3 time-points:
immediately post-placement, time of loading, and least 1-year post-loading. RBLs (mm/year) were
computed using these 3 time-frames and corresponding MBL changes as: RBL-placement-loading,
RBL-loading-review, RBL-placement-review. Exploratory ordination of 3 RBLs, corresponding time-
durations, and 17 background factors was used for visualization. Hierarchical linear mixed effects
models (MEM) with predictor selection were applied to RBL outcomes. The correlation of actual MBL
with MBLs predicted by RBL placement-loading and RBL loading-review was tested. Results: Median
RBL placement-loading was 0.9 mm/year (IQR=2.02), loading-review was 0.06 mm/year (IQR=0.16)
and overall RBL placement-review was 0.21 mm/year (IQR=0.33). Among patient variance was highest
for RBL placement-loading (sd=0.66). Longer time predicted lower RBL in all time-frames. Shorter
time of loading significantly predicted lower RBL placement-review. Augmentation predicted lower
RBL placement-loading, while anterior location and older age predicted lower RBLs placement-loading
placement-review. Only MBL projected using RBL-placement-loading significantly correlated with
actual MBL. Conclusions: Exploratory analysis indicated RBL varied with the time-duration used for
calculation in pre-, post-loading, and overall periods. In each period, RBL declined with increasing time.
Earlier loading predicted lower overall RBL. Higher pre-loading RBL predicted worse actual bone level.
3
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
4
3
Introduction:
Maintenance of peri-implant marginal bone level is a key criterion for implant success. Surgical
placement is usually followed by remodelling and peri-implant marginal bone loss Many thresholds of
acceptable marginal bone loss are reported. Widely adopted thresholds are; 0.1-0.2 mm of bone loss per
year (Albrektsson, Zarb, Worthington, & Eriksson, 1986) or loss of 2 mm (Misch et al., 2008) after the
first year of loading. Other reported thresholds include; 2.5 mm bone loss after 5 years (Berglundh,
Persson, & Klinge, 2002) and, 1-1.5mm (Derks & Tomasi, 2015) or 0.4mm (Koldsland, Scheie, & Aass,
2010) from the time-point of loading. Although these bone loss thresholds provide easy clinical ‘cut-
offs', they do not predict future bone loss. A common reference point for measuring bone loss is
prosthetic loading. With advances in implant dentistry, earlier loading is more common than before. On
the other hand, large augmentation/GBR procedures are also increasing. In such cases, implants are
more likely to be loaded later than usual. Taken together these suggest a wide range of loading times
exist in current clinical scenarios. If bone loss is measured from the loading event but the time of
loading is ignored, such differences could be confounding. Marginal bone remodeling is a dynamic
process. Thus, its rate changes over time. This is evident in the poor correlation of total marginal bone
loss with time in function (Hasegawa, Hotta, Hoshino, Ito, Komatsu, & Saito, 2015). Therefore, the rate
of bone loss (RBL) has been proposed as a better index of implant success than bone loss or bone level
values (Galindo-Moreno, León-Cano, Ortega-Oller, Monje, O′Valle & Catena, 2015). RBL may have
predictive use. High early RBL correlated to worse marginal bone levels, and a ‘bone loser’ phenotype
could be identified (Galindo-Moreno, et al., 2015). However, in practice, the reported annual bone loss
may be based on annual measurements or calculated from change in bone levels measured over varying
time intervals. A previous study used four time-intervals and found different RBL values for each,
leading to different implant success rates (Geraets, Zhang, Liu & Wismeijer, 2014).
5
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
6
4
Many patient, site, and implant-related factors have are associated with peri-implant bone loss. These
include the duration of healing (Naert, Koutsikakis, Duyck, Quirynen, Van Steenberghe & Jacobs, 2002;
Rouck, Collys, & Cosyn, 2008), the depth of implant placement (Valles, Rodríguez-Ciurana, Clementini,
Baglivo, Paniagua, & Nart, 2018), location and type of implant-abutment interface (Atieh, Ibrahim, &
Atieh, 2010; van Eekeren, Tahmaseb, & Wismeijer, 2015), abutment height (Nóvoa, Batalla, Caneiro,
Pico, Liñares & Blanco, 2017 ; Blanco, J., Pico, A., Caneiro, L., Nóvoa, L., Batalla, P., & Martín-
Lancharro, 2018), implant geometry and surface characteristics (Zechner et al., 2004; Şener-Yamaner,
Yamaner, Sertgöz, Çanakçi, & Özcan, 2017), type and timing of occlusal load (Schincaglia, et al., 2016),
bruxism and occlusal overload (Kozlovsky, et al., 2007; Zhou, Gao, Luo, & Wang, 2015), soft-tissue at
implant site (Linkevicius, Puisys, Linkeviciene, Peciuliene, & Schlee, 2013; Suárez-López del Amo,
Lin, Monje, Galindo-Moreno & Wang 2016), bone density (Chow, Chow, Chai, & Mattheos, 2016),
augmentation procedures (Huang, Ogata, Hanley, Finkelman, & Hur, 2014). Inflammatory peri-implant
disease and its risk factors; periodontal disease susceptibility (Corcuera-Flores, Alonso-Domínguez,
Serrera-Figallo, Torres-Lagares, Castellanos-Cosano & Machuca-Portillo, 2016), smoking (Peñarrocha,
Palomar, Sanchis, Guarinos & Balaguer, 2004; Levin, Hertzberg, Har-Nes, & Schwartz-Arad, 2008) and
diabetes (Alrabiah et al., 2018) also influence marginal bone loss. Factors that influence how the annual
rate of bone loss can change with time are not well understood.
Determining the best predictors is a challenge in exploratory studies. There are many potential
explanatory variables, some of which may be correlated. Data-mining and multivariate exploratory
statistics attempt to resolve this complexity. They are increasingly used to gain insights from clinical
data. Ordination is an exploratory method which can visually depict the relationship of multiple
variables, by reducing data-complexity. In implant dentistry, this technique has been used to select
classifiers of peri-implantitis affected and resistant clusters (Papantonopoulos, Gogos, Housos, Bountis,
7
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
8
5
& Loos, 2016). As peri-implant bone levels are also affected by patient-level factors, when many
implants per patient are analysed such data is hierarchical. With appropriate analytics and mixed models,
one can model such complex data and select the most relevant predictors.
The aims of the current study were: i) to explore how multiple background features were related to rates
of peri-implant bone loss (RBLs) computed over different time-frames in reference to loading, ii) to
identify amongst these background features, the most robust predictors of these RBLs iii) to analyse the
impact of patient-level variation on these RBLs, and, iv) to analyse correlation, if any, between projected
marginal bone levels using different RBLs in reference to loading, and the actual marginal bone levels
measured at review.
Materials and Methods:
The study was approved by the Institutional Review Board of the University of Hong Kong/Hospital
Authority Hong Kong and the Ethics Committee of the Faculty of Dentistry, University of Hong Kong
(reference number: UW 15-609). Retrospective data were sourced from patients treated by postgraduate
students in the Implant Dentistry clinic, The University of Hong Kong (Ng, Fan, Leung, Fokas, &
Mattheos, 2018).
Subject Recruitment: In brief, patients who had received implant treatment from 2009 to 2014 in the
Centre of Advanced Dental Care, Prince Philip Dental Hospital, The University of Hong Kong had been
invited to participate in the study and attend a recall visit. Prior to screening and clinical examination,
patients had been informed about the research, and written informed consent was obtained. All
procedures were aligned with the ICH-GCP guidelines.
9
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
10
6
Inclusion and exclusion criteria: 

 !
"#
!
$%
! 
& Implants with diagnosis of peri-implantitis based on the current
case-definition, (Berghlundh, et al., 2018; Renvert, Persson, Pirih & Camargo; 2018) were excluded. In
brief, peri-implantitis was diagnosed where radiographic measurement of bone level of 3 mm and/or
probing depth 6 mm was found concomitant with bleeding on probing.
Study Procedures: Demographic, dental and medical history; history of diabetes mellitus, other
controlled systemic disease, medication status, smoking, parafunctional habits and past history of
periodontal disease and details relevant to implant treatment were sourced from patient records. Clinical
examination data included full-mouth periodontal examination including Probing Pocket Depth (PPD),
full mouth plaque score (FMPS), and bleeding on probing (BoP). One Periapical radiograph with
parallel technique had been taken for each of the implants.
Radiographic Measurement of marginal bone levels (MBL) and computation of rates of bone loss
(RBL): Periapical Radiographs from each implant at the 3 time points were scanned at a resolution of
400 dpi and digitized by Epson Perfection V700 Photo Dual lenses scanner. After digitalization,
Image J (Wayne Rasband National Institutes of Health, USA) was used for obtaining precise
11
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
12
7
measurements on the radiographs. Dimensions of the digitized radiographs were calibrated to account
for distortion by using known distance between implant threads and implant diameter. Two examiners
were calibrated with 6 random sets of radiographs with an intraclass correlation coefficients documented
as 0.99 and then conducted all measurements (Ng et al., 2018).
MBLs were assessed by defining 2 specific landmarks on the 3 radiographs of each patient : The implant
platform of tissue level (or shoulder of bone level implant) (PL) and the most coronal point of crestal
bone (CB) in contact with the implant. A line parallel to the axis of the implant was drawn between CB
and PL (CB-PL). Measures on mesial and distal aspects were obtained. To determine the RBLs in the
placement-loading, loading-review, and overall (placement-review) period, the change in corresponding
mesial CB-PL(mCB-PL) and distal CB-PL (dCB-PL) were computed, averaged to determine change
in MBL per implant (MBL), and divided by the duration of the corresponding time-frame in years, to
obtain the 3 RBL values (in mm/year) per implant: A) RBL placement-loading, B) RBL loading-review,
C) Overall RBL placement-review.
Statistical Analysis: All statistical analyses were performed in the R statistical environment
(https://www.r-project.org/) using R version 3.1.3. 17 background (implant-, site-, and subject- related)
variables were documented and categorised as described in Table S1 (supplementary file). These
included continuous (Age, Time at loading, Time since placement, Time since loading, PPD, Full mouth
plaque score), ordinal (BOP, Diabetes, Smoking, Periodontal Disease, Bruxism) and categorical
(Gender, Soft tissue Biotype, Implant abutment interface, Type of Implant surface, Placement Type,
Augmentation at implant site, Arch location; Antero-posterior location; Prosthesis Type; Retention Type,
Antagonist Type). The 3 RBLs and background variables were used for ordination analysis using the
‘Multiple Factor Analysis’ (MFA) function in the R package ‘FactoMineR (Husson, Josse, Le & Mazet,
13
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
14
8
2016). MFA is a dimensionality reduction method. It reduces complexity of multivariate data and allows
visual interpretation of major patterns. It is suited to data that contains both continuous and categorical
variables. MFA also allows grouping of variables where each group is normalized individually, in order
to balance their influence. An MFA correlation circle plot depicts the continuous variables and factors
plot depicts the categorical variables. These were drawn to visualize the inter-relationships of
background variables and RBLs. To determine the best predictors of each RBL, a ‘best subset’ approach
was applied to all possible predictors. The ‘regsubsets’ function in the ‘leaps package (Lumley, 2009)
was applied. A combination of ‘model adjusted R squared’ (indicating model-fit) and ‘Bayesian
Information Criteria’ (BIC) (based on both model complexity and fit) were used. The predictor set that
gave the best combination of highest adjusted R squared and lowest BIC values was selected. Linear-
mixed effects regression models (MEM) were made with these selected predictors for each RBL as
outcome (Pinheiro, Bates, DebRoy, Sarkar & R Core team, 2018). MEM is suitable for hierarchical or
nested data. The patient was incorporated as a grouping factor (as random intercept) considering there
were multiple implants in several patients. This accounted for among-patient variation in RBLs. Lack of
multicollinearity of MEM predictors was confirmed by variable inflation factors (VIF). VIF<2 was
indicated the predictors were not correlated. MEM parameters were bias- corrected by simulation
(bootstrapping, n=1000) (Loy and Steele, 2016). Simulation-based correction of bias and model
validation is particularly relevant to models with small sample sizes (Van der Leeden, Meijer, & Busing,
2008). The full MEMs were compared to null models with patient-level random intercept only models
(parametric bootstrapped test, n=1000) (Halekoh & Højsgaard, 2014).
Lastly, 2 projected MBLs (p-MBLs) (using RBL placement-loading and RBL loading-review each),
were determined by multiplying each RBL with the overall time in function and adding the obtained
values to MBL measured at placement. Correlations of the 2 p-MBL with the actual MBL at review were
assessed using Spearman’s correlation test.
15
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
16
9
Results:
A total of 243 Straumann implants from 119 patient records were screened and 154 implants from 86
patients were included. Table 1 presents the descriptive statistics for all the background parameters and
computed RBL values. The time from placement to loading ranged 1 month-2 years & 3 months
(median=5 months, IQR=4 months), time from loading to review ranged 1 year-6 years & 1 month
(median=3 years IQR=1 year & 1 month) and the total time from placement to review was 1 year & 7
months -6 years & 11 months (median=3 years & 10 months, IQR=10 months) (Figure 1, Table 1). Bone
loss averaged per implant (MBL) from placement to loading ranged from -1.56 to 3.16 mm
(median=0.53, IQR=0.85), from loading-review ranged from -0.77 to 1.77 mm (median=0.22,
IQR=0.46), and total bone loss from placement to review ranged from -1.08 to 3.98 mm (median=0.78,
IQR=1.09).
Figure 2 depicts the ordination of variables along the first two MFA components. Together these
components explained 18.92% of the total variance in the data, suggesting much variation remained
unexplained. However several inter-relationships among the measured variables were evident. RBL
values and their corresponding time-frames had high projections in opposing directions (Figure 2). Thus,
they were inversely related to each other. History of periodontitis was closely grouped with implant-
level PPD and risk factors of periodontitis, smoking and diabetes. BOP, age, and full-mouth plaque
scores were similarly related and aligned with RBLs suggesting these are positively related. The
qualitative variable levels projection showed posterior and mandibular location, transmucosal
placement, no augmentation at implant site, tissue-level implant-abutment interface, single crown type
17
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
18
10
prosthesis, cement retained type retention and SLA type surface as correlated. On the opposing aspect,
anterior and maxillary location, submerged placement, bone level, augmentation at implant site, bone-
level implant-abutment interface, SLActive type surface and cantilever type prosthesis were clustered
(Figure 2). From among these, ‘best subset’ based predictor selection gave 4 predictors for RBL
placement-loading (Time: placement to loading, Augmentation, Age and Anteroposterior location), 3
predictors for RBL loading-review (Time Loading-Review, Anteroposterior location, Placement type)
and 5 predictors for RBL placement-review (Time Placement-Loading, Time Placement-Review,
Augmentation, Age and Anteroposterior location) (Figure S1).
The MEM outcomes are summarized in Table 3. Overall similar patterns were seen for RBL placement-
loading and RBL placement-review models. . The length of each corresponding time-frame had a
significantly negative impact on all 3 RBL outcomes (Table 2). Per unit time, RBL declined the fastest
during placement-loading but this effect was not significant (bias corrected β= -0.63, p=0.13) (Figure 3).
It was the lowest for RBL-Loading-Review (bias corrected β= -0.05, p=0.04*). For RBL placement-
review, an opposite effect of the time from placement to loading was seen. Here, greater time from
placement to loading predicted greater RBL (bias corrected β=0.1, p=0.03*). Thus, earlier loading was
independently associated with lower overall RBL at review. Age and anterior location significantly
predicted lower RBLs placement-loading and placement-review. Scatter plots showed the relationship
between Age and RBL values showed a bimodal relationship peaking at 50-60 years, followed by a
decline (Figure S2). Similarly, Augmentation and anterior location significantly predicted lower RBL
placement-loading but did not reach significance for RBL placement-review (Table 2). The variation in
the patient-level intercept reflects the among-patient RBL variation. The highest value was noted for
RBL placement-loading (bias corrected sd=0.66). Model R-squared value is a measure of model-fit or its
explanatory value. It was highest for the RBL placement-review (R squared=0.48) and lowest for RBL
loading-review (R squared=0.08). Thus, the best explained was RBL placement-review and the least
19
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
20
11
explained was RBL loading-review. When the full MEM and patient-level random-intercept only models
were compared, significant differences were noted for all 3 models. All patient-level intercept-only
models had relatively lower R-squared values. However, for RBL placement-review, the patient-level
intercept-only model (R squared=0.40) also performed relatively well compared to the full-model (R
squared=0.48), showing inter-patient variation notably explained this RBL outcome.
Correlation tests showed p-MBL using RBL placement-loading was significantly and strongly correlated
to actual MBL at review (Spearman’s rho=0.79, 95% CI: 0.74-0.91, p<0.001). No significant correlation
was noted between p-MBL using RBL loading-review and actual MBL at review (Spearman’s rho=-
0.09, 95% CI: -0.24-0.07, p=0.27) (Figure 4).
Discussion:
The present study used the rate of bone loss as an outcome. This was computed over 3 different time-
frames. We found the actual bone level at review was poorly predicted using the post-loading rate of
bone loss but closely predicted using the pre-loading rate. Overall, these findings support the annual rate
of bone loss as an index implant success as suggested earlier (Galindo-Moreno, et al., 2015), as it can
have value in predicting the future bone loss. These findings also confirm that the annual rate of bone
loss varies with the time over which it is calculated (Gearaets, et al. 2014). Together, they strongly
support the need to standardize how annual rate of peri-implant bone loss is measured, reported, and
interpreted.
A primary finding was that the rate of bone loss was not a stable linear trait but de-accelerated with time.
Ordination showed RBLs and time-durations projected in opposing directions (Figure 2). The MEM
validated this inverse relationship in both pre-, post- loading and overall measurement periods, after
21
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
22
12
accounting for other predictors, time-frame, and inter-subject variation. These findings have important
implications for future research. Meta-analyses that summarise annual peri-implant bone loss reported in
cross-sectional studies should be pool it only when it is calculated over similar times since placement.
Alternatively, they must account for differences in total time in function, time-point of loading, and
other influencing factors. This message is also relevant to future retrospective studies. As implants
increase in both numbers and time in function, more retrospective assessments can be anticipated. These
should define the annual rate of bone loss as a function of the total time over which it was calculated. In
addition, if only the loading event is used as reference but its time-point is variable or ignored, simple
comparisons may be erroneous.
The second main finding is that inter-patient variability contributed to much variation in the early or pre-
loading rate of bone loss. Moreover, this pre-loading rate of bone loss was strongly predictive of later
bone levels and the models for RBL placement-loading and RBL placement-review were very similar.
Together these indicate an ‘early bone loser type’ who is more predisposed to worse marginal bone
levels (Galindo-Moreno, et. al, 2015). As there were multiple implants per subject in many cases, a
hierarchical model was used to predict RBL. The role of inter-patient variability was quantified and
accounted for. A number of factors could explain such individual variation. These include host-bone
tissue characteristics (Merheb, et. al, 2015) and myriad local or systemic influences.
The third key finding is that earlier loading predicted a lower rate of bone loss in the long-term.
Mechanical loading can modulate bone turnover so earlier loading may slow marginal bone loss (Bilhan,
Mumcu & Arat, 2010; Schincaglia, et al., 2016). A tendency to lower bone loss at early versus
conventionally loaded implants is reported (Helmy, Alqutaibi & Shawky, 2018). Biologically, loading
mechanically stimulates osteocytes leading to increased bone mass (Klein-Nulend, Bakker, Bacabac,
Vatsa, & Weinbaum, 2013). Notably, no immediate implants were included in the present analysis,
23
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
24
13
which precludes any conclusions regarding these. Others have found immediate loading did not show
any positive influence on bone levels (Elsyad, Al-Mahdy & Fouad, 2012; Elsyad, Elsaih & Khairallah,
2014).
In the placement-loading period, augmentation had a significant protective effect on RBL. As all
augmentations were grouped in this analysis, including different defect size and type, materials, GBR,
and sinus floor augmentation, it is difficult to draw clear conclusions. GBR has been associated with
lower marginal bone loss after loading (Jung, Herzog, Wolleb, Ramel, Thoma & Hämmerle, 2016). No
difference or less bone loss was found by some (Bazrafshan & Darby, 2018; Zumstein; Billström &
Sennerby, 2010). Others found augmented sites had greater bone loss in the pre-loading (Huang et al.,
2014) and long-term durations (Zitzmann, Schärer & Marinello, 2001). Sinus floor augmentation has
also shown greater marginal bone loss in the first year (Galindo-Moreno, Fernández-Jiménez, Avila-
Ortiz, Silvestre, Hernández-Cortés & Wang, 2013). The type of augmentation material and protocols can
differently influence bone loss (Benic, Bernasconi, Jung & Hämmerle, 2017; Schwarz, Schmucker &
Becker, 2016). Bone loss was lower in bone substitute grafted or untreated sites but increased in sites
treated with membranes (Zambon, Mardas, Horvath, Petrie, Dard & Donos, 2011). How augmentation
parameters impact bone turnover before and after loading needs greater study. Implant-abutment
connection type, abutment height, and patient factors also have interaction effects (Galindo-Moreno, et
al., 2013). Augmentation was also clustered with implant submergence, anterior and maxillary location,
bone level type implant in the ordination plot (Figure 2). These reflect that a need to augment was linked
to clinical choices of submergence and bone level interface. Among these, the factor which drives the
clinical decision may be the more important predictor in reality. Future studies should assess the driving
factors of clinical decisions. A tendency of submerged implants for lower RBL loading-review was
noted. This conflicted with past reports (Sanz et al., 2013; Paul, Petsch, & Held, 2017; ) but was similar
to Flores-Guillen et al. (FloresGuillen, ÁlvarezNovoa, Barbieri, Martín & Sanz, 2018). Bone turnover
25
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
26
14
is slower before abutment connection in the case of submerged implants (Hermann, Cochran,
Nummikoski & Buser, 1997). However, this is not consistent with the observation that an effect was
observable after loading. Older age weakly predicted slower rates of bone loss. Aging negatively affects
bone density and cancellous bone mass, increases cell apoptosis, but also reduces osteoblast activity
(Boskey & Coleman, 2010). Others found aging was associated with greater marginal bone loss at
implants which peaked at 50-60 years (Negri, et al., 2014). These contradictions may be explained by
the large spread of age in the present cohort. A bimodal relationship, with the rate of bone loss peaking
at the age of 50-60, but slow in more advanced age was seen (Supplementary File, Figure S2). A
limitation is that the 18-21 age group was not included, as we considered active jaw-growth may be a
factor (Björk, 1963).
'(
&&)#*+&)#! 
&, 
+&&)#& 
-".
"%&/
"%
&0

&
1&he ordination (Figure
2) depicted bleeding on probing and RBLs projected in similar directions. This indicated a positive
relationship existed.
27
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
28
15
Major limitations of this study include its exploratory nature, retrospective design, single time-point
assessment, modest sample size, short observation periods and a limited number of potential predictors
addressed. For example, implants of less than 8mm are common alternatives to augmentation and so
these may affect bone loss of in clinical populations (Tabrizi; Arabion; Aliabadi & Hasanzadeh, 2016;
Nielsen, Schou, Isidor, Christensen & Starch-Jensen, 2018). The current cohort did not include these.
All the implants analyzed were placed by trainees, thus were cases that were deemed suitable for
training. Operator experience and competency are variables that can impact surgical accuracy. The lack
of cases treated by experts in this population is another possible source of bias (Cushen & Turkyilmaz,
2013). Many implant-related variables such as abutment height and implant surface (Spinato,
Bernardello, Sassatelli & Zaffe, 2017; Blanco, et al., 2018) or soft-tissue thickness (Suárez-López del
Amo et al., 2016), which have been related to bone loss were not evaluated. While many known patient
and implant level factors were explored, only a small set of the most influential variables were evaluated
as explanatory variables. Considering the modest size of this cohort, the inclusion of large numbers of
predictors would weaken statistical robustness. For the same reason, interactions between predictors
were not examined. The model for RBL loading-review was notably weak. Another limitation is that we
determined the correlates of annual bone loss rates using a single time-point clinical observation.
Annually obtained clinical and radiographic assessments are needed to accurately understand how bone
loss changes over time and other factors.
The main strength of this study is the analytical strategy which showed an integrated view of multiple
variables’ relation to RBL. The insights from this exploratory study caution against simplistic
conclusions based on retrospectively determined annual bone loss rates. Prospective studies with very
well-characterized cohorts are essential to confirm the true risk factors of RBL. These findings are
indicative and best viewed as a basis for hypothesis development. They do, however, demonstrate the
29
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
30
16
value of clinical data mining in implant dentistry. Larger data repositories from clinical populations of
multi-center origin can be collected and similarly analyzed.
Conclusion: i) Background factors were explored in relation to annual rates of peri-implant bone loss
computed using 3 different measurement time-frames; pre-, post- loading, and overall time since
placement. ii) For each period, the average annual rate of bone loss declined with increasing time used
for calculation. Shorter loading time predicted a lower overall rate of bone loss. Anterior location and
older age predicted lower rate of bone loss in pre-loading and overall periods, while augmentation
predicted a lower pre-loading rate. iii) Between-patient variability was highest for the early/pre-loading
rate of bone loss and iv) marginal bone levels predicted using this pre-loading/early rate strongly
correlated to the actual marginal bone levels at review.
References
Albrektsson, T., Zarb, G., Worthington, P., & Eriksson, A. R. (1986). The long-term efficacy of
currently used dental implants: a review and proposed criteria of success. International Journal of
Oral Maxillofacial Implants, 1, 11-25.
Alrabiah, M., Al-Aali, K. A., Al-Sowygh, Z. H., Binmahfooz, A. M., Mokeem, S. A., & Abduljabbar,
T. (2018). Association of advanced glycation end products with peri-implant inflammation in
prediabetes and type 2 diabetes mellitus patients. Clinical Implant Dentistry and Related Research,
https://doi.org/10.1111/cid.12607
31
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
32
17
Atieh, M. A., Ibrahim, H. M., & Atieh, A. H. (2010). Platform Switching for Marginal Bone
Preservation Around Dental Implants: A Systematic Review and Meta-Analysis. Journal of
Periodontology, 81(10), 1350–1366. https://doi.org/10.1902/jop.2010.100232
Bazrafshan, N., & Darby, I. (2013). Retrospective success and survival rates of dental implants
placed with simultaneous bone augmentation in partially edentulous patients. Clinical Oral Implants
Research, 25, 768–773. https://doi.org/10.1111/clr.12185
Benic, G. I., Bernasconi, M., Jung, R. E., & Hämmerle, C. H. F. (2017). Clinical and radiographic
intra-subject comparison of implants placed with or without guided bone regeneration: 15-year
results. Journal of Clinical Periodontology, 44, 315–325. https://doi.org/10.1111/jcpe.12665
Berglundh, T., Persson, L., & Klinge, B. (2002). A systematic review of the incidence of biological
and technical complications in implant dentistry reported in prospective longitudinal studies of at
least 5 years. Journal of Clinical Periodontology, 29, 197–212. https://doi.org/10.1034/j.1600-
051x.29.s3.12.x
Berglundh, T., Armitage, G., Araujo, M.G., AvilaOrtiz, G., Blanco, J., Camargo P.M.,….Zitzman, N.
(2018). Peri-implant diseases and conditions: Consensus report of workgroup 4 of the 2017 World
Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions. Journal of
Periodontolology, 89, S313–S318. https://doi.org/10.1002/JPER.17-0739
Bilhan, H., Mumcu, E., & Arat, S. (2010). The Role of Timing of Loading on Later Marginal Bone
Loss Around Dental Implants: A Retrospective Clinical Study. Journal of Oral Implantology, 36,
363–376. https://doi.org/10.1563/aaid-joi-d-09-00078
Björk, A. (1963). Variations in the growth pattern of the human mandible: longitudinal radiographic
study by the implant method. Journal of Dental Research, 42, 400-411.
33
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
34
18
Blanco, J., Pico, A., Caneiro, L., Nóvoa, L., Batalla, P., & Martín-Lancharro, P. (2018). Effect of
abutment height on interproximal implant bone level in the early healing: A randomized clinical trial.
Clinical Oral Implants Research, 29, 108-117. http://doi.org/10.1111/clr.13108.
Boskey, A. L., & Coleman, R. (2010). Aging and Bone. Journal of Dental Research, 89, 1333–1348.
http://doi.org/10.1177/0022034510377791
Chow, L., Chow, T. W., Chai, J., & Mattheos, N. (2016). Bone stability around implants in elderly
patients with reduced bone mineral density - a prospective study on mandibular overdentures.
Clinical Oral Implants Research, 28, 966–973. https://doi.org/10.1111/clr.12907
Corcuera-Flores, J. R., Alonso-Domínguez, A. M., Serrera-Figallo, M. Á., Torres-Lagares, D.,
Castellanos-Cosano, L., & Machuca-Portillo, G. (2016). Relationship Between Osteoporosis and
Marginal Bone Loss in Osseointegrated Implants: A 2-Year Retrospective Study. Journal of
Periodontology, 87, 14–20. https://doi.org/10.1902/jop.2015.150229
Cushen, S. E., & Turkyilmaz, I. (2013). Impact of operator experience on the accuracy of implant
placement with stereolithographic surgical templates: an in vitro study. The Journal of Prosthetic
Dentistry, 109, 248-254. https://doi.org/10.1016/S0022-3913(13)60053-0.
Derks, J., & Tomasi, C. (2015). Peri-implant health and disease. A systematic review of current
epidemiology. Journal of Clinical Periodontology, 42, S158–S171.
https://doi.org/10.1111/jcpe.12334
Elsyad, M. A., Al-Mahdy, Y. F., & Fouad, M. M. (2012). Marginal bone loss adjacent to conventional
and immediate loaded two implants supporting a ball-retained mandibular overdenture: a 3-year
randomized clinical trial. Clinical Oral Implants Research, 23, 496–503.
https://doi.org/10.1111/j.1600-0501.2011.02173.x
35
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
36
19
Elsyad, M. A., Elsaih, E. A., & Khairallah, A. S. (2014). Marginal bone resorption around immediate
and delayed loaded implants supporting a locator-retained mandibular overdenture. A 1-year
randomised controlled trial. Journal of Oral Rehabilitation, 41, 608–618.
https://doi.org/10.1111/joor.12182
Flores-Guillen, J., Álvarez-Novoa, C., Barbieri, G., Martín, C., & Sanz, M. (2017). Five-year
outcomes of a randomized clinical trial comparing bone-level implants with either submerged or
transmucosal healing. Journal of Clinical Periodontology, 45, 125–
135.https://doi.org/10.1111/jcpe.12832
Galindo-Moreno, P., Fernández-Jiménez, A., O’Valle, F., Silvestre, F. J., Sánchez-Fernández, E.,
Monje, A., & Catena, A. (2013). Marginal Bone Loss in Implants Placed in Grafted Maxillary Sinus.
Clinical Implant Dentistry and Related Research, 17, 373–383. https://doi.org/10.1111/cid.12092
Galindo-Moreno, P., León-Cano, A., Ortega-Oller, I., Monje, A., O′Valle, F., & Catena, A. (2015).
Marginal bone loss as success criterion in implant dentistry: beyond 2 mm. Clinical Oral Implants
Research, 26, e28–e34. https://doi.org/10.1111/clr.12324
Galindo-Moreno P., Fernández-Jiménez A., Avila-Ortiz, G., Silvestre, F.J, Hernández-Cortés, P.,
Wang, H.L. (2013). Marginal bone loss around implants placed in maxillary native bone or grafted
sinuses: a retrospective cohort study. Clinical Oral Implant Research 00:2013;000–000. doi:
10.1111/clr.12122
Geraets, W., Zhang, L., Liu, Y., & Wismeijer, D. (2014). Annual bone loss and success rates of dental
implants based on radiographic measurements. Dentomaxillofacial Radiology, 43, 20140007. doi:
20140007.10.1259/dmfr.20140007]
37
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
38
20
Halekoh, U. & Højsgaard, S. (2014). A Kenward-Roger Approximation and Parametric Bootstrap
Methods for Tests in Linear Mixed Models - The R Package pbkrtest. Journal of Statistical Software,
59, 9, 1-30. http://www.jstatsoft.org/v59/i09/.
Hasegawa, M., Hotta, Y., Hoshino, T., Ito, K., Komatsu, S., & Saito, T. (2015). Long-term
radiographic evaluation of risk factors related to implant treatment: suggestion for alternative
statistical analysis of marginal bone loss. Clinical Oral Implants Research, 27, 1283–1289.
https://doi.org/10.1111/clr.12734
Helmy, M. H. E.-D., Alqutaibi, A. Y., El-Ella, A. A., & Shawky, A. F. (2018). Effect of implant
loading protocols on failure and marginal bone loss with unsplinted two-implant-supported
mandibular overdentures: systematic review and meta-analysis. International Journal of Oral and
Maxillofacial Surgery, 47, 642–650. https://doi.org/10.1016/j.ijom.2017.10.018
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 the Canine Mandible. Journal of Periodontology, 68, 1117–1130.
https://doi.org/10.1902/jop.1997.68.11.1117
Huang, H., Ogata, Y., Hanley, J., Finkelman, M., & Hur, Y. (2014). Crestal bone resorption in
augmented bone using mineralized freeze-dried bone allograft or pristine bone during submerged
implant healing: a prospective study in humans. Clinical Oral Implants Research, 27, e25–e30.
https://doi.org/10.1111/clr.12512
Husson, F., Josse, J., Le, S. & Mazet, J. (2016). FactoMineR: Multivariate Exploratory Data
Analysis and Data Mining. R package version 1.32. http://CRAN.R-
project.org/package=FactoMineR
39
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
40
21
Jung, R. E., Herzog, M., Wolleb, K., Ramel, C. F., Thoma, D. S., & Hämmerle, C. H. F. (2016). A
randomized controlled clinical trial comparing small buccal dehiscence defects around dental
implants treated with guided bone regeneration or left for spontaneous healing. Clinical Oral
Implants Research, 28, 348–354. https://doi.org/10.1111/clr.12806
Klein-Nulend, J., Bakker, A. D., Bacabac, R. G., Vatsa, A., & Weinbaum, S. (2013).
Mechanosensation and transduction in osteocytes. Bone, 54, 182-190.
https://doi.org/10.1016/j.bone.2012.10.013
Koldsland, O. C., Scheie, A. A. & Aass, A. M. (2010) Prevalence of peri-implantitis related to
severity of the disease with different degrees of bone loss. Journal of Periodontology 81, 231–238.
https://doi.org/10.1902/jop.2009.090269
Kozlovsky, A., Tal, H., Laufer, B.-Z., Leshem, R., Rohrer, M. D., Weinreb, M., & Artzi, Z. (2007).
Impact of implant overloading on the peri-implant bone in inflamed and non-inflamed peri-implant
mucosa. Clinical Oral Implants Research, 18, 601–610. https://doi.org/10.1111/j.1600-
0501.2007.01374.x
Levin, L., Hertzberg, R., Har-Nes, S., & Schwartz-Arad, D. (2008). Long-Term Marginal Bone Loss
Around Single Dental Implants Affected by Current and Past Smoking Habits. Implant Dentistry, 17,
422–429. https://doi.org/10.1097/id.0b013e31818c4a24
Linkevicius, T., Puisys, A., Linkeviciene, L., Peciuliene, V., & Schlee, M. (2013). Crestal Bone
Stability around Implants with Horizontally Matching Connection after Soft Tissue Thickening: A
Prospective Clinical Trial. Clinical Implant Dentistry and Related Research, 17, 497–508.
https://doi.org/10.1111/cid.12155
Loy, A. & Steele S. (2016). lmeresampler: Bootstrap Methods for Nested Linear Mixed-Effects
Models. R package version 0.1.0.9000. https://github.com/aloy/lmeresampler
41
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
42
22
Lumley, T. (2009) Package ‘LEAPS’: Regression subset selection. R package version 2.
http://CRAN.R-project.org/package=leaps
Merheb, J., Graham, J., Coucke, W., Roberts, M., Quirynen, M., Jacobs, R., & Devlin, H. (2015).
Prediction of implant loss and marginal bone loss by analysis of dental panoramic radiographs.
International Journal of Oral & Maxillofacial Implants, 30, 372-327.
https://doi.org/10.11607/jomi.3604.
Misch, C.E., Perel, M.L., Wang, H.L., Sammartino, G., Galindo-Moreno, P., Trisi, P., Steigmann, M.,
Rebaudi, A., Palti, A., Pikos, M.A., Schwartz-Arad, D., Choukroun, J., Gutierrez-Perez, J.L.,
Marenzi, G. & Valavanis, D.K. (2008) Implant success, survival, and failure: the international
congress of oral implantologists (icoi) pisa consensus conference. Implant Dentistry, 17, 5–15.
Naert, I., Koutsikakis, G., Duyck, J., Quirynen, M., Van Steenberghe, D & Jacobs, R. (2002).
Biologic outcome of implant‐supported restorations in the treatment of partial edentulism. Clinical
Oral Implants Research, 13, 381-389.
Negri, M., Galli, C., Smerieri, A., Macaluso, G. M., Manfredi, E., Ghiacci, G., … Lumetti, S. (2014).
The Effect of Age, Gender, and Insertion Site on Marginal Bone Loss around Endosseous Implants:
Results from a 3-Year Trial with Premium Implant System. BioMed Research International, 2014,
369051. http://doi.org/10.1155/2014/369051
Ng K.T., Fan M.H.M., Leung M.C., Fokas G. & Mattheos N. (2018). Peri-implant inflammation and
marginal bone level changes around dental implants in relation to proximity with and bone level of
adjacent teeth. Australian Dental Journal, in press
Nielsen, H. B., Schou, S., Isidor, F., Christensen, A. E., & Starch-Jensen, T. (2018). Short implants
(≤ 8 mm) compared to standard length implants (> 8 mm) in conjunction with maxillary sinus floor
43
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
44
23
augmentation: a systematic review and meta-analysis. International Journal of Oral and
Maxillofacial Surgery, in-press, http://doi.org/10.1016/j.ijom.2018.05.010
Nóvoa, L., Batalla, P., Caneiro, L., Pico, A., Liñares, A., & Blanco, J. (2017). Influence of Abutment
Height on Maintenance of Peri-implant Crestal Bone at Bone-Level Implants: A 3-Year Follow-up
Study. The International Journal of Periodontics & Restorative Dentistry, 37, 721–727.
https://doi.org/10.11607/prd.2762
Papantonopoulos, G., Gogos, C., Housos, E., Bountis, T., & Loos, B. G. (2016). Prediction of
individual implant bone levels and the existence of implant “phenotypes.” Clinical Oral Implants
Research, 28, 823–832. https://doi.org/10.1111/clr.12887
Paul, S., Petsch, M., & Held, U. (2017). Modeling of Crestal Bone After Submerged vs
Transmucosal Implant Placement: A Systematic Review with Meta-Analysis. The International
Journal of Oral & Maxillofacial Implants, 32, 1039–1050. https://doi.org/10.11607/jomi.5505
Peñarrocha, M., Palomar, M., Sanchis, J. M., Guarinos, J., & Balaguer, J. (2004). Radiologic study
of marginal bone loss around 108 dental implants and its relationship to smoking, implant location,
and morphology. The International Journal of Oral & Maxillofacial Implants, 19, 861.
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. & R Core Team (2018). nlme: Linear and Nonlinear
Mixed Effects Models. R package version 3.1-137, https://CRAN.R-project.org/package=nlme.
Renvert S, Persson GR, Pirih FQ, Camargo PM (2018). Peri‐implant health, peri‐implant mucositis,
and peri-implantitis: Case definitions and diagnostic considerations. Journal of Periodontology, 89,
(Suppl 1):S304–S312. https://doi.org/10.1002/JPER.17-0588
Sanz, M., Ivanoff, C.-J., Weingart, D., Wiltfang, J., Gahlert, M., Cordaro, L., … Hammerle, C.
(2013). Clinical and Radiologic Outcomes after Submerged and Transmucosal Implant Placement
45
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
46
24
with Two-Piece Implants in the Anterior Maxilla and Mandible: 3-Year Results of a Randomized
Controlled Clinical Trial. Clinical Implant Dentistry and Related Research, 17, 234–246.
https://doi.org/10.1111/cid.12107
Schincaglia, G., Rubin, S., Thacker, S., Dhingra, A., Trombelli, L., & Ioannidou, E. (2016). Marginal
Bone Response Around Immediate- and Delayed-Loading Implants Supporting a Locator-Retained
Mandibular Overdenture: A Randomized Controlled Study. The International Journal of Oral &
Maxillofacial Implants, 448–458. https://doi.org/10.11607/jomi.4118
Schwarz, F., Schmucker, A., & Becker, J. (2016). Long-term outcomes of simultaneous guided bone
regeneration using native and cross-linked collagen membranes after 8 years. Clinical Oral Implants
Research, 28, 779–784. https://doi.org/10.1111/clr.12881
Şener-Yamaner, I. D., Yamaner, G., Sertgöz, A., Çanakçi, C. F., & Özcan, M. (2017). Marginal Bone
Loss Around Early-Loaded SLA and SLActive Implants. Implant Dentistry, 26, 592–599.
https://doi.org/10.1097/id.0000000000000625
Spinato S, Bernardello F, Sassatelli P, Zaffe D. (2017) Hybrid and fully‐etched surface implants in
periodontally healthy patients: A comparative retrospective study on marginal bone loss. Clin
Implant Dentistry and Related Research. 19, 663–670. https://doi.org/10.1111/cid.12504
Suárez-López del Amo, F. , Lin, G. , Monje, A. , Galindo‐Moreno, P. and Wang, H. (2016), Influence
of Soft Tissue Thickness on Peri‐Implant Marginal Bone Loss: A Systematic Review and Meta-
Analysis. Journal of Periodontology, 87: 690-699. https://doi.org /10.1902/jop.2016.150571
Tabrizi, R., Arabion, H., Aliabadi, E., & Hasanzadeh, F. (2016). Does increasing the number of short
implants reduce marginal bone loss in the posterior mandible? A prospective study. British Journal
of Oral and Maxillofacial Surgery, 54, 731-735. https://doi.org /10.1016/j.bjoms.2016.04.010
47
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
48
25
Valles, C., Rodríguez-Ciurana, X., Clementini, M., Baglivo, M., Paniagua, B., & Nart, J. (2018).
Influence of subcrestal implant placement compared with equicrestal position on the peri-implant
hard and soft tissues around platform-switched implants: a systematic review and meta-analysis.
Clinical Oral Investigations, 22, 555–570. https://doi.org/10.1007/s00784-017-2301-1
Van der Leeden, R., Meijer, E., & Busing, F. M. (2008). Resampling multilevel models. In
Handbook of multilevel analysis (pp. 401-433). Springer, New York, NY.
https://doi.org/10.1007/978-0-387-73186-5_11
van Eekeren, P., Tahmaseb, A., & Wismeijer, D. (2015). Crestal bone changes in macrogeometrically
similar implants with the implant-abutment connection at the crestal bone level or 2.5 mm above: a
prospective randomized clinical trial. Clinical Oral Implants Research, 27, 1479–1484.
https://doi.org/10.1111/clr.12581
Zambon, R., Mardas, N., Horvath, A., Petrie, A., Dard, M., & Donos, N. (2011). The effect of
loading in regenerated bone in dehiscence defects following a combined approach of bone grafting
and GBR. Clinical Oral Implants Research, 23, 591–601. https://doi.org/10.1111/j.1600-
0501.2011.02279.x
Zechner, W., Trinkl, N., Watzak, G., Busenlechner, D., Tepper, G., Haas, R., & Watzek, G. (2004).
Radiologic follow-up of peri-implant bone loss around machine-surfaced and rough-surfaced
interforaminal implants in the mandible functionally loaded for 3 to 7 years. International Journal of
Oral & Maxillofacial Implants, 19, 216-221.
Zhou, Y., Gao, J., Luo, L., & Wang, Y. (2015). Does Bruxism Contribute to Dental Implant Failure?
A Systematic Review and Meta-Analysis. Clinical Implant Dentistry and Related Research, 18, 410–
420. https://doi.org/10.1111/cid.12300
49
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
50
26
Zitzmann, N. U., Schärer, P., & Marinello, C. P. (2001). Long-term results of implants treated with
guided bone regeneration: a 5-year prospective study. International Journal of Oral & Maxillofacial
Implants, 16, 355-366.
Zumstein, T., Billström, C., & Sennerby, L. (2010). A 4- to 5-Year Retrospective Clinical and
Radiographic Study of Neoss Implants Placed with or without GBR Procedures. Clinical Implant
Dentistry and Related Research, 14, 480–490. https://doi.org/10.1111/j.1708-8208.2010.00286.x
51
589
590
591
592
593
594
595
596
597
598
599
52
27
0
3
6
9
1 2
1 5
1 8
2 1
2 4
2 7
3 0
0 3 6 9 1 2 1 5 1 8 2 1 2 4 2 7
T i m e f r o m P l a c e m e n t t o L o a d i n g ( m o n t h s )
c o u n t
0
1
2
3
4
5
6
7
8
9
1 2 3 4 5 6
T i m e f r o m L o a d i n g t o R e v i e w ( y e a r s )
c o u n t
1. Figure 1: Frequency distribution of the 156 implants according to a) Time from Placement-Loading (months) and b)
Time from Loading-Review (years).
53
600
601
602
54
28
- 2 . 0 - 1 . 5 - 1 . 0 - 0 . 5 0 . 0 0 . 5 1 . 0
- 1 . 5 - 1 . 0 - 0 . 5 0 . 0 0 . 5 1 . 0 1 . 5
I n d i v i d u a l f a c t o r m a p
D i m 1 ( 1 2 . 2 8 % )
D i m 2 ( 1 1 . 4 7 % )
A u g m e n t a t i o n
N o . A u g m e n t a t i o n
B L
T L
S L A
S L A c t i v e
S u b m e r g e d
T r a n s m u c o s a l
T h i c k . o r . M e d i u m
T h i n
F e m a l e
M a l e
- 1 . 0 - 0 . 5 0 . 0 0 . 5 1 . 0
- 1 . 0 - 0 . 5 0 . 0 0 . 5 1 . 0
C o r r e l a t i o n c i r c l e
D i m 1 ( 1 2 . 2 8 % )
D i m 2 ( 1 1 . 4 7 % )
R B L
T i m e
P e r i o d o n t a l . s t a t u s . a n d . R i s k . f a c t o r s
B r u x i s m
R B L . P l a c e m e n t . t o . L o a d i n g
R B L . P l a c e m e n t . t o . R e v i e w
R B L . L o a d i n g . t o . R e v ie w
T i m e . P l a c e m e n t . t o . L o a d i n g
T i m e . P l a c e m e n t . t o . R e v i e w
T i m e . L o a d i n g . t o . R e v i e w
A g e
P P D B O P
F u l l . M o u t h . P l a q u e . S c o r e
D i a b e t e s
S m o k i n g
H i s t o r y . P e r i o d o n t i t i s
B r u x i s m
Figure 2: Plots of variables in multiple factor analysis (MFA). Variables that are closely clustered are positively correlated
and those that project in opposing directions are negatively correlated. The individual factors map shows the inter-
relationships of categorical variable levels. The correlation circle plot plot shows the inter-relationships of the continuous
variables, where the length of the vector corresponds to the importance of the variable in the MFA projection. Variables close
to the center of the plots have low weightage in the projection.
55
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
56
29
Time From Placement-Review(yr)
Time From Loading-Review(yr)
Time From Placement-Loading(yr)
-5
0
5
0 2 4 6 0 2 4 6 0 2 4 6
Time (years)
Rate of bone loss (mm/year)
Figure 3: RBLs as a function of time in years within: a) the total time frame of Placement-Review, b) Loading-Review, c)
Placement-Loading. In each time-frame RBL declined with increase in time. The decline was greatest in the Placement-
Loading time-frame.
57
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
58
30
-10
0
10
20
30
0 2 4 6
Actual MBL at Review (mm)
Projected MBL (p-MBL) (mm)
Type p-MBL Loading to Review p-MBL Placement to Loading
Figure 4: Correlation of Actual MBL at review with projected-MBL (p-MBL) using i) RBL Placement-Loading (p-MBL
Placement to Loading) and ii) RBL Loading-Review (p-MBL Loading to Review). Actual MBL strongly correlated with p-
MBL Placement to Loading (r=0.88, p<0.001) but did not significantly correlate with p-MBL Loading to Review.
59
644
645
646
647
648
649
650
651
652
653
654
655
656
60
31
T r a n s m u c o s a l S u b m e r g e d
- 4 - 2 0 2 4 6 8
R B L P l a c e m e n t - L o a d i n g
T r a n s m u c o s a l S u b m e r g e d
- 1 . 5 - 0 . 5 0 . 5 1 . 5
R B L L o a d i n g - R e v i e w
T L B L
- 1 . 5 - 0 . 5 0 . 5 1 . 5
R B L L o a d i n g - R e v i e w
Figure 5: RBL Placement-Loading and RBL Loading-Review categorized by Augmentation at implant site, Placement Type
(Transmucosal/Submerged) and Implant abutment interface {Tissue level (TL)/ Bone level(BL)}. Greater variation was
evident in RBL Placement-Loading as compared to RBL Loading-Review, especially in Transmucosal placement and TL
interface.
61
657
658
659
660
661
662
663
62
32
Table 1: Descriptive statistics for RBL values and background variables
Variable Descriptive Statistics
Age (years)* 57 (15), 26-79
Gender Male=55, Female=101
Time at loading (years since placement)* 0.42 (0.34), 0.08-2.25
Time since placement (years)* 3.75 (0.82), 1.58-6.75
Time since loading (years)* 3.00 (1.01), 1.00-6.08
PPD (average of mesial and distal
measures/implant) (mm)*
3.00 (1.13), 1.00-5.00
BOP (average of mesial and distal scores/implant)* 0.00 (0.50), 0.00-1.00
Full mouth plaque score (%)* 34.6 (32.8), 6.00-37.5
Diabetes# 0=146,, 1=10
Smoking# 0= 146, 1=10
History of Periodontitis# 0=71, 1=85
Bruxism# 0=127, 1=29
Implant abutment interface# Bone Level (BL)= 31, Tissue Level
(TL)=125
Implant Surface# SLA=152, SLActive=4
Placement Type# Submerged= 13, Transmucosal= 143
Augmentation at implant site# Augmentation=82, No
Augmentation=74
Soft tissue Biotype# Thick/Medium =128, Thin=28
* Median (Inter-quartile range), Range; # Counts for variable response levels
63
664
665
666
667
668
669
670
671
64
1
Table 2: Bootstrapped Linear Mixed-Effects Regression Models for RBL outcomes
Predictors
RBL Placement- Loading1RBL Loading-Review2RBL Placement-Review3
Fixed Effects
(Bias corrected, Bootstrap, n=1000)
Estimate (95% ci) p value Estimate (95% ci) p value Estimate (95% ci) p value
(Intercept) 3.62 (2.79,5.55) <0.001 0.26 (0.04,0.42) 0.02 0.65 (0.40,1.02)) <0.001
Time Placement-Loading -0.63 (-1.37,0.19) 0.13 --- --- 0.10 (0.01,0.30) 0.03
Time Loading-Review --- --- -0.05 (-0.08, -0.01) 0.04 --- ---
Time Placement-Review --- --- --- --- -0.07 (-0.12, -0.03) <0.001
Age -0.03 (-0.06,-0.02) <0.001 --- --- -0.003 (-0.008, -0.0005) 0.02
Anteroposterior Location
(Reference:Posterior)
-0.78 (-0.50,-1.75) <0.001 0.10 (-0.30,0.00) 0.18 -0.06 (-0.02,-0.21) 0.02
Augmentation (Reference level: No) -0.57 (-0.03,-1.22) 0.04 --- --- -0.006(-0.1,0.09) 0.86
Placement Type (Reference level:
Transmucosal)
--- --- -0.10 (-0.03,0.34) 0.21 --- --
Random Effects
(Bias corrected, Bootstrap n=1000)
sd (95% ci) sd (95% ci) sd (95% ci)
Patient (Intercept) 0.66 (0.0,1.07) 0.02 (0.0,374.8) 0.21 (0.11,0.22)
Residual 1.64 (1.36,1.84) 0.28 (0.23,0.32) 0.21 (0.17,0.24)
Model Comparison*
(Bootstrap, n=1000)
MEM vs Random effect
only model
MEM vs Random effect
only model
MEM vs Random effect
only model
Model AIC/R squared4
622.8/23.0 vs 626.8/14.4
LRT, p-value
17.6, <0.001
Model AIC/R squared4
63.9/0.08 vs 60.0/0.04
LRT, p-value
10.8, 0.01
Model AIC/R squared4
59.2/0.48 vs 38.5/0.40
LRT, p-value
19.8, 0.003
Significant p-values (<0.05) in bold font. *Model comparisons: likelihood ratio and parametric bootstrap tests, MEM against Model with random patient level intercept only.
Variable inflation factors:
1. Model RBL-placement- loading: Time placement to loading =1.07, Age=1.03, Antero-Posterior location=1.07, Augmentation=1.02
2. Model RBL-loading-review=Time.loading.to.review=1.01, Antero-Posterior location=1.34, Placement Type=1.35
3. Model RBL-placement-review=Time placement.to.loading=1.07, Time placement to review=1.06, Age=1.04, Antero-Posterior location=1.07, Augmentation=1.08
65
672
673
674
675
676
677
678
66
1
Supplementary Information:
1. Multiple Factor Analysis:
The R package ‘FactoMineR’ (https://cran.r-project.org/web/packages/FactoMineR/index.html) was
used for Multiple Factor Analysis based ordination using the function ‘MFA. In the current analysis,
variable grouping was as follows: i) "RBL" including the 3 RBL values, ii) "Time" including the 3 time-
frames, iii) "Periodontal history and Risk factors" including Age and the ordinal variables BOP,
Diabetes, Smoking, Periodontal Disease, iv) "Implant type and placement" including the categorical
variables Implant abutment interface, Placement Type, and v-viii) "Bruxism", "Augmentation",
"Gender" and "Biotype", “Arch location”, “Antero-posterior location”, “Prosthesis Type”, “Retention
Type”, “Antagonist Type” as individual groups.
2. Predictor selection using ‘best subsets’:
The R package ‘leaps’ (https://cran.r-project.org/web/packages/leaps/index.html) was used for best
subset based predictor selection using the function ‘regsubsets’ to generate one best subset of predictors
for each possible number of predictors. Final selection was based on both Adjusted R squared and
Bayesian Information Criterion (BIC) values to balance for modelfit, complexity and likelihood of a true
model. Predictor sets that gave the best possible combination of highest Adjusted R squared with lowest
BIC values were selected.
3. Mixed Effects Modelling (MEM):
MEM were made using ‘nlme’ and ‘lme4’ packages in R and bootstrapping was done using
‘lmeresampler’ (with n=1000) and ‘boot’. A conditional R sqaured statistic as an indicator of the MEM
model-fit was computed using the function ‘r.squaredGLMM’ in the R package ‘MuMIn’. A ‘cases’ type
bootstrapping resampled for both patients and implants. Bias was subtracted from the parameter estimate
67
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
68
2
to determine bias corrected estimates. Bootstrapped confidence intervals were determined and bootstrap
plots were visually inspected. Random-effect (Patient-Level) only model was compared with the full
MEM with Likelihood Ratio Tests and p-values were generated using parametric bootstrap test with
1000 simulations (R package: ‘pbkrtest’).
69
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
70
3
Table S1: Background variables used for ordination and predictor selection
Variable Type of variable and levels
Age (years) Quantitative, Continuous
Gender Qualitative, Categorical, Levels: Male, Female
Time at loading (years since
placement)
Quantitative, Continuous
Time since placement (years) Quantitative, Continuous
Time since loading (years) Quantitative, Continuous
PPD (average of mesial and distal
measures/implant) (mm)
Quantitative, Continuous
BOP (average of mesial and distal
scores/implant)
Quantitative, Continuous
Full mouth plaque score (%) Quantitative, Continuous
Diabetes Quantitative, Ordinal, 0=No history,
1=positive history
Smoking Quantitative, Ordinal, 0=No history,
1=positive history
History of Periodontitis Quantitative, Ordinal, 0=No history,
1=positive history
Bruxism Quantitative, Ordinal, 0=No history,
1=positive history
Implant abutment interface Qualitative, Categorical, Levels: Bone Level
(BL), Tissue Level (TL)
Implant Surface Qualitative, Categorical, Levels: SLA, SLActive
Placement Type Qualitative, Categorical, Levels:
Transmucosal, Submerged
Augmentation at implant site Qualitative, Categorical, Levels: No
Augmentation, Augmentation (GBR/Sinus
graft)
Soft tissue Biotype Qualitative, Categorical, Levels:
Thick/Medium, Thin
Anteroposterior Location Qualitative, Categorical, Levels:
Anterior/Posterior
71
729
730
72
4
Arch Location Qualitative, Categorical, Levels:
Maxillary/Mandibular
Prosthesis Type Qualitative, Categorical, Levels:
Cantilever/Fixed Dental Prosthesis/Single
Crown
Retention Type Qualitative, Categorical, Levels: Cement
Retained /Screw Retained
Antagonist Type Qualitative, Categorical, Levels: Antagonist
Tooth, Antagonist tooth supported (dental)
Fixed Dental Prosthesis, Antagonist Implant
supported Fixed Dental Prosthesis,
73
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
74
5
Figure S1: Model Adjusted r squared and BIC values for each best possible combinations of n (max=9) predictors for the 3
RBL outcomes (included predictors: coloured squares, greyscale: parameter strength)
75
749
750
751
752
753
76
6
Figure S2: Scatter plots showing relationship of Age and RBL values. A ‘loess’ regression fit line depicts a bimodal
relationship with RBL values peaking at 50-60 years, followed by decline
77
754
755
756
78
... Not only does the need to define the etiology of MBL remain, but also to distinguish between physiological and pathological peri-implant bone losses. Recent studies [6,7], highlighting that MBL rates rather than raw MBL data might improve the ability of clinicians to predict a change in health status of peri-implant tissues, offered some important clues on this issue. However, to date, it still seems impossible to generalize a parameter or a measurement that allows one to define a physiological or pathological MBL without an evaluation of the specific characteristics of each single implant/patient, with each playing an important role in the prognosis [6,7]. ...
... Recent studies [6,7], highlighting that MBL rates rather than raw MBL data might improve the ability of clinicians to predict a change in health status of peri-implant tissues, offered some important clues on this issue. However, to date, it still seems impossible to generalize a parameter or a measurement that allows one to define a physiological or pathological MBL without an evaluation of the specific characteristics of each single implant/patient, with each playing an important role in the prognosis [6,7]. ...
... Although these bone loss thresholds provide easy clinical 'cut-offs', they do not predict future MBL. Since marginal bone remodeling is a dynamic process, the rate of MBL has been proposed as a better index of implant success than bone loss or bone level values [6,7]. The rate of MBL has been correlated with the time-duration (pre-, post-loading, and overall periods) [7]. ...
Article
Full-text available
Objectives: The aim of this retrospective study was to analyze peri-implant marginal bone loss levels/rates and peri-implant sulcular fluid levels/rates of metalloproteinase-8 in three timeframes (6 months post-surgery-restoration delivery (T0)-and 6 (T6) and 24 (T24)-months post-loading) and to evaluate if there is a correlation between peri-implant sulcular fluid levels of metalloproteinase-8 and peri-implant marginal bone loss progression. Materials and methods: Two cohorts of patients undergoing implant surgery between January 2017 and January 2019 were selected in this retrospective study. A total of 39 patients received 39 implants with a laser-microtextured collar surface, and 41 subjects received 41 implants with a machined/smooth surface. For each patient, periapical radiographs and a software package were used to measure marginal bone loss rates. Implant fluid samples were analyzed by an enzyme-linked immunosorbent assay (ELISA) test. The modified plaque index, probing depth, and bleeding on probing were also recorded. Results: High marginal bone rates at T24 were strongly associated with elevated rates between T0 and T6. The levels of metalloproteinase-8 were significantly more elevated around implants with marginal bone loss, in relation to implants without marginal bone loss. Marginal bone loss (MBL) rates at 24 months were associated with initial bone loss rates and initial levels of metalloproteinase-8. Conclusions: Peri-implant marginal bone loss progression is statistically correlated to peri-implant sulcular fluid levels of metalloproteinase-8. Moreover, the initial high levels of marginal bone loss and metalloproteinase-8 can be considered as indicators of the subsequent progression of peri-implant MBL: implants with increased marginal bone loss rates and metalloproteinase-8 levels at 6 months after loading are likely to achieve additional marginal bone loss values.
... Accordingly, the rate of MBL as a new index was proposed by Galindo-Moreno et al. recently [69]. As remodelling of marginal bone is a dynamic process, the rate of MBL is calculated in millimetre/month (mm/m) and could change over time [69,71]. Galindo-Moreno et al. [69] found that the progression of MBL tends to be higher and the risk of implant failure could be significantly increased when the rates of MBL was higher than 0.44 mm at six months postloading. ...
Article
Full-text available
Background Immediate loading has recently been introduced into unsplinted mandibular implant-retained overdentures for the management of edentulous patients due to their increasing demand on immediate aesthetics and function. However, there is still a scarcity of meta-analytical evidence on the efficacy of immediate loading compared to delayed loading in unsplinted mandibular implant-retained overdentures. The purpose of this study was to compare the marginal bone loss (MBL) around implants between immediate and delayed loading of unsplinted mandibular implant-retained overdentures. Methods Randomized controlled trials (RCTs), controlled clinical trials (CCTs), and cohort studies quantitatively comparing the MBL around implants between immediate loading protocol (ILP) and delayed loading protocol (DLP) of unsplinted mandibular overdentures were included. A systematic search was carried out in PubMed, EMBASE, and CENTRAL databases on December 02, 2020. “Grey” literature was also searched. A meta-analysis was conducted to compare the pooled MBL of two different loading protocols of unsplinted mandibular overdentures through weighted mean differences (WMDs) with 95% confidence intervals (95% CIs). The subgroup analysis was performed between different attachment types (i.e. Locator attachment vs. ball anchor). The risk of bias within and across studies were assessed using the Cochrane Collaboration’s tool, the Newcastle–Ottawa scale, and Egger’s test. Results Of 328 records, five RCTs and two cohort studies were included and evaluated, which totally contained 191 participants with 400 implants. The MBL of ILP group showed no significant difference with that of DLP group (WMD 0.04, CI − 0.13 to 0.21, P > .05). The subgroup analysis revealed similar results with Locator attachments or ball anchors ( P > .05). Apart from one RCT (20%) with a high risk of bias, four RCTs (80%) showed a moderate risk of bias. Two prospective cohort studies were proved with acceptable quality. Seven included studies have reported 5.03% implant failure rate (10 of 199 implants) in ILP group and 1.00% failure rate (2 of 201 implants) in DLP group in total. Conclusions For unsplinted mandibular implant-retained overdentures, the MBL around implants after ILP seems comparable to that of implants after DLP. Immediate loading may be a promising alternative to delayed loading for the management of unsplinted mandibular implant-retained overdentures. PROSPERO registration number : CRD42020159124.
... ent study, the consensus case definition used for peri-implantitis diagnosis was BoP and/or ≥ 6 mm probing depth together with bone loss ≥ 3 mm. If one of these findings was not evident, the implant was included in the study group and only implants with a peri-implantitis diagnosis were excluded, similar to very recent studies.34 Changes in condition of the peri-implant tissues were analyzed in terms of PPD and BoP values for different implant-abutment complexes in this study.Basically, it was noted that the type had influenced the BoP values whereas PPD values were not affected(Table 3). ...
Article
Full-text available
Purpose: The design of the implant-abutment complex is thought to be responsible for marginal bone loss (MBL) and might affect the condition of the peri-implant tissues. This the present study aimed to evaluate the influence of the implant-abutment complex on MBL and the peri-implant tissues in partially edentulous patients treated with dental implants and determine the most advantageous design. Materials and methods: A total of ninety-one endosseous implants with different designs of implant-abutment complex [tissue level-TL (n = 30), platform switch-PS (n = 18), and platform match-PM (n = 43)] were reviewed for MBL, Probing Pocket Depth (PPD) and Bleeding on Probing (BoP). MBL was calculated for first year of the insertion and the following years. Results: The median MBL for the PM implants (2.66 ± 1.67 mm; n = 43) in the first year was significantly higher than those for the other types (P =.033). The lowest rate of MBL (0.61 ± 0.44 mm; n = 18) was observed with PS implants (P =.000). The position of the crown-abutment border showed a statisticallysignificant influence (P =.019) and a negative correlation (r=-0.395) on MBL. BoP was found significantly higher in PM implants (P =.006). The lowest BoP scores were detected in PS implants, but the difference was not significant (P =.523). The relation between PPD and connection type revealed no statistically significant influence (P >.05). Conclusion: Within the limitations of the present study, it may be concluded that PS implants seem to show better peri-implant soft tissue conditions and cause less MBL.
... Accordingly, the rate of MBL as a new index was proposed by Galindo-Moreno et al. recently [69]. As remodelling of marginal bone is a dynamic process, the rate of MBL is calculated in millimetre/month (mm/m) and could change over time [69,71]. Galindo-Moreno et al. [69] found that the progression of MBL tends to be higher and the risk of implant failure could be signi cantly increased when the rates of MBL was higher than 0.44 mm at six months post-loading. ...
Preprint
Full-text available
Background: Immediate loading has recently been introduced into unsplinted mandibular implant-retained overdentures for the management of edentulous patients due to their increasing demand on immediate aesthetics and function. However, there is still a scarcity of meta-analytical evidence on the efficacy of immediate loading compared to delayed loading in unsplinted mandibular implant-retained overdentures. The purpose of this study was to compare the marginal bone loss (MBL) around implants between immediate and delayed loading of unsplinted mandibular implant-retained overdentures. Methods: Randomized controlled trials (RCTs), controlled clinical trials (CCTs), and cohort studies quantitatively comparing the MBL around implants between immediate loading protocol (ILP) and delayed loading protocol (DLP) of unsplinted mandibular overdentures were included. A systematic search was carried out in PubMed, EMBASE, and CENTRAL databases on December 02, 2020. “Grey” literature was also searched. A meta-analysis was conducted to compare the pooled MBL of two different loading protocols of unsplinted mandibular overdentures through weighted mean differences (WMDs) with 95% confidence intervals (95% CIs). The subgroup analysis was performed between different attachment types (i.e. Locator attachment vs. ball anchor). The risk of bias within and across studies were assessed using the Cochrane Collaboration’s tool, the Newcastle-Ottawa scale, and Egger’s test. Results: Of 328 records, five RCTs and two cohort studies were included and evaluated, which totally contained 191 participants with a follow-up of no less than 12 months. The MBL of ILP group showed no significant difference with that of DLP group (WMD 0.04, CI -0.13 to 0.21, P > .05). The subgroup analysis revealed similar results with Locator attachments or ball anchors (P > .05). Apart from one RCT (20%) with a high risk of bias, four RCTs (80%) showed a moderate risk of bias. Two prospective cohort studies were proved with acceptable quality. Seven included studies have reported ten implant failures in ILP groups and two implant failures in DLP groups in total. Conclusions: For unsplinted mandibular implant-retained overdentures, the MBL around implants after ILP seems comparable to that of implants after DLP. Immediate loading may be a promising alternative to delayed loading for the management of unsplinted mandibular implant-retained overdentures. PROSPERO registration number: CRD42020159124
... Accordingly, the rate of MBL as a new index was proposed by Galindo-Moreno et al. recently [67]. As remodelling of marginal bone is a dynamic process, the rate of MBL is calculated in millimetre/month (mm/m) and could change over time [67,68]. Galindo-Moreno et al. [67] found that the progression of MBL tends to be higher and the risk of implant failure could be signi cantly increased when the rates of MBL was higher than 0.44 mm at six months post-loading. ...
Preprint
Full-text available
Background: Immediate loading has recently been introduced into unsplinted mandibular implant-retained overdentures for the management of edentulous patients due to their increasing demand on immediate aesthetics and function. However, there is still a scarcity of meta-analytical evidence on the efficacy of immediate loading compared to delayed loading in unsplinted mandibular implant-retained overdentures. The purpose of this study was to compare the marginal bone loss (MBL) around implants between immediate and delayed loading of unsplinted mandibular implant-retained overdentures. Methods: Randomized controlled trials (RCTs), controlled clinical trials (CCTs), and cohort studies quantitatively comparing the MBL around implants between immediate loading protocol (ILP) and delayed loading protocol (DLP) of unsplinted mandibular overdentures were included. A systematic search was carried out in PubMed, EMBASE, and CENTRAL databases on April 28, 2020. “Grey” literature was also searched. A meta-analysis was conducted to compare the pooled MBL of two different loading protocols of unsplinted mandibular overdentures through weighted mean differences with 95% confidence intervals. The subgroup analysis was performed between different attachment types ( i.e. Locator attachment vs. ball anchor). The risk of bias within and across studies were assessed using the Cochrane Collaboration’s tool, the Newcastle-Ottawa scale, and Egger's test. Results: Of 305 records, five RCTs and two cohort studies were included and evaluated, which totally contained 191 participants with a follow-up of no less than 12 months. The MBL of ILP group showed no significant difference with that of DLP group (WMD0.04, CI -0.13 to 0.21, P > .05). Subgroup analysis revealed similar results between the two different loading groups restored with Locator attachments or ball anchors ( P > .05). Apart from one RCT (20%) with a high risk of bias, four RCTs (80%) showed a moderate risk of bias. Two prospective cohort studies were proved with acceptable quality. Furthermore, the Egger's test indicated no significant bias among seven included studies ( P > .05). Conclusions: For unsplinted mandibular implant-retained overdentures, the MBL around implants after ILP seems comparable to that of implants after DLP. Immediate loading may be a promising alternative to delayed loading for the management of unsplinted mandibular implant-retained overdentures.
... Accordingly, the rate of MBL as a new index was proposed by Galindo-Moreno et al. recently [67]. As remodelling of marginal bone is a dynamic process, the rate of MBL is calculated in millimetre/month (mm/m) and could change over time [67,68]. Galindo-Moreno et al. [67] found that the progression of MBL tends to be higher and the risk of implant failure could be signi cantly increased when the rates of MBL was higher than 0.44 mm at six months post-loading. ...
Preprint
Full-text available
Background: Immediate loading has recently been introduced into unsplinted mandibular implant-retained overdentures for the management of edentulous patients due to their increasing demand on immediate aesthetics and function. However, there is still a scarcity of meta-analytical evidence on the efficacy of immediate loading compared to delayed loading in unsplinted mandibular implant-retained overdentures. The purpose of this study was to compare the marginal bone loss (MBL) around implants between immediate and delayed loading of unsplinted mandibular implant-retained overdentures. Methods: Randomized controlled trials (RCTs), controlled clinical trials (CCTs), and cohort studies quantitatively comparing the MBL around implants between immediate loading protocol (ILP) and delayed loading protocol (DLP) of unsplinted mandibular overdentures were included. A systematic search was carried out in PubMed, EMBASE, and CENTRAL databases on April 28, 2020. “Grey” literature was also searched. A meta-analysis was conducted to compare the pooled MBL of two different loading protocols of unsplinted mandibular overdentures through weighted mean differences (WMDs) with 95% confidence intervals (95% CIs). The subgroup analysis was performed between different attachment types (i.e. Locator attachment vs. ball anchor). The risk of bias within and across studies were assessed using the Cochrane Collaboration’s tool, the Newcastle-Ottawa scale, and Egger’s test. Results: Of 305 records, five RCTs and two cohort studies were included and evaluated, which totally contained 191 participants with a follow-up of no less than 12 months. The MBL of ILP group showed no significant difference with that of DLP group (WMD 0.04, CI -0.13 to 0.21, P > .05). The subgroup analysis revealed similar results between the two different loading groups restored with Locator attachments or ball anchors (P > .05). Apart from one RCT (20%) with a high risk of bias, four RCTs (80%) showed a moderate risk of bias. Two prospective cohort studies were proved with acceptable quality. Furthermore, the Egger’s test indicated no significant bias among seven included studies (P > .05). Conclusions: For unsplinted mandibular implant-retained overdentures, the MBL around implants after ILP seems comparable to that of implants after DLP. Immediate loading may be a promising alternative to delayed loading for the management of unsplinted mandibular implant-retained overdentures. PROSPERO registration number: CRD42020159124
... Overall, the rate of marginal bone loss reduced with increasing time. Implants that showed greater bone loss levels preloading had worse actual bone levels post-loading [147]. In addition, molar sites are more prone to bone loss than premolar sites [148]. ...
Article
Full-text available
Purpose of Review This manuscript reviews the literature concerning risk factors associated with the development and/or progression peri-implantitis, focusing on studies published within the last 3 years (2017–2020). For the purpose of this review, all factors that can potentially contribute to the development of peri-implantitis will be considered “risk factors.” Recent Findings Recent studies have focused on evaluating various risk factors associated with the development of peri-implantitis. Research shows that peri-implantitis lesions are associated with complex microbial biofilms consisting of not only periodontal pathogens, but also certain unique bacteria and other microorganisms such as viruses, yeasts, and parasites. Recent evidence reinforces the role of previously well-established risk factors in the pathogenesis of peri-implantitis such as smoking, diabetes, lack of oral hygiene and maintenance, history of periodontitis, and poor peri-implant soft tissue quality. Bone quality, obesity, metabolic syndrome, implant surface characteristics, and placement depth have also been reported to be predisposing factors for the development of peri-implantitis. Few studies suggest that factors like certain medications, age, gender, vitamin D, and autoimmune diseases also play a role, but are currently not well-understood. The role of genetics is still unclear, but studies show that certain polymorphisms may be associated with peri-implantitis. Prosthetic risk factors such as improper restorative design, occlusal overload, microgap, and residual cement are significant as well. A recently emerging risk factor for peri-implantitis is the presence of peri-implant tissue-bound titanium particles. Summary Several risk factors have been associated with peri-implantitis in the literature published over the past 3 years. While some risk factors such as smoking, diabetes, history of periodontitis, and some restorative factors are well-recognized, there is still a need for well-designed randomized controlled trials and longitudinal studies in order to establish the association of some other more recently emerging risk factors for peri-implantitis.
... A recent retrospective study on 174 patients [37] that evaluated the marginal bone loss around implants with laser micro-grooved collar found mean peri-implant bone resorption of 0.18 ± 0.7 mm at the mesial aspect and 0.19 ± 0.6 mm at the distal aspect. Similar results were also reported from the study of Acharya et al. [38], in which the authors performed an exploratory analysis of annual rates of peri-implant marginal bone loss using the same three radiographical intervals used by the present retrospective study (imme- [39]. Another study evaluating peri-implant bone level changes around surface-modified implants reported a mean bone loss of 0.36 mm from the time of implant placement for implants in function for a mean 32 months [40]. ...
Article
Full-text available
Purpose: The aim of the present retrospective study was to evaluate clinical and radiological outcomes, in terms of implant survival rate, marginal bone loss, and peri-implantitis incidence, of a titanium implants with an innovative laser-treated surface. Materials and methods: A total of 502 dental implants were inserted in four dental practices (Udine, Arezzo, Frascati, Roma) between 2008 and 2013. All inserted implants had laser-modified surface characterized by a series of 20-μm-diameter holes (7-10 μm deep) every 10 μm (Synthegra®, Geass srl, Italy). The minimum follow-up period was set at 1 year after the final restoration. Radiographs were taken after implant insertion (T0), at time of loading (T1), and during the follow-up period (last recall, T2). Marginal bone loss and peri-implant disease incidence were recorded. Results: A total of 502 implants with a maximum follow-up period of 6 years were monitored. The mean differential between T0 and T2 was 0.05 ± 1.08 mm at the mesial aspect and 0.08 ± 1.11 mm at the distal with a mean follow-up period of 35.76 ± 18.05 months. After being in function for 1 to 6 years, implants reported varying behavior: 8.8% of sites did not show any radiographic changes and 38.5% of sites showed bone resorption. The bone appeared to have been growing coronally in 50.7% of the sites measured. Conclusion: Implants showed a maintenance of marginal bone levels over time, and in many cases, it seems that laser-modified implant surface could promote a bone growth. The low peri-implant disease incidence recorded could be attributed to the laser titanium surface features that seem to prevent bacterial colonization. Future randomized and controlled studies are needed to confirm the results of the present multi-centrical retrospective analysis.
Article
Full-text available
Dental implant macro- and micro-shape should be designed to maximize the delivery of optimal favorable stresses in the surrounding bone region. The present study aimed to evaluate the stress distribution in cortical and cancellous bone surrounding two models of dental implants with the same diameter and length (4.0 × 11 mm) and different implant/neck design and thread patterns. Sample A was a standard cylindric implant with cylindric neck and V-shaped threads, and sample B was a new conical implant with reverse conical neck and with “nest shape” thread design, optimized for the favorable stress distribution in the peri-implant marginal bone region. Materials and methods: The three-dimensional model was composed of trabecular and cortical bone corresponding to the first premolar mandibular region. The response to static forces on the samples A and B were compared by finite element analysis (FEA) using an axial load of 100 N and an oblique load of 223.6 N (resulting from a vertical load of 100 N and a horizontal load of 200 N). Results: Both samples provided acceptable results under loadings, but the model B implant design showed lower strain values than the model A implant design, especially in cortical bone surrounding the neck region of the implant. Conclusions: Within the limitation of the present study, analyses suggest that the new dental implant design may minimize the transfer of stress to the peri-implant cortical bone.
Article
Full-text available
Objectives This retrospective study assessed radiographic bone changes and prevalence of peri‐implant inflammation around teeth and neighboring implants supporting a single‐unit fixed dental prosthesis (FDP), in relation to implant‐ positioning and characteristics. Material and Methods Patients with an implant‐supported FDP in function for at least 1 year were recruited. The radiographic horizontal and vertical position of the implants was identified. Probing depth (PD), bleeding on probing (BOP) and radiographic bone level around implants and adjacent teeth at the time of placement, prosthesis delivery, and the most recent review were assessed. Results 98 patients with 195 implants were evaluated for a mean of 37.8 months. Survival rate was 99.6% and success ranged from 31.3% to 91.3% when different success criteria were utilized. Significantly greater interproximal bone loss around teeth and higher prevalence of interproximal peri‐implant inflammation occurred when the horizontal distance of BL implants was <1mm, but not with TL implants. There was no significant impact of the corono‐apical positioning of the implants on marginal bone loss. Conclusion Proximity of implants to adjacent teeth of <1mm leads to increased prevalence of inflammation and interproximal bone resorption at the teeth adjacent to bone level implants. This article is protected by copyright. All rights reserved.
Article
Full-text available
A classification for peri‐implant diseases and conditions was presented. Focused questions on the characteristics of peri‐implant health, peri‐implant mucositis, peri‐implantitis, and soft‐ and hard‐tissue deficiencies were addressed. Peri‐implant health is characterized by the absence of erythema, bleeding on probing, swelling, and suppuration. It is not possible to define a range of probing depths compatible with health; Peri‐implant health can exist around implants with reduced bone support. The main clinical characteristic of peri‐implant mucositis is bleeding on gentle probing. Erythema, swelling, and/or suppuration may also be present. An increase in probing depth is often observed in the presence of peri‐implant mucositis due to swelling or decrease in probing resistance. There is strong evidence from animal and human experimental studies that plaque is the etiological factor for peri‐implant mucositis. Peri‐implantitis is a plaque‐associated pathological condition occurring in tissues around dental implants, characterized by inflammation in the peri‐implant mucosa and subsequent progressive loss of supporting bone. Peri‐implantitis sites exhibit clinical signs of inflammation, bleeding on probing, and/or suppuration, increased probing depths and/or recession of the mucosal margin in addition to radiographic bone loss. The evidence is equivocal regarding the effect of keratinized mucosa on the long‐term health of the peri‐implant tissue. It appears, however, that keratinized mucosa may have advantages regarding patient comfort and ease of plaque removal. Case definitions in day‐to‐day clinical practice and in epidemiological or disease‐surveillance studies for peri‐implant health, peri‐implant mucositis, and peri‐implantitis were introduced. The proposed case definitions should be viewed within the context that there is no generic implant and that there are numerous implant designs with different surface characteristics, surgical and loading protocols. It is recommended that the clinician obtain baseline radiographic and probing measurements following the completion of the implant‐supported prosthesis.
Article
Full-text available
Aim: The aim of this article is to systematically review the effect of subcrestal implant placement compared with equicrestal position on hard and soft tissues around dental implants with platform switch. Material and methods: A manual and electronic search (National Library of Medicine and Cochrane Central Register of Controlled Trials) was performed for animal and human studies published up to December 2016. Primary outcome variable was marginal bone level (MBL) and secondary outcomes were crestal bone level (CBL), soft tissue dimensions (barrier epithelium, connective tissue, and peri-implant mucosa), and changes in the position of soft tissue margin. For primary and secondary outcomes, data reporting mean values and standard deviations of each study were extracted and weighted mean differences (WMDs) and 95% confidence intervals (CIs) were calculated. Results: A total of 14 publications were included (7 human studies and 7 animal investigations). The results from the meta-analyses have shown that subcrestal implants, when compared with implants placed in an equicrestal position, exhibited less MBL changes (human studies: WMD = - 0.18 mm; 95% CI = - 1.31 to 0.95; P = 0.75; animal studies: WMD = - 0.45 mm; 95% CI = - 0.66 to - 0.24; P < 0.001). Furthermore, the CBL was located at a more coronal position in subcrestal implants with respect to the implant shoulder (WMD = - 1.09 mm; 95% CI = - 1.43 to - 0.75; P < 0.001). The dimensions of the peri-implant mucosa seem to be affected by the positioning of the microgap and were greater at implants placed in a subcrestal position than those inserted equicrestally (WMD = 0.60 mm; 95% CI = 0.26 to 0.95; P < 0.001). While the length of the barrier epithelium was significantly greater in implants placed in a subcrestal position (WMD = 0.39 mm; 95% CI = 0.19 to 0.58; P < 0.001), no statistical significant differences were observed between equicrestal and subcrestal implant positioning for the connective tissue length (WMD = 0.17 mm; 95% CI = - 0.03 to 0.36; P = 0.10). Conclusion: This systematic review suggests that PS implants placed in a subcrestal position have less MBL changes when compared with implants placed equicrestally. Furthermore, the location of the microgap seems to have an influence on the dimensions of peri-implant soft tissues. Clinical relevance When compared with PS placed in an equicrestal position, subcrestal implant positioning demonstrated less peri-implant bone remodeling.
Article
Full-text available
Objective: The aim of this randomized clinical trial was to compare the effect on the interproximal implant bone loss (IBL) of two different heights (1 and 3 mm) of definitive abutments placed at bone level implants with a platform switched design. Material and methods: Twenty-two patients received forty-four implants (6.5-10 mm length and 3.5-4 mm diameter) to replace at least two adjacent missing teeth, one bridge set to each patient-two implants per bridge. Patients were randomly allocated, and two different abutment heights, 1 and 3 mm using only one abutment height per bridge, were used. Clinical and radiological measurements were performed at 3 and 6 months after surgery. Interproximal bone level changes were compared between treatment groups. The association between IBL and categorical variables (history of periodontitis, smoking, implant location, implant diameter, implant length, insertion torque, width of keratinized mucosa, bone density, gingival biotype and antagonist) was also performed. Results: At 3 months, implants with a 1-mm abutment had significantly greater IBL (0.83 ± 0.19 mm) compared to implants with a 3-mm abutment (0.14 ± 0.08 mm). At 6 months, a greater IBL was observed at implants with 1-mm abutments compared to implants with 3-mm abutments (0.91 ± 0.19 vs. 0.11 ± 0.09 mm). The analysis of the relation between patient characteristics and clinical variables with IBL revealed no significant differences at any moment except for smoking. Conclusions: Abutment height is an important factor to maintain interproximal implant bone level in early healing. Short abutments led to a greater interproximal bone loss in comparison with long abutments after 6 months. Other variables except smoking showed no relation with interproximal bone loss in early healing.
Article
Full-text available
The aim of this study was to compare implant failure and radiographic bone level changes with different loading protocols for unsplinted two-implant-supported mandibular overdentures. An electronic search of two databases (PubMed, Cochrane Library) was performed, without language restriction, to identify randomized controlled trials (RCTs) comparing immediate or early versus conventional dental implant loading protocols for unsplinted two-implant-supported mandibular overdentures. Data were extracted independently by two reviewers. The Cochrane tool was used to assess the quality of included studies. A meta-analysis was performed. Eight RCTs were identified, seven of which were included; one trial was excluded because related outcomes were not measured. Four of the seven studies were considered to have a high risk of bias and three an unclear risk. Meta-analysis revealed no difference between immediate versus conventional or early versus conventional implant loading protocols regarding implant failure (risk difference (RD) -0.02, 95% confidence interval (CI) -0.13 to 0.10; RD 0.09, 95% CI -0.03 to 0.20) or marginal bone loss (mean difference (MD) 0.09, 95% CI -0.10 to 0.28; MD -0.05, 95% CI -0.12 to 0.02) for implants supporting mandibular overdentures. These findings should be interpreted with great caution given the serious numerical limitations of the studies included.
Article
Full-text available
In this 3-year follow-up study, peri-implant bone loss at bone-level implants was evaluated with two definitive abutment heights: 1 mm and 2.5 mm. Peri-implant bone loss was defined as the distance between the implant shoulder and the first bone-to-implant contact from the time of loading to the 36-month follow-up, estimated using periapical radiographs. The bone loss was increased at the time of follow-up, to 1.30 mm (95% confidence interval [CI]: 0.70-1.89 mm; SD = 1.89) and 0.33 mm (95% CI: 0.11-0.55; SD = 0.59) at 36 months in short and long abutments, respectively. Placement of short abutments induced higher periimplant bone loss at bone-level implants during a peri-implant recall program.
Article
The objective was to test the hypothesis of no difference in the treatment outcome after the installation of short implants (≤8mm) in the posterior part of the maxilla compared to standard length implants (>8mm) in conjunction with maxillary sinus floor augmentation (MSFA) using the lateral window technique, after an observation period of ≥3years. A search of the MEDLINE, Embase, and Cochrane Library databases, in combination with a hand-search of relevant journals, was conducted. The search yielded 1102 titles. Finally, three studies that fulfilled the inclusion criteria were included. All were considered to have a low risk of bias. Meta-analyses revealed no significant differences in implant survival or peri-implant marginal bone loss between the two treatment modalities. However, the use of standard length implants in conjunction with MSFA was characterized by a tendency towards more peri-implant marginal bone loss. There was no statistically significant difference between the two treatment modalities with regard to overall patient satisfaction. Short implants seem to be a suitable alternative to standard length implants in conjunction with MSFA. However, further randomized controlled trials with larger patient samples and an observation period of more than 3years are needed before one treatment modality might be considered superior to the other.
Article
The objective of this review is to identify case definitions and clinical criteria of peri‐implant healthy tissues, peri‐implant mucositis, and peri‐implantitis. The case definitions were constructed based on a review of the evidence applicable for diagnostic considerations. In summary, the diagnostic definition of peri‐implant health is based on the following criteria: 1) absence of peri‐implant signs of soft tissue inflammation (redness, swelling, profuse bleeding on probing), and 2) the absence of further additional bone loss following initial healing. The diagnostic definition of peri‐implant mucositis is based on following criteria: 1) presence of peri‐implant signs of inflammation (redness, swelling, line or drop of bleeding within 30 seconds following probing), combined with 2) no additional bone loss following initial healing. The clinical definition of peri‐implantitis is based on following criteria: 1) presence of peri‐implant signs of inflammation, 2) radiographic evidence of bone loss following initial healing, and 3) increasing probing depth as compared to probing depth values collected after placement of the prosthetic reconstruction. In the absence of previous radiographs, radiographic bone level ≥3 mm in combination with BOP and probing depths ≥6 mm is indicative of peri‐implantitis.
Article
Background It is postulated that peri‐implant sulcular fluid (PISF) levels of advanced glycation end products (AGEs) are higher with high glycemic levels. Purpose In the present clinico‐biochemical study, we explored the clinical and radiographic peri‐implant parameters and levels of AGEs among prediabetic, type 2 diabetic (T2DM), and non‐diabetic patients and to evaluate the correlation of AGEs with clinical peri‐implant parameters. Materials and Methods Ninety patients were divided into three groups of 30 patients each; group 1: patients with prediabetes; group 2: patients with T2DM; and group 3: non‐diabetic individuals. Clinical and radiographic peri‐implant parameters assessed included plaque index (PI), bleeding on probing (BOP), probing depth (PD), and marginal bone loss (MBL). PISF was collected and analyzed for AGEs levels using enzyme‐linked immunosorbent assay. Between‐group comparison of means was verified with Kruskal‐Wallis test and Pearson correlation coefficient for correlations of AGE levels with peri‐implant parameters. Results Mean peri‐implant PI, BOP, PD, and MBL was significantly higher in group 1 and 2 as compared with non‐diabetic patients (P < .05). Mean PI, BOP, PD, and MBL were comparable between group 1 and group 2 patients (P > .05). Mean levels of AGEs in PISF were significantly higher among prediabetic and T2DM patients as compared with non‐diabetic patients (P < .05). Between group 1 and group 2, mean levels of AGEs was significantly higher in group 2 (P < .05). A significant positive correlations were found between levels of AGEs and PD (P = .0371) and MBL (P = .0117) in T2DM patients, respectively. Conclusion Clinical and radiographic peri‐implant parameters were worse and levels of AGEs in PISF were increased in individuals with prediabetes and T2DM. AGEs may play an important role in peri‐implant inflammation in prediabetes and T2DM.