Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
ScienceDirect
journal homepage: www.intl.elsevierhealth.com/journals/jden
Review
article info
abstract
Article history:
Objectives: To test the null hypothesis of no difference in the implant failure rates, postop-
erative infection, and marginal bone loss for the insertion of dental implants in periodon-
28 August 2014
Keywords:
estimates of relative effect were expressed in risk ratio (RR) and mean difference (MD) in
Dental implants
millimetres. All studies were judged to be at high risk of bias, none were randomized. A total
Periodontal disease
of 10,927 dental implants were inserted in PCPs (587 failures; 5.37%), and 5881 implants in
Periodontitis
PHPs (226 failures; 3.84%). The difference between the patients significantly affected the
implant failure rates (RR 1.78, 95% CI 1.502.11; P < 0.00001), also observed when only
Postoperative infection
the controlled clinical trials were pooled (RR 1.97, 95% CI 1.382.80; P = 0.0002). There were
Meta-analysis
infections (RR 3.24, 95% CI 1.696.21; P = 0.0004) and in marginal bone loss (MD 0.60, 95% CI
0.330.87; P < 0.0001) when compared to PHPs.
Conclusions: The present study suggests that an increased susceptibility for periodontitis
may also translate to an increased susceptibility for implant loss, loss of supporting bone,
and postoperative infection. The results should be interpreted with caution due to the
presence of uncontrolled confounding factors in the included studies, none of them
randomized.
Clinical Significance: There is some evidence that patients treated for periodontitis may
experience more implant loss and complications around implants including higher bone loss
and peri-implantitis than non-periodontitis patients. As the philosophies of treatment may
alter over time, a periodic review of the different concepts is necessary to refine techniques
and eliminate unnecessary procedures. This would form a basis for optimum treatment.
# 2014 Elsevier Ltd. All rights reserved.
* Corresponding author at: Department of Prosthodontics, Faculty of Odontology, Malmo University, Carl Gustafs vag 34, SE-205 06, Malmo,
Sweden. Tel.: +46 725 541 545; fax: +46 40 6658503.
E-mail addresses: bruno.chrcanovic@mah.se, brunochrcanovic@hotmail.com (B.R. Chrcanovic).
http://dx.doi.org/10.1016/j.jdent.2014.09.013
0300-5712/# 2014 Elsevier Ltd. All rights reserved.
1510
1.
Introduction
2.
2.1.
Objective
2.2.
Search strategies
2.3.
2.4.
Study selection
2.5.
Quality assessment
2.6.
1511
3.
Results
3.1.
Literature search
3.2.
Study
Published
Study design
Patients (n)
(number per group)
1996
RA (unicenter)
1672 (32.9)
2000
RA (multicenter)
1690 (53)
2000
CCT (multicenter)
143 (NM)
2000
CCT (unicenter)
2002
RA (unicenter)
2003
CCT (unicenter)
2004
Healing period/
loading
Failed/placed
implants (n)
Implant
failure
rate (%)
6 months (maxilla)
4 months (mandible)
6 months
5/62 (G1)
2/47 (G2)
39/375 (G1)
29/647 (G2)
8.06 (G1)
4.26 (G2)
10.40 (G1)
4.48 (G2)
14/98 (G1)
3/166 (G2)
0/7 (G1)
0/26 (G2)
8/100 (G1)
3/92 (G2)
2/21 (G1)
14.29 (G1)
1.81 (G2)
0 (G1)
0 (G2)
8 (G1)
3.26 (G2)
9.52 (G1)
3/91 (G2)
16/77 (G1)
6/72 (G2)
86/923 (G1)
37/588 (G2)
2/120 (G1a)
3.30 (G2)
20.78 (G1)
8.33 (G2)
9.32 (G1)
6.29 (G2)
1.67 (G1)
0/30 (G2)
10/122 (G1)
67/1803 (G2)
0 (G2)
8.20 (G1)
3.72 (G2)
1/41 (G1)
0/13 (G2)
17/68 (G1)
2.44 (G1)
0 (G2)
25 (G1)
0/2 (G2)
2/155 (G1b)
0/72 (G2)
16/252 (G1a)
8/261 (G2)
61/1171 (G1)
0 (G2)
1.29 (G1)
0 (G2)
6.35 (G1)
3.07 (G2)
5.21 (G1)
16/455 (G2)
204/5346 (G1c)
15/497 (G2)
4/40 (G1)
2/40 (G2)
5/34 (G1)
9/97 (G2)
3.52 (G2)
3.82 (G1)
3.02 (G2)
10 (G1)
5 (G2)
14.71 (G1)
9.28 (G2)
3/20 (G1)
1/20 (G2)
70/1512 (G1d)
23/747 (G2)
16/185 (G1b)
2/61 (G2)
15 (G1)
5 (G2)
4.63 (G1)
3.08 (G2)
8.65 (G1)
3.28 (G2)
6/198 (G1b)
0/54 (G2)
3.03 (G1)
0 (G2)
Antibiotics/
mouth rinse
(days)
10/NM
NM (47, females)
NM (40, males)
2263 (NM)
1, 3, and 5 years
NM
4 years
NM
6 months (maxilla)
34 months (mandible)
delayed
5 years
NM
NM
NM (53.5, G1)
NM (57.3, G2)
NM
10 years
NM
46 months
RA (unicenter)
NM
184030 days
NM
46 months
2004
RA (unicenter)
13 years
710/28 days
59 months
2005
CCT (unicenter)
NM (61.1, G1)
NM (49.5, G2)
1959 (3234, G1; 31, G2)
3 years
NM
NM
Flores-de-Jacoby [17]
Wagenberg
and Froum [28]
2006
RA (unicenter)
891 (NM)
1494 (57.9)
Mean 71 months
(range 12193)
12/NM
2007
CCT (unicenter)
NM
2008
RA (multicenter)
NM (48)
Every 3 months
over a 3-year period
Mean 13.4 years
2008
CCT (multicenter)
2009
CCT (unicenter)
2010
RA (unicenter)
NM (52)
(range 819)
5 years
Mean 46.8 (G1) and
48.1 months (G2)
Mean 30 months
NM
NM/14
6 months (maxilla)
3 months (mandible)
6 months (maxilla)
3 months (mandible)
(64 immediately loaded)
6 months (maxilla)
3 months (mandible)
NM
0/7
NM
NM
(range 1114)
5 years
35/14
10 years
NM
2010
RA (multicenter)
2010
RA (multicenter)
2010
RA (unicenter)
55 (NM)
2988 (68.7)
1016 years
710/NM
34 months
2011
RA (multicenter)
NM (51)
10 years
NM
46 months
2011
CCT (unicenter)
NM
2012
CCT (unicenter)
Mean 54 months
(up to 144 months)
10 years
NM
NM (54, G1)
NM (46, G2)
NM
NM
36 months
2014
CCT (unicenter)
NM (53, G1)
NM (43, G2)
10 years
NM
612 weeks
Rosenquist and
Grenthe [23]
Brocard et al. [24]
Follow-up visits
(or range)
1512
Study
Published
Study design
P value
(for failure
rate)
Postoperative
infection
P value (for
postoperative
infection)
Marginal
bone loss
(mean SD) (mm)
Periodontal
therapy
Periodontal
disease
definitions
Implant surface
modification
(brand)
Rosenquist and
Grenthe [23]
1996
RA (unicenter)
NM
4 (G1)
1 (G2)
NM
NM
Turned
(Nobelpharma,
Nobelpharma AB,
Goteborg, Sweden
Extraction indication
had been
periodontitis
NM
Brocard
et al. [24]
2000
RA (multicenter)
NM
NM
NM
NM
Prior to implant
placement, some
patients were treated
for periodontal
disease
Prior to implant
placement, the G1
patients were
treated for
periodontal disease.
This involved a
Waldenburg,
Switzerland)
Observations
All implants in
fresh extraction
sockets, use of
membranes in 5
patients
132 smokers, 66
bruxers, 177
sites with GBR
followed in some
cases by periodontal
surgery. All patients
were enrolled in a
periodontal
maintenance
programme with
Polizzi
et al. [14]
2000
CCT (multicenter)
NM
NM
NM
NM
Turned (Branemark,
Nobel Biocare AB,
Goteborg, Sweden)
Periodontitis cited as
a reason for tooth
extraction, history
of periodontitis before
regular professional
plaque control
NM
tooth extraction
Watson
et al. [15]
2000
CCT (unicenter)
NM
NM
NM
NM
Hydroxyapatitiecoated (Calcitek
omniloc, Carlsbad,
Chronic periodontitis:
pockets 4 mm and
radiographic bone
USA)
loss
NM
146 implants in
fresh extraction
sockets,
membranes
used in 64
implants, 8
grafts
Smokers were
included, but the
exact number
hygienic phase
consisting of
scaling, root
planning, and oral
hygiene
instructions,
was not
informed, no
grafts, only
single-tooth
restorations
1513
1514
Table 1 (Continued )
Study
Hardt
et al. [25]
2002
2003
Study design
RA (unicenter)
CCT (unicenter)
P value
(for failure
rate)
Postoperative
infection
P value (for
postoperative
infection)
NM
NM
NM
NM
NM
NM
et al. [16]
Evian
et al. [26]
2004
RA (unicenter)
NM
NM
NM
Marginal
bone loss
(mean SD) (mm)
2.2 0.8 (G1)
1.7 0.8 (G2)
Implant surface
modification
(brand)
Periodontal
disease
definitions
Turned (Branemark,
Nobel Biocare AB,
Goteborg, Sweden)
An overall descriptor
of the patients
experience of
periodontal
destruction before the
time of implant
therapy was
generated through
the calculation of an
age-related
periodontal bone loss
score
Patients having lost
ITI, Straumann,
Waldenburg,
Switzerland)
NM
? (Paragon, Zimmer
Dental, Carlsbad,
USA)
Periodontal disease
was diagnosed if
probing depths were
5 mm or greater and
associated with
radiographic signs of
bone loss. Patients
who exhibited 1 or
more teeth with
periodontal disease,
or who originally lost
their teeth as a result
of periodontitis, were
considered to have
periodontal disease
Periodontal
therapy
Observations
NM
Fixed partial
dentures in the
maxillary
posterior
segments, no
grafts
28 implants
placed in 12
smokers (10 in
G1, 18 in G2) and
in 41 nonsmokers (11 in
G1, 73 in G2)
implants
Periodontal
treatment was
performed prior to
or in conjunction
with implant
placement
Only patients
who received a
single implant
Karoussis
Published
Rosenberg
et al. [27]
2005
RA (unicenter)
CCT (unicenter)
NM
NM
NM
NM
NM
NM
NM
Turned (Branemark,
Nobel Biocare AB,
Goteborg, Sweden),
TPS and SLA (ITI,
Straumann,
Waldenburg,
Switzerland), TPS
Patients were
classified as
periodontally
compromised if they
had a history of
periodontal disease
that resulted in tooth
Prior to implant
placement, all
necessary
periodontal,
restorative, and
endodontic
treatment was
completed,
including extraction
of hopeless teeth
coated (Swede-vent,
Screw-vent,
Corevent, Paragon,
Encino, USA)
of attachment (with
the exception of facial
or lingual recession)
or probing depth
greater than 34 mm
was present at the
time of implant
Turned (Mk II
Branemark, Nobel
Biocare AB,
Goteborg, Sweden;
n = 83), acid-etched
(G1 GA)
0.18 0.11 (G1 GC)
0.12 0.08 (G2)
(Osseotite, 3i
Implant
Innovations, Palm
Beach Gardens,
USA; n = 67)
placement
The diagnosis of
generalized chronic
and aggressive
periodontitis was
based on the
American Academy of
Periodontology
criteria
All patients
underwent
periodontal surgery
and were entered
into a 3-month
No smokers
Wagenberg
and Froum
[28]
2006
RA (unicenter)
0.02
NM
NM
NM
Turned (Branemark,
Nobel Biocare AB,
Goteborg, Sweden,
n = 1398), acidetched (Osseotite, 3i
Implant
Innovations, Palm
Beach Gardens,
USA; n = 527)
Mengel and
Flores-deJacoby [17]
2004
All implants
placed in fresh
extraction
sockets, bone
grafts were
utilized in all
cases in which
there was a
residual space
around the
implant, 13
implants in
sinus-lifts, 323
1515
implants in
smokers
Study
Mengel
et al. [18]
Published
2007
Study design
CCT (unicenter)
P value
(for failure
rate)
Postoperative
infection
P value (for
postoperative
infection)
NM
NM
NM
Periodontal
therapy
Observations
Marginal
bone loss
(mean SD) (mm)
Implant surface
modification
(brand)
Periodontal
disease
definitions
Acid-etched
(Osseotite, BIOMET/
3i, Palm Beach
Gardens, USA)
The diagnosis of
generalized
aggressive
periodontitis was
based on the
All generalized
aggressive
periodontitis
patients underwent
periodontal
Edentulous
patients. G1
patients were
fitted with
removable
American Academy of
Periodontology
criteria
implant-tooth
supported
superstructures,
G2 patients
received either
fixed cemented
Fardal and
Linden [5]
2008
2008
RA (multicenter)
CCT (multicenter)
NM
NM
5 (G1)
0 (G2)
4 (G1)
0 (G2)
NM
NM
NM
NM
Patients received
initial periodontal
therapy, followed by
at least 8 years of
maintenance
treatment in the
degeneration of the
periodontium despite
ongoing sanative,
surgical and/or
pharmacological
therapy
specialist practice
Several (Nobel
Biocare,
Gothenburg,
Sweden; Zimmer
The periodontal
conditions were
assessed using a
modification of the
Dental, Carlsbad,
USA; Mathys,
Bettlach,
Switzerland;
Straumann,
Waldenburg,
Switzerland;
Periodontal Screening
and Recording index
modification of the
Periodontal
Screening and
Recording index,
and subsequently
periodontal therapy
Dentsply Friadent,
Mannheim,
Germany)
implantsupported
dentures in the
maxilla or
single-tooth
implants. No
smokers.
1516
Table 1 (Continued )
De Boever
et al. [20]
Anner et al.
2009
2010
CCT (unicenter)
RA (unicenter)
NM
0.1498
NM
NM
NM
NM
NM
NM
Periodontally
susceptible patients
with tooth loss due to
periodontal disease
and patients with
periodontal disease
Before implant
placement, all
patients received, if
necessary,
periodontal nonsurgical and/or
11.4% of the
patients were
smokers (13
smokers in G1, 9
in G2), 10.3%
former smokers,
134 ridge
augmentations
(75, G1; 59, G2)
NM
surgical therapy.
Periodontally
susceptible patients
were enrolled in a
strict maintenance
programme
246 patients (51.7%)
participated of a
structured
supportive
periodontal
programme
Periodontal therapy
smokers
[29]
2010
RA (multicenter)
NM
NM
NM
NM
Periodontal
conditions were
assessed using a
modification of the
Periodontal Screening
and Recording (PSR)
index
The classification of
(non-surgical and
surgical) was
administered as
required
in G1, 63 in G2)
The patients
No smokers,
received
individualized
periodontal
treatment before
implant surgery. On
the basis of the
only dental
implants in a
single-unit gap
et al. [30]
Matarasso
2010
RA (multicenter)
NM
NM
NM
et al. [8]
results achieved
after the
periodontal
treatment, the
patients were
placed on an
Simonis
et al. [31]
2010
RA (unicenter)
0.327
13 (G1)
10 (G2)
0.006
TPS (ITI,
Straumann,
Waldenburg,
Switzerland; solid
screw, n = 116;
hollow screw,
n = 15)
NM
individually tailored
maintenance care
programme
All patients were
instructed on how
to maintain
appropriate oral
Gianserra
49 diabetics, 63
9 smokers, only
implantsupported fixed
restorations
1517
1518
Table 1 (Continued )
Study
Aglietta
et al. [32]
Roccuzzo
et al. [2]
2011
2011
2012
Study design
RA (multicenter)
CCT (unicenter)
CCT (unicenter)
P value
(for failure
rate)
Postoperative
infection
P value (for
postoperative
infection)
NM
NM
NM
NM
NM
NM
NM
NM
NM
Marginal
bone loss
(mean SD) (mm)
Turned, 3.47 1.09 (G1)
2.65 0.31 (G2)
TPS, 3.77 1.43 (G1)
2.51 0.31 (G2)
NM
Implant surface
modification
(brand)
Turned (Branemark,
Nobel Biocare AB,
Goteborg, Sweden;
n = 20), TPS (ITI,
Straumann,
Waldenburg,
Switzerland; n = 20)
NM
TPS (ITI,
Straumann,
Waldenburg,
Switzerland)
Periodontal
disease
definitions
Periodontal
therapy
Observations
Patients were
treated for
periodontitis
All periodontally
involved patients
had undergone
cause-related as
81 diabetics, 103
smokers (71 in
G1, 32 in G2)
periodontal diagnosis
that was based on a
classification of
periodontal diseases
18 smokers (15
in G1, 3 in G2)
Levin
et al. [21]
Published
Roccuzzo
et al. [22]
2014
CCT (unicenter)
NM
NM
NM
NM
Sandblasted and
acid-etched (SLA,
Straumann,
Waldenburg,
Switzerland)
S = Number of
pockets (57 mm) + 2
Number of pockets
(8 mm)
hygiene instruction
and scaling and root
planing, with the
aim to reduce to a
minimal level
periodontal
pathogens.
21 smokers (16
in G1, 5 in G2)
Here the implants and the patients with severe chronic periodontitis and moderate chronic periodontitis were put together in G1.
Here the number of patients and implants were considered for the patients followed for 5 years. The implants and the patients with severe periodontitis (569 patients, 2938 implants, 130 failures) and moderate periodontitis (712
patients, 2408 implants, 74 failures) at the 5-year follow-up were put together in G1. The numbers at baseline were different (1727 patients; 1469, G1, 258, G2), (severe periodontitis: 630 patients, 3260 implants, 130 failures; moderate
periodontitis: 839 patients, 2813 implants, 74 failures).
d
Here the implants and the patients with severe chronic periodontitis and moderate chronic periodontitis were put together in G1.
c
1519
1520
3.3.
Quality assessment
All studies except one23 were high quality. The scores are
summarized in Table 2.
Published
Selection
Representativeness
of the exposed
cohort
Selection of
external
control
Comparability
Ascertainment
of exposure
Outcome of
interest not
present at
start
Comparability of
cohorts
Rosenquist and
Grenthe [23]
Brocard et al. [24]
Polizzi et al. [14]
Watson et al. [15]
Hardt et al. [25]
Karoussis et al. [16]
Evian et al. [26]
Rosenberg et al. [27]
Mengel and
Flores-de-Jacoby [17]
Wagenberg and
Froum [28]
Mengel et al. [18]
Fardal and Linden [5]
Gatti et al. [19]
De Boever et al. [20]
Anner et al. [29]
Gianserra et al. [30]
Matarasso et al. [8]
Simonis et al. [31]
Aglietta et al. [32]
Levin et al. [21]
Roccuzzo et al. [2]
Roccuzzo et al. [22]
Assessment
of outcome
Follow-up
long
enougha
Total
(9/9)
Adequacy of
follow-up
Additional
factor
1996
5/9
2000
2000
2000
2002
2003
2004
2004
2005
$
0
$
0
$
0
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
0
0
0
$
$
0
$
$
$
$
$
$
$
$
$
$
0
$
$
$
$
0
0
0
$
$
0
0
0
0
8/9
7/9
7/9
7/9
7/9
7/9
8/9
6/9
2006
7/9
2007
2008
2008
2009
2010
2010
2010
2010
2011
2011
2012
2014
0
$
$
0
$
$
$
$
$
$
0
0
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
0
0
0
$
$
0
$
0
$
$
0
0
$
$
$
$
$
$
$
$
$
$
$
$
0
$
$
0
0
$
$
$
$
0
$
$
$
0
$
0
0
$
$
0
$
0
$
0
6/9
7/9
8/9
6/9
7/9
8/9
9/9
7/9
9/9
7/9
7/9
6/9
Main
factor
Outcome
a
Five years was chosen to be enough for the outcome implant failure to occur. This time point was chosen due to the fact that Roccuzzo et al.2 showed that the difference between PHP and PCP is
negligible during the first 5 years, but becomes more pronounced later on, being in accordance with the findings of Karoussis et al.,16 who first demonstrated that a 5-year follow-up is usually not
sufficient to evaluate the differences in the clinical outcomes of the various groups of patients.
1521
1522
Fig. 2 Forest plot for the event implant failure in the comparison between periodontally compromised vs. periodontally
healthy patients.
3.4.
Meta-analysis
Fig. 3 Forest plot for the event implant failure in the comparison between periodontally compromised vs. periodontally
healthy patients, when only the CCTs were pooled.
1523
Fig. 4 Forest plot for the event postoperative infection in the comparison between periodontally compromised vs.
periodontally healthy patients.
Fig. 5 Forest plot for the event marginal bone loss comparing PCPs and PHPs. 1y 1 year; 3y 3 years; turned turned
implants; TPS titanium plasma-sprayed implants.
3.5.
Publication bias
4.
Discussion
Narrowing the inclusion criteria of studies increases homogeneity but also excludes the results of more trials and thus risks
the exclusion of significant data.33 The issue is important
because meta-analyses are frequently conducted on a limited
number of RCTs. In meta-analyses such as these, adding more
information from observational studies may aid in clinical
reasoning and establish a more solid foundation for causal
inferences.33 However, potential biases are likely to be greater
for non-randomized studies compared with RCTs, so results
should always be interpreted with caution when they are
included in reviews and meta-analyses.34 The search strategy
adopted here did not find any randomized study on the
subject. Thus, the results must be interpreted carefully.
The statistical heterogeneity stands for the variability in
the intervention effects being evaluated in the different
studies, and is a consequence of clinical or methodological
diversity, or both, among the studies. The low level of
heterogeneity observed when the outcomes implant failure
and postoperative infection were analyzed is surprising,
given the variability of the included studies (varying lengths of
follow-up, patient ages, number of implants, classification of
severity of periodontitis etc.). For this reason, a randomeffects model was also used to incorporate heterogeneity
among studies, resulting in the same significance of the
treatment effects. However, it is important to stress that care
1524
1525
5.
Conclusion
Acknowledgements
This work was supported by CNPq, Conselho Nacional de
Desenvolvimento Cientfico e Tecnologico Brazil. We would
like to thank Dr. Ricardo Trindade.
references
1526
1527